Outcomes of dental implants placed in a surgical training programme


  • LP Smith,

    1. Department of Oral and Maxillofacial Surgery, The Royal Dental Hospital of Melbourne, Victoria.
    2. Department of Oral and Maxillofacial Surgery, Western Hospital, Footscray, Victoria.
    3. Melbourne Dental School, The University of Melbourne, Parkville, Victoria.
    Search for more papers by this author
  • M Ng,

    1. Department of Oral and Maxillofacial Surgery, The Royal Dental Hospital of Melbourne, Victoria.
    Search for more papers by this author
  • D Grubor,

    1. Department of Oral and Maxillofacial Surgery, The Royal Dental Hospital of Melbourne, Victoria.
    2. Melbourne Dental School, The University of Melbourne, Parkville, Victoria.
    Search for more papers by this author
  • A Chandu

    1. Department of Oral and Maxillofacial Surgery, The Royal Dental Hospital of Melbourne, Victoria.
    2. Department of Oral and Maxillofacial Surgery, Western Hospital, Footscray, Victoria.
    3. Melbourne Dental School, The University of Melbourne, Parkville, Victoria.
    Search for more papers by this author

Mr Arun Chandu
1st Floor
665 Mt Alexander Road
Moonee Ponds VIC 3039
Email: chandua@unimelb.edu.au


Background:  This study evaluates surgical outcomes and survival rates of implants placed in a multidisciplinary implant teaching programme.

Methods:  A retrospective review of all implant surgery performed over a 6-year period by accredited oral and maxillofacial surgery trainees at the Royal Dental Hospital of Melbourne was undertaken. Patients were reviewed for a minimum of 6 months post-implant placement. Implant survival was defined as those implants which were not removed, were clinically integrated as assessed by torque testing and in an appropriate position to receive a subsequent prosthesis. Kaplan-Meier analysis was used to assess overall survival and univariate factors affecting survival. Multivariate analysis used Cox proportional hazards models.

Results:  Over 6 years, 127 patients were treated. Follow-up data were present for 105 patients with 236 implants placed. Survival of implants at 1 and 5 years was 94 per cent and 92.8 per cent, respectively. The only univariate and multivariate factor which affected implant survival was perioperative bone grafting. All failed implants were single stage. Other factors such as patient age, smoking status, implant site, anaesthetic type, immediate or delayed placement, implant length and diameter, and medical comorbidities did not significantly affect implant survival.

Conclusions:  A satisfactory implant survival rate was found in a tertiary teaching centre. Perioperative bone grafting significantly increased the risk of implant failure.


Royal Dental Hospital of Melbourne


Prior to the implementation of a structured implant training programme, accredited Victorian oral and maxillofacial trainees who needed implant training had to do so through different providers such as: (i) university affiliated centres; (ii) manufacturer and distributor organized or sponsored seminars; (iii) interest groups; or (iv) assisting private surgeons. These findings have been previously reported in the literature.1 While these avenues of training are still incorporated in the current programme, the increasing scope and complexity of implant treatment required implementation of a standardized implant training programme.

One of the current challenges to the surgical registrar in general public hospitals is the inability to provide implant treatment in the environment of readily accessible specialist dental support.2 The Royal Dental Hospital of Melbourne and its affiliation with the University of Melbourne’s Dental School provides the opportunity for accredited trainees to treatment plan and perform procedures in a multidisciplinary fashion with other dental colleagues. There is little information presented in the literature concerning the outcomes of implant placement in the hands of trainee surgeons. This study aims to evaluate the clinical outcomes of patients undergoing implant placement, as part of a multidisciplinary implant training programme, performed by trainee oral and maxillofacial surgeons at a tertiary referral dental teaching hospital.

Materials and methods

This study was a retrospective audit of outcomes of patients treated over a six-year period (2002–2007 inclusive) as a part of a formalized implant education programme for accredited Victorian trainees in their first year of oral and maxillofacial surgery training. The study was approved by the Ethics/HREC of the Royal Dental Hospital of Melbourne (RDHM). Treatment of patients was carried out at the RDHM whose adult dental clinics provide general and specialized dental care for government health concession cardholders and dependants. The Oral and Maxillofacial Surgery Department is a specialist referral-based unit of the RDHM and is associated with the University of Melbourne in its role as a teaching hospital.

