Implant-guided vertical bone growth in

Implant-guided vertical bone growth inthe mini-pig

Martin FreilichBo WenDavid ShaferPeter SchleierMichel DardDavid PendrysDenise OrtizLiisa Kuhn

Authors’ affiliations:Martin Freilich, Liisa Kuhn, Bo Wen, Denise Ortiz,Department of Reconstructive Sciences, Center forBiomaterials, School of Dental Medicine, University ofConnecticut, Farmington, CT, USADavid Shafer, Department of Craniofacial Sciences,Division of Oral & Maxillofacial Surgery, School ofDental Medicine, University of Connecticut,Farmington, Ct, USADavid Pendrys, Department of ReconstructiveSciences, School of Dental Medicine, University ofConnecticut, Farmington, CT, USAPeter Schleier, Oralfacial Surgery Section, Clinic forSpecial Medicine, Stavanger, NorwayMichel Dard, Institut Straumann, Basel, Switzerland

Corresponding author:Dr Martin FreilichDepartment of Reconstructive SciencesSchool of Dental MedicineUniversity of Connecticut263 Farmington AvenueFarmingtonCT USATel.: 860 679 2649Fax: 860 679 1370e-mail: [email protected]

Key words: biomaterial scaffold, ng/rhBMP-2, scaffold retainer, vertical supracrestal bone growth

Abstract

Objective: To attain and describe guided vertical bone regeneration around titanium (Ti) and titanium

zirconium (Ti–Zr) dental implants utilizing non-glycosylated recombinant human bone morphogenetic

protein-2 (ng/rhBMP-2), biomaterial scaffolds and a scaffold retainer.

Materials and methods: Thirty-two modified Straumann TE implants were partially embedded in the

mandibles of eight adult mini-pigs. Pre-shaped resorbable scaffolds were placed around the implant and

shielded and stabilized with a newly developed Ti custom scaffold retainer (umbrella) or wide-neck (WN)

healing caps to stabilize the scaffold. Ng/rhBMP-2 (50mg) was applied to the supracrestal portion of the

implant or incorporated within the scaffold. At 9 weeks, soft tissue healing was assessed. Vertical bone

regeneration outcomes including bone height, bone-to-implant contact (BIC) and bone volume were

assessed by micro-computed tomography and histology.

Results: Soft tissue healing at the test sites (þng/rhBMP-2/þ scaffold) appeared to be substantially

better than the control sites (�ng/rhBMP-2/� scaffold). Bone height, BIC percentage and bone volume

were all similar regardless of whether WN healing caps or umbrella scaffold stabilization was used for all

biomaterial scaffolds tested. WN healing cap test sites showed greater new bone height and BIC as

compared with aggregate data from the control sites (P¼ 0.05). Comparison of aggregate data from the

umbrella test sites showed greater BIC and new bone volume as compared with aggregate data from the

control sites(P¼ 0.05).

Conclusion: Vertical bone regeneration was successfully attained utilizing ng/rhBMP-2, biomaterial

scaffolds and a scaffold retainer.

Current clinical methods for increasing vertical

bone height include grafting from intra- and

extra-oral sites and distraction osteogenesis. A

successful outcome is site dependent and gener-

ally associated with substantial levels of morbid-

ity (Feichtinger et al. 2007). The overall goal of

the study described here is to use dental implants,

implant components and osteoinductive factors

to guide a new layer of bone height while mini-

mizing the technique sensitivity, risks and pro-

blems seen with current clinical methods. The

development of methods to predictably regener-

ate 2–3 mm of additional bone height of good

density and width would have important clinical

implications at both anterior and posterior sites.

The simultaneous placement of a partially in-

serted dental implant, an inductive agent and a

resorbable three-dimensional scaffold has been

attempted to treat the vertical defect in experimen-

tal dog/large animal models with varying degrees

of success (Jovanovic et al. 1995; Renvert et al.

1996; Sigurdsson et al. 1997; Roos-Jansaker et al.

2002). For this approach, inductive agents have

included allograft materials (Jung et al. 2008) or

growth factors such as bone morphogenetic pro-

tein-2 (BMP-2).

A demineralized bone matrix (DBM) allograft

has osteoinductive qualities and can serve as the

three-dimensional scaffold to support new tissue

growth. DBM is able to maintain space for bone

growth, exhibit a framework for cell and matrix

protein adhesion, and has the potential to deliver

osteogenic cell signaling agents. The osteoinduc-

tive value of DBM was recognized almost four

decades ago after its placement into muscle

pouches of animals and the subsequent formation

of ectopic bone (Urist 1965; Lindholm et al. 1988).

It has been shown that the osteogenic components

Date:Accepted 14 March 2011

To cite this article:Freilich M, Wen B, Shafer D, Schleier P, Dard M, Pendrys D,Ortiz D, Kuhn L. Implant guided vertical bone growth in themini-pig.Clin. Oral Impl. Res. xx, 2011; 000–000doi: 10.1111/j.1600-0501.2011.02199.x

c� 2011 John Wiley & Sons A/S 1

mailto:[email protected]

of DBM are a mixture of growth factors including

many from the transforming growth factor-b fa-

mily (Urist & Strates 1971) such as BMP-2.

