Inernational tic Journal IEJ 2010 Issue January

EDITORIAL

I would like to acknowledge and thank the Associate

Editors for their outstanding contribution during 2009:

Kishor Gulabivala, Matt German, Jeremy Hayes,

Michael Hulsmann, Yuan-Ling Ng, Ove Peters,

Min-Kai Wu, Matthias Zehnder.

I would also like to acknowledge and thank the

following referees for their critical appraisal of papers

received:

Paul Abbott, Guido Aesaert, Michael Ahlquist, Anas

Al-jadaa, Antonio Apicella, Saeed Asgary, Paul Ashley,

Phil Atkin, Graham Bailey, Rafael Yague Ballester,

Michael Baumann, Michael Behr, Sema Belli, Lars

Bergmans, Matthias Bickel, Lars Bjørndal, Gilles Blu-

teau, Emre Bodrumlu, Patrick Bogaerts, George Bogen,

Fiona Boissonade, Peter Bolhuis, Tatiana Botero, Serge

Bouillaguet, Martha Brackett, Peter Briggs, Paul Brun-

ton, Josette Camilleri, Jean Camps, Dermot Canavan,

Peter Carrotte, Bruno Cavalcanti, Nick Chandler, Gary

Cheung, Bun San Chong, David Cohen, Ben Cole, Georg

Conrads, Ian Corbett, Margaret Corson, Bill Costerton,

Elisabetta Cotti, Francesco D’Aiuto, Till Dammaschke,

Camillo D’Arcangelo, Peter Day, Mieke De Bruyne,

Roeland De Moor, Carlos de Souza Costa, Gustavo

De-Deus, Chris Deery, Anibal Diogenes, Nick Donos,

Nicholas Drage, Peter Duckmanton, Johannes Ebert,

Martin Ehrbar, Ashraf ElAyouti, Paul Eleazer, George

Eliades, Chris Emery, Unni Endal, Carlos Estrela, Marco

Ferrari, Jose Figueiredo, Ashraf Fouad, Richard Foxton,

Roland Frankenberger, Inge Fristad, Massimo Gagliani,

Gianluca Gambarini, Jennifer Gibbs, Alan Gluskin,

Michel Goldberg, Brenda Gomes, Harold Goodis,

Simone Grandini, Rene Gruythuysen, James Gutmann,

Markus Haapasalo, Gunnar Hasselgren, Sivakami

Haug, Jianing He, Brian Henderson, Michael Hofmann,

Christopher Hope, Keith Horner, Preben Horsted-

Bindslev, Tony Hoskinson, George Huang, Bart

Huybrechts, Richard Kahan, Asma Khan, Andrej

Kielbassa, Eun-Cheol Kim, Lise-Lotte Kirkevang, Anil

Kishen, Elisabeth Koulaouzidou, Thomas Kvist, Paul

Lambrechts, Rachel Leeson, Jim Lewsey, Ludwig

Limbach, Shaul Lin, Christina Lindh, Howard Lloyd,

Matthew Locke, Claus Lost, Robert Love, Hans Ulrich

Luder, Phil Lumley, Pierre Machtou, Iain Mackie,

Francesco Mannocci, Monika Marending, Phil Marsh,

Paul McCabe, John McCabe, Robert McConnell, Harold

Messer, Thimios Mitsiadis, Dirk Mohn, Anders Moland-

er, Francesca Monticelli, Nicky Mordan, Peter Murray,

Akhila Muthukrishnan, P.N.R. Nair, Mohammad

Nekoofar, Jacques Nor, Takashi Okiji, Dag Ørstavik,

Ahmet Ozok, Cornelis Pameijer, Frank Paque, Peter

Parashos, Shanon Patel, Jorge Perdigao, Hiran Perin-

panayagam, Christine Peters, Linda Peters, Kerstin

Petersson, David Pitt, Heather Pitt Ford, Gianluca

Plotino, Carlo Prati, Jonathan Pratten, Alison Qualt-

rough, Ivana Radovic, Derren Ready, John Regan,

Kathrin Reichenmiller, Claes Reit, John Rhodes, Adam

Roberts, Sarah Rolland, Martin Rosentritt, Vivian

Rushton, Kamran Safavi, Chankhrit Sathorn, Julian

Satterthwaite, Bill Saunders, Edgar Schafer, Jorg

Schirrmeister, Patrick Schmidlin, Helmut Schweikl,

Geoffrey Seccombe, Christine Sedgley, Bilge Sen, Ann

Shearer, Hagay Shemesh, Sharan Sidhu, Asgeir Sig-

urdsson, Nick Silikas, Ulf Sjogren, Alastair Sloan, Carlos

Soares, David Sonntag, Manoel Sousa-Neto, Erick

Souza, Valerie Sparkes, Dave Spratt, Vidya Srinivasan,

Hideaki Suda, Pia Titterud Sunde, Mario Tanomaru-

Filho, Franklin R Tay, Peter Taylor, Fabricio Teixeira,

Leo Tjaderhane, Mirek Tolar, Muhittin Toman, Phillip

Tomson, Mahmoud Torabinejad, Dimitrios Tziafas,

lucas van der Sluis, Peter Velvart, Frank Vertucci,

Morgana Vianna, Thomas von Arx, William Walker,

Angus Walls, Damien Walmsley, Tuomas Waltimo,

Rick Walton, John Wataha, Paula Waterhouse, Roland

Weiger, Richard Welbury, Paul Wesselink, John Whit-

worth, David Witherspoon, Karl Thomas Wrbas, Peter

Yaman.

Without their commitment, dedication and expertise,

the International Endodontic Journal could not maintain

its position as the leading publication in the field of

Endodontology.

Paul M. H. Dummer

Editor-in-Chief

doi:10.1111/j.1365-2591.2009.01656.x

1ª 2010 International Endodontic Journal International Endodontic Journal, 43, 1, 2010

REVIEW

The smear layer in endodontics – a review

D. R. Violich1 & N. P. Chandler2

1Private Endodontic Practice, Tauranga, New Zealand; and 2Sir John Walsh Research Institute, School of Dentistry, University of

Otago, Dunedin, New Zealand

Abstract

Violich DR, Chandler NP. The smear layer in endodontics –

a review. International Endodontic Journal, 43, 2–15, 2010.

Root canal instrumentation produces a layer of organic

and inorganic material called the smear layer that may

also contain bacteria and their by-products. It can

prevent the penetration of intracanal medicaments into

dentinal tubules and influence the adaptation of filling

materials to canal walls. This article provides an

overview of the smear layer, focusing on its relevance

to endodontics. The PubMed database was used

initially; the reference list for smear layer featured

1277 articles, and for both smear layer dentine and

smear layer root canal revealed 1455 publications.

Smear layer endodontics disclosed 408 papers. A

forward search was undertaken on selected articles

and using some author names. Potentially relevant

material was also sought in contemporary endodontic

texts, whilst older books revealed historic information

and primary research not found electronically, such

that this paper does not represent a ‘classical’ review.

Data obtained suggests that smear layer removal

should enhance canal disinfection. Current methods

of smear removal include chemical, ultrasonic and

laser techniques – none of which are totally effective

throughout the length of all canals or are universally

accepted. If smear is to be removed, the method of

choice seems to be the alternate use of ethylenedi-

aminetetraacetic acid and sodium hypochlorite solu-

tions. Conflict remains regarding the removal of the

smear layer before filling root canals, with investiga-

tions required to determine the role of the smear layer

in the outcomes of root canal treatment.

Keywords: dentine, ethylenediaminetetraacetic acid,

endodontic treatment, smear layer.

Received 20 June 2007; accepted 21 July 2009

Introduction

Whenever dentine is cut using hand or rotary

instruments, the mineralized tissues are not shredded

or cleaved but shattered to produce considerable

quantities of debris. Much of this, made up of very

small particles of mineralized collagen matrix, is

spread over the surface to form what is called the

smear layer. Identification of the smear layer was

made possible using the electron microprobe with

scanning electron microscope (SEM) attachment, and

first reported by Eick et al. (1970). These workers

showed that the smear layer was made of particles

ranging in size from less than 0.5–15 lm. Scanning

electron microscope studies of cavity preparations by

Brannstrom & Johnson (1974) demonstrated a thin

layer of grinding debris. They estimated it to be

2–5 lm thick, extending a few micrometres into the

dentinal tubules.

The smear layer in a cavity and in the root canal

may not be directly comparable. Not only are the tools

for dentine preparation different in coronal cavities, but

in the root canal the dentinal tubule numbers show

greater variation and there are likely to be more soft

tissue remnants present. The first researchers to

describe the smear layer on the surface of instrumented

root canals were McComb & Smith (1975). They

Correspondence: Nicholas Chandler, Associate Professor,

School of Dentistry, University of Otago, P.O. Box 647,

Dunedin 9054, New Zealand (Tel.: 0064 3 479 7124; fax:

0064 3 479 5079; e-mail [email protected]).

doi:10.1111/j.1365-2591.2009.01627.x

International Endodontic Journal, 43, 2–15, 2010 ª 2010 International Endodontic Journal2

http://endodontic.ws


suggested that the smear layer consisted not only of

dentine as in the coronal smear layer, but also the

remnants of odontoblastic processes, pulp tissue and

bacteria. Lester & Boyde (1977) described the smear

layer as ‘organic matter trapped within translocated

inorganic dentine’. As it was not removed by sodium

hypochlorite irrigation, they concluded that it was

primarily composed of inorganic dentine. Goldman

et al. (1981) estimated the smear thickness at 1 lm

and agreed with previous investigators that it was

largely inorganic in composition. They noted its pres-

ence along instrumented canal surfaces. Mader et al.

(1984) reported that the smear layer thickness was

generally 1–2 lm. Cameron (1983) and Mader et al.

(1984) discussed the smear material in two parts: first,

superficial smear layer (Fig. 1) and second, the material

packed into the dentinal tubules. Packing of smear

debris was present in the tubules to a depth of 40 lm.

Brannstrom & Johnson (1974) and Mader et al. (1984)

concluded that the tubular packing phenomenon was

due to the action of burs and instruments. Components

of the smear layer can be forced into the dentinal

tubules to varying distances (Moodnik et al. 1976,

Brannstrom et al. 1980, Cengiz et al. 1990) to form

smear plugs (Fig. 2). However, Cengiz et al. (1990)

proposed that the penetration of smear material into

dentinal tubules could also be caused by capillary

action as a result of adhesive forces between the

dentinal tubules and the material. This hypothesis of

capillary action may explain the packing phenomenon

observed by Aktener et al. (1989), who showed that the

penetration could increase up to 110 lm when using

surface-active reagents in the canal during endodontic

instrumentation. The thickness may also depend on the

type and sharpness of the cutting instruments and

whether the dentine is dry or wet when cut (Barnes

1974, Gilboe et al. 1980, Cameron 1988). In the early

stages of instrumentation, the smear layer on the walls

of canals can have a relatively high organic content

because of necrotic and/or viable pulp tissue in the root

canal (Cameron 1988). Increased centrifugal forces

resulting from the movement and the proximity of the

instrument to the dentine wall formed a thicker layer

which was more resistant to removal with chelating

agents (Jodaikin & Austin 1981). The amount pro-

duced during motorized preparation, as with Gates-

Glidden or post drills, has been reported as greater in

volume than that produced by hand filing (Czonstkow-

sky et al. 1990). However, McComb & Smith (1975)

observed under SEM that instrumentation with

K-reamers, K-files and Giromatic reciprocating files

created similar surfaces. Additional work has shown

that the smear layer contains organic and inorganic

substances that include fragments of odontoblastic

processes, microorganisms and necrotic materials

(Pashley 1992). The generation of a smear layer is

almost inevitable during root canal instrumentation.

Whilst a noninstrumentation technique has been

described for canal preparation without smear forma-

tion, efforts rather focus on methods for its removal,

such as chemical means and methods such as ultra-

sound and hydrodynamic disinfection for its disruption.

Root canal preparation without the creation of a smear

Figure 1 Scanning electron micrograph of smeared surface of

dentine. The crack shapes are processing artefacts overlying

dentinal tubules.

Figure 2 Scanning electron micrograph of dentine surface

showing smear plugs occluding tubules. The surface has been

treated for 60 s with Tubulicid Blue Label (Dental Therapeu-

tics AB, Nacka, Sweden).

3

Violich & Chandler Smear layer in endodontics

ª 2010 International Endodontic Journal International Endodontic Journal, 43, 2–15, 2010

layer may be possible. A noninstrumental hydrody-

namic technique may have future potential (Lussi et al.

1993), and sonically driven polymer instruments with

tips of variable diameter are reported to disrupt the

smear layer in a technique called hydrodynamic

disinfection (Ruddle 2007).

When viewed under the SEM, the smear layer often

has an amorphous irregular and granular appearance

(Brannstrom et al. 1980, Yamada et al. 1983, Pashley

et al. 1988) (Fig. 3). The appearance is thought to be

formed by translocating and burnishing the superficial

components of the dentine walls during treatment

(Baumgartner & Mader 1987).

The significance of the smear layer

Root canal treatment usually involves the chemome-

chanical removal of bacteria and infected dentine from

within the root canals. The process is often followed by

an intracanal dressing and a root filling. Amongst

important factors affecting the prognosis of root canal

treatment is the seal created by the filling against the

walls of the canal. Considerable effort has been made to

understand the effect of the smear layer on the apical

and coronal seal (Madison & Krell 1984, Goldberg et al.

1985, 1995, Evans & Simon 1986, Kennedy et al.

1986, Cergneux et al. 1987, Saunders & Saunders

1992, 1994, Gencoglu et al. 1993a, Karagoz-Kucukay

& Bayirli 1994, Tidswell et al. 1994, Lloyd et al. 1995,

Behrend et al. 1996, Chailertvanitkul et al. 1996,

Vassiliadis et al. 1996, Taylor et al. 1997, Timpawat

& Sripanaratanakul 1998, Economides et al. 1999,

2004, von Fraunhofer et al. 2000, Froes et al. 2000,

Goya et al. 2000, Timpawat et al. 2001, Clark-Holke

et al. 2003, Cobankara et al. 2004, Park et al. 2004).

Workers have reached different conclusions, with

current knowledge of interactions between the smear

layer and factors such as filling technique and sealer

type being limited. In addition, the methodology of

studies, the type and site of leakage tests, and the

sample size should be taken into account and consid-

eration given to these variables before conclusions are

reached (Shahravan et al. 2007).

Some authors suggest that maintaining the smear

layer may block the dentinal tubules and limit bacterial

or toxin penetration by altering dentinal permeability

(Michelich et al. 1980, Pashley et al. 1981, Safavi et al.

1990). Others believe that the smear layer, being a

loosely adherent structure, should be completely

removed from the surface of the root canal wall

because it can harbour bacteria and provide an avenue

for leakage (Mader et al. 1984, Cameron 1987a,

Meryon & Brook 1990). It may also limit the effective

disinfection of dentinal tubules by preventing sodium

hypochlorite, calcium hydroxide and other intracanal

medicaments from penetrating the dentinal tubules.

Should the smear layer be removed?

The question of keeping or removing the smear layer

remains controversial (Drake et al. 1994, Shahravan

et al. 2007). Some investigations have focussed on its

removal (Garberoglio & Brannstrom 1976, Outhwaite

et al. 1976, Pashley 1985), whilst others have consid-

ered its effects on apical and coronal microleakage

(Madison & Krell 1984, Goldberg et al. 1995, Cha-

ilertvanitkul et al. 1996), bacterial penetration of the

tubules (Pashley 1984, Williams & Goldman 1985,

Meryon & Brook 1990) and the adaptation of root

canal materials (White et al. 1987, Gencoglu et al.

1993a, Gutmann 1993). In support of its removal are:

1. It has an unpredictable thickness and volume,

because a great portion of it consists of water (Cerg-

neux et al. 1987).

2. It contains bacteria, their by-products and necrotic

tissue (McComb & Smith 1975, Goldberg & Abramo-

vich 1977, Wayman et al. 1979, Cunningham &

Martin 1982, Yamada et al. 1983). Bacteria may

survive and multiply (Brannstrom & Nyborg 1973)

and can proliferate into the dentinal tubules (Olgart

et al. 1974, Akpata & Blechman 1982, Williams &

Figure 3 Scanning electron micrograph of dentine surface

with typical amorphous smear layer with granular appear-

ance and moderate debris present (courtesy of Dr Artika

Soma).

4

Smear layer in endodontics Violich & Chandler

International Endodontic Journal, 43, 2–15, 2010 ª 2010 International Endodontic Journal



Goldman 1985, Meryon et al. 1986, Meryon & Brook

1990), which may serve as a reservoir of microbial

irritants (Pashley 1984).

3. It may act as a substrate for bacteria, allowing their

deeper penetration in the dentinal tubules (George et al.

2005).

4. It may limit the optimum penetration of disinfecting

agents (McComb & Smith 1975, Outhwaite et al. 1976,

Goldberg & Abramovich 1977, Wayman et al. 1979,

Yamada et al. 1983). Bacteria may be found deep

within dentinal tubules (Bystrom & Sundqvist 1981,

1983, 1985) and smear layer may block the effects of

disinfectants in them (Goldberg & Abramovich 1977,

Wayman et al. 1979, Yamada et al. 1983, Baumgart-

ner & Mader 1987). Haapasalo & Ørstavik (1987)

found that in the absence of smear layer, liquid

camphorated monochlorophenol disinfected the den-

tinal tubules rapidly and completely, but calcium

hydroxide failed to eliminate Enterococcus faecalis even

after 7 days of incubation. A subsequent study con-

cluded that the smear layer delayed but did not abolish

the action of the disinfectant (Ørstavik & Haapasalo

1990). Brannstrom (1984) had previously stated that

following the removal of the smear layer, bacteria in

the dentinal tubules can easily be destroyed.

5. It can act as a barrier between filling materials and

the canal wall and therefore compromise the formation

of a satisfactory seal (Lester & Boyde 1977, White et al.

1984, Cergneux et al. 1987, Czonstkowsky et al. 1990,

Foster et al. 1993, Yang & Bae 2002). Lester & Boyde

(1977) found that zinc oxide – eugenol based root

canal sealers failed to enter dentinal tubules in the

presence of smear. In two consecutive studies, White

et al. observed that plastic filling materials and sealers

penetrated dentinal tubules after removal of smear

layer (White et al. 1984, 1987). Oksan et al. (1993)

also found that smear prevented the penetration of

sealers into dentinal tubules, whilst no penetration of

sealer was observed in control groups. Penetration in

their smear-free groups ranged from 40 to 60 lm. It

may be concluded that such tubular penetration

increases the interface between the filling and the

dentinal structures, which may improve the ability of a

filling material to prevent leakage (White et al. 1984).

If the aim is maximum penetration into the dentinal

tubules to prevent microleakage, root canal filling

materials should be applied to a surface free of smear

and either a low surface activity or, alternatively, an

adequate surface-active reagent must be added to them

(Aktener et al. 1989). However, there are no reports of

a correlation between microleakage and penetration of

filling materials into dentinal tubules, whilst the basis

of leakage studies remains questionable. Pashley et al.

(1989) observed an extensive network of microchan-

nels around restorations that had been placed in

cavities with smear layer. The thickness of these

channels was 1–10 lm. Smear layer may thus present

a passage for substances to leak around or through its

particles at the interface between the filling material

and the tooth structure. Pashley & Depew (1986)

reported that, when experimenting with class 1 cavi-

ties, microleakage decreased after the removal of smear

layer, but dentinal permeability increased. Saunders &

Saunders (1992) concluded that coronal leakage of

root canal fillings was less in smear-free groups than

those with a smear layer.

6. It is a loosely adherent structure and a potential

avenue for leakage and bacterial contaminant passage

between the root canal filling and the dentinal walls

(Mader et al. 1984, Cameron 1987b, Meryon & Brook

1990). Its removal would facilitate canal filling

(McComb & Smith 1975, Goldman et al. 1981, Cam-

eron 1983).

Conversely, some investigators believe in retaining

the smear layer during canal preparation, because it

can block the dentinal tubules, preventing the ex-

change of bacteria and other irritants by altering

permeability (Michelich et al. 1980, Pashley et al.

1981, Safavi et al. 1990, Drake et al. 1994, Galvan

et al. 1994). The smear layer serves as a barrier to

prevent bacterial migration into the dentinal tubules

(Drake et al. 1994, Galvan et al. 1994, Love et al.

1996, Perez et al. 1996). Pashley (1985) suggested

that if the canals were inadequately disinfected, or if

bacterial contamination occurred after canal prepara-

tion, the presence of a smear layer might stop bacterial

invasion of the dentinal tubules. Bacteria remaining

after canal preparation are sealed into the tubules by

the smear layer and subsequent filling materials. Some

studies provide evidence to support the hypothesis that

the smear layer inhibits bacterial penetration (Pashley

et al. 1981, Safavi et al. 1989). A major limitation is

that the experiments were undertaken with dentine

discs or root cross-sections, models with little relevance

in terms of simulating the clinical conditions of root

canal treatment. Drake et al. (1994) developed a more

clinically relevant model to determine the effect of the

presence or absence of the smear layer on bacterial

colonization of root canals.

Williams & Goldman (1985) reported that the smear

layer was not a complete barrier and could only delay

bacterial penetration. In their experiment, using the

5



motile, swarming bacterium Proteus vulgaris, the smear

layer delayed the passage of the organisms through the

tubules. Madison & Krell (1984) using ethylenedi-

aminetetraacetic acid (EDTA) solution in a dye pene-

tration study found that the smear layer made no

difference to leakage. Goldberg et al. (1995) studied the

sealing ability of Ketac Endo and Tubliseal in an India

ink study with and without smear layer and found no

difference. Chailertvanitkul et al. (1996) found no

difference in leakage with or without smear layer,

however the time period was short. When the smear

layer is not removed, the durability of the apical seal

should be evaluated over a long period. Since the smear

layer is nonhomogenous and may potentially be

dislodged from the underlying tubules (Mader et al.

1984), it may slowly disintegrate, dissolving around a

leaking filling material to leave a void between the

canal wall and sealer. Meryon & Brook (1990) found

the presence of smear layer had no effect on the ability

of three oral bacteria to penetrate dentine discs. All

were able to digest the layer, possibly stimulated by the

nutrient-rich medium below the discs.

The adaptation of root canal materials to canal walls

has been studied. White et al. (1987) found that

pHEMA, silicone and Roth 801 and AH26 sealers

extended into tubules consistently when smear layer

was removed. Gencoglu et al. (1993b) found removing

the smear layer enhanced the adaptation of gutta-

percha in both cold laterally compacted and thermo-

plastic root fillings without sealer. Gutmann (1993)

also showed that after removing the smear layer,

themoplastic gutta-percha adapted with or without

sealer.

A systematic review and meta-analysis by Shahra-

van et al. (2007) set out to determine whether smear

layer removal reduced leakage of root filled teeth ex

vivo. Using 26 eligible papers with 65 comparisons,

54% of the comparisons reported no significant differ-

ence, 41% reported in favour of removing the smear

layer and 5% reported a difference in favour of keeping

it. They concluded that smear layer removal improved

the fluid-tight seal of the root canal system, whereas

other factors such as filling technique or the type of

sealer did not produce significant effects.

Urethane dimethacrylate (UDMA) based root canal

sealers have been introduced. Their aim is to provide a

better bond to allow less microleakage and increase the

fracture resistance of root filled teeth through the

creation of monoblocks, when a core material such as

Resilon replaces gutta-percha. Whilst some studies

indicate that smear layer removal leads to higher

tubule penetration, increased sealer to dentine bond

strength and enhanced fluid-tight seal, a recent report

concluded that smear layer removal did not necessarily

equate to improved resistance to bacterial penetration

along these and older types of sealers (Saleh et al.

2008).

Methods to remove the smear layer

Chemical removal

The quantity of smear layer removed by a material is

related to its pH and the time of exposure (Morgan &

Baumgartner 1997). A number of chemicals have

been investigated as irrigants to remove the smear

layer. According to Kaufman & Greenberg (1986), a

working solution is the one which is used to clean the

canal, and an irrigation solution the one which is

essential to remove the debris and smear layer created

by the instrumentation process. Chlorhexidine, whilst

popular as an irrigant and having a long lasting

antibacterial effect through adherence to dentine, does

not dissolve organic material or remove the smear

layer.

Sodium hypochlorite

The ability of NaOCl to dissolve organic tissues is well-

known (Rubin et al. 1979, Wayman et al. 1979,

Goldman et al. 1982) and increases with rising tem-

perature (Moorer & Wesselink 1982). However, its

capacity to remove smear layer from the instrumented

root canal walls has been found to be lacking. The

conclusion reached by many authors is that the use of

NaOCl during or after instrumentation produces super-

ficially clean canal walls with the smear layer present

(Baker et al. 1975, Goldman et al. 1981, Berg et al.

1986, Baumgartner & Mader 1987).

Chelating agents

Smear layer components include very small particles

with a large surface : mass ratio, which makes them

soluble in acids (Pashley 1992). The most common

chelating solutions are based on EDTA which reacts

with the calcium ions in dentine and forms soluble

calcium chelates (Fig. 4). It has been reported that

EDTA decalcified dentine to a depth of 20–30 lm in

5 min (von der Fehr & Nygaard-Ostby 1963); however,

Fraser (1974) stated that the chelating effect was

almost negligible in the apical third of root canals.

6



Different formulations of EDTA have been used as

root canal irrigants. In a combination, urea peroxide is

added to encourage debris to float out of the root canal

(Stewart et al. 1969). This product (RC-Prep, Premier

Dental Products, Plymouth Meeting, PA, USA) also

includes a wax that left a residue on the root canal

walls despite further instrumentation and irrigation

and which may compromise the ability to obtain a

hermetic seal (Biesterfeld & Taintor 1980). Many

studies have shown that paste-type chelating agents,

whilst having a lubricating effect, do not remove the

smear layer effectively when compared to liquid EDTA.

A recent experiment examining the addition of two

surfactants to liquid EDTA did not result in better smear

layer removal (Lui et al. 2007).

A quaternary ammonium bromide (cetrimide) has

been added to EDTA solutions to reduce surface tension

and increase penetrability of the solution (von der Fehr

& Nygaard-Ostby 1963). McComb & Smith (1975)

reported that when this combination (REDTA) was

used during instrumentation, there was no smear layer

remaining except in the apical part of the canal. After

using REDTA in vivo, it was shown that the root canal

surfaces were uniformly occupied by patent dentinal

tubules with very little superficial debris (McComb et al.

1976). When used during and after instrumentation, it

was possible to still see remnants of odontoblastic

processes within the tubules even though there was no

smear layer present (Goldman et al. 1981). Goldberg &

Abramovich (1977) observed that the circumpulpal

surface had a smooth structure and that the dentinal

tubules had a regular circular appearance with the use

of EDTAC (EDTA and cetavlon). The optimal working

time of EDTAC was suggested to be 15 min in the root

canal and no further chelating action could be expected

after this (Goldberg & Spielberg 1982). This study also

showed that REDTA was the most efficient irrigating

solution for removing smear layer. In a study using a

combination of 0.2% EDTA and a surface-active

antibacterial solution, Brannstrom et al. (1980) ob-

served that this mixture removed most of the smear

layer without opening many dentinal tubules or

removing peritubular dentine. Bis-dequalinium-acetate

(BDA), a dequalinium compound and an oxine deriv-

ative has been shown to remove the smear layer

throughout the canal, even in the apical third (Kauf-

man et al. 1978, Kaufman 1981). BDA is well tolerated

by periodontal tissues and has a low surface tension

allowing good penetration. It is considered less toxic

that NaOCl and can be used as a root canal dressing. A

commercial form of BDA called Solvidont (De Trey,

A.G., Zurich, Switzerland) was available in the 1980s

and its use as an alternative to NaOCl was supported

experimentally (Kaufman 1983a,b, Chandler & Lilley

1987, Lilley et al. 1988, Mohd Sulong 1989). Salvizol

(Ravens Gmbh, Konstanz, Germany) is a commercial

brand of 0.5% BDA and possesses the combined actions

of chelation and organic debridement. Kaufman et al.

(1978) reported that Salvizol had better cleaning

properties than EDTAC. When comparing Salvizol with

5.25% NaOCl, both were found comparable in their

ability to remove organic debris, but only Salvizol

opened dentinal tubules (Kaufman & Greenberg 1986).

Berg et al. (1986) found that Salvizol was less effective

at opening dentinal tubules than REDTA.

Calt & Serper (2000) compared the effects of ethylene

glycol-bis (ß-aminoethyl ether)-N,N,N¢, N¢-tetraacetic

acid (EGTA) with EDTA. The smear layer was com-

pletely removed by EDTA, but it caused erosion of the

peritubular and intertubular dentine, whilst EGTA was

not as effective in the apical third of root canals. EGTA

is reported to bind calcium more specifically (Schmid &

Reilley 1957).

Tetracylines (including tetracycline hydrochloride,

minocycline and doxycycline) are antibiotics effective

against a wide range of microorganisms. Tetracyclines

have unique properties in addition to their antimicro-

bial aspect. They have low pH in concentrated solution,

and because of this can act as a calcium chelator and

cause enamel and root surface demineralization (Bjor-

vatn 1982). The surface demineralization of dentine is

comparable with that of citric acid (Wikesjo et al.

1986). Barkhordar et al. (1997) reported that doxycy-

Figure 4 Scanning electron micrograph of dentine following

60 s exposure to 18% ethylenediaminetetraacetic acid solu-

tion (Ultradent Products Inc., South Jordan, UT, USA).

7



cline hydrochloride (100 mg mL-1) was effective in

removing the smear layer from the surface of instru-

mented canals and root-end cavity preparations. They

speculated that a reservoir of active antibacterial agents

might remain, because doxycycline readily attaches to

dentine and can be subsequently released (Baker et al.

1983, Wikesjo et al. 1986). Haznedaroglu & Ersev

(2001) showed that 1% tetracycline hydrochloride or

50% citric acid can be used to remove the smear layer

from surfaces of root canals. Although they reported no

difference between the two groups, it appeared that the

tetracycline demineralized less peritubular dentine than

the citric acid.

In an effort to produce an irrigant capable of both

removing the smear layer and disinfecting the root

canal system, Torabinejad et al. (2003) developed a

new irrigating solution containing a mixture of a

tetracycline isomer, an acid, and a detergent (MTAD).

Their work concluded MTAD to be an effective solution

for the removal of the smear layer. It does not

significantly change the structure of the dentinal

tubules when the canals are irrigated with sodium

hypochlorite and followed with a final rinse of MTAD.

This irrigant demineralizes dentine faster than 17%

EDTA (De-Deus et al. 2007) and bacterial penetration

into filled canals is similar with both solutions (Ghod-

dusi et al. 2007).

Organic acids

The effectiveness of citric acid as a root canal irrigant

has been demonstrated (Loel 1975, Tidmarsh 1978)

and confirmed to be more effective than NaOCl alone in

removing the smear layer (Baumgartner et al. 1984).

Citric acid removed smear layer better than polyacrylic

acid, lactic acid and phosphoric acid but not EDTA

(Meryon et al. 1987). Wayman et al. (1979) showed

that canal walls treated with 10%, 25% and 50% citric

acid solution were generally free of the smeared

appearance, but they had the best results in removing

smear layer with sequential use of 10% citric acid

solution and 2.5% NaOCl solution, then again followed

by a 10% solution of citric acid. However, Yamada

et al. (1983) observed that the 25% citric acid–NaOCl

group was not as effective as a 17% EDTA–NaOCl

combination. To its detriment, citric acid left precipi-

tated crystals in the root canal which might be

disadvantageous to the root canal filling. With 50%

lactic acid, the canal walls were generally clean, but

with openings of dentinal tubules that did not appear to

be completely patent (Wayman et al. 1979). Bitter

(1989) introduced 25% tannic acid solution as a root

canal irrigant and cleanser. Canal walls irrigated with

this solution appeared significantly cleaner and

smoother than walls treated with a combination of

hydrogen peroxide and NaOCl, and the smear layer was

removed. Sabbak & Hassanin (1998) refuted these

findings and explained that tannic acid increased the

cross-linking of exposed collagen with the smear layer

and within the matrix of the underlying dentine,

therefore increasing organic cohesion to the tubules.

McComb & Smith (1975) compared the efficacy of

20% polyacrylic acid with REDTA and found that it

was no better than REDTA in removing or preventing

the build up of smear layer, thought to be as a result of

its higher viscosity. McComb et al. (1976) also used 5%

and 10% polyacrylic acid as an irrigant and observed

that it could remove smear layer in accessible regions.

Polyacrylic acid (Durelon liquid and Fuji II liquid) at

40% has been reported to be very effective, and because

of its potency users should not exceed a 30 s applica-

tion (Berry et al. 1987).

Sodium hypochlorite and EDTA

When irrigating a root canal the purpose is twofold: to

remove the organic component, the debris originating

from pulp tissue and microorganisms, and the mostly

inorganic component, the smear layer. As there is no

single solution which has the ability to dissolve organic

tissues and to demineralize the smear layer, the

sequential use of organic and inorganic solvents has

been recommended (Koskinen et al. 1980, Yamada

et al. 1983, Baumgartner et al. 1984). Numerous

authors have agreed that the removal of smear layer

as well as soft tissue and debris can be achieved by the

alternate use of EDTA and NaOCl (Yamada et al. 1983,

White et al. 1984, Baumgartner & Mader 1987, Cengiz

et al. 1990). Goldman et al. (1982) examined the effect

of various combinations of EDTA and NaOCl, and the

most effective final rinse was 10 mL of 17% EDTA

followed by 10 mL of 5.25% NaOCl, a finding con-

firmed by Yamada et al. (1983). Used in combination

with EDTA, NaOCl is inactivated with the EDTA

remaining functional for several minutes.

Ultrasonic smear removal

Following the introduction of dental ultrasonic devices

in the 1950s, ultrasound was investigated in end-

odontics (Martin et al. 1980, Cunningham & Martin

1982, Cunningham et al. 1982). A continuous flow of

8



NaOCl activated by an ultrasonic delivery system was

used for the preparation and irrigation of canals.

Smear-free canal surfaces were observed using this

method (Cameron 1983, 1987a,b, Griffiths & Stock

1986, Alacam 1987). Whilst concentrations of 2–4%

sodium hypochlorite in combination with ultrasonic

energy were able to remove smear layer, lower

concentrations of the solutions were unsatisfactory

(Cameron 1988). However, Ahmad et al. (1987a)

claimed that their technique of modified ultrasonic

instrumentation using 1% NaOCl removed the debris

and smear layer more effectively than the technique

recommended by Martin & Cunningham (1983). The

apical region of the canals showed less debris and

smear layer than the coronal aspects, depending on

acoustic streaming, which was more intense in

magnitude and velocity at the apical regions of the

file. Cameron (1983) also compared the effect of

different ultrasonic irrigation periods on removing

smear layer and found that a 3- and 5-min irrigation

produced smear-free canal walls, whilst an 1-min

irrigation was ineffective. In contrast to these results,

other investigators found ultrasonic preparation un-

able to remove smear layer (Cymerman et al. 1983,

Baker et al. 1988, Goldberg et al. 1988).

Researchers who found the cleaning effects of

ultrasonics beneficial used the technique only for the

final irrigation of root canal after completion of hand

instrumentation (Ahmad et al. 1987a, Alacam 1987,

Cameron 1988). This is given the term passive ultra-

sonic irrigation and has been the subject of a recent

review (van der Sluis et al. 2007). Ahmad et al.

(1987a,b) claimed that direct physical contact of the

file with the canal walls throughout instrumentation

reduced acoustic streaming. Acoustic streaming is

maximized when the tips of the smaller instruments

vibrate freely in a solution. Lumley et al. (1992)

recommended that only size 15 files be used to

maximize microstreaming for the removal of debris.

Prati et al. (1994) also achieved smear layer removal

with ultrasonics. Walker & del Rio (1989, 1991)

showed no significant difference between tap water and

sodium hypochlorite when used with ultrasonics, but

they reported that neither solution was effective at any

level in the canal to remove the smear layer ultrason-

ically. Baumgartner & Cuenin (1992) also observed

that ultrasonically energized NaOCl, even at full

strength, did not remove the smear layer from root

canal walls. Guerisoli et al. (2002) evaluated the use of

ultrasonics to remove the smear layer and found it

necessary to use 15% EDTAC with either distilled water

or 1% sodium hypochlorite to achieve the desired

result.

Laser removal

Lasers can be used to vaporize tissues in the main

canal, remove the smear layer and eliminate residual

tissue in the apical portion of root canals (Takeda et al.

1998a,b, 1999). The effectiveness of lasers depends on

many factors, including the power level, the duration of

exposure, the absorption of light in the tissues, the

geometry of the root canal and the tip-to-target

distance (Dederich et al. 1984, Onal et al. 1993, Tewfik

et al. 1993, Moshonov et al. 1995).

Dederich et al. (1984) and Tewfik et al. (1993) used

variants of the neodymium–yttrium-aluminium-gar-

net (Ne:YAG) laser and reported a range of findings

from no change or disruption of the smear layer to

actual melting and recrystallization of the dentine.

This pattern of dentine disruption was observed in

other studies with various lasers, including the carbon

dioxide laser (Onal et al. 1993), the argon fluoride

excimer laser (Stabholz et al. 1993), and the argon

laser (Moshonov et al. 1995, Harashima et al. 1998).

Takeda et al. (1998a,b, 1999) using the erbium-

yttrium-aluminium-garnet (Er:YAG) laser, demon-

strated optimal removal of the smear layer without

melting, charring or recrystallization associated with

other laser types. Kimura et al. (2002) also demon-

strated the removal of the smear layer with an Er:YAG

laser. Although they showed removal of the smear

layer, photomicrographs showed destruction of peri-

tubular dentine. The main difficulty with laser

removal of the smear layer is access to the small

canal spaces with the relatively large probes that are

available.

Conclusion

Contemporary methods of root canal instrumentation

produce a layer of organic and inorganic material called

the smear layer that may also contain bacteria and their

by-products. This layer covers the instrumented walls

and may prevent the penetration of intracanal medica-

ments into the dentinal tubules and interfere with the

close adaptation of root filling materials to canal walls.

The data presented indicate removal of the smear layer

for more thorough disinfection of the root canal system

and better adaptation of materials to the canal walls.

There are, however, no clinical trials to demonstrate

this. Current methods of smear layer removal include

9



chemical, ultrasonic and laser techniques – none of

which are totally effective throughout the length of all

canals or are used universally. However, if the smear

layer is to be removed the method of choice seems to be

the alternate use of EDTA and sodium hypochlorite

solutions. Whilst much is known about individual

irrigants, their use in combination and their interac-

tions (and in some cases precipitates) is less well

understood. Conflicting reports exist regarding the

removal of the smear layer before filling root canals.

As several new sealer and core materials have recently

been introduced, further investigations are required to

determine the role of the smear layer in the outcome of

treatment.

References

Ahmad M, Pitt Ford TR, Crum LA (1987a) Ultrasonic

debridement of root canals: acoustic streaming and its

possible role. Journal of Endodontics 13, 490–9.

Ahmad M, Pitt Ford TR, Crum LA (1987b) Ultrasonic

debridement of root canals: an insight into the mechanisms

involved. Journal of Endodontics 13, 93–101.

Akpata ES, Blechman H (1982) Bacterial invasion of pulpal

dentin wall in vitro. Journal of Dental Research 61, 435–8.

Aktener BO, Cengiz T, Piskin B (1989) The penetration of

smear material into dentinal tubules during instrumenta-

tion with surface-active reagents: a scanning electron

microscopic study. Journal of Endodontics 15, 588–90.

Alacam T (1987) Scanning electron microscope study com-

paring the efficacy of endodontic irrigating systems. Inter-

national Endodontic Journal 20, 287–94.

Baker NA, Eleazer PD, Averbach RE, Seltzer S (1975)

Scanning electron microscopic study of the efficacy of

various irrigating solutions. Journal of Endodontics 1, 127–

35.

Baker PJ, Evans RT, Coburn RA, Genco RJ (1983) Tetracycline

and its derivatives strongly bind to and are released from the

tooth surface in active form. Journal of Periodontology 54,

580–5.

Baker MC, Ashrafi SH, Van Cura JE, Remeikis NA (1988)

Ultrasonic compared with hand instrumentation: a scan-

ning electron microscope study. Journal of Endodontics 14,

435–40.

Barkhordar RA, Watanabe LG, Marshall GW, Hussain MZ

(1997) Removal of intracanal smear by doxycycline in vitro.

Oral Surgery, Oral Medicine, Oral Pathology, Oral Radiology

and Endodontics 84, 420–3.

Barnes IE (1974) The production of inlay cavity bevels. British

Dental Journal 137, 379–90.

Baumgartner JC, Cuenin PR (1992) Efficacy of several

concentrations of sodium hypochlorite for root canal

irrigation. Journal of Endodontics 18, 605–12.

Baumgartner JC, Mader CL (1987) A scanning electron

microscopic evaluation of four root canal irrigation regi-

mens. Journal of Endodontics 13, 147–57.

Baumgartner JC, Brown CM, Mader CL, Peters DD, Shulman

JD (1984) A scanning electron microscopic evaluation of

root canal debridement using saline, sodium hypochlorite,

and citric acid. Journal of Endodontics 10, 525–31.

Behrend GD, Cutler CW, Gutmann JL (1996) An in-vitro

study of smear layer removal and microbial leakage along

root-canal fillings. International Endodontic Journal 29, 99–

107.

Berg MS, Jacobsen EL, BeGole EA, Remeikis NA (1986) A

comparison of five irrigating solutions: a scanning electron

microscopic study. Journal of Endodontics 12, 192–7.