Potential implant patients are referred to the RDHM Combined Implant Clinic from all departments and community clinics of Dental Health Services Victoria. Once referrals are screened, patients are initially seen in a treatment planning clinic staffed by a consultant oral and maxillofacial surgeon, prosthodontist and periodontist. Once the initial treatment plan is agreed to, patients are then allocated to one or more of the specialist departments for further planning and management. Those patients allocated to our department are seen in one of the twice-weekly implant clinics by both a consultant and a trainee allocated to the clinic. Implants placed are restored by postgraduate students in prosthodontics or students undertaking the Graduate Diploma in Implantology. Liaison between the restoring dentist and the surgeon occurs prior to the finalization of the treatment plan.

A limited number of patients are accepted into the implant programme with entry criteria strictly applied. These include the following: (i) patients who have lost an anterior tooth, or have a tooth of poor prognosis particularly following trauma, and who otherwise have intact arches and a sound dentition; (ii) patients missing several teeth due to trauma, non-development or disease of the jaws; (iii) edentulous patients with severe alveolar resorption who are otherwise unable to retain a functional denture; (iv) patients with one or more posterior teeth missing from aesthetic zones or one or more adjacent posterior teeth missing in otherwise intact arches; (v) when the provision of a conventional fixed posterior prosthesis will necessitate the preparation of otherwise untreated sound teeth. Patients are charged a nominal fee for the placement and restoration of the implants.

For the purpose of the study, the inclusion criteria were all cases of surgical implants placed under the auspices of the Combined Implant Clinic in the Department of Oral and Maxillofacial Surgery, RDHM. All patients, regardless of age, medical history, smoking status, race and gender were included. Exclusion criteria for the study were those patients who underwent surgical implant placement or preoperative grafting under the care of other specialist departments. Patients were also excluded if a majority of their treatment was not planned or performed by a trainee surgeon or if the patient records were unavailable or inadequate. Patients with significant cardiovascular or pulmonary disease, insulin-dependent diabetes or those requiring overnight admission were unable to be treated in the RDHM Day Procedure Unit and could not be offered treatment under general anaesthesia. These patients were still offered treatment under local anaesthesia or referred to an external general hospital for inpatient care.

Surgical placement of the implants was carried out under local anaesthesia in the Oral and Maxillofacial Surgery Unit and under general anaesthesia in the Day Procedure Unit. Patients undergoing general anaesthesia received local anaesthetic infiltration or blockade at the time of surgery. Selection for general anaesthesia included those patients having complex surgery including extensive bone grafting and multiple implant placement, and those patients unable to tolerate surgical procedures under local anaesthesia due to anxiety. All procedures were carried out by surgical registrars under the supervision of a consultant oral and maxillofacial surgeon. A standardized sterile surgical protocol was carried out using sterile surgical drapes without the use of skin preparation.3,4 Perioperative gargles with chlorhexidine were not used. Intravenous perioperative antibiotics were given to those patients receiving treatment under general anaesthesia as per established RDHM day surgery protocol. The antibiotic given was either a broad spectrum penicillin or cephalosporin. Postoperative antibiotics and analgesics were given to all patients. Patients were reviewed postoperatively at 2 weeks, 3 month and 6 month intervals after implant surgery. Some patients underwent longer review periods as required. The review consisted of patient interviews regarding postoperative period, clinical examination and assessment of postoperative X-rays. X-rays were taken at 2 weeks postoperatively to assess the position of the implant and at the conclusion of the surgical postoperative follow-up period (6 months) to establish baseline peri-implant bone levels.

Data collection and analysis

The study group was identified initially by review of electronic records held by the RDHM by searching for all implant and bone graft related treatment codes. Further analysis then allowed isolation of the procedures performed by the Department of Oral and Maxillofacial Surgery as compared to those performed by other specialist dental departments. The medical and dental records of patients identified for inclusion were then reviewed.

The primary outcome variable was implant failure and this was defined as removal of the implant after insertion for any reason.5,6 Time to failure was determined as the duration in days from implant placement to removal of the implant. Implant failure was further categorized as intraoperative, those implants removed during the insert appointment; early failure, those implants removed within 6 weeks of insertion; and late failure, those implants removed beyond 6 weeks. Differentiation between early and late failures was made on a time basis rather than related to restoration with the final prosthesis. This is because some patients had prolonged time periods between implant surgery and restoration due to patient handover between postgraduate students at the end of university years. Implant survival was defined as those implants which were not removed, were clinically integrated as assessed by torque testing and in an appropriate position to receive a subsequent prosthesis.

Secondary variables were risk factors that may have affected implant survival. These were categorized as follows: (i) demographic variables; (ii) health variables including smoking status, diabetes mellitus, osteoporosis, immunosuppression, bisphosphonate and antibiotic use;7–14(iii) implant variables including location, number, length and diameter;15–19 (iv) implant timing variables including immediate or delayed surgical placement and number of stages of surgery; (v) and whether perioperative autogenous bone grafting was used.20 Smoking status was assigned based on medical records taken at the preoperative surgical and anaesthetic consultation.

