Quantity of Research Available
A total of 1671 citations were identified in the literature search. Following screening of titles and abstracts, 1522 citations were excluded and 149 potentially relevant reports from the electronic search were retrieved for full-text review. Three potentially relevant publications were retrieved from the grey literature search. Of these potentially relevant articles, 123 publications were excluded for various reasons, while 29 publications met the inclusion criteria and were included in this report. Twenty-two studies provided answers to the first question about the longevity of all-ceramic crowns, two studies provided answers to the second question about the longevity of metal-ceramic crowns, four studies provided information about the comparative longevity of all-ceramic and metal-ceramic crowns, and one study provided cost evaluation. One of the comparative studies provided partial information about some contextual factors affecting crown survival. Appendix 1 describes the PRISMA flowchart of the study selection.
Appendix 5 presents a list studies that were already included in at least one of the included systematic reviews.
Summary of Study Characteristics
A summary of individual study characteristics is presented in Appendix 2.
What is the clinical evidence on the longevity of all-ceramic/porcelain crowns?
A total of ten systematic reviews,6–15 one RCT,16 and eleven non-randomized uncontrolled studies provided answers to this question (see Table 2 and Table 3).17–27
The ten systematic reviews included more than 60 unique primary studies; some of which were included in more than one systematic review (Table 2). All except four primary studies were observational uncontrolled studies published in the period between 1992 and 2013. The total number of crowns included the systematic reviews ranged from 12 crowns in Larsson et al.7 to 696 crowns in Pieger et al.6 The majority of the included primary studies had short-term follow-up below five years; one systematic review by Pieger et al.6 included a study with 10-year follow. Authors of all systematic reviews did not report from where patients were recruited, and they did not systematically report information on tooth vitality or the presence of post and core for endodontically treated teeth.
The included systematic reviews varied in terms of the evaluated all-ceramic material. Some of them evaluated one specific all-ceramic crown material; for example, Pieger et al.6 evaluated lithium disilicate crowns only, Larsson et al.7 evaluated zirconia-based crowns, while Heintze et al.9 and El-Mowafy et al.13 evaluated leucite-reinforced ceramic (IPS Empress). On the other hand Wang et al.8 Kassem et al.,10 Wittneben et al.,11 Wassermann et al.,12 and Ho et al.14 were not specific to the crown material and included different all-ceramic crown materials.
All included systematic review evaluated the longevity of all-ceramic crowns; however, they varied in their definition of success and failure. They also differed in reporting of the results; some of them simply reported the survival rate at the end of follow-up such as Heintze et al. 2010,9 Kassem et al. 2010,10 El-Mowafy et al. 2002,13 Ho et al. 2012.14 The remaining systematic reviews were more thorough and reported the cumulative survival rate which takes into account the time each crown was exposed to the risk of failure.
Rammelsberg et al.16 published the only randomized-controlled trial that evaluated longevity of all-ceramic crowns (Table 3). The trial tested the effect of preparation finishing line (chamfer versus shoulder finish lines) on the survival of metal free polymer crowns (Artglass). The authors included 71 patients and 117 single crowns in the trial, and followed patients for three years. During this period the authors counted crown failures, which was defined as fractures or decementation.
Five prospective uncontrolled studies provided information about the longevity of all-ceramic crowns (Table 3).17–21 Three studies recruited patients from a university-based practice,17,19,20 while the other two studies had patients from private dental practices.18,21 These were relatively small studies with a sample size ranging from 34 patients (41 crowns)20 to 50 patients (155 crowns).19 Ceramic material varied in these studies; two studies evaluated lithium disilicate all-ceramic crowns,17,20 and the remaining three studies evaluated one material each: zirconia-based crowns,18 densely sintered aluminum oxide,19 and leucite glass-ceramic.21 The main outcome in these studies was cumulative survival rate at two20,21 to nine years.17,19
The remaining six studies were retrospective uncontrolled studies (Table 3).22–27 These studies were conducted in Europe22,24,25,27,28 and the USA26 between the 2013 and 2015. Three studies were based on data collected from private dental practices,22,24,27 two were based on university-based patient data,25,26 and one study had a mixed population of university and private practices.23 The sample size ranged from 88 single crowns (from 70 patients)24 to 618 crowns (from 148 patients).23 Two studies evaluated lithium disilicate all-ceramic crowns,22,25 two studies evaluated zirconia-based crowns,23,24 and the last two studies had both zirconia- and alumina-based crowns.26,27 The authors of these studies used cumulative survival rates as the primary outcome.