Berry EA III, von der Lehr WN, Herrin HK (1987) Dentin

surface treatments for the removal of the smear layer: an

SEM study. Journal of the American Dental Association 115,

65–7.

Biesterfeld RC, Taintor JF (1980) A comparison of periapical

seals of root canals with RC-Prep or Salvizol. Oral Surgery,

Oral Medicine and Oral Pathology 49, 532–7.

Bitter NC (1989) A 25% tannic acid solution as a root canal

irrigant cleanser: a scanning electron microscope study.

Oral Surgery, Oral Medicine and Oral Pathology 67, 333–7.

Bjorvatn K (1982) Antibiotic compounds and enamel demin-

eralization. An in vitro study. Acta Odontologica Scandinavica

40, 341–52.

Brannstrom M (1984) Communication between the oral

cavity and the dental pulp associated with restorative

treatment. Operative Dentistry 9, 57–68.

Brannstrom M, Johnson G (1974) Effects of various condi-

tioners and cleaning agents on prepared dentin surfaces: a

scanning electron microscopic investigation. Journal of

Prosthetic Dentistry 31, 422–30.

Brannstrom M, Nyborg H (1973) Cavity treatment with a

microbicidal fluoride solution: growth of bacteria and effect

on the pulp. Journal of Prosthetic Dentistry 30, 303–10.

Brannstrom M, Nordenvall KJ, Glantz P-O (1980) The effect of

EDTA-containing surface-active solutions on the morphol-

ogy of prepared dentin: an in vivo study. Journal of Dental

Research 59, 1127–31.

Bystrom A, Sundqvist G (1981) Bacteriologic evaluation of the

efficacy of mechanical root canal instrumentation in end-

odontic therapy. Scandinavian Journal of Dental Research 89,

321–8.

Bystrom A, Sundqvist G (1983) Bacteriologic evaluation of the

effect of 0.5 percent sodium hypochlorite in endodontic

therapy. Oral Surgery, Oral Medicine and Oral Pathology 55,

307–12.

Bystrom A, Sundqvist G (1985) The antibacterial action of

sodium hypochlorite and EDTA in 60 cases of endodontic

therapy. International Endodontic Journal 18, 35–40.

Calt S, Serper A (2000) Smear layer removal by EGTA. Journal

of Endodontics 26, 459–61.

10



Cameron JA (1983) The use of ultrasonics in the removal of

the smear layer: a scanning electron microscope study.

Journal of Endodontics 9, 289–92.

Cameron JA (1987a) The synergistic relationship between

ultrasound and sodium hypochlorite: a scanning electron

microscope evaluation. Journal of Endodontics 13, 541–5.

Cameron JA (1987b) The use of 4 per cent sodium hypochlo-

rite, with or without ultrasound, in cleansing of uninstru-

mented immature root canals; SEM study. Australian Dental

Journal 32, 204–13.

Cameron JA (1988) The use of ultrasound for the removal of

the smear layer. The effect of sodium hypochlorite concen-

tration; SEM study. Australian Dental Journal 33, 193–200.

Cengiz T, Aktener BO, Piskin B (1990) Effect of dentinal tubule

orientation on the removal of smear layer by root canal

irrigants. A scanning electron microscopic study. Interna-

tional Endodontic Journal 23, 163–71.

Cergneux M, Ciucchi B, Dietschi JM, Holz J (1987) The

influence of the smear layer on the sealing ability of canal

obturation. International Endodontic Journal 20, 228–32.

Chailertvanitkul P, Saunders WP, MacKenzie D (1996) The

effect of smear layer on microbial coronal leakage of gutta-

percha root fillings. International Endodontic Journal 29,

242–8.

Chandler NP, Lilley JD (1987) Clinical trial of a bis-dequalin-

ium-acetate solution as an endodontic irrigant. Journal of

Dental Research 66, 842.

Clark-Holke D, Drake D, Walton R, Rivera E, Guthmiller JM

(2003) Bacterial penetration through canals of endodonti-

cally treated teeth in the presence or absence of the smear

layer. Journal of Dentistry 31, 275–81.

Cobankara FK, Adanir N, Belli S (2004) Evaluation of the

influence of smear layer on the apical and coronal sealing

ability of two sealers. Journal of Endodontics 30, 406–9.

Cunningham WT, Martin H (1982) A scanning electron

microscope evaluation of root canal debridement with the

endosonic ultrasonic synergistic system. Oral Surgery, Oral

Medicine and Oral Pathology 53, 527–31.

Cunningham WT, Martin H, Forrest WR (1982) Evaluation of

root canal debridement by the endosonic ultrasonic syner-

gistic system. Oral Surgery, Oral Medicine and Oral Pathology

53, 401–4.

Cymerman JJ, Jerome LA, Moodnik RM (1983) A scanning

electron microscope study comparing the efficacy of hand

instrumentation with ultrasonic instrumentation of the root

canal. Journal of Endodontics 9, 327–31.

Czonstkowsky M, Wilson EG, Holstein FA (1990) The smear

layer in endodontics. Dental Clinics of North America 34, 13–

25.

Dederich DN, Zakariasen KL, Tulip J (1984) Scanning electron

microscopic analysis of canal wall dentin following neo-

dymium-yttrium-aluminum-garnet laser irradiation. Journal


De-Deus G, Reis C, Fidel S, Fidel R, Paciornik S (2007) Dentin

demineralization when subjected to BioPure MTAD: a

longitudinal and quantitative assessment. Journal of End-

odontics 33, 1364–1368.

Drake DR, Wiemann AH, Rivera EM, Walton RE (1994)

Bacterial retention in canal walls in vitro: effect of smear

layer. Journal of Endodontics 20, 78–82.

Economides N, Liolios E, Kolokuris I, Beltes P (1999) Long-

term evaluation of the influence of smear layer removal on

the sealing ability of different sealers. Journal of Endodontics

25, 123–5.

Economides N, Kokorikos I, Kolokouris I, Panagiotis B, Gogos

C (2004) Comparative study of apical sealing ability of a

new resin-based root canal sealer. Journal of Endodontics 30,

403–5.

Eick JD, Wilko RA, Anderson CH, Sorensen SE (1970)

Scanning electron microscopy of cut tooth surfaces and

identification of debris by use of the electron microprobe.

Journal of Dental Research 49(Suppl), 1359–68.

Evans JT, Simon JHS (1986) Evaluation of the apical seal

produced by injected thermoplasticized gutta-percha in the

absence of smear layer and root canal sealer. Journal of

Endodontics 12, 100–7.

von der Fehr FR, Nygaard-Ostby B (1963) Effect of EDTAC and

sulfuric acid on root canal dentine. Oral Surgery, Oral


Foster KH, Kulild JC, Weller RN (1993) Effect of smear

layer removal on the diffusion of calcium hydroxide

through radicular dentin. Journal of Endodontics 19, 136–

40.

Fraser JG (1974) Chelating agents: their softening effect on

root canal dentin. Oral Surgery, Oral Medicine and Oral

Pathology 37, 803–11.

von Fraunhofer JA, Fagundes DK, McDonald NJ, Dumsha TC

(2000) The effect of root canal preparation on microleakage

within endodontically treated teeth: an in vitro study.

International Endodontic Journal 33, 355–60.

Froes JA, Horta HGP, da Silveira AB (2000) Smear layer

influence on the apical seal of four different obturation

techniques. Journal of Endodontics 26, 351–4.

Galvan DA, Ciarlone AE, Pashley DH, Kulild JC, Primack PD,

Simpson MD (1994) Effect of smear layer removal on the

diffusion permeability of human roots. Journal of Endodontics

20, 83–6.

Garberoglio R, Brannstrom M (1976) Scanning electron

microscopic investigation of human dentinal tubules.

Archives of Oral Biology 21, 355–62.

Gencoglu N, Samani S, Gunday M (1993a) Dentinal wall

adaptation of thermoplasticized gutta-percha in the absence

or presence of smear layer: a scanning electron microscopic

study. Journal of Endodontics 19, 558–62.

Gencoglu N, Samani S, Gunday M (1993b) Evaluation of

sealing properties of Thermafil and Ultrafil techniques in the

absence or presence of smear layer. Journal of Endodontics

19, 599–603.

George S, Kishen A, Song KP (2005) The role of environmen-

tal changes on monospecies biofilm formation on root canal

11



wall by Enterococcus faecalis. Journal of Endodontics 31, 867–

72.

Ghoddusi J, Rohani A, Rashed T, Ghaziani P, Akbari M (2007)

An evaluation of microbial leakage after using MTAD as a

final irrigation. Journal of Endodontics 33, 173–176.

Gilboe DB, Svare CW, Thayer KE, Drennon DG (1980)

Dentinal smearing: an investigation of the phenomenon.

Journal of Prosthetic Dentistry 44, 310–6.

Goldberg F, Abramovich A (1977) Analysis of the effect of

EDTAC on the dentinal walls of the root canal. Journal of


Goldberg F, Spielberg C (1982) The effect of EDTAC and the

variation of its working time analyzed with scanning

electron microscopy. Oral Surgery, Oral Medicine and Oral

Pathology 53, 74–7.

Goldberg F, Bernat MI, Spielberg C, Massone EJ, Piovano SA

(1985) Analysis of the effect of ethylenediaminetetraacetic

acid on the apical seal of root canal fillings. Journal of


Goldberg F, Soares I, Massone EJ, Soares IM (1988) Compar-

ative debridement study between hand and sonic instru-

mentation of the root canal. Endodontics and Dental

Traumatology 4, 229–34.

Goldberg F, Artaza LP, De Silvio A (1995) Apical sealing

ability of a new glass ionomer root canal sealer. Journal of


Goldman LB, Goldman M, Kronman JH, Lin PS (1981) The

efficacy of several irrigating solutions for endodontics: a

scanning electron microscopic study. Oral Surgery, Oral


Goldman M, Goldman LB, Cavaleri R, Bogis J, Lin PS (1982)

The efficacy of several endodontic irrigating solutions: a

scanning electron microscopic study: Part 2. Journal of


Goya C, Yamazaki R, Tomita Y, Kimura Y, Matsumoto K

(2000) Effects of pulsed Nd:YAG laser irradiation on smear

layer at the apical stop and apical leakage after obturation.


Griffiths BM, Stock CJR (1986) The efficiency of irrigants in

removing root canal debris when used with ultrasonic

preparation technique. International Endodontic Journal 19,

277–84.

Guerisoli DMZ, Marchesan MA, Walmsley AD, Lumley PJ,

Pecora JD (2002) Evaluation of smear layer removal by

EDTAC and sodium hypochlorite with ultrasonic agitation.


Gutmann JL (1993) Adaptation of injected thermoplasticized

gutta-percha in the absence of the dentinal smear layer.


Haapasalo M, Ørstavik D (1987) In vitro infection and

disinfection of dentinal tubules. Journal of Dental Research

66, 1375–9.

Harashima T, Takeda FH, Zhang C, Kimura Y, Matsumoto K

(1998) Effect of argon laser irradiation on instrumented root

canal walls. Endodontics and Dental Traumatology 14, 26–30.

Haznedaroglu F, Ersev H (2001) Tetracycline HCl solution

as a root canal irrigant. Journal of Endodontics 27, 738–

40.

Jodaikin A, Austin JC (1981) Smear layer removal with

chelating agents after cavity preparation. Journal of Pros-

thetic Dentistry 46, 171–4.

Karagoz-Kucukay I, Bayirli G (1994) An apical leakage study

in the presence and absence of the smear layer. International

Endodontic Journal 27, 87–93.

Kaufman AY (1981) The use of dequalinium acetate as a

disinfectant and chemotherapeutic agent in endodontics.

Oral Surgery, Oral Medicine and Oral Pathology 51, 434–

41.

Kaufman AY (1983a) Solvidont – a new chemotherapeutic

and bacteriocidal agent for endodontic use (I). Quintessence

International 14, 71–9.

Kaufman AY (1983b) Solvidont – a new chemotherapeutic

and bacteriocidal agent for endodontic use (II). Quintessence

International 14, 235–44.

Kaufman AY, Greenberg I (1986) Comparative study of the

configuration and the cleanliness level of root canals

prepared with the aid of sodium hypochlorite and bis-

dequalinium-acetate solutions. Oral Surgery, Oral Medicine

and Oral Pathology 62, 191–7.

Kaufman AY, Binderman I, Tal M, Gedalia I, Peretz G (1978)

New chemotherapeutic agent for root canal treatment. A

preliminary electron microscopic study on an in vivo and in

vitro endodontically treated tooth. Oral Surgery, Oral Med-

icine and Oral Pathology 46, 283–95.

Kennedy WA, Walker WA III, Gough RW (1986) Smear layer

removal effects on apical leakage. Journal of Endodontics 12,

21–7.

Kimura Y, Yonaga K, Yokoyama K, Kinoshita J, Ogata Y,

Matsumoto K (2002) Root surface temperature increase

during Er:YAG laser irradiation of root canals. Journal of


Koskinen KP, Meurman JH, Stenvall H (1980) Appearance of

chemically treated root canal walls in the scanning electron

microscope. Scandinavian Journal of Dental Research 88, 397–

405.

Lester KS, Boyde A (1977) Scanning electron microscopy of

instrumented, irrigated and filled root canals. British Dental

Journal 143, 359–67.

Lilley JD, Russell C, Chandler NP (1988) Comparison of bis-

dequalinium-acetate and sodium hypochlorite solutions as

endodontic irrigants. Journal of Dental Research 67, 300.

Lloyd A, Thompson J, Gutmann JL, Dummer PMH (1995)

Sealability of the Trifecta technique in the presence or

absence of a smear layer. International Endodontic Journal 28,

35–40.

Loel DA (1975) Use of acid cleanser in endodontic therapy.

Journal of the American Dental Association 90, 148–51.

Love RM, Chandler NP, Jenkinson HF (1996) Penetration of

smeared or nonsmeared dentine by Streptococcus gordonii.


12



Lui J-N, Kuah H-G, Chen N-N (2007) Effects of EDTA with and

without surfactants or ultrasonics on removal of smear

layer. Journal of Endodontics 33, 472–5.

Lumley PJ, Walmsley AD, Walton RE, Rippin JW (1992) Effect

of precurving endosonic files on the amount of debris and

smear layer remaining in curved root canals. Journal of


Lussi A, Nussbacher U, Grosrey J (1993) A novel noninstru-

mented technique for cleansing the root canal system.


Mader CL, Baumgartner JC, Peters DD (1984) Scanning

electron microscopic investigation of the smeared layer on

root canal walls. Journal of Endodontics 10, 477–83.

Madison S, Krell KV (1984) Comparison of ethylenediamine

tetraacetic acid and sodium hypochlorite on the apical seal

of endodontically treated teeth. Journal of Endodontics 10,

499–503.

Martin H, Cunningham MJ (1983) Endosonic endodontics, the

ultrasonic synergistic system. In: Gerstein H, ed. Techniques

in Clinical Endodontics. Philadelphia, PA, USA: WB Saunders,

316–22.

Martin H, Cunningham WT, Norris JP, Cotton WR (1980)

Ultrasonic versus hand filing of dentin: a quantitative

study. Oral Surgery, Oral Medicine and Oral Pathology 49,

79–81.

McComb D, Smith DC (1975) A preliminary scanning electron

microscopic study of root canals after endodontic proce-

dures. Journal of Endodontics 1, 238–42.

McComb D, Smith DC, Beagrie GS (1976) The results of in vivo

endodontic chemomechanical instrumentation-a scanning

electron microscopic study. Journal of the British Endodontic

Society 9, 11–8.

Meryon SD, Brook AM (1990) Penetration of dentine by three

oral bacteria in vitro and their associated cytotoxicity.


Meryon SD, Jakeman KJ, Browne RM (1986) Penetration

in vitro of human and ferret dentine by three bacterial

species in relation to their potential role in pulpal inflam-

mation. International Endodontic Journal 19, 213–20.

Meryon SD, Tobias RS, Jakeman KJ (1987) Smear removal

agents: a quantitative study in vivo and in vitro. Journal of

Prosthetic Dentistry 57, 174–9.

Michelich VJ, Schuster GS, Pashley DH (1980) Bacterial

penetration of human dentin in vitro. Journal of Dental

Research 59, 1398–403.

Mohd Sulong MZA (1989) The incidence of postoperative pain

after canal preparation of open teeth using two irrigation

regimes. International Endodontic Journal 22, 248–51.

Moodnik RM, Dorn SO, Feldman MJ, Levey M, Borden BG

(1976) Efficacy of biomechanical instrumentation: a scan-

ning electron microscopic study. Journal of Endodontics 2,

261–6.

Moorer WR, Wesselink PR (1982) Factors promoting the

tissue dissolving capability of sodium hypochlorite. Interna-


Morgan LA, Baumgartner JC (1997) Demineralization of

resected root-ends with methylene blue dye. Oral Surgery,

Oral Medicine, Oral Pathology, Oral Radiology and Endodontics

84, 74–8.

Moshonov J, Sion A, Kasirer J, Rotstein I, Stabholz A (1995)

Efficacy of argon laser irradiation in removing intracanal

debris. Oral Surgery, Oral Medicine, Oral Pathology, Oral

Radiology and Endodontics 79, 221–5.

Oksan T, Aktener BO, Sen BH, Tezel H (1993) The penetration

of root canal sealers into dentinal tubules. A scanning

electron microscopic study. International Endodontic Journal

26, 301–5.

Olgart L, Brannstrom M, Johnson G (1974) Invasion of

bacteria into dentinal tubules. Experiments in vivo and in

vitro. Acta Odontologica Scandinavica 32, 61–70.

Onal B, Ertl T, Siebert G, Muller G (1993) Preliminary report

on the application of pulsed CO2 laser radiation on root

canals with AgCl fibers: a scanning and transmission

electron microscopic study. Journal of Endodontics 19,

272–6.

Ørstavik D, Haapasalo M (1990) Disinfection by endodontic

irrigants and dressings of experimentally infected dentinal

tubules. Endodontics and Dental Traumatology 6, 142–9.

Outhwaite WC, Livingston MJ, Pashley DH (1976) Effects of

changes in surface area, thickness, temperature and post-

extraction time on human dentine permeability. Archives of

Oral Biology 21, 599–603.

Park DS, Torabinejad M, Shabahang S (2004) The effect of

MTAD on the coronal leakage of obturated root canals.


Pashley DH (1984) Smear layer: physiological considerations.

Operative Dentistry Supplement 3, 13–29.

Pashley DH (1985) Dentin-predentin complex and its perme-

ability: physiologic overview. Journal of Dental Research 64

Spec Iss, 613–20.

Pashley DH (1992) Smear layer: overview of structure and

function. Proceedings of the Finnish Dental Society 88(Suppl

1), 215–24.

Pashley DH, Depew DD (1986) Effects of the smear layer,

Copalite, and oxalate on microleakage. Operative Dentistry

11, 95–102.

Pashley DH, Michelich V, Kehl T (1981) Dentin permeability:

effects of smear layer removal. Journal of Prosthetic Dentistry

46, 531–7.

Pashley DH, Tao L, Boyd L, King GE, Horner JA (1988)

Scanning electron microscopy of the substructure of smear

layers in human dentine. Archives of Oral Biology 33, 265–

70.

Pashley DH, Depew DD, Galloway SE (1989) Microleakage

channels: scanning electron microscopic observation. Oper-

ative Dentistry 14, 68–72.

Perez F, Calas P, Rochd T (1996) Effect of dentin treatment on

in vitro root tubule bacterial invasion. Oral Surgery, Oral

Medicine, Oral Pathology, Oral Radiology and Endodontics 82,

446–51.

13



Prati C, Selighini M, Ferrieri P, Mongiorgi R (1994) Scanning

electron microscopic evaluation of different endodontic

procedures on dentin morphology of human teeth. Journal


Rubin LM, Skobe Z, Krakow AA, Gron P (1979) The effect of

instrumentation and flushing of freshly extracted teeth in

endodontic therapy: a scanning electron microscope study.


Ruddle CJ (2007) Hydrodynamic disinfection: tsunami end-

odontics. Dentistry Today 26(5), 114–7.

Sabbak SA, Hassanin MB (1998) A scanning electron micro-

scopic study of tooth surface changes induced by tannic

acid. Journal of Prosthetic Dentistry 79, 169–74.

Safavi KE, Spangberg LSW, Costa NS Jr, Sapounas G (1989)

An in vitro method for longitudinal evaluation of toxicity of

endodontic sealers. Journal of Endodontics 15, 484–6.

Safavi KE, Spangberg LSW, Langeland K (1990) Root canal

dentinal tubule disinfection. Journal of Endodontics 16, 207–

10.

Saleh IM, Ruyter IE, Haapasolo M, Ørstavik D (2008) Bacterial

penetration along different root canal filling materials in the

presence or absence of smear layer. International Endodontic

Journal 41, 32–40.

Saunders WP, Saunders EM (1992) The effect of smear layer

upon the coronal leakage of gutta-percha root fillings and a

glass ionomer sealer. International Endodontic Journal 25,

245–9.

Saunders WP, Saunders EM (1994) Influence of smear layer

on the coronal leakage of Thermafil and laterally condensed

gutta-percha root fillings with a glass ionomer sealer.


Schmid R, Reilley C (1957) New complexion for titration of

calcium in the presence of magnesium. Annals of Chemistry

29, 264–8.

Shahravan A, Haghdoost AA, Adl A, Rahimi H, Shadifar F

(2007) Effect of smear layer on sealing ability of canal

obturation: a systematic review and meta-analysis. Journal


van der Sluis LWM, Versluis M, Wu MK, Wesselink PR (2007)

Passive ultrasonic irrigation of the root canal: a review of

the literature. International Endodontic Journal 40, 415–26.

Stabholz A, Neev J, Liaw LH, Stabholz A, Khayat A,

Torabinejad M (1993) Effect of ArF-193 nm excimer laser

on human dentinal tubules. A scanning electron micro-

scopic study. Oral Surgery, Oral Medicine and Oral Pathology

75, 90–4.

Stewart GG, Kapsimalas P, Rappaport H (1969) EDTA and

urea peroxide for root canal preparation. Journal of the

American Dental Association 78, 335–8.

Takeda FH, Harashima T, Kimura Y, Matsumoto K (1998a)

Comparative study about the removal of smear layer by

three types of laser devices. Journal of Clinical and Laser

Medical Surgery 16, 117–22.

Takeda FH, Harashima T, Kimura Y, Matsumoto K (1998b)

Efficacy of Er:YAG laser irradiation in removing debris and

smear layer on root canal walls. Journal of Endodontics 24,

548–51.

Takeda FH, Harashima T, Kimura Y, Matsumoto K (1999) A

comparative study of the removal of smear layer by three

endodontic irrigants and two types of laser. International


Taylor JK, Jeansonne BG, Lemon RR (1997) Coronal leakage:

effects of smear layer, obturation technique, and sealer.


Tewfik HM, Pashley DH, Horner JA, Sharawy MM (1993)

Structural and functional changes in root dentin following

exposure to KTP/532 laser. Journal of Endodontics 19, 492–

7.

Tidmarsh BG (1978) Acid-cleansed and resin-sealed root

canals. Journal of Endodontics 4, 117–21.

Tidswell HE, Saunders EM, Saunders WP (1994) Assessment

of coronal leakage in teeth root filled with gutta-percha and

a glass of ionomer root canal sealer. International Endodontic

Journal 27, 208–12.

Timpawat S, Sripanaratanakul S (1998) Apical sealing ability

of glass ionomer sealer with and without smear layer.


Timpawat S, Vongsavan N, Messer HH (2001) Effect of

removal of the smear layer on apical microleakage. Journal


Torabinejad M, Khademi AA, Babagoli J et al. (2003) A new

solution for the removal of the smear layer. Journal of


Vassiliadis L, Liolios E, Kouvas V, Economides N (1996) Effect

of smear layer on coronal microleakage. Oral Surgery, Oral

Medicine, Oral Pathology, Oral Radiology and Endodontics 82,

315–20.

Walker TL, del Rio CE (1989) Histological evaluation of

ultrasonic and sonic instrumentation of curved root canals.


Walker TL, del Rio CE (1991) Histological evaluation of

ultrasonic debridement comparing sodium hypochlorite and

water. Journal of Endodontics 17, 66–71.

Wayman BE, Kopp WM, Pinero GJ, Lazzari EP (1979) Citric

and lactic acids as root canal irrigants in vitro. Journal of


White RR, Goldman M, Lin PS (1984) The influence of the

smeared layer upon dentinal tubule penetration by plastic

filling materials. Journal of Endodontics 10, 558–62.

White RR, Goldman M, Lin PS (1987) The influence of the

smeared layer upon dentinal tubule penetration by end-

odontic filling materials. Part II. Journal of Endodontics 13,

369–74.

Wikesjo UM, Baker PJ, Christersson LA et al. (1986) A

biochemical approach to periodontal regeneration: tetracy-

cline treatment conditions dentin surfaces. Journal of Peri-

odontal Research 21, 322–9.

Williams S, Goldman M (1985) Penetrability of the smeared

layer by a strain of Proteus vulgaris. Journal of Endodontics

11, 385–8.



Yamada RS, Armas A, Goldman M, Lin PS (1983) A scanning

electron microscopic comparison of a high volume final

flush with several irrigating solutions: Part 3. Journal of


Yang SE, Bae KS (2002) Scanning electron microscopy

study of the adhesion of Prevotella nigrescens to the

dentin of prepared root canals. Journal of Endodontics 28,

433–7.


ª 2010 International Endodontic Journal International Endodontic Journal, 43, 2–15, 2010 15

Comparison of working length determination withradiographs and two electronic apex locators

J. P. Vieyra1, J. Acosta2 & J. M. Mondaca2

1School of Dentistry, Universidad Autonoma de Baja California, Tijuana, Baja California, Mexico; and 2Private Practice in

Endodontics, 71OE San Ysidro Blvd., 1513 San Ysidro, California 92173, USA

Abstract

Vieyra JP, Acosta J, Mondaca JM. Comparison of working

length determination with radiographs and two electronic apex

locators. International Endodontic Journal, 43, 16–20, 2010.

Aim To evaluate the accuracy of the Root ZX and

Elements-Diagnostic electronic apex locators when

compared with radiographs for locating the canal

terminus or minor foramen.

Methodology The canal terminus of 482 canals

in 160 maxillary and mandibular teeth was located

in vivo with both locators and radiographically.

After extraction, the actual location of the minor

foramen was determined visually and with magnifica-

tion. A paired samples t-test, chi-square test and a

repeated measure anova at the 0.05 level of signifi-

cance were used to determine differences between the

groups.

Results The Root ZX located the minor foramen

correctly 68% of the time in anterior and premolar

teeth, and 58% of the time in molar teeth. The

Elements-Diagnostic located the minor foramen

correctly 58% of the time in anterior and premolar

teeth and 49% of the time in molar teeth. Radio-

graphs located the minor foramen correctly 20% of

the time in anterior and premolar teeth and 11% of

the time in molar teeth. There was no statistically

significant difference between the two locators, but

there was a significant difference between them and

radiographs. For all teeth, the measurements made by

the apex locators were within ±0.5 mm of the minor

foramen 100% of the time, whereas for the radio-

graphs, the measurements were within this range only

15% of the time. This difference was significant

(P = 0.05).

Conclusion Measuring the location of the minor

foramen using the two apex locators was more

accurate than radiographs and would reduce the risk

of instrumenting and filling beyond the apical foramen.

Keywords: apical constriction, electronic apex loca-

tor, elements-diagnostic, Root ZX, working length

determination.

Received 20 November 2008; accepted 8 July 2009

Introduction

Root canal preparation and filling should not extend

beyond the tooth root nor leave uninstrumented areas

inside the root canal. Anatomically, the apical con-

striction (AC), also called the minor apical diameter or

minor diameter (Kuttler 1955), is a logical location for

working length (WL), as it often coincides with the

narrowest diameter of the root canal (AAE 2003).

However, locating the AC clinically is problematic.

Dummer et al. (1984) concluded that it is impossible to

locate the minor foramen clinically with certainty

because of its position and topography. The cemento-

dentinal junction (CDJ) has also been suggested as the

location for WL, because it represents the transition

between pulpal and periodontal tissue (Grove 1931).

The location of the CDJ is widely accepted as being

0.50–0.75 mm coronal to the apical foramen (Ricucci

& Langeland 1998) but, as with the AC, the exact

location of the CDJ is impossible to identify clinically. In

general, the CDJ is considered to be co-located with the

minor foramen (Stein et al. 1990); however, this is not

always the case (Dummer et al. 1984).

Correspondence: Dr Jorge Paredes Vieyra, PMB#1513, 710E,

San Ysidro Blvd., Suite ‘‘A’’, San Ysidro, CA 92173, USA (Tel.:

+1 619 946 0459; fax: +1 664 687 2207; e-mail:

[email protected]).

doi:10.1111/j.1365-2591.2009.01620.x


Working length is defined as ‘the distance from a

coronal reference point to the point at which canal

preparation and filling should terminate’ (American

Association of Endodontists (AAE) 2003). Radiographic

determination of WL has limitations such as distortion,

shortening and elongation, interpretation variability

and lack of three-dimensional representation. Even

when a paralleling technique is used, elongation of

images has been found to be approximately 5% (Van de

Voorde & Bjondahl 1969).

A WL 1 mm short of the radiographic apex may

result in over or under instrumentation because of the

variability in distance between the terminus of the root

canal (minor foramen) and the radiographic apex

(Gutierrez & Aguayo 1995). Thus, this often used ‘rule’

is not predictable or reliable.

Custer (1918) was the first to determine WL

electronically. Suzuki (1942) investigated the electrical

resistance properties of oral tissues and developed the

first electronic apex locator (EAL). The device was

resistance-based and measured the resistance between

two electrodes to determine the location of an instru-

ment in the canal. Later devices were impedance-based

(Nekoofar et al. 2006) and used multiple frequencies.

More recently, resistance- and capacitance-based

devices emerged that measure resistance and capa-

citance, directly and independently.

The Root ZX (J. Morita Corp., Tokyo, Japan) uses the

‘ratio method’ to locate the minor foramen (Kobayashi

& Suda 1994) by the simultaneous measurement of

impedance using two frequencies. The Root ZX claims

to work in the presence of electrolytes and nonelec-

trolytes and requires no calibration (Kobayashi 1995).

The Elements-Diagnostic (Sybron Endo, Sybron Den-

tal, Orange, CA, USA) uses multiple frequencies, in an

attempt to eliminate the influence of canal conditions.

In addition to improving WL accuracy (Nekoofar

et al. 2006), EALs address concerns about radiation,

as they have the potential to reduce the number of

radiographs taken during root canal treatment

(Pagavino et al. 1998).

The purpose of this study was to evaluate in vivo the

accuracy and predictability of two EALs for determining

WL as compared with radiographs.

Materials and methods

One hundred and sixty teeth (482 canals) with fully

formed apices and without apical resorption were used

(Table 1). All teeth gave positive responses to hot and

cold tests and were extracted for periodontal or

prosthodontic reasons. Ethical approval for the study

and an informed consent to participate was signed by

the patients.

After local anaesthesia, rubber dam isolation and

access cavity preparation were performed, the canals

were flared coronally with size 1 and 2 Orifice Shapers

(Dentsply Tulsa Dental, Tulsa, OK, USA) using 3%

sodium hypochlorite (NaOCl) for irrigation. The final

rinse was aspirated, but no attempt was made to dry

the canals.

The AC of each tooth was located with two EALs and

radiographically.

The minor foramen was located with the Root ZX by

advancing a size15 stainless steel K-file in the canal,

until the locator indicated that the minor foramen had

been reached, according to the manufacturer’s instruc-

tions (J. Morita Corp. 2004). The LCD showed a

flashing bar between APEX and 1 and a flashing tooth.

The silicone stop on the file was positioned at the

reference point. The file was removed from the canal

and the length was measured to the nearest 0.01 mm

with a digital caliper. This was the insertion length.

The AC was located with the Elements-Diagnostic

EAL by advancing the same size 15 K-file in the canal,

until the locator indicated that the minor foramen had

been reached, as per the manufacturer’s instructions

(Sybron Endo 2003). The stop was positioned at the

reference point and the insertion length measured. The

sequence of testing alternated between the two loca-

tors.

The minor foramen was located radiographically by

advancing the size15 K-file, until its tip was 1.0 mm

from the radiographic apex (determined from a pre-

treatment parallel technique radiograph). A radiograph

was exposed and if the file tip was seen not to be

1.0 mm from the radiographic apex, the file was

repositioned and another radiograph taken to ensure

that it was. The distance from the stop to the tip was

the insertion length. The file was then re-inserted to

the insertion length (1 mm from the radiographic

Table 1 Distribution of 160 teeth (482 canals)

Tooth (n)

Number of canals

Maxillary Mandibular

Central incisor (10) 7 3

Lateral incisor (8) 6 2

Canine (5) 3 2

Premolar (17) 11 6

Molar (120) 225 217

Total (160) 252 230

Vieyra et al. Working length determination


apex) and cemented in place with Fuji II LC dual-cure

glass ionomer cement (GC Corp, Tokyo, Japan). The

file handle was sectioned with a high-speed bur and

the tooth was extracted without disturbing the file,

placed in 6% NaOCl for 15 min to clean the root

surface and stored in a 0.2% thymol solution. All of the

clinical procedures were conducted by the principal

investigator.

After the tooth was removed from the solution and

with the file still in place, the apical 5 mm of the root

was ground parallel to the long axis of the canal with a

fine diamond bur and abrasive discs. When the file

became visible, additional dentine was removed under

20 · magnification (OPMI Pico microscope; Carl Zeiss,

Munich, Germany) until the file tip, the canal terminus,

and the foramen were in focus. A digital photograph

was taken and stored in Adobe Photoshop 5.5 (Adobe

Systems Inc., San Jose, CA, USA) and the distance of

the file tip to the minor foramen was measured. This

distance was recorded as being: )1.0 mm from the

minor foramen; )0.5 mm from the minor foramen; at

the minor foramen; +0.5 mm from the minor foramen

or +1.0 mm from the minor foramen. A minus symbol

()) indicated a file short of the minor foramen; a plus

symbol (+) indicated it was long.

Once the actual length to the minor foramen was

measured visually, the distance from the minor fora-

men determined by the two EALs was also completed

()1.0 mm from the minor foramen; )0.5 mm from the

minor foramen, etc.), by comparing their insertion

lengths to the actual length (distance to the AC)

(Tables 2–4).

The measurements obtained by the two EALs and

radiographs relative to the actual location of the minor

foramen were compared using a paired samples t-test,

chi-square test and a repeated measure. anova evalu-

ation was conducted at the 0.05 level of significance.

Results

For anterior teeth, the Root ZX, Elements and radio-

graphs located the minor foramen 74%, 65% and 22%

of the time, respectively. For premolar teeth, the Root

ZX, Elements and radiographs located the minor

foramen 53%, 41% and 35% of the time, respectively.

For molar teeth, the Root ZX, Elements and radiographs

located the minor foramen 58%, 49% and 11% of the

time, respectively. There was no statistically significant

difference between the two EALs, but there was a

difference when the EALs and radiographs were com-

pared (Tables 2–4).

Table 2 Distance of file tip from minor foramen determined by

Root ZX, Elements and radiograph (anteriors)

Distance from minor

foramen (mm)

Root ZX Elements Radiograph

n = 23 (%) n = 23 (%) n = 23 (%)

)1.0 – – –

)0.5 – – –

MF 17 (73.9) 15 (65.2) 5 (21.7)

+0.5 6 (26.08) 8 (34.7) 10 (43.47)

+1.0 8 (34.78)

MF, minor foramen.

(+) and ()) values indicate file tip beyond (+) or short ()) of the

AC.

Table 3 Distance of file tip from minor foramen determined by

Root ZX, Elements and radiograph (premolars)

Distance from minor

foramen (mm)

Root ZX Elements Radiograph

n = 17 (%) n = 17 (%) n = 17 (%)

)1.0 – – –

)0.5 – – –

MF 9 (52.94) 7 (41.17) 6 (35.29)

+0.5 8 (47.05) 10 (58.82) 5 (29.41)

+1.0 6 (35.29)

MF, minor foramen.

(+) and ()) values indicate file tip beyond (+) or short ()) of the

AC.

Table 4 Distance of file tip from minor foramen determined by Root ZX, Elements and radiograph (molars)

Distance from minor foramen (mm)

Root ZX (n = 444) Elements (n = 423) Radiograph (n = 414)

Canal Canal Canal

MB ML D DB DL Pa MB ML D DB DL Pa MB ML D DB DL Pa

)1.0 – – – – – – – – – – – – – – – – – –

)0.5 2 3 – – – – 6 5 18 – – 8 1 – 2 – – 1

MF 65 61 58 19 19 38 66 59 54 18 19 2 11 8 12 6 5 5

+0.5 53 45 27 16 16 22 48 41 15 22 21 21 61 41 47 17 17 28

+1.0 – – – – – – – – – – – – 43 34 26 11 12 26

MF, minor foramen.

(+) and ()) values indicate file tip beyond (+) or short ()) of the AC.

Working length determination Vieyra et al.


For anterior, premolar and molar teeth, none of the

measurements were 1.0 mm short of the minor fora-

men. For anterior and premolar teeth, none of the

measurements were 0.5 mm short of the minor fora-

men, but for molar teeth 1%, 8% and 1% of the

measurements using the Root ZX, Elements and radio-

graphs, respectively, were short.

For anterior teeth, the Root ZX, Elements and

radiographs were 0.5 mm long of the minor foramen

a 26%, 35% and 39% roots, respectively. For premolar

teeth, the Root ZX, Elements and radiographs were

0.5 mm long of the minor foramen 47%, 59% and 29%

roots, respectively, and for molar teeth it was 41%,

42% and 48%, respectively.

No EAL measurements were 1.0 mm long of the

minor foramen for anterior, premolar and molar teeth,

but for radiographs it was 35% for anterior teeth, 35%

for premolar teeth and 37% for molar teeth. There

was no statistically significant difference between the

two EALs, but there was a significant difference

(P = 0.05) when the EALs and radiographs were

compared.

Discussion

The use of electronic devices to determine WL

has gained in popularity. When using them, an

important consideration is being aware of the possi-

ble sources of error such as metallic restorations,

salivary contamination, dehydration, etc. However, as

shown in this and other studies, the accuracy of EALs is

superior to radiographs (Van de Voorde & Bjondahl

1969, Pratten & McDonald 1996, Venturi & Breschi

2007).

One of the reasons why a radiographically deter-

mined WL lacks accuracy is that it is based on the

radiographic apex rather than the canal terminus – the

minor foramen. WL is obtained with a radiograph by

positioning the tip of a file a certain distance (usually

1.0 mm) from the radiographic apex. However, WL

should be based on the location of the minor foramen

rather than the apex, because the foramen frequently is

not at the apex (Wrbas et al. 2007). In this study,

radiographs correctly located the minor foramen 15%

of the time, whereas for the Root ZX and Elements it

was 63% and 53% of the time, respectively. Both EALs

were within ±0.5 mm from the minor foramen 100%

of the time, whereas radiographs were within ±0.5 mm

of 63% of cases. An in vivo study by Shabahang et al.

(1996) reported that the Root ZX was within 0.5 mm

from the minor foramen 96% of the time, a value

similar to the present findings. In general, this study

also agrees with others (Usun et al. 2007, 2008) that

EALs are more accurate than radiographs and greatly

reduce the risk of instrumenting and filling short or

beyond the canal terminus.

As the minor foramen varies in location and anat-

omy (sharply defined, parallel, or missing) (Nekoofar

et al. 2006), caution should be used to avoid over-

estimating WL. According to Gutierrez & Aguayo

(1995), over-instrumentation of the root canal must

be a common and unnoticed occurrence. An instru-

ment passing through a necrotic pulp and through the

foramen most likely carries bacteria and toxins into the

apical tissues (Siqueira et al. 2002, Siqueira & Barnett

2004). An indication by an EAL of reaching the minor

foramen or foramen is very helpful in avoiding

mishaps. Indeed this study showed that WL obtained

with radiographs was 1.0 mm long of the AC 37% of

the time, but 0% for the two EALs. This high incidence

of error is clinically important, because a WL 1.0 mm

long would result in canals being instrumented beyond

the foramen.

Conclusion

Under clinical conditions, the EALs identified the minor

foramen with high degree of accuracy. EALs were more

accurate compared with radiographs with the potential

to greatly reduce the risk of instrumenting and filling

beyond the apical foramen.

Acknowledgements

We thank Dr E. Steve Senia and Dr Michael Hulsmann

for their valuable assistance in reviewing this manu-

script.

References

American Association of Endodontists (AAE) (2003) Glossary

of Endodontic Terms, 7th edn. Chicago, IL: American

Association of Endodontists.

Custer LE (1918) Exact method of locating the apical foramen.

Journal of the National Dental Association 5, 815–9.

Dummer PMH, McGinn JH, Rees DG (1984) The position and

topography of the apical canal constriction and apical

foramen. International Endodontic Journal 17, 192–8.

Grove CJ (1931) The value of the dentinocemental junction

in pulp canal surgery. Journal of Dental Research 11,

466–8.

Gutierrez JH, Aguayo P (1995) Apical foraminal openings in

Human teeth. Number and location. Oral Surgery, Oral

Vieyra et al. Working length determination


Medicine, Oral Pathology, Oral Radiology, and Endodontics 79,

769–77.

J. Morita Corp. (2004) Fully Automatic Root Canal Measuring

Device. Root ZX Operation. Tokyo: J. Morita Corp.

Kobayashi C (1995) Electronic canal length measurement.