All data collected were entered into a computer database. This implant database has become the basis of a prospective database of implants placed in our unit. Statistical analysis was carried out using the Statview Statistical Package Version 5 (SAS institute Inc, Cary, NC, USA). Kaplan Meier survival analysis was performed to assess implant survival. For the purpose of assessing factors affecting survival, intraoperative failures were excluded from the analysis as these implants did not achieve primary stability. Certain factors were also grouped less than or equal to the mean versus above the mean to facilitate analysis. This included patient age and implant length. For implant length, analysis was also undertaken for implants equal and below 10 mm in length. Ten millimetres was used as only a few 8 mm implants were placed. Position of implant placement was assessed either anterior or posterior to the second premolar.

Factors affecting survival of implants were assessed using Kaplan Meier survival analysis and log rank tests for univariate analysis. Cox proportional hazard models were used for multivariate analysis of univariate variables that were able to be assessed by log rank tests. Patients for whom no follow-up information was available were excluded from the statistical analysis. A p value less than 0.05 was taken to be significant.


One hundred and twenty-seven patients were identified as having implants placed by our unit. From this, 105 patients had follow-up data for 236 implants (average of 2.24 implants per patient inserted) and these were included for the analysis. The remaining 22 patients were excluded from analysis due to incomplete records, incomplete treatment due to patient withdrawal, ineligibility to continue treatment because of change in income status or the patients were lost to follow-up. The implants were placed by 9 oral and maxillofacial surgery trainees (mean 26.2 implants) during 6 months training at RDHM. Figure 1 shows the distribution of implant cases, fixture placements and failures over the time period of implants that were included in the study.

Figure 1.

 Distribution of implant cases, implant number and failures over the 6-year period of formalized implant training programme.

The mean age of patients was 43.2 years (range 17–72, SD 15.5) with a male to female ratio of 1:1.7. In regards to medical factors, 33.3 per cent of patients smoked, 3.8 per cent were on oral steroids while 4.8 per cent of patients were diabetic. No patients had a history of osteoporosis or bisphosphonate use. Table 1 presents the types of preoperative investigations used while Table 2 describes the types of implants and the implant surface, their length and diameter. The mean implant length was 11.9 mm (median 12, range 8–18). Thus, 12 mm was used as a cut-off for the purposes of survival analysis. Overall, 53 per cent of implants were placed anterior to the second premolars. Placement of implants in either mandible or maxilla was evenly distributed with 51 per cent of implants placed in the maxilla. Only 17 per cent of implants were placed as immediate implants at the time of tooth extraction with the rest being delayed implant placement after a healing period from the time of extraction. Fifteen per cent of implants received perioperative bone grafting. The proportion of implants placed as a single stage procedure was 61.3 per cent. Two stage procedures contributed in 36.6 per cent of implants placed. Procedures were evenly distributed for type of anaesthesia (i.e., 50 per cent having local anaesthesia).

Table 1.   Preoperative diagnostic imaging
PeriapicalOPGi-CAT†Cone Beam‡Dentascan§
  1. †Imaging Sciences International, Hatfield, USA.

  2. ‡3D Diagnostix Inc, Brighton, USA.

  3. §GE Healthcare, Giles, UK.

Table 2.   Distribution of implants placed by oral and maxillofacial surgery trainees between 2001–2007, implant type and surface, length and diameter described
Implant and surfaceTotalLengthDiameter
≤12 mm>12 mmNarrow <3.3 mmRegular >3.3–<4.8mmWide >4.8 mm
  1. †Nobel Biocare. Gothenburg, Sweden.

  2. ‡The Straumann Group. Basel, Switzerland.

  3. ¶Nobel Biocare. Gothenburg, Sweden.

  4. §Astra Tech. Mölndal, Sweden.

Brånemark† (anodic oxidation)1526785713015
Straumann‡ (grit blasted/acid etched)3733421916
3i§ (acid etched)211381137
Replace¶ (anodic oxidation)227151813
Astra§ (titanium blasted)413121

Survival and factors affecting survival were assessed by Kaplan Meier analysis and was based per implant. The mean review time from insertion was 18.3 months (range 10.0–61.6, SD 10.2). Survival of implants at 1 year was 94 per cent (95% CI 0.91–0.97) and this reduced to 92.8 per cent (95% CI 0.89–0.97) at 5 years (Fig 2). There were a total of 15 failures (6.3 per cent of implants placed). Five occurred intraoperatively (2.1 per cent), 8 in the early postoperative period (3.4 per cent) and 2 late failures developed (0.8 per cent) (Table 3).