Recombinant human (rh)BMP-2 is a growth

factor that has been used frequently to induce

bone growth in a variety of experimental animal

models (Wikesjo et al. 2008). Furthermore, rhBMP-

2 is now FDA approved and commercially available

for use in the jaws of humans. Most previously

reported work utilized a variety of dosages of the

glycosylated formulation of this agent. For our

studies, we chose to use the non-glycosylated

rhBMP-2 (ng/rhBMP-2) version of this molecule,

which has reduced solubility and inherent delayed

release relative to glycosylated rhBMP-2 (Sampath

et al. 1990; Schmoekel et al. 2004; Sachse et al.

2005). Ng/rhBMP-2 shows promise in terms of

allowing a reduction of BMP-2 doses necessary for

the efficient induction of bone (Woo et al. 2001),

which is important, given the possible safety issues

associated with BMP-2 (Poynton & Lane 2002).

A non-resorbable titanium (Ti) implant surface

can be used to carry an osteogenic agent such as

BMP-2 to enhance local bone formation and os-

teointegration (Hall et al. 2007; Leknes et al. 2008;

Wikesjo et al. 2008). Ng/rhBMP-2 has been re-

leased from miniaturized Ti implant components

to grow a robust new layer of bone in mouse

calvaria (Freilich et al. 2008). Synthetic resorbable

biomaterial scaffolds may also be used effectively

to release an agent to guide bone growth. Healos, a

collagen/hydroxyapatite (HA) composite scaffold

biomaterial (Jþ J, DePuy, Rayhnam, MA, USA),

is composed of cross-linked bovine Type I collagen

fibers coated with approximately 30% HA exhibit-

ing pore sizes of 4–200mm (Jahng et al. 2004).

Outcomes resembling the use of autogenous bone

grafts have been described when this scaffold was

used in conjunction with bone marrow aspirate or

recombinant human growth/differentiation factor-

5 (GDF-5) (Lutolf et al. 2003; Jahng et al. 2004;

Miranda et al. 2005; Neen et al. 2006).

Dental implant surface roughness and positive

charge (SLActivet, Institut Straumann, Basel,

Switzerland) are associated with osteopromotive or

osteoconductive function (Cochran et al. 1996).

These characteristics have been associated with

enhanced osteoblast function and matrix production

as well as increased levels of endogenous growth

factors in cell culture studies (Kieswetter et al. 1996)

and are well suited for guiding new bone growth.

Implants with these surface characteristics may be

machined from commercially pure Ti and more

recently a Titania–zirconia (Ti–Zr) alloy. Enhanced

design features used to stabilize the adjacent resorb-

able scaffold may help to guide vertical osteogenesis.

As a follow-up to earlier mouse studies, a newly

developed custom scaffold retainer (umbrella) was

used to successfully grow a new layer of bone in

rabbit mandibles (Freilich et al. 2009).

The present descriptive study builds upon

previous small-animal research by testing a vari-

ety of inductive agents and scaffolds in an in-

traoral large-animal model. The purpose was to

generate outcomes that would then influence the

therapy selection and inform sample size/power

calculation for subsequent studies. This study

explored the effect of two primary variables on

new bone growth in the mini-pig: (1) the type of

scaffold retainer used (wide–neck [WN] healing

abutment vs. a custom ‘‘umbrella’’ scaffold re-

tainer) and (2) the release of ng/rhBMP-2 from

pre-made fibrous scaffolds (DBM or Healos) or

dental implant surfaces.

Materials and methods

Study design

The study timeline can be seen in Fig. 1. This

study protocol was approved by the ethics com-

mittee for animal research at Lund University in

Sweden. Four implant sites were used in each of

eight adult Gottingen mini-pigs ranging in weight

from 30 to 50kg and 2 years of age. The study

design includes four study groups: one negative

control (two animals and eight implant sites) and

three treatment groups each with two animals and

four implant sites for a total of 32 study sites. Each

treatment group animal (groups 2, 3 and 4) pro-

vided a split-mouth comparison between oversized

WN healing caps and newly developed custom Ti

scaffold retainer (umbrella) abutments. The speci-

fic treatment groups can be seen in Table 1 and

they include: (1) Ti-negative controls without

BMP or scaffold; (2) Ti/BMPþDBM; (3) Ti–Zr/

BMPþHealos; and (4) Ti–ZrþHealos/BMP. The

effects of two primary independent variables were

evaluated. These were: (1) the use of an umbrella

vs. a WN healing cap to stabilize the biomaterial

scaffold and (2) the site of ng/rhBMP-2 release.

Secondarily, the use of Ti vs. a Ti–Zr alloy and

scaffold type was also evaluated. For all sites, the

dental implants were partially embedded in poster-

ior mandible sites such that the implant shoulder

was located 2.5 mm above the bone crest. The

Fig. 1. Time schedule schematic.