Oral Surgery, Oral Medicine, Oral Pathology, Oral Radiology,

and Endodontics 79, 177–9.

Kobayashi C, Suda H (1994) New electronic canal measuring

device based on the ratio method. Journal of Endodontics 20,

111–4.

Kuttler Y (1955) Microscopic investigation of root apexes.


Nekoofar MN, Ghandi MM, Hayes SJ, Dummer PMH (2006)

The fundamental operating principles of electronic root

canal length measurement devices. International Endodontic

Journal 39, 595–609.

Pagavino G, Pace R, Baccetti T (1998) An SEM study of in vivo

accuracy of the Root ZX electronic apex locator. Journal of


Pratten DH, McDonald NJ (1996) Comparison of radiographic

and electronic working lengths. Journal of Endodontics 22,

173–6.

Ricucci D, Langeland K (1998) Apical limit of root canal

instrumentation and obturation, part 2: a histological study.


Shabahang S, Goon WWY, Gluskin AH (1996) An in vivo

evaluation of Root ZX electronic apex locator. Journal of


Siqueira JF, Barnett F (2004) Interappointment pain: mecha-

nisms, diagnosis, and treatment. Endodontic Topics 7, 93–109.

Siqueira JF Jr, Rocas IN, Favieri A, et al. (2002) Incidence of

postoperative pain after intracanal procedures based on an

antimicrobial strategy. Journal of Endodontics 6, 457–60.

Stein TJ, Corcoran JF, Zillich RM (1990) The influence of the

major and minor foramen diameters on apical electronic

probe measurements. Journal of Endodontics 16, 520–2.

Suzuki K (1942) Experimental study on iontophoresis. Journal

of the Japanese Stomatology 16, 411.

Sybron Endo (2003) Elements Diagnostic: Instruction Guidelines.

Glendora, CA: Sybron Endo, pp. 1–3.

Usun O, Topuz O, Tinaz AC, Sadik B (2007) Accuracy of the

apex locating function of TCM Endo V in simulated

conditions: a comparison study. Oral Surgery, Oral Medicine,

Oral Pathology, Oral Radiology, and Endodontics 103, e73–6.

Usun O, Topuz O, Tinaz C, Nekoofar MH, Dummer PMH

(2008) Accuracy of two root canal length measurement

devices integrated into rotary endodontic motors when

removing gutta-percha from root-filled teeth. International


Van de Voorde HE, Bjondahl AM (1969) Estimating endodon-

tic ‘‘working length’’ with paralleling radiographs. Oral

Surgery, Oral Medicine, and Oral Pathology 27, 106–10.

Venturi M, Breschi L (2007) A comparison between two

electronic apex locators: an ex vivo investigation. Interna-


Wrbas KT, Ziegler AA, Altenburger MJ, Schirrmeister JF

(2007) In vivo comparison of working length determination

with two electronic apex locators. International Endodontic

Journal 40, 133–8.

Working length determination Vieyra et al.


Evaluation of the radiopacity of calcium silicatecements containing different radiopacifiers

J. Camilleri1,2 & M. G. Gandolfi3,4

1Department of Building and Civil Engineering, Faculty for the Built Environment, University of Malta, Msida; 2Department of

Dental Surgery, Faculty of Dental Surgery, University of Malta, Msida, Malta; 3Department of Earth Sciences, University of

Bologna, Bologna; and 4Department of Odontostomatological Sciences-Endodontic Section, University of Bologna, Bologna, Italy

Abstract

Camilleri J, Gandolfi MG. Evaluation of the radiopacity of

calcium silicate cements containing different radiopacifiers.

International Endodontic Journal, 43, 21–30, 2010.

Aim To identify the suitable ratio of alternative

radiopacifiers to impart the necessary radiopacity to

calcium silicate cements (CSC) and assess the purity of

the radiopacifying agents.

Methodology Alternative radiopacifying materials

for incorporation into CSC included barium sulphate,

titanium oxide, zinc oxide, gold powder and silver/tin

alloy. The chemical composition of the alternative

radipacifying materials and bismuth oxide, which is

used in mineral trioxide aggregate (MTA), was deter-

mined using energy dispersive X-ray analysis. In

addition, using an aluminium step-wedge and densi-

tometer, the radiopacity of each material was evaluated

as recommended by international standards. The opti-

cal density was compared with the relevant thickness of

aluminium (Al). A commercial MTA and CSC were

used as controls. Statistical analysis comparing the

radiodensity of the different cements to MTA was

performed using anova with P = 0.05 and post hoc

Tukey test.

Results All percentage replacements of bismuth oxide,

gold and silver–tin alloy powder, and the 25% and 30%

replacements with barium sulphate and zinc oxide had

radiopacities greater than 3 mm thickness of aluminium

(Al) recommended by ISO 6876 (2002). The 25%

replacement of cement with gold powder and 20%

replacement of cement with silver/tin alloy powder

exhibited radiopacity values of 8.04 mm Al and

7.52 mm Al, respectively, similar to MTA (P > 0.05).

The cement replaced with 20% bismuth oxide showed a

radiopacity of 6.83 mm Al, lower than MTA (P = 0.003).

Conclusions Silver/tin alloy and gold powder im-

parted the necessary radiopacity to a calcium silicate-

based cement. Barium sulphate was also a suitable

radiopacifier together with a lower concentration of

silver/tin alloy and gold powder that achieved the

radiodensity recommended by ISO 6876. Further

research is required to investigate the broader proper-

ties of the calcium silicate-based cement with the

different radiopacifiers.

Keywords: bismuth oxide, calcium silicate-based

cement, chemical composition, mineral trioxide aggre-

gate, radiopacity.

Received 20 April 2009; accepted 30 June 2009

Introduction

Calcium silicate-based cements [white Portland cement

and mineral trioxide aggregate (MTA)] are hydraulic

cements composed primarily of tricalcium silicate,

dicalcium silicate and tricalcium aluminate (Taylor

1997, Camilleri et al. 2005, Camilleri 2008b). In

dentistry, MTA is used amongst other things to seal

lateral root perforations (Lee et al. 1993, Pitt Ford et al.

1995) and as a root-end filling material (Torabinejad

et al. 1995a, 1997, Chong et al. 2003, Saunders

2008). Calcium silicate cements (CSC) without radiop-

acifying additives have intrinsic radiopacity values

ranging from 0.86 to 2.02 mm aluminium (Al) (Islam

Correspondence: Dr Josette Camilleri PhD, Department of

Building and Civil Engineering, Faculty for the Built Environ-

ment, University of Malta, Msida MSD 2080, Malta (Tel.: 356

2340 2870; fax: 356 21330190; e-mail: josette.camiller-

[email protected]).

doi:10.1111/j.1365-2591.2009.01621.x


et al. 2006, Kim et al. 2008, Saliba et al. 2009), values

lower than the 3 mm aluminium recommended by the

International Standards for dental root canal sealing

materials (ISO 6876 Section 7.8 2002). Thus, a

radiopacifying material has to be added to calcium

silicate-based cements to allow the cement to be

detected radiographically and thus distinguished from

surrounding anatomical structures (Beyer-Olsen &

Ørstavik 1981).

Mineral trioxide aggregate is said to be composed of a

mixture of CSC and bismuth oxide in 4 : 1 ratio

(Torabinejad & White 1995). Bismuth oxide is added

to the MTA to increase the radiopacity of the material.

MTA is commercially available as white and grey

ProRoot MTA (Dentsply, Tulsa Dental Products, Tulsa,

OK, USA) and white and grey MTA-Angelus (Angelus

Solucoes Odontologicas, Londrina, Brazil). The addition

of bismuth oxide increased the radiopacity of the

material to higher levels than the equivalence of

3 mm Al suggested by ISO 6876 (2002). ProRoot

was reported to have a radiopacity ranging from

5.34 mm Al to 6.92 mm Al (Laghios et al. 2000, Chng

et al. 2005, Danesh et al. 2006, Islam et al. 2006, Kim

et al. 2008). White ProRoot MTA showed higher

radiopacity than the grey version (Chng et al. 2005,

Islam et al. 2006, Tanomaru-Filho et al. 2008). MTA-

Angelus demonstrated a radiopacity of 3–3.3 mm Al

(Tanomaru-Filho et al. 2008).

Most of the materials used in endodontics have

radiopacifying agents added to them such as barium or

bismuth compounds. There have been few investiga-

tions on the effect that radiopacifiers have on the

properties of materials. The bismuth oxide added to

white ProRoot MTA has been shown to affect its

hydration mechanism. The bismuth formed part of the

structure of calcium silicate hydrate, replacing the

silica in its structure. Approximately 5% by weight of

bismuth was attached to the calcium silicate hydrate

structure. Bismuth oxide reduced the precipitation of

calcium hydroxide in the hydrated paste (Camilleri

2007) and was also leached out from the material

together with calcium hydroxide (Camilleri 2008b). It

has been reported that bismuth is toxic (Bloodworth &

Render 1992) and induces cell death (Camilleri et al.

2004). Other researchers demonstrated that CSC con-

taining bismuth oxide induced cytotoxicity in dental

pulp cells (Min et al. 2007). Conversely, most research

performed on the biocompatibility of MTA has proved

that this material is biocompatible and induces cell

growth and activity. This may demonstrate that

addition of bismuth oxide to CSC does not seem to

affect the biocompatibility of the material (Kim et al.

2008, Koulaouzidou et al. 2008).

The use of bismuth oxide with CSC has been shown

to be deleterious to the physical properties of the

material, particularly the compressive strength in a

concentration-related manner (Coomaraswamy et al.

2007). This is in accordance with other reports where

Portland cement clinker was used (Camilleri 2008a)

but in opposition to studies reporting no significant

difference in the strength of Portland cement with

varying additions of bismuth oxide (Saliba et al. 2009).

The difference in the results of the studies performed

could be due to nonstandardization of testing when

performing compressive strength tests (Camilleri et al.

2006).

The bismuth oxide in MTA can be replaced by other

radiopacifying materials. Ideally, an alternative radi-

opacifier should only impart the necessary radiopacity

to the cement and should be inert, free from any

contaminants, colourless and nontoxic and be added in

minimal amounts. Addition of minimal amounts of any

material necessitates the use of elements that have a

high relative atomic mass. A number of materials with

high relative atomic masses are already used in dental

practice. Such materials include zinc oxide, which is

used in restorative dentistry, endodontics and peri-

odontology as a base material, root canal sealer and as

a dressing, respectively. Silver–tin alloy is the alloy used

in dental amalgam, gold is used in alloyed form in cast

restorations, and titanium is used for the construction

of endosseous implants. Barium sulphate is used

extensively in the medicine as a radiopacifier for

colonoscopies. The extensive use of these materials in

both medicine and dentistry indicates that the materi-

als have been well researched and thus their interac-

tion with the host tissues should be favourable.

However, not all have been evaluated for implantation

into deep sites.

The addition of a radiopacifier even in minimal

amounts can affect the physical properties of the

resultant material. The replacement of the cement with

a noncementitious material affects the water to cement

ratio (Neville 1981). In turn, variations to the water to

cement ratio affects the workability and the strength of

the resultant material. Other factors that affect the

water required to achieve a workable mixture include

particle size distribution and particle shape.

This study aimed at identifying the suitable ratio of

alternative radiopacifiers for CSC and assessing the

purity and physical properties of the radiopacifiying

agents.

Radiopacity of calcium silicate cements Camilleri & Gandolfi



The materials used in this study were white calcium

silicate-based cement (Aalborg White, Aalborg, Den-

mark manufactured to BS EN 197-1: British Standard

Institution 2000, type CEM I), and six radiopacifying

materials, which included titanium dioxide (rutile;

Sigma-Aldrich, Gillingham, UK), zinc oxide (Fischer

Scientific, Leicester, UK), barium sulphate (Sigma-

Aldrich), gold powder (Sigma-Aldrich), bismuth oxide

(Fischer Scientific), and silver/tin alloy powder (De-

gussa Dental GmbH, Hanau, Germany). The radiopa-

cifiers and the cement were placed in plastic containers

and were blended by placing the container on a rotary

shaker for 15 min (Luckham 4RT, Burgess Hill, UK).

White ProRoot MTA (Dentsply, Tulsa Dental Products)

and white MTA-Angelus (Angelus Solucoes Odontolog-

icas) were used as controls and received no additional

radiopacifier.

Scanning electron microscopy and elemental

analysis

Scanning electron microscopy of the radiopacifiers was

performed in order to determine the particle shape of the

materials. A thin layer of powder was dispersed onto an

aluminium stub (Agar Scientific, Stansted, UK) over

double-sided carbon tape and then carbon coated (Agar

Scientific) for electrical conductivity. The specimens

were then observed by scanning electron microscope

(SEM; Leo 1430, Philips, Cambridge, UK) and photomi-

crographs were recorded. Energy dispersive X-ray anal-

ysis (EDX) was performed to determine the constituent

elements. Semi-quantitative analysis of cements and

radiopacifiers was performed using a cobalt standard.

Two samples for each material were prepared and the

analysis was performed twice for each sample.

Particle size distribution of powders

The particle size distribution of the radiopacifying

materials was determined using a laser particle size

analyser (CILAS 1180, Orleans, France) having a range

of 0.04–2500 lm.

Evaluation of radiopacity

The radiopacifying materials were added to the calcium

silicate-based cement by replacing 10%, 15%, 20%,

25% and 30% of the cement by weight. Calcium

silicate-based cement without additive, ProRoot MTA

and MTA-Angelus were used as controls. The experi-

mental protocol was based on ISO 6876 Section 7.8

(2002) for dental root canal sealing materials. The

cement and the cement containing the radiopacifiers

were mixed with water, at a water to cement ratio of

0.30 measured by weight of materials. The MTAs were

mixed according to the manufacturer’s instructions.

The materials were compacted incrementally using

hand pluggers into stainless steel ring moulds 10 mm

in diameter and 1 mm high, and pressed against two

glass cover slips to make the specimens 1 mm thick.

Three specimens of each material were prepared. The

cements were allowed to cure for 24 h at 37 �C and

100% relative humidity covered by a plastic sheet to

avoid cement desiccation. After removal from the

moulds, they were stored in distilled water at 37 �C

for 7 days.

The cements were placed directly on a cassette

loaded with a cephalostat type film with an intensifying

screen (Kodak, Rochester, NY, USA) adjacent to a 10-

step aluminium step-wedge made of aluminium, where

each step measured 1 mm in height (Agfa Mamoray,

Agfa Gevaert, Mortsel, Belgium) and X-ray irradiated

using a standard X-ray machine (GEC Medical Equip-

ment Ltd., Middlesex, UK) at tube voltage of 50 kV, and

current 50 mA and exposure time of 0.05 s. The target

to film distance was set at 100 cm. Three specimens per

material under test were arranged on the cassette and

two radiographs were taken of the specimens. Eight

layers of lead foil covered a small area of each film to

obtain a small area of nonexposure. The radiographs

were processed in an automatic processing machine

(Clarimat 300, Gendex Dental Systems, Medivance

Instruments Ltd., London, UK). A photographic densi-

tometer (PTWdensix, Freiburg, Germany) was used to

measure the density of the radiographic images of the

specimens, of each aluminium step and the un-exposed

part of the film. Three density values of each material

were obtained for each radiograph of each specimen

and the mean density was calculated. The net radio-

graphic density was calculated by subtracting the base

and fog value from the gross radiographic density. The

base and fog value is the inherent optical transmission

density (lowest density) of a film base plus the

nonimage density contributed by the developed emul-

sion. Graphs were plotted for net radiographic density

of the aluminium steps (NRDAL) versus the logarithm of

the thickness of aluminium (log d) for each radiograph.

From the resultant plots, the gradient and the intercept

were calculated for each film. Linear regression of the

data was obtained using the following formula:

Camilleri & Gandolfi Radiopacity of calcium silicate cements


NRDAL ¼ m: log d þ I

where NRDAL was the net radiographic density of the

aluminium step wedge, m was the gradient, log d was

the logarithm of the step height, I was the intercept.

By rearranging the above equation into:

log d ¼ I � NRD

m

The logarithm of the relevant thickness of alumin-

ium for each material could be calculated from its net

radiographic density for each film taking into consid-

eration that specimen thickness was 1 mm. Logarithms

of step height were then converted to thicknesses of

aluminium (Watts & McCabe 1999).

Statistical analysis

The data was evaluated using Statistical Package for

the Social Sciences (SPSS) software (SPSS Inc., Chicago,

IL, USA). The distribution was first evaluated to

determine what kind of statistical test would be

performed. The Kolmogorov–Zmirnov test revealed

normal distribution and parametric statistics using

anova with post hoc Tukey test was performed.

Results

Scanning electron microscopy and elemental

analysis

Scanning electron micrographs of the radiopacifying

materials are shown in Fig. 1(a–f). Both zinc oxide and

barium sulphate (Fig. 1b,c respectively) were composed

of very fine particles that were difficult to discern by

SEM even at a magnification of ·2000. Thus, the

magnification was increased until the individual par-

ticles could easily be identified on the micrographs. The

bismuth oxide was composed of elongated needle-

shaped particles (Fig. 1e). The particle shape of gold

(Fig. 1d) and titanium (Fig. 1a) were spherical and the

silver/tin alloy was composed of lathe cut particles

(Fig. 1f). The results of semi-quantitative analysis

(a) (b)

(c) (d)

(e) (f)Figure 1 Scanning electron micro-

graphs of (a) titanium oxide (b) zinc

oxide (c) barium sulphate (d) gold

powder (e) bismuth oxide (f) silver/tin

alloy powder (2000· magnification).

The micrographs for barium sulphate

and zinc oxide were at higher magnifi-

cations because of the small size of the

powders.



(elemental analysis percentage) of the cements and

radiopacifying materials are shown in Table 1, and the

EDX spectra are shown in Fig. 2. The radiopacifying

materials were mostly pure. The silver/tin alloy powder

had an inclusion of approximately 4% copper. The

calcium silicate-based cement and MTAs were com-

posed of calcium, silicon and aluminium with MTAs

including bismuth. The bismuth loading of MTA-

Angelus was lower than that of ProRoot MTA.

Particle size distribution

The particle size distribution of the radiopacifying

materials is shown in Fig. 3. The barium sulphate

and zinc oxide had very fine particles. Most of the

gold particles ranged between 2 and 4 lm. The

silver/tin alloy powder and bismuth oxide had a

wider range of particle sizes ranging from 0.1 to

53 lm for the silver/tin alloy and 5–100 lm for the

bismuth oxide with the bismuth oxide exhibiting the

largest particle sizes compared with the other radi-

opacifying agents.

Evaluation of radiopacity

The radiopacity values of the materials tested are

shown in Fig. 4. The calcium silicate-based cement

exhibited a low intrinsic radiopacity value of

Table 1 Semi-quantitative analysis of powders using cobalt standard

Material

Elemental analysis (%)

Ag Al Au Ba Bi Ca Cu O S Si Sn Ti Zn

Calcium silicate-based cement 0 0.7 0 0 0 19 0 40 0 9 0 0 0

Titanium dioxide 0 0 0 0 0 0 0 77 0 0 0 22 0

Zinc oxide 0 0 0 0 0 0 0 22 0 0 0 0 90

Silver/tin alloy powder 66 0 0 0 0 0 4 0 0 0 26 0 0

Barium sulphate 0 0 0 7 0 0 0 90 3 0 0 0 0

Gold powder 0 0 100 0 0 0 0 0 0 0 0 0 0

Bismuth oxide 0 0 0 0 62 0 0 12 0 0 0 0 0

White MTA (Dentsply) 0 0.6 0 0 8 29 0 45 0 6 0 0 0

White MTA (Angelus) 0 1.7 0 0 5 32 0 42 0 6 0 0 0

(a) (b)

(c) (d)

(e) (f)

Figure 2 Elemental analysis of (a) tita-

nium oxide (b) zinc oxide (c) barium

sulphate (d) gold powder (e) bismuth

oxide (f) silver/tin alloy powder.



1.62 ± 0.29 mm Al. The bismuth oxide, gold and

silver/tin alloy replaced cements at all the percentage

replacements and 25–30% cement replaced with bar-

ium sulphate and zinc oxide showed radiopacity values

greater than 3 mm thickness of Al. Thus, these

replaced cements complied with the recommendations

by ISO 6876 Section 7.8 (2002). The radiopacity of

calcium silicate-based cement, all additions of titanium

oxide, the 10% and 15% replacement with barium

sulphate and 10%, 15% and 20% replacement with

zinc oxide demonstrated a radiopacity lower than

3 mm thickness of aluminium.

MTA-Angelus had a lower radiopacity value than

ProRoot MTA (P < 0.001). The MTA-Angelus dis-

played a similar radiopacity to CSC with 10% gold

and silver/tin alloy, to 10% and 15% bismuth oxide

and 25% and similar to 30% barium sulphate and zinc

oxide (P > 0.05). The 25% replacement of cement with

gold powder and 20% replacement of cement with

silver/tin alloy powder displayed radiopacity values of

8.04 ± 0.67 mm Al and 7.52 ± 0.20 mm Al, respec-

tively, similar to ProRoot MTA (P > 0.05). The

cement replaced with 20% bismuth oxide revealed a

radiopacity of 6.83 ± 0.48 mm Al. The radiopacity of

this cement was lower than that of ProRoot MTA

(P = 0.003).

Discussion

In the present study, the radiopacity of calcium silicate-

based cement with varying additions of different

radiopacifying materials was investigated. The different

0

10

20

30

40

50

60

70

80

90

100

0.04 0.

10.

30.

50.

70.

91.

11.

31.

6 22.

4 3 56.

57.

58.

5 10 12 14 16 18 20 25 32 38 45 53 63 75 85 95 106

125

140

150

Diameter µm

Cu

mu

lati

ve v

alu

e %

Titanium oxide

Zinc Oxide

Barium sulphate

Gold

Bismuth oxide

Silver/Tin

Figure 3 Particle size analysis of

radiopacifying materials using laser

particle size distribution.

0.00

1.00

2.00

3.00

4.00

5.00

6.00

7.00

8.00

9.00

10.00

11.00

12.00

No addition Titaniumoxide

Zinc oxide Bariumsulphate

Goldpowder

Bismuthoxide

Silver/tinalloy

MTADentsply

MTA-Angelus

Addition of radiopacifier to calcium silicate-based cement %

Mea

n t

hic

knes

s o

f al

um

iniu

m m

m

10% 15% 20% 25% 30%

Figure 4 Radiopacity of cement with

varying additions of radiopacifying

materials expressed as mean thicknesses

of aluminium ± SD (n = 2). The dotted

line shows the minimum value for

radiopaque restorative material.



radiopacifying materials were added by replacing the

amount by weight in percentages varying from 10–

30%.

The materials used to render the cement radiopaque

were powdered metals, metal oxides or salts. The

materials used in this study had no major contami-

nants; thus, the radiopacity increase was due to the

radiopacifier itself and not caused by contamination.

The lack of contaminants was checked on the material

data sheets provided with the chemicals and verified by

semi-quantitative analysis (EDX). The alternative rad-

iopacifying materials replaced the cement portion by

weight. This was carried out to standardize the amount

of radiopacifier added to the cement. The radiopacifier

materials used had different relative atomic masses

which affected the amount of material added to make

up the necessary weight of material used. The water to

cement ratio was standardized and other properties

affecting the physical properties of the resultant mate-

rial were investigated. These included the particle

shape and the particle size distribution. Addition of

small-sized particles increases the specific surface area

and would potentially make the mixture less workable

if the water to cement ratio is kept constant. The effect

that particle shape and particle size distribution of a

cement-replacing material have on the physical prop-

erties of the resultant material are still to be investi-

gated.

The radiopacifiers were chosen because of their easy

availability and the long-term use in dental clinical

practice. In dentistry, titanium is used in dental

composites and porcelain to add whiteness and opacity;

zinc oxide as a base materials under plastic restorative

materials, as temporary restorative materials, as root

canal sealers, as a dressing in periodontal surgery and

is the main constituent compound in gutta-percha

points; silver/tin alloy powder is the c phase of dental

amalgam (Van Noort 2002). The alloy used in the

present study was composed of lathe-cut particles. The

amount of copper was lower than that anticipated for

high copper silver/tin alloys; however, the tests carried

out were semi-quantitative, and thus the precise

amount of copper present could not be determined.

Barium sulphate is a compound characterized by an

extremely low solubility and is clinically used as a

radio-contrast agent for X-ray imaging and other

diagnostic procedures (Ott & Gelfand 1983) such as

imaging of the gastrointestinal tract. It is also used in

root canal filling materials. It has been reported that

the addition of barium sulphate to glass ionomer

cement at low concentrations reduced working and

initial setting times, but further addition delayed the

setting reaction of glass ionomer cements. However,

both compressive strength and surface hardness

decreased with increasing concentrations of the radi-

opacifier (Prentice et al. 2006). The effects of barium

sulphate on the physical–chemical properties of Port-

land cement are yet to be investigated.

In dentistry, bismuth oxide is used as a radiopacify-

ing agent for dental material such as dental acrylic

resin and MTAs. ProRoot MTA has a 20% loading of

bismuth oxide as reported by the manufacturer. The

amount present in MTA-Angelus is not specified by the

manufacturer. The 20% loading for the ProRoot was

verified by Rietveld X-ray diffraction analysis (Camilleri

2008b). Other researchers reported both higher (Song

et al. 2006) and lower bismuth oxide loading for

ProRoot MTA (Oliveira et al. 2007). The bismuth oxide

loading of MTA-Angelus varied from 38.8% to 9%

reported in different studies (Song et al. 2006, Oliveira

et al. 2007, respectively). In the current study, EDX

analysis was used to determine both the purity of the

radopacifiers and also the bismuth oxide loading of

both ProRoot and MTA-Angelus. EDX provides only

semi-quantitative analysis which would explain the

lower level of bismuth oxide detected in ProRoot MTA

when compared with other studies (Camilleri 2008b)

and the data provided by the manufacturer. Bismuth

oxide used in this study was composed of elongated

needle-shaped crystals similar to that previously

observed in ProRoot MTA (Camilleri et al. 2005,

Camilleri 2007).

In the present study, calcium silicate-based cement

revealed an intrinsic radiopacity of 1.62 mm Al, in

accordance with previous studies that reported radio-

pacity values equivalent to 0.95 mm Al (Islam et al.

2006), 2.02 mm Al (Saliba et al. 2009) and 0.86 mm

Al (Kim et al. 2008). This intrinsic value does not

satisfy the recommendation of the International Stan-

dards for dental root canal sealing materials (ISO 6876

Section 7.8 2002). Others have reported radiopacities

equivalent to 3.32 mm Al (Danesh et al. 2006), which

is higher than the recommended 3 mm Al. Thus,

unmodified calcium silicate-based cement is not suit-

able as a root-end filling material or as a sealer as its

presence will not be detected easily on a radiograph. In

the present study, the cement with 20% replaced

bismuth oxide revealed a radiographic density equiva-

lent to 6.83 mm thickness of Al in accordance with

previous studies that reported that the cement with

20% bismuth oxide showed a radiopacity of 6.81 mm

Al (Kim et al. 2008) and 6.62 mm Al (Saliba et al.



2009). In the present study, the values reported for CSC

replaced with 20% bismuth oxide were not similar to

those obtained for ProRoot MTA (8.26 mm Al). The

radiopacity of MTA-Angelus in the present study was

similar to that reported by others (Tanomaru-Filho

et al. 2008) who reported the radiopacity of MTA-

Angelus to be equivalent to 3 mm Al, which is lower

than that reported for ProRoot MTA by other groups

(Torabinejad et al. 1995b, Laghios et al. 2000, Chng

et al. 2005, Danesh et al. 2006, Islam et al. 2006, Kim

et al. 2008) and also in the present study.

The variation in the results obtained in different

studies is likely because of differences in the cements

tested. Calcium silicate-based cement can be obtained

from a wide range of manufacturers, and thus results

cannot be compared. The full details of the types of

materials used are not always reported nor are the

water/cement ratios specified and other details that

affect the material properties. Particle size and shape

affect the water absorption of the material. Addition of

radiopacifiers composed of very fine particles causes an

increase in water uptake of the material as the specific

surface area is increased, and thus using the same

water/powder ratio the consistency of the resultant mix

can vary. Radiopacifiers that do not absorb water cause

a decrease in the water/cement ratio (Neville 1981).

High water/cement ratios have been reported to cause

a reduction in the radiopacity of the material (Coom-

araswamy et al. 2008).

Specimen size can also affect the resultant radiopac-

ity of the material. ISO 6876 (2002) suggests the use of

specimens 10 mm in diameter and 1 mm thick. Other

researchers (Laghios et al. 2000) have used thicker

specimens thus making comparisons between different

studies difficult. Variations may also arise from differ-

ences in the techniques used to evaluate radiopacity. In

the present study, a technique adopted by Watts &

McCabe (1999) was used to convert the optical density

to thickness of aluminium. The ISO 6876 (2002) does

not give any details in this regard. Other researchers

used linear regression (Laghios et al. 2000) or con-

verted the radiographs to digital images (Kim et al.

2008) and measured the grey pixel value. In most

publications, no details are given whether the base and

fog values were reduced in optical density calculations,

in fact, no technical detail is given on parameters used

(Chng et al. 2005, Islam et al. 2006).

The radiopacity values of a material are related to the

relative atomic mass of constituent elements. The

presence of elements with low relative atomic mass

like titanium and zinc caused low radiopacity values

and increase in the percentage of the material added to

the cement did not increase proportionally the radio-

pacity values of the cement. Materials containing

elements with a high relative atomic mass like bismuth

and gold exhibited high radiopacity values, which were

proportional to the increase in the quantity of material

added to the cement. Barium sulphate, although

having a high atomic number, conferred low radio-

pacity values to the cement. This is due to low levels of

barium element present in the material BaSO4, as

confirmed by EDX analysis.

Gold and silver/tin alloy could potentially be alter-

native radiopacifying materials for use with CSC.

Cement replaced with 25% gold and 20% silver/tin

alloy displayed radiopacity values comparable to Pro-

Root MTA but higher than MTA-Angelus. Thus, gold

and silver/tin alloy powders can both be used to replace

bismuth oxide in MTAs as these alternative materials

impart the necessary radiopacity to the resultant

cement. The use of gold powder could be prohibitive

because of the high cost of the material. In addition,

gold and silver/tin alloy impart a dark colour to the

material and can thus cause tattooing of the adjacent

tissues produced by corrosion products of silver from

the silver/tin alloy. Barium sulphate and zinc oxide

used in 25–30% replacement and lower loadings of

gold, silver/tin alloy and bismuth oxide exhibited

similar radiopacity values to MTA-Angelus. Different

radiopacifier materials and lower loadings can be used

in conjunction with CSC to achieve radiopacity values

greater than 3 mm Al, which is the value recom-

mended by ISO 6876 Section 7.8 (2002). Further

research is required to establish the optimal loading

and the effects that the radiopacifiers have on the other

properties and compatibilities of the calcium silicate-

based cement.

Conclusions

Silver/tin alloy and gold powder impart the necessary

radiopacity to calcium silicate-based cement. In addi-

tion, barium sulphate and zinc oxide represent suitable

radiopacifiers able to confer the radiodensity recom-

mended by ISO 6876. Further research is required to

investigate the properties of the calcium silicate-based

cement with the different radiopacifiers.

Acknowledgements

The University of Malta Research Fund Committee for

funding; Mr J. Sand Damtoft of Aalborg Cement Denmark



for providing the cement. Mr L. Spiteri of Heritage Malta

for his assistance with the electron microscopy; Mr R.

Mallett of the Biomaterials Department at King’s College

London Dental Institute at Guy’s, King’s and St Thomas’

Hospitals, London for his assistance with the particle size

distribution; Mr E. Grupetta for access to equipment and

Mr R. Spiteri and Ms G. Bonnici, radiographers at Mater

Dei and St Luke’s Hospitals Malta for their help with the

radiography of the samples.

References

Beyer-Olsen EM, Ørstavik D (1981) Radiopacity of root canal

sealers. Oral Surgery, Oral Medicine, and Oral Pathology 51,

320–8.

Bloodworth KE, Render PJ (1992) Dental acrylic resin

radiopacity: literature review and survey of practitioners’

attitudes. The Journal of Prosthetic Dentistry 67, 121–3.

British Standard Institution (2000) Cement-Part 1: Composition,

Specifications and Conformity Criteria for Common Cements.

BS EN 197–1. London: British Standard Institution.

Camilleri J (2007) Hydration mechanisms of mineral trioxide

aggregate. International Endodontic Journal 40, 462–70.

Camilleri J (2008a) The physical properties of accelerated

Portland cement for endodontic use. International Endodontic

Journal 41, 151–7.

Camilleri J (2008b) Characterization of hydration products of

mineral trioxide aggregate. International Endodontic Journal

41, 408–17.

Camilleri J, Montesin FE, Papaioannou S, McDonald F, Pitt

Ford TR (2004) Biocompatibility of two commercial forms of

mineral trioxide aggregate. International Endodontic Journal

37, 699–704.

Camilleri J, Montesin FE, Brady K, Sweeney R, Curtis RV, Pitt

Ford TR (2005) The constitution of mineral trioxide

aggregate. Dental Materials 21, 297–303.

Camilleri J, Montesin FE, Curtis RV, Pitt Ford TR (2006)

Characterization of Portland cement for use as a dental

restorative material. Dental Materials 22, 569–75.

Chng HK, Islam I, Yap AUJ, Tong YW, Koh ET (2005)

Properties of a new root-end filling material. Journal of


Chong BS, Pitt Ford TR, Hudson MB (2003) A prospective

clinical study of mineral trioxide aggregate and IRM when

used as root-end filling materials in endodontic surgery.


Coomaraswamy KS, Lumley PJ, Hofmann MP (2007) Effect of

bismuth oxide radioopacifier content on the material

properties of an endodontic Portland cement-based (MTA-

like) system. Journal of Endodontics 33, 295–8.

Coomaraswamy KS, Lumley PJ, Hofmann MP (2008) Effect of

Cement Paste Dilution on the Radiopacity of MTA, Abstract

0613. London: Pan European Federation for Dental

Research.

Danesh G, Dammaschke T, Gerth HU, Zandbiglari T, Schafer E

(2006) A comparative study of selected properties of

ProRoot mineral trioxide aggregate and two Portland

cements. International Endodontic Journal 39, 213–9.

International Standards Organization (2002) Dental Root Canal

Sealing Materials. ISO 6876 Section 7.8.

Islam I, Chng HK, Yap AU (2006) Comparison of the physical

and mechanical properties of MTA and Portland cement.


Kim EC, Lee BC, Chang HS, Lee W, Hong CU, Min KS (2008)

Evaluation of the radiopacity and cytotoxicity of Portland

cements containing bismuth oxide. Oral Surgery, Oral

Medicine, Oral Pathology, Oral Radiology, and Endodontics

105, 54–7.

Koulaouzidou EA, Economides N, Beltes P, Geromichalos G,

Papazisis K (2008) In vitro evaluation of the cytotoxicity of

ProRoot MTA and MTA Angelus. Journal of Oral Science 50,

397–402.

Laghios CD, Benson BW, Gutmann JL, Cutler CW (2000)

Comparative radiopacity of tetracalcium phosphate and

other root-end filling materials. International Endodontic

Journal 33, 311–5.

Lee SJ, Monsef M, Torabinejad M (1993) Sealing ability of a

mineral trioxide aggregate for repair of lateral root perfora-

tions. Journal of Endodontics 19, 541–4.

Min KS, Chang HS, Bae JM, Park SH, Hong CU, Kim EC (2007)

The Induction of heme oxygenase-1 modulates bismuth

oxide-induced cytotoxicity in human dental pulp cells.


Neville AM (1981) Properties of Concrete, 3rd edn. Essex:

Longman Scientific and Technical.

Oliveira MG, Xavier CB, Demarco FF, Pinheiro AL, Costa AT,

Pozza DH (2007) Comparative chemical study of MTA and

Portland cements. Brazilian Dental Journal 18, 3–7.

Ott DJ, Gelfand DW (1983) Gastrointestinal contrast agents.

Indications, uses, and risks. The Journal of the American

Medical Association 249, 2380–4.

Pitt Ford TR, Torabinejad M, McKendry DJ, Hong CU,

Kariyawasam SP (1995) Use of mineral trioxide aggregate

for repair of furcal perforations. Oral Surgery, Oral Medicine,

and Oral Pathology 79, 756–63.

Prentice LH, Tyas MJ, Burrow MF (2006) The effect of

ytterbium fluoride and barium sulphate nanoparticles on

the reactivity and strength of a glass-ionomer cement.

Dental Materials 22, 746–51.

Saliba E, Abassi Ghadi S, Vowles R, Camilleri J, Hooper S,

Camilleri J (2009) Evaluation of the strength and radio-

opacity of Portland cement with varying additions of

bismuth oxide. International Endodontic Journal 42, 322–8.

Saunders WP (2008) A prospective clinical study of perira-

dicular surgery using mineral trioxide aggregate as a root-

end filling. Journal of Endodontics 34, 660–5.

Song JS, Mante FK, Romanow WJ, Kim S (2006) Chemical

analysis of powder and set forms of Portland cement,

gray ProRoot MTA, white ProRoot MTA, and gray



MTA-Angelus. Oral Surgery, Oral Medicine, Oral Pathology,

Oral Radiology and Endodontics 102, 809–15.

Tanomaru-Filho M, da Silva GF, Duarte MA, Goncalves M,

Tanomaru JM (2008) Radiopacity evaluation of root-end

filling materials by digitization of images. Journal of Applied

Oral Science 16, 376–9.

Taylor HFW (1997) Cement Chemistry. London: Thomas Telford.

Torabinejad M, White DJ (1995) Tooth filling material and

use. US Patent Number 5769638.

Torabinejad M, Hong CU, Lee SJ, Monsef M, Pitt Ford TR

(1995a) Investigation of mineral trioxide aggregate for root-

end filling in dogs. Journal of Endodontics 21, 603–8.

Torabinejad M, Hong CU, McDonald F, Pitt Ford TR (1995b)

Physical and chemical properties of a new root-end filling

material. Journal of Endodontics 21, 349–53.

Torabinejad M, Pitt Ford TR, McKendry DJ, Abedi HR, Miller

DA, Kariyawasam SP (1997) Histologic assessment of

mineral trioxide aggregate as a root-end filling in monkeys.


Van Noort R (2002) Introduction to Dental Materials, 2nd edn.

London: Mosby.

Watts DC, McCabe JF (1999) Aluminium radiopacity stan-

dards for dentistry: an international survey. Journal of

Dentistry 27, 73–8.



The effect of sodium hypochlorite andethylenediaminetetraacetic acid irrigation,individually and in alternation, on toothsurface strain

R. Rajasingham1, Y.-L. Ng1, J. C. Knowles2 & K. Gulabivala1

1Unit of Endodontology, and 2Unit of Biomaterials Science, Divisions of Restorative Dental Sciences and Biomaterials Science and

Tissue-Engineering, UCL Eastman Dental Institute, University College London, London, UK

Abstract

Rajasingham R, Ng Y.-L, Knowles JC, Gulabivala K. The

effect of sodium hypochlorite and ethylenediaminetetraacetic

acid irrigation, individually and in alternation, on tooth surface

strain. International Endodontic Journal, 43, 31–40, 2010.

Aim To evaluate the effect of irrigation regimens on

tooth surface strain using saline, sodium hypochlorite

(3% and 5% NaOCl) and ethylenediaminetetraacetic

acid (17% EDTA), individually and in alternating

combinations.

Methodology Single-rooted premolar teeth with

single canals prepared to standardized dimensions

were grouped by anatomical features and randomly

distributed amongst six experimental groups (n = 12

each). The six groups were: (1) saline; (2) 5% NaOCl;

(3) 3% NaOCl; (4) 17% EDTA; (5) 3% NaOCl and

17% EDTA; (6) 5% NaOCl and 17% EDTA. All

groups underwent four (group 1) or five (groups 2, 3,

4, 5, 6) sequential 30-min irrigation periods follow-

ing each of which the tooth was subjected to a

standard regime of cyclic, nondestructive, occlusal

loading. Tooth surface strain was measured during

each loading cycle using electrical strain gauges

mounted cervico-proximally. The data were analysed

by Hierarchical anova and post hoc multiple com-

parisons.

Results Irrigation with 5% NaOCl alone or alternat-

ing with 17% EDTA significantly (P < 0.001) increased

the peak strain values for each of the irrigation periods

compared with that of saline (group 1). The data for the

other groups revealed no significant differences com-

pared with those of saline. The strain increase after the

fourth irrigation cycle was significantly higher for group

6 than for group 2. The measured canal morphology

and dentine thickness parameters did not prove to have

a significant effect on tooth surface strain.

Conclusions Irrigation with 5% NaOCl acting alone

or alternated with 17% EDTA (used in 30 min cycles),

significantly increased tooth surface strain. The alter-

nated regimen showed significantly greater changes in

tooth surface strain than NaOCl alone. Irrigation with

3% NaOCl and 17% EDTA individually or in combina-

tion did not significantly alter the tooth surface strain.

Keywords: ethylenediaminetetraacetic acid, irriga-

tion, sodium hypochlorite, tooth surface strain.

Received 28 January 2009; accepted 15 July 2009

Introduction

It is widely believed that root filled teeth are more

susceptible to fracture than teeth with vital pulps

(Rosen 1961, Johnson et al. 1976, Gher et al. 1987)

but conclusive evidence is lacking. Nevertheless, there

is circumstantial evidence for putative causes of non-

vital and root filled tooth fracture (Burke 1992). The

main causes could be: loss of tooth tissue, altered

physical properties of dentine, and altered propriocep-

tive/nociceptive properties (Gutmann 1992, Gulabivala

1995, Kinney et al. 2003). These factors probably

interact cumulatively to influence tooth loading, stress

distribution and, ultimately result in catastrophic

failure.

