Sven Rinke Matthias Roediger Peter Eickholz Katharina Lange Dirk Ziebolz
Technical and biological complications of single-molar implant restorations
Authors’ affiliations: Sven Rinke, Private Practice Hanau, Germany Matthias Roediger, Department of Prosthodontics, Georg-August-University, Goettingen, Germany Peter Eickholz, Department of Periodontology, Center for Dental Medicine, Oral and Maxillofacial Medicine, Johann Wolfgang Goethe-University, Frankfurt/Main, Germany Katharina Lange, Department of Medical Statistics, Georg-August-University, Goettingen, Germany Dirk Ziebolz, Department of Operative Dentistry, Preventive Dentistry and Periodontology, GeorgAugust-University, G€ ottingen, Germany
Key words: cement, ceramics, clinical studies, dental implant(s), peri-implant infections(s),
Corresponding author: Priv.-Doz. Dr. Sven Rinke, MSc, MSc Private Practice Geleitstr. 68 63456 Hanau Germany Tel.: +49 6181 189 0950 Fax: +49 6181 189 0959 e-mail: [email protected]
dependent peri-implantitis rates were calculated, and the influencing factors were identified using
tobacco Abstract Objectives: Retrospective evaluation of the biological and technical complications in implantsupported single-tooth molar restorations performed in a private practice after functional periods of ≥4 years. Material and Methods: Sixty-five patients (34 females, age 51.7 10.6 years) with 112 implants received annual follow-up examinations and participated in a maintenance program. The survival (in situ) and success (complication-free) rates of implants and superstructures were evaluated. Timea multiple Cox regression. Results: The implant survival rate was 100%. Three of 112 crowns required replacement (prosthetic survival rate = 98.1%). Thirty technical complications were observed: loss of retention (16), ceramic fracture (10), and screw loosening (4). The success rate of the superstructures was 79.0% after 7 years. Overall, 9.2% of the patients developed peri-implantitis (probing depth ≥5 mm, BOP, suppuration, bone loss ≥3.5 mm); (smokers: 41.6%, non-smokers: 1.8%). After 7 years, the time-dependent implant success rate (free of peri-implantitis) was 100% for non-smokers and 58.6% for smokers. Multiple analysis showed a significant effect of smoking (hazard ratio, 19.5; P = 0.008) on peri-implantitis. Conclusions: Implants with cemented single-tooth restorations in the molar region constitute a reliable treatment in private practice. Smokers have a significantly increased peri-implantitis rate.
Date: Accepted 22 February 2014 To cite this article: Rinke S, Roediger M, Eickholz P, Lange K, Ziebolz D. Technical and biological complications of single-molar implant restorations. Clin. Oral Impl. Res. 00, 2014, 1–7 doi: 10.1111/clr.12382
Implant-supported single crowns have been shown to exhibit a high predictability for good clinical long-term results (Pjetursson & Lang 2008; Jung et al. 2012). Implant survival rates were shown to vary between 96% and 97%, and the survival of the prosthetic construction was shown to be between 87% and 97% after 5 years. Even the 10-year survival rate of single-tooth implants was reported to be 95%, while the survival rate of the prosthetic restorations was shown to be 89% (Jung et al. 2012). Implant-supported single-tooth restorations have mainly been examined with regard to their use in the anterior region (Jung et al. 2012). Although some studies have specifically examined their use in the molar region, the interpretation of these results is limited by the relatively short observational periods of no longer than 5 years and the heterogeneity of the study designs. Cumulative implant survival rates for single-molar implants were
© 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
reported to range between 91.1% and 100% for mean observational periods ranging from 2 to 5 years (Becker et al.1999; Schwartz-Arad et al. 1999; Polizzi et al. 2000; Romanos & Nentwig 2000; Simon 2003; Levin et al. 2006; Levine et al. 2007; Kim et al. 2010; Koo et al. 2010; Annibali et al. 2011). Long-term data with observational periods >5 years are still sparse for implant-borne molar crowns (Fugazzotto 2001; Blanes et al. 2007). Technical complications are common in single-tooth implant restorations (Theoharidou et al. 2008; Chaar et al. 2011; Albrektsson & Donos 2012; Jung et al. 2012). Detailed information on the technical and biological complications (periimplantitis) for single-tooth implants in the molar region is lacking for extended evaluation periods (Jung et al. 2012). Thus, the aim of the present study was to retrospectively evaluate implants using a morse taper connection for single-tooth implants to replace molars under private practice conditions.
1
Rinke et al Complications of single-molar implants
Materials and methods Patients
This retrospective study evaluated partially edentulous patients who were restored with single-tooth implants and metal-ceramic crowns in the molar region in a private dental office between January 1, 1999 and June 30, 2007. The following inclusion criteria were applied:
• •
•
• • •
single-tooth implants in the molar region of the maxilla and mandible of the same type (Ankylos, Dentsply Friadent GmbH, Mannheim, Germany); metal-ceramic crowns composed of a high-precious metal alloy (Degunorm, DeguDent GmbH, Hanau, Germany) and luted with a provisional cement (Temp Bond, Kerr Hawe GmbH, Karlsruhe, Germany); at least annual prophylaxis or supportive periodontal therapy (SPT) at the same dental office where the implants were inserted, including a periodontal examination (probing pocket depth [PPD] and bleeding on probing [BOP]) complete radiographic documentation with panoramic radiographs (PT); complete medical history including smoking habits; complete documentation of the superstructure-related complications or failures.