Table 1. Chronology of study procedures

Groupn Side A Side B

1 (n¼2) Ti SLA with Ti SLActive with a Controla WN healing cap WN healing cap

2 (n¼2) Ti implant/BMP (50 mg) Ti implant/BMP (50mg) TestDBM scaffold DBM scaffoldUmbrella WN healing cap

3 (n¼2) Ti–Zr implant/BMP (50 mg) Ti–Zr implant/BMP (50mg)Healos scaffold Healos scaffoldUmbrella WN healing cap

4 (n¼2) Ti–Zr implant Ti–Zr implantHealos scaffold/BMP (50 mg) Healos scaffold/BMP (50 mg)Umbrella WN healing cap

nTwo animals per group, two implants per side per animal.

Ti, titanium; BMP, non-glycosylated rhBMP-2; DBM: demineralized bone matrix; Ti–Zr, titanium zirconium; WN,

wide-neck; BMP, bone morphogenetic protein.

Freilich et al � Implant-guided bone growth in mini-pig

2 | Clin. Oral Impl. Res. 10.1111/j.1600-0501.2011.02199.x c� 2011 John Wiley & Sons A/S

system was designed to guide the regeneration of

new bone along the aspect of the implant surface

left outside of bone at the time of placement. Ng/

rhBMP-2 was used as an osteoinductive signaling

agent, which was released from either the osteo-

genic portion of the implant surface or the resorb-

able scaffold adjacent to the implant at the

supracrestal region. A titanium umbrella abutment

or a WN healing cap stabilized the scaffold and

provided protected space for new bone formation.

Implant design and manufacturing

The 9 mm long � 4.1 mm solid screw implants

made of commercially pure titanium or a tita-

nium–zirconium (Ti–Zr) alloy (Roxolidt) were

obtained from Institut Straumann. The implants

were cylindrically shaped, except for the coronal-

most 2.5 mm of the implant, which exhibited a

slight taper and a diameter of 4.8 mm at the head of

the implant. A roughened, sand-blasted, large grit,

acid-etched surface (SLAt) or a hydrophilic SLA

surface with a positive charge (SLActivet) was

placed to the finish line at the head of each

implant. There was no polished collar.

Custom-designed titanium scaffold retainer

abutments (umbrella) 9 � 9 mm in size, a thick-

ness of 1 mm with polished surfaces and four

‘‘L’’-shaped openings, and oversized WNHCs

with a diameter of 7 mm and a thickness of

2 mm were also obtained from Institut Strau-

mann. The design of the implant and umbrella

can also be seen in Fig. 1.

Experimental scaffolds

DBM allograft material using procedures similar

to those for producing a commercially available

human allograft material (Graftons

Flex) (Osteo-

tech, Eatontown, NJ, USA) and HA-coated col-

lagen (Healoss

) (DePuy Spine, Raynham, MA,

USA) were both prepared up to a size of

10 � 10 mm, thickness of 3 mm and a central

opening with a diameter of 5 mm, which allowed

for the scaffolds to be placed around the implants.

BMP-2 delivery

Escherichia coli expressed ng/rhBMP-2 was used

as an osteogenic signaling agent in these studies

as described previously in Sachse et al. (2005).

The three different ng/rhBMP-2 delivery systems

tested included a resorbable scaffold (adjacent to

the SLActive surface Zr–Ti implant) and two

different implant surfaces adjacent to non-BMP

carrying scaffolds.

For BMP-2 released from Ti and Zr–Ti SLActive

implant surface (Ti/BMP), lyophilized ng/rhBMP-2

powder was reconstituted in 50% acetonitrile/

0.1% trifluoroacetic acid (ATFA) at a concentra-

tion of 5 mg/ml and placed aseptically onto the

osteogenic aspect (coronal most 2.5 mm) of an

SLActive-treated implant by evenly wetting the

surface of each implant with 10ml of the ng/

rhBMP-2 solution. All samples were dried under

a laminar air flow in a biological safety cabinet.

Each Ti SLActive implant (100mg/cm2 surface)

had an ng/rhBMP-2 dosage of 50mg. As the volume

of the Healos is ten times larger than that of Ti or

the Zr–Ti implant surface, a stock ng/rhBMP-2/

ATFA Solution (5 mg/ml) was diluted to a con-

centration of 1 mg/ml and 50ml of this solution

was applied to the Healos with a Hamilton syringe

(Hamilton, Reno, NV, USA) to obtain the

same dose as that on the Ti or the Zr–Ti implant.

Each scaffold (200 ng/mm3 scaffold) had an ng/

rhBMP-2 dosage of 50mg. All samples were dried

under a laminar airflow in a biological safety

cabinet.

Surgical procedure

The animals were pre-medicated with an intra-

muscular injection of atropine (atropinum sulfur-

icum, 0.05 mg/kg IM) and then given a

combination of ketamine (10 mg/kg, Ketalar

Vet 50 mg/ml, Pfizer AB, Sollentuna, Sweden)

and midazolam (0.5 mg/kg Dormicums

5 mg/

ml: Roche, Basel, Switzerland) for general an-

esthesia. During the surgery, ketamine (10 mg/

kg) and midazolam (0.5 mg/kg) were reinjected

when needed. All animals received 1.8 ml local

anesthesia (Xylocain Dental adrenalin 20 mg/

mlþ12.5 mg/ml, Astra AB, Sodertalje, Sweden)

at each surgical site just before surgery.

