Outcome of Endodontic Surgery: A Meta-analysis of the Literature—Part 3: Comparison of Endodontic Microsurgical Techniques with 2 Different Root-end Filling Materials

The aim of the present study was to investigate the influence of root-end preparation and filling material on endodontic surgery outcome. A systematic review and meta-analysis was conducted to determine the outcome of resin-based endodontic surgery (RES, the use of high-magnification preparation of a shallow and concave root-end cavity and bonded resin-based root-end filling material) versus endodontic microsurgery (EMS, the use of high-magnification ultrasonic root-end preparation and root-end filling with SuperEBA [Keystone Industries, Gibbstown, NJ], IRM [Dentsply Sirona, York, PA], mineral trioxide aggregate [MTA], or other calcium silicate cements). An exhaustive literature search was conducted to identify prognostic studies on the outcome of root-end surgery. Human studies conducted from 1966 to the end of December 2016 in 5 different languages (ie, English, French, German, Italian, and Spanish) were searched in 4 electronic databases (ie, Medline, Embase, PubMed, and Cochrane Library). Relevant review articles on the subject were scrutinized for cross-references. In addition, 5 dental and medical journals (Journal of Endodontics; International Endodontic Journal; Oral Surgery, Oral Medicine, Oral Pathology, Oral Radiology, and Endodontics; Journal of Oral and Maxillofacial Surgery; and International Journal of Oral and Maxillofacial Surgery) were hand checked dating back to 1975. All abstracts were screened by 3 independent reviewers (H.B., M.K., and F.S.). Strict inclusion-exclusion criteria were defined to identify relevant articles. Raw data were extracted from the full-text review of these selected articles independently by each of the 3 reviewers. In case of disagreement, an agreement was reached by discussion, and qualifying articles were assigned to group RES. For EMS, the same search strategy was performed for the time frame October 2009 to December 2016, whereas up to October 2009 the data were obtained from a previous systematic review with identical criteria and search strategy. Weighted pooled success rates and a relative risk assessment between RES and EMS were calculated. To make a comparison between groups, a random effects model was used. Sixty-eight articles were eligible for full-text review. Of these, per strict inclusion exclusion criteria, 14 studies qualified, 3 for RES (n = 862) and 11 for EMS (n = 915). Weighted pooled success rates for RES were 82.20% (95% confidence interval [CI], 0.7965–0.8476) and 94.42% for EMS (95% CI,