Details of study characteristics, critical appraisal, and study findings are located in Appendix 2, Appendix 3, and Appendix 4.
Quantity of Research Available
A total of 527 citations were identified in the literature search. Following screening of titles and abstracts, 480 citations were excluded and 47 potentially relevant reports from the electronic search were retrieved for full-text review. Thirty potentially relevant publications were retrieved from the grey literature search. Of these potentially relevant articles, 70 publications were excluded for various reasons, while seven publications met the inclusion criteria and were included in this report. Appendix 1 describes the PRISMA flowchart of the study selection.
Additional references of potential interest are provided in Appendix 5.
Summary of Study Characteristics
Detailed study characteristics and findings are presented in Table 1, Appendix 2.
The seven included studies addressed the effectiveness of crowns on root canal treated teeth. The literature search did not provide any evidence-based guidelines or cost-effectiveness studies.
What is the clinical-effectiveness and prognosis of placing crowns on endodontically treated teeth?
Three systematic reviews and four non-randomized studies assessed the survival of root canal treated teeth restored with crowns or direct restorations.19–24
Fedorowicz et al. published a Cochrane systematic review to evaluate the effects of restoration of root canal treated teeth by crowns compared or conventional filling materials.19 The authors searched for randomized and quasi-randomized controlled trials published prior to February 2012; one randomized controlled trial (RCT) conducted by Mannocci et al. in 2002 was included in this systematic review.19 The included study randomized 117 patients (117 premolar teeth) in two groups; the first received metal-ceramic crowns after the endodontic treatment, and the other group received composite fillings. Both groups received finer posts and composite core build up. The included patients had class II carious lesions on the premolar teeth with no cuspal involvement, and they had to have no more than moderate (<40%) periodontal attachment loss. The primary outcome was survival at three years after treatment; the authors defined treatment failure as root fracture, post fracture, post decementation, evidence of material gap between tooth and restoration, or the presence of secondary caries at the margins of the restoration.19
In their systematic review, Ng at al. investigated the effect of study characteristics on reported tooth survival and the effect of clinical factors on the proportion of root filled teeth surviving after root canal treatment.20 The authors included 14 studies in the review, but only four retrospective studies compared tooth survival when restored with crowns versus direct fillings.20 The four studies were published by Lazarski (2001), Aquilino (2002), Alley (2004), and Lynch (2004), and their sample size ranged from 176 to 44,613 teeth.20 Ng’s review grouped the results of these studies in two groups; the first group included patients treated with crowns after root canal treatment, and the second had patients treated with direct fillings. The authors did not report the type of crowns and fillings used in these studies.20 The authors reported the survival probability for both groups after one to ten years of follow-up.
The third systematic review by Stavropoulou et al. tested the hypothesis that the placement of a crown after root canal treatment is associated with improved (long term) survival.21 The authors included ten clinical studies, but they did not report the design and characteristics for each study. Two of these studies25,26 were also included in the previous systematic review. Teeth were grouped according to the final restoration with full coverage crowns or direct restorations; however, the exact types of crowns and materials were not reported.21 The authors estimated the survival rate for each group after 1, 2, 3, 5, and 10 years of root canal treatment.21 Failure was defined as fracture of the restoration, post fracture, post decementation, dislodgment of the restoration, marginal leakage of the restoration and tooth loss.
The four non-randomized studies treated to various extents the same question: what are the factors that affect the survival of root canal treated teeth.22–24,27 To answer this question, the authors of these studies used retrospective cohorts that ranged from 196 patients22 to 1,126,288 patients23. The follow up duration ranged from five years22 up to 20 years24. The authors of these studies evaluated several clinical factors affecting survival of root canal treated teeth including the type of restoration. All studies reported the percentage of failures for teeth restored by crowns or direct restorations. However, failure varied in each study; Tickle et al. defined failures by tooth extraction, replacement of the root filling, or periradicular surgery performed on the tooth;22 Cheung et al. defined failure as tooth extraction, retreatments, or presence of periapical radiolucency,24 and Salehrabi et al. considered tooth retention in the oral cavity as evidence of treatment success and the occurrence of untoward events (not defined) as treatment failure.24 Piovesan et al. did not provide a specific definition for failure.27
Summary of Critical Appraisal
Details of the critical appraisal of individual studies are presented in Appendix 3.