The surgical procedures can be seen in Fig. 2

and were performed 6 months after the extraction

of the second, third and fourth pre-molars and the

Fig. 2. Clinical photographs of the surgical procedure. (a and b) Two implants placed in each side of the posterior mandible after

a gentle reduction of the alveolar crest to create a flat surface. Implants placed with the shoulder located 2.5 mm above bone level.

(c, d and e) Scaffolds placed around the supracrestal part of the implant and covered by a umbrella or a wide-neck healing abutment.


c� 2011 John Wiley & Sons A/S 3 | Clin. Oral Impl. Res. 10.1111/j.1600-0501.2011.02199.x

first molar. Incisions were made with a 15-size

blade on the edentulous alveolar crest to allow

full-thickness flap elevation. Osteotomy and

implant placement procedures followed the sur-

gical protocol described elsewhere (Dard et al.

2008). The custom-designed 9 � 4.1 mm im-

plants were placed bilaterally to a depth of

6.5 mm, leaving 2.5 mm of each implant beyond

the level of bone. Partial bone decortication

surrounding the implant was made at four evenly

placed sites to enhance blood clot development

and access to intraosseous vasculature and cells.

For test sites with scaffolds, the scaffold was

placed in direct contact with the underlying

partially decorticated bone and maintained in

position by a custom-designed scaffold retention

screw or a WN healing cap fastened to the screw

channel at the head of each implant. Once all

elements of the implant system construct were

placed, the mucoperiosteal flap was closed with

4.0 suture material (Vicryl, Ethicon Inc. Johnson

& Johnson company, Somerville, NJ, USA).

A veterinary technician monitored all vital signs

of each animal during surgery. Postsurgery, all

animals were placed on a soft diet.

Terminal procedures and soft tissue evaluation

All animals were sacrificed 9 weeks after dental

implant placement by inducing a cardiac arrest

with an intracardiac injection of a 20% solution

of pentobarbital (Pentobarbitalnatrium, Apoteket

AB; Stockholm, Sweden, 60 mg/ml) after general

anesthesia. Soft tissue healing was quantified

with scores of 0 (no dehiscence) to 4 (complete

dehiscence) as can be seen in Fig. 3. The mand-

ibles were excised and the left and right hemi-

mandibles were separated with a band saw, fixed

(4% formaldehyde) for 2 weeks and prepared for

micro-computed (micro-CT) tomography ima-

ging and histological processing by transfer into

an ethanol solution.

Micro-CT evaluation

Following fixation, all samples were imaged

three-dimensionally using conebeam X-ray mi-

cro-CT (CT80, Scanco Medical AG, Bassersdorf,

Switzerland). A trilinear interpolation method

was implemented to rotate the micro-CT images.

This resulted in alignment of the implant with

the z-axis in voxel space and acted as a smoothing

filter. Serial tomographic images were acquired

transverse to the implant longitudinal axis at

90 kVp and 77mA, collecting 1000 projec-

tions per rotation at 400 ms integration time.

Three-dimensional cone beam images were re-

constructed with 50 mm nominal resolution (iso-

tropic). The images were segmented to separate

the implant and bone from the background using

a global thresholding procedure. Newly regener-

ated supracrestal bone height and volume were

measured directly from the segmented images.

Histopathologic and histomorphometric analyses

To identify the position of the sites, X-rays were

performed before histological preparation. The

samples were dehydrated in alcohol of increasing

concentrations, cleared in xylene and embedded

in polymethylmetacrylate resin. For each hemi-

mandible, one sagittal mesio-distal and two

bucco-lingual histological sections were prepared.

The sections were obtained using a micro-cutting

and grinding technique adapted from Donath

(Donath & Breuner 1982). The sections were

then stained for qualitative and quantitative his-

Fig. 3. Soft tissue summary and scores (groups 2, 3 and 4 – Umbrella vs. wide-neck healing cap [WNHC]).

Fig. 4. Representative histology images (groups 2, 3 and 4 – Umbrella vs. wide-neck healing cap [WNHC]).



tology with a modified polychromatic Paragon

staining. The histological sections were observed

using a Nikon microscope (Eclipse E600, Nikon,

Melville, NY, USA) fitted with � 2, � 4, � 10,

� 20 and � 40 lenses. Photomicrographs were

performed. The stained ground sections were

observed using a Zeiss Axioscope microscope

(Thornwood, NY, USA) fitted with � 5, � 10,

� 20 and � 40 objectives and equipped with a

color image-analyzing system SAMBAs

(SAMBA

Technologies, Chatillon, France). The outcomes

adjacent to each implant were assessed by analyz-

ing bone growth, osseointegration and the

presence or absence of neovascularization. Quali-

tative and semi-quantitative analyses were per-

formed to measure the height of supracrestal bone

at the implant surface (mm), which was measured

from just external to the cortical surface of the

underlying bone to the highest point of bone-to-

implant contact (BIC) % to new bone. Addition-

ally, the BIC was determined at both the osteo-

genic (supracrestal bone) and the anchoring

(subcrestal) portions of the implant within one

mm of the implant surface.

