Abstract
Objective
To investigate the longevity and reasons for failure of posterior cast Class II gold inlays and onlays among a group of Norwegian adults. The term inlay was used for both inlays and onlays.
Methods
A cohort of 138 patients regularly attending a general practice for check-up were examined in 2016. The patients had a total of 391 posterior gold inlays placed in the period 1970–2015. The inlays were categorized as successful, repaired or failed. Reasons for failure were classified as either “secondary caries”, “fractures”, “lost inlay” or “other”. Participation was voluntary and no compensation was given.
Results
The mean age of the patients at placement was 50.8 years (SD: 12.7 yr). Most gold inlays were placed in molars (85.9%) and 14.1% in premolars; 49.4% of the inlays were in the maxilla and the 50.6% in the mandibula. Average length of follow-up was 11.6 years (range: 1–46 years, SD: 7.9); 82.9% were classified as successful, 10.7% as repaired and 6.4% as failed. Reasons for failure were secondary caries (41.3%), lost inlay (25.4%), fractures (23.8%) and other (9.5%). Mean annual failure rate (AFR) was 1.69% for repaired and failed inlays combined. However, if repaired inlays were considered as success, the AFR decreased to 0.57%. Multi-level Cox regression analyses identified low age of the patient and high number of restored surfaces as risk factors for failure.
Conclusion
The present retrospective clinical study demonstrated an acceptable annual failure rate for Class II cast gold inlays.
Clinical significance
All dental restorations have a limited lifespan and will eventually need to be repaired or replaced. It is important to have in mind that the longevity of gold restorations have not been shown to be superior to that of resin composite, which should still be the first material of choice. However, gold inlays could be considered when the right indications are present.
1
Introduction
Changes in restorative treatment and ideology have forced dentists and patients to be faced with an overwhelming number of dental material options for posterior restorations. However, failure of restorations continues to be a problem in dental practice, and it is estimated that about 60% of a dentist’s working day still is attributed to placement and replacement of restorations [ ]. Use of amalgam as a dental restorative material has been banned in Norway since 2008, and resin composites have become the almost exclusively used restorative material [ , ]. In many parts of the world, the use of amalgam is also decreasing [ ]. However, for some patients an alternative to resin composites could be desirable [ ].
Indirectly placed gold cast restorations have for many years been considered as a durable restorative material with longevity superior to other restorative materials. This assumption may not be sufficiently supported by scientific evidence. First of all, there are few – if any – recent papers on longevity of gold inlays. Second, the longevity of gold inlays presented in earlier studies may not be as advantageous compared with what can be achieved with today’s restorative materials and cementation procedures. A review by Hickel and Manhart [ ] found annual failure rates (AFR) for cast gold inlays to range from 0% to 5.9%, with a median AFR of 1.2%. In the same study, the median AFR for amalgam, composite and glass-ionomer restorations was 3.3%, 2.2% and 7.7%, respectively. In another review, Manhart et al. [ ] found a mean AFR for posterior stress-bearing restorations to be: 3.0% for amalgam, 2.2% for composite, 2.9% for composite inlays, 1.7% for CAD/CAM ceramic restorations and 1.4% for cast gold inlays and onlays. These AFR values may have been impressive at that time, but shade in the light of recent studies presenting AFR’s of resin composites around 1–3% [ ]. Mjör and Medina [ ] found that the longevity of cast gold restorations exceeded that of other dental materials present at that time by a factor of two to four. However, that study is based on data on ‘age of failed restorations’; an analysis which has been shown to overestimate the longevity of older materials (e.g. gold inlays) compared to newer materials (e.g. resin composites) [ ]. Survival analyses in which the life time of both restorations that have failed and those that remain in service is taken into account, such as Kaplan-Meier statistics, is hitherto the best method for calculating longevity of dental restorations [16,17]. Thus, if gold cast inlays should continue to be a part of dentists’ arsenal of restorative materials, newer studies performed by Kaplan-Meier analyses should be performed.
The use of cast gold restorations have been low lately [ , ]. Advances in adhesive technique and escalation in aesthetic demands have increased indications for tooth-coloured restorations. In a recent questionnaire study presenting a patient case with a fractured amalgam restoration in a premolar, only 3.1% of the Norwegian dentists considered to restore the tooth with a gold inlay. Other, more aesthetic, prosthetic options were chosen by 10.9% of the dentists [ ]. The manufacturers are steadily offering new tooth-coloured materials with improving properties, both for direct and indirect techniques, as well as materials for CAD/CAM techniques. To ad knowledge to the quality of this old technique, in order to compare and be aware of what to expect from the new materials, the aim of the present study was to investigate the longevity and reasons for failure of posterior Class II gold cast restorations among a group of Norwegian adults in a private practice.