The systematic review by Fedorowicz et al.19 followed the rigorous Cochrane methodology for literature search, data extraction, analysis and reporting. Authors of Cochrane reviews search multiple databases; screen, select and extract data in duplicate; and follow a priori defined statistical analyses. However, the generalizability of the review findings might be limited because it included one randomized controlled trial of 117 patients. Furthermore, the trial was performed by one dentist who included only premolar teeth that had class II carious lesions without cuspal involvement.19 These factors limit the generalizability of findings to anterior and posterior teeth and even to posterior teeth with more severe destruction. Finally, the follow up period was limited to three years which limits the extrapolation of long term survival.
Ng et al.20 reported that they followed a priori defined protocol, performed a systematic literature search, and evaluated the quality of the included studies. Despite the acceptable literature search methodology, the authors did not use an appropriate analysis to the extracted data. In the review, the authors reported odd ratio estimates for tooth failure based on various prognostic factors; however, they did not account for the different follow-up periods in the included studies that ranged from 1 to 11 years. This type of analysis masks the effect of time on survival estimates.20
The third systematic review by Stavropoulou et al.21 had a priori protocol, and the authors followed standard methodology in literature search, data extraction and analysis with one exception; that the authors search one database (MEDLINE) and did not include other databases of grey literature search. The authors did not report important information such as design of the included studies, type of restoration (exact type of crowns and fillings), restored teeth, and degree of tooth destruction. Therefore, the generalizability of the findings is uncertain.21
Tickle et al. clearly reported their study objectives and the variables that they wanted to evaluate for their effects on tooth survival.22 The authors used an appropriate statistical analysis for survival data. The generalizability this study might be limited because the authors collected data relative to mandibular first molar teeth only, and this might limit the extrapolation of findings to other teeth. Furthermore, the authors did not report how the participating dentists were selected, and whether they (and their patients) were representative to NHS funded practices.22
Piovesan et al. reported their objectives and the evaluated interventions, but the authors did not report on patient characteristics or the degree of tooth destruction.27 Furthermore, the authors did not provide explanation on how the final restorations were selected. In the statistical analysis plan, the authors reported survival for each group, but they did not conduct any comparative analysis between the treatment groups.27
Salehrabi et al. retrospectively analyzed a databased of more than 1.4 million patients from a health insurance provider with eight years of follow up.23 Despite the large number of patients, it is not clear if one insurance company database would be representative to non-insured patients or to patients insured with other companies. The authors did not report on patient characteristics or degree of tooth loss. Furthermore, the reporting of results was not appropriate; the authors reported the distribution of extracted teeth based on the type of restoration, but they did not account for the total number of treated teeth for each type of restoration. Therefore, these results could lead to inaccurate conclusions, and the true distribution of failure (extracted teeth) could not be known.23
Cheung et al. conducted a retrospective analysis for which they clearly reported the objectives and methodology.24 The authors reported some key characteristics for patients and teeth of interest, and they included them in the analysis. The statistical analysis was appropriate for the type of data collected by authors.24 The main limitation in this study was that the period of follow up ranged from 1981 and 1989; techniques and materials used during this period might not be generalizable to the modern dental practice.
The aforementioned four non-randomized trials used retrospective data from various databases.22–24,27 The quality of retrospective studies is inherently dependent on the quality of database itself, and this could not be evaluated through the published articles.
Summary of Findings
Detailed study characteristics and findings are presented in Appendix 4.
What is the clinical-effectiveness and prognosis of placing crowns on endodontically treated teeth?