Data analyses

Data analyses were performed using SPSS soft-

ware, version 18 (IBM Corporation, Armonk,

NY, USA). Descriptive statistics, including

means, standard deviations and distribution

shape, were generated for all outcome measures.

The unit for all analyses was the mini-pig. As

each mini-pig had two implants per side, the

average values for each implant pair were used

for all analyses. Between-group analyses were

conducted for both micro-CT and histology out-

comes using the Mann–Whitney U-test for in-

dependent samples.

Results

Uneventful healing and recovery of all animals

followed the surgical procedures. All outcomes

were assessed at 9 weeks postimplant placement.

All of the negative control sites had a soft tissue

dehisence whereas only four of the 24 of the

implants with BMP and scaffold (groups 2, 3, 4)

sites had dehisence. The semi-quantitative

findings and a summary of the scores for soft tissue

dehiscence comparing umbrella and WN healing

caps can be seen in Fig. 3. There did not appear to

be any meaningful difference between umbrella and

WN healing cap sites for any of the treatment pairs.

Qualitiative histology showed that the newly

formed bone tissue was characterized by woven

bone covered by a thin layer of cortical bone. The

newly formed bone merged with the original

alveolar bone, with substantial vasculature infil-

trating the marrow and bone tissues. At the

cellular level, osteoblasts actively depositing

bone were observed throughout the newly formed

bone as were few inflammatory cells. Represen-

tative histology sections of ‘‘test’’ treatment

groups 2, 3 and 4 all showing substantial vertical

bone regeneration can be seen in Fig. 4.

Bone height

Fig. 5a shows new bone height data as deter-

mined via quantitative histology. The findings

for negative control group sides 1A and 1B were

similar and thus combined. Umbrella and WN

healing cap findings within the test sites (groups

2, 3 and 4) were also generally similar. A com-

parison of aggregate data for mean new bone

height (mean¼0.62 mm, SD¼0.07) from the

negative control sites (n¼2) with aggregate data

from BMP/scaffold test sites in groups 2, 3 and

4 that used a WN healing cap (n¼6) demon-

strated more than a three-fold increase for the test

sites (mean¼1.97 mm, SD¼ .74, P¼0.04).

New mean bone height for umbrella test sites

in groups 2, 3 and 4 (n¼6) was 1.75 mm

(SD¼0.28), slightly less than a threefold increase

as compared with the control sites (P¼0.05).

BIC percentage

Fig. 5b shows BIC data as determined via histol-

ogy. As with bone height, the BIC outcomes for

negative control group sides 1A and 1B were

similar and thus combined. All active treatment

sites (groups 2, 3 and 4) showed similar BIC

values. New mean BIC for aggregate data from

the WN healing cap test sites for groups 2, 3 and

4 (mean¼49.33, SD¼9.22) was 62% greater

than that for the aggregate data of the control

(group 1A and 1B) sites (mean¼30.5%, SD¼0.71)

(P¼0.04). New mean BIC for the group 2, 3 and 4

umbrella test sites (mean¼47.83%, SD¼8.18)

was 57% greater than the negative controls

(P¼0.04).

Bone volume

Fig. 5c shows new mean bone volume data as

determined via micro-CT. As with bone height

and BIC, bone volumes for negative control group

sides 1A and 1B were similar and were thus

combined. Umbrella and WN healing cap find-

ings within the test treatment sites (groups 2, 3

and 4) showed similar new mean bone volume.

New mean bone volume for aggregate data from

the WN healing cap test sites for groups 2, 3 and

4 (mean¼106.41 mm3, SD¼22.39) was 65%

greater than that of the aggregate data of the

control groups (mean¼64.31, SD¼14.70)

(P¼0.1). New mean bone volume for the group

2, 3 and 4 test sites with umbrella was 59%

greater (mean¼102.39 mm3, SD¼13.24) than

that seen with the negative controls (P¼0.05).

Representative micro-CT images for the control

and the active treatment groups, 2, 3 and 4 can be

seen in Figs 6 and 7, respectively.

Discussion

In this descriptive study, implant-guided bone

regeneration in association with an osteogenic

Fig. 5. Histology and micro-CT Outcomes, (a) bone height as measured by histology, (b) bone-to-implant contact percent as

measured by histology, (c) bone volume as determined by micro-CT.