1
Introduction
Changes in restorative treatment and ideology have forced dentists and patients to be faced with an overwhelming number of dental material options for posterior restorations. However, failure of restorations continues to be a problem in dental practice, and it is estimated that about 60% of a dentist’s working day still is attributed to placement and replacement of restorations [ ]. Use of amalgam as a dental restorative material has been banned in Norway since 2008, and resin composites have become the almost exclusively used restorative material [ , ]. In many parts of the world, the use of amalgam is also decreasing [ ]. However, for some patients an alternative to resin composites could be desirable [ ].
Indirectly placed gold cast restorations have for many years been considered as a durable restorative material with longevity superior to other restorative materials. This assumption may not be sufficiently supported by scientific evidence. First of all, there are few – if any – recent papers on longevity of gold inlays. Second, the longevity of gold inlays presented in earlier studies may not be as advantageous compared with what can be achieved with today’s restorative materials and cementation procedures. A review by Hickel and Manhart [ ] found annual failure rates (AFR) for cast gold inlays to range from 0% to 5.9%, with a median AFR of 1.2%. In the same study, the median AFR for amalgam, composite and glass-ionomer restorations was 3.3%, 2.2% and 7.7%, respectively. In another review, Manhart et al. [ ] found a mean AFR for posterior stress-bearing restorations to be: 3.0% for amalgam, 2.2% for composite, 2.9% for composite inlays, 1.7% for CAD/CAM ceramic restorations and 1.4% for cast gold inlays and onlays. These AFR values may have been impressive at that time, but shade in the light of recent studies presenting AFR’s of resin composites around 1–3% [ ]. Mjör and Medina [ ] found that the longevity of cast gold restorations exceeded that of other dental materials present at that time by a factor of two to four. However, that study is based on data on ‘age of failed restorations’; an analysis which has been shown to overestimate the longevity of older materials (e.g. gold inlays) compared to newer materials (e.g. resin composites) [ ]. Survival analyses in which the life time of both restorations that have failed and those that remain in service is taken into account, such as Kaplan-Meier statistics, is hitherto the best method for calculating longevity of dental restorations [16,17]. Thus, if gold cast inlays should continue to be a part of dentists’ arsenal of restorative materials, newer studies performed by Kaplan-Meier analyses should be performed.
The use of cast gold restorations have been low lately [ , ]. Advances in adhesive technique and escalation in aesthetic demands have increased indications for tooth-coloured restorations. In a recent questionnaire study presenting a patient case with a fractured amalgam restoration in a premolar, only 3.1% of the Norwegian dentists considered to restore the tooth with a gold inlay. Other, more aesthetic, prosthetic options were chosen by 10.9% of the dentists [ ]. The manufacturers are steadily offering new tooth-coloured materials with improving properties, both for direct and indirect techniques, as well as materials for CAD/CAM techniques. To ad knowledge to the quality of this old technique, in order to compare and be aware of what to expect from the new materials, the aim of the present study was to investigate the longevity and reasons for failure of posterior Class II gold cast restorations among a group of Norwegian adults in a private practice.
2
Material and methods
The patient‘s electronic dental records (EDR) and radiological records of a general practice in Norway were used for collecting the data for this study. All Class II gold inlays in premolars and molars of patients attending the dental practice for check-up and follow-up treatment in 2015–2016 were included in the study. The inlays were placed in the time period 1970–2015. In addition, data on 23 patients that were no longer attending the dental practice were extracted from the patient records and included in the study. In these cases the last registered examination at the dental clinic was used as the censoring date. In the present study, the term inlay is used for both inlays and restorations including cusps (onlays). Full golden crowns were excluded. The same practitioner (GS) evaluated the inlays included in the present study as part of the patients’ regular dental and radiological examination. The gold inlays were evaluated based on clinical and radiological examination, and categorized as successful, repaired or failed. Inlays that had previously been repaired or failed were assessed retrospectively based on listings in the dental records, and the dates of replacement and reason for failure were extracted. Reasons for failure were classified as either “secondary caries”, “fractures”, “lost inlay” or “other” (extracted tooth or endodontically treated tooth). Patient-related information, such as gender and age at the time of placement of the restoration, was extracted from the electronic dental records (EDR). The patients’ general risk of failure (due to caries, fractures and other incidences) was estimated by their accumulated DMFT-score over a 20-years period (1996–2016), which covered most of the observation period. The reason for this time interval was that the dental clinic started using EDR’s in 1996, providing access to DMFT data.