The following applied:
• • • •
exclusion
criteria
were
aggressive periodontitis (Lang et al. 1999); no prophylaxis or periodontal therapy prior to the implant therapy; inadequate radiographs; other missing data.
The study was approved by the Ethics Committee of the Medical Faculty of the Georg-August University, Germany G€ ottingen (application no. 3/2/10). The recommendations for strengthening the reporting of observational studies in epidemiology (STROBE) were followed. Treatment course
Patients with a history of chronic periodontal disease (Armitage 1999), (scaling and root planning, or surgical therapy within 5 years prior to implant placement) received supportive therapy prior to implant placement while the remaining patients received at least one prophylaxis treatment prior to implant placement. Patients exhibiting ineffective plaque control (plaque control record (PCR) >35%)
2 |
Clin. Oral Impl. Res. 0, 2014 / 1–7
were seen for additional professional hygiene appointments until a PCR < 20% was established (Rinke et al. 2011). Treatment was completed at a single private practice in Germany by two dentists with several years of experience and specific training in oral implant therapy. Implant placement, second stage surgery, and prosthetic treatment were completed according to the manufacturer’s standard protocol. Following the insertion of the prosthetic restoration, the patients were instructed in dental hygiene home care. They were then seen for post-implant maintenance every 3– 12 months. Prophylaxis and SPT were performed in patients with and without a history of periodontal disease, respectively. The treatment encompassed the following elements for all patients at each appointment: Assessment of gingival bleeding index (Ainamo & Bay 1975) and PCR (O’Leary et al. 1972), re-instruction, and remotivation to effective individual plaque control, professional tooth cleaning, and polishing of all teeth using rubber cups and polishing paste, application of a fluoride gel. Once a year (in case of prophylaxis) or twice a year (in case of SPT), a dental status and PPD measurements were obtained at four sites per tooth/ implant. Sites exhibiting PPD = 4 mm and BOP as well as sites with PPD ≥5 mm were scaled subgingivally using sonic and hand instruments. In case of a peri-implant mucositis, the scaling around implants was performed with a special tip for sonic scalers (Kavo Sonicflex implant, Kavo Dental GmbH, Biberach, Germany) followed by manual instrumentation and the subgingival application of a chlorhexidine gel (Corsodyl 1% Dental Gel, GlaxoSmithKline, Hamburg, Germany). Prophylaxis/SPT was rendered to most patients in 3-month intervals during the first year after implant insertion and later on in 6month intervals. Patients exhibiting ineffective plaque control during the maintenance phase (PCR > 35%) were seen four times a year for SPT (3 months intervals). This close recall was maintained until a PCR < 20% was established for three consecutive SPT visits. Data collection
Using patient records, the patients were evaluated according to the following parameters: age (at implant placement), gender, medication, smoking, implant location (according to the FDI scheme), implant length, number of implants, implant loss, time of crown placement, and observation period. All
prosthetic-technical complications, including implant and abutment fracture, screw loosening, retention loss, and fracture of the veneering ceramic were documented. Each patient’s clinical examination was completed at annual intervals during prophylaxis or SPT, and the following parameters were assessed:
• •
Survival analysis of the implant and prosthetic restoration and Technical complications.
Probing pocket depth and BOP were assessed using a millimeter-scaled periodontal probe (PCP 15; Hu-Friedy, Chicago, IL, USA). PPDs were measured at four measurement sites per implant. The deepest PPD of each implant was used as the diagnostic criterion for peri-implantitis. In case of an increased PPD (≥5 mm) and BOP, a PT was used to assess the radiographic bone loss to confirm the peri-implantitis diagnosis. Bone loss was determined by a metric analysis of the panoramic radiographs. All PTs were obtained using the same digital X-ray device (Orthophos, Sirona Dental Systems GmbH, Bensheim, Germany). Data were analyzed with the respective computer program (Sidexis XG, Sirona Dental Systems GmbH, Bensheim, Germany) and a calibrated screen (SyncMaster 2443SW, Samsung Electronics GmbH, Bad Schwalbach, Germany). The distance between the implant shoulder and the marginal bone level (MBL) was measured at the mesial and distal aspect of each implant. MBL was defined as the most apical location of the bone margin adjacent to an implant surface. The site with the most pronounced bone loss was chosen to represent the implant. Threshold level for a progressive bone loss was a bone level located at least 3.5 mm more apically related to the implant shoulder compared with the last available radiograph. Baseline radiographs after implant placement were used to check the original bone level around the implant. The original protocol for the implant system applied in this study recommends a position of the implant shoulder slightly subcrestal (0.5 mm). Implants placed more than 1 mm subcrestally, thus violating the surgical protocol, were excluded from the examination. All radiographs were read by the same calibrated operator (DZ). Radiographs from 10 patients with 16 implant sites were selected for a second analysis of the peri-implant bone level to assess the intra-examiner variability. These radiographs were chosen using a table of random numbers. In 93.7% of the analyzed implants, the intra-examiner analysis demonstrated a difference of the measurements
© 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
Rinke et al Complications of single-molar implants
Case definition
The respective survival and success of both the implant and the prosthetic reconstruction were the dependent variables of the study. Survival was defined as the implant or the prosthetic reconstruction remaining in situ at the follow-up examination visit without presenting an absolute failure (in situ criterion). Implant success was defined as being free of technical or biological complications (periimplantitis) for the entire follow-up period. Peri-implantitis was defined according to (Karoussis et al. 2004; Fransson et al. 2005; Roos-Jansaker et al. 2006a,b): PPD ≥ 5 mm, with BOP/suppuration and radiographic bone loss with a distance of at least 3.5 mm between implant shoulder and bone level. The date of the X-ray on which bone loss ≥3.5 mm was assessed for the first time was used for the time-dependent analysis of the event “peri-implantitis.” Prosthetic success was defined as being free of technical complications (i.e., the restoration did not require any intervention to maintain function during the entire followup) (Albrektsson & Donos 2012). Statistical analysis