agent, ng/rhBMP-2, a biomaterial scaffold and a

scaffold retention device was evaluated for the

first time in a large-animal intraoral mini-pig

model. Previously, successful results were ob-

tained when testing this dental implant bone

regeneration system in several small-animal

models (e.g. mouse, rat and rabbit models [Frei-

lich et al. 2008, 2009]); however, those models

were not intraoral models. The present study

shows that the release of ng/rhBMP-2 from the

dental implant surface or scaffold materials in

combination with the use of a scaffold-stabilizing

device reliably induced approximately 2 mm of

vertical bone growth of good density and width

along the length of the dental implant. Many

clinical situations requiring implant placement

would best be managed by the addition of new

alveolar bone height and volume. Based on these

studies, the use of a custom scaffold retainer

(umbrella) or a WN healing abutment to stabilize

and retain pre-made fibrous scaffolds (DBM and

Healos) in the presence of ng/rhBMP-2 around Ti

and Ti–Zr implants appears to be a suitable

option to increase alveolar bone height. In this

study, histology and micro-CT analysis deter-

mined that the negative control sites (�BMP

� scaffold) utilizing WN healing abutments and

implants without a scaffold and ng/rhBMP-2

regenerated very little new bone. In contrast,

the test groups with ng/rhBMP-2 and any

of the resorbable scaffold biomaterials (þBMP

þ scaffold) were able to effectively support a new

layer of wide, dense vertical bone growth.

The findings of this study suggest that supra

alveolar bone can be grown with a high level of

predictability as demonstrated by new bone aug-

mentation across all of our umbrella and WN

healing cap test groups. Quantitative histology

confirmed that new supracrestal bone growth for

all þBMPþ scaffold groups was of good quality

and density and was well adapted to the Ti and

Ti–Zr implant surfaces. Micro-CT showed that

considerable new bone volume regenerated at all

of these test sites. The data further show that the

type of scaffold stabilizing device (WN healing

cap vs. umbrella) and site of ng/rhBMP-2 release

(implant surface vs. scaffold) had minimal prac-

tical influence on bone height, BIC percentage or

bone volume at the test sites. A striking differ-

ence was also seen in soft tissue healing for the

experimental sites in contrast to the control sites.

All of the control sites, which had neither BMP nor

a scaffold (group 1A and 1B), exhibited soft tissue

dehisence, whereas only four of the 24 sites with

BMP and scaffold (groups 2, 3, 4) had dehisence.

No practical difference was seen in the osteoinduc-

tive efficacy of bone formation between two types

of scaffold holding devices in this study, which

used fibrous, pre-made scaffolds. This might be

explained by the fact that the 6.5mm WN healing

cap also served as a scaffold retention screw to

maintain space and keep scaffold biomaterials

stable, while at the same time providing protection

from soft tissue collapse and traumatic overload-

ing. The additional size and extension of the

umbrella beyond that of the WN healing abutment

would allow it to confer effective retention and

stability to particulate scaffold materials that may

be selected for use in future studies.

Previous studies have made attempts to utilize

dental implants in the absence of autogenous

grafts for the purpose of guiding vertical bone

growth. These studies have used occlusive mem-

branes such as polytetrafluoroethylene (PTFE)

with and without titanium reinforcement to

maintain space and prevent soft tissue cells

from repopulating the area of attempted bone

regeneration. Unsuccessful attempts have in-

volved the use of a blood clot in the absence of

a scaffold or an agent to induce new bone forma-

tion in a dog model (Roos-Jansaker et al. 2002;

Polimeni et al. 2005). Later testing utilizing the

same approach, but with the addition of glycosy-

lated BMP-2, produced a narrow dimension of

vertical bone growth of low density (Wikesjo

et al. 2008). Studies in the dog model (Jovanovic

et al. 1995; Renvert et al. 1996; Sigurdsson et al.

1997) and case reports in humans (Tinti et al.

1997; Simion et al. 2007) have reported better

outcomes for this approach. These previous stu-

dies utilize the technique-sensitive PTFE mem-

brane, which is likely to result in the failure of

new bone regeneration when soft tissue dehis-

cence results in membrane exposure.

The present study utilized two osteoinductive

carrier systems: (1) a rough Ti surface and (2)

Healos. It has been demonstrated previously that

21mg of ng/rhBMP-2 released from the roughened

(SLA) surface of a non-resorbable Ti scaffold can

induce substantial bone formation in a murine

calvarial model (Freilich et al. 2008). The positive

outcomes with 50mg of the ng/rhBMP-2 released

from rough Ti surfaces seen in the current study

are consistent with the outcomes achieved in an

earlier canine intraoral study that demonstrated

alveolar ridge augmentation using 200 or 600mg

of glycosylated rhBMP-2 coated on the titanium

porous-oxide surface (Leknes et al. 2008; Wikesjo

et al. 2008). The use of ng/rhBMP-2 allowed

positive results to be achieved using a substan-

tially lower dosage. This is confirmed in work by

others utilizing release from a fibrin matrix

(Schmoekel et al. 2005). Healos has been suc-

cessfully used in previous work as a scaffold in

Fig. 7. Representative microcomputed tomography images (groups 2, 3 and 4 – Umbrella vs. wide-neck Healing cap

[WNHC]). The green line represents the approximate original bone level.

Fig. 6. Representative microcomputed tomography images (group 1). WNHC, wide-neck healing cap.



conjunction with the infusion of bone marrow

aspirate or by releasing recombinant human

GDF-5 (Jahng et al. 2004; Magit et al. 2006).

However, this study is the first to demonstrate

that Healos can successfully carry ng/rhBMP-2.