The mean annual failure rate of the investigated restorations was calculated according to the formula (1- y ) z = (1- x ), in which “ y ” is the mean annual failure rate and “ x ” is the total failure rate at z years [ ]. Empirical failure rates at 3, 5, 7, 10 and 18 years with 95% CI were calculated using Kaplan-Meier statistics ( Table 1 ). Kaplan-Meier analyses were also used to create survival curves ( Fig. 1 ). The possible dependence of multiple inlays placed in the same patients was taken into account by use of a multi-level approach. The multi-level Cox regression models were fitted with gamma-distributed heterogeneity using the coxph- function in the software R, presenting a frailty model with time effects on the patient level given by:
Here, h 0 ( t ) is the baseline hazard that describes the shape of the hazard rate as a function of time, exp ( β i X ij ) is the hazard ratio that describes how the size of the hazard rate depends on the covariates, while α j represents the patient frailty term. The multi-level Cox regression analyses were performed with failure of gold inlay as the dependent variable and the following independent variables: patient’s age, gender, patient‘s risk (total accumulated DMFT-score during 20 years), jaw (mandible vs. maxilla), tooth type (molar vs. premolar), number of surfaces of the inlay and cuspal coverage (no/yes). Collinearity was checked using the criterion VIF <5 and no independent variables were found to invalidate the analysis. A significance level of 5% was used throughout. IBM SPSS 24.0.0.2 (Statistical Package for the Social Sciences) and R version 3.4.1 (The R Foundation for Statistical Computing) were used to analyse the data. STATA 14.2 (Statistics/Data Analysis) was used to create hazard plots.
| Time | Failure rate | 95% CI |
|---|---|---|
| 3 years | 0.99 | 0.98–1.00 |
| 5 years | 0.98 | 0.96–1.00 |
| 7 years | 0.97 | 0.95–0.99 |
| 10 years | 0.95 | 0.92–0.98 |
| 18 years | 0.88 | 0.80–0.96 |
Participation was voluntary and no compensation was given to the respondents. The implementation of the study was approved by the Regional Committee for Medical Research Ethics in Norway (ID: 2015/1324).
3
Results
From 1970–2015 a total of 391 Class II posterior gold inlays were placed in 138 patients (81 males and 57 females). The number of inlays in each patient ranged from one to thirteen. About one-third of the patients (n = 44) received only one inlay, 27 (19.6%) patients received two inlays, 26 (18.8%) patients received three and 41 (29.7%) patients received four or more inlays. The distribution of the 391 gold inlays according to gender was 67.8% placed in male (n = 265) and 32.2% placed in female patients (n = 126). The mean age of the patients at placement of the inlays was 50.8 years (SD: 12.7 yr). Most gold inlays were placed in molars (85.9%, n = 334) and 14.1% were placed in premolars. The distribution of inlays was quite similar in the maxilla (49.4%, n = 192) and the mandible (50.6%, n = 197). Almost two-thirds of the inlays had cuspal coverage (69.8%). The number of restored surfaces varied from two to five.
Average length of follow-up of the inlays was 11.6 years (range: 1–46 years, SD: 7.9). After the individual follow-up period, 82.9% (n = 324) were classified as successful, 10.7% (n = 42) had been repaired and 6.4% (n = 25) were deemed to have failed. Reasons for failure were secondary caries (41.3%), lost inlay (25.4%), fractures (23.8%) and other (9.5%), e.g. endodontic complications. The main reasons for repair (n = 42) were secondary caries and fractures. Mean annual failure rate (AFR) was calculated to be 1.69% for repaired and failed inlays combined. However, if repaired inlays were considered as success, the AFR decreased to 0.57%. This difference is shown in Fig. 1 , presenting separate survival curves for these two approaches. Empirical failure rates at different time points are given in Table 1 .
Multi-level Cox-regression analyses were performed to identify factors related to failure of the gold restorations ( Table 2 ). In the adjusted analyses, increasing age of the patient and increasing number of restored surfaces remained statistically significant. Thus, the patients’ gender, patient‘s risk, the inlay’s location in the jaw and tooth type, and whether there was cuspal coverage or not were not found to be significantly related to the survival of the gold restorations. Hazard plots show that the hazard (risk of failure of gold inlays) for patients older than 50 years was lower than for patient aged below 50 years. Thus, lower risk of failure is expected as the patient gets older ( Fig. 2 ). There was also a clear indication that if the number of surfaces was five, then the hazard (risk of failure) was higher than in inlays with <5 surfaces. Thus, as the number of surfaces increases, so does the risk of failure and the risk is more prominent if the number of surfaces is above four ( Fig. 3 ).