Correspondence: K. Gulabivala, Professor and Head of Endod-

ontology, UCL Eastman Dental Institute, 256 Gray’s Inn Road,

London WC1X 8LD, UK (Tel.: 020-7915-1033; fax: 020-

7915-2371; e-mail: [email protected]).

doi:10.1111/j.1365-2591.2009.01625.x


Tooth tissue loss reduces the force required to strain

and ultimately fracture teeth in vitro. The pattern of

tooth tissue loss influences the magnitude of the

induced strain (Mondelli et al. 1980, Larson et al.

1981, Panitvisai & Messer 1995, Lang et al. 2006);

clinical studies confirm these observations (Cavel et al.

1985, Hansen et al. 1990). The relative importance of

intact marginal ridge and the width/depth character-

istics of cavities are under debate, but original tooth

anatomy may also play a part (Khera et al. 1990).

Endodontic access cavities potentially weaken teeth

further (Reeh et al. 1989, Howe & McKendry 1990,

Panitvisai & Messer 1995) as does wide preparation of

canals (Hansen & Asmussen 1993, Lang et al. 2006).

The properties of dentine have been investigated

intensively and may be influenced by many factors

(Kinney et al. 2003, Kishen 2006). Factors investigated

include: changes in moisture content (Helfer et al.

1972, Huang et al. 1992, Jameson et al. 1994, Papa

et al. 1994, Kishen 2006), nature of collagen (Rivera

et al. 1988) and standard laboratory physical proper-

ties (Lewinstein & Grajower 1981, Carter et al. 1983,

Huang et al. 1992, Sedgley & Messer 1992). Unfortu-

nately, the findings have been contradictory or equiv-

ocal and no definitive proof of mechanical weakening of

dentine exists, except for the influence of water content.

The fundamental problem is that all ex vivo tests are by

definition on teeth without vital pulps and it is

debatable whether teeth can become significantly

dehydrated in the mouth.

Two studies (Loewenstein & Rathkamp 1955, Ran-

dow & Glantz 1986) suggest that pulp necrosis or loss

compromises the tooth’s proprioceptive/nociceptive

properties, predisposing to greater loading in function

with consequent increased likelihood of fracture (Ran-

dow & Glantz 1986).

In addition to these factors, there is increasing

evidence that intracanal irrigants, medicaments and

materials also influence the physical and mechanical

properties of dentine. The implicated materials include

sodium hypochlorite (NaOCl) (Grigoratos et al. 2001,

Sim et al. 2001, O’Driscoll et al. 2002, Oyarzun et al.

2002, Slutzky-Goldberg et al. 2004), hydrogen perox-

ide (Chng et al. 2002), MTAD (Machnick et al. 2003),

chloroform, xylene and halothane (Rotstein et al.

1999), calcium hydroxide (Grigoratos et al. 2001,

Andreasen et al. 2002, Rosenberg et al. 2007) and

eugenol (Biven et al. 1972).

Sodium hypochlorite and ethylenediaminetetraacetic

acid (EDTA) are widely used during root canal treat-

ment for established and sound biological reasons

(Bystrom & Sundqvist 1985). However, their potential

influence on the biomechanical properties of teeth and

dentine has only recently gained significant attention

(Yamada et al. 1983, Dogan & Calt 2001, Grigoratos

et al. 2001, Sim et al. 2001, Calt & Serper 2002). A

laboratory study by Sim et al. (2001) found that

irrigation with a 5.25% solution of NaOCl significantly

increased the tooth surface strain of dentine using

cyclical nondestructive loading in a whole tooth model.

The possible mechanisms involved in dentine weaken-

ing were shown to be due to the disintegration of the

organic element, leaving the mineral component intact

(O’Driscoll et al. 2002). Sim et al. (2001) found that

sequential, repeated 30-min irrigation steps with

5.25% NaOCl did not result in a linear increase in

tooth surface strain but one that plateaued after the

first two steps. This may be because the mineral

component of dentine is not depleted and consequently

poses a barrier to further NaOCl penetration. It may be

further hypothesized that the use of EDTA, an agent

that chelates the calcium may deplete the inorganic

component of dentine and expose more of the organic

structure for further depletion. Therefore, alternate

irrigation with NaOCl and EDTA (Yamada et al.1983,

Marending et al. 2007) may eliminate the ‘plateauing’

tendency evident when irrigating solely with NaOCl

(Dogan & Calt 2001, Calt & Serper 2002, Yoshioka

et al. 2002).

The aims of this study were to evaluate the effect of

irrigation on tooth surface strain when using saline,

NaOCl (3%, 5%) and EDTA (17%) individually, and in

alternate combination as follows; irrigation with NaOCl

(either 3% or 5%) followed by EDTA (17%).


Preparation of teeth and allocation to experimental

groups and anatomical sub-groups

Human extracted teeth were obtained with consent

from patients undergoing routine extractions for ortho-

dontic or restorative reasons. Seventy-two single-rooted

premolars with single root canals, all of which were

noncarious and crack-free (confirmed by transillumi-

nation under a microscope), were used after temporary

storage in 4% formal-saline (Lam & Gulabivala 1996)

immediately following extraction. Gross debris on the

external surfaces of the teeth was removed with a sharp

scalpel. The crowns of the teeth were removed 4 mm

coronal to the cemento-enamel junction with a dia-

mond bur in a high-speed hand-piece leaving a

Effect of irrigants on tooth strain Rajasingham et al.


flattened surface that was perpendicular to the long

axis of the tooth. The remaining enamel was removed

with a high-speed hand-piece (Sim et al. 2001).

The access cavity size and shape were dictated by the

form of the pulp chamber and was not standardized.

Access to the canals was standardized with Profile�

orifice shapers (Maillefer Instruments, Ballaigues, Swit-

zerland), the pulps were extirpated with a Hedstrom�

file (Kerr UK, Peterborough, UK) and the canals were

prepared to a standard maximum apical size (30) and

taper (0.06), where preoperative canal size allowed

engagement of dentine; no attempt was made to gauge

canal diameter and enlarge further by a standard

number of instruments. Each canal was prepared to the

apical foramen, the position of which was determined

by the emergence of a size 10 file through it. A crown-

down sequence of Profile� nickel–titanium endodontic

instruments (Maillefer Instruments) was used with

saline as the irrigant. The teeth were maintained in a

hydrated state within saline-soaked tissue paper during

all procedures. Following canal preparation, the apices

were sealed with two coats of nail varnish (Boots No.17

Clear Nail Varnish, Nottingham, England).

Periapical radiographs in two planes (bucco-lingual

and mesio-distal) were taken for all teeth. The width of

the dentine at the cemento-enamel junction on the two

radiographic views was measured with digital callipers

and the average obtained. The teeth were grouped into

like types by anatomical features; the main criterion

was the apico-coronal position at which the root canal

started to narrow as follows: (i) immature apex or no

canal narrowing; (ii) coronal 1/3; (iii) middle 1/3; and

(iv) apical 1/3. The middle and coronal groups were

further partitioned by root length. This resulted in six

anatomically distinct groups with 12 teeth in each

group; teeth from each anatomical group were ran-

domly assigned amongst six experimental irrigation

groups to give a sample size of 12 per group.

Mounting of teeth and bonding of strain gauges

Clear acrylic resin (Specifix-20, Struers Epoxy resins;

Struers A/S, Copenhagen, Denmark), mixed according to

the manufacturer’s instructions was used to secure each

tooth centrally within circular plastic moulds (2.5 cm

high). The long axes of the teeth were strictly aligned

parallel to the walls of the plastic moulds and 2–3 mm of

the root was left exposed below the cemento-enamel

junction, simulating the level of the alveolar bone.

Constantan strain gauges (gauge factor = 2.05;

Measurements Group UK Ltd, Basingstoke, UK) with

short attached copper leads (resistance = 120 X; type

EA-06-062AP-120, option LE) were used. The gauge

polyimide backing was trimmed to an 1-mm border to

enable easier positioning on the tooth (Fig. 1). The

proximal bonding site on each tooth was prepared by

rinsing with water, drying with air (3-in-1 syringe) and

application of a thin layer of cyanoacrylate adhesive

(M-Bond 200 Adhesive, Measurements Group UK Ltd).

The gauge was positioned with the top edge of the

backing 1 mm below the occlusal surface and vertically

aligned along the long axis of the tooth, using an

engineering T-square. Gentle but firm pressure was

applied until the adhesive had set. The strain gauge and

exposed wire leads were protected with M-Coat D air-

drying acrylic varnish (Measurements Group UK Ltd).

The teeth were wrapped in damp gauze and placed in a

sealed polythene bag for storage until use. Every effort

was made to prevent dehydration of the teeth during

testing using damp gauze. Shortly before use, a length

Straingauge

Instron grip

Premolar with flattened coronal face

Point at whichload is applied

Figure 1 Experimental set-up of tooth,

acrylic holder and strain gauge.

Rajasingham et al. Effect of irrigants on tooth strain


of shielded twin cable (RS Components Ltd, Corby, UK)

was soldered to the copper leads and the resistance of

the gauge checked with an Ohm-meter to ensure

absence of component burnt out.

Preparation of strain gauge circuit

The ‘active’ strain gauge on the tooth was incorporated

as one arm of a ‘quarter bridge’ Wheatstone bridge

circuit; two arms consisted of a precision wirewound

1 kX resistor (RS Components Ltd) and the fourth arm

incorporated a ‘dummy’ strain gauge bonded to a tooth

for temperature compensation. A 100 kX precision

wirewound resistor (RS Components Ltd) with a switch

in parallel with the active gauge, served to calibrate the

strain gauge circuit. Shielded cable was used to

minimize electro-magnetic interference from proximal

equipment. The excitation voltage to the Wheatstone

bridge circuit was supplied by an RDP Transducer

Indicator E308 (RDP Electronics Ltd, Wolverhampton,

UK) and was set at 1.45V continuously for the duration

of the experimental period. This also amplified and

displayed the output voltage from the circuit and

allowed zeroing prior to loading. The strain-induced

voltage could then be measured relative to this base-

line. The output from the Wheatstone bridge circuit

was also logged via the transducer by a computer. The

software used (Waveview for Dos 1.24; Eagle Appli-

ances Pty Ltd, Brighton, UK) was set to sample the

output at intervals of 0.1 s.

Preparation of experimental solutions

The physiologic saline was commercially obtained

(Baxter Healthcare Ltd, Thetford, UK), whilst the 3%

and 5% solutions of NaOCl were diluted from 12%

NaOCl stock (BDH Laboratory Supplies, Poole, UK) and

the concentrations verified by iodometric titration. A

17% solution of EDTA (BDH Laboratory Supplies) was

obtained by adjusting the appropriate mass in water at

pH 7.8. Freshly made solutions were stored in opaque

bottles under controlled room temperature until use

but never longer than 2 weeks.

Irrigation regimen per experimental group

The teeth in groups 1–6 were exposed to four or five

sequential 30-min standardized irrigation cycles

(Table 1). Each 30-min irrigation cycle consisted of

delivering 10 mL of test solution as follows: (i) irriga-

tion with 3 mL over 1 min followed by agitation with

size 25 Flexo-file (Maillefer Instruments) for 10 s and

leaving undisturbed for 50 s; (ii) irrigation with 0.5 mL

over 20 s, followed by agitation for 10 s, and then

leaving undisturbed for 90 s; (iii) cycle 2 was repeated

13 more times. At the end of this period, the canals

were flushed with 9 mL saline over 3 min. In total

therefore, each irrigation cycle consisted of 33 min

(30 min test solution and 3 min saline).

Cyclic loading

Following each 30-min irrigation period, each tooth

was subjected to a standard regime of cyclic nonde-

structive occlusal loading (Goldsmith et al. 2002).

Tooth surface strain was measured during each loading

cycle. To perform the loading, the acrylic block

containing the test tooth was secured in a brass

receptacle with four restraining screws. The receptacle

was clamped via an attachment to the crosshead of a

Universal loading machine (Instron Ltd, High

Wycombe, UK). A ball bearing of 5.0 mm diameter

fixed to the end of the loading arm (with 1kN load cell)

delivered the load accurately to the centre of the tooth

access preparation along its long axis, by adjustment of

the receptacle position. Once positioned, the tooth was

not moved during the entire testing procedure; access

for irrigation procedures being achieved by lowering

the crosshead to leave 5 cm between the end of the

loading arm and the tooth.

Each loading cycle consisted of 3 stages: (1) Loading

from 0 N to 20 N (the pre-load); (2) Five cycles of

loading from 20 N up to 110 N and unloading down to

20 N at a crosshead speed of 0.7 mm/s (usually

completed over 2–3 mins); and (3) Unloading from

20 N to 0 N. The pre-load was set at the start of each

cycle through the Instron console, whilst the loading

regime was controlled by the Instron Series XII

software (Instron Ltd).

Strain measurement

At completion of the first irrigation cycle, the bridge

was balanced to zero and output logging was com-

menced. After 15–20 s, the calibration resistor was

activated for 20 s and the zero dial adjusted to record a

second baseline value before the loading procedure was

commenced; the baselines facilitated later calculations.

The data sampled from the output voltage were saved

electronically at the end of the loading period. The

irrigation and loading procedures were repeated four

(Group 1) or five (Groups 2, 3, 4, 5, 6) times for each



tooth, generating four or five data files per tooth. The

entire experimental period was completed in a single

sitting for each tooth.

Five peak outputs were obtained from the recorded

data for each loading period and a mean value

calculated (peak strain value). The data were analysed

using the STATA software programme (stata version 9

2005; STATA Corporation, College Station, USA). The

change in the mean strain values of teeth following

each irrigation period from the baseline were calculated

for each experimental group and analysed using the

hierarchical anova analysis followed by the Bonferroni

analysis.

Results

The mean peak strain values of the teeth after each

consecutive irrigation period are summarized by each

experimental group in Table 1. In group 1 (negative-

control saline), there was little change in these values

following each consecutive irrigation period. In groups

2 (positive control 5% NaOCl) and 6 (5% NaOCl/17%

EDTA), there was marked increase in strain values

following the second and third irrigation periods, this

increase plateaued at the fourth irrigation period for

teeth without the alternate irrigation with 17% EDTA

(Group 2). In groups 3 (3% NaOCl) and 5 (3% NaOCl/

17% EDTA), there was a proportionately smaller

increase in the strain values following the second and

the third irrigation periods, but again this increase

plateaued for the teeth without the alternate irrigation

with 17% EDTA (Group 3). In group 4 (17% EDTA),

there was a continuous but small increase in the strain

values following each irrigation period. For test groups

2–4, there was no significant difference between the

mean peak strain values obtained after the fourth (test

solution) and fifth (saline) irrigation cycles. In contrast,

the difference was significant (P = 0.002) for the teeth

in group 6, although the magnitude was small (4.8 le;

95% CI 2.8 le, 6.7 le) (Table 1). Therefore the values

obtained following the fifth irrigation cycle were not

considered when comparing the changes in mean peak

strain.

The difference in the mean peak strain values for teeth

after each irrigation period and the baseline are plotted

by each experimental group in Fig. 2. The differences in

the changes in the mean peak strain values were found

to be normally distributed. Hierarchical anova analysis

showed that there was a significant interaction

between test groups and irrigation period (P < 0.001)

with significant differences between the test groups

(P < 0.001). This indicated that the rates of change of

Table 1 Mean peak strain values

[microstrains (le)] and standard devia-

tions (n = 12) for each group and irri-

gation cycle depicted by time and cycle

number. The irrigant used at each

30 min cycle is also shown

Time (min)

1 Baseline 2 3 4 5

30 60 90 120 150

Group 1

Irrigant Saline Saline Saline Saline

Mean 507.6 520 510.5 515.4

SD 343.4 359.9 337.8 344.8

Group 2

Irrigant Saline 5%NaOCl 5%NaOCl 5%NaOCl Saline

Mean 319.0 404.8 450.6 431.1 427.6

SD 93.1 125.6 144.9 162.4 162.5

Group 3

Irrigant Saline 3%NaOCl 3%NaOCl 3%NaOCl Saline

Mean 259.0 273.8 294.2 284.7 273.1

SD 136. 8 141.4 142.2 132.3 137.5

Group 4

Irrigant Saline 17%EDTA 17%EDTA 17%EDTA Saline

Mean 229.3 234.7 241.2 245.6 247.2

SD 140.3 138.6 137.0 135.8 136.4

Group 5

Irrigant Saline 3%NaOCl 17%EDTA 3%NaOCl Saline

Mean 370.0 386.3 396.3 409.3 409.0

SD 233. 9 234.9 235.5 235.9 238.8

Group 6

Irrigant Saline 5%NaOCl 17%EDTA 5%NaOCl Saline

Mean 231.9 317.3 355.7 434.5 439.3

SD 129.2 176.5 190.3 246.9 247.7



tooth surface strain between groups were significantly

different. In contrast, the irrigation cycle (P = 0.3),

average dentine thickness (P = 0.8) and canal mor-

phology (P = 0.5) had no significant influence on the

changes in the strain values. Multiple comparisons

between test groups for each period were subsequently

carried out using the Bonferroni test. Following the

second, third and fourth irrigation periods, the changes

in mean peak strain values of teeth in groups 2 (5%

NaOCl) and 6 (5% NaOCl/17% EDTA) were signifi-

cantly (P < 0.001) higher than those in group 1

(negative-control saline) and other experimental

groups. There were no significant differences

(P = 1.0) in the changes between teeth in group 1

(saline) and those in groups 3 (3% NaOCl), 4 (17%

EDTA) and 5 (3% NaOCl/17% EDTA). Following the

second and third irrigation periods, the differences in

the change of mean peak strain values between groups

2 and 6 (P = 1.0) and between groups 3 and 5

(P = 1.0) were not significant. However, the change

was significantly (P = 0.02) higher in group 6 com-

pared with group 2 following the fourth irrigation

period.

Discussion

Previous work (Goldsmith et al. 2002, Sim et al. 2001)

had suggested that root canal irrigation with suffi-

ciently concentrated NaOCl could alter tooth surface

strain, potentially predisposing teeth to fracture. This

effect was attributed to depletion of the organic content

of the root canal surface dentine (O’Driscoll et al.

2002). It has been suggested that alternation of

irrigation with 17% EDTA could facilitate greater

bacterial load reduction (Bystrom & Sundqvist 1985),

removal of the smear layer and bacterial biofilm from

noninstrumented surfaces (Gulabivala et al. 2005). The

potential disadvantage of this combination may be that

alternate depletion of organic and inorganic substrate

may allow the tooth to be weakened further as

demonstrated by an increase in tooth surface strain.

To test this hypothesis, the previous work was repeated

with a relevant study design and stringent attention to

detail in order to overcome the data variation seen in

the experiments of Goldsmith et al. (2002).

The 72 teeth used in this study were single-rooted

premolars with single root canals. The teeth were

grouped anatomically according to the location of root

canal narrowing and root length. Sim et al. (2001) had

standardized their sample by using mandibular second

premolars with single mature roots, whilst Goldsmith

et al. (2002) used a mixture of single and two-rooted

premolars with mature or immature apices. As would

be expected, the latter study showed greater variability

of the data. It is difficult to ‘quantify’ tooth anatomy but

the adopted method indirectly described the dentine

distribution within the tooth. The potential confound-

ing effect of tooth anatomy was reduced by stratified

randomized allocation of teeth and accounted for in the

hierarchical anova analysis by incorporating the

average thickness of dentine at the cemento-enamel

junction as well as canal morphology as co-variates.

The remaining enamel from the coronal 4 mm

of the prepared teeth was removed to show the effect

of the irrigant on dentine, since it has been shown that

the band of enamel around the cervical dentine has a

significant effect on tooth stiffness (Meredith 1992, Sim

et al. 2001). Equally, this measure should be taken into

Change in strain from first irrigation

–50

0

50

100

150

200

250

300

2 3 4 5 6

Irrigation period

Str

ain

(u

nit

s)Group 1Group 2Group 3Group 4Group 5Group 6

Figure 2 Graph depicting change in

mean peak strain values (microstrains

[le]) from baseline, with confidence

interval lines (colour coded); the obvi-

ously different groups are 2 and 6.



account when considering the clinical implications of

the results. The root canals were prepared to a

standardized taper to allow the irrigating needle to

penetrate to a realistic depth towards the apices; the

precise method of preparation was not considered

important (Sim et al. – balanced force with hand files;

Goldsmith et al. – Quantec� 2000 instruments), as

long as stresses were not induced in the residual

dentine.

A clear acrylic resin (epoxy) was used for mounting

in this study because it showed minimal dimensional

change after setting compared to self-curing acrylic

resin used by Goldsmith et al. (2002).

The single-element strain gauge used by Goldsmith

et al. (2002) was again chosen for this study in

contrast to the rosette gauge used by Sim et al.

(2001) because the latter found that the main axial

loading stress was compressive and was mainly

recorded by the gauge element orientated in the

longitudinal axis. The location of the single-element

strain gauge was determined by the outcome of full

field stress pattern analysis (SPATE), which showed the

chosen site to be a zone of stress concentration during

loading of posterior teeth (Meredith 1992). Other

studies also concur regarding the distribution of strains

within loaded teeth (Asundi & Kishen 2000, Palamara

et al. 2002).

The choice of solutions and their concentrations were

informed by clinical usage and prior work (Grigoratos

et al. 2001, Sim et al. 2001, Goldsmith et al. 2002). A

9 mL saline final flush immediately prior to occlusal

loading, as per Sim et al. (2001) was repeated here

because Goldsmith’s work was suggestive of possible

tooth dehydration during NaOCl irrigation. Figure 2 did

not indicate significant alteration due to the final saline

irrigation phase but minor fluctuations during record-

ing confirmed the probability of some minor individual

tooth-related effect due to ‘dehydration’.

The maximum load applied (110 N) and the rate of

loading and unloading was the same as that used by

Goldsmith et al. (2002). The actual time taken to

complete a loading cycle differed between sample teeth

(range: 20–30 s) but remained constant for individual

teeth at each of the five loading periods. The range of

20 N to 110 N is within ‘physiological limits’ for the

human dentition (De Boever et al. 1978).

Values of peak strain reported in this study were

slightly higher than those found by Sim et al. (2001),

but were of the same order of magnitude as those

reported by Meredith (1992) and Goldsmith et al.

(2002). The wide range of baseline strain values seen

in all experimental groups can be accounted for at least

partly by the position of the loading arm on the

flattened surface of the resected crown. Once the tooth

position was fixed in the Universal loading machine, it

remained constant throughout the test, allowing com-

parison between irrigation cycles and for each tooth to

act as its own control. The magnitude of strain

recorded at the cervical target site was dependent on

loading position. Other factors expected to influence

strain might be tooth anatomy and dentine thickness.

Although the precise thickness of coronal dentine was

not measured after tooth preparation, one would expect

higher strain values to be found in teeth with the

thinnest walls (Goldsmith et al. 2002). In the present

study, neither of the tooth anatomy parameters (aver-

age dentine thickness at the cemento-enamel junction

and canal morphology) were found to have a signifi-

cant influence on the change in tooth surface strain.

This negative finding may be true or could possibly be

attributed to the crude and possibly inaccurate estimate

of dentine thickness obtained from radiographic

images. It may also be hypothesized that whilst there

may be straining of the dentine at the point of loading,

the transmission of this strain through the entire

thickness of dentine may be limited at certain surfaces

(given the complex nature of resolution of tensile and

compressive stresses and strains), resulting in absence

of detection at the point of measurement (Lertichira-

karn et al. 2003).

The potential issue of recovery time between cycles

was addressed by Sim et al. (2001) and data from this

study confirmed that on the whole, the possibility of

accumulative strain was not a problem. Occasional

unexpected changes in strain in individual samples

were attributed to micro-fractures propagating with

each loading cycle or errors in strain gauge positioning.

It was evident that the method was sufficiently sensitive

and transparent to detect such aberrant readings.

The effect of 5% NaOCl irrigation had yielded

contrasting results in the two previous reports (Sim

et al. 2001, Goldsmith et al. 2002), the latter of which

had considerable variation in their data. The present

study clearly confirmed that the findings of Sim et al.

(2001) were reliable and that consistent patterns may

only be revealed with precise attention to detail in the

experimental set-up. The key steps of influence were:

(i) alignment of the strain gauges to the long axis of the

tooth; (ii) flattening of the occlusal surface of the tooth

to allow true perpendicular loading; and (iii) mounting

of the tooth in a resin with minimal dimensional

changes. All 12 teeth showed an increase in peak strain



values after irrigation with 5% NaOCl followed in the

majority by a plateauing of strain increase.

Alternate irrigation with 5% NaOCl and 17% EDTA

resulted in increases in tooth surface strain that were

highly significantly different from the saline control; the

plateauing of strain increase (evident with irrigation

with 5% NaOCl alone) was eliminated as predicted by

the test hypothesis. The results therefore support the

hypothesis that alternate irrigation with NaOCl (5%)

and EDTA (17%) probably allows the alternate deple-

tion of organic and inorganic material, with a greater

accumulative depth of effect on dentine and therefore

tooth surface strain. Teeth undergoing root canal

treatment, already compromised by loss of tooth

structure, particularly with breaks in the continuity

of the band of circumferential enamel, could be further

weakened by 5% NaOCl irrigation acting alone but

particularly when alternated with 17% EDTA. The

increase in strain, although significant, does not yet

indicate whether it is sufficient to result in increased

risk of tooth fracture due to fatigue. It could be

hypothesized that since such irrigation patterns pro-

duce surface flaws in dentine (Calt & Serper 2002),

cyclic loading at the normal stresses of mastication may

allow fatigue crack growth to catastrophic proportions

(Kinney et al. 2003, Kishen 2006).

Irrigation with 3% NaOCl and 17% EDTA individu-

ally or in combination showed similar trends as for the

higher concentration sodium hypochlorite solution

(Fig. 2) but the difference was not significant and the

groups were comparable to the saline control. The

negligible changes in strain from baseline values for all

teeth irrigated with 17% EDTA alone implies the effect

on inorganic component of dentine is confined enough

to not affect tooth surface strain significantly. Alternate

irrigation with 3% NaOCl and 17% EDTA produced

small but noticeable increases in strain values between

the third (17% EDTA) and fourth (3% NaOCl) irrigation

cycles in some teeth. The duration, concentration and

irrigant combination appear to be critical as shown in

other in vitro work (Marending et al. 2007). The slight

reduction in peak strain values for some teeth, follow-

ing the final saline irrigation could be explained by

rehydration of the dentine, as anticipated from Gold-

smith et al. (2002).

The irrigation regime as used during the cyclic

nondestructive loading phase may not be directly

comparable to the clinical practice of all practitioners;

nevertheless it was well defined and provided key

insight into the potential clinical effects. In the clinical

scenario, the irrigant would have to contend with more

organic tissue in the root canal system and would

therefore become spent more rapidly. In this study,

pulpal tissue was removed during the canal preparation

stage, prior to irrigation with NaOCl and EDTA. The

intention was to test the worst case scenario to find any

differences, if they existed.

Irrigation with 3% NaOCl on its own, despite

showing some increase in strain values, was not

significantly different from irrigation with saline and

may be a ‘safer’ concentration to use. Its antimicro-

bial and tissue-dissolving properties should be ade-

quate (Bystrom & Sundqvist 1985, Baumgartner &

Cuenin 1992). It would be prudent to select a

suitable lower concentration of NaOCl that would

have minimal undesirable effects on the physical

properties of dentine.

Likewise, the use of EDTA on its own did not result in

any significant difference from irrigation with saline.

The experiment provided no evidence for depletion of

structural components that could induce alteration in

the mechanical properties of the teeth.

Conclusions

This study showed that within the confines of its

design, irrigation with 5% NaOCl alone but especially

when alternated with 17% EDTA for sufficient duration

may significantly increase tooth surface strain.

References

Andreasen JO, Farik B, Munksgaard EC (2002) Long-term

calcium hydroxide as a root canal dressing may increase

risk of root fracture. Dental Traumatology 18, 134–7.

Asundi A, Kishen A (2000) A strain gauge and photoelastic

analysis of in vivo strain and in vitro stress distribution in

human dental supporting structures. Archives of Oral Biology

45, 543–50.

Baumgartner JC, Cuenin PR (1992) Efficacy of several

concentrations of sodium hypochlorite for root canal

irrigation. Journal of Endodontics 18, 605–12.

Biven GM, Bapna MS, Heuer MA (1972) Effect of eugenol

and eugenol-containing root canal sealers on the microh-

ardness of human dentine. Journal of Dental Research 51,

1602–9.

Burke FJT (1992) Tooth fracture in vivo and in vitro. Journal of


Bystrom A, Sundqvist G (1985) The antibacterial action of

sodium hypochlorite and EDTA in 60 cases of endodontic

therapy. International Endodontic Journal 18, 35–40.

Calt S, Serper A (2002) Time dependent effects of EDTA on

dentine structures. Journal of Endodontics 28, 17–9.



Carter JM, Sorensen SE, Johnson RR, Teitelbaum RL, Levine MS

(1983) Punch shear testing of extracted vital and endodon-

tically treated teeth. Journal of Biomechanics 16, 841–8.

Cavel WT, Kelsey WP, Blankenhall RJ (1985) An in vivo study

of cuspal fracture. Journal of Prosthetic Dentistry 53, 38–41.

Chng HK, Palamara EA, Messer HH (2002) Effect of hydrogen

peroxide on biomechanical properties of human dentine.


De Boever JA, McCall WD, Holdens S, Ash MM (1978)

Functional occlusal forces : an investigation by telemetry.

Journal of Prosthetic Dentistry 40, 326–33.

Dogan H, Calt S (2001) Effects of chelating agents and sodium

hypochlorite on mineral content of root dentin. Journal of


Gher ME, Dunlap RM, Anderson MH, Kuhl LV (1987) Clinical

survey of fractured teeth. Journal of American Dental Asso-

ciation 114, 174–7.

Goldsmith M, Gulabivala K, Knowles J (2002) The effect of

sodium hypochlorite irrigant concentration on tooth surface

strain. Journal of Endodontics 28, 575–9.

Grigoratos D, Knowles J, Ng Y-L, Gulabivala K (2001) Effect of

exposing dentine to sodium hypochlorite and calcium

hydroxide on its flexural strength and elastic modulus.


Gulabivala K (1995) Chapter 15 – Restoration of the root-filled

tooth. In: Stock CJR, Gulabivala K, Walker RT, Goodman JR,

eds. Color Atlas and Text of Endodontics, 2nd edn. London:

Mosby-Wolfe, 241–72.

Gulabivala K, Patel B, Evans G, Ng Y-L (2005) Effects of

mechanical and chemical procedures on root canal surfaces.

Endodontic Topics 10, 103–22.

Gutmann JL (1992) The dentine-root complex: anatomic and

biologic considerations in restoring endodontically treated

teeth. Journal of Prosthetic Dentistry 67, 458–67.

Hansen EK, Asmussen E (1993) Cusp fracture of endodonti-

cally treated posterior teeth restored with amalgam – Teeth

restored in Denmark before 1975 versus after 1979. Acta

Odontologica Scandinavica. 51, 73–7.

Hansen EK, Asmussen E, Christiansen NC (1990) In vivo

fractures of endodontically treated teeth restored with

amalgam. Endondontics and Dental Traumatology 6, 49–55.

Helfer AR, Melnick S, Schilder H (1972) Determination of the

moisture content of vital and pulpless teeth. Oral Surgery,

Oral Medicine, Oral Pathology 34, 661–70.

Howe CA, McKendry DJ (1990) Effect of endodontic access

preparation on resistance to crown-root fracture. Journal of

the American Dental Association 121, 712–5.

Huang TG, Schilder H, Nathanson D (1992) Effects of moisture

content and endodontic treatment on some mechanical

properties of human dentine. Journal of Endodontics 18,

209–15.

Jameson MW, Tidmarsh BG, Hood JAA (1994) Effect of storage

media on subsequent water loss and regain by human and

bovine dentine and on mechanical properties of human

dentine in vitro. Archives of Oral Biology 39, 759–67.

Johnson JK, Schwartz NL, Blackwell RT (1976) Evaluation &

restoration of endodontic treated posterior teeth. Journal of

the American Dental Association 93, 597–605.

Khera SC, Carpenter CW, Vetter JD, Staley RN (1990)

Anatomy of cusps of posterior teeth and their fracture

potential. Journal of Prosthetic Dentistry 64, 139–47.

Kinney JH, Marshall SJ, Marshall GW (2003) The mechanical

properties of human dentin: a critical review and re-

evaluation of the dental literature. Critical Reviews in Oral

Biology & Medicine 14, 13–29.

Kishen A (2006) Mechanisms and risk factors for fracture

predilection in endodontically treated teeth. Endodontic

Topics 13, 57–83.

Lam MCJA, Gulabivala K (1996) Influence of storage condi-

tions of extracted teeth on dentine removal by a standar-

dised method of filing. Endodontics and Dental Traumatology

12, 25–32.

Lang H, Korkmaz Y, Schneider K, Raab WHM (2006) Impact

of endodontic treatments on the rigidity of the root. Journal

of Dental Research 85, 364–8.

Larson TD, Douglas WH, Geistfeld RE (1981) Effect of prepared

cavities on the strength of teeth. Operative Dentistry 6, 2–5.

Lertichirakarn V, Palamara JEA, Messer HH (2003) Finite

element analysis and strain-gauge studies of vertical root

fracture. Journal of Endodontics 29, 529–34.

Lewinstein I, Grajower R (1981) Root dentine hardness of

endodontically treated teeth. Journal of Endodontics 7, 421–2.

Loewenstein NR, Rathkamp R (1955) A study on the presso-

receptive sensibility of the tooth. Journal of Dental Research

34, 287–94.

Machnick TK, Torabinejad M, Munoz CA, Shabahang S

(2003) Effect of MTAD on flexural strength and modulus

of elasticity of dentin. Journal of Endodontics 29, 747–50.

Marending M, Paque F, Fischer J, Zehnder M (2007) Impact of

irrigant sequence on mechanical properties of human root

dentin. Journal of Endodontics 33, 1325–8.

Meredith N (1992) An in-vitro analysis of the stresses in natural and

restored human teeth. PhD Thesis, University of London.

Mondelli J, Steagall I, Ishikiriama A, Navarro MF, Soares FB

(1980) Fracture strength of human teeth with cavity

preparations. Journal of Prosthetic Dentistry 43, 419–22.

O’Driscoll CO, Dowker S, Anderson P, Wilson R, Gulabivala K

(2002) Effects of sodium hypochlorite solution on root

dentine composition. Journal of Materials Science: Materials in

Medicine 13, 219–23.

Oyarzun A, Cordero AM, Whittle M (2002) Immunohisto-

chemical evaluation of the effects of sodium hypochlorite on

dentin collagen and glycosaminoglycans. Journal of End-

odontics 28, 152–6.

Palamara JEA, Palamara D, Messer HH (2002) Strains in the

mariginal ridge during occlusal loading. Australian Dental

Journal 47, 218–22.

Panitvisai P, Messer HH (1995) Cuspal deflection in molars in

relation to endodontic and restorative procedures. Journal of




Papa J, Cain C, Messer HH (1994) Moisture content of vital vs

endodontically treated teeth. Endodontics and Dental Trau-

matology 10, 91–3.

Randow K, Glantz PO (1986) On cantilever loading of vital

and non-vital teeth – An experimental clinical study. Acta

Odontologica Scandinavica 44, 271–7.

Reeh ES, Messer HH, Douglas WH (1989) Reduction in tooth

stiffness as a result of endodontic & restorative procedures.


Rivera E, Yamaguchi G, Chandler G, Bergenholtz G (1988)

Dentine collagen cross-links of root filled and normal teeth.

Journal of Endodontics 14, 195. (Abstract 13).

Rosen H (1961) Operative procedures on mutilated endodonti-

cally treated teeth. Journal of Prosthetic Dentistry 11, 973–86.

Rosenberg B, Murray PE, Namerow K (2007) The effect of

calcium hydroxide root filling on dentin fracture strength.

Dental Traumatology 23, 26–9.

Rotstein I, Cohenca N, Teperovich E et al. (1999) Effect of

chloroform, xylene, halothane on enamel and dentine

microhardness of human teeth. Oral Surgery, Oral Medicine,

Oral Pathology 87, 366–8.

Sedgley CM, Messer HH (1992) Are endodontically treated

teeth more brittle? Journal of Endodontics 18, 332–5.

Sim TP, Knowles J, Ng YL, Shelton J, Gulabivala K (2001)

Effect of sodium hypochlorite irrigant concentration on

strain in teeth. International Endodontic Journal 34, 120–

32.

Slutzky-Goldberg I, Maree M, Liberman R, Heling I (2004)

Effect of sodium hypochlorite on dentin microhardness.


Yamada RS, Armas A, Goldman M, Sun LinP (1983) A

scanning electron microscopic comparison of high volume

final flush with several irrigating solutions: part 3. Journal of


Yoshioka WN, Kobayashi C, Suda H (2002) A scanning

electron microscopic study of dentinal erosion by final

irrigation with EDTA and NaOCl solutions. International

Journal of Endodontics. 35, 934–9.



Effect of restorations on pulpal blood flow in molarsmeasured by laser Doppler flowmetry

N. P. Chandler1, T. R. Pitt Ford2 & B. D. Monteith1

1Department of Oral Rehabilitation, School of Dentistry, University of Otago, Dunedin, New Zealand; and 2Department of

Conservative Dentistry, Dental Institute, King’s College London, London, UK

Abstract

Chandler NP, Pitt Ford TR, Monteith BD. Effect of resto-

rations on pulpal blood flow in molars measured by laser

Doppler flowmetry. International Endodontic Journal, 43, 41–46,

2010.

Aim To: (i) compare laser Doppler pulpal blood flow

(PBF) signals from restored and unrestored first molar

teeth, (ii) investigate PBF in teeth with large and small

restorations, and (iii) to relate PBF to pulp chamber

dimensions on radiographs.

Methodology Bitewing radiographs of young adults

with restored first molars were obtained and pulp

chamber dimensions measured. Subjects were divided

into 2 groups: group A with a restored tooth and an

unrestored contralateral (43 subjects) and group B,

those with a molar with a small (usually occlusal)

restoration whilst the contralateral tooth had an

extensive occlusal restoration (or restorations) or

restored proximal surface(s) and/or cuspal overlay

(31 subjects). The 148 teeth responded to electric pulp

testing, and their PBF was recorded using a laser

Doppler flowmeter. Data were analysed using Student’s

t-test.

Results In group A the PBF in the restored teeth was

significantly lower than in unrestored contralaterals

(P = 0.028) and the total pulp chamber area and that

in the clinical crown were smaller (P = 0.039 and

0.021 respectively). The group B molars with large

restorations had significantly lower PBF than contra-

laterals with small restorations (P = 0.001), and their

total pulp chamber area and pulp chamber width at

cervix were reduced significantly (P = 0.003 and

0.032 respectively).

Conclusions In molars the size of the pulp chamber

was influenced by the presence of restorations and the

PBF was reduced when restorations were present. Size

and extent of restorations had a significant effect on PBF.

Keywords: bitewing radiographs, endodontics, laser

Doppler flowmetry, pulp vitality.

Received 20 August 2008; accepted 5 August 2009

Introduction

Conventional pulp sensitivity tests rely on an intact

nerve supply and a patient response to a noxious

stimulus. Problems of electric pulp testing restored

posterior teeth include limited access to enamel and/or

dentine and the leakage of current via metallic resto-

rations or adjacent teeth (Pantera et al. 1992, Myers

1998). A test involving blood perfusion such as laser

Doppler flowmetry (LDF) eliminates the need for a

patient response. LDF has been used to assess the pulp

condition of anterior teeth, but few investigators have

focused on posterior teeth or teeth with restorations.

Whilst endodontic diagnosis involves integration of

clinical and radiographic findings, in a study of

treatment need only one-third of molars with disease

could be diagnosed from radiographs alone, compared

with half of the incisors (Petersson et al. 1986).

Improvements to pulp testing of posterior teeth are

therefore important, especially when there is a risk of

misdiagnosis and the economics of molar root canal

treatment are considered.

The aims of this study were to investigate pulpal

blood flow (PBF) in first molar teeth of subjects of

Correspondence: Associate Professor Nicholas Chandler,

Department of Oral Rehabilitation, School of Dentistry, Uni-

versity of Otago, P.O. Box 647, Dunedin 9054, New Zealand

(Tel.: 0064 3 479 7124; fax: 0064 3 479 5079; e-mail:

[email protected]).

doi:10.1111/j.1365-2591.2009.01630.x


similar age, and to relate this to the restorations present

and to pulp chamber dimensions on radiographs.


Ethical approval was granted by the Otago Ethics

Committee. An examination of the bitewing radio-

graphs of 275 dental students was carried out to screen

for appropriate combinations of healthy and restored

maxillary and mandibular first molar teeth. The

bitewings were available and none were taken for the

study. The images were all taken with the same X-ray

machine (Source 1, Belmont Co., Osaka, Japan:

70 kVp, 10 mA) using E speed film (Kodak Co.,

Rochester, NY, USA) held in a Rinn bitewing holder/

beam alignment device (Dentsply Rinn, Weybridge, UK)

and film processing was automatic (All-Pro 2000M,

Hicksville, NY, USA). Seventy-four individuals had

radiographs which indicated suitable dentitions and

they gave their written informed consent for investiga-

tion of PBF. The subjects were examined clinically.

