Copyrig

ht

by

N

otfor

Qu

in

tessence

Not

forPublication

� 209

Eur J Oral Implantol 2010;3(3):209–219

RANDOMISED CONTROLLED CLINICAL TRIAL

Matteo Capelli, DDSAssistant, Section of Implan-tology and Oral Rehabilita-tion, Dental Clinic, IRCCS Galeazzi Orthopaedic Insti-tute, Department of Health Technologies, Università degli Studi di Milano, and private practice, Milan, Italy

Marco Esposito, DDS, PhDAssociated Professor, Department of Biomaterials, The Sahlgrenska Academy at Göteborg University, Göteborg, Sweden

Francesco Zuffetti, MD, DDSAssistant, Section of Implan-tology and Oral Rehabilita-tion, Dental Clinic, IRCCS Galeazzi Orthopaedic Insti-tute, Department of Health Technologies, Università degli Studi di Milano, and private practice, Milan, Italy

Fabio Galli, MD, DDSAssistant, Section of Implan-tology and Oral Rehabilita-tion, Dental Clinic, IRCCS Galeazzi Orthopaedic Insti-tute, Department of Health Technologies, Università degli Studi di Milano, and private practice, Monza, Italy

Massimo Del Fabbro, Bsc, PhDResearcher, Head of Section of Oral Physiology, Dental Clinic, IRCCS Galeazzi Ortho-paedic Institute, Department of Health Technologies, Università degli Studi di Milano, Milan, Italy

Tiziano Testori, MD, DDSHead of the Section of Implan tology and Oral Rehabilitation, IRCCS Galeazzi Orthopaedic Institute, Depart-ment of Health Technologies, Università degli Studi di Milano, and private practice, Como, Italy

Matteo Capelli, Marco Esposito, Francesco Zuffetti, Fabio Galli, Massimo Del Fabbro, Tiziano Testori

A 5-year report from a multicentre randomised clinical trial: immediate non-occlusal versus early loading of dental implants in partially edentulous patients

Key words dental implants, early loading, follow-up, immediate loading, partial edentulism

Conflict-of-interest statement: This trial was independently designed and initiated by the investiga-tors, however BIOMET 3i provided partial economic support at a later stage.

Purpose: To compare peri-implant bone and soft-tissue levels of immediately non-occlusally loaded versus non-submerged early loaded implants in partially edentulous patients 5 years after implant placement.Materials and methods: Fifty-two patients were randomised in five Italian private practices: 25 in the immediately loaded group and 27 in the early loaded group. To be immediately loaded, single implants had to be inserted with a torque of > 30 Ncm, and splinted implants with a torque of > 20 Ncm. Immediately loaded implants were provided with non-occluding temporary restorations within 48 hours. After 2 months, the provisional restorations were put in full occlusion. Implants were early loaded after 2 months. Final restorations were provided 8 months after implant place-ment. Outcome measures were prosthesis and implant failures as well as biological and prosthetic complications recorded by non-blinded assessors. Blinded assessors evaluated peri-implant bone and soft-tissue levels. Results: Fifty-two implants were immediately loaded and 52 early loaded. One patient of the early loaded group dropped out after the 1-year recall. One single immediately loaded implant failed 2 months after placement. Only one complication (iatrogenic peri-implantitis) occurred in one patient of the early loading group. Both groups gradually lost peri-implant bone in a highly statistically sig-nificant way at 2, 8 and 14 months and at 4 and 5 years. After 5 years, patients of both groups had lost an average of 1.2 mm of peri-implant marginal bone. There were no statistically significant dif-ferences in peri-implant bone and soft-tissue level changes between the 2 groups. At 5 years, there was a statistically significant recession (0.2 mm) of the vestibular soft tissues from baseline (delivery of the final restorations 8 months after implant placement) only for immediately loaded implants.Conclusions: In well maintained patients, complications are uncommon and healthy and stable peri-implant tissues can be maintained for 5 years around immediately and early loaded implants.

Copyrig

ht

by

N

otfor

Qu

in

tessence

Not

forPublication

Capelli et al Immediate versus early loading: bone and soft-tissue levels210 �

Eur J Oral Implantol 2010;3(3):209–219

� Introduction

Immediate loading of implant-supported prostheses is highly appreciated by patients. This approach can reduce their treatment periods drastically while they live a normal life with minimal discomfort due to edentulism1,2. In carefully selected cases, immedi-ate and early loading procedures can be success-ful3, though not all authors have shown predict-ably high success rates4-8. To date no published trial has reported on the stability of peri-implant soft tissues when comparing immediately versus early loaded implants3, though a few trials have provided data on marginal bone-level changes when com-paring immediately versus conventionally loaded implants9-14 and early versus conventionally loaded implants4,15. Meta-analyses of these RCTs3 were not able to demonstrate any statistically significant dif-ferences or even trends between the compared pro-cedures for marginal bone-level changes.