For the statistical evaluation, information on the survival and success portions of the implants and prosthetic reconstructions was used. The implant/restoration was defined as the observational unit. The survival/success time of a restoration/implant was defined as the period between the cementation and the last follow-up or, in the case of a failure or clinical intervention, the appointment scheduled to address the failure/complication, as documented in the patient’s file. The timedependent survival rates (in situ criterion) and the success rates of the implants (no peri-implantitis) and prosthetic restorations (intervention-free) were calculated according to the Kaplan–Meier analysis. Implant position (maxilla/mandible), history of periodontal treatment, smoking, and the percentage of pocket probing depths >5 mm (more than 30%) in the remaining natural teeth were included as covariables for calculating the time-dependent peri-implantitis rates. A Cox regression analysis was performed for the
influencing factors. Different observations in the same patient (several implants per patient) were dependent. This dependence was allowed by adjusting the variance estimations in the Cox regression model. Thus, for data analysis, a marginal model was applied (Gerds et al. 2009). A type one error a of 5 mm” could only be seen in the analysis of this covariable (Fig. 4), but diminishes in the multiple analysis (Table 4).
The survival rate of the implants placed in this study was 100%, and it was 98.1% (95% CI, 0.955–1) for the prosthetic restoration after a mean follow-up of 7 years. This result is in accordance with a systematic review that analyzed the long-term results of singletooth implants (Jung et al. 2012). Clinical data for single-tooth replacements in the molar region with follow-ups of more than 5 years are still limited. Fugazzotto (2001) reported a cumulative survival rate for single-tooth molar implants in a retrospective practice-based study of 96.3% after up to 13 years. Blanes et al. (2007) reported a 10year cumulative survival rate of 97.7% for implants placed in the molar and premolar regions. Both findings are in good agreement with the results of the present study with survival rates for implants and superstructures of more than 98% after a mean followup of 7 years. Based on available systematic reviews (Theoharidou et al. 2008; Chaar et al. 2011; Jung et al. 2012), the most frequent technical complications with single-tooth implant restorations are the loss of retention in cemented superstructures, loosening of the abutment screw, and fracture of the veneering ceramic. The most frequent technical complication in the present study was loss of
Table 4. Results of the multivariate analysis of potential risk factors for biological implant-related complication (peri-implantitis) (Cox regression model) Factor Smoking habit Smoker vs. non-smoker Periodontal disease Perio treatment vs. No perio treatment Residual pockets 5 mm vs. >30% of natural teeth with PPD >5 mm JAW Maxilla vs. mandible
4 |
Clin. Oral Impl. Res. 0, 2014 / 1–7
SE
Hazard (exp (coef)
P
2.971
0.886
19.5
0.0008
0.299
0.999
1.349
0.760
0.210
1.353
1.234
0.880
1.460
0.947
0.232
0.120
Coefficient
retention (n = 16) followed by ceramic fractures (n = 10), while abutment screw loosening occurred in only four of the 112 implants. In the present study, 16 of 112 crowns (14%) had to be recemented. The prosthetic success rate reported for implant-supported single crowns cemented with temporary cement in other studies ranged between 72% and 84% for follow-ups of 2–5 years (Krennmair et al. 2002; Simon 2003; Schwarz et al. 2012). The high incidence of recementations for the crowns luted with temporary cement in the present study is in good accordance with other studies. The second most frequent technical complications in the present study were the fracture of the veneering ceramic, with an incidence of 8.9% after 7 years. The high incidence of ceramic fractures is in line with the findings of two systematic reviews in which the prevalence of this technical complication ranged between 3.5% and 5% for single crowns within 5-year observation periods (Chaar et al. 2011; Jung et al. 2012). In two systematic reviews, the incidence of screw loosening in single crowns was reported to be up to 6% within 3 years (Theoharidou et al. 2008) and 8.8% within 5 years (Jung et al. 2012). In the present study, a loosening of the abutment screw occurred in only four of the 112 single-molar crown restorations, resulting in a complication rate of 4% after 7 years. A possible explanation for this relatively low rate may be the morse taper connection of the implant system used in this study. Low rates of screw loosening have been reported in other studies that used implant systems with a morse taper connection (Romanos & Nentwig 2000; Mangano et al. 2009). Peri-implantitis is known as a serious biological complication that, if untreated, can result in implant loss (Mombelli et al. 2012). Based on the published results, it may be stated that the prevalence of peri-implantitis seems to be in the range of 10% of all implants and 20% of all patients during the 5–10 years after the implant placement (Mombelli et al. 2012). Smoking and a history of periodontal disease, particularly inadequate oral hygiene, are among the risk factors linked to the occurrence of peri-implantitis (Roos-Jansaker et al. 2006a,b; Rinke et al. 2011; Pjetursson et al. 2012; Costa et al. 2012; Mombelli et al. 2012). This observation was confirmed for the risk factor “smoking” by the present study. An implant-based time-to-event analysis revealed a success rate of 58.6% after 7 years
© 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
Rinke et al Complications of single-molar implants
Fig. 2. Success probability (no peri-implantitis) of single-tooth molar implants (red line = smokers, black line = non-smokers) using the Kaplan–Meier plot.