What is the clinical evidence on the longevity of porcelain-fused-to-metal crowns?
Two uncontrolled studies provided information about the longevity of porcelain-fused-to-metal crowns (Table 4).29,30 Both studies were conducted in university-based settings in Germany. The prospective study by Hey et al.30 included 21 patients and 41 crowns, while the retrospective study by Behr et al.29 was based on the records of 997 crowns treated between 1984 and 2009.29 Hey et al. were interested in the longevity of crowns made with titanium coping using computer-aided design/computer-aided manufacturing (CAD/CAM) technology.30 Whereas Behr at al. evaluated the longevity of crowns that had precious-metal cores.29 Hey et al. followed-up the patients for six years,30 while Behr et al. used patients data that had a follow-up up to 14 years (median 4.3 years).29
What is the clinical evidence of the longevity of porcelain-fused-to-metal crowns compared with all-ceramic crowns?
Two systematic reviews,2,31 one randomized controlled trial,32 and one non-randomized-controlled study provided information about the comparative longevity of porcelain-fused-to-metal and all-ceramic crowns (Table 5 and Table 6).33
The systematic review by Sailer et al.2 included 67 primary studies published between 1991 and 2013 (Table 5). The majority (63 out of 67) of the included primary studies were uncontrolled studies, and four of them were randomized-controlled trials. Only one randomized controlled trial compared porcelain-fused-to-metal and all-ceramic crowns; the remaining three randomized trials compared different types of either all-ceramic crowns or metal-ceramic crowns. The included studies evaluated different types of all-ceramic crowns; these were densely sintered zirconia (9 studies); densely sintered alumina (8 studies); glass-infiltrated alumina (15 studies); leucite/lithium disilicate reinforced glass ceramics (12 studies), and feldspathic/silica-based ceramic (10 studies). The remaining studies evaluated metal-ceramic crowns (12 studies), and one study included both types of crowns. The included studies were either based on private practice (20 studies) or university-based practice (47 studies), and they included from 10 to 456 patients with a total of 14,156 single crowns and a mean follow-up of 5.8 years. The main outcome was the cumulative survival rate at 5 years.
A total of nineteen primary studies were included in the systematic review by Takeichi et al.31 six of which were also included in Sailer’s review (Table 5). The included studies were published between 1993 and 2011; four studies evaluated all-ceramic crowns, and 15 studies evaluated metal-ceramic crowns.31 The authors did not provide information about the setting from which patients were included, and they did not report the design of each included study. The authors were interested in comparing all-ceramic crowns (zirconia-based crowns) with metal-ceramic crowns. They included a total of 3621 crowns in their analyses of annual failure rate during 24 to 39 months of observation.
Ohlmann et al. conducted a randomized-controlled trial to compare the clinical performance of posterior, metal-free polymer with metal–ceramic crowns (Table 6).32 A total of 66 patients and 120 teeth were randomized to receive one of three crown types: polymer composite resin with a glass–fibre framework (40 crowns), polymer composite resin without a glass–fibre framework (40 crowns), and metal-ceramic crowns (40 crowns). Patients were recruited from a university setting, and they were followed for up to six years.
Burke et al. conducted a database study and compared the survival rates of different types of crowns (Table 6).33 Data were obtained from the National Health Service (NHS) General Dental Services (GDS) in England and Wales, and it included the records of 88,000 patients and 47,474 crown restorations installed between 1990 and 2002. The authors grouped crown types into four categories: metal-crowns (7,817), porcelain jacket or all-ceramic crowns (1,434), porcelain-fused to-metal crowns (38,166), and synthetic resin full crowns (57).