The aim of the present randomised clinical trial was to compare the efficacy of immediate non-occlusal loading versus early loading in partially edentulous patients. The null hypothesis that there would be no difference in clinical outcomes between the two procedures was tested against the alterna-tive hypothesis of a difference.

Immediate non-occlusal loading was defined as the seating within 48 hours after implant placement of a provisional restoration that would not be in occlusal contact for approximately 2 months. Early loading was defined as placing the provisional resto-rations after an initial 2-month healing period.

The present report presents clinical data at 5 years after implant loading. Previous publications reported on 14-month data16,17. The present article is reported according to the CONSORT statement for improving the quality of reports of parallel-group randomised trials (http://www.consort-statement.org/).

� Materials and methods

Any partially dentate patient requiring dental implants who was 18 years or older and able to sign an informed consent form was eligible for inclusion in the present trial. All participants were informed of the nature of the study. For patients with mul-

tiple edentulous areas to be restored, the operator was free at the screening visit to select one area to be included in the trial. Smokers were included and patients were grouped according to what they declared: 1) non smokers; 2) light smokers (if smok-ing up to 10 cigarettes per day); 3) heavy smokers (if smoking more than 10 cigarettes per day). Patients were not included in the study if any of the following exclusion criteria were present: 1) general contraindi-cations to implant surgery, 2) irradiation in the head and neck area, 3) poor oral hygiene and motivation, 4) periodontal disease; 5) uncontrolled diabetes, 6) pregnancy or lactation, 7) substance abuse, 8) psy-chiatric problems, 9) lack of opposing occluding den-tition in the area intended for implant placement, 10) severe bruxism or clenching, 11) infection or severe inflammation in the area intended for implant place-ment, 12) a need for bone-augmentation procedures including sinus lifting or 13) a gap between one of the bone walls and the surface of a post-extraction implant of more than 1.5 mm.

Patients were recruited and treated in five pri-vate dental clinics located in northern Italy: Como (two centres), Milan (two centres) and Monza (one centre), all having extensive experience in the treatment of patients with immediate loading pro-cedures. The investigators have worked, and are still working together, in the same team for at least 5 years prior to the initiation of the present study at the IRCCS Galeazzi Orthopaedic Institute so they developed simi-lar clinical operative procedures. All patients signed a written informed consent form to participate in the trial. One experienced surgeon at each centre per-formed all of the operations. Patients were randomised to have implants non-occlusally loaded immediately (test group) or early (control group) in a parallel group study design. Only one prosthesis in the area selected at screening was considered for each patient.

Patients were instructed to use chlorhexidine mouthwash 0.2% for 1 minute, twice a day, starting 3 days prior to the intervention and thereafter for 2 weeks. All patients received prophylactic antibiotic therapy: amoxicillin 2 g 1 hour prior to the inter-vention and 2 g 6 hours postoperatively. Patients allergic to penicillin were given clarithromycin 500 mg 1 hour prior to the intervention. Ibuprofen 600 mg was given 1 hour prior to intervention and then twice a day for 3 days.

Correspondence to: Dr Marco Esposito, PO Box 34, 20043, Arcore (MB), Italy E-mail: espositomarco@hotmail.com

Copyrig

ht

by

N

otfor

Qu

in

tessence

Not

forPublication

Capelli et al Immediate versus early loading: bone and soft-tissue levels � 211

Eur J Oral Implantol 2010;3(3):209–219

Full-thickness crestal flaps were elevated after crestal incision with a minimal extension to minimise patient discomfort. Teeth extractions were performed as atraumatically as possible to preserve the buccal alveolar bone. The choice of the implant diameter and length was left up to the surgeon. At four centres, Full OSSEOTITE® Tapered Implants (FOSS) (BIOMET 3i, Palm Beach, FL, USA) were used exclusively and inserted according to the manufacturer’s instructions. At the Como 1 centre, 16 of the 35 implants placed were the FOSS configuration, while 19 out of the 35 were prototypes of tapered implants from the same manufacturer with identical surface characteristics. Implant diameters used were 4, 5 and 6 mm; lengths used were 8.5, 10, 11.5, 13 and 15 mm.

Bone density at drilling was subjectively evalu-ated and the bone at the implant site was classi-fied as either ‘hard’, ‘medium’ or ‘soft’18. Resistance to implant insertion was objectively recorded with Osseocare equipment (Nobel Biocare, Göteborg, Sweden). In the protocol-formulation phase, it was decided that single implants with a torque resist-ance of ≤ 30 Ncm or splinted implants with a torque resistance of ≤ 20 Ncm that were randomised to the immediately loaded group should instead be treated as belonging to the early loaded group. In the event that the planned insertion torque could not be achieved an intention-to-treat analysis was to be applied, however all implants achieved the minimal insertion torque required.