Figure 2.

 Kaplan-Meier cumulative survival plot for survival time (months). Censor variable: fixture failure. Survival of implants at 1 and 5 years was 94% and 92.8%, respectively.

Table 3.   Failure of implants placed by oral and maxillofacial surgery trainees (15 implants), with the distribution between intraoperative failure (5 implants), early failure within 6 weeks of placement (8 implants) and late failure beyond 6 weeks (2 implants). Site, patient age, type of implant, diameter and clinical indication for removal of the implant are described
SitePatient ageTypeDiameterClinical indication for removal
  1. N: narrow diameter (N < 3.3 mm); R: regular diameter (3.3 mm > R < 4.8 mm), W: wide diameter (W > 4.8 mm).

Intraoperative implant failure 2.1% of implants placed
1650Brånemark Mk IIIRAbort due to insufficient bone
2146AstraRAbort due to insufficient bone
2129Brånemark Mk IVRAbort due to insufficient bone
4561StraumannRLack primary stability
4633Brånemark Mk IVWLack primary stability
Early implant failure 3.4% of implants placed
2352Brånemark Mk IIINFailure to integrate
3127Brånemark Mk IVRFailure to integrate
1432Brånemark Mk IVRFailure to integrate
1646Brånemark Mk IVWFailure to integrate
4650Brånemark Mk IIIRFailure to integrate
2452Brånemark Mk IIINFailure to integrate
2534Brånemark Mk IVRFailure to integrate
2646Brånemark Mk IVWFailure to integrate
Late implant failure 0.8% of implants placed
2123Brånemark Mk IVRIncorrect position
4117StraumannNIncorrect position

In regards to factors affecting survival, the 5 implants that did not achieve primary stability intraoperatively, were excluded from the analysis. Therefore for both univariate and multivariate analysis, the only factor that was found to significantly affect implant survival was perioperative bone grafts (p < 0.001 both for univariate and multivariate analysis). The increased relative risk of failure was 2.95 for perioperative bone grafts (95% CI 1.89–4.59). All other multivariate and univariate factors were not found to be significant.

Significant surgical complications were defined as those requiring further surgical intervention or admission to hospital and did not include implant failure. In this patient group there was a low incidence, with 3.8 per cent (4 patients) experiencing significant complication. This constituted significant complication in 2.5 per cent (6 implants) of implants. One patient experienced haemorrhage requiring diathermy cauterization to achieve haemostasis with subsequent failure of two implants. One patient experienced significant sinus and infraorbital infection after simultaneous Caldwell-Luc sinus lift and implant placement in the 15 and 16 region, requiring admission, incision and drainage plus parenteral antibiotics, however the implants were retained. Two patients experienced mental nerve paraesthesia, one of which was transient and the other permanent. Both of these were associated with implants placed in the second premolar site. This complication is attributed to the variation in course of the inferior alveolar nerve as it approaches the mental foramen, which has been previously described.21

At the time of review, 5 patients had implants that were not restored due to them leaving the treatment programme prior to completion of treatment. Of the remaining patients, 4 received an implant retained overdenture, 24 received an implant retained fixed partial denture and 94 were restored by one or more single unit implant retained crowns.


The changing scope of service provision by Australian dentists indicates that there is an increasing community demand to maintain a functional dentition in the presence of partially dentate dental arches.22 Long-term data demonstrating predictable clinical outcomes with implant retained dental prosthesis indicates that implant treatment should be considered when restoring deficient dental arches.23 Referral patterns within the dental community show high rates of referral for dental implants.24 Cottrell et al. reported only 10.3 per cent of general dentists undertake implant surgery themselves, with the rest being referred to oral and maxillofacial surgeons (50.1 per cent) and periodontists (31.0 per cent).24 Ongoing technological advances require a formalized approach to the training of oral and maxillofacial trainees in implantology to enable the maintenance of high levels of patient care.2,5,25

During the period of review, there were an increasing number of implants placed by the trainee surgeons (Fig 1). This reflects the initial difficulties that are encountered in instituting a formalized implant training programme in a public hospital setting. The relative plateau in the number of implants placed over the last 3 years represents the constraints placed on training by budget allocation and resource limitations to implant training.