Their first molars were caries and symptom-free and

there was no history of operative dental work, ortho-

dontics or trauma in the previous 6 months. Restora-

tions were recorded and their radiographic appearances

verified; none of the teeth had carious lesions. There

were two groups:

Group A. Unrestored teeth were compared with

contralateral restored teeth (Fig. 1). The 43 subjects

were 26 females and 17 males with a mean age of

22 years 0 months (range 19 years 0 month to

24 years 7 months). The teeth were 19 maxillary

and 24 mandibular first molars. Twenty-eight of the

restored teeth had single or multiple occlusal restora-

tions. Four teeth had both occlusal and buccal

restorations. Eleven teeth had mesio-occlusal (MO) or

disto-occlusal (DO) restorations, with two having a

buccal extension or separate buccal restoration.

Group B. Teeth with small and large restorations

were compared (Fig. 2). Restoration size was related to

the numbers of restored tooth surfaces, otherwise this

judgement was more subjective. Many small restora-

tions seen on radiographs during the screening

process proved to be extensive when the subjects

were examined clinically. Similarly, several very

extensive composite resin overlays were found that

were nearly invisible on the bitewings. The 31

subjects were 23 females and 8 males with a mean

age of 21 years 4 months (range 19 years 10 months

to 25 years 2 months). There were 16 maxillary and

15 mandibular first molars. Two teeth had buccal

restorations and two had occlusopalatal restorations.

One tooth had an occlusobuccal restoration, and there

were four teeth with small DO restorations. In the

group of heavily restored contralateral teeth were six

large or multiple occlusal restorations and 25 teeth

with either an MO, a DO or a mesio-occlusal-distal

restoration. One of these proximally restored teeth was

overlaid and three of these teeth also had restored

buccal surfaces.

The teeth were tested with an electric pulp tester

(Vitality Scanner Model 2001; Analytic Technology,

Redmond, WA, USA) with the probe tip on the

mesiobuccal cusp tip whenever possible (Lin et al.

2007). All the subjects gave unambiguous positive

responses for both teeth under investigation; as dental

students, they had a good understanding of electric

testing and responses to the stimuli so further pulp tests

were considered unnecessary. From the dental history,

clinical examination and radiological evidence avail-

able all the pulps were considered healthy. A laser

Doppler blood flow monitor (MBF3-D, Moor Instru-

ments, Axminster, UK) was used to display and store

(a)

(b)

Figure 1 Group A; example of unrestored (a) and restored (b)

contralateral mandibular first molar teeth.

Blood flow in restored molars Chandler et al.


PBF signals from the teeth. The laser probe (Moor P5a,

Axminster, UK) was positioned 2 mm from the gingival

margin of the teeth (Roebuck et al. 2000) on the mid-

buccal surface using a splint made of a dark brown

silicone putty (President, Coltene AG, Altstatten, Swit-

zerland). This stabilized the probe and excluded light.

The same probe was used throughout the experiment

and regularly recalibrated. For the small number of

teeth featuring restorative material on the buccal

surface a site to the mesial was adopted for both teeth

in that individual. Two subjects had discrete opaque

hypoplastic regions of buccal enamel which were also

avoided. An accurate, clean hole was cut in the putty

to support the probe using a 1.5 mm dermal biopsy

punch (Miltex, Bethpage, NY, USA). The subjects lay

supine for 10 min in a quiet, draft-free room and then a

resting PBF trace was recorded at 20 Hz for 3 min for

each tooth. All the teeth provided clearly pulsatile

traces at heart beat frequency (Fig. 3). The mean flux

(PBF, machine units) was calculated for each trace by

the flowmeter’s processor.

The radiographs were code numbered and scanned

(DSR-1000 scanner; Electromedical Systems, Nyon,

Switzerland) and pulp measurements were made on a

desktop computer by a single operator using the Scion

Image Program (Scion Corp., Frederick, MD, USA). A

calibration radiograph of a measuring grid with 1 mm

squares (PHIL-X Grid; Medidenta International, Wood-

side, NY, USA) was also made and scanned. Prior to

measurement three pilot studies involving linear and

area assessments of bitewings were performed. These

revealed a high correlation of repeat measurements by

the operator (r = 0.90, 0.92 and 0.92 respectively)

with a regression slope not significantly different to 1.0.

For each tooth a line was drawn across the image

between the mesial and distal enamel-cementum mar-

gins, the area above this being termed the clinical

crown. Two areas were measured, the total pulp

chamber area (above a line across the most superior

part of the pulpal floor) and the pulp chamber area in

the clinical crown. The pulp chamber width at the

cervix and the heights of the mesial and distal pulp

horns (from the enamel-cementum line) were also

measured (Fig. 4).

Following the PBF recordings the radiograph codes

were broken and the data matched for each tooth.

Statistical significance was determined using Student’s

t-test (2-tailed in related samples).

Results

Means and standard deviations of the pulp chamber

measurements and PBF values are shown for group A

in Table 1 and for group B in Table 2.

In group A the presence of restorations was associ-

ated with a significant reduction of total pulp chamber

area, pulp chamber area in the clinical crown, mesial

pulp horn height and PBF (P-values all <0.05,

(a)

(b)

Figure 2 Group B; example of contralateral mandibular first

molar teeth with small (a) and large (b) restorations, in this

patient gold inlays.

Figure 3 Pulpal blood flow trace as seen on flowmeter screen.

Chandler et al. Blood flow in restored molars


Table 1). In group B the presence of large restorations

was associated with a significant reduction in total pulp

chamber area, pulp chamber width at the cervix and

PBF compared with teeth with smaller restorations

(P-values all <0.05, Table 2).

In group A the maxillary teeth revealed a significant

reduction in PBF (P = 0.008). In group B both the

maxillary and mandibular teeth demonstrated reduc-

tions in PBF, with this change more pronounced in the

maxillary teeth (P-values 0.003 and 0.047).

Discussion

Previous blood flow studies have made use of radio-

graphs when available (Odor et al. 1994a,b, Premdas &

Pitt Ford 1995). Other workers have taken radiographs

to determine that the pulp chamber of the tooth was

visible and that the periapical condition was normal

(Hartmann et al. 1996). The present study is unique in

including radiographic measurements alongside PBF

data. Radiographs had already been taken; ethical

approval would not have been granted to take radio-

graphs purely for the experiment. Digitization allowed

accurate measuring of magnified images of the pulp

spaces. The images and their measurements were not

available to the clinical researcher when the PBF

recordings were made. This was a deliberate aspect of

the experimental design in order to avoid bias. Had

radiographs been observed it might have been possible

to direct the LDF probe towards a maximal region of

pulp in an attempt to achieve stronger PBF signals.

Undergraduate students provided a group with a

narrow age range to reduce major variation due to

dentine deposition (Woods et al. 1990) and the

decrease in pulp chamber size (Ketterl 1983) and PBF

(Ikawa et al. 2003), which accompany increasing age.

A maximum of 6 months elapsed between taking

radiographs and recording PBF, with most readings

made within a few weeks.

There are few LDF studies of molars, and no

consensus on optimum probe positions. In studies of

local anaesthesia on mandibular first molars the probes

were positioned on the mid-buccal surfaces of the teeth

(Odor et al. 1994a,b).

Laser flowmetry studies of maxillary incisors reveals

an increase in PBF signal as probes are moved from

incisal to gingival (Ramsay et al. 1991, Ingolfsson et al.

1994, Hartmann et al. 1996). This could be due to a

larger volume of tissue being sampled as the probe is

Figure 4 Diagram of measurements made from radiographs.

(1) Total pulp area (area above a line drawn across the most

superior part of the pulpal floor). (2) Pulp area in the clinical

crown (area above a line drawn between mesial and distal

enamel-cementum junctions). (3) Pulp width at cervix

(between mesial and distal enamel-cementum junctions). (4)

Mesial horn height (above a line drawn between mesial and

distal enamel-cementum junctions). (5) Distal horn height

(above a line drawn between mesial and distal enamel-

cementum junctions).

Table 1 Group A. Mean pulp

dimensions and PBF for unrestored and

restored first molars (standard deviations

in parentheses)

Unrestored (n = 43) Restored (n = 43) t-test P-value

Total pulp area (mm2) 8.92 (2.59) 8.32 (2.55) 0.039*

Pulp area clinical crown (mm2) 3.41 (1.58) 2.89 (1.66) 0.021*

Pulp width at cervix (mm) 3.67 (0.71) 3.40 (1.15) 0.075

Mesial horn height (mm) 1.57 (0.57) 1.34 (0.58) 0.012*

Distal horn height (mm) 0.99 (0.47) 0.90 (0.46) 0.200

PBF (machine units) 98 (47) 83 (46) 0.028*

PBF, pulpal blood flow.

*Significant at the 0.05 level.



moved apically. No data on this effect in molar teeth

have been published.

It has been shown that light from an LDF probe

placed 2 mm above the buccal enamel-cementum

junction in molars is transmitted apically towards the

radicular pulp (Odor et al. 1996). It should therefore be

possible to use LDF to investigate the health of molar

pulps which have diminished size or are situated

unusually deeply.

The pulp chamber width at the cervix of mandibular

first molar teeth is significantly larger than for maxil-

lary teeth (Chandler et al. 2003). A significant decrease

in pulp chamber width was found in group B, and this

may have a role in the success of LDF if an optimal

probe position cannot be achieved. A study of maxillary

central incisors found PBF measurements made at

various mesiodistal locations at the same level on the

tooth did not differ (Ramsay et al. 1991). Forty teeth

(38%) in the present study had proximal restorations.

In a previous study, 26% of the teeth had proximal

restorations (Chandler et al. 2003). In keeping with the

present findings, those restorations were also related to

a significant reduction in total pulp chamber area. The

pulp chamber area in the clinical crown was also

smaller amongst the restored teeth.

Most investigations of the dimensions of opposite

pairs of teeth have been done in man (Black 1980). The

differences are small and probably because of complex

genetic and environmental factors and known as

fluctuating asymmetry. Despite radiographic standard-

ization, minor changes in film position and beam

geometry across the jaws are likely in the present

study, together with anatomical variations such as

tooth rotation. Fluctuating asymmetry of the pulp

space has not been reported; the variations in PBF

signals were in any event much larger. Data may have

been recorded from the gingival crest (Vongsavan &

Matthews 1996) and it is assumed this was comparable

on both sides of the mouth. There appears to be no

published data on differences in blood perfusion on

different sides of the jaws, or LDF studies which

compare signal strengths between healthy maxillary

and mandibular first molars. In radiographic studies

pulp horn location is more reliable in mandibular

molars with bitewings, with the mesiolingual horn

responsible for clarity of the image and longer than that

of the mesiobuccal horn of maxillary teeth (Kandemir

1998, Chandler et al. 2003). The maxillary first molar

has the largest pulp chamber volume of human teeth,

followed by the mandibular first molar; in one study the

difference was 23% (Fanibunda 1986). This difference

in volume may account for the more noticeable

reductions in PBF in maxillary molars in both groups

in the present experiment.

Conclusion

In first molars, the pulp chamber size was influenced by

the presence of restorations. PBF was reduced when

restorations were present. The size and extent of the

restorations had a significant effect on the blood flow

recordings.

References

Black TK (1980) An exception to the apparent relationship

between stress and fluctuating dental asymmetry. Journal of

Dental Research 59, 1168–9.

Chandler NP, Pitt Ford TR, Monteith BD (2003) Coronal pulp

size in molars: a study of bitewing radiographs. International


Fanibunda KB (1986) A method of measuring the volume of

human dental pulp cavities. International Endodontic Journal

19, 194–7.

Hartmann A, Azerad J, Boucher Y (1996) Environmental

effects on laser Doppler pulpal blood-flow measurements in

man. Archives of Oral Biology 41, 333–9.

Ikawa M, Komatsu H, Ikawa K, Mayanagi H, Shimauchi H

(2003) Age-related changes in the human pulpal blood flow

measured by laser Doppler flowmetry. Dental Traumatology

19, 36–40.

Table 2 Group B. Mean pulp dimensions

and PBF for first molars with small

restorations and large restorations

(standard deviations in parentheses)

Small restoration

(n = 31)

Large restoration

(n = 31)

t-test

P-value

Total pulp area (mm2) 8.20 (2.47) 7.20 (2.73) 0.003*

Pulp area clinical crown (mm2) 2.43 (1.34) 2.09 (1.43) 0.134

Pulp width at cervix (mm) 3.54 (1.02) 3.19 (1.25) 0.032*

Mesial horn height (mm) 1.19 (0.50) 1.08 (0.60) 0.218

Distal horn height (mm) 0.74 (0.40) 0.63 (0.44) 0.162

PBF (machine units) 102 (55) 73 (44) 0.001*

PBF, pulpal blood flow.

*Significant at the 0.05 level.

Chandler et al. Blood flow in restored molars


Ingolfsson AER, Tronstad L, Riva CE (1994) Reliability of laser

Doppler flowmetry in testing vitality of human teeth.

Endodontics and Dental Traumatology 10, 185–7.

Kandemir S (1998) The radiographic determinability of the

distance between the pulp horns in the permanent first and

second molar teeth. Journal of Oral Science 40, 143–6.

Ketterl W (1983) Age-induced changes in the teeth and their

attachment apparatus. International Dental Journal 33, 262–

71.

Lin J, Chandler NP, Purton DG, Monteith B (2007) Appropri-

ate electrode placement site for electric pulp testing first

molar teeth. Journal of Endodontics 33, 1296–8.

Myers JW (1998) Demonstration of a possible source of error

with an electric pulp tester. Journal of Endodontics 24, 199–

200.

Odor TM, Pitt Ford TR, McDonald F (1994a) Effect of inferior

alveolar nerve block anaesthesia on the lower teeth.


Odor TM, Pitt Ford TR, McDonald F (1994b) Adrenaline in

local anaesthesia: the effect of concentration on dental

pulpal circulation and anaesthesia. Endodontics and Dental


Odor TM, Watson TF, Pitt Ford TR, McDonald F (1996)

Pattern of transmission of laser light in teeth. International


Pantera EA, Anderson RW, Pantera CT (1992) Use of dental

instruments for bridging during electric pulp testing. Journal


Petersson K, Wennberg A, Olsson B (1986) Radiographic and

clinical estimation of endodontic treatment need. Endodon-

tics and Dental Traumatology 2, 62–4.

Premdas CE, Pitt Ford TR (1995) Effect of palatal injections on

pulpal blood flow in premolars. Endodontics and Dental


Ramsay DS, Artun J, Martinen SS (1991) Reliability of pulpal

blood-flow measurements utilizing laser Doppler flowmetry.

Journal of Dental Research 70, 1427–30.

Roebuck EM, Evans DJP, Stirrups D, Strang R (2000) The

effect of wavelength, bandwidth, and probe design and

position on assessing the vitality of anterior teeth with laser

Doppler flowmetry. International Journal of Paediatric


Vongsavan N, Matthews B (1996) Experiments in pigs on the

sources of laser Doppler blood-flow signals recorded from

teeth. Archives of Oral Biology 41, 97–103.

Woods MA, Robinson QC, Harris EF (1990) Age-progressive

changes in pulp widths and root lengths during adulthood:

a study of American blacks and whites. Gerodontology 9, 41–

50.



Analysis of resin tags formation in root canaldentine: a cross sectional study

Y. Malyk, C. Kaaden, R. Hickel & N. IlieDepartment of Restorative Dentistry and Periodontology, Ludwig-Maximilians University, Munich, Germany

Abstract

Malyk Y, Kaaden C, Hickel R, Ilie N. Analysis of resin tags

formation in root canal dentine: a cross sectional study.


Aim To evaluate the length, density and quality of

resin tags formed by penetration of various types of

adhesive systems into dentinal tubules at various cross

section levels of the root canal in correlation to the

density of dentinal tubules.

Methodology Thirty mandibular premolars were

instrumented and fibre posts were inserted with three

different adhesive systems with and without activator:

etch & rinse XP Bond and XP Bond/Self Cure Activator;

self-etch (two-step) AdheSE and AdheSE/AdheSE DC

Activator and self-etch (one-step) Hybrid Bond and

Hybrid Bond/Hybrid Brushes. The resin tags were

evaluated from slices obtained from sections perpen-

dicular to the long axis of the teeth at 3, 6, and 9mm

from the root apex under a Confocal Laser Scanning

microscope.

Results In all groups, lack of continuity of resin tag

length, density and quality was observed not only

from the cervical to the apical region of each root

canal, but also in a mesio-distal direction to the long

axis of the root. Application of etch & rinse adhesive

in contrast to the self-etch adhesives provided the

formation of the shorter, but considerably denser,

more homogeneous and not interrupted resin

tags with similar length. Use of the activator for all

types of adhesives significantly increased the com-

pleteness (P=0.014) and continuity (P=0.024) of

resin tags.

Conclusions None of the investigated adhesives

were able to completely infiltrate the dentinal tubules

in the entire root canal. Use of the etch & rinse adhesive

system and the activators significantly increased the

density and the quality of resin tags.

Keywords: adhesive systems, confocal laser scanning

microscopy, resin tags, root canal dentine.

Received 3 February 2009; accepted 5 August 2009

Introduction

The bonding principle of dental adhesives is based on

the formation of a hybrid layer (Nakabayashi et al.

1991) as well as the penetration of adhesive into

dentine tubules and the formation of ‘resin tags’ (Titley

et al. 1995, Ferrari & Davidson 1996). Adaptation of

adhesive systems for fibre post bonding in root canal is

an attractive clinical concept, but its implementation is

controversial for several reasons: influence of the end-

odontic procedure, polymerization shrinkage (Feilzer

et al. 1993, Carvalho et al. 1996), unfavourable cavity

configuration factor (C-Factor) (Carvalho et al. 1996,

Tay et al. 2005), poor control of moisture (Bouillaguet

et al. 2003) or polymerization difficulties in the apical

regions (Roberts et al. 2004).

So far there are no conclusive and specific reports on

the importance of the resin tags in the quality of the

bond. While some authors measured a higher bond

strength in the coronal section of the root canal

because of the higher density of dentinal tubules and

the longer resin tags formed in this area (Patierno et al.

1996, Kurtz et al. 2003, Mallmann et al. 2005), other

authors found no correlation between bond strength

Correspondence: Yuriy Malyk, Department of Restorative

Dentistry and Periodontology, Ludwig-Maximilians University,

Goethe st. 70, 80336, Munich, Germany (Tel.:

+498951609337; fax: +498951509302; e-mail: ymalyk@

dent.med-uni.muenchen.de).

doi:10.1111/j.1365-2591.2009.01631.x


and resin penetration into dentinal tubules (Tay &

Pashley 2001, Kremeier et al. 2008). Duret et al.

(1990) and Pegoretti et al. (2002) noted that the

resultant homogeneous biomechanical unit allows a

more uniform stress distribution, which better pre-

serves the weakened tooth structure and reduces

microleakage at the dentine-cement interface as well

as reinfection of the peri-apical area (Bachicha et al.

1998, Reid et al. 2003).

Currently, various types of adhesives systems that

either follow an etch & rinse or self-etch approach, can

be used to bond fibre posts into root canals. Previous

studies showed that self-etching primers are more

advantageous for bonding of fibre posts, since they

contain a high concentration of acidic monomers that

demineralize the substrate, that acid does not need to

be removed with water, and the bond forms simulta-

neously to dentine (Yoshiyama et al. 1998, Tay et al.

2003). A recently published investigation showed that

bond strength was not influenced by the demineraliza-

tion of root canal dentine neither with phosphoric acid

nor with self-etching systems, but was affected by the

luting agent and the region of the root canal (Bitter

et al. 2006). As the number of tubules decreases from

the crown to the apex (Carrigan et al. 1984), the

response to acid etching and, consequently, dentine

bonding can vary among different areas of the same

root canal (Ferrari et al. 2000a). The application of

adhesive systems in the root canal usually shows a

non-uniform resin tag formation with consequent

decrease of retention towards the apex (Ferrari &

Mannocci 2000, Mannocci et al. 2004).

Although bonding to the root dentine wall has made

undeniable progress in recent years, the loss of adhe-

sion at the adhesive/root dentine interface is still the

main reason for leakage (Ferrari et al. 1994, Ferrari &

Davidson 1996), decrease in bond strength (Bouillag-

uet et al. 2003, Bolhuis et al. 2005) and hence, failure

of restorations (Ferrari et al. 2000b, Bouillaguet et al.

2003, Mannocci et al. 2003). To decrease these neg-

ative influences, some manufactures recommend the

use of an activator, which triggers the self-curing

reaction of the adhesives, to produce an additional set

that should improve the bonding to root canal walls.

On the basis of this consideration the aim of this study

was to analyze the resin tags formed by various types of

adhesive systems in relation to the density of dentinal

tubules in the cervical, middle and apical third of the

root canal using Confocal Laser Scanning Microscopy

(CLSM). Three types of adhesive systems were com-

pared in this study: an etch & rinse, a self-etch two-step

and a self-etch one-step system each with and without

addition of self-cure activators. Moreover, the density of

dentinal tubules and the quality of resin tag formation

were recorded. The null hypothesis tested was that both

the type of adhesive system and the morphology of the

dentinal tubules do not affect the formation of the resin

tags.

Material and methods

Thirty extracted human mandibular premolars were

used. The criteria for tooth selection were the presence

of a single root canal, verified radiographically, along

with a fully formed apex and no visible root caries or

fractures, verified by examination with a 3.2· magni-

fying glass.

Each root was separated from the crown with a low-

speed diamond saw (Isomet, Buehler, Lake Bluff, IL,

USA) to obtain a 12mm long specimen. The root canals

were shaped with K- and H-files (Dentsply Maillefer,

Ballaigues, Switzerland) to size 50, 1mm short of the

apex. After each instrument size, the canals were

irrigated with 1mL of 1% sodium hypochlorite (NaOCl).

The teeth were then randomly divided into six

groups (n=5). The roots were enlarged with low-speed

DT Drills sizes 0–3 designated for the respective post

system (VDW, Munich, Germany). The depth of the

post space preparation was 11mm. Irrigation was

performed after change of each drill size with 1mL of

1% NaOCl. Afterwards, the root specimens were

irrigated with 9% EDTA for 1min to remove the smear

layer, and then rinsed for 1min with 0.9% NaCl. The

fibre posts DT Light size 3 (VDW) were shortened to a

length of 15mm with water-cooled diamond rotary

cutting instrument and inserted into the root specimens

using six different adhesive systems. Inserted adhesive

systems: XP Bond (Dentsply DeTrey, Konstanz, Ger-

many), XP Bond/Self Cure Activator (SCA) (Dentsply

DeTrey), AdheSE (Ivoclar Vivadent, Schaan, Lichten-

stein), AdheSE/AdheSE DC Activator (Ivoclar Viva-

dent), Hybrid Bond (Sun Medical, Shiga, Japan) and

Hybrid Bond/Hybrid Brushes (Sun Medical). The fluo-

rescent dye 0.1% Rhodamine B isothiacyanate (RITC)

(Merck, Darmstadt, Germany) was mixed into the

components of the adhesive systems, to highlight the

resin tags under CLSM. The labelled adhesives were

applied into the canal space using microbrush tips for

the cervical and middle third of the canal and paper

points for the apical third. The adhesive systems and

application method in this study are summarized in

Table 1. The dual-cured cement Calibra Esthetic Resin

Analysis of resin tags Malyk et al.


Tab

le1

Th

ea

dh

esiv

esy

stem

sa

nd

ap

pli

cati

on

met

ho

du

sed

inth

isin

ves

tig

ati

on

Gro

up

Ad

hesi

ve

Man

ufa

ctu

reLo

tN

o

Ad

hesi

ve

typ

eB

on

dP

rim

er

Act

ivato

rA

pp

lica

tio

n

1X

PB

on

dD

en

tsp

ly

DeT

rey;

Ko

nst

an

z,

Germ

an

y

0609001329

To

tal-

etc

hC

arb

oxyli

caci

dm

od

ified

dim

eth

acr

yla

te(T

CB

resi

n),

PE

NT

A,

UD

MA

,

TE

GD

MA

,H

EM

A

––

To

tal

etc

hw

ith

34%

ph

osp

ho

ric

aci

dfo

r15s

Wate

rsp

ray,

dry

wit

hair

an

d

pap

er

po

ints

Bo

nd

ap

ply

Exce

ssre

mo

ve

usi

ng

pap

er

po

ints

Lig

ht

cure

for

10s

2X

PB

on

d/

Self

Cu

re

Act

ivato

r

Den

tsp

ly

DeT

rey;

Ko

nst

an

z,

Germ

an

y

060516

To

tal-

etc

hC

arb

oxyli

caci

dm

od

ified

dim

eth

acr

yla

te(T

CB

resi

n),

PE

NT

A,

UD

MA

,

TE

GD

MA

,H

EM

A

–U

DM

A,

HE

MA

,

cata

lyst

,

ph

oto

init

iato

r,

ace

ton

e,

wate

r

To

tal

etc

hw

ith

34%

ph

osp

ho

ric

aci

dfo

r15s

Wate

rsp

ray,

dry

wit

hair

an

d

pap

er

po

ints

Bo

nd

mix

wit

hact

ivato

r

(1:1

)an

dap

ply

3A

dh

eS

E

Pri

mer/

Ad

heS

E

Bo

nd

Ivo

clar

Viv

ad

en

t;

Sch

aan

,

Lic

hte

nst

ein

Pri

mer:

JO

6075

Bo

nd

:

AK

03345

Self

-etc

hH

EM

A,

dim

eth

acr

yla

te,

sili

con

dio

xid

e,

init

iato

rs

Dim

eth

acr

yla

te,

ph

osp

ho

nic

aci

d

acr

yla

te,

wate

r

–P

RIM

ER

ap

ply

for

30s

Dry

wit

hair

an

dp

ap

er

po

ints

Bo

nd

ap

ply

Exce

ssre

mo

ve

usi

ng

pap

er

po

ints

Lig

ht

cure

for

10s

4A

dh

eS

E/

Ad

heS

ED

C

Act

ivato

r

Ivo

clar

Viv

ad

en

t;

Sch

aan

,

Lic

hte

nst

ein

K00997

Self

-etc

hH

EM

A,

dim

eth

acr

yla

te,

sili

con

dio

xid

e,

init

iato

rs

Dim

eth

acr

yla

te,

ph

osp

ho

nic

aci

d

acr

yla

te,

wate

r

Init

iato

rs,

solv

en

ts,

eth

an

ol

Pri

mer

ap

ply

for

30s

Dry

wit

hair

an

dp

ap

er

po

ints

Bo

nd

mix

wit

hact

ivato

r

(1:1

)an

dap

ply

5H

yb

rid

Bo

nd

Su

nM

ed

ical;

Sh

iga,

Jap

an

LF-2

Self

-etc

h

‘on

e-b

ott

le’

4-m

eth

acr

ylo

xyeth

ylt

rim

ell

itate

an

hyd

rid

e

––

Bo

nd

ap

ply

for

20s

Exce

ssre

mo

ve

usi

ng

pap

er

po

ints

Lig

ht

cure

for

10s

6H

yb

rid

Bo

nd

/Hyb

rid

Bru

shes

Su

nM

ed

ical;

Sh

iga,

Jap

an

LF-2

Self

-etc

h

‘on

e-b

ott

le’

4-m

eth

acr

ylo

xyeth

ylt

rim

ell

itate

an

hyd

rid

e

–H

yb

rid

Bru

shes

con

tain

a

bo

nd

ing

pro

mo

ter

Bo

nd

mix

wit

hb

rush

an

d

ap

ply

for

20s

Exce

ssre

mo

ve

usi

ng

pap

er

po

ints

Lig

ht

cure

for

10s

PE

NT

A,

Di-

Pen

taery

trit

ho

l-P

en

ta-A

cryla

te-M

on

op

ho

sp

hate

;U

DM

A,

ure

than

ed

imeth

acr

yla

te;

TE

GD

MA

,tr

ieth

yle

ne

gly

col

dim

eth

acr

yla

te;

HE

MA

,h

yd

roxyeth

yl-

meth

acr

yla

t.

Malyk et al. Analysis of resin tags


Cement (Dentsply DeTrey) was mixed for 20s and

spread onto the surface of the fibre post and into the

root canal with a Lentulo Spiral, as recommended by

manufacture. The posts were inserted into the root

canal and manually held in place throughout the self-

cure set time of 5min. After initial set, the light source

(Freelight 2; 1241mWcm)2; 3M-ESPE, Seefeld, Ger-

many) was placed on the post surface and the resin

luting cement was polymerized for 20s. All procedures

were performed by the same operator.

After storing for 24h at 37�C and 100% humidity

the samples were sectioned in 1mm thin slices at

9mm (cervical), 6mm (middle) and 3mm (apical) from

the root apex with a microtome saw (Leica SP 1600;

Leica, Nussloch, Germany). The slices were then

polished with a series of silicon carbide abrasive

papers (1200, 2400 and 4000 grit) on an automatic

polishing device (Exact 40 CS; Exact Apparatebau,

Norderstedt, Germany). The samples were kept moist

during the study.

CLSM examination

The density, length and quality of resin tags, as well as

the density of dentinal tubules and the amount of non-

infiltrated dentinal tubules were recorded in the cervi-

cal, middle and apical regions of the root canal using

LSM-510 Meta microscope (Carl Zeiss, Jena, Germany)

equipped with a water-immersion objective (Achro-

plan·63/0.95W). Microscopy was performed at four

standardized areas (buccal, lingual, mesial, and distal

to the long axis of the root) of each root slice (Fig. 1).

The visualized layer was selected 10lm below the slice

surface and was 15lm thick (15 images, distance

between images was 1lm). The image size was

150·150lm with resolution of 2656·2924 pixels.

The image analysis and 3D reconstruction were carried

out using LSM Image Browser 4.6 (Carl Zeiss).

The slices were scanned with ‘multitracking’, allow-

ing the simultaneous evaluation of labelled adhesive

and the density of the dentinal tubules. To observe the

RITC labelled adhesive, the 488nm laser line for

excitation (beamsplitter: HFT 405/488) was used, thus

allowing emitted fluorescent light to pass through and

hit the detector (filter: BP 530–560nm). To identify the

dentinal tubules, which were not infiltrated by adhesive

labelled with fluorescent dye a second channel was

added. The reflection mode (excitation 488nm) was

used with a special beam splitter (NT 80/20) and a

long-pass filter with a detection window for the

reflected light (long pass 420nm). In reflection, all

dentinal tubules give a strong and clear signal because

of a change in the optical medium between the dentine

and the medium (distilled water). Counting of dentinal

tubules was performed on images taken on an area

15·100lm. The results were indicated as tubules/

mm2.

Additionally, to determine the quality of resin

penetration into the dentinal tubules, the three factors

of completeness, continuity and evenness of resin tags

were evaluated using the classification system with

scores from 0 to 3 indicating best to worst quality.

Completeness (homogeneity of adhesive penetration

into dentinal tubules): 0=90–100% of the dentinal

tubules were homogenously filled with adhesive;

1=50–90%; 2=10–50%; 3=0–10%.

Continuity (uninterruptedly of the resin tags): 0=90–

100% of resin tags were not interrupted; 1=50–90%;

2=10–50%; 3=0–10%.

Evenness (equability of the resin tags length): 0=90–

100% of resin tags were of equal length; 1=50–90%;

2=10–50%; 3=0–10%.

The general linear model procedure of sas/stat 9.2

Software (SAS Institute, Cary, NC, USA) was used to

quantify the influence of the adhesive and root canal

region on the quality of the adhesive/root dentine

interface (a=0.05). One- and multiple-way anova and

Tukey post hoc tests for the morphology of dentinal

tubules were used (a=0.05).

Results

The results obtained regarding density of resin tags,

dentinal tubules and resin tags length are summarized

1 m

m

Figure 1 Preparation of cross sectional slice. Microscopy was

performed at four standardized areas (buccal, lingual, mesial,

and distal to the long axis of the root) of each root slice.



Tab

le2

Den

tin

al

tub

ule

sd

ensi

ty,

resi

nta

gs

len

gth

an

da

mo

un

to

fd

enti

na

ltu

bu

les

no

tp

enet

rate

dw

ith

ad

hes

ive

com

po

nen

ts(m

ean

+S

D)

Po

siti

on

Ad

hesi

ve

syst

em

s

Nu

mb

er

of

den

tin

al

tub

ule

s

(mm

2)

Len

gth

of

resi

nta

gs

(lm

)

Len

gth

of

resi

nta

gs

(bu

ccal,

lm

)

Len

gth

of

resi

nta

gs

(lin

gu

al,

lm

)

Len

gth

of

resi

nta

gs

(mesi

al,

lm

)

Len

gth

of

resi

nta

gs

(dis

tal,

lm

)

No

t

infi

ltra

ted

den

tin

al

tub

ule

sw

ith

ad

hesi

ve

(%)

No

tin

filt

rate

d

den

tin

al

tub

ule

sw

ith

ad

hesi

ve

(bu

ccal,

%)

No

tin

filt

rate

d

den

tin

al

tub

ule

s

wit

had

hesi

ve

(lin

gu

al,

%)

No

tin

filt

rate

d

den

tin

al

tub

ule

s

wit

had

hesi

ve

(mesi

al,

%)

No

tin

filt

rate

d

den

tin

al

tub

ule

sw

ith

ad

hesi

ve

(dis

tal,

%)

Cerv

ival

XP

Bo

nd

37985

(2608)

26.4

(5.3

)a27.0

(7.5

)27.3

(5.9

)22.3

(5.9

)21.5

(3.9

)6.6

3(2

.8)b

6.6

(2.7

)6.4

(2.2

)7.4

(3.5

)7.2

(2.5

)

XP

Bo

nd

/SC

A27.5

(6.8

)a28.1

(6.4

)27.4

(6.5

)22.9

(4.2

)23.8

(7.3

)0.4

(0.2

)a0

00

1.8

(0.4

)

Ad

heS

E93.7

(26.2

)c95.1

(20.0

)92.2

(30.0

)80.6

(38.8

)84.1

(36.0

)9.9

1(4

.8)c

5.2

(3.9

)8.4

(4.6

)9.2

(4.3

)11.3

(10.8

)

Ad

heS

E65.5

(11.2

)b66.1

(11.0

)67.3

(11.0

)65.3

(10.2

)64.6

(12.2

)1.6

(0.7

)a2.2

(1.1

)1.8

(0.7

)2.1

(1.6

)2.0

(0.6

)

HB

79.1

(19.6

)b,c

78.2

(20.8

)81.5

(18.0

)77.8

(16.6

)78.7

(18.2

)2.4

(0.7

)a0

00

6.8

(0.4

)

HB

/Hyb

rid

Bru

shes

152.3

(29.2

)d156.6

(24.8

)152.7

(21.5

)150.2

(27.1

)149.8

(26.1

)0.5

2(0

.2)a

00

1.8

(0.4

)0

Mid

dle

XP

Bo

nd

31028

(3671)

19.5

(4.5

)a20.0

(4.4

)19.4

(4.2

)18.4

(3.9

)18.8

(3.8

)5.6

6(2

.1)b

5.0

(1.6

)4.3

(3.0

)7.7

(2.0

)8.8

(6.2

)

XP

Bo

nd

/SC

A23.6

(5.7

)a24.0

(3.8

)24.0

(6.1

)18.2

(4.0

)20.7

(5.8

)2.8

8(1

.2)a

0.8

(0.4

)1.7

(0.8

)4.8

(2.7

)4.0

(2.9

)

Ad

heS

E76.6

(21.4

)b77.6

(19.0

)82.8

(22.9

)69.1

(17.9

)74.6

(26.8

)5.7

4(2

.7)c

3.8

(3.1

)4.2

(2.5

)6.2

(1.7

)15.0

(11.2

)

Ad

heS

E70.8

(15.2

)b79.6

(14.2

)64.5

(12.3

)61.8

(13.2

)60.8

(13.1

)5.3

6(2

.8)a

,b3.2

(1.6

)3.4

(2.4

)10.2

(6.8

)3.8

(1.6

)

HB

64.6

(12.8

)b65.7

(11.7

)69.1

(14.6

)61.4

(13.7

)62.2

(10.9

)3.6

6(1

.9)a

1.7

(0.8

)0

5.3

(2.4

)4.9

(2.2

)

HB

/Hyb

rid

Bru

shes

101.1

(34.8

)c96.5

(35.4

)95.4

(38.0

)84.0

(29.2

)95.3

(35.5

)1.0

2(0

.4)a

00

3.3

(0.6

)0.9

(0.4

)

Ap

ical

XP

Bo

nd

26042

(2792)

16.1

(3.3

)a15.9

(3.1

)17.0

(3.5

)14.4

(4.7

)13.8

(5.3

)4.4

3(2

.1)b

3.4

(0.8

)3.2

(2.0

)5.0

(1.2

)6.2

(5.2

)

XP

Bo

nd

/SC

A18.9

(6.4

)a19.5

(2.8

)19.6

(12.0

)11.8

(3.2

)12.2

(3.0

)4.1

7(3

.2)a

0(0

)0

(0)

10.2

(9.2

)5.4

(3.2

)

Ad

heS

E19.1

9(1

2.7

)a22.4

(14.2

)18.1

(10.5

)16.3

(12.9

)16.1

(12.8

)7.8

9(3

.2)c

5.8

(1.6

)5.4

(2.1

)9.6

(6.7

)14.4

(9.0

)

Ad

heS

E13.8

(4.7

)a13.4

(2.8

)14.4

(3.1

)11.3

(5.1

)12.2

(2.6

)6.5

5(3

.1)a

,b2.2

(1.7

)2.2

(1.8

)12.5

(11.4

)12.4

(10.2

)

HB

46.2

(10.2

)b47.1

(7.0

)45.4

(6.1

)44.7

(9.1

)42.7

(7.4

)5.4

7(3

.1)a

1.7

(1.1

)1.9

(0.8

)12.2

(8.8

)9.8

(2.7

)

HB

/Hyb

rid

Bru

shes

58.2

(15.6

)c59.6

(16.7

)59.1

(13.9

)54.4

(15.0

)55.7

(16.9

)4.8

8(2

.6)a

02.1

(1.1

)9.5

(3.8

)9.0

(6.6

)

Su

pers

crip

tle

tters

ind

icate

stati

stic

all

yh

om

og

en

eo

us

sub

gro

up

s(a

=0.0

5).



in Table 2. The typical bonding morphology between

adhesive systems and root dentine is shown in Fig. 2.

The dentinal tubule density varied depending on the

region examined. In the cervical region the tubule

density was significantly higher (37985 per mm2) than

those observed in the middle (31028 per mm2) and

apical (about 26042 per mm2) thirds. Differences in

dentinal tubule densities between the four standardized

areas (buccal, lingual, mesial and distal) within one

part of a root slice were not significant.

Significant differences of the resin tag length were

found between groups (P<0.001). In all groups also a

discontinuity of resin tag length from cervical to

apical region was observed. The resin tags formed by

Hybrid Bond/Hybrid Brushes (Group 6) were found to

be longer than in the other five groups and consisted

of 152.3lm for the cervical, 101.1lm for the middle

and 58.2lm for the apical thirds. In contrast, the

lengths of resin tags for XP Bond (Group 1) samples

were the shortest and reached lengths of up to

26.4lm, 19.5lm and 16.1lm, accordingly. Bonfer-

roni adjustment indicated significant differences in the

depth of adhesive penetration within the apical and

middle regions of the root canal at four standardized

areas (buccal, lingual, mesial and distal) (P<0.001).

The adhesive penetration was deeper at buccal and

lingual areas to the long axis of the root than at

mesial and distal.

None of the adhesive systems tested was able to

infiltrate and flow inside all the dentinal tubules along

the entire root canal. An increase in non-infiltrated

dentinal tubules from cervical to apical region was

observed for all groups (P<0.001). The lowest numbers

of non-infiltrated dentinal tubules were observed for XP

Bond/SCA (Group 2), Hybrid Bond (Group 5) and

Hybrid Bond/Hybrid Brushes (Group 6). Non-infiltrated

dentinal tubules were significantly more frequently

observed at the mesial and distal areas than at the

buccal and lingual (P=0.04).

The quality of resin penetration into the dentinal

tubules was significantly different between the study

groups (Table 3). A significant reduction in complete-

ness, continuity and evenness of resin tags from the

cervical to the apical region of the root canal for all

adhesive systems tested was observed. The resin tags

formed by XP Bond/SCA (Group 2) were more complete

and even than those of other adhesives. In the cervical

region the dentinal tubules were 100% homogenously

filled with adhesive components, declining to 90% in

the middle and 80% in the apical. The least homoge-

nous adhesive penetration was observed for AdheSE

(Group 3) and consisted 55%, 25% and 0%, accord-

ingly. For all types of adhesive systems the use of an

activator influenced significantly the completeness

(P=0.014) and continuity (P=0.024) of resin tags,

but not the evenness of resin tags.

Discussion

Good quality dentine bonding is obtained when a

continuous hybrid layer forms with regular, dense resin

tags (Mjor & Nordahl 1996, Mannocci et al. 1998). The

current investigation compared resin tags length,

quality and density created by various types of adhesive

systems in relation to the density of dentinal tubules in

root canals under CLSM. Because of the lack of

information regarding the adhesive infiltration in the

apical part of the root canal (Tay et al. 2005) the fibre

post was inserted intentionally 1mm from the working

length to examine the resin tags formation along the

whole root canal wall. The null hypotheses tested in

this study – that the type of adhesive material and the

morphology of the dentinal tubules will not affect the

formation of the resin tags – were both rejected.