What is the long-term (eight years and longer) cost-effectiveness of porcelain-fused-to-metal crowns compared with all-ceramic crowns?
Kelly et al.34 evaluated and compared the cost-effectiveness of alternative methods for restoring large tooth substance loss in adults (Table 7). PFM crowns and all-ceramic (porcelain jacket) crowns were included in the compared methods; Class I amalgam restorations were used as a reference for the comparison between the other methods. The analysis was based on patients’ record data with all restorations performed before 1985 and followed-up for at least 10 years. The study assumed that crown removal due to endodontic or periodontal diseases was not related to crown type; therefore, the study excluded these crowns from the survival analyses. The authors considered the mean costs of restoration placement in South Australian metropolitan in 1992; the costs were obtained from Australian Dental Association fee survey in 1992.
What are the contextual considerations for all-ceramic crowns or porcelain-fused-to-metal crowns that may affect their clinical or cost-effectiveness?
The study by Burke et al. was described above, and it provided partial information about some contextual considerations of interest (Table 6).33 The authors evaluated the influence of forty clinical factors on crown survival.
Summary of Critical Appraisal
A summary of the critical appraisal of individual studies is presented in Appendix 3.
What is the clinical evidence on the longevity of all-ceramic/porcelain crowns?
The ten reviewed systematic reviews had some shared strengths and limitations (Table 8). In seven systematic reviews, for example, the literature search was conducted by several investigators who used more than one database and clear inclusion criteria.6–9,11,12,14,15 The three remaining systematic reviews used one database without any hand search or grey literature screening.9,10,13 The quality of the included studies was evaluated in two systematic reviews only.14,15 Furthermore, all the included systematic reviews were based mainly on observational uncontrolled studies, and the authors of these reviews did not evaluate or discuss the potential selective reporting of the most successful cases. Selective reporting could be evaluated through the rate of missing information and the rate of patients who were lost to follow-up. The authors of five systematic reviews had acceptable survival analyses methods; these methods accounted for the time crowns were exposed to the risk of failure.6–8,11,12 The remaining reviews either reported the numbers or crude rates of failure crowns. This kind of reporting does not provide an accurate survival estimates because it does not account for the time of failure and the time during which each crown was exposed to the risk of failure.
Rammelsberg et al.16 conducted the only randomized controlled trial to answer this question. The authors managed to follow-up all included patients, but they did not report where these patients were recruited from or the inclusion criteria. In this study, the authors used appropriate survival analyses, but it was not clear if they used a statistical power calculation to determine the sample size. Blinding was not possible in this trial which might lead to differential treatment and outcome assessment.
The external validity (generalizability) was questionable in almost all the eleven uncontrolled studies. For instance, five studies included patients from private dental practices;18,21,22,24,27 the issue with such studies is that they rely on the training and expertise of individual dentists which may not apply to other dentists. In some studies, the authors failed to report the inclusion criteria.17,19–22,27 The survival analyses of all these studies accounted for time at risk for each crown; in one study however, the authors failed to apply imputation or censoring methods for missing data.18
What is the clinical evidence on the longevity of porcelain-fused-to-metal crowns?
Hey et al.30 and Behr et al.29 applied acceptable statistical analyses to account for time at risk and survival time of the evaluated crowns (Table 9). Hey et al. however, did not apply any imputation method to account for the patients lost to follow-up; instead, the authors excluded them for the analyses. A better approach could be including these patients in the survival analysis and censoring them at the time they stopped to show up for the follow-up visits. Furthermore, Hey et al. did not report specific inclusion criteria, and therefore, the external validity of their study could not be evaluated. The external validity of Behr’s study might be limited due to the fact that the authors included patients who were treated since 1984; materials and techniques used in the earlier period of the study might not be representative to materials used more recently.
What is the clinical evidence of the longevity of porcelain-fused-to-metal crowns compared with all-ceramic crowns?
Sailer et al.2 used an acceptable literature search strategy (multiple investigators screening several databases and using clear inclusion criteria) (Table 10). Takeichi et al., in contrast, used one database, and they did not complement it with grey literature search.31 Both reviews used acceptable survival analyses methods that account for time at risk, but Takeichi et al. estimated the survival rates for each type of crowns separately without conducting any comparison between the two estimates.