Fig. 3. Success probability (no peri-implantitis) of single-tooth molar implants (red line = perio treatment, black line = no perio treatment) using the Kaplan–Meier plot.
for the implants placed in smokers. In the non-smokers, no implants showed signs of peri-implantitis after the same observation period. According to the statistical analysis of the present study, smoking was associated with a significantly HR for peri-implantitis (HR 19.5, P = 0.0008). This result is also clearly marked by the difference in the patient-based peri-implantitis rate (nonsmokers, 1.8%; smokers, 41.6%). However, in the present study, a history of periodontal disease (HR: 1.349; P = 0.76) and the percentage of pocket probing depth >5 mm at the remaining natural teeth (HR: 1.234; P = 0.88) could not be determined as a risk factor for peri-implantitis. These findings may be explained by two specific aspects of
the present research: (i) Only patients with chronic periodontitis (Armitage 1999) were included. It seems plausible that patients with aggressive periodontitis are likely to exhibit a higher risk for peri-implant diseases. (ii) All patients followed up in this study participated at least irregularly in posttherapeutic maintenance. Another possible explanation for the missing association of residual pockets of >5 mm and the outcome variable can be seen in a confounding of this variable with the smoking status. Only 8% of the non-smokers showed a probing depth of >5 mm on more than 30% of the natural teeth, while in 50% of the smokers, more than 30% of the sites showed a probing depth of >5 mm. This might explain the missing
© 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
significance in the multivariate testing procedure applied in the present study, and the results of the analysis for the influencing factor “residual pockets” (Fig. 4). The method used for radiographic evaluation is another specific issue of the present study. While in most clinical studies on periimplant disease, intraoral radiographs were used; in this study, PTs were evaluated for the detection of marginal bone loss. There are only a few studies that have assessed the extent to which similar diagnostic results can be obtained comparing readings from intraoral and PT radiographs. Persson et al. (2003) found a high agreement between panoramic and intraoral radiographs when evaluating periodontal bone height. Taking into account the increasing concern about patient exposure to radiation and thus replacing a series of intraoral films (for a patient with multiple implants) with a single panoramic radiograph, this method certainly reduces the amount of exposure. Accounting the findings of Persson et al. (2003) and the high intraexaminer agreement of the measurements together with the advantage of a reduced radiation exposure, the usage of PT for the detection of a progressive bone loss seems to be justified, especially for practice-based investigations. As a positive aspect of this method, valid assessments of peri-implant disease can be provided from pre-existing radiographs. Another important aspect for the interpretation of the study is the definition of the endpoints. For the survival of implant and suprastructure (in situ criteria), commonly applied success criteria (intervention-free survival) were selected (Pjetursson et al. 2012). Nevertheless, the definition of implantrelated biological complications (peri-implantitis) shows a great heterogeneity in the studies published up to now, especially the threshold level for progressive bone loss is an important factor for the definition of peri-implantitis. In the present study, the threshold level for progressive bone loss was defined as a distance of at least 3.5 mm between implant shoulder and bone level. Other clinical studies using different implant systems have used a comparable threshold level of > 3 mm distance between the implant-abutment connection and the bone level (Fransson et al. 2005; Roos-Jansaker et al. 2006a,b); Rinke et al. 2011). Nevertheless, it is has to be stated that this represent a high threshold of radiographic bone loss marking a pronounced progression of disease. As treatment of peri-implantitis should be initiated in the early progression of bone loss, lower threshold
5 |
Clin. Oral Impl. Res. 0, 2014 / 1–7
Rinke et al Complications of single-molar implants
with single-tooth restorations placed in the molar area using the same implant type. Moreover, this practice-based study reports biological as well as technical complications on the implant and superstructure levels with a follow-up of more than 5 years (6.8 years). As this type of result is rare, it can make a valuable contribution to the evaluation of the prognosis of implant-based restorations placed under the typical conditions of a private practice.