Although the sample size for the present

study was small and power was low, it was

possible to demonstrate a new layer of vertical

bone growth, with clear differences seen between

the test and the control groups. The findings

obtained by this descriptive study help to estab-

lish the foundation for future studies with a larger

sample size and fewer variables in the same

animal model.

Acknowledgments: This study was

supported by Institut Straumann AG, Basel,

Switzerland, which included the manufacture

of custom experimental titanium implants and

components. We appreciate the gift of ng/

rhBMP-2 provided by Dr Peter Hortschansky of

the Hans Knoll Institute, Jena, Germany, and

the excellent micro-CT imaging performed by

Dr Martin Stauber at b-cube AG, Zurich,

Switzerland. We also thank Dr Antoine Alves

(NAMSA) for histological and

histomorphometric analysis. Finally, we are

highly indebted to Mr Marcel Obrecht at

Institut Straumann for his careful organization

of all activities during surgical implantation

and important contributions to this

manuscript.

References

Cochran, D.L., Nummikoski, P.V., Higginbottom, F.L.,

Hermann, J.S., Makins, S.R. & Buser, D. (1996)

Evaluation of an endosseous titanium implant with

a sandblasted and acid-etched surface in the canine

mandible: radiographic results. Clinical Oral Im-

plants Research 7: 240–252.

Dard, M., Carlsson, U.B. & Obrecht, M. (2008) The

mepidor minipig model: a step forward in oral surgical

research. British Journal of Surgery 95: 10–11.

Donath, K. & Breuner, G. (1982) A method for the

study of undecalcified bones and teeth with attached

soft tissues. The sage-schliff (sawing and grinding)

technique. Journal of Oral Pathology 11: 318–326.

Feichtinger, M., Mossbock, R. & Karcher, H. (2007)

Assessment of bone resorption after secondary alveo-

lar bone grafting using three-dimensional computed

tomography: a three-year study. Cleft Palate Cranio-

facial Journal 44: 142–148.

Freilich, M., Patel, C.M., Wei, M., Shafer, D., Schleier,

P., Hortschansky, P., Kompali, R. & Kuhn, L. (2008)

Growth of new bone guided by implants in a murine

calvarial model. Bone 43: 781–788.

Freilich, M., Shafer, D., Wei, M., Adams, D. & Kuhn,

L. (2009) Implant system for guiding a new layer of

bone- computed microtomography and histomorpho-

metric analysis in the rabbit mandible. Clinical Oral

Implants Research 20: 201–207.

Hall, S.L., Lau, K.H., Chen, S.T., Wergedal, J.E., Sri-

vastava, A., Klamut, H., Sheng, M.H., Gridley, D.S.,

Mohan, S. & Baylink, D.J. (2007) Sca-1(þ ) hemato-

poietic cell-based gene therapy with a modified fgf-2

increased endosteal/trabecular bone formation in

mice. Molecular Therapy 15: 1881–1889.

Jahng, T.A., Fu, T.S., Cunningham, B.W., Dmitriev,

A.E. & Kim, D.H. (2004) Endoscopic instrumented

posterolateral lumbar fusion with healos and recom-

binant human growth/differentiation factor-5. Neuro-

surgery 54: 171–180.

Jovanovic, S.A., Schenk, R.K., Orsini, M. & Kenney, E.B.

(1995) Supracrestal bone formation around dental im-

plants: an experimental dog study. The International

Journal of Oral & Maxillofacial Implants 10: 23–31.

Jung, R.E., Thoma, D.S. & Hammerle, C.H. (2008)

Assessment of the potential of growth factors for loca-

lized alveolar ridge augmentation: A systematic review.

Journal of Clinical Periodontology 35: 255–281.

Kieswetter, K., Schwartz, Z., Hummert, T.W., Co-

chran, D.L., Simpson, J., Dean, D.D. & Boyan,

B.D. (1996) Surface roughness modulates the local

production of growth factors and cytokines by osteo-

blast-like mg-63 cells. Journal of Biomedical Materi-

als Research 32: 55–63.

Leknes, K.N., Yang, J., Qahash, M., Polimeni, G.,

Susin, C. & Wikesjo, U.M. (2008) Alveolar ridge

augmentation using implants coated with recombi-

nant human bone morphogenetic protein-2: radio-

graphic observations. Clinical Oral Implants

Research 19: 1027–1033.

Lindholm, T.C., Lindholm, T.S., Alitalo, I. & Urist,

M.R. (1988) Bovine bone morphogenetic protein

(bbmp) induced repair of skull trephine defects in

sheep. Clinical Orthopaedics and Related Research

227: 265–268.

Lutolf, M.P., Lauer-Fields, J.L., Schmoekel, H.G., Met-

ters, A.T., Weber, F.E., Fields, G.B. & Hubbell, J.A.

(2003) Synthetic matrix metalloproteinase-sensitive

hydrogels for the conduction of tissue regeneration:

engineering cell-invasion characteristics. Proceedings

of the National Academy of Sciences USA 100:

5413–5418.

Magit, D.P., Maak, T., Trioano, N., Raphael, B., Ha-

mouria, Q., Polzhofer, G., Drespe, I., Albert, T.J. &

Grauer, J.N. (2006) Healos/recombinant human

growth and differentiation factor-5 induces poster-

olateral lumbar fusion in a new zealand white rabbit

model. Spine 31: 2180–2188.