In soft bone, under-preparation was performed using a shaping drill one size smaller than the final implant diameter. In general, implants were placed at crestal level in healed edentulous ridges and slightly subcrestally in immediate post-extraction sockets. In cases where a residual gap of ≤ 1.5 mm was present between the implant surface and the bone wall, the gap was filled with autogenous bone chips. A non-submerged technique was employed. Before placing the abutments, the envelope containing the randomisation code was opened, allowing the sur-geon to know whether the implant was to be loaded immediately or early. Impression copings or heal-ing screws were placed accordingly, and interrupted sutures were placed using a monofilament thread. An impression with the pick-up impression copings was made for the implants to be immediately loaded, and healing abutments were placed. Ice packs were

provided, and a soft diet was recommended. Smok-ers were told to avoid smoking for 48 hours post-operatively.

Acrylic resin provisional restorations were pro-vided to patients of the immediately loaded group the following day. The occlusal surface of the provi-sional restoration was ground to avoid any occlusal contact with the opposite dentition. All provisional restorations of the immediately loaded group were placed within 48 hours. Sutures were removed 2 weeks after implant placement.

Two months after implant placement, resin was added to the immediate non-occlusally loaded res-torations to put these in full occlusion. Patients of the early loaded group received provisional restora-tions identical to those of the immediately loaded groups with full occlusal contacts 2 months after implant placement. Final metal-ceramic restora-tions were cemented 8 months after implant place-ment.

Patients were recalled for oral hygiene mainte-nance and prosthetic controls every 3 months up to the first year and thereafter every 6 months. Out-come measures were:• Prosthesis failure: the planned prosthesis could

not be placed or was lost because of implant failure.

• Implant failure: the presence of any mobility of the individual implant (assessed manually by rotating the implant) at insertion of the provi-sional or definitive prostheses, and/or any infec-tion dictating implant removal. After insertion of the definitive restorations, prostheses were not removed to assess clinical mobility of individual implants.

• Any biological or prosthetic complications. Exam-ples of possible biological complications were numbness of the lower lip and chin, peri-implant mucositis (heavily inflamed soft tissue without bone loss), peri-implantitis (bone loss with sup-puration or heavily inflamed tissues, fistulas), etc. Examples of possible prosthetic complications were fracture of the implant, abutment screw, framework, occlusal material, etc.

The following outcome measures were assessed by the treating dentists who were therefore not blinded.

Copyrig

ht

by

N

otfor

Qu

in

tessence

Not

forPublication

Capelli et al Immediate versus early loading: bone and soft-tissue levels212 �

Eur J Oral Implantol 2010;3(3):209–219

• Peri-implant marginal bone-level changes on intraoral radiographs made with the paralleling technique. Periapical radiographs were taken at implant placement, 2, 8, 14, 48 and 60 months after implant placement. Radiographs were scanned (HP Scanjet 3c/t, Hewlett Packard, Cernusco sul Naviglio, Milan, Italy), digitised in JPG format, converted to TIFF format at 600 dpi resolution, and stored in a personal computer. Peri-implant marginal bone levels were measured using Scion Image software (Scion Corporation, Frederick, MD, USA). The software was cali-brated for every single image using the known distance of 2 consecutive treads (0.9 mm for FOSS implants, and 0.6 mm for the prototype implants). Measurements of the mesial and distal bone-crest levels adjacent to each implant were made to the nearest 0.01 mm. Reference points for the linear measurements were the coronal margin of the implant collar and the most coronal point of bone-to-implant contact.

• Soft-tissue stability (height of the clinical crown) measured with a digital calliper (Calliper IP54, Sham, Guillin, China) on hard plaster models

(Fig 1). Impressions were taken at 8 months (after delivery and adjustment of the final restorations), and 14 and 60 months using a high dimensional stability alginate (Zhermack, Hidrogum 5, Badia Polesine (RO), Italy). Plaster models were imme-diately prepared using extra-hard plaster (type 4). For incisors the reference point was the mid-dle of the incisal margin, for canines and bicus-pids it was the tip of the cuspid, and for molars the deepest occlusal vestibular margin between the two cusps. Measurements were done ves-tibularly from the occlusal reference point per-pendicular to the marginal gingiva.

