IntroductionThe objective of this retrospective study was to assess the outcome of periapical surgery in a large number of molars in order to identify possible variables that might affect the outcome. MethodsThe healing outcome of patients undergoing periapical surgery of molars from October 1999 to October 2019 was retrospectively evaluated. Outcome was dichotomized into “healed” and “nonhealed” using well-established clinical and radiographic healing criteria. The potential influence of patient-, tooth-, and treatment-related parameters on the healing outcome was analyzed. ResultsA total of 424 molars in the same number of patients (45.5% male and 54.5% female) were evaluated. Three hundred seventy-two molars were classified as healed (87.7%). Three significant outcome predictors were identified: 1-year follow-up versus >1–5 years, >5–10 years, and >10 years (95.3% vs 82.2%, 76.3%, and 76.5% healed, respectively; P < .0001); root end filling material with bioceramic root repair material versus mineral trioxide aggregate (96.9% vs. 86.3% healed, respectively; P = .001); and preoperative evaluation based on cone-beam computed tomographic imaging versus 2-dimensional radiography (90.2% vs 81.4% healed, respectively; P = .02). Sex, age, tooth location, type of restoration, attachment level, presence of a post, quality of the root canal filling, technique of root end preparation, administration of antibiotics, and type of surgery had no significant impact on the healing outcome. ConclusionsThe healed rate for the concave (Retroplast) and cavity (mineral trioxide aggregate, SuperEBA [Staident International, Staines, UK], and bioceramic root repair material) root end preparation technique over all follow-up periods was 84% and 88.5%, respectively. The follow-up period, root end filling material, and preoperative evaluation based on cone-beam computed tomographic imaging had a significant influence on the healing outcome.