Analysis of failures revealed that a majority of implant failures (53 per cent) occurred in the early postoperative period with the clinical indication for implant removal being failure to osseointegrate. There were a relatively high proportion of technical failures due to errors in treatment planning or surgical technique, which accounted for the intraoperative failures (33 per cent) and late failures (14 per cent). This can be explained by the experience levels of the surgical trainees. Lambert et al. investigated the effects of surgical experience on implant survival, with inexperienced surgeons having a significantly higher failure rate (4.4 per cent) as compared to experienced surgeons (2.1 per cent) in the treatment of their first 15 cases.25 In the preparation of the present study, we found it difficult to quantify the previous implant surgery experience, due to diverse clinical backgrounds of trainees prior to entering the training programme, however a majority of implants were placed by trainees in their first year of training, with no or limited implant experience. The absence of peri-implant mucosal conditions contributing to failure is presumably due to the duration of follow-up and is likely to be under-represented in this analysis. Of the records reviewed, no patients had been treated for peri-implant mucosal conditions. However, most patients had not undergone formal periodontal assessment within the dental hospital after discharge from the specialist departments (formal review period) as patients are referred back to regional clinics for routine care.

The present study reported an overall 1-year survival of 94 per cent and a 5-year survival of 92.8 per cent for the 236 fixtures placed by oral and maxillofacial surgery trainees. This result is in keeping with data for long-term survival available from the broader literature.26–28Analysis of the factors affecting implant survival have been extensively reported in the literature.12–18 In the present study, the only factor shown to significantly affect implant survival was implant placement in the presence of perioperative bone grafting. Of the 236 implants placed, 36 had autogenous particulate alveolar ridge augmentation at the time of implant surgery, with 3 subsequent failures. Although successful clinical outcomes have been demonstrated with perioperative bone grafting in the past,30,31 in the present study and the broader literature, it remains unclear whether the bone from such procedures is of benefit to the patient.32,33 Cristenssen et al. in a prospective cohort study of 45 implants also reported increased risk of bone loss found around implants placed with concomitant bone grafting.30

All other univariate and multivariate factors assessed in the present study were not found to be significant. These included demographic factors, implant site, implant type and size, and anaesthetic type. Important risk factors for implant failure including diabetes and smoking have been previously shown to be significant risk factors in implant failure.8,12,34,35 In the present study, the effect of non-insulin requiring type 2 diabetes was not found to be a significant risk factor in implant failure. This is supported by more recent studies demonstrating that type 2 diabetics that were well controlled (HBA1c < 7.0) and those that were moderately controlled (HBA1c 7.0–9.0), did not have a significantly increased risk in implant failure compared to the normal population.36 Despite the broader literature demonstrating the link between smoking and implant failure, this study failed to demonstrate significance which is likely to be a function of the small numbers of implant failures for statistical analysis.

Single stage surgery has been shown to be an effective treatment modality with the benefits of decreased cost, time, and operative discomfort with increased convenience.37 Several studies have demonstrated predictable treatment outcomes with this technique.37–40 Esposito et al. has previously found no clinical significance between single and two-stage surgeries but concluded that more research into this area is required.40 Interestingly in the present study, all implant failures occurred in single-stage procedures which can be attributed to the large proportion (63.4 per cent) of single-stage procedures within the study population; however this did not constitute statistical significance. This confirmed the findings of Fenlon et al. who demonstrated poorer rates of implant survival for single-stage implant surgery.41

The limitations of this study include the retrospective nature and the small number of patients compared to other analysis of implant survival.26–29 Although not a specific aim of the current study, separate assessment of implant failures, due to technical and biological reasons, was attempted in the analysis for factors affecting survival as all intraoperative implant failures were removed for the purpose of analysis. It would be our aim in the future when sufficient numbers are present, to reassess the data more specifically differentiating between biological and technical failures. With regards to bone grafting, only perioperative bone grafts were considered in terms of overall implant survival. Failures of bone grafts prior to implant placement were not within the limits of this study but will also be analysed in future publications. Despite these issues, meaningful results were found from the data collected. The implant database has now become prospective, allowing further analysis in a prospective fashion.

Overall, this study demonstrated that predictable surgical outcomes are achievable for implants placed by trainee surgeons in an adequately structured and conducted multidisciplinary implant training programme yielding satisfactory long-term surgical outcomes that are comparable to those placed in larger studies.26–29 Perioperative bone grafting was the only factor found to significantly increase the relative risk of implant failure and thus decrease survival. Many of the failures can be attributable to surgeon inexperience with outcomes associated with biological and patient causes alone demonstrating similar survival rates to that of the broader literature. Increases in the number of patients receiving dental implants as part of overall treatment plans indicates that further research into implant training is required to enhance patient outcomes and quality assurance.