(a) (b) (c)

Figure 2 CLSM images of the adhesive/root dentine interface under 63· magnifications in reflection and fluorescence mode: RT –

resin tags, DT – dentinal tubules, RD – root dentine, RC – resin composite. (a) Penetration of the adhesive system AdheSE/AdheSE

DC Activator into the dentinal tubules in the cervical region of the root canal. The adhesive system labelled with RITC appears red

in fluorescent mode. (b) Dentinal tubules appear white in reflection mode. (c) Adhesive/root dentine interface in fluorescent and

reflection mode.



The CLS microscopy offers multidimensional access

to different structures of the same sample by staining

them with different markers. In this study, a combina-

tion of the reflection mode with fluorescence markers

allowed 3D-analysis of the adhesive resin penetration

revealing the morphological structure (Fig. 2). For the

visualization of detailed information of the penetration

and distribution of resin tags several advantages of the

confocal technique, compared with SEM imaging were

reported (Bitter et al. 2009). There is no additional

sample preparation necessary, which could cause

shrinking, swelling or similar artefacts by drying or

freezing the sample, as is essential for high resolution

electron microscopy. The RITC used as a fluorochrome

marker in this study is effective in very low concentra-

tions, soluble in water as well as in organic solutions

such as dentine primers (Pioch et al. 1997), moves

freely across the bonded interface, and is stable under

various pH-levels (Sidhu & Watson 1998).

In accordance with the findings of Ferrari & Mann-

occi (2000) and Mjor et al. (2001) who observed

significantly higher density of dentinal tubules in the

cervical third of the root canal than in the middle and

apical thirds, the results of this study revealed decreases

in the density of the dentinal tubules from 37985 per

mm2 in the cervical region to 31028 per mm2 in the

middle and 26042 per mm2 at the apex. Disparity of

dentinal tubule density in four standardized areas

(buccal, lingual, mesial and distal) within one slice of

the root canal was not significant.

The length of resin tags in this study decreased from

cervical to apical within the root canal. The type of

adhesive system significantly affected the resin tags

length (P<0.001). The self-etch adhesive Hybrid Bond/

Hybrid Brushes (one-step) created the deepest penetra-

tion into the dentinal tubules compared with the self-

etch (two-step) and to etch & rinse adhesives. The

deeper penetration of the self-cure adhesive might be

explained not only by dentine conditioning, but also

with material permeability and the conduct of the

adhesive component.

The morphological reasons that can impact on resin

penetration in an apical direction are the diameter of

the dentinal tubules, which are larger cervically than

apically (Marion et al. 1991) as well as sclerotic

processes, that hamper the access to the dentinal

tubules (Mjor 1985, Wang & Weiner 1998). It should

be highlighted that resin tag length was significantly

dependent not only on the root region observed, but

also on the area of observation within one slice. Longer

resin tag formation was observed at buccal and lingual

areas to the long axis than at mesial and distal in the

apical and middle regions of the root canal. It is because

of more prominent intratubular calcifications of the

dentinal tubules in the mesial and distal root directions

compared with buccal and lingual (Paque et al. 2006).

The significant difference in the density of the resin

tags was highly dependent on the type of adhesive

system used (P<0.001) and on the root region

(P<0.001). Numerous resin tags might provide a more

durable bond of the post to the root canal dentine

(Bitter et al. 2004) and prevent leakage (Mannocci

et al. 2001). Conditioning of the root canal dentine

with phosphoric acid revealed considerably more resin

tags than observed after the application of self-etching

adhesives. Lowest numbers of infiltrated dentinal

Table 3 Quality of adhesive penetration into the dentinal tubule (%). Completeness (I), continuity (II) and evenness (III) of resin

tags were evaluated using the classification system with scores from 0 to 3 indicating best to worst quality

Position Scale

XP Bond XP Bond/SCA AdheSE

AdheSE/Acti-

vator Hybrid Bond

Hybrid Bond/

hybrid

brushes

I II III I II III I II III I II III I II III I II III

Cervical 0 40 80 75 100 100 85 55 35 0 80 70 5 5 5 0 10 80 0

1 50 10 25 0 0 10 40 60 40 20 30 90 65 80 45 80 20 80

2 10 10 0 0 0 5 5 5 60 0 0 5 30 15 55 10 0 20

3 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

Middle 0 25 70 60 90 90 65 25 5 0 80 30 80 0 0 0 5 40 0

1 65 20 30 10 10 20 30 30 50 15 65 15 30 70 30 45 55 45

2 10 10 10 0 0 15 30 50 55 5 5 5 70 30 70 50 5 55

3 0 0 0 0 0 0 15 15 5 0 0 0 0 0 0 0 0 0

Apical 0 5 20 70 90 80 75 0 0 0 65 50 65 0 0 0 0 30 0

1 45 55 15 10 20 15 50 15 0 15 30 25 20 55 5 65 60 15

2 40 20 10 0 0 10 40 70 90 20 20 10 75 40 90 35 10 85

3 10 5 5 0 0 0 10 15 10 0 0 0 5 5 5 0 0 0



tubules for the self-etch (two-step) adhesive AdheSE,

when compared with etch & rinse XP Bond adhesive

were observed. The reason for this might be that the

multiple-stage adhesive system was able to produce a

more uniform and thick resin-dentine interdiffusion

zone than the self-etching primer (Bitter et al. 2004). In

the apical region, the amount of dentinal tubules that

were not infiltrated with adhesive was more frequent

than in the cervical region. Just like resin tag length,

non-infiltrated dentinal tubules were more frequently

observed at mesial and distal areas than at buccal and

lingual.

The completeness of resin infiltration into dentinal

tubules is responsible for the long-term monomer/

dentine strength and stability (Zicari et al. 2008). In

this study, significant differences were found in the

qualitative evaluation of the resin tags by scores with

reference to the root region, area of observation and

adhesive type (Fig. 3). Despite the deepest penetrations

of Hybrid Bond, the interaction of this adhesive system

with the wall of dentinal tubule was irregular: the

dentinal tubules were incompletely infiltrated and often

interrupted. Only 5% of the dentinal tubules were

infiltrated completely when using Hybrid Bond in the

cervical region, declining to none in the apical region.

The worst completeness and continuity of the resin tags

with various lengths was recorded in the specimens in

which self-etch adhesive AdheSE was applied. The

self-etch (two-step) adhesive AdheSE Primer is charac-

terized by less acidity (pH=1.7), compared with the

self-etch (one-step) Hybrid Bond (pH=1.0). Therefore,

the acidic monomers of the self-etch AdheSE Primer

may not solubilize enough mineral to achieve homog-

enous resin penetration. In addition, the mineral

components from the smear layer may neutralize the

acidity of these self-etch systems (Tay & Pashley 2001).

Non and/or insufficiently resin infiltrated dentinal

tubules may cause the passage of fluids whereas

homogenous resin tags might be responsible for a tight

seal. Application of the conventional adhesive XP

Bond/SCA after dentine etching provided mainly for-

mation of homogeneous and complete tags of similar

length. This was probably caused by the etching effect

that dissolved the smear layer, thus allowing better

access and complete filling of dentinal tubule by the

monomers of the etch & rinse adhesive. Use of the

activator for all types of adhesives significantly

increased the completeness and continuity of resin tags

(P=0.014). A better quality of resin penetration was

observed at buccal and lingual areas to the long root

axis than at mesial and distal.

It has to be considered that bonding to root canal

dentine might be hampered by a lack of direct vision

and luting agent application techniques (D’Arcangelo

et al. 2008). In this study, poorer tag formation in the

apical third might be due to the fact that conditioning

with a microbrush would be better in the cervical

region whereas in the apical third the contact and fluid

exchange with paper cones might be reduced, resulting

in resin penetration less deeply into the tubules. These

findings agree with the results of Ferrari et al. (2002),

who reported that the microbrush promoted a higher

number of resin tags.

Conclusion

The application of the etch & rinse adhesive system

resulted in shorter, but more dense and more complete

resin tags compared with the two- or one-step self-etch

adhesives. There was a significant difference in resin

(a) (b) (c)

Figure 3 CLSM images of the resin tags formed by various types of adhesive systems in root canal under 63·magnification in

fluorescence mode: RT – resin tags, RD – root dentine, HL – hybrid layer, RC – resin composite. (a) Penetration of the adhesive

system Hybrid Bond into the dentinal tubules in the cervical region of the root canal. The resin tags are not homogenously, often

interrupted and not of equal length. (b) Penetration of the adhesive system AdheSE into the dentinal tubules in the middle region of

the root canal. The resin tags have an equal length, are continuous but not homogenous. (c) Penetration of the adhesive system

XP Bond into the dentinal tubules in the apical region of the root canal. The resin tags are not completeness but equal length.



tags formation not only among different regions of the

root canal (cervical, middle and apical), but also in the

direction of dentinal tubules (buccal, lingual, mesial

and distal to the long axis of the root). Use of the

activator for all types of adhesive systems significantly

increased density and quality of resin tags. The clinical

importance of the length of resin tags and/or their

quality in relation to bonding should be further

investigated.

Acknowledgements

The authors would like to thank Prof. Matthias

Folwazcny for permitting the use of the CLS micro-

scope, Dr. Jan-Erik Heil for CLSM technical support and

Dr. Alexander Crispin from IBE for statistical analysis.

References

Bachicha WS, DiFiore PM, Miller DA, Lautenschlager EP,

Pashley DH (1998) Microleakage of endodontically treated

teeth restored with posts. Journal of Endodontics 24, 703–8.

Bitter K, Paris S, Martus P, Schartner R, Kielbassa AM (2004)

A Confocal Laser Scanning Microscope investigation of

different dental adhesives bonded to root canal dentine.

International Endodntic Journal 37, 840–8.

Bitter K, Meyer-Lueckel H, Priehn K, Kanjuparambil JP,

Neumann K, Kielbassa AM (2006) Effects of luting agent

and thermocycling on bond strengths to root canal dentine.


Bitter K, Paris S, Mueller J, Neumann K, Kielbassa AM (2009)

Correlation of scanning electron and confocal laser scan-

ning microscopic analyses for visualization of dentin/adhe-

sive interfaces in the root canal. Journal Adhesive Dentistry

11, 7–14.

Bolhuis P, de Gee A, Feilzer A (2005) The influence of fatigue

loading on the quality of the cement layer and retention

strength of carbon fiber post-resin composite core restora-

tions. Operative Dentistry 30, 220–7.

Bouillaguet S, Troesch S, Wataha JC, Krejci I, Meyer JM,

Pashley DH (2003) Microtensile bond strength between

adhesive cements and root canal dentin. Dental Materials

19, 199–205.

Carrigan PJ, Morse DR, Furst ML, Sinai IH (1984) A scanning

electron microscopic evaluation of human dentinal tubules

according to age and location. Journal of Endodontics 10,

359–63.

Carvalho RM, Pereira JC, Yoshiyama M, Pashley DH (1996) A

review of polymerization contraction: the influence of stress

development versus stress relief. Operative Dentistry 21, 17–

24.

D’Arcangelo C, Zazzeroni S, D’Amario M et al. (2008) Bond

strengths of three types of fibre-reinforced post systems in

various regions of root canals. International Endodntic Journal

41, 322–8.

Duret B, Reynaud M, Duret F (1990) New concept of

coronoradicular reconstruction: the Composipost. Chirurgie

Dentaire Francaise 60, 131–41.

Feilzer AJ, De Gee AJ, Davidson CL (1993) Setting stresses in

composites for two different curing modes. Dental Materials

9, 2–5.

Ferrari M, Davidson CL (1996) In vivo resin-dentin interdif-

fusion and tag formation with lateral branches of two

adhesive systems. Journal of Prosthetic Dentistry 76, 250–3.

Ferrari M, Mannocci F (2000) A ‘one-bottle’ adhesive system

for bonding a fibre post into a root canal: an SEM evaluation

of the post-resin interface. International Endodntic Journal 33,

397–400.

Ferrari M, Cagidiaco MC, Mason PN (1994) Micromorphologic

relationship between resin and dentin in Class II restora-

tions: an in vivo and in vitro investigation by scanning

electron microscopy. Quintessence International 25, 861–6.

Ferrari M, Mannocci F, Vichi A, Cagidiaco MC, Mjor IA

(2000a) Bonding to root canal: structural characteristics of

the substrate. American Journal of Dentistry 13, 255–60.

Ferrari M, Vichi A, Mannocci F, Mason PN (2000b) Retro-

spective study of the clinical performance of fiber posts.

American Journal of Dentistry 13, 9–13.

Ferrari M, Vichi A, Grandini S, Geppi S (2002) Influence of

microbrush on efficacy of bonding into root canals. American

Journal of Dentistry 15, 227–31.

Kremeier K, Fasen L, Klaiber B, Hofmann N (2008) Influence

of endodontic post type (glass fiber, quartz fiber or gold) and

luting material on push-out bond strength to dentin in vitro.


Kurtz JS, Perdigao J, Geraldeli S, Hodges JS, Bowles WR (2003)

Bond strengths of tooth-colored posts, effect of sealer, dentin

adhesive, and root region. American Journal of Dentistry 16,

31A–3A.

Mallmann A, Jacques LB, Valandro LF, Mathias P, Muench A

(2005) Microtensile bond strength of light- and self-cured

adhesive systems to intraradicular dentin using a translu-

cent fiber post. Operative Dentistry 30, 500–6.

Mannocci F, Innocenti M, Ferrari M (1998) Stereomicroscopic

and scanning electron microscopic study of roots obturated

with vertically condensed gutta-percha, epoxy resin cement,

and dentin bonding agent. Journal of Endodontics 24, 397–

400.

Mannocci F, Ferrari M, Watson TF (2001) Microleakage of

endodontically treated teeth restored with fiber posts and

composite cores after cyclic loading: a confocal microscopic

study. Journal of Prosthetic Dentistry 85, 284–91.

Mannocci F, Bertelli E, Watson TF, Ford TP (2003) Resin-

dentin interfaces of endodontically-treated restored teeth.

American Journal of Dentistry 16, 28–32.

Mannocci F, Pilecki P, Bertelli E, Watson TF (2004) Density of

dentinal tubules affects the tensile strength of root dentin.




Marion D, Jean A, Hamel H, Kerebel LM, Kerebel B (1991)

Scanning electron microscopic study of odontoblasts and

circumpulpal dentin in a human tooth. Oral Surgery, Oral

Medicine, Oral Pathology, Oral Radiology and Endodontology

72, 473–8.

Mjor IA (1985) Dentin-predentin complex and its permeabil-

ity: pathology and treatment overview. Journal of Dental

Research. 64, 621–7.

Mjor IA, Nordahl I (1996) The density and branching of

dentinal tubules in human teeth. Archives of Oral Biology 41,

401–12.

Mjor IA, Smith MR, Ferrari M, Mannocci F (2001) The

structure of dentine in the apical region of human teeth.


Nakabayashi N, Nakamura M, Yasuda N (1991) Hybrid layer

as a dentin-bonding mechanism. Journal of Esthetic Dentistry

3, 133–8.

Paque F, Luder HU, Sener B, Zehnder M (2006) Tubular

sclerosis rather than the smear layer impedes dye penetra-

tion into the dentine of endodontically instrumented root

canals. International Endodntic Journal 39, 18–25.

Patierno JM, Rueggeberg FA, Anderson RW, Weller RN,

Pashley DH (1996) Push-out strength and SEM evaluation

of resin composite bonded to internal cervical dentin.


Pegoretti A, Fambri L, Zappini G, Bianchetti M (2002) Finite

element analysis of a glass fibre reinforced composite

endodontic post. Biomaterials 23, 2667–82.

Pioch T, Stotz S, Staehle HJ, Duschner H (1997) Applications

of confocal laser scanning microscopy to dental bonding.

Advances in Dental Research 11, 453–61.

Reid LC, Kazemi RB, Meiers JC (2003) Effect of fatigue testing

on core integrity and post microleakage of teeth restored

with different post systems. Journal of Endodontics 29, 125–

31.

Roberts HW, Leonard DL, Vandewalle KS, Cohen ME, Charlton

DG (2004) The effect of a translucent post on resin

composite depth of cure. Dental Materials 20, 617–22.

Sidhu SK, Watson TF (1998) Interfacial characteristics of

resin-modified glass-ionomer materials: a study on fluid

permeability using confocal fluorescence microscopy. Jour-

nal of Dental Research 77, 1749–59.

Tay FR, Pashley DH (2001) Aggressiveness of contemporary

self-etching systems. I: depth of penetration beyond dentin

smear layers. Dental Materials 17, 296–308.

Tay FR, Suh BI, Pashley DH, Prati C, Chuang SF, Li F (2003)

Factors contributing to the incompatibility between simpli-

fied-step adhesives and self-cured or dual-cured composites.

Part II. Single-bottle, total-etch adhesive. Journal of Adhesive


Tay FR, Loushine RJ, Lambrechts P, Weller RN, Pashley DH

(2005) Geometric factors affecting dentin bonding in root

canals: a theoretical modeling approach. Journal of Endodon-

tics 31, 584–9.

Titley K, Chernecky R, Chan A, Smith D (1995) The

composition and ultrastructure of resin tags in etched

dentin. American Journal of Dentistry 8, 224–30.

Wang RZ, Weiner S (1998) Strain-structure relations in

human teeth using Moire fringes. Journal of Biomechanics

31, 135–41.

Yoshiyama M, Matsuo T, Ebisu S, Pashley D (1998) Regional

bond strengths of self-etching/self-priming adhesive systems.

Journal of Dentistry 26, 609–16.

Zicari F, Couthino E, De Munck J et al. (2008) Bonding

effectiveness and sealing ability of fiber-post bonding. Dental

Materials 24, 967–77.



Haemostatic effect and tissue reactions ofmethods and agents used for haemorrhagecontrol in apical surgery

S. S. Jensen1,2, P. M. Yazdi3, E. Hjørting-Hansen1,4, D. D. Bosshardt1 & T. von Arx1

1Department of Oral Surgery and Stomatology, School of Dental Medicine, University of Bern, Bern, Switzerland; 2Department of

Oral & Maxillofacial Surgery, Copenhagen University Hospital, Glostrup, Denmark; 3Department of Oral & Maxillofacial Surgery,

School of Dental Medicine, University of Arhus, Arhus, Denmark; and 4Department of Oral & Maxillofacial Surgery, School of

Dentistry, Faculty of Health Sciences, University of Copenhagen, Copenhagen, Denmark

Summary

Jensen SS, Yazdi PM, Hjørting-Hansen E, Bosshardt DD,

von Arx T. Haemostatic effect and tissue reactions of methods

and agents used for haemorrhage control in apical surgery.


Aim To compare the haemostatic effect and tissue

reactions of different agents and methods used for

haemorrhage control in apical surgery.

Methodology Six standardized bone defects were

prepared in the calvaria of six Burgundy rabbits. Five

haemostatic modalities were tested for their haemo-

static effect and tissue reactions, and were compared

with untreated control defects: ExpasylTM + Stasis�,

ExpasylTM + Stasis� + freshening of the bone defect

with a bur, Spongostan�, Spongostan� + epinephrine,

and electro cauterization. The haemostatic effect was

analysed visually and compared using Wilcoxon’s

signed rank test. Two groups of three animals were

evaluated histologically for hard and soft tissue reac-

tions related to the different haemostatic measures,

after 3 and 12 weeks of healing respectively.

Results ExpasylTM + Stasis� and electro cauteriza-

tion proved most effective in reducing bleeding

(P < 0.05), but were accompanied by unfavourable

tissue reactions, as indicated by the presence of necrotic

bone, inflammatory cells and the absence of bone

repair. These adverse tissue reactions did not recover

substantially over time. However, adverse reactions

were not observed when the superficial layer of bone

had been removed with a rotary instrument. In

contrast, Spongostan� + epinephrine showed only a

moderate haemostatic effect, but elicited also only mild

adverse tissue reactions.

Conclusions Haemostasis in experimental bone

defects is most effectively accomplished by using

ExpasylTM + Stasis� or electro cauterization. However,

the bone defects should be freshened with a rotary

instrument before suturing so as not to compromise

healing.

Keywords: animal study, apical surgery, haemor-

rhage control, haemostatic agent.

Received 15 July 2009; accepted 25 August 2009

Introduction

Haemorrhage control is important in apical surgery to

facilitate inspection of the root-end surface and to allow

placement and setting of the root-end filling. Usually,

one or more local agents are needed to achieve

sufficient haemostasis. These agents should be either

removed completely or should be fully biocompatible

and degrade without interfering with periapical healing.

The haemostatic effect and tissue reactions of bone

wax, ferric sulphate (Stasis�, Belport Co, Camarillo, CA,

USA), and an aluminium chloride-containing paste

(ExpasylTM, Pierre Rolland, Merignac, France) intended

for application into the sulcus prior to impression-taking

has been reported (von Arx et al. 2006). The combina-

tion of Stasis� and ExpasylTM or ExpasylTM alone proved

most efficient in controlling haemorrhage. However,

their use was accompanied by an inflammatory and a

Correspondence: Dr Simon Storgard Jensen, Department of

Oral & Maxillofacial Surgery, Copenhagen University Hospital

Glostrup, Ndr. Ringvej, DK-2600 Glostrup, Denmark (Tel.:

+45 43 23 32 08; fax: +45 43 23 39 63; e-mail:

[email protected]).

doi:10.1111/j.1365-2591.2009.01637.x


foreign body tissue reaction at the histological level.

Based on these findings, the clinical use of ExpasylTM has

been modified to include freshening the bony surface of

the periapical crypt with a bur after placement and

initial setting of the root-end filling material.

Resorbable gelatin-based sponges, such as Spongo-

stan�, Spongostan�, Dental, Johnson & Johnson

medical Ltd., Ascot, UK, are frequently used for

haemostasis in several surgical specialties (Petersen

et al. 1984, Finn et al. 1992, Schonauer et al. 2004).

Spongostan� can be used alone, but is often combined

with a vasoconstrictor to enhance the haemostatic

effect (Rud et al. 2001). Tissue reactions to Spongo-

stan� are generally considered to be mild (Alpaslan

et al. 1997). However, when Spongostan� is left inside

osseous defects, delayed healing has been reported

(Liening et al. 1997, Schonauer et al. 2004). It is not

known how the tissue reacts if the gelatin sponge is

removed after haemorrhage control has been

achieved.

Electro cauterization is an effective method for

producing haemostasis by coagulation and vesicular

clumping. Most often, electro cauterization is used to

stop localized bleeding in the soft tissues, but it has also

been reported to be efficient when used on oozing bone

surfaces (Jensen et al. 2002). With this approach, no

foreign substance is introduced into the bony crypt.

However, concern has been raised about the influence

on healing due to the thermal damage to the bone

tissue (Eriksson et al. 1982).

The purpose of the present study was twofold:

• To compare the haemostatic effects of ExpasylTM +

Stasis�, Spongostan�, Spongostan� + epinephrine and

electro cauterization in standardized bone defects.

• To evaluate the tissue reactions after using Expa-

sylTM + Stasis� with and without freshening of the

bone defect with a bur, after electro cauterization, and

after using Spongostan� alone or in combination with

epinephrine.

Material and methods

Study design

Approval to perform the study was granted by the

authorities of the Canton of Bern, Department of

Agriculture, Section Veterinary Service, Experimental

Animal Studies (study number 100/06). The experi-

mental study was conducted in six adult Burgundy

rabbits, each at least 5 months old and weighing

between 3 and 4.5 kg.

The surgical procedures were performed under intra-

venous general anaesthesia using the medication and

surgical protocol presented by von Arx et al. (2006).

In each rabbit, six standardized monocortical bone

defects were created in the calvarium. The defects were

prepared using a trephine with an outer diameter of

4 mm. The depth of the defects depended on the

thickness of the outer cortical bone layer. Each defect

then received one of the following treatments in a

randomized sequence (Fig. 1), with a randomization

scheme generated using http://www.randomiza-

tion.com (seed: 2604):

• Control: no haemostatic agent was placed.

• ExpasylTM and Stasis�: ExpasylTM (Pierre Rolland,

Merignac, France) was placed into the bone defect with

a spatula, flush with the adjacent outer cortex; after

2 min the paste was removed with a dental curette.

(a) (c)(b)

Figure 1 a) Standardized monocortical bone defects in the rabbit calvarium before application of haemostatic agents. Example of

photograph used for visual assessment of initial bleeding score. b) Schematic illustrations used for visual assessment of bleeding. c)

Presentation after application of haemostatic agents. Example of photograph used for visual assessment of final bleeding score.

Haemostasis in apical surgery Jensen et al.


Subsequently, a small sponge soaked with Stasis�

(Belport Co, Camarillo, CA, USA) was placed for 5 s

into the bone defect.

• ExpasylTM and Stasis� with freshening of the bone

defect: ExpasylTM and Stasis� were applied as described

above. Before primary closure, the bone defect was

freshened using a small round bur (Ø: 1.2 mm) under

copious saline irrigation to remove all macroscopically

visible remnants of ExpasylTM.

• Spongostan�: A Spongostan� sponge, 1 cm3

(Spongostan� Dental; Johnson & Johnson Medical Ltd.,

Ascot, UK) was compressed into the defect for 2 min

using a gauze tampon, and then removed.

• Spongostan� and epinephrine: A Spongostan�

sponge 1 cm3 was soaked in three drops of epinephrine

1%, compressed into the defect for 2 min using a gauze

tampon, and then removed.

• Electro cauterization: Any visible bleeding within the

defect was cauterized using a spatula-shaped cauteriza-

tion head (straight, 2.35 · 19 mm) (ERBOTOM ICC,

ERBE Swiss AG, Winterthur, Switzerland. Setting: Soft

coagulation 60 Watt). Before primary closure, the defect

was curetted using a surgical spoon.

Sacrifice

One group of three animals was allowed to heal for

3 weeks, and a second group of three animals for

12 weeks. Following each designated healing period,

sacrifice was performed as previously described (von

Arx et al. 2006). The retrieved calvarial specimens

were immediately immersed in a solution of 4%

formaldehyde and 1% calcium chloride.

Histological analysis

The non-decalcified specimens were embedded in

methyl-methacrylate and stained with combined basic

fuchsin and toluidine blue. Transversal sections with a

thickness of approximately 80 lm were obtained for

descriptive histology (Schenk et al. 1984). The histo-

logic examination for the description of qualitative

tissue reactions included absence or presence of (i)

remnants of anticoagulation agents; (ii) new bone; (iii)

necrotic bone; (iv) an inflammatory cell infiltrate; and

(v) multinucleated giant cells.

Visual analysis of haemostatic effect

Photos were taken before application and after removal

of the haemostatic agents (Fig. 1). The amount of blood

per site was assessed on a scale from 0 (completely dry

defect) to 7 (profuse bleeding from the defect) (von Arx

et al. 2006). Three evaluators independently examined

the photos and determined the bleeding score per site. A

mean bleeding score was calculated per treatment for the

different sites before application (=initial score) and after

removal (=final score) of the haemostatic agents. The

difference between the two means determined the mean

haemostatic effect per agent (reduction of bleeding).

Statistics

The results of the visual analysis of haemostatic effect

were compared using Wilcoxon’s signed rank test for

paired samples. Exact two-sided P-values were com-

puted to detect differences between the various treat-

ment options. As pair wise comparisons were

completed on the same data, the P-values would have

needed to be adjusted to compensate for the multiple

testing situation. However, because of the explorative

nature of the study and the small sample size, no

adjustment was carried out. Cohen’s weighted kappa

values were calculated to evaluate inter-observer vari-

ations (Fleiss & Cohen 1973).

Results

One animal in the 12-week group died immediately

postoperatively because of an anaesthetic complication.

An additional animal was therefore included, resulting

in seven animals being included in the visual evalua-

tion of the haemostatic effect. Another animal in the

12-week group died 7 weeks postoperatively. The

calvarium of this animal was evaluated histologically

and was found to demonstrate tissue reactions compa-

rable to the two remaining animals in the 12-week

group. This animal was therefore included in the

qualitative histologic evaluation.

Haemostatic effect

In the visual quantification of the bleeding, there was

strong agreement between the three observers

(weighted kappa values: 0.71 to 0.98) (Fleiss & Cohen

1973). The initial bleeding scores, final bleeding scores,

and reduction of bleeding for the individual test groups

are presented in Table 1. Pair wise comparisons of the

different test groups regarding final bleeding score and

bleeding reduction are given in Tables 2 and 3.

No significant differences were found between the

initial mean bleeding scores for the different treatment

Jensen et al. Haemostasis in apical surgery


modalities (P between 0.14 and 1.00). The mean final

bleeding score was significantly smaller final than

the ‘control’ for all groups except ‘Spongostan�’

(P = 0.798). ‘ExpasylTM + Stasis�’ and ‘ExpasylTM +

Stasis� + freshening’ both had significantly smaller

mean bleeding scores than ‘Spongostan�’ and ‘Spongo-

stan� + epinephrine’ (P < 0.05). ‘Electro cauterization’

exhibited borderline significantly smaller final mean

bleeding scores than ‘Spongostan�’ and ‘Spongostan� +

epinephrine’ (P = 0.059 and P = 0.051, respectively).

With regard to the mean bleeding reduction scores, all

but the two ‘Spongostan�’ groups (P = 0.866 and

P = 0.295, respectively) demonstrated significantly

higher bleeding reduction than the control defects

(P < 0.05). ‘ExpasylTM + Stasis�’ showed borderline

significantly higher bleeding reduction than ‘Spongo-

stan�’ (P = 0.050). ‘ExpasylTM + Stasis� + freshening’

reduced the bleeding significantly more than ‘Spongo-

stan�’ (P = 0.031) and borderline significantly more

than ‘Spongostan� + epinephrine’ (P = 0.051). Electro

cauterization resulted in significantly higher bleeding

reduction than ‘Spongostan�’ (P = 0.034).

Histology

No attempt was made to preserve the volume of the

original bone defects by covering them with a barrier

membrane. Therefore, herniation of soft tissues into the

defects was a frequent finding, irrespective of the

haemostatic agent applied (Fig. 2).

Control sites

3 Weeks: Vivid bone formation was observed extending

from the defect walls. Ongoing osteogenic activity was

observed throughout the defects with woven bone

trabeculae with osteoblastic seams.

12 Weeks: Maturation of the newly formed bone was

generally observed (Fig. 2). However, pressure from the

covering soft tissue caused surface resorption of some of

the newly formed bone.

Table 1 Mean bleeding scores ( ± SD)

before application and after removal of

haemostatic agents, and mean reduc-

tion in bleeding scores (n = 7)

Mean initial

score ( ± SD)

Mean final

score ( ± SD)

Mean

reduction ( ± SD)

Control 3.81 ( ± 1.68) 3.33 ( ± 1.60) 0.48 ( ± 1.87)

ExpasylTM + Stasis� 3.24 ( ± 0.83) 0.43 ( ± 0.29) 2.81 ( ± 0.71)

ExpasylTM + Stasis� +

freshening of defect

4.24 ( ± 1.74) 0.29 ( ± 0.21) 3.95 ( ± 1.84)

Electro cauterization 4.57 ( ± 2.12) 1.05 ( ± 0.90) 3.52 ( ± 1.74)

Spongostan� 3.62 ( ± 1.33) 3.10 ( ± 0.90) 0.52 ( ± 1.32)

Spongostan� + epinephrine 3.86 ( ± 1.59) 2.24 ( ± 0.92) 1.62 ( ± 0.93)

Table 2 P-values of pairwise comparisons of the final bleeding scores using Wilcoxon’s signed rank test

Control

ExpasylTM

+ Stasis�ExpasylTM + Stasis�

+ freshening of defect

Electro

cauterization Spongostan�

Expasyl + Stasis� 0.036 – – – –

Expasyl + Stasis� +

freshening of defect

0.022 0.586 – – –

Electro cauterization 0.034 0.100 0.181 – –

Spongostan� 0.798 0.022 0.016 0.059 –

Spongostan� + epinephrine 0.034 0.031 0.016 0.051 0.104

Table 3 P-values of pairwise comparisons of the scores of calculated bleeding reduction using Wilcoxon’s signed rank test

Control

ExpasylTM

+ Stasis�ExpasylTM + Stasis�


Electro

cauterization Spongostan�

ExpasylTM + Stasis� 0.036 – – – –

ExpasylTM + Stasis�


0.031 0.106 – – –

Electro cauterization 0.031 0.444 0.400 – –

Spongostan� 0.866 0.050 0.031 0.034 –

Spongostan� + epinephrine 0.295 0.141 0.051 0.150 0.204



ExpasylTM and Stasis�

3 Weeks: The entire bone surface was necrotic, without

signs of repair activity. The osteocyte lacunae were

empty along the surfaces of the cavities, and numerous

macrophages and multinucleated giant cells were

observed close to the defect walls and to remnants of

ExpasylTM, which were particularly evident in concav-

ities of the defect wall.

12 Weeks: Necrotic areas could still be identified

within the defect walls with areas of undermining

resorption, nearly forming a sequester (Fig. 3). Rem-

nants of ExpasylTM were often observed, and were

always surrounded by an extensive foreign body

reaction (Fig. 3).

ExpasylTM and Stasis� with freshening of the bone defect

3 Weeks: Moderate amounts of woven bone formation

were observed close to the defect walls, particularly at

sites where the bone marrow spaces were opened

widely. Remnants of ExpasylTM were seen rarely.

However, if present, they were always accompanied

by multinucleated giant cells.

12 Weeks: The amount of osseous repair was

limited, but no signs of necrotic bone or foreign body

reaction were present.

Spongostan�

3 Weeks: Woven bone formation was observed along

the entire surface of the bone cavities. However, the

osteoblastic activity along the defect surfaces was

reduced compared with the control defects.

12 Weeks: The defects were dominated by matura-

tion of the newly formed bone, with layers of parallel-

fibered bone reinforcing the woven bone trabeculae.

Spongostan� and epinephrine

3 Weeks: Vivid new bone formation could be seen, with

woven bone throughout the former bone defect

(Fig. 4). Ongoing osteogenic activity, as indicated by

the presence of osteoid seams and osteoblasts, was a

dominating feature.

12 Weeks: Maturation of woven bone formed earlier

was seen, with only sparse signs of ongoing bone

formation. In general, the surfaces demonstrated

lamellar bone coverage.

Figure 2 Control defect after 12 weeks of healing. Most of the

original defect volume (marked with lines) is occupied by soft

tissue with the character of loose connective tissue dominated

by epidermal fat cells. Bone formed form the defects walls has

the character of mature lamellar bone.

Figure 3 Defect 12 weeks after application of ExpasylTM +

Stasis� without freshening the defect with a round bur.

Remnants of ExpasylTM (E) are surrounded by lots of multi-

nucelated giant cells (large arrows). Empty osteocyte lacunae

can be see in a bone trabecula undermined by resorption

cavities (Small arrows).

Figure 4 Defect three weeks after application of Spongostan�.

The former defect (marked with lines) is completely occupied

by newly formed woven bone. The ePTFE suture used to close

the periosteal layer can be recognized (asterisks).



Electro cauterization

3 Weeks: The surfaces of the entire cavities appeared

necrotic, without any signs of bone repair. The

osteocyte lacunae near to the surface were empty

(Fig. 5). The cell populations adjoining the surfaces

consisted almost exclusively of inflammatory cells,

mainly macrophages and some multinucleated giant

cells. In addition, many erythrocytes were often seen on

the cauterized bone surface. The opened bone marrow

often revealed signs of degeneration.

12 Weeks: The bone surface appeared vital, but

jagged after extensive osteoclastic activity. The size and

shape of the cavities were nearly identical to the

situation immediately postoperatively with very limited

signs of new bone formation.

Discussion

This study evaluated the haemostatic efficacy and

tissue reactions of different methods for local haemor-

rhage control used in apical surgery. Overall, Expa-

sylTM + Stasis� and electro cauterization proved most

efficient in reducing bleeding, while the use of Spongo-

stan� alone did not demonstrate any significant hae-

mostatic effect as compared with the control defects.

The addition of epinephrine to the Spongostan� had

some effect on the final bleeding score, without

reaching the efficacy of ExpasylTM + Stasis�.

The efficacy of ExpasylTM + Stasis� in reducing

bleeding was in accordance with a previous study

using the same model (von Arx et al. 2006). However,

a concern was the localized foreign body reaction

elicited by remnants of ExpasylTM in the bone defects.

Only limited documentation exists on tissue reactions

to aluminium chloride in paste form, but studies which

have evaluated topical application of aluminium chlo-

ride in liquid form have also reported inflammatory

reactions (Barr et al. 1993, Kopac et al. 2002). The

results of the present study suggest that these tissue

reactions can be significantly reduced by freshening the

defect with a bur. In the clinical case of using a root-

end filling material that does not set during the surgical

procedure [e.g. mineral trioxide aggregate (MTA)],

there is a risk of flushing out the material during the

freshening of the bony cavity. To avoid this, it has

proved important to use a relatively small round bur to

reduce the risk of touching the cut root surface and

MTA filling, and to prevent direct water spray on the

cut root surface.

Spongostan� is widely used in several surgical

specialties to control bleeding, but most often in

surgical sites where it can be left in situ, such as in

dental extraction sockets or in donor sites after bone

graft harvesting (Petersen et al. 1984, Finn et al. 1992,

Blinder et al. 1999). In the present study, the sponge

was removed after 2 min to conform with the typical

protocol in apical surgery (Rud et al. 2001). This

eliminated the compressive element, and the resulting

intrinsic haemostatic effect proved to be limited. Histo-

logic analysis revealed slightly delayed bone healing

that was qualitatively comparable to the control

defects. Similar findings have been reported in previous

experimental and clinical studies, where the gelatin

sponges were left in situ (Petersen et al. 1984, Finn

et al. 1992). Addition of epinephrine 1% to a gelatin

sponge or cotton pellet has been reported to provide

sufficient haemostasis to allow undisturbed placement

and setting of dentine-bonded composite resin root-end

fillings (Rud et al. 2001, Jensen et al. 2002). In the

present experimental setting, the addition of epineph-

rine to Spongostan� only marginally increased the

haemostatic effect, without reaching statistical signifi-

cance. No histologic difference in healing pattern was

observed with the addition of epinephrine. As epineph-

rine is a naturally occurring circulating hormone in the

organism, disturbance of healing was not to be

expected.

Electro cauterization provided a haemostatic effect

similar to that of ExpasylTM + Stasis�. However, bone

healing was delayed when compared to control defects

and to ExpasylTM + Stasis�-treated defects that were

Figure 5 Defect three weeks after using electro cauterization

for haemorrhage control. Multiple empty osteocyte lacunae

(small arrows) can be seen close to the defect walls (large

arrows). The soft tissue occupying the defect has the character

of immature granulation tissue. A coagulum containing many

erythrocytes is observed along the defect walls.



freshened with a bur. Limited bone formation was

observed after 12 weeks of healing, following initial

signs of superficial necrosis. This can presumably be

ascribed to thermal injury, as has previously been

documented (Eriksson et al. 1982). In the early healing

phase, an adverse tissue reaction was seen in relation

to the necrotic zones. This inflammatory and foreign

body reaction was not observed after 12 weeks of

healing. Coagulated tissue remnants were removed

with a curette before suturing. It can be speculated that

the osseous healing conditions could have been

improved by removing the superficial bone layer with

a rotary instrument, as was observed with the fresh-

ened ExpasylTM + Stasis�-treated defects.

Conclusion

ExpasylTM + Stasis� and electro cauterization proved

most efficient in the reduction of bleeding from stan-

dardized bone defects. However, the same measures

were accompanied by the most pronounced adverse

tissue reactions. It is recommended to thoroughly

remove the superficial layer in the bone defect with a

rotary instrument after application of ExpasylTM + Sta-

sis� or electro cauterization in apical surgery to reduce

these unfavourable tissue reactions. Defects treated

with Spongostan� demonstrated no adverse tissue

reactions but delayed bone healing. Despite the reduced

haemostatic effect of Spongostan� and epinephrine

compared with ExpasylTM + Stasis� and electro cau-

terization, this combination will often clinically ensure

sufficient haemostasis for the undisturbed placement

and setting of a root-end filling.

Acknowledgements

The authors gratefully acknowledge the assistance of

Dr. med. vet. Daniel Mettler and the veterinarian

team of the Department of Experimental Surgery,

Bern University Hospital, Inselspital, Bern, Switzer-

land. We also thank Mrs. Britt Hoffmann and Mr.

David Reist, Department of Oral Surgery and Stoma-

tology, University of Bern, for the histologic prepara-

tion of the specimens. The statistical assistance of Mr.

D. Klingbiel, Institute of Mathematical Statistics and

Actuarial Science, University of Bern, was highly

appreciated.

Funding: The study was generously funded by a

grant from the Foundation for Dental Research and

Education (FDR), Basel, Switzerland (Grant 1–04/17).

The authors declare no conflicts of interest.

References

Alpaslan C, Alpaslan GH, Oygur T (1997) Tissue reaction to

three subcutaneously implanted local hemostatic agents.

British Journal of Oral and Maxillofacial Surgery 35, 129–32.

von Arx T, Jensen SS, Hanni S, Schenk RK (2006) Haemo-

static agents in periradicular surgery: an experimental study

of their efficacy and tissue reactions. International Endodontic

Journal 39, 800–8.

Barr RJ, Alpern KS, Jay S (1993) Histiocytic reaction associ-

ated with topical aluminium chloride (Drysol reaction). The

Journal of Dermatologic Surgery and Oncology 19, 1017–21.

Blinder D, Manor Y, Martinowitz U, Taicher S (1999) Dental

extractions in patients maintained on continued oral

anticoagulant comparison of local hemostatic modalities.

Oral Surgery Oral Medicine Oral Pathology Oral Radiology and


Eriksson A, Albrektsson T, Grane B, McQueen D (1982)

Thermal injury to bone A vital-microscopic description of

heat effects. International Journal of Oral Surgery 11, 115–22.