Conclusion Considering the limitations of the present study, the following conclusions can be drawn. Fig. 4. Success probability (no peri-implantitis) of single-tooth molar implants (red line = PPD > 5 mm more than 30%, black line = PPD > 5 mm
Technical and biological complications of single-molar implant restorations
Authors’ affiliations: Sven Rinke, Private Practice Hanau, Germany Matthias Roediger, Department of Prosthodontics, Georg-August-University, Goettingen, Germany Peter Eickholz, Department of Periodontology, Center for Dental Medicine, Oral and Maxillofacial Medicine, Johann Wolfgang Goethe-University, Frankfurt/Main, Germany Katharina Lange, Department of Medical Statistics, Georg-August-University, Goettingen, Germany Dirk Ziebolz, Department of Operative Dentistry, Preventive Dentistry and Periodontology, GeorgAugust-University, G€ ottingen, Germany
Key words: cement, ceramics, clinical studies, dental implant(s), peri-implant infections(s),
Corresponding author: Priv.-Doz. Dr. Sven Rinke, MSc, MSc Private Practice Geleitstr. 68 63456 Hanau Germany Tel.: +49 6181 189 0950 Fax: +49 6181 189 0959 e-mail: [email protected]
dependent peri-implantitis rates were calculated, and the influencing factors were identified using
tobacco Abstract Objectives: Retrospective evaluation of the biological and technical complications in implantsupported single-tooth molar restorations performed in a private practice after functional periods of ≥4 years. Material and Methods: Sixty-five patients (34 females, age 51.7 10.6 years) with 112 implants received annual follow-up examinations and participated in a maintenance program. The survival (in situ) and success (complication-free) rates of implants and superstructures were evaluated. Timea multiple Cox regression. Results: The implant survival rate was 100%. Three of 112 crowns required replacement (prosthetic survival rate = 98.1%). Thirty technical complications were observed: loss of retention (16), ceramic fracture (10), and screw loosening (4). The success rate of the superstructures was 79.0% after 7 years. Overall, 9.2% of the patients developed peri-implantitis (probing depth ≥5 mm, BOP, suppuration, bone loss ≥3.5 mm); (smokers: 41.6%, non-smokers: 1.8%). After 7 years, the time-dependent implant success rate (free of peri-implantitis) was 100% for non-smokers and 58.6% for smokers. Multiple analysis showed a significant effect of smoking (hazard ratio, 19.5; P = 0.008) on peri-implantitis. Conclusions: Implants with cemented single-tooth restorations in the molar region constitute a reliable treatment in private practice. Smokers have a significantly increased peri-implantitis rate.
Date: Accepted 22 February 2014 To cite this article: Rinke S, Roediger M, Eickholz P, Lange K, Ziebolz D. Technical and biological complications of single-molar implant restorations. Clin. Oral Impl. Res. 00, 2014, 1–7 doi: 10.1111/clr.12382
Implant-supported single crowns have been shown to exhibit a high predictability for good clinical long-term results (Pjetursson & Lang 2008; Jung et al. 2012). Implant survival rates were shown to vary between 96% and 97%, and the survival of the prosthetic construction was shown to be between 87% and 97% after 5 years. Even the 10-year survival rate of single-tooth implants was reported to be 95%, while the survival rate of the prosthetic restorations was shown to be 89% (Jung et al. 2012). Implant-supported single-tooth restorations have mainly been examined with regard to their use in the anterior region (Jung et al. 2012). Although some studies have specifically examined their use in the molar region, the interpretation of these results is limited by the relatively short observational periods of no longer than 5 years and the heterogeneity of the study designs. Cumulative implant survival rates for single-molar implants were
© 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
reported to range between 91.1% and 100% for mean observational periods ranging from 2 to 5 years (Becker et al.1999; Schwartz-Arad et al. 1999; Polizzi et al. 2000; Romanos & Nentwig 2000; Simon 2003; Levin et al. 2006; Levine et al. 2007; Kim et al. 2010; Koo et al. 2010; Annibali et al. 2011). Long-term data with observational periods >5 years are still sparse for implant-borne molar crowns (Fugazzotto 2001; Blanes et al. 2007). Technical complications are common in single-tooth implant restorations (Theoharidou et al. 2008; Chaar et al. 2011; Albrektsson & Donos 2012; Jung et al. 2012). Detailed information on the technical and biological complications (periimplantitis) for single-tooth implants in the molar region is lacking for extended evaluation periods (Jung et al. 2012). Thus, the aim of the present study was to retrospectively evaluate implants using a morse taper connection for single-tooth implants to replace molars under private practice conditions.
1
Rinke et al Complications of single-molar implants
Materials and methods Patients
This retrospective study evaluated partially edentulous patients who were restored with single-tooth implants and metal-ceramic crowns in the molar region in a private dental office between January 1, 1999 and June 30, 2007. The following inclusion criteria were applied:
• •
•
• • •
single-tooth implants in the molar region of the maxilla and mandible of the same type (Ankylos, Dentsply Friadent GmbH, Mannheim, Germany); metal-ceramic crowns composed of a high-precious metal alloy (Degunorm, DeguDent GmbH, Hanau, Germany) and luted with a provisional cement (Temp Bond, Kerr Hawe GmbH, Karlsruhe, Germany); at least annual prophylaxis or supportive periodontal therapy (SPT) at the same dental office where the implants were inserted, including a periodontal examination (probing pocket depth [PPD] and bleeding on probing [BOP]) complete radiographic documentation with panoramic radiographs (PT); complete medical history including smoking habits; complete documentation of the superstructure-related complications or failures.