Up to 1 year, all bone and soft tissue measurements were taken twice, after at least a 15-day interval, and the two values were averaged. After the first year, since the difference between the two assess-ments was negligible, only one assessment was made. These outcome measures were assessed by two independent, blinded, calibrated outcome asses-sors. One measured all radiographs and the other the clinical crown heights on plaster models up to 1 year after loading. After the 1-year assessment, the asses-sors were switched, i.e. the one who performed the bone level assessment up to 1 year performed the soft tissue assessment and vice-versa. All radiographs were coded so that the outcome assessor was blind to which group the implants belonged, nor was the assessor informed of the aims of the study. Since the assessor did not know which crown(s) to measure on the plaster models, the implant-supported crown(s) were marked with a black spot. The outcome asses-sors did not know the dates on which the radio-graphs or the plaster models were created.

The sample size was chosen based on calculations of the number of patients likely to have at least one restoration failure (primary outcome measure). In a recent study5 of partially edentulous patients, the pro-portion of failures in the immediately loaded group was 0.39 compared to 0.04 in the conventionally loaded group. A two-group continuity-corrected chi-square test with a 0.050 two-sided significance level will have 80% power to detect the difference between a proportion of 0.39 and a proportion of 0.04 (odds ratio of 0.065) when the sample size in each group is 26 patients. It was planned to include 30 patients in each group to compensate for possible dropouts.

Fig 1 Plaster model with its code number unknown to the asses-sor. The black spot indicates the implant-supported crown included in the trial. The clinical crown height was measured with a digital calliper.

Copyrig

ht

by

N

otfor

Qu

in

tessence

Not

forPublication

Capelli et al Immediate versus early loading: bone and soft-tissue levels � 213

Eur J Oral Implantol 2010;3(3):209–219

A manually generated restricted randomisation list was used to create two groups with equal num-bers of patients. Only one of the investigators, not involved in the selection and treatment of the patients, was aware of the randomisation sequence and had access to the randomisation list stored in a password-protected portable computer. The randomised codes were enclosed in sequentially numbered, identical, opaque, sealed envelopes. Envelopes were opened sequentially only after the implants to be included in the trial were inserted, thereby concealing treatment allocation from the investigators in charge of enroll-ing and treating the patients.

All data analysis was carried out according to a pre-established analysis plan. The patient was the statistical unit of the analyses. Two biostatisticians with expertise in dentistry analysed the data, without knowing the group allocation. Differences in the pro-portion of failures and other complications between the groups were compared using Fisher’s exact prob-ability test. The average radiographic values for the mesial and distal surfaces were calculated and aver-aged for each patient. Average crown-length values were also calculated for each patient. Comparisons between each time point and the baseline meas-urement were made by paired tests, to detect any changes in bone or soft-tissue levels. An analysis of covariance was used to compare the mean radio-graphic and soft-tissue values at 5 years (outcome variable), with the baseline value as a covariate. A general linear model using the treating centre, the loading time and the covariate (baseline values) as explanatory variables, was employed. Initially, the possible interaction between the centres and loading time was considered, however the interaction was eliminated from the model when it was found not to be significant. Possible differences between centres were investigated using the Tukey-Kramer Honestly Significant Difference Test. All statistical comparisons were conducted at the 0.05 level of significance.

� Results

All patients eligible for the present trial agreed to participate. Fifty-two patients were consecutively enrolled in the trial and randomised: 25 to the imme-diately loaded group and 27 to the early loaded

group. The planned number of 30 patients per group was not achieved since the centres decided to stop patient recruitment at the end of May 2005. All patients were treated according to the allocated interventions. All implants achieved the required minimal insertion torque. One patient from the early loaded group dropped out after the 14-month evaluation and moved to another town. The data of all patients were evaluated in the statistical analyses.

Patients were recruited and treated from Octo-ber 2004 to May 2005 and they were followed for 5 years after implant placement. Initially, six centres agreed to participate by treating 10 patients each, however two centres withdrew before start-ing the trial and a new centre was added as a partial replacement. The remaining envelopes containing the randomisation codes were redistributed among the centres with higher recruitment capacities. The patient distribution across the various centres is shown in Table 1. The main baseline patient char-acteristics are presented in Table 2. Patients were generally healthy. Three patients suffered from hypertension and one from hepatitis C. By chance, these four patients were all randomised to the early loaded group.

Fifty-two implants were placed in the imme-diately loaded group and 52 in the early loaded group. The lengths and diameters of the inserted implants are presented in Table 3, whereas the bone density, subjectively evaluated, and the maximum insertion torque (primary implant stability) appear in Table 4. There were no apparent significant base-line imbalances between the two groups.

Deviations from the operative protocol were the following: while four centres used exclusively

Centre Immediate (n = 25) Early (n = 27)

Comi 1 (Dr Testori) 9 6*

Como 2 (Dr Ritzmann) 2 5

Milano 1 (Dr Zuffetti) 6 6

Milano 2 (Dr Capelli) 4** 4

Monza (Dr Galli) 4 6***

*1 patient dropped out (moved away) after 14 months**1 patient experienced an early implant failure***1 patient experienced a complication (iatrogenic peri-implantitis)

Table 1 Patient distribution across centres. Originally, six centres were supposed to treat 10 patients each.