Finn MD, Schow SR, Schneiderman ED (1992) Osseous

regeneration in the presence of four common hemostatic

agents. Journal of Oral and Maxillofacial Surgery 50, 608–12.

Fleiss JL, Cohen J (1973) The equivalence of weighted kappa

and the intraclass correlation coefficient as measures of

reliability. Educational and Psychological Measurement 33,

613–9.

Jensen SS, Nattestad A, Egdø P, Sewerin I, Munksgaard EC,

Schou S (2002) A prospective, randomized, comparative

clinical study of resin composite and glass ionomer cement for

retrograde root filling. Clinical Oral Investigations 6, 236–43.

Kopac I, Cvetko E, Marion L (2002) Gingival inflammatory

response induced by chemical retraction agents in beagle

dogs. International Journal of Prosthodontics 15, 14–9.

Liening DA, Lundy L, Silberberg B, Finstuen K (1997) A

comparison of the biocompatibility of three absorbable

hemostatic agents in the rat middle ear. Otolaryngology

Head and Neck Surgery 16, 454–7.

Petersen JK, Krogsgaard J, Nielsen KM, Nørgaard EB (1984) A

comparison between two absorbable hemostatic agents:

gelatin sponge (Spongostan) and oxidized regenerated

cellulose (Surgicel). International Journal of Oral Surgery 13,

406–10.

Rud J, Rud V, Munksgaard EC (2001) Periapical healing of

mandibular molars after root-end sealing with dentine-

bonded composite. International Endodontic Journal 34, 285–

92.

Schenk RK, Olah AJ, Herrmann W (1984) Preparation of

calcified tissues for light microscopy. In: Dickson GR, ed.

Methods of Calcified Tissue Preparation. Amsterdam, Nether-

lands: Elsevier, pp. 1–56.

Schonauer C, Tessitore E, Barbagallo G, Albanese V, Moraci A

(2004) The use of local agents: bone wax, gelatin, collagen,

oxidized cellulose. European Spine Journal 13(Suppl. 1), S89–

96.



Volumetric analysis of root fillings using spiralcomputed tomography: an in vitro study

R. Anbu, S. Nandini & N. VelmuruganDepartment of Conservative Dentistry & Endodontics, Meenakshi Ammal Dental College, Maduravoyal, Chennai, Tamilnadu, India

Abstract

Anbu R, Nandini S, Velmurugan N. Volumetric analysis of

root fillings using spiral computed tomography: an in vitro

study. International Endodontic Journal, 43, 64–68, 2010.

Aim To analyse volumetrically using spiral computed

tomography (SCT) the efficacy of various techniques to

fill root canals.

Methodology Root canals in 40 maxillary central

incisors were instrumented with K-files to size 60 and

the volume of the canal measured using SCT. The teeth

were divided into four groups of 10 each and root filled

by lateral compaction, Thermafil, Obtura II and System

B techniques, respectively. AH plus was used as sealer

with all techniques. The filled volume in each canal

was measured using SCT and the percentage of

obturated volume (POV) was calculated. The data were

statistically analysed using Kruskal–Wallis test and

Mann–Whitney U-test.

Results The four groups were comparable in canal

volume. The overall POV was 80.4%, 93.3%, 84.8%

and 93.7% for lateral compaction, Thermafil, Obtura II

and System B, respectively (P < 0.05).

Conclusion The greatest POV was obtained with

System B and Thermafil. Voids were seen in all root

fillings.

Keywords: Obtura II, obturation, percentage of

obturated volume, spiral CT, three dimensional, volu-

metric analysis.

Received 9 May 2009; accepted 25 August 2009

Introduction

A good root filling is essential to prevent bacteria and/

or their by-products from reaching the periapical

region. There are various methods available for filling

canals ranging from cold lateral compaction to

thermoplasticized techniques. Various experimental

methods have been used to assess the quality of root

fillings, such as radioisotope (Haıkel et al. 2000), dye

penetration (McRobert & Lumley 1997, Venturi 2006),

fluid filtration (Kontakiotis et al. 2007), bacterial

leakage (Jacobson et al. 2002), microscopic analysis

(De-Deus et al. 2006, 2008), clearing techniques

(Oliver & Abbott 2001) and Micro CT (Hammad et al.

2009). It has been reported that spiral computerized

tomography (SCT) has been a useful tool in various

in vivo and laboratory studies. It was concluded that

with SCT three-dimensional volume measurements are

possible without sectioning specimens thus avoiding

loss of material (Nandini et al. 2006, Hammad et al.

2009).

The aim of this study was to assess the efficacy of

different root filling techniques by calculating the

percentage of obturated volume (POV).


Tooth specimens

Forty single rooted maxillary central incisors were

selected. Soft tissue remnants and calculus were

removed. Collection, storage, sterilization and handling

of extracted teeth followed the Occupational Safety and

Health Administration (OSHA) guidelines and regula-

tions (Reuben et al. 2008).

Correspondence: Dr Natanasabapathy Velmurugan, (Prof and

Head), Department of Conservative Dentistry & Endodontics,

Meenakshi Ammal Dental college, Allapakkam main road,

Maduravoyal, Chennai 95, Tamil nadu, India (Tel.:

09840164167; fax: 91 44 23781631; e-mail: Vel9911@

yahoo.com).

doi:10.1111/j.1365-2591.2009.01638.x


All specimens were checked for number and curva-

ture of root canals radiographically. Teeth with single

straight canals were chosen whilst teeth with incom-

pletely formed apices, calcified canals, fractures or

resorption were excluded.

Tooth preparation

An access cavity was prepared in each tooth and a size

10 K-file introduced until the tip was just visible at the

foramen of the root. The working length was derived by

subtracting 0.5 mm from the measured length. Orifice

enlargement was achieved with sizes 2, 3 and 4 Gates

Glidden drills (Dentsply Maillefer, Ballaigues, Switzer-

land). Apical enlargement was completed to size 60

with K-files and a step-back procedure was used until

the preparation blended with the coronal flaring. Two

millilitres of 1% sodium hypochlorite and 2 mL of

saline were used as irrigants between each file size.

Ethylene-Diamine Tetraacetic acid (EDTA) paste (RC

Help; Prime Dental Products, Thane, India) was used as

a lubricant during instrumentation. Before filling, root

canals were dried using a size 60 paper point.

The teeth were numbered from 1 to 40. Specimens

were scanned using a Light Speed VCT Scanner (GE

Electricals, Milwaukee, WI, USA). They were then

viewed under high resolution, both cross-sectionally

and longitudinally with a constant thickness of

0.625 mm/slice and a constant spiral or table speed

of 0.5 and 140 kVp. The scanned data was then

transferred to Advantage windows work station

(GE system, Milwaukee, WI, USA) image analysis and

evaluated. The area of prepared root canal in each slice

was measured from cemento-enamel junction (CEJ) to

the apex of the root. The volume of root canal in each

slice was calculated by multiplying the root canal area

by the slice thickness (0.625 mm). Finally, the volume

of each canal was calculated.

The root length was divided into three equal parts;

coronal, middle and apical thirds and the volume of

each segment was calculated separately.

Forty teeth were randomly divided into four groups

of 10 each.

In all the teeth AH Plus sealer (Dentsply Maillefer)

was placed into the canal using a lentulo-spiral filler.

Group-I: lateral compaction (LC)

A size 60 gutta-percha (GP) (Dentsply Maillefer) was

coated with the sealer and placed in the canal to

working length with tug-back. Lateral compaction was

achieved using additional accessory GP cones and

standardized finger spreaders (Dentsply Maillefer) start-

ing 1 mm short of working length. When the points

prevented the spreader penetration beyond the coronal

third of the canal, the canal was considered to be

adequately filled and excess GP was removed at the CEJ

using a heated condenser. The GP at the CEJ was

compacted using a cold plugger.

Group-II: Thermafil

A 60 size Thermafil (Dentsply Maillefer) verifier was

used to check the size of the canal. A 60 size Thermafil

cone was heated (ThermaPrep Plus Oven, Tulsa Dental

Products, OK, USA) according to the manufacturer’s

instruction and introduced into each canal using firm

apical pressure within 0.5 mm short of working length.

An inverted cone bur was used to cut the plastic shaft

1–2 mm within the access cavity. Excess GP was

removed using pluggers (Gencoglu et al. 2008).

Group-III: Obtura II

A 20 gauge Obtura (Obtura Spartan, Fenton, MI, USA)

needle tip was selected. The tip was inserted into the

canal 3–5 mm short of the working length. The

temperature was set at 200 �C, the trigger was pressed

so that the molten GP flowed and the tip was

withdrawn slowly out of the canal. The apical segment

was compacted using appropriate Obtura pluggers.

Backfilling was achieved by the application of thermo-

plasticized GP in 4–5-increments, followed by uniform

compaction with pluggers.

Group-IV: System B

A medium-large nonstandardized GP cone was placed

to within 0.5 mm of the working length. A medium-

large System B (EIE/Analytic Technology, Redmond,

WA, USA) insert tip, which bound in the canal 3 mm

from the working length, was used for the down-pack.

The heat was preset to 200 �C during the first down-

pack. An accessory GP cone was then placed into the

canal and the heat was preset to 100 or 250 �C during

the second down-pack. This procedure was repeated

until the entire root canal was filled (Gencoglu et al.

2008).

A second SCT scan was performed to determine the

volume of GP and sealer. POV in each tooth was

calculated. POV in the coronal, middle and apical thirds

for each tooth was calculated separately (Fig. 1).

Anbu et al. Volumetric analysis of root fillings


Statistical analysis

Statistical analysis was performed with nonparametric

tests (Kruskal–Wallis and Mann–Whitney U-test). The

software used was SPSS (Statistical Package for Social

Sciences, Chicago, IL, USA) version 11.5. The level of

significance was set at P < 0.05.

Results

The volume (mean & standard deviation) of root canals

after cleaning and shaping for each group is given in

Table 1. The four groups were statistically comparable

in respect of canal volume (P > 0.05).

All POV values are summarized in Table 2. The

System B and Thermafil gave the highest POV values

and were significantly different in comparison with

Obtura II and lateral compaction (P < 0.05).

Discussion

Traditional methods of evaluating root fillings have

disadvantages. On sectioning the root, there could be

loss of material which might mimic voids. Radiographs

give only two-dimensional interpretations (Robinson

et al. 2002). The time taken for fluid filtration (Pommel

& Camps 2001) and clearing techniques (Oliver &

Abbott 2001) may be a concern. Dye penetration

studies do not correlate clinically (Oliver & Abbott

2001) whereas dye extraction studies evaluate only the

apical third of the tooth (Camps & Pashley 2003).

Bacterial leakage studies do not simulate exact clinical

conditions, need long periods of observation and do not

allow quantification of the number of penetrating

bacteria (Siqueira et al. 1999, 2000).

A literature search revealed that only sectioning

studies have been undertaken to assess thermoplasti-

cized root fillings at various levels (Jung et al. 2003,

ElAyouti et al. 2005, De-Deus et al. 2006). SCT, a

noninvasive technique gives a 3D interpretation (Nair

& Nair 2007) at various levels, avoids loss of material

(Hammad et al. 2009), yields reproducible results and

the specimens can be used for further research. The

specific location of voids can be determined accurately.

Recent studies have proved that SCT provides volu-

metric analysis of root fillings and remaining remnants

of root fillings (Bartletta et al. 2008, Hammad et al.

2008). Hence, SCT was chosen as the tool for inves-

tigating the efficacy of fillings in this study. The only

limitation of SCT is that it is difficult to differentiate GP

and sealer.

In this study, Thermafil and System B gave the

highest overall POV. The reason could be that the use

of heat softened GP had created a better homogenous

mass with less voids and better adaptation of the GP

to the canal wall. This is in accordance to the studies

reported by De-Deus et al. (2006) and Gencoglu et al.

(a) (b)

Figure 1 (a) SCT slice showing canal

obturated with Obtura II. (b) SCT slice

showing canal obturated by System B.

203 · 123 mm (300 · 300 DPI).

Table 1 The volume (in cm3) of root canals for each group

Groups Mean Standard deviation

LC 0.033 0.0055

Thermafil 0.035 0.0043

Obtura II 0.034 0.0051

System B 0.035 0.0047

LC, lateral compaction.

Volumetric analysis of root fillings Anbu et al.


(2002), who showed that Thermafil produced a

significantly higher POV. In this study two specimens

in the System B group had voids between the coronal-

middle and middle-apical segments. These voids were

not through and through but were found only in a

few sections, which could have been missed in a

conventional radiograph. Obtura II, a thermoplastic

injectable technique had only 85% of overall POV.

Presence of voids in the apical region had reduced the

overall POV, which may be due to failure of the needle

tip to reach the apical third, poor compaction and

entrapment of air. The use of a 20-guage Obtura

needle which has an outer diameter of 0.81 mm could

have prevented the tip from reaching the appropriate

depth.

Lateral compaction had 83% of overall filling, the

lowest value of all the fillings. This was mainly because

this technique does not produce a homogenous mass

and may leave spaces between the GP and the dentinal

walls or accessory cones. According to Schilder (2006),

in lateral compaction the final filling had the appear-

ance of numerous GP cones that had been tightly

pressed together and joined by frictional grip and the

cementing substance. Spreader tracts can be devoid of

sealer or the sealer can resorb later leading to voids. In

this study, voids were seen between the accessory cones

throughout the length of the canal.

When comparing the POV of coronal, middle and

apical thirds of Thermafil, System B and lateral

compaction techniques, there was no difference in

the efficacy of fillings. It was seen that the POV of

coronal third of lateral compaction, Thermafil and

System B was slightly less than apical third. This

could be due to the use of coronal orifice enlargement

with Gates Glidden drills and lack of additional

vertical condensation with pluggers in the Thermafil

group. Even though pluggers were used in case of

System B for coronal compaction there was a decrease

in efficacy. This could be due to the mismatch in

taper of the instrument to that of the enlarged canal

orifice.

Conclusion

Within the limitations of this study, voids were seen in

all the root fillings. The greatest POV was obtained with

System B and Thermafil techniques; lateral compac-

tion, produced the least POV. SCT appears to be a

valuable tool to locate voids and to assess the efficacy of

obturation at various levels.

References

Bartletta F, Reis M, Wagner M, Borges J, Dall’Agnol C (2008)

Computed tomography assessment of three techniques for

removal of filling material. Australian Endodontic Journal 34,

102–5.

Camps J, Pashley D (2003) Reliability of the dye penetration

studies. Journal of Endodontics 29, 592–4.

De-Deus G, Gurgel-Filho ED, Magalhaes KM, Coutinho-Filho

T (2006) A laboratory analysis of gutta-percha-filled

area obtained using Thermafil, System B and lateral

condensation. International Endodontic Journal 39, 378–

83.

De-Deus G, Reis C, Beznos D, de Abranches AMG, Coutinho-

Filho T, Paciornik S (2008) Limited ability of three

commonly used thermoplasticized gutta-percha techniques

in filling oval-shaped canals. Journal of Endodontics 34,

1401–5.

ElAyouti A, Achleithner C, Lost C, Weiger R (2005) Homo-

geneity and adaptation of a new gutta-percha paste to root

canal walls. Journal of Endodontics 31, 687–90.

Gencoglu N, Garip Y, Bas M, Samani S (2002) Comparison of

different gutta-percha root filling techniques: thermafil,

Quick-Fill, System B and lateral condensation. Oral Surgery

Oral Medicine Oral Pathology, Oral Radiology and Endodontics

9, 333–6.

Gencoglu N, Yildirim T, Garip Y, Karagenc B, Yilmaz H (2008)

Effectiveness of different gutta-percha techniques when

filling experimental internal resorptive cavities. International


Haıkel Y, Freymann M, Fanti V, Claisse A, Poumier F, Watson

M (2000) Apical microleakage of radiolabeled lysozyme over

time in three techniques of root canal obturation. Journal of


Table 2 POV values for each groupGroups Overall Coronal third Middle third Apical third

LC 80.4 ± 1.6 (X) 80.7 ± 3.6 (a) 81.5 ± 3.1 (a) 83 ± 3.6 (a)

Thermafil 93.3 ± 2.4 (Y) 92.5 ± 3.4 (b) 95.5 ± 4.7 (b) 97.4 ± 5.4 (b)

Obtura II 84.8 ± 6 (X) 92.9 ± 8.5 (b) 95.6 ± 6.1 (b) 54.3 ± 19.3 (c)

System B 93.7 ± 3.6 (Y) 92.1 ± 5.7 (b) 94.7 ± 6 (b) 96.2 ± 6.2 (b)

Cells with the same letter denote no statistical significance between them (P > 0.05).

Capital letters were used for the overall data.

POV, percentage of obturated volume; LC, lateral compaction.

Anbu et al. Volumetric analysis of root fillings


Hammad M, Qualtrough A, Silikas N (2008) Three-dimen-

sional evaluation of effectiveness of hand and rotary

instrumentation for retreatment of canals filled with differ-

ent materials. Journal of Endodontics 34, 1370–3.

Hammad M, Qualtrough A, Silikas N (2009) Evaluation of

root canal obturation: a three dimensional in vitro study.


Jacobson HLJ, Xia T, Baumgartner JC, Marshall JG, Beeler WJ

(2002) Microbial leakage evaluation of the continuous

wave of condensation. Journal of Endodontics 28, 269–71.

Jung IY, Lee S, Kim E, Lee C, Lee S (2003) Effect of different

temperatures and penetration depths of a system B plugger

in the filling of artificially created oval canals. Oral Surgery


96, 453–7.

Kontakiotis EG, Tzanetakis GN, Loizides AL (2007) A 12-

month longitudinal in vitro leakage study on a new silicon-

based root canal filling material (gutta-flow). Oral Surgery


103, 854–9.

McRobert AS, Lumley PJ (1997) An in vitro investigation of

coronal leakage with three gutta-percha backfilling tech-

niques. International Endodontic Journal 30, 413–7.

Nair MK, Nair UP (2007) Digital and advanced imaging in

endodontics: a review. Journal of Endodontics 33, 1–6.

Nandini S, Velmurugan N, Kandaswamy D (2006) Removal

efficiency of calcium hydroxide intracanal medicament

with two calcium chelators: volumetric analysis using

spiral CT, an in vitro study. Journal of Endodontics 32,

1097–101.

Oliver CM, Abbott PV (2001) Correlation between clinical

success and apical dye penetration. International Endodontic

Journal 34, 637–44.

Pommel L, Camps J (2001) Effect of pressure and measure-

ment time on the fluid filtration method in endodontics.


Reuben J, Velmurugan N, Kandaswamy D (2008) The

evaluation of root canal morphology of the mandibular first

molar in an Indian population using spiral computed

tomography scan: an in vitro study. Journal of Endodontics

34, 212–5.

Robinson S, Czerny C, Gahleitner A, Bernhart T, Kainberger FM

(2002) Dental CT evaluation of mandibular first premolar root

configurations and canal variations. Oral Surgery Oral Medi-

cine Oral Pathology, Oral Radiology and Endodontics 93, 328–

32.

Schilder H (2006) Filling root canals in three dimensions.


Siqueira JF Jr, Rocas IN, Lopes HP, De Uzeda M (1999) Coronal

leakage of two root canal sealers containing calcium

hydroxide after exposure to human saliva. Journal of


Siqueira JF Jr, Rocas IN, Favieri A, Abad EC, Castro AJR,

Gahyva SM (2000) Bacterial leakage in coronally unsealed

root canals obturated with 3 different techniques. Oral

Surgery Oral Medicine Oral Pathology 90, 647–50.

Venturi M (2006) Evaluation of canal filling after using two warm

vertical gutta-percha compaction techniques in vivo: a pre-

liminary study. International Endodontic Journal 39, 538–46.

Volumetric analysis of root fillings Anbu et al.


Influence of the shape of artificial canals on thefatigue resistance of NiTi rotary instruments

G. Plotino1, N. M. Grande1, M. Cordaro2, L. Testarelli1 & G Gambarini1

1University of Roma ‘La Sapienza’, Rome, Italy; and 2Catholic University of Sacred Heart, Rome, Italy

Abstract

Plotino G, Grande NM, Cordaro M, Testarelli L, Gam-

barini G. Influence of the shape of artificial canals on the

fatigue resistance of NiTi rotary instruments. International

Endodontic Journal, 43, 69–75, 2010.

Aim To investigate the influence of the trajectory of

NiTi rotary instruments on the outcome of cyclic

fatigue tests.

Methodology Ten ProFile and Mtwo instruments tip

size 20, taper 0.06 and tip size 25, taper 0.06 were

tested in two simulated root canals with an angle of

curvature of 60� and radius of curvature of 5 mm but

with different shape. Geometrical analysis of the angle

and radius of the curvature that each instrument

followed inside the two different artificial canals was

performed on digital images. The instruments were

then rotated until fracture at a constant speed of

300 rpm to calculate the number of cycles to failure

(NCF) and the length of the fractured fragment. Mean

values were calculated and analysed using two different

multivariate linear regression models and an indepen-

dent sample t-test.

Results The shape of the artificial root canal used in

cyclic fatigue studies influenced the trajectory of the

instrument. This difference is reflected by the NCF

measured for the same instrument in the different

artificial root canals and by the impact of the type of

canal on both the NCF (St.b = 0.514) and fragment

length (St.b = )0.920).

Conclusions Small variations in the geometrical

parameters of the curvature of an instrument subjected

to flexural fatigue could have a significant influence on

the results of fatigue tests.

Keywords: angle of curvature, artificial canal, cyclic

fatigue test, radius of curvature.

Received 27 June 2009; accepted 2 September 2009

Introduction

Fracture of instruments used in rotary motion occurs in

two different ways: fracture due to torsion and fracture

due to flexural fatigue (Serene et al. 1995, Sattapan

et al. 2000, Ullmann & Peters 2005). Torsional fracture

occurs when an instrument tip or another part of the

instrument is locked in a canal whilst the shank

continues to rotate. When the torque exerted by the

hand-piece exceeds the elastic limit of the metal,

fracture of the tip becomes inevitable (Peters 2004,

Parashos & Messer 2006). Instruments fractured

because of torsional loads often carry specific signs

such as plastic deformation (Sattapan et al. 2000).

Fracture due to fatigue through flexure occurs

because of metal fatigue. The instrument does not bind

in the canal but it rotates freely in a curvature,

generating tension/compression cycles at the point of

maximum flexure until the fracture occurs (Pruett et al.

1997, Haikel et al. 1999). As an instrument is held in a

static position and continues to rotate, one half of the

instrument shaft on the outside of the curve is in

tension, whilst the half of the shaft on the inside of the

curve is in compression. This repeated tension–

compression cycle, caused by rotation within curved

canals, increases cyclic fatigue of the instrument over

time and may be an important factor in instrument

fracture (Pruett et al. 1997).

Correspondance: Dr Gianluca Plotino, Via Eleonora Duse 22,

00197 Rome, Italy (Tel.: +393396910098; fax:

+3968072289; e-mail: [email protected]).

doi:10.1111/j.1365-2591.2009.01641.x


Resistance of rotary instruments to cyclic fatigue is

affected by the angle and radius of canal curvature and

the size and taper of the instrument. Increased severity

in the angle and radius of the curves around which the

instrument rotates decreases instrument lifespan

(Pruett et al. 1997, Haikel et al. 1999, Grande et al.

2006). Instruments have been tested in canals having

radii of 2 mm, 5 mm and 10 mm, with the conclusion

that the smaller the radius, the shorter the life of the

instrument when rotating (Pruett et al. 1997, Haikel

et al. 1999, Grande et al. 2006). Similarly, several

studies have shown that increased diameter at the

point of maximum curvature of the instrument, which

is determined by tip size and taper, reduces the time to

fracture (Pruett et al. 1997, Haikel et al. 1999, Plotino

et al. 2006, 2007). Only the study by Yared et al.

(2000) did not support these findings. Ruddle (2002)

has asserted that the position of the curvature of a

canal is a factor in instrument safety, a point that was

demonstrated in an earlier study (Malagnino et al.

1999). When the curvature is localized in a coronal

portion of the root canal, the instrument is subjected to

the maximum stress in the area in which its diameter is

largest.

In nearly all studies reported in the endodontic

literature, the rotating instrument was either confined

in a glass or metal tube, in a grooved block-and-rod

assembly or in a sloped metal block (Plotino et al.

2009); there has been no mention of the ‘fit’ of the

instrument in the tube or groove. As the instrument is

likely to be fitting loosely, the description of the radius

of curvature in those studies is likely to be overstated;

that is, the file was actually bent less severely than

reported. Previous studies using cylindrical metallic

tubes to test the cyclic fatigue life of NiTi rotary

instruments reported that the tubes do not sufficiently

constrain the shafts of the smaller instruments (Pruett

et al. 1997, Mize et al. 1998, Yared et al. 1999, 2000).

The aim of the present study was to investigate the

influence of the trajectory of NiTi rotary instruments on

the outcome of cyclic fatigue tests. The null hypothesis

tested was that there was no difference in the cyclic

fatigue resistance of the same instrument tested in two

artificial canals with the same radius and angle of

curvature but with different shapes.


Ten ProFile NiTi rotary instruments (Dentsply Maillefer,

Ballaigues, Switzerland) tip size 20, 0.06 taper, ten

ProFile instruments tip size 25, 0.06 taper, ten Mtwo

NiTi rotary instruments (Sweden & Martina, Padova,

Italy) tip size 20, 0.06 taper and ten Mtwo instruments

tip size 25, 0.06 taper were selected.

Two simulated root canals with an angle of curva-

ture of 60� and radius of curvature of 5 mm were

constructed for each instrument size. The centre of the

curvature was approximately 5 mm from the tip of the

instrument, the curved segment of the canal was

approximately 5 mm in length and the linear segment

between the tip of the instrument and the end-point of

the curvature was approximately 2.5 mm. The artifi-

cial canals with two different shapes were milled in

stainless-steel blocks with a precision milling machine.

An artificial canal (A) was constructed with a

tapered shape corresponding to the dimensions of the

instruments tested (tip size and taper) (Fig. 1a), thus

providing the instrument with a suitable trajectory. To

ensure the accuracy of the size of each canal a copper

duplicate of each instrument was milled increasing the

original size of the instrument by 0.1 mm using a

computer numerical control machining bench (Bridge-

port VMC 760XP3; Hardinge Machine Tools Ltd.,

Leicester, UK). The copper duplicates were constructed

according to the curvature parameters that were

chosen for the study. With these negative moulds the

artificial canals were made using a die-sinking electri-

cal-discharge machining process (Agietron Hyperspark

3, AGIE Sa, Losone, Switzerland) in a stainless-steel

block. The depth of each artificial canal was machined

to the maximum diameter of the instrument +0.2 mm,

allowing the instrument to rotate freely inside the

artificial canal. The blocks were hardened through

annealing.

A second artificial canal (B) was constructed with a

tapered shape but with larger dimensions that did not

match the instrument size and taper; the artificial canal

was machined increasing the original size of the

instrument by 0.3 mm (Fig. 1b).

Each artificial canal was mounted on a stainless-steel

block that was connected to a frame to which a mobile

plastic support for the hand-piece was also connected.

The dental hand-piece was mounted upon a mobile

device that allowed for precise and simple placement of

each instrument inside the artificial canal, ensuring

three-dimensional alignment and positioning of the

instruments to the same depth. The artificial canal was

covered with tempered glass to prevent the instruments

from slipping out and to allow for observation of the

instrument.

Geometrical analysis of the trajectory that each

instrument followed inside the two different artificial

Influence of trajectory on fatigue resistance Plotino et al.


canals was performed on digital images, determining

two parameters: angle and radius of the curvature

described by the instruments as measured by Pruett

et al. (1997). A straight line (PQ) was drawn along the

long axis of the coronal straight portion of the

instrument. A second line (TS) was drawn along the

long axis of the apical straight portion of the instru-

ment. There was a point on each of these lines at which

the instrument deviated to begin (Q) or end (S) the

curvature. The curved portion of the instrument was

represented by a segment of a circle (C) with tangents

at these points. The most precise circumference that

lied on the trajectory of the instrument was geometri-

cally determined using the osculating circumference

method (Gray 1997). The osculating circle of a curve at

a given point is the circle that best approximate the

curve at that point and it is unique. This method was

chosen because the curvature followed by an instru-

ment not constrained in a precise trajectory is not a

circumference, but a plain curve with a different

equation. In these cases, the above described is the

most precise method to define a curvature by the

parameters of radius and angle of a circumference. It is

possible to determine the osculating circle passing

through three points of a curve. The points that were

chosen on the trajectory of the instruments were the

beginning (Q) and the end (S) of the curve and a point B

that was chosen as the centre of mass of the triangle

resulting from the points Q, S and R that was the point

in which the straight coronal and apical portions of the

instrument met. The angle of curvature was defined as

the number of degrees on the arc of the circle between

the beginning and end-points of the curvature; the

radius of the circle was defined as the radius of the

canal curvature in millimetres (Fig. 2).

The calculation of the radius and angle of curvature

determined by the osculating circumference method

was repeated for each instrument analysed in the two

(a) (b)

Figure 1 The artificial canals used in the

present study. (a) canal A; (b) canal B.

Figure 2 The osculating circumference method to determine

the geometrical parameters of the trajectory of the instrument.

Plotino et al. Influence of trajectory on fatigue resistance


different artificial canals. Mean values were then

calculated for each instrument size.

ProFile and Mtwo instruments were then tested

within the two different artificial canals. The instru-

ments were rotated at a constant speed of 300 rpm

using a 6:1 reduction hand-piece (Sirona Dental

Systems GmbH, Bensheim, Germany) powered by a

torque-controlled motor (Silver, VDW GmbH, Munich,

Germany). To reduce the friction of the file as it

contacted the artificial canal walls, high-flow synthetic

oil designed for lubrication of mechanical parts (Super

Oil, Singer, Elizabethport, NJ, USA) was applied. All

instruments were rotated until fracture occurred.

Fracture was easily detectable because the instruments

were visible through the glass window. The time to

fracture for each file was recorded visually with a 1/

100 s chronometer, and the number of rotations was

calculated to the nearest whole number. The time to

fracture was multiplied by the number of rotations per

minute to obtain the number of cycles to failure (NCF)

for each instrument. Mean values were then calculated.

The length of the fractured tip was also recorded for

each instrument and the mean values were then

calculated for each instrument type in each group.

Analysed data consisted of NCF and the length of the

fractured tip for each instrument tested under the

specified artificial canal, and the radius and angle of

curvature curvature followed by the instruments in

both the artificial root canals tested. The data were

processed using spss software (SPSS, Oakbrook, IL,

USA). Means and standard deviations (SD) were

calculated. Two different multivariate linear regression

models were performed to investigate the effects of the

independent variables considered in the model (size of

the instrument, type of the instrument, type of the

artificial root canal) on the dependent variables anal-

ysed (NCF and fragment length). An independent

sample t-test was used to analyse significant differences

for the angle and radius of curvature measured

between the two artificial root canals for each instru-

ment tested. Significance was determined at the 95%

confidence level.

Results

Mean values and SD of the radius and angle of

curvature described by the instruments in the different

artificial canals are displayed in Table 1.

Mean values ± SD expressed as NCF and the mean

length of the fractured segment are displayed in

Table 2.

In the first model, considering the NCF as dependent

variable, the overall regression model was statistically

significant (F = 13.4; P = 0.000; R = 0.586). Further-

more, amongst the independent variables canal type

(A, B) and instrument size (20, 0.06; 25, 0.06) were

statistically significant (P < 0.05), whilst the instru-

ment type (Mtwo, ProFile) was not (P = 0.371). The

Table 1 Mean values ± SD of the radius and angle of curvature described by the instruments in the different artificial canals and

P-values (t-test between canal A and B)

ProFile size 20/0.06 taper Angle (�) Radius (mm) Mtwo size 20/0.06 taper Angle (�) Radius (mm)

Canal A 60 ± 0.1 4.9 ± 0.3 Canal A 60 ± 0.01 5 ± 0.1

Canal B 51 ± 0.1 5.7 ± 0.3 Canal B 55 ± 0.1 5.9 ± 0.2

ProFile size 25/0.06 taper Angle (�) Radius (mm) Mtwo size 25/0.06 taper Angle (�) Radius (mm)

Canal A 60 ± 0.1 5 ± 0.2 Canal A 60 ± 0.1 4.9 ± 0,3

Canal B 50 ± 0.1 6.6 ± 0.4 Canal B 54 ± 0.3 6 ± 0,4

Table 2 Mean ± SD expressed in number of cycles to failure (NCF) registered during the cyclic fatigue testing, mean length of the

fragments ± SD registered for each group (in mm) and increase of the lifespan between the two groups

ProFile size 20/0.06 taper NCF mm Mtwo size 20/0.06 taper NCF mm

Canal A 605 ± 52 4.9 ± 0.4 Canal A 617 ± 73 5 ± 0.2

Canal B 677 ± 55 3.3 ± 0.4 Canal B 703 ± 61 3.5 ± 0.3

Difference* 10% Difference* 12%

ProFile size 25/0.06 taper NCF mm Mtwo size 25/0.06 taper NCF mm

Canal A 564 ± 63 4,9 ± 0.3 Canal A 566 ± 84 5.1 ± 0.3

Canal B 645 ± 75 3.1 ± 0.3 Canal B 659 ± 82 3,6 ± 0.3

Difference* 12.5% Difference* 14%

*Indicates an increase of the lifespan for canal B compared to canal A.



multivariate linear regression showed that canal type

was the independent variable with the greatest impact

in the model; canal B positively affected the NCF value

(St.b = 0.514, P < 0.000) when compared with canal

A, whilst an increasing in size negatively affected the

outcome variable (St.b = )0.260, P < 0.000).

In the second model, considering the fragment

length as dependent variable, the overall regression

model was statistically significant (F = 178.9;

P = 0.000; R = 0.935). Even considering the fragment

length the variable with the greater impact in the

model was the canal type, canal B negatively affected

the length of the fragment (St.b = )0.920, P < 0.000)

whilst size of the instrument was not statistically

significant (P = 0.844) and type of instrument has a

lower impact on the output variable (St.b = )0.179,

P < 0.000).

For all the instruments tested a statistically signifi-

cant difference was found between canal A and canal B

for both angle and radius of curvature (P < 0.000).

Discussion

Clinically, NiTi rotary instruments are subjected to both

torsional load and cyclic fatigue (Gambarini 2001,

Ullmann & Peters 2005), and ongoing research aims to

clarify the relative contributions of both factors to

instrument separation (Peters 2004).

Both cyclic fatigue tests (Pruett et al. 1997, Haikel

et al. 1999) and torsion tests (Camps & Pertot 1995,

Yared 2004, Ullmann & Peters 2005) have been

performed to investigate how these factors may influ-

ence the behaviour of NiTi rotary instruments in vitro.

In addition, torsional properties of used instruments

have been investigated (Yared et al. 2003, Yared 2004,

Ullmann & Peters 2005) to analyse how the combina-

tion of these two factors may influence instrument

failure.

The results of the present study showed that the

shape of the artificial root canal used in cyclic fatigue

studies influenced the trajectory of the instrument; for

all the instrument tested both angle and radius of the

curve statistically varied between canal A and B

(P < 0.000). This difference is reflected by the number

of cycles to failure measured for the same instrument in

the different artificial root canals and in the high

impact of the type of canal on both the NCF

(St.b = 0.514) and fragment length (St.b = )0.920).

The results of the present study showed a statistically

significant increase in the number of cycles to failure

when instruments were tested in an artificial canal that

does not sufficiently restrict the instrument shaft (canal

B). In this canal, the instrument would tend to regain

its original straight shape, aligning into a trajectory of

greater radius and reduced angle; that is, the file was

actually bent less severely than reported.

The results of the present study confirmed that the

size of the instrument at the point of maximum

curvature influenced resistance to fracture for cyclic

fatigue: bigger instruments are less resistant than

smaller instruments. That is, NCF decreased as the

diameter of the instrument increased (Pruett et al.

1997, Haikel et al. 1999, Grande et al. 2006, Plotino

et al. 2006, 2007). This is due to the fact that when a

curved root canal instrument rotates, any points within

it in the segment subjected to the maximum stress,

except those in the centre (neutral axis), are subjected

to repeated tensile or compressive strains. The farther

away from the central axis, the greater the imposed

strain at that point (Craig 1997). This explains why

instruments of a larger diameter are affected by fatigue

more than smaller ones.

Analysis of the data regarding the length of the

fractured segment revealed a statistically significant

difference in the mean size between canal A and canal

B for all of the instrument sizes. The centre of the

curvature was constructed approximately 5 mm from

the tip of the instrument for both canal A and canal B.

Instruments subjected to cyclic fatigue fractured at the

centre of the curvature or just below this point (Pruett

et al. 1997, Fife et al. 2004). The results of the present

study demonstrated that when the instruments were

tested in a precise artificial root canal they followed

precisely the trajectory established in the construction

of the canal. In fact, in the present study instruments

tested in canal A fractured at the established point of

maximum stress, as expected. This confirms previous

findings (Fife et al. 2004, Grande et al. 2006, Plotino

et al. 2006, 2007). On the contrary, data demonstrated

a significant decrease in the mean length of the

fragments for instruments tested in canal B. This was

due to the fact that instruments tested in canal B did

not followed the trajectory established by parameters

with which the artificial canal was constructed and

consequently the point of maximum stress were the

instrument fracture may vary.

Furthermore, considering the fragment length, there

was a minor but statistically significant impact of the

type of instrument on this variable (St.b = )0.179).

This was because different instruments followed an

unpredictable trajectory if the canal in which they were

tested did not guide them in a precise trajectory.



As above mentioned, bending properties of different

files may determine a different trajectory if the file is not

constrained in a precise trajectory. If testing is com-

pleted for all different files at a given angle to ensure

consistency, the bending properties of the different files

determining different angles of curvature, thus bias the

results and the comparisons. To limit these problems,

Cheung & Darvell (2007a,b,c) constrained the instru-

ment into a curvature using three stainless-steel pins.

They used three smooth cylindrical pins of 2 mm

diameter from a high hardness stainless steel mounted

in acrylic shims, which were adjustable in the hori-

zontal direction; the position of the pins determines the

curvature of the instrument. A small V-shaped groove

prepared on the lowest pin maintained the position of

the tip of the instrument during rotation. The authors

reported in detail on the effect of surface strain ampli-

tude on fatigue failure using it as a different indicator of

the stress on instruments instead of radius and angle of

curvature. It has been reported in a three-point bending

test of NiTi wires that such constraints will produce a

curvature that is circular (Wick et al. 1995). The

authors affirmed that although this cannot actually be

true, the approximation should be reasonable. Unfortu-

nately, NiTi endodontic files are tapered and with

different cross-sectional design. The different bending

properties of the different files and the different bending

properties between the coronal and apical portion of the

same file may determine a different trajectory between

the pins, if the file is not constrained precisely.

The present study sought to overcome the limitations

of some laboratory studies in terms of the model used

for testing. The artificial canal was specifically designed

for each instrument in terms of size and taper, giving it

a precise trajectory. Cylindrical metallic tubes used in

previous studies (Pruett et al. 1997, Mize et al. 1998,

Yared et al. 1999, 2000, Melo et al. 2002) did not

sufficiently restrict the instrument shaft, which would

tend to regain its original straight shape, aligning into a

trajectory of greater radius and reduced angle (Yared

et al. 1999, 2000, Melo et al. 2002, Bahia & Buono

2005). The results of a previous study (Plotino et al.

2009b) reported that an artificial canal manufactured

as described in the present to riproduce instrument size

and taper seems to guarantee that different NiTi rotary

instruments may follow a precise and repeatable

trajectory in terms of radius and angle of curvature.

On the contrary, if the artificial canal is not identical (in

shape and size) to the instrument, its trajectory will not

respond to the established parameters, thus having a

reduced curvature during the test. The results of the

present study demonstrated that the variation in the

trajectory followed by the instruments in the artificial

canals used to test fatigue resistance could influence

the results of cyclic fatigue tests.

Conclusions

The null hypothesis tested in the present study has

been rejected. Results of the present study reported that

even small variations of the geometrical parameters of

the curvature of an instrument subjected to flexural

fatigue could determine a significant influence on the

results of fatigue tests. The standardization of the

parameters and devices used for cyclic fatigue testing of

NiTi rotary instruments is lacking. A more precise

regulation is required to obtain more consistent and

comparable results in different studies.

References

Bahia M, Buono V (2005) Decrease in the fatigue resistance of

nickel-titanium rotary instruments after clinical use in

curved root canals. Oral Surgery Oral Medicine Oral Pathology

Oral Radiology and Endodontology 100, 249–55.

Camps JJ, Pertot WJ (1995) Torsional and stiffness properties

of nickel-titanium K files. International Endodontic Journal 28,

239–43.

Cheung GS, Darvell BW (2007a) Fatigue testing of a NiTi

rotary instrument. Part 1: strain-life relationship. Interna-


Cheung GS, Darvell BW (2007b) Fatigue testing of a NiTi

rotary instrument. Part 2: fractographic analysis. Interna-


Cheung GS, Darvell BW (2007c) Low-cycle fatigue of NiTi

rotary instruments of various cross-sectional shapes. Inter-

national Endodontic Journal 40, 626–32.

Craig RG (1997) Restorative Dental Materials, 10th edn. St.

Louis, MO, USA: Mosby.

Fife D, Gambarini G, Britto LR (2004) Cyclic fatigue testing of

ProTaper NiTi rotary instruments after clinical use. Oral

Surgery Oral Medicine Oral Pathology Oral Radiology and

Endodontology 97, 251–6.