The following applied:
• • • •
exclusion
criteria
were
aggressive periodontitis (Lang et al. 1999); no prophylaxis or periodontal therapy prior to the implant therapy; inadequate radiographs; other missing data.
The study was approved by the Ethics Committee of the Medical Faculty of the Georg-August University, Germany G€ ottingen (application no. 3/2/10). The recommendations for strengthening the reporting of observational studies in epidemiology (STROBE) were followed. Treatment course
Patients with a history of chronic periodontal disease (Armitage 1999), (scaling and root planning, or surgical therapy within 5 years prior to implant placement) received supportive therapy prior to implant placement while the remaining patients received at least one prophylaxis treatment prior to implant placement. Patients exhibiting ineffective plaque control (plaque control record (PCR) >35%)
2 |
Clin. Oral Impl. Res. 0, 2014 / 1–7
were seen for additional professional hygiene appointments until a PCR < 20% was established (Rinke et al. 2011). Treatment was completed at a single private practice in Germany by two dentists with several years of experience and specific training in oral implant therapy. Implant placement, second stage surgery, and prosthetic treatment were completed according to the manufacturer’s standard protocol. Following the insertion of the prosthetic restoration, the patients were instructed in dental hygiene home care. They were then seen for post-implant maintenance every 3– 12 months. Prophylaxis and SPT were performed in patients with and without a history of periodontal disease, respectively. The treatment encompassed the following elements for all patients at each appointment: Assessment of gingival bleeding index (Ainamo & Bay 1975) and PCR (O’Leary et al. 1972), re-instruction, and remotivation to effective individual plaque control, professional tooth cleaning, and polishing of all teeth using rubber cups and polishing paste, application of a fluoride gel. Once a year (in case of prophylaxis) or twice a year (in case of SPT), a dental status and PPD measurements were obtained at four sites per tooth/ implant. Sites exhibiting PPD = 4 mm and BOP as well as sites with PPD ≥5 mm were scaled subgingivally using sonic and hand instruments. In case of a peri-implant mucositis, the scaling around implants was performed with a special tip for sonic scalers (Kavo Sonicflex implant, Kavo Dental GmbH, Biberach, Germany) followed by manual instrumentation and the subgingival application of a chlorhexidine gel (Corsodyl 1% Dental Gel, GlaxoSmithKline, Hamburg, Germany). Prophylaxis/SPT was rendered to most patients in 3-month intervals during the first year after implant insertion and later on in 6month intervals. Patients exhibiting ineffective plaque control during the maintenance phase (PCR > 35%) were seen four times a year for SPT (3 months intervals). This close recall was maintained until a PCR < 20% was established for three consecutive SPT visits. Data collection
Using patient records, the patients were evaluated according to the following parameters: age (at implant placement), gender, medication, smoking, implant location (according to the FDI scheme), implant length, number of implants, implant loss, time of crown placement, and observation period. All
prosthetic-technical complications, including implant and abutment fracture, screw loosening, retention loss, and fracture of the veneering ceramic were documented. Each patient’s clinical examination was completed at annual intervals during prophylaxis or SPT, and the following parameters were assessed:
• •
Survival analysis of the implant and prosthetic restoration and Technical complications.
Probing pocket depth and BOP were assessed using a millimeter-scaled periodontal probe (PCP 15; Hu-Friedy, Chicago, IL, USA). PPDs were measured at four measurement sites per implant. The deepest PPD of each implant was used as the diagnostic criterion for peri-implantitis. In case of an increased PPD (≥5 mm) and BOP, a PT was used to assess the radiographic bone loss to confirm the peri-implantitis diagnosis. Bone loss was determined by a metric analysis of the panoramic radiographs. All PTs were obtained using the same digital X-ray device (Orthophos, Sirona Dental Systems GmbH, Bensheim, Germany). Data were analyzed with the respective computer program (Sidexis XG, Sirona Dental Systems GmbH, Bensheim, Germany) and a calibrated screen (SyncMaster 2443SW, Samsung Electronics GmbH, Bad Schwalbach, Germany). The distance between the implant shoulder and the marginal bone level (MBL) was measured at the mesial and distal aspect of each implant. MBL was defined as the most apical location of the bone margin adjacent to an implant surface. The site with the most pronounced bone loss was chosen to represent the implant. Threshold level for a progressive bone loss was a bone level located at least 3.5 mm more apically related to the implant shoulder compared with the last available radiograph. Baseline radiographs after implant placement were used to check the original bone level around the implant. The original protocol for the implant system applied in this study recommends a position of the implant shoulder slightly subcrestal (0.5 mm). Implants placed more than 1 mm subcrestally, thus violating the surgical protocol, were excluded from the examination. All radiographs were read by the same calibrated operator (DZ). Radiographs from 10 patients with 16 implant sites were selected for a second analysis of the peri-implant bone level to assess the intra-examiner variability. These radiographs were chosen using a table of random numbers. In 93.7% of the analyzed implants, the intra-examiner analysis demonstrated a difference of the measurements
© 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
Rinke et al Complications of single-molar implants
Case definition
The respective survival and success of both the implant and the prosthetic reconstruction were the dependent variables of the study. Survival was defined as the implant or the prosthetic reconstruction remaining in situ at the follow-up examination visit without presenting an absolute failure (in situ criterion). Implant success was defined as being free of technical or biological complications (periimplantitis) for the entire follow-up period. Peri-implantitis was defined according to (Karoussis et al. 2004; Fransson et al. 2005; Roos-Jansaker et al. 2006a,b): PPD ≥ 5 mm, with BOP/suppuration and radiographic bone loss with a distance of at least 3.5 mm between implant shoulder and bone level. The date of the X-ray on which bone loss ≥3.5 mm was assessed for the first time was used for the time-dependent analysis of the event “peri-implantitis.” Prosthetic success was defined as being free of technical complications (i.e., the restoration did not require any intervention to maintain function during the entire followup) (Albrektsson & Donos 2012). Statistical analysis