Copyrig

ht

by

N

otfor

Qu

in

tessence

Not

forPublication

Capelli et al Immediate versus early loading: bone and soft-tissue levels214 �

Eur J Oral Implantol 2010;3(3):209–219

Immediate (n = 25) Early (n = 27)

Females 12 17

Mean age at implant insertion (range) 51.6 (27-74) 51.3 (34-73)

Smokers 9 light 3 light and 1 heavy

Total number of inserted implants 52 52

Total number of inserted prototype implants (only by 1 centre) 11 8

Implants inserted in mandibles 38 21

Implants inserted in anterior areas (canine to canine) 3 3

Implants inserted in fresh extraction sockets 6 9 (1 grafted)

Number of patients receiving single implants 7 10

Number of patients receiving two implants 10 9

Number of patients receiving three implants 7 8

Number of patients receiving four implants 1 0

Table 2 Patient and intervention characteristics.

Implant length Immediate (n = 52) Early (n = 52)

8.5 mm 8 2

10 mm 18 16

11.5 mm 13 15

13 mm 11 17

15 mm 2 2

Implant diameter Immediate (n = 52) Early (n = 52)

4 mm 38 33

5 mm 12 18

6 mm 2 1

Tactile bone density Immediate (n = 52) Early (n = 52)

Hard 1 2

Medium 43 40

Soft 8 10

Insertion torque Immediate (n = 52) Early (n = 52)

20 Ncm 3 4

30 Ncm 6 14

40 Ncm 18 20

50 Ncm 22 14

60 Ncm 3 0

Table 3 Length and diameters of the implants.

Table 4 Bone density evaluated clinically and with the Osseocare device (primary implant stability).

FOSS NT implants, the Como 1 centre also used some prototypes of tapered implants with identical surface characteristics from the same manufacturer (19 out of 35 inserted implants). One patient from the immediately loaded group was provided with the final restoration after 2 months instead of the planned 8 months. Because of a cardiac disease, one patient from the early loaded group received the final prosthesis 14 months after implant placement, however that patient’s implants were assessed for stability at the planned 8-month interval.

A single implant and its provisional prosthesis, inserted with a primary stability of 50 Ncm in medium bone density, failed 2 months after placement in the immediately loaded group. The failed implant was suc-cessfully replaced with another implant after 6 months of healing. There were no statistically significant differences in prosthesis and implant failures among the two interventions (Fisher’s exact test P = 1.0).

Only one complication occurred in one patient: peri-implantitis around two early loaded implants detected 33 months after implantation. The patient did not show up for regular recalls between months 14 and 33. On the control periapical radiograph, bone loss up to the fourth thread could be observed (Fig 2a), however clinically, the peri-implant tissues appeared healthy with no evident sign of pathology (Fig 2b). A flap was immediately elevated (Fig 2c) which revealed the presence of radio-transparent

Copyrig

ht

by

N

otfor

Qu

in

tessence

Not

forPublication

Capelli et al Immediate versus early loading: bone and soft-tissue levels � 215

Eur J Oral Implantol 2010;3(3):209–219

Fig 2 Sequence of treatment of the only complication, which occurred in one patient of the early loaded group and was detected 33 months after implant placement: a) on the periapical radiograph bone loss is visible up to the fourth thread; b) peri-implant tissues appeared healthy even after prosthesis removal; c) a flap was elevated; d) the excessive cement was removed and the implant surface was carefully cleaned; e) the defects were filled with granular anorganic bovine bone and covered with a resorbable bar-rier; f) clinical view after uneventful healing and suture removal; g) clinical situation and h) periapi-cal radiograph at the 5-year follow-up, show-ing a good outcome of the complication (iatro-genic peri-implantitis).

a b

c d

e f

g h

Copyrig

ht

by

N

otfor

Qu

in

tessence

Not

forPublication

Capelli et al Immediate versus early loading: bone and soft-tissue levels216 �

Eur J Oral Implantol 2010;3(3):209–219

cement around the exposed margins of the prosthe-sis. The excessive cement was removed (Fig 2d), the defect was carefully debrided and the implant sur-face thoroughly cleaned with a jet of bicarbonates. The defect was filled with granular anorganic bovine bone (Bio-Oss®, Geistlich Pharma, Wolhusen, Swit-zerland) and covered with a resorbable collagen membrane (Bio-Gide®, Geistlich Pharma) (Fig 2e). The post-operative healing was uneventful (Fig 2f) and health was maintained over time (Fig 2g and h). There were no statistically significant differences in complications between the two interventions (Fisher’s exact test P = 1.0).