Gambarini G (2001) Cyclic fatigue of ProFile rotary instru-

ments after prolonged clinical use. Internatiomal Endodontic

Journal 34, 386–9.

Grande NM, Plotino G, Pecci R, Bedini R, Somma F (2006)

Cyclic fatigue resistance and three-dimensional analysis of

instruments from two nickel-titanium rotary systems.


Gray A (1997) Osculating circles to plane curves. In: Gray A,

Abbena E, Salamon S, eds. Modern Differential Geometry of

Curves and Surfaces with Mathematic, 2nd edn. Boca Raton,

FL: CRC Press, pp. 111–5.



Haikel Y, Serfaty R, Bateman G, Senger B, Allemann C (1999)

Dynamic and cyclic fatigue of engine-driven rotary nickel

titanium endodontic instruments. Journal of Endodontics 25,

434–40.

Malagnino VA, Passariello P, Corsaro S (1999) The influence

of root canal trajectory on the risk of cyclic fatigue failure on

Ni-Ti engine driven endodontic instruments. Italian Journal


Melo MC, Bahia MGA, Buono VTL (2002) Fatigue resistance of

engine-driven rotary nickel-titanium endodontic instru-

ments. Journal of Endodontics 28, 765–9.

Mize SB, Clement DJ, Pruett JP, Carnes DL Jr (1998) Effect of

sterilization on cyclic fatigue of rotary nickel-titanium

endodontic instruments. Journal of Endodontics 24, 843–7.

Parashos P, Messer HH (2006) Rotary NiTi instrument fracture

and its consequences. Journal of Endodontics 32, 1031–43.

Peters OA (2004) Current challenges and concepts in the

preparation of root canal systems: a review. Journal of


Plotino G, Grande NM, Cordaro M, Testarelli L, Gambarini G.

(2009) A review on cyclic fatigue test of nickel-titanium

rotary instruments. Journal of Endodontics. In press.

Plotino G, Grande NM, Sorci E, Malagnino VA, Somma F

(2006) A comparison of cyclic fatigue between used and

new Mtwo NiTi rotary instruments. International Endodontic

Journal 39, 716–23.

Plotino G, Grande NM, Sorci E, Malagnino VA, Somma F

(2007) Influence of a brushing working motion on the

fatigue life of NiTi rotary instruments. International End-

odontic Journal 40, 45–51.

Plotino G, Grande NM, Cordaro M, Testarelli L, Gambarini G

(2009b) Measurement of the trajectory of different NiTi

rotary instruments in an artificial canal specifically designed

for cyclic fatigue tests. Oral Surgery, Oral Medicine, Oral

Pathology, Oral Radiology and Endodontics 108, e152–6. DOI:

10.1016/j.tripleo.2009.05.046.

Pruett JP, Clement DJ, Carnes DL (1997) Cyclic fatigue of

nickel-titanium endodontic systems. Journal of Endodontics

23, 77–85.

Ruddle C (2002) Cleaning and shaping the root canal system.

In: Cohen S, Burns RC, eds. Pathways of the Pulp, 8th edn. St.

Louis: Mosby, pp. 231–92.

Sattapan B, Nervo G, Palamara J, Messer H (2000) Defects in

nickel titanium endodontic rotary files after clinical usage.


Serene TP, Adams JD, Saxena A (1995) Nickel-Titanium

Instruments: Applications in Endodontics. St Louis, MO, USA:

Ishiyaku EuroAmerica.

Ullmann CJ, Peters OA (2005) Effect of cyclic fatigue on static

fracture loads in ProTaper nickel-titanium rotary instru-

ments. Journal of Endodontics 31, 183–6.

Wick A, Vohringer O, Pelton AR (1995) The bending behavior

of NiTi. Journal de Physique IV, Colloque C8 (ICOMAT-95) 5,

789–94.

Yared GM (2004) In vitro study of the torsional properties of

new and used profile nickel titanium rotary files. Journal of


Yared GM, Bou Dagher FE, Machtou P (1999) Cyclic fatigue of

profile rotary instruments after simulated clinical use.


Yared GM, Bou Dagher FE, Machtou P (2000) Cyclic fatigue of

profile rotary instruments after clinical use. International


Yared G, Kulkarni GK, Ghossayn F (2003) Torsional properties

of new and used rotary K3 NiTi files. Australian Endodontic

Journal 29, 75–8.



CASE REPORT

Apexogenesis after initial root canaltreatment of an immature maxillaryincisor – a case report

S. R. Kvinnsland1, A. Bardsen2 & I. Fristad2

1Clinic of Dentistry – Endodontics; and 2Department of Clinical Dentistry – Endodontics,

University of Bergen, Bergen, Norway

Abstract

Kvinnsland SR, Bardsen A, Fristad I. Apexogenesis after initial root canal treatment of an

immature maxillary incisor – a case report. International Endodontic Journal, 43, 76–83, 2010.

Aim To present a case where a traumatized, immature tooth still showed capacity for

continued root development and apexogenesis after root canal treatment was initiated

based on an inaccurate pulpal diagnosis.

Summary Traumatic dental injuries may result in endodontic complications. Treatment

strategies for traumatized, immature teeth should aim at preserving pulp vitality to

ensure further root development and tooth maturation. A 9-year-old boy, who had

suffered a concussion injury to the maxillary anterior teeth, was referred after

endodontic treatment was initiated in tooth 21 one week earlier. The tooth had

incomplete root length, thin dentinal walls and a wide open apex. The pulp chamber

had been accessed, and the pulp canal instrumented to size 100. According to the

referral, bleeding from the root made it difficult to fill the root canal with calcium

hydroxide. No radiographic signs of apical breakdown were recorded. Based on

radiographic and clinical findings, a conservative treatment approach was followed to

allow continued root development. Follow-up with radiographic examination every 3rd

month was performed for 15 months. Continued root formation with apical closure

was recorded. In the cervical area, a hard tissue barrier developed, which was sealed

with white mineral trioxide aggregate (MTA). Bonded composite was used to seal the

access cavity. At the final 2 years follow-up, the tooth showed further root

development and was free from symptoms.

Key learning points

• Endodontic treatment of immature teeth may result in a poor long-term prognosis.

• The pulp of immature teeth has a significant repair potential as long as infection is

prevented.

doi:10.1111/j.1365-2591.2009.01645.x

Correspondence: Inge Fristad, Department of Clinical Dentistry – Endodontics, University of

Bergen, Arstadveien 17, N-5009 Bergen, Norway (Tel.: + 47 55 58 66 04; e-mail: inge.fristad@

iko.uib.no).


• Treatment strategies of traumatized, immature permanent teeth should aim at

preserving pulp vitality to secure further root development and tooth maturation.

• Radiographic interpretation of the periapical area of immature teeth may be confused

by the un-mineralized radiolucent zone surrounding the dental papilla.

Keywords: apexogenesis, diagnosis, endodontic treatment, immature tooth, pulp

necrosis, root development.

Received 5 June 2009; accepted 20 September 2009

Introduction

Traumatic dental injuries may jeopardize pulp survival in affected teeth. Luxation injuries

and avulsions are the most frequent traumatic causes for pulp necrosis resulting in the

need for endodontic treatment. In immature teeth, preservation of pulp vitality is crucial

for continued dentine formation and root development. Thus, treatment strategies for the

immature young dentition are important for the long-term prognosis of teeth and should

aim at preserving pulp vitality to secure tooth maturation and root development. In

immature teeth with pulp necrosis and bacterial infection, the long-term prognosis is

related to the stage of root development and the amount of root dentine present at time of

injury (Cvek 1992).

In teeth with an open apex, luxation may occur without disruption of the pulpal blood

and nerve supply. Moreover, pulp revascularization and repair will more readily occur in

teeth with a wide apical foramen (Andreasen et al. 1986). Consequently, a more

conservative treatment approach is recommended during follow-up of traumatized

immature teeth. Bacterial control is important and decisive to avoid infection resulting in

arrested root development.

The repair potential of immature teeth following luxation injuries is reflected in a more

favourable outcome after injury compared to mature teeth (Andreasen & Pedersen 1985).

Two factors have been found to be significantly related to the development of pulp

necrosis; the type of luxation injury and stage of root development (Andreasen 1970). The

frequency of pulp necrosis after luxation injuries in the permanent dentition has been

found to range from 5% to 59% (Andreasen & Andreasen 2007). Concussion and

subluxation injuries seldom results in pulp necrosis in immature teeth, whereas pulp

necrosis occurs in approximately 5% of teeth with complete root development

(Andreasen & Pedersen 1985). Following more serious luxation injuries, such as extrusive

and lateral luxation, approximately 10% of teeth with an open apex will develop pulp

necrosis (Andreasen et al. 1987, Andreasen 1989).

From previous studies, there appears to be a general agreement that lack of pulp

sensitivity or coronal discolouration alone is not sufficient diagnostic criteria to justify pulp

necrosis (Magnusson & Holm 1969, Bhaskar & Rappaport 1973, Zadik et al. 1979,

Jacobsen 1980). Periapical radiolucency has so far been considered to be the ‘safe’ sign of

pulp necrosis. However, investigations have questioned the validity of this assumption

(Andreasen 1989). In teeth with incomplete root formation, the radiographic interpretation

of the periapical area may be confused by the un-mineralized radiolucent zone surrounding

the dental papilla (Andreasen 1989). Even the concomitant presence of all three classical

signs of pulp necrosis; coronal discolouration, loss of pulp sensitivity and periapical

radiolucency, can in rare cases be followed by pulp repair (Andreasen 1989).

Pulp necrosis should be confirmed by sensitivity tests, keeping in mind that false

positive or negative result may be recorded. Pulp diagnosis is decisive for appropriate

CA

SE

RE

PO

RT


treatment and long-term prognosis. Infected teeth left untreated (false positive) might be

lost because of infectious related resorptions (Fuss et al. 2003). On the other hand, the

initiation of endodontic treatment of vital immature teeth (false negative) will impair

dentine formation and root development, thus substantially reducing the chances of long-

term survival (Cvek 1992).

The aim of this report is to present a case where a traumatized, immature tooth still

showed capacity for further root development and apexogenesis even after endodontic

instrumentation of the root canal. The treatment was based on an inaccurate pulp

diagnosis.

Case report

A 9-year-old boy was referred from the public dental health service to the clinic for post-

graduate endodontic training, University of Bergen, Bergen, Norway. The referral was

based on the following information: the maxillary central incisors were subjected to a

traumatic dental injury during ice-skating. Immediately after the accident, the patient was

examined at a public dental emergency clinic where concussion of the maxillary central

incisors was diagnosed. No emergency treatment was performed, and the patient was

referred to the public dental health service for follow-up. One month later, the patient

claimed weak and diffuse symptoms in the maxillary anterior region. An appointment at

the public dental health service was organized. Based on radiographic and clinical findings,

apical periodontitis was diagnosed (Fig. 1a). Vital pulp tissue with normal bleeding was

recorded when the pulp chamber was accessed. The root canal was then instrumented to

size 100 and irrigated with sodium hypochlorite 0.5% (Fig. 1b). According to the patient

record, there was profound bleeding with difficulties applying calcium hydroxide paste in

the instrumented root canal.

Following the referral, a clinical and radiographic examination was performed 1 week

after initial endodontic treatment. Tooth 21 was free from symptoms. Radiographs

revealed an immature tooth with incomplete root length, thin dentinal walls and a wide

open apex (Fig. 2a). No radiographic signs of apical breakdown were recorded. A radio-

(a) (b)

Figure 1 Radiographs taken 1 month after subluxation of the anterior teeth. The diagnosis apical

periodontitis form an infected root canal was set based on radiograph (a). The instrumentation length

was set according to radiograph (b), followed by instrumentation of the root canal to reamer ISO 100.

CA

SE

RE

PO

RT


opaque material (calcium hydroxide) was visible only in the coronal part of the root canal

(Fig. 2a). Based on the radiographic and clinical findings, the diagnosis ‘previous initiated

root canal treatment’ (vital tooth) was recorded. Because of the insufficient introduction of

calcium hydroxide into the root canal, a conservative approach was decided upon, thereby

allowing observation of any further continued root development. Completion of the

endodontic treatment at this stage would result in a weak tooth with poor prognosis. The

patient and his parents were informed and agreed to the proposed treatment strategy.

At follow-up, 1 month later, (Fig. 2b) the tooth was still free of symptoms. The colour of

the tooth was normal, and signs of slight growth of the root could be noticed from the

radiographs.

Four months later, (Fig. 2c) the radiographs showed continued root formation and

thickening of the dentinal walls. The calcium hydroxide dressing was removed with

sodium hypochlorite 0.5%. A calcified bridge of hard tissue was verified in the cervical 1/3

of the root by visual inspection through a dental surgical microscope. The coronal part of

the tooth was dried and packed with calcium hydroxide paste, and IRM� was placed as a

temporary filling (Fig. 3a).

The tooth was then followed with radiographic examination every 3rd month for the

following 15 months (Figs 2d–f). Continued root formation and apical closure were

registered. No clinical symptoms were recorded. Finally, a 2–3 mm thick plug of white

mineral trioxide aggregate (MTA, Angulus) was placed in contact with the hard tissue

bridge (Fig. 2g, and the access cavity filled with bonded composite (Tetric flow/Tetric

Ceram, Ivoclar Vivadent AG, Liechtenstein). Follow-up was performed 7 months later

(a) (b) (c) (d)

(e) (f) (g) (h)

Figure 2 Radiographs showing continued root development during a 2 year follow-up period. (a)

Radiograph taken at the first appointment shows an immature tooth with incomplete root length, thin

dentinal walls and an open apex. Calcium hydroxide is visible in the coronal part of the root canal. (b)

One month later, slight growth of the root and mineralization in the cervical area is noted. (c–f)

Continued root formation and apical closure is observed during 15 months follow-up. (g) Radiograph

taken after application of mineral trioxide aggregate (MTA). (h) Final follow-up 2 years after the first

appointment. Bonded composite is used to seal the access cavity.

CA

SE

RE

PO

RT


(Fig. 2h). The tooth was free from symptoms, and radiographs showed further root

development. Slight discolouration was noted in the cervical area (Fig. 3b).

Discussion

This case report illustrates the repair potential of a tooth with incomplete root formation.

The capacity for continued root development was preserved after traumatic injury and

treatment complications. Furthermore, it underlines the importance of an accurate pulp

diagnosis and a proper plan for treatment and follow-up of these teeth.

Development of pulp necrosis after dental trauma can be associated with symptoms

such as spontaneous pain or tenderness to percussion (Andreasen 1989). From previous

studies, it appears to be a general agreement that lack of pulp sensitivity (Magnusson &

Holm 1969, Bhaskar & Rappaport 1973, Zadik et al. 1979, Jacobsen 1980) or coronal

discolouration alone is not sufficient to justify pulp necrosis (Magnusson & Holm 1969,

Jacobsen 1980). Diagnosing traumatized, immature teeth may be a challenge to the

dentist, as demonstrated in the present case. The radiolucent zone surrounding the apical

dental papilla was interpreted as a periapical lesion from an infected necrotic pulp. The

initial endodontic treatment was based on misinterpretation of clinical and radiographic

findings. Although the root canal was instrumented to size 100, some odontoblasts and

pulp cells may have been left intact. The incomplete administration of calcium hydroxide

paste into the pulp canal was in this sense favourable. In addition, the copious solid

bleeding from the pulp tissue may have favoured reorganization of surviving pulpal tissue.

Different traumatic injuries may interfere with the pulpal neurovascular supply and

give rise to various defence and repair responses, ranging from localized or generalized

pulpal inflammation, tissue regeneration, reparative dentine formation or bone metapla-

sia and internal resorption, as well as pulp necrosis with or without bacterial

(a)

(b)

Figure 3 Clinical situation before placement of mineral trioxide aggregate (MTA) (a). Follow-up

7 months after application of white MTA showing light grey discolouration in the cervical area of tooth

number 21 (b).

CA

SE

RE

PO

RT


contamination (Andreasen et al. 1988). A common pattern of the repair process is

reorganization of the damaged pulp tissue, formation of new vessels and recruitment of

pulp progenitor cells to the injured area, whereby a tissue loss is gradually replaced by

new tissue (Andreasen et al. 1988).

The character of the pulpal responses varies, not only according to the type and severity

of the traumatic injury, but also on the origin of the progenitor cells involved in the process.

Tissue repair may be initiated from progenitor cells of pulpal origin, from periodontal

tissues or from a combination of the two. If damaged pulp tissue is renewed by progenitor

cells of pulpal origin, differentiation of new odontoblasts, forming reparative dentine, may

occur. The new dentine formed may even be re-innervated by sensory nerves (Kvinnsland

et al. 1991). In contrast, when the damaged tissue is restored by cells from periodontal

tissues, periodontal stem cell progenitors may invade the root canal resulting in collagen

and hard tissue formation. In the present case, continued normal root formation was seen,

indicating repair based on cells of pulpal origin.

If a pulp exposure site is covered with a suitable capping material, limiting or

preventing bacterial contamination, a hard tissue barrier is normally established (Watts

& Paterson 1981, Cvek 2007). In the present case, radiographic examination indicated

the presence of calcium hydroxide paste only in the coronal part of the tooth. The

initial application of calcium hydroxide paste may have initiated the formation of a hard

tissue bridge in the canal entrance. The type and quality of this hard tissue bridge

cannot be evaluated by radiographic or clinical inspection. The cells responsible for the

formation of this hard tissue barrier include mesenchymal, paravascular cells that

differentiate into odontoblasts like cells (Ruch 1945, Sveen & Hawes 1968, Zach et al.

1969, Feit et al. 1970, Luostarinen 1971, Yamamura 1985). Hard tissue bridges formed

after calcium hydroxide application are often incomplete with multiple tunnel defects

that may lead to micro leakage (Cox et al. 1996). As a consequence, a bacteria tight

seal should be established over the bridge. In this case, white MTA followed by

bonded composite was used for this purpose. Discolouration of the tooth because of

the use of Grey MTA in the cervical region has been reported (Glickman & Koch 2000).

As an attempt to overcome this problem, white MTA has recently been introduced.

The major difference between grey and white MTA is the concentration of Al2O3, MgO

and, especially, FeO, with the observed values for each of these oxides being

considerably lower in the white MTA (Asgary et al. 2005). Differences in the observed

FeO concentration are thought to be primarily responsible for the variation in colour of

the white MTA when compared to gray MTA. The present case showed slight grey

discolouration even after filling with white MTA in the cervical part of the root canal,

indicating that the aesthetic properties of MTA are not completely solved. Although the

cervical discolouration of the crown was noted, it was accepted by the patient and his

parents.

Conclusion

Special care should be taken during the evaluation and follow-up of traumatized immature

teeth, and more then one sign indicating pulp necrosis should be recorded before

endodontic treatment is started. In this case, the apical un-mineralized apical area

surrounding the developing dental papilla was unintentionally interpreted as apical

pathosis from an infected necrotic pulp. An observation strategy is recommended and no

intervention should be carried out before pulp necrosis is properly verified. Initially, a

frequent follow-up regime should be used for periodontal injuries at high risk of

inflammatory resorption to allow early identification of this pathology. The long-term

prognosis of immature teeth is dependent on continued root formation.

CA

SE

RE

PO

RT


Disclaimer

Whilst this article has been subjected to Editorial review, the opinions expressed, unless

specifically indicated, are those of the author. The views expressed do not necessarily

represent best practice, or the views of the IEJ Editorial Board, or of its affiliated Specialist

Societies.

References

Andreasen JO (1970) Luxation of permanent teeth due to trauma. A clinical and radiographic follow-up

study of 189 injured teeth. Scandinavian Journal of Dental Research 78, 273–86.

Andreasen FM (1989) Pulpal healing after luxation injuries and root fracture in the permanent dentition.


Andreasen FM, Andreasen JO (2007) Luxation injuries of permanent teeth: general findings. In:

Andreasen JO, Andreasen FM, Andreasen L, eds. Traumatic Injuries to the Teeth, 4th edn.

Copenhagen: Blackwell Munksgaard, pp. 372–97.

Andreasen FM, Pedersen BV (1985) Prognosis of luxated permanent teeth – the development of pulp

necrosis. Endodontics and Dental Traumatology 1, 207–20.

Andreasen FM, Zhijie Y, Thomsen BL (1986) Relationship between pulp dimensions and development

of pulp necrosis after luxation injuries in the permanent dentition. Endodontics and Dental


Andreasen FM, Zhijie Y, Thomsen BL, Andersen PK (1987) Occurrence of pulp canal obliteration after

luxation injuries in the permanent dentition. Endodontics and Dental Traumatology 3, 103–15.

Andreasen JO, Kristerson L, Andreasen FM (1988) Damage of the Hertwig’s epithelial root sheath:

effect upon root growth after autotransplantation of teeth in monkeys. Endodontics and Dental


Asgary S, Parirokh M, Eghbal MJ, Brink F (2005) Chemical differences between white and gray

mineral trioxide aggregate. Journal of Endodontics 31, 101–3.

Bhaskar SN, Rappaport HM (1973) Dental vitality tests and pulp status. Journal of American Dental

Association 86, 409–11.

Cox CF, Subay RK, Ostro E, Suzuki S, Suzuki SH (1996) Tunnel defects in dentin bridges: their

formation following direct pulp capping. Operative Denistryt 21, 4–11.

Cvek M (1992) Prognosis of luxated non-vital maxillary incisors treated with calcium hydroxide and

filled with gutta-percha. A retrospective clinical study. Endodontics and Dental Traumatology 8,

45–55.

Cvek M (2007) Endodontic Management and the use of Calcium Hydroxide in Traumatized Permanent

Teeth. In: Andreasen JO, Andreasen FM, Andersson L, eds. Traumatic Injuries to the Teeth, 4th

edn. Copenhagen: Blackwell Munksgaard, pp. 598–647.

Feit J, Metelova M, Sindelka Z (1970) Incorporation of 3H thymidine into damaged pulp of rat incisors.

Journal of Dental Research 49, 783–6.

Fuss Z, Tsesis I, Lin S (2003) Root resorption – diagnosis, classification and treatment choices based

on stimulation factors. Dental Traumatology 19, 175–82.

Glickman GN, Koch KA (2000) 21st-century endodontics. Journal of American Dental Association

131(Suppl), 39S–46S.

Jacobsen I (1980) Criteria for diagnosis of pulp necrosis in traumatized permanent incisors.

Scandinavian Journal of Dental Research 88, 306–12.

Kvinnsland I, Heyeraas KJ, Byers MR (1991) Regeneration of calcitonin gene-related peptide

immunoreactive nerves in replanted rat molars and their supporting tissues. Archives of Oral Biology

36, 815–26.

Luostarinen V (1971) Dental pulp response to trauma. An experimental study in the rat. Suomi

Hammaslaak Toim 67(Suppl 2), 3–74.

Magnusson B, Holm AK (1969) Traumatised permanent teeth in children – a follow-up. I. Pulpal

complications and root resorption. Svenske Tandlakare Tidskrift 62, 61–70.

CA

SE

RE

PO

RT


Ruch J (1945) Odontoblast differentiation and the formation of the odontoblast layer. Journal of Dental

Research 64, 489–98.

Sveen OB, Hawes RR (1968) Differentiation of new odontoblasts and dentine bridge formation in rat

molar teeth after tooth grinding. Archives of Oral Biology 13, 1399–409.

Watts A, Paterson RC (1981) Cellular responses in the dental pulp: a review. International Endodontic

Journal 14, 10–9.

Yamamura T (1985) Differentiation of pulpal cells and inductive influences of various matrices with

reference to pulpal wound healing. Journal of Dental Research 64, 530–40.

Zach L, Topal R, Cohen G (1969) Pulpal repair following operative procedures. Radioautographic

demonstration with tritiated thymidine. Oral Surgery Oral Medicine Oral Pathology 28, 587–97.

Zadik D, Chosack A, Eidelman E (1979) The prognosis of traumatized permanent anterior teeth with

fracture of the enamel and dentin. Oral Surgery Oral Medicine Oral Pathology 47, 173–5.

CA

SE

RE

PO

RT


CASE REPORT

Non-resolving periapical inflammation:a malignant deception

D. Saund, S. Kotecha, J. Rout & T. DietrichBirmingham Dental Hospital, University of Birmingham, Birmingham, UK

Abstract

Saund D, Kotecha S, Rout J, Dietrich T. Non-resolving periapical inflammation: a malignant

deception. International Endodontic Journal, 43, 84–90, 2010.

Aim To report a case of oral non-Hodgkin’s lymphoma with a delayed diagnosis.

Summary Non-Hodgkin’s lymphoma of the oral cavity is an uncommon but important

condition. Early diagnosis is complicated when the presenting signs and symptoms are

similar to those of odontogenic infections. This report describes the case of a 38-year-old

female patient who presented to her dentist complaining of pain in her upper jaw.

Subsequent dental treatment, including extraction, root canal treatment and apicectomy

including biopsy were carried out by the patient’s dentist and local dental hospital. Nine

months elapsed before a more extensive surgical exploration established a diagnosis of

lymphoma.

Key learning points

• To appreciate the importance of recognizing discrepancies between the clinical

scenario and histopathological findings.

• To appreciate subtle radiographic changes that may accompany malignant disease of

the jaw bones.

• To appreciate the need for early referral when a patient’s symptoms do not

satisfactorily respond to conventional dental therapies.

• To appreciate lymphoma should be considered in the differential diagnosis of non-

healing periapical inflammation and non-healing socket.

Keywords: lymphoma, non-healing socket, periapical infection.

Received 16 December 2008; accepted 7 September 2009

Introduction

The incidence of non-Hodgkin’s lymphoma (NHL) has increased by over 50% in the

20-year period between 1986 and 2005 (Cancer Research UK 2008a). NHL now accounts

for 4% of all malignant neoplasms in the UK (Cancer Research UK 2008b). It is the third

most common malignancy to affect the head and neck region, after squamous cell

doi:10.1111/j.1365-2591.2009.01644.x

Correspondence: Daniel Saund, Birmingham Dental Hospital, St Chad’s Queensway, Birming-

ham, B4 6NN, UK (e-mail: [email protected]).


carcinoma and salivary gland tumours (Barker 1984). NHL can be classified according to

the cell of origin into T-cell or B-cell lymphoma; the latter being more common in the oral

cavity (Neville et al. 2009). The incidence of NHL is rare in patients under the age of 40,

with approximately 70% of all cases being diagnosed in people over 60 years. It

predominately occurs in lymph nodes but 20–40% arise in extra-nodal sites (Neville et al.

2009). The gut is the commonest site for extra-nodal lymphoma, but bone and the mouth

are other frequently affected sites.

Non-Hodgkin’s lymphoma can present in a number of different forms within the oral

cavity, the more frequent being palatal (Tomich and Shafer 1975) and gingival swellings

(Spatafore et al. 1989, Payne and al-Damouk 1993). It is reported that 36–45% of oral NHL

can affect the jaw bones (Keyes et al. 1988). The initial diagnosis of oral lymphomas can

be challenging as they may resemble pyogenic granulomas, ulcers, sinusitis (Spatafore et

al. 1989), a non-healing socket (Thomas et al. 1991) or mimic an acute dental abscess

(Spatafore et al. 1989, Rog 1991, Payne & al-Damouk 1993). Patients may complain of

non-specific pain, which may be misdiagnosed as periapical inflammatory disease.

This report presents a case of malignant NHL which was originally diagnosed and

treated as an odontogenic infection.

Case report

A 38 year-old Afro-Caribbean female referred herself to the Birmingham Dental Hospital,

UK primary care unit in November 1998. She described a 6-month history of a

spontaneous intermittent dull ache in the upper left canine region. There was no

disturbance to her sleep pattern. She had visited her GDP on several occasions over the

preceding 6 months without resolution of her discomfort, despite extraction of tooth 22

and root canal treatment to teeth 23 and 24. Her medical history was unremarkable. She

was a non-smoker and drank 2 units of alcohol per week.

On clinical examination, there was no lymphadenopathy of the head and neck region.

An intra oral inspection revealed a non-healing socket where the 22 had been removed

4 months previously. The buccal sulcus was tender to palpation over the apex of the 23,

but no swelling or ulceration was apparent. The 23 and 24 were not mobile and were non-

tender to percussion. She had a good standard of oral hygiene and there was no evidence

of periodontal disease.

A periapical radiograph revealed a radiolucency and a dense radiopaque foreign body,

probably amalgam, at the base of the 22 socket (Fig. 1). The 23 and 24 (Fig. 2) had

satisfactory root canal fillings with no associated apical radiolucencies and good

periodontal support. She was prescribed a course of amoxicillin and reviewed a week

later. As her symptoms had not improved the 22 socket was surgically investigated.

During this procedure, granulation tissue was removed but not submitted for histopa-

thology.

Curettage of the 22 socket and a further two courses of amoxicillin failed to resolve her

symptoms so the 23 and 24 region was investigated and an apicectomy performed on

both teeth. Soft tissue was curetted from around the apices of 23 and 24 resulting in an

oro-antral communication due to loss of bone. The histology revealed chronically inflamed

granulation tissue.

Three months after her presentation to the dental hospital and 9 months following

onset of her symptoms, radiographic follow-up revealed destructive bony changes (Fig. 3).

The 23 and 24 apical radiolucency had increased in size, showed perforation of the cortical

bone and loss of the bony antral floor (Fig. 3).

As a consequence, a further biopsy was performed, which revealed extensive soft

tissue replacement of the left maxillary alveolar process extending from close to the

CA

SE

RE

PO

RT


palatal midline, posteriorly to the left maxillary buttress and superiorly towards the floor of

the nose. Extraction of the 23 led to the simultaneous removal of the 24 encased in a

loose segment of surrounding maxillary alveolus.

The histology of the soft tissue component again showed a granulomatous foreign

body reaction consistent with a periapical granuloma, whereas examination of the

tissue block containing bone and teeth revealed a dense infiltrate of cells that had the

appearance of malignant lymphoid cells. A preliminary diagnosis of lymphoma was

made and the patient referred for further assessment and management. A comput-

erized tomography (CT) scan showed destruction of the left maxillary alveolus (Fig. 4).

Further biopsy confirmed a diagnosis of non-Hodgkin’s lymphoma of the left maxilla.

Additional investigations showed that there were no other lesions elsewhere in the

body and thus the disease was classified stage 1AE (single extranodal site without

systemic signs of disease). She was treated with radiotherapy to the left maxilla and

chemotherapy. Seven years later the patient has no signs of recurrence and remains

under annual review.

Discussion

Inflammatory processes of the jaws may present in an indolent manner or show a more

aggressive behaviour. Typically odontogenic inflammation results in pain, widening of the

periodontal ligament space, and the development of a periapical radiolucency that is

usually well defined. Occasionally developmental anomalies, metabolic diseases and

malignancies can resemble dental inflammatory disease but do not respond to root canal

treatment or tooth extraction. In this situation the clinician should review the accuracy of

the diagnosis so that the appropriate treatment is not delayed.

Figure 1 Periapical radiograph taken at presentation of UL2 socket demonstrating foreign body.

CA

SE

RE

PO

RT


The presented case and several others in the literature have demonstrated that

extranodal lymphoma of the jaws may initially present, particularly in the early stages, with

unspecific signs and symptoms mimicking periapical disease (Slootweg et al. 1985,

Macintyre 1986). Lymphomas can become secondarily infected and present with swelling

mimicking a dental abscess (Rog 1991, Bavitz et al. 1992, Ardekian et al. 1996). The initial

clinical impression of inflammatory disease was supported when antibiotic therapy

appeared to reduce symptoms (Keyes et al. 1988). Whilst many malignant lesions are

easily recognized there are situations when they resemble other conditions. Although

Figure 2 Periapical radiograph taken at presentation showing root fillings UL34 and periradicular

bone.

Figure 3 Periapical radiograph taken 3 months after presentation showing destructive features with

cortical bony destruction above UL3.

CA

SE

RE

PO

RT


certain clinical features such as increased tooth mobility in the absence of advanced

periodontal disease and neurosensory disturbances may point towards non-odontogenic

disease, they may not be present initially (Gusenbauer et al. 1990). Similarly, radiographs

used to investigate dental disease may demonstrate findings such as poorly defined or

‘moth-eaten’ osteolytic lesions (Macintyre 1986), root resorption and erosion of crestal

bone, which are not typical for odontogenic lesions. However, destructive radiographic

changes may not be evident in slow growing lymphomatous lesions of the jaws (Keyes

et al. 1988, Rog 1991).

Malignant disease involving bone can resemble periapical inflammatory disease

particularly when the latter is infected changing its margin so it is less well defined.

Periapical inflammation is common whilst lymphoma in the jaws is not thus one tends not

to consider it in the diagnosis. It is important to review the clinical features and

radiological findings, and when these are unusual the diagnosis needs to be reconsidered

rather than persisting with inappropriate treatment. In addition, early referral to a

secondary setting for specialist opinion must always be considered. However, cases such

as the one presented here have to be cautiously interpreted with respect to missed

clinical and radiographic signs, as post-hoc interpretation may be misleading because

primary manifestation of lymphoma or other malignancies as periapical pathology is

uncommon. Therefore, a significant proportion of cases presenting with one or more of

the ‘atypical’ signs and symptoms discussed above may still represent odontogenic

pathologies.

Whilst the authors cannot comment on the indications for, and the sequence of

treatment undertaken by the general practitioner during the 6-month period prior to the

patient’s attendance in the hospital, it is quite clear that several therapeutic attempts

aimed at what was thought to be an odontogenic problem had failed. When initially seen

at our clinic, the non-healing extraction socket was attributed to the foreign body visible

on the radiograph and surgically revised without obtaining any material for histopath-

ological examination. Although an unlikely cause for the patient’s symptoms, it is not

unreasonable to remove a foreign body from a non-healing socket. However, the long

history and failure to respond to previous treatment should have raised the suspicion

that the condition was not infective and a tissue sample for histopathological

examination should have been retrieved (Rog 1991). In the present case it is however

unclear if this would have resulted in an earlier diagnosis, particularly when the

Figure 4 Axial CT image showing erosion of left maxillary alveolus following diagnosis of lymphoma.

CA

SE

RE

PO

RT


subsequent histology suggested an inflammatory process and the age of the patient was

younger than one expects for lymphoma.

Histologically, the distinction between lymphoma and periapical inflammation is often

challenging, and as in this and other cases, lymphoma has been interpreted as being

inflammatory in nature (Keyes et al. 1988, Richards et al. 2000). These difficulties have

been attributed to inadequate biopsy specimens and poor handling of the tissue by the

clinician leading to ‘crush artefact’ which obscures the fine cytological detail needed to

distinguish between benign and malignant lymphocytes. It is often difficult to obtain an

adequate tissue sample because of the close location to roots (Wannfors and Hammar-

strom 1990). To increase the chances of accurate diagnosis large specimens represen-

tative of the tumour are required (Rog 1991). One biopsy may be insufficient to make a

diagnosis and re-biopsy of non-healing lesions including bone within the sample may be

required. The diagnostic difficulty increases when lymphomas become inflamed,

obscuring the neoplastic nature of the infiltrate (Wright and Radman 1995). Indeed, the

soft tissue samples retrieved during the course of treatment were found to be consistent

with a chronic inflammatory lesion. The histopathological diagnosis of lymphoma was

made from the hard tissue block accidentally retrieved during the extraction of the

associated teeth. This illustrates that, in order to obtain a correct diagnosis earlier, a block

biopsy of the affected bone may have been required to yield the true nature of the disease

process.. However, this constitutes a rather invasive, if not destructive procedure, and

given that lymphoma is extremely rare, adoption of a practice of ‘early block biopsy’ would

result in unnecessary morbidity in many cases. However, in light of diagnostic difficulties

with histopathological examination and plain film radiography earlier referral for more

advanced imaging techniques such as computed tomography or magnetic resonance

imaging should have been considered.

Over the course of her treatment, the patient received three courses of antibiotics

without resolution of her symptoms. Antibiotics have no role in the treatment of persistent

non-healing sockets. Lymphomas and other non-odontogenic diseases may become

secondarily infected. In this situation a reduction of symptoms with antibiotic treatment

may delay proper diagnosis (Keyes et al. 1988).

Although extranodal lymphoma of the jaws is uncommon, perhaps with the increasing

incidence of HIV infection (UNAIDS 2008), a continued increase in the incidence of

lymphoma will be observed. As demonstrated, lymphoma may masquerade as common

dental inflammatory disease and clinicians should be alert to the possibility of sinister

pathology. In the study by Maxymiw et al. 2001 a high percentage of patients with NHL

had dental symptoms. These cases often demonstrate recurrent or protracted disease

patterns, as seen in this case with a delay of 9 months before making the correct

diagnosis. This is longer than the average 2.5 months between presentation and

treatment reported in other cases (Gusenbauer et al. 1990). NHL of the head and neck

has a good prognosis with a median survival rate of 10–15 years but the prognosis is

improved with early diagnosis (Payne & al-Damouk 1993).

Conclusion

Despite its rare occurance, dentists must consider lymphoma in the differential diagnosis

of pain, swelling, ulceration and non-healing periapical inflammation. In general, dentists

should have a high index of suspicion for lesions (including periapical lesions) that do not

respond to conventional therapy or appear unusual in other ways and as such have a role

in early diagnosis and prompt referral of patients for specialist secondary care. Finally, the

possibility of false negative biopsy results must be considered and referral to specialist

care may be warranted even in a case of a negative initial biopsy result.

CA

SE

RE

PO

RT


Disclaimer

Whilst this article has been subjected to Editorial review, the opinions expressed, unless

specifically indicated, are those of the author. The views expressed do not necessarily

represent best practice, or the views of the IEJ Editorial Board, or of its affiliated Specialist

Societies.

References

Ardekian L, Peleg M, Samet N, Givol N, Taicher S (1996) Burkitt’s lymphoma mimicking an acute

dentoalveolar abscess. Journal of Endodontics 22, 697–8.

Barker GR (1984) Unifocal lymphomas of the oral cavity. British Journal of Oral & Maxillofacial Surgery

22, 426–30.

Bavitz JB, Patterson DW, Sorensen S (1992) Non-Hodgkin’s lymphoma disguised as odontogenic pain.


Cancer Research UK (2008a) UK Non-Hodgkin lymphoma incidence statistics. http://info.cancerre-

searchuk.org/cancerstats/types/nhl/incidence/?a=5441 [Accessed on December 6, 2008].

Cancer Research UK (2008b) UK Non-Hodgkin Lymphoma (NHL) statistics. http://info.cancerresear-

chuk.org/cancerstats/types/nhl/?a=5441 [Accessed on December 6, 2008].

Gusenbauer AW, Katsikeris NF, Brown A (1990) Primary lymphoma of the mandible: report of a case.

Journal of Oral and Maxillofacial Surgery 48, 409–15.

Keyes GG, Balaban FS, Lattanzi DA (1988) Periradicular lymphoma: differentiation from inflammation.

Oral Surgery Oral Medicine Oral Pathology 66, 230–5.

Macintyre DR (1986) Lymphomas of the mandible presenting as acute alveolar swellings. British

Dental Journal 161, 253–4.

Maxymiw WG, Goldstein M, Wood RE (2001) Extranodal non-Hodgkin’s lymphoma of the maxillofacial

region: analysis of 88 consecutive cases. South African Dental Journal 56, 524–7.

Neville BW, Damm DD, Allen CM, Bouquot JE (2009) Hemotologic Disorders. In: BW Neville, DD

Damm, CM Allen, JE Bouquot eds. Oral and Maxillofacial Pathology, 3rd edn. St Louis: Saunders

Elsevier, pp. 592–8.

Payne M, al-Damouk JD (1993) Gingival swelling as a manifestation of non-Hodgkin’s lymphoma.

British Dental Journal 175, 293–4.

Richards A, Costelloe MA, Eveson JW, Scully C, Irvine GH, Rooney N (2000) Oral mucosal non-

Hodgkin’s lymphoma – a dangerous mimic. Oral Oncology 36, 556–8.

Rog RP (1991) Beware of malignant lymphoma masquerading as facial inflammatory processes. Oral

Surgery Oral Medicine Oral Pathology 71, 415–9.

Slootweg PJ, Wittkampf AR, Kluin PM, de Wilde PC, van Unnik JA (1985) Extranodal non-Hodgkin’s

lymphoma of the oral tissues. An analysis of 20 cases. Journal of Maxillofacial Surgery 13, 85–92.

Spatafore CM, Keyes G, Skidmore AE (1989) Lymphoma: an unusual oral presentation. Journal of


Thomas DW, Gray W, Tate RJ (1991) Non-Hodgkin’s lymphoma presenting at the site of a recent

dental extraction: a report of two cases. British Journal of Oral and Maxillofacial Surgery 29, 34–7.

Tomich CE, Shafer WG (1975) Lymphoproliferative disease of the hard palate: a clinicopathologic

entity. A study of twenty-one cases. Oral Surgery Oral Medicine Oral Pathology 39, 754–68.

UNAIDS (2008) North America, Western and Central Europe: AIDS epidemic updates regional

summary. http://data.unaids.org/pub/Report/2008/jc1532_epibriefs_namerica_europe_en.pdf [Ac-

cessed on December 6, 2008].

Wannfors K, Hammarstrom L (1990) Periapical lesions of mandibular bone: difficulties in early

diagnostics. Oral Surgery Oral Medicine Oral Pathology 70, 483–9.

Wright JM, Radman WP (1995) Intrabony lymphoma simulating periradicular inflammatory disease.

Journal of American Dental Association 126, 101–5.

CA

SE

RE

PO

RT


Documents

Inernational tic Journal IEJ 2010 Issue January