For the statistical evaluation, information on the survival and success portions of the implants and prosthetic reconstructions was used. The implant/restoration was defined as the observational unit. The survival/success time of a restoration/implant was defined as the period between the cementation and the last follow-up or, in the case of a failure or clinical intervention, the appointment scheduled to address the failure/complication, as documented in the patient’s file. The timedependent survival rates (in situ criterion) and the success rates of the implants (no peri-implantitis) and prosthetic restorations (intervention-free) were calculated according to the Kaplan–Meier analysis. Implant position (maxilla/mandible), history of periodontal treatment, smoking, and the percentage of pocket probing depths >5 mm (more than 30%) in the remaining natural teeth were included as covariables for calculating the time-dependent peri-implantitis rates. A Cox regression analysis was performed for the
influencing factors. Different observations in the same patient (several implants per patient) were dependent. This dependence was allowed by adjusting the variance estimations in the Cox regression model. Thus, for data analysis, a marginal model was applied (Gerds et al. 2009). A type one error a of 5 mm” could only be seen in the analysis of this covariable (Fig. 4), but diminishes in the multiple analysis (Table 4).
The survival rate of the implants placed in this study was 100%, and it was 98.1% (95% CI, 0.955–1) for the prosthetic restoration after a mean follow-up of 7 years. This result is in accordance with a systematic review that analyzed the long-term results of singletooth implants (Jung et al. 2012). Clinical data for single-tooth replacements in the molar region with follow-ups of more than 5 years are still limited. Fugazzotto (2001) reported a cumulative survival rate for single-tooth molar implants in a retrospective practice-based study of 96.3% after up to 13 years. Blanes et al. (2007) reported a 10year cumulative survival rate of 97.7% for implants placed in the molar and premolar regions. Both findings are in good agreement with the results of the present study with survival rates for implants and superstructures of more than 98% after a mean followup of 7 years. Based on available systematic reviews (Theoharidou et al. 2008; Chaar et al. 2011; Jung et al. 2012), the most frequent technical complications with single-tooth implant restorations are the loss of retention in cemented superstructures, loosening of the abutment screw, and fracture of the veneering ceramic. The most frequent technical complication in the present study was loss of
Table 4. Results of the multivariate analysis of potential risk factors for biological implant-related complication (peri-implantitis) (Cox regression model) Factor Smoking habit Smoker vs. non-smoker Periodontal disease Perio treatment vs. No perio treatment Residual pockets 5 mm vs. >30% of natural teeth with PPD >5 mm JAW Maxilla vs. mandible
4 |
Clin. Oral Impl. Res. 0, 2014 / 1–7
SE
Hazard (exp (coef)
P
2.971
0.886
19.5
0.0008
0.299
0.999
1.349
0.760
0.210
1.353
1.234
0.880
1.460
0.947
0.232
0.120
Coefficient
retention (n = 16) followed by ceramic fractures (n = 10), while abutment screw loosening occurred in only four of the 112 implants. In the present study, 16 of 112 crowns (14%) had to be recemented. The prosthetic success rate reported for implant-supported single crowns cemented with temporary cement in other studies ranged between 72% and 84% for follow-ups of 2–5 years (Krennmair et al. 2002; Simon 2003; Schwarz et al. 2012). The high incidence of recementations for the crowns luted with temporary cement in the present study is in good accordance with other studies. The second most frequent technical complications in the present study were the fracture of the veneering ceramic, with an incidence of 8.9% after 7 years. The high incidence of ceramic fractures is in line with the findings of two systematic reviews in which the prevalence of this technical complication ranged between 3.5% and 5% for single crowns within 5-year observation periods (Chaar et al. 2011; Jung et al. 2012). In two systematic reviews, the incidence of screw loosening in single crowns was reported to be up to 6% within 3 years (Theoharidou et al. 2008) and 8.8% within 5 years (Jung et al. 2012). In the present study, a loosening of the abutment screw occurred in only four of the 112 single-molar crown restorations, resulting in a complication rate of 4% after 7 years. A possible explanation for this relatively low rate may be the morse taper connection of the implant system used in this study. Low rates of screw loosening have been reported in other studies that used implant systems with a morse taper connection (Romanos & Nentwig 2000; Mangano et al. 2009). Peri-implantitis is known as a serious biological complication that, if untreated, can result in implant loss (Mombelli et al. 2012). Based on the published results, it may be stated that the prevalence of peri-implantitis seems to be in the range of 10% of all implants and 20% of all patients during the 5–10 years after the implant placement (Mombelli et al. 2012). Smoking and a history of periodontal disease, particularly inadequate oral hygiene, are among the risk factors linked to the occurrence of peri-implantitis (Roos-Jansaker et al. 2006a,b; Rinke et al. 2011; Pjetursson et al. 2012; Costa et al. 2012; Mombelli et al. 2012). This observation was confirmed for the risk factor “smoking” by the present study. An implant-based time-to-event analysis revealed a success rate of 58.6% after 7 years
© 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
Rinke et al Complications of single-molar implants
Fig. 2. Success probability (no peri-implantitis) of single-tooth molar implants (red line = smokers, black line = non-smokers) using the Kaplan–Meier plot.