Miranda, D.A., Blumenthal, N.M., Sorensen, R.G.,

Wozney, J.M. & Wikesjo, U.M. (2005) Evaluation of

recombinant human bone morphogenetic protein-2

on the repair of alveolar ridge defects in baboons.

Journal of Periodontology 76: 210–220.

Neen, D., Noyes, D., Shaw, M., Gwilym, S., Fairlie, N.

& Birch, N. (2006) Healos and bone marrow aspirate

used for lumbar spine fusion: a case controlled study

comparing healos with autograft. Spine 31: E636–640.

Polimeni, G., Albandar, J.M. & Wikesjo, U.M. (2005)

Prognostic factors for alveolar regeneration: effect of

space provision. Journal of Clinical Periodontology

32: 951–954.

Poynton, A.R. & Lane, J.M. (2002) Safety profile for the

clinical use of bone morphogenetic proteins in the

spine. Spine 27: S40–48.

Renvert, S., Claffey, N., Orafi, H. & Albrektsson, T.

(1996) Supracrestal bone growth around partially in-

serted titanium implants in dogs. A pilot study.

Clinical Oral Implants Research 7: 360–365.

Roos-Jansaker, A.M., Franke-Stenport, V., Renvert, S.,

Albrektsson, T. & Claffey, N. (2002) Dog model for

study of supracrestal bone apposition around partially

inserted implants. Clinical Oral Implants Research

13: 455–459.

Sachse, A., Wagner, A., Keller, M., Wagner, O., Wetzel,

W.-D., Layher, F., Venbrocks, R.-A., Hortschansky,

P., Pietraszczyk, M., Wiederanders, B., Hempel, H.J.,

Bossert, J., Horn, J., Schmuck, K. & Mollenhauer, J.

(2005) Osteointegration of hydroxyapatite–titanium

implants coated with nonglycosylated recombinant

human bone morphogenetic protein-2 (bmp-2) in aged

sheep. Bone 37: 699–710.

Sampath, T.K., Coughlin, J.E., Whetstone, R.M., Banach,

D., Corbett, C., Ridge, R.J., Ozkaynak, E., Oppermann,

H. & Rueger, D.C. (1990) Bovine osteogenic protein is

composed of dimers of op-1 and bmp-2a, two members

of the transforming growth factor-beta superfamily.

Journal Biological Chemistry 265: 13198–13205.

Schmoekel, H.G., Schense, J.C., Weber, F.E., Graetz,

K.W., Gnaegi, D., Mueller, R. & Hubbell, J.A. (2004)

Healing of critical size defects in the rat and the dog

with nonglycosylated rhbmp-2 with enhanced reten-

tion in fibrin ingrowth matrices. Journal of Orthope-

dic Research 22: 376–381.

Schmoekel, H.G., Weber, F.E., Hurter, K., Schense, J.C.,

Seiler, G., Spreng, D., Schawalder, P. & Hubbell, J.

(2005) Enhancement of bone healing using non-glyco-

sylated rhbmp-2 released from a fibrin matrix in dogs

and cats. Journal of Small Animal Practice 46: 17–21.

Sigurdsson, T.J., Fu, E., Tatakis, D.N., Rohrer, M.D. &

Wikesjo, U.M. (1997) Bone morphogenetic protein-2

for peri-implant bone regeneration and osseointegra-

tion. Clinical Oral Implants Research 8: 367–374.

Simion, M., Rocchietta, I. & Dellavia, C. (2007) Three-

dimensional ridge augmentation with xenograft and

recombinant human platelet-derived growth factor-bb

in humans: report of two cases. International Journal

of Periodontics Restorative Dentistry 27: 109–115.

Tinti, C., Parma-Benfenati, S. & Manfrini, F. (1997)

Spacemaking metal structures for nonresorbable

membranes in guided bone regeneration around

implants. Two case reports. International Journal of

Periodontics Restorative Dentistry 17: 53–61.

Urist, M.R. (1965) Bone formation by autoinduction.

Science 150: 893–899.

Urist, M.R. & Strates, B.S. (1971) Bone morphogenetic

protein. Journal of Dental Research 50: 1392–1406.

Wikesjo, U.M., Qahash, M., Polimeni, G., Susin, C.,

Shanaman, R.H., Rohrer, M.D., Wozney, J.M. &

Hall, J. (2008) Alveolar ridge augmentation using

implants coated with recombinant human bone mor-

phogenetic protein-2: histologic observations. Journal

of Clinical Periodontology 35: 1001–1010.

Woo, B.H., Fink, B.F., Page, R., Schrier, J.A., Jo, Y.W.,

Jiang, G., DeLuca, M., Vasconez, H.C. & DeLuca,

P.P. (2001) Enhancement of bone growth by sustained

delivery of recombinant human bone morphogenetic

protein-2 in a polymeric matrix. Pharmaceutical

Research 18: 1747–1753.



Documents

Implant-guided vertical bone growth in