Data sources: Medline, Embase and reference lists of identified articles were used to find relevant publications; searches were limited to English language.
Study selection: Randomised control trials (RCT) or prospective case series of surgical endodontic treatments performed by a modern technique using magnification, root-end resection with minimal or no bevel, retrograde cavity preparation with ultrasonic tips, and retrograde root canal filling where the lesion was located at the periapical area, with a minimum of 1-year followup and outcomes assessed according to the Rud¹ or Molven² criteria, were included. Studies that did not use a modern surgical approach, involved teeth with apico-marginal defects, periodontal disease, root fractures and root perforations or undergoing repeat surgery were excluded, as were those with less than 1-year followup and with outcomes not evaluated using the Rud¹ or Molven² criteria.
Data extraction and synthesis: Studies were reviewed and data extracted independently by three observers, with disagreements resolved by discussion. Studies included were combined to estimate the pooled successes and 95% confidence intervals (CI), where success was considered to be complete healing or incomplete healing, by using generalised linear models, assuming a binomial (proportion) distribution.
Results: Eleven studies (six RCT and five prospective case series) met the criteria. The pooled proportion of treatments deemed to be a success (complete healing and incomplete healing) was found to be 91.6% (95% CI, 85.9-95.1%). The pooled proportion deemed to be failures was found to be 4.7% (95% CI, 2.24-9.50%). The remainder were categorised as of uncertain healing (3.7%).
Conclusions: Surgical endodontic treatment carried out using a modern technique carries predictable outcomes. Additional large-scale prospective clinical studies are needed to evaluate further predictors of success and failure.