Ohlmann et al.32 recruited patients from a university setting, and patients were treated by several dentists (Table 10). The benefit of such studies is that they provide a better generalizability and external validity than single dentist-based studies. The authors used a statistical power calculation to estimate the sample size. However, the authors did not report the method of randomization or how randomization was concealed. On the statistical analysis plan, the authors excluded 9/120 teeth from analysis because patients did not keep regular appointments, and was not clear how these exclusions affected the statistical power of the trial. However, it would have been more appropriate if these patients were included in the analysis and were censored at their last known status. Furthermore, authors used one of the tested interventions (polymer composite resin with glass–fibre framework) as a reference for the other interventions, and it would be more appropriate to consider metal-ceramic as the reference. The impact of this analysis was the absence of direct testing of the relative efficacy of metal-ceramic versus polymer composite resin without glass–fibre framework.
Burke et al.33 randomly selected patients’ records from a comprehensive database (National Health Service- General Dental Services in England and Wales) (Table 10). The authors used appropriate statistical analysis to estimate crown survival. However, the findings of this study might not be generalizable because the database included crowns made before 2002. Therefore, newer ceramic materials might not be available or familiar to investigators during the evaluation period used by this study.
What is the long-term (eight years and longer) cost-effectiveness of porcelain-fused-to-metal crowns compared with all-ceramic crowns?
The economic evaluation by Kelly et al. did not report a clear definition for survival (Table 11).34 Furthermore, the provision of porcelain-fused-to-metal or all-ceramic crowns was not randomized in this study; the decision to use a specific crown type may be based on patient preference, cost, criteria set by the funding agency, or other factors which may have an impact on the performance of the crown (e.g., the remaining tooth structure). The authors included restorations performed before 1985; dental materials used in these crowns fabrication were changed considerably since the installation of these crowns. This may affect the generalizability of the study results.
Summary of Findings
A summary of individual study findings is presented in Appendix 4.
What is the clinical evidence on the longevity of all-ceramic/porcelain crowns?
Short-term survival (less than five years)
Twelve studies provided short-term survival data and reported all-ceramic crown survival rates ranging from 69.8%24 and 100% (Table 12).6 The survival rates varied between each type of ceramic and from one study to another for the same type of ceramic. For example, the survival rate for lithium disilicate crowns ranged from 92%13 to 99.4%.6 Wassermann et al. reported survival rates ranging from 91.7% to 100% for In-Ceram Spinell (MgAl2O4) crowns and 92.7% to 100% for the In-Ceram Alumina (Al2O3) crowns.12 The survival rates for zirconia-based crowns ranged from 69.8%24 to 95.1%.26 One study reported a 96% survival rate of polymer crowns.16
Mid-term survival (five years to eight years)
The overall mid-term survival ranged from 87.1%10 to 98.1% (Table 12).23 One study reported a mid-term survival of 97.9% for lithium disilicate crowns.6 Kassem et al. reported a survival rate of 94.6% for In-Ceram crowns.10 Zirconia-based crowns had survival rates that ranged from 89.9%10 to 98.1%.23
Long-term survival (eight years or more)
Five studies reported long-term survival for all-ceramic crowns (Table 12). Three studies reported survival rates for lithium disilicate crowns that ranged from 87.4%17 to 100%.22 Alumina crowns had a survival rate 83.9% in one study,19 and another study reported a survival rate of 92.8% for zirconia-based crowns.
What is the clinical evidence on the longevity of porcelain-fused-to-metal crowns?
Mid-term survival (five years to eight years)
Behr et al. reported a mid-term survival rate of 96.4% and 97.5% for anterior and posterior porcelain-fused-to-metal crown,29 while Hey et al. reported a survival rate of 67.8% (Table 13).30
Long-term survival (eight years or more)
Behr et al. reported a survival rate of 92.3% and 95.9% for anterior and posterior porcelain-fused-to-metal crowns (Table 13).29
What is the clinical evidence of the longevity of porcelain-fused-to-metal crowns compared with all-ceramic crowns?