Fig. 3. Success probability (no peri-implantitis) of single-tooth molar implants (red line = perio treatment, black line = no perio treatment) using the Kaplan–Meier plot.
for the implants placed in smokers. In the non-smokers, no implants showed signs of peri-implantitis after the same observation period. According to the statistical analysis of the present study, smoking was associated with a significantly HR for peri-implantitis (HR 19.5, P = 0.0008). This result is also clearly marked by the difference in the patient-based peri-implantitis rate (nonsmokers, 1.8%; smokers, 41.6%). However, in the present study, a history of periodontal disease (HR: 1.349; P = 0.76) and the percentage of pocket probing depth >5 mm at the remaining natural teeth (HR: 1.234; P = 0.88) could not be determined as a risk factor for peri-implantitis. These findings may be explained by two specific aspects of
the present research: (i) Only patients with chronic periodontitis (Armitage 1999) were included. It seems plausible that patients with aggressive periodontitis are likely to exhibit a higher risk for peri-implant diseases. (ii) All patients followed up in this study participated at least irregularly in posttherapeutic maintenance. Another possible explanation for the missing association of residual pockets of >5 mm and the outcome variable can be seen in a confounding of this variable with the smoking status. Only 8% of the non-smokers showed a probing depth of >5 mm on more than 30% of the natural teeth, while in 50% of the smokers, more than 30% of the sites showed a probing depth of >5 mm. This might explain the missing
© 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
significance in the multivariate testing procedure applied in the present study, and the results of the analysis for the influencing factor “residual pockets” (Fig. 4). The method used for radiographic evaluation is another specific issue of the present study. While in most clinical studies on periimplant disease, intraoral radiographs were used; in this study, PTs were evaluated for the detection of marginal bone loss. There are only a few studies that have assessed the extent to which similar diagnostic results can be obtained comparing readings from intraoral and PT radiographs. Persson et al. (2003) found a high agreement between panoramic and intraoral radiographs when evaluating periodontal bone height. Taking into account the increasing concern about patient exposure to radiation and thus replacing a series of intraoral films (for a patient with multiple implants) with a single panoramic radiograph, this method certainly reduces the amount of exposure. Accounting the findings of Persson et al. (2003) and the high intraexaminer agreement of the measurements together with the advantage of a reduced radiation exposure, the usage of PT for the detection of a progressive bone loss seems to be justified, especially for practice-based investigations. As a positive aspect of this method, valid assessments of peri-implant disease can be provided from pre-existing radiographs. Another important aspect for the interpretation of the study is the definition of the endpoints. For the survival of implant and suprastructure (in situ criteria), commonly applied success criteria (intervention-free survival) were selected (Pjetursson et al. 2012). Nevertheless, the definition of implantrelated biological complications (peri-implantitis) shows a great heterogeneity in the studies published up to now, especially the threshold level for progressive bone loss is an important factor for the definition of peri-implantitis. In the present study, the threshold level for progressive bone loss was defined as a distance of at least 3.5 mm between implant shoulder and bone level. Other clinical studies using different implant systems have used a comparable threshold level of > 3 mm distance between the implant-abutment connection and the bone level (Fransson et al. 2005; Roos-Jansaker et al. 2006a,b); Rinke et al. 2011). Nevertheless, it is has to be stated that this represent a high threshold of radiographic bone loss marking a pronounced progression of disease. As treatment of peri-implantitis should be initiated in the early progression of bone loss, lower threshold
5 |
Clin. Oral Impl. Res. 0, 2014 / 1–7
Rinke et al Complications of single-molar implants
with single-tooth restorations placed in the molar area using the same implant type. Moreover, this practice-based study reports biological as well as technical complications on the implant and superstructure levels with a follow-up of more than 5 years (6.8 years). As this type of result is rare, it can make a valuable contribution to the evaluation of the prognosis of implant-based restorations placed under the typical conditions of a private practice.
Conclusion Considering the limitations of the present study, the following conclusions can be drawn. Fig. 4. Success probability (no peri-implantitis) of single-tooth molar implants (red line = PPD > 5 mm more than 30%, black line = PPD > 5 mm