Both groups gradually lost marginal peri-implant bone in a highly statistically significant way (P < 0.001) at 2, 8, 14, 48 and 60 months (Table 5). After 2 months, patients in the immediately loaded group lost an average of 0.5 mm of peri-implant bone versus 0.6 mm for the patients in the early

loaded group (Table 6). After 8 months, patients in the immediately loaded group lost an average of 0.9 mm of peri-implant bone versus 1.0 mm for the patients in the early loaded group (Table 6). After 14, 48 and 60 months, patients of both groups lost an average of 1.1, 1.1 and 1.2 mm of peri-implant bone, respectively (Table 6). There was no statistically significant difference between the two loading strat-egies for peri-implant bone-level changes, when an analysis of covariance was applied (P = 0.4130). The difference between the two interventions adjusted for the centres and the covariate was 0.12 mm (CI 95% from -0.17 to 0.41) in favour of immedi-ate loading. There was a significant centre effect (P = 0.0384). In particular, centre Milan 1 had a statistically significantly lower mean bone level dif-ference (0.63 mm) compared to centre Como 1 (CI 95% from 0.07 to 1.19). There were no other statistically significant differences between centres.

Baseline 2 months* 8 months* 14 months* 48 months* 60 months*

N; Mean (SD) N; Mean (SD) N; Mean (SD) N; Mean (SD) N; Mean (SD) N; Mean (SD)

Immediate loading 25; 0.03 (0.09) 25; 0.55 (0.50) 24; 0.95 (0.56) 24; 1.13 (0.56) 24; 1.15 (0.55) 24; 1.18 (0.56)

Early loading 27; 0.07 (0.16) 27; 0.67 (0.46) 27; 1.04 (0.52) 27; 1.18 (0.51) 26; 1.25 (0.50) 26; 1.28 (0.50)

*All changes from baseline statistically significantly different (P < 0.001)

Table 5 Mean radiographic peri-implant marginal bone levels between groups and time periods.

Baseline - 2 months Baseline - 8 months Baseline - 14 months Baseline - 48 months Baseline - 60 months

N; Mean (SD) N; Mean (SD) N; Mean (SD) N; Mean (SD) N; Mean (SD)

Immediate loading 25; 0.52 (0.48) 24; 0.92 (0.55) 24; 1.10 (0.58) 24; 1.12 (0.54) 24; 1.15 (0.56)

Early loading 27; 0.60 (0.46) 27; 0.97 (0.50) 27; 1.11 (0.54) 26; 1.18 (0.49) 26; 1.21 (0.49)

P value 0.41*

*Analysis of covariance for 60 months with baseline as a covariate

Table 6 Comparison of mean changes in peri-implant marginal bone levels at different time periods between groups.

8 - 14 months 8 - 60 months P value comparing 8 with 60 months

P value comparing immediate vs early loading

N; Mean Difference (SD) N; Mean Difference (SD)

Immediate loading 24; 0.13 (0.40) 24; 0.20 (0.40) 0.01*

0.31Early loading 27; 0.01 (0.37) 26; 0.05 (0.37) 0.64

*Change statistically significantly different from baseline (8 months after implant placement)

Table 7 Comparison of soft-tissue recessions between groups and time periods.

Copyrig

ht

by

N

otfor

Qu

in

tessence

Not

forPublication

Capelli et al Immediate versus early loading: bone and soft-tissue levels � 217

Eur J Oral Implantol 2010;3(3):209–219

After 60 months, there was a statistically signifi-cant recession (0.2 mm; P = 0.01) of the vestibular soft tissues from baseline (delivery of the final res-torations 8 months after implant placement; Table 7) only for immediately loaded implants. There were no statistically significant differences between the two loading strategies for soft tissue changes (P = 0. 3156; Table 7). The difference between the two interventions adjusted for the centres and the covariate was 0.14 mm (CI 95% from -0.14 to 0.41) in favour of early loaded implants. No statisti-cally significant differences between centres were observed (P = 0.3043).