Short-term survival (less than five years)
Three studies reported short-term survival rates for both all-ceramic and porcelain-fused-to-metal crowns (Table 14).31–33 Takeichi et al. reported survival rates of 95.9% for zirconia-based crowns and 95.4% for porcelain-fused-to-metal crowns.31 Burke et al. reported survival rates of 92% for all-ceramic crowns and 93% for porcelain-fused-to-metal crowns.33 Ohlmann et al. reported the only statistical comparison between porcelain-fused-to-metal crowns and all-ceramic crowns.32 The authors reported a hazard ratio of failure of 0.74 [95% confidence interval 0.29 to 1.87] for porcelain-fused-to-metal crowns relative to polymer crowns with glass-fiber framework. The hazard ratio showed that porcelain-fused-to-metal crowns had numerically lower failure rate, but the difference did not reach statistical significance.32
Mid-term survival (five years to eight years)
Sailer et al. reported a mid-term survival rate for porcelain-fused-to-metal crowns of 96% (Table 14).31 The authors also reported the survival rates for several all-ceramic crown types; these were feldspathic/silica-based ceramic (90.7%), leucite or lithium-disilicate reinforced glass ceramic (96.6%), glass-infiltrated ceramic (94.6%), densely sintered alumina (96%), densely sintered zirconia (92%), and composite crowns (83.4%).31 Burke et al.33 reported lower survival rates for both porcelain-fused-to-metal crowns (76%) and all-ceramic crowns (68%).33
Long-term survival (eight years or more)
Burke et al. reported a long-term survival rate of 62% for porcelain-fused-to-metal crowns and 48% for all-ceramic crowns (Table 14).33
What is the long-term (eight years and longer) cost-effectiveness of porcelain-fused-to-metal crowns compared with all-ceramic crowns?
Kelly et al.34 reported that the cost-effectiveness values at 5 and 10 years of all-ceramic (porcelain jacket) crowns relative to Class I amalgam were higher than those of PFM crowns relative to Class I amalgam (Table 15). However, this relationship was reversed at the 15 year evaluation, and PFM crowns were more cost-effective than porcelain jacket crowns due to their increased failure rates beyond 15 years. Interpretation of these finding should be in light of the fact that there were a limited number of porcelain jacket crowns (18) compared to PFM crowns (212).
What are the contextual considerations for all-ceramic crowns or porcelain-fused-to-metal crowns that may affect their clinical or cost-effectiveness?
Burke et al. used Cox-regression modelling to test the statistical significance of forty clinical factors that have a potential effect on crown survival.33 The authors reported that the following twenty-two factors had a statistically significant influence on crown survival:
View in own window
| Mean annual fees for patient | Patient age group |
| Use of a core and post | Mouth quadrant |
| Pin or screw retention | Region |
| Median attendance interval for patient (days) | Dentist gender |
| Change of dentist after crown placement | Associated examination |
| Charge-paying status | Associated resin composite restoration |
| Associated periodontal treatment | Associated amalgam restoration |
| Type of crown | Dentist country of qualification |
| Tooth position | Associated bridgework |
| Associated radiographs | Associated inlay |
| Patient gender | Age of dentist |
The authors of this study did not report separate analyses for different crown materials, so it is unclear whether these factors may influence the survival of porcelain fused to metal or all-ceramic crowns differently.
Limitations
The majority of the included studies were non-randomized studies; the decision to use a specific crown type may be based on patient preference, cost, criteria set by the funding agency, or other clinical factors which may have an impact on the performance of the crown (e.g., the remaining tooth structure). Furthermore, a very limited number of direct comparative studies was identified, and comparisons across studies might not be appropriate because of differences in patient populations, dentist skills, and variations in the availability of different restorative materials. Another limitation factor was the heterogeneity in defining crown failure across studies.
The cost-effectiveness study was based on Australian prices of dental restorations in 1992; the current review did not attempt the adjustment for currency change of inflation rates since 1992. Therefore, these prices might not be representative to the Canada prices of dental restorations in 2013.