Fedorowicz et al. reported that in the one included study, after three-year follow up, there were 1/53 (1.9%) and 3/54 (5.6%) restoration failures in root canal treated premolars restored with crowns and composite fillings respectively.19 The difference between the two groups was not statistically significant with a relative risk (95% confidence interval [CI]) of 0.34 (0.04 to 3.16). The authors concluded that there was insufficient evidence supporting or refuting the use of crowns to restore root canal treated premolars.19
Ng et al. pooled the survival probabilities that were reported in four studies, and they reported that root canal treated teeth restored with crowns were associated statistically significantly higher odds of survival; the odds ratio (95% CI) was 3.9 (3.5 to 4.3).20 The authors concluded that crown restoration after root canal treatment was a significant prognostic factor. However, this analysis did not account for the different follow-up duration in the included studies that ranged from one to eleven years.20
Stavropoulou et al. pooled the cumulative survival root canal treated teeth from ten studies. At year one, the cumulative survivals in the crown and direct restoration groups were 99% and 95% respectively.21 The difference between the two groups continued to increase up to the fifth year; at which, the cumulative survival rates were 94% and 63% respectively. Between the fifth and tenth year post-treatment, the cumulative survival of the crown group dropped from 94% to 81%, while the direct restorations group had a stable cumulative survival of 63%. The statistical differences between groups were not tested. The authors concluded that root canal treated teeth restored with crowns had an acceptable 10-year survival, while direct restorations had satisfactory survival for the short term only.21
Tickle et al. reported the failure rate at five years post treatment in 196 patients.22 The failure rate per 100 years with a root filled tooth was null for the teeth restored with crowns (specific crown types were not reported), and it was 5.0 for teeth restored with plastic fillings. The statistical difference between the two groups was not reported. The authors concluded that restoration with crown might provide a lower risk of tooth fracture and better coronal seal, but they suggested that this conclusion needed further investigation to be confirmed.22
Piovesan et al. reported the number of failed root canal treated teeth in 69 patients followed up for up to 97 months.27 The authors reported 5/68 (7.4%) and 1/42 (2.1%) tooth failures in the crown and composite restoration groups respectively. On anterior teeth, 2/11 (18.2%) metal-ceramic and 1/9 (11.1%) all-ceramic crowns failed; on the posterior teeth 2/30 (6.7%) all-ceramic crowns failed. The authors concluded that the type of final restoration (resin composite, all-ceramic, and ceramic-fused-to-metal) did not affect the tooth survival, but they admitted that longer-term results are uncertain.27
Salehrabi et al. reported the number of extracted teeth during the eight years following root canal therapy and restoration with crowns or direct restorations.23 The authors reported that 1344 and 6565 anterior teeth were extracted in the two groups respectively; for premolars and molar teeth, they reported 1560 versus 8879 tooth extractions and 3373 versus 20253 tooth extractions for crowns versus direct restorations respectively.23 The authors did not report the total number of treated teeth in each group, and therefore, the comparison between groups was not possible from the reported results.
Cheung et al. reported the mean survival time 10 to 20 years after root canal therapy restored with crowns or direct restorations.24 It was reported that the mean survival for direct restorations was 95 months, while the mean survival for crowns ranged from 130 months to 150 months for the full gold and ceramic-metal crown respectively.24 The differences between the crown groups and direct filling groups were statistically significant.24 The authors concluded that the provision of a crown restoration after treatment was associated with a better tooth survival. They also concluded that anterior teeth and mandibular premolars had higher survival than other teeth.24
Limitations
The included studies did not consider a high caries risk populations, and the risk of caries was not evaluated explicitly as an outcome in the include trials. Information about high caries risk populations and caries outcomes are crucial in the perspective of health insurance plans. The results obtained from studies conducted on the general public may not be generalizable to the patients of special interest for public health plans, especially patients who have a higher risk of caries and patients who live in remote areas where regular care provision is not feasible.
The included studies did not evaluate toot survival based on the amount of its destruction which may play a major role in tooth survival. Furthermore, the current review was interested in the cost-effectiveness and evidence-based practice guidelines. However, the searched literature search did not identify any relevant cost-effectiveness studies or evidence-based guidelines.