� Discussion

The primary aim of the present trial was to evalu-ate the effectiveness of immediate and early loading procedures in partially edentulous patients, how-ever the long-term tissue stability of peri-implant tissues was also evaluated. The data presented cover a follow-up period of 5 years after implant placement. Both loading procedures achieved good results which were maintained over a 5-year period. The mean peri-implant marginal bone loss was in the range of 1.2 mm whereas buccal soft tissues did not recede in a visible way. These results are strengthened by the fact that the methodology of this report was robust: patients were randomised and group allocation was concealed to clinicians. All treated patients were accounted for with no exclu-sions. Only one dropout occurred when the patient moved to another town. Assessment of the peri-implant tissue stability was performed by calibrated, independent, blinded outcome assessors, using objective, validated, and reproducible outcome measures. Only one complication occurred and it was classified as ‘iatrogenic peri-implantitis’. This was classified as iatrogenic because during cemen-tation of the definitive bridge, excess cement was not thoroughly removed and over time induced a peri-implantitis-like bone destruction around two implants. It is interesting to observe that this com-plication developed between months 14 and 33, the period in which the patient stopped coming in for regular maintenance visits. It was only detected after probing and was finally confirmed by a periapical

radiograph. Surprisingly, the peri-implant soft tissue looked firm, pink and healthy. The lesion was suc-cessfully treated with a regenerative procedure and peri-implant health was maintained up to the 5-year recall interval. The authors therefore suggest the use of a radiopaque cement when cementing prostheses so that any excess cement will be visible on periapical radiographs and can be promptly removed.

In the present trial, the pattern of peri-implant bone loss was virtually identical for both groups. Approximately 0.5 mm of bone was lost during the first 2 months. Eight months after baseline, the average bone loss was approximately 1 mm. An additional 0.1 mm of bone was lost over the next 6 months and after 5 years an average of 1.2 mm of bone was lost by both groups (Table 6). The authors were unable to compare these good long-term results with other investigations using implants with similar characteristics because these could not be identified. While the original plan was to follow these patients for 5 years, it would be inter-esting to prolong the follow-up to 10 years or more.

When answering the question of whether there are any differences in marginal bone-level changes between immediately and early loaded implants, the present findings suggest that no difference should be expected. This is in agreement with a meta-analysis3 that included two other RCTs2,19, showing no statis-tically significant differences between the two load-ing strategies (mean difference -0.06 mm, 95% CI from -0.16 to 0.03).

When interpreting the findings of a statistically significant recession of vestibular soft-tissue levels after delivery of the definitive restorations in patients of the immediate loading group, it should be con-sidered that 0.2 mm is virtually not detectable clini-cally and that there was no difference in soft-tissue changes between the two groups. It is possible though that some soft-tissue recessions may have occurred during the first 8 months when the provi-sional restorations were in use.

The main limitations of the present trial are the small sample size and the use by one of the centres of some implant prototypes instead of the implants decided upon at the protocol stage. The prototypes were similar to the commercially available implants used in the trial and were randomly placed without the clinician knowing at placement into which group

Copyrig

ht

by

N

otfor

Qu

in

tessence

Not

forPublication

Capelli et al Immediate versus early loading: bone and soft-tissue levels218 �

Eur J Oral Implantol 2010;3(3):209–219

they were going to be included since group alloca-tion was concealed. In fact, they were randomly dis-tributed between the two groups in approximately equal numbers (Table 2), therefore, it is unlikely that this protocol violation significantly impacted the out-come of the present study.

The decision not to put the provisional resto-rations of immediately loaded implants in direct occlusion with the opposite dentition for 2 months, though the restorations were actually loaded during chewing, was dictated by the desire to minimise the risk of early implant failures. This decision was purely based on clinical reasoning. Surprisingly, the only two RCTs that have investigated this hypothesis7,8 did not find any statistically significant difference or even a clinical trend when comparing immediate occlusal versus non-occlusal loading. Also, 1 year after loading, marginal bone loss in the two groups was virtually identical. From a clinical point of view, it is still tempting to recommend that provisional restorations not be placed in direct contact with the opposite dentition, particularly in those situations where the clinician fears that problems might arise (i.e. posterior regions), but the limited evidence available does not support this clinical practice.

With respect to the generalisability (external valid-ity) of these findings, on one hand patient inclusion criteria were broad, and both techniques were tested in real clinical conditions, therefore the results can be easily generalised to a wider population. On the other hand, the dentists who delivered the interven-tions were highly trained, therefore caution should be exercised by less experienced dentists when applying immediate or early loading procedures.

� Conclusions

The timing of loading (immediate or early) does not appear to have any significant clinical impact on marginal peri-implant bone or soft-tissue lev-els. While an average of 1.2 mm of peri-implant bone was lost over 5 years, virtually no buccal soft-tissue recession occurred from the placement of the definitive restorations up to 5 years after implant placement.

� Acknowledgements

The authors wish to thank Dr Michele Nieri for the additional help with the statistical analysis, Dr Joerg Ritzmann for treating seven patients included in this trial, Ms Ilaria Spreafico for collecting data from the various centres, and Dr Marco Fossati and Dr Mat-teo De Florian for the blind evaluation of periapical radiographs and clinical crown heights.

