RESUMO
INTRODUCTION: A systematic review and meta-analysis were conducted to report combined and individual weighted pooled outcome rates for crown resection (CR) and root resection (RR) procedures. METHODS: Three electronic databases (PubMed [MEDLINE], Scopus, and the Cochrane Library) were searched to identify human studies in 12 languages on CR (hemisection, trisection, and premolarization) and RR (amputations and RRs without removal of crown portions). Five peer-reviewed journals, references of relevant publications, and reviews were hand searched. Assessment by 3 independent reviewers was based on the following predefined Population, Intervention, Comparison, Outcome, Study Design question: "For teeth in patients undergoing surgical therapy by CR versus RR, what is the expected probability of survival according to longitudinal studies with strictly defined outcome measurements and inclusion/exclusion criteria?" Clinical investigations with at least 12 months of follow-up were included. Studies and level of evidence were appraised using the Newcastle-Ottawa Scale and Grading of Recommendations, Assessment, Development and Evaluations. RESULTS: Thirty-four articles were obtained for final analysis. Data could be extracted from 19 studies (CR and RR OVERALL: N = 2667 [19 studies], CR: n = 111 [3 studies], and RR: n = 1127 [9 studies]). A random effects model showed weighted mean survival rates of 85.6% (95% confidence interval [CI], 76.7-91.5) for CR and RR procedures OVERALL. Individual data showed weighted mean survival rates of 81.9% (95% CI, 72.0-88.8) for CR and 87.2% (95% CI, 71.7-94.8) for RR. There was no statistically significant difference between CR and RR (P = .89, odds ratio calculation) or between maxillary and mandibular molars (P = .81, Fisher exact test). CONCLUSIONS: Overall, CR and RR procedures showed good outcome rates. Large-scale randomized controlled trials should be conducted to strengthen the evidence.