� References

1. Cannizzaro G, Leone M, Esposito M. Immediate versus early loading of two implants placed with a flapless tech-nique supporting mandibular bar-retained overdentures: a single-blinded, randomised controlled clinical trial. Eur J Oral Implantol 2008;1:33-43.

2. Cannizzaro G, Torchio C, Leone M, Esposito M. Immediate versus early loading of flapless-placed implants supporting maxillary full-arch prostheses: a randomised controlled clini-cal trial. Eur J Oral Implantol 2008;1:127-139.

3. Esposito M, Grusovin MG, Achille H, Coulthard P, Worthing-ton HV. Interventions for replacing missing teeth: different times for loading dental implants. Cochrane Database Syst Rev 2008:CD003878.

4. Tawse-Smith A, Payne AGT, Kumara R, Thomson WM. Early loading of unsplinted implants supporting mandibular overdentures using a one-stage operative procedure with two different implant systems: A 2-year report. Clin Implant Dent Relat Res 2002;4:33-42.

5. Ottoni JM, Oliveira ZF, Mansini R, Cabral AM. Correlation between placement torque and survival of single-tooth implants. Int J Oral Maxillofac Implants 2005;20:769-776.

6. Oh TJ, Shotwell JL, Billy EJ, Wang HL. Effect of flapless implant surgery on soft tissue profile: a randomized control-led clinical trial. J Periodontol 2006;77:874-882.

7. Lindeboom JA, Frenken JW, Dubois L, Frank M, Abbink I, Kroon FH. Immediate loading versus immediate provisionali-zation of maxillary single-tooth replacements: a prospective randomized study with BioComp implants. J Oral Maxillofac Surg 2006;64:936-942.

8. Cannizzaro G, Torchio C, Felice P, Leone M, Esposito M. Immediate occlusal versus non-occlusal loading of single zir-conia implants. A multicenter pragmatic randomised clinical trial. Eur J Oral Implantol 2010;3:111-120.

9. Cannizzaro G, Leone M. Restoration of partially edentulous patients using dental implants with a microtextured surface: a prospective comparison of delayed and immediate full occlu-sal loading. Int J Oral Maxillofac Implants 2003;18:512-522.

10. Hall JA, Payne AG, Purton DG, Torr B. A randomized con-trolled clinical trial of conventional and immediately loaded tapered implants with screw-retained crowns. Int J Prostho-dont 2006;19:17-19.

11. Crespi R, Cappare P, Gherlone E, Romanos GE. Immediate ver-sus delayed loading of dental implants placed in fresh extrac-tion sockets in the maxillary esthetic zone: a clinical compara-tive study. Int J Oral Maxillofac Implants 2008;23:753-758.

12. Schincaglia GP, Marzola R, Giovanni GF, Chiara CS, Scotti R. Replacement of mandibular molars with single-unit restora-tions supported by wide-body implants: immediate versus delayed loading. A randomized controlled study. Int J Oral Maxillofac Implants 2008;23:474-480.

Copyrig

ht

by

N

otfor

Qu

in

tessence

Not

forPublication

Eur J Oral Implantol 2010;3(3):209–220

Capelli et al

13. Turkyilmaz I, Tumer C. Early versus late loading of unsplinted TiUnite surface implants supporting mandibular overden-tures: a 2-year report from a prospective study. J Oral Reha-bil 2007;34:773-780.

14. Güncü MB, Aslan Y, Tümer C, Güncü GN, Uysal S. In-patient comparison of immediate and conventional loaded implants in mandibular molar sites within 12 months. Clin Oral Implants Res 2008;19:335-341.

15. Payne AG, Tawse-Smith A, Duncan WD, Kumara R. Con-ventional and early loading of unsplinted ITI implants supporting mandibular overdentures: Two-year results of a prospective randomized clinical trial. Clin Oral Implants Res 2002;13:603-609.

16. Testori T, Galli F, Capelli M, Zuffetti F, Esposito M. Immediate non-occlusal versus early loading of dental implants in par-tially edentulous patients: 1-year results from a multicenter, randomized controlled clinical trial. Int J Oral Maxillofac Implants 2007;22:815-822.

17. Galli F, Capelli M, Zuffetti F, Esposito M, Testori T. Immediate non-occlusal vs early loading of dental implants in partially edentulous patients: a multicentre randomized clinical trial. Peri-implant bone and soft-tissue levels Clin Oral Implants Res 2008;19:546-552.

18. Trisi P, Rao W. Bone classification: clinical-histomorphometric comparison. Clin Oral Implants Res 1999;10:1-7.

19. Cannizzaro G, Leone M, Torchio C, Viola P, Esposito M. Immediate versus early loading of 7 mm long flapless-placed single implants: a split-mouth randomised controlled clinical trial. Eur J Oral Implantol 2008;1:277-292.