Effect of various prognostic factors on the outcome of endodontic microsurgery : propensity score matching analysis

Abstract

1. Objectives The purpose of this retrospective clinical study is to evaluate effect of various potential prognostic factors related to endodontic microsurgery on the clinical outcome with PSM analysis. Five factors are included in this study. Lesion type(isolated endodontic lesion vs. endodontic-periodontal combined lesion), surgery type(initial surgery vs. re-surgery), the existence of through-and-through lesion, Endocem as root-end filling material, and sinus perforation was evaluated as prognostic factors of the healing outcomes following endodontic microsurgery using propensity score matching (PSM) based statistical analyses. 2. Materials and Methods The clinical and radiographic records were collected from the Microscope Center of the Department of Conservative Dentistry at the Dental College of Yonsei University, Seoul, Korea. Data were from patients who underwent endodontic microsurgery in the center between March 2001 and March 2014. All patients had postoperative follow-up period of at least 1 year. The surgical outcomes of treatment group and control group were compared for each factor. To reduce the effect of selection bias in this study, the estimated propensity scores were used to match the cases. After matching, Cox proportional hazard regreesion analysis was performed for the estimation of the effect of potential prognostic factors on the outcome.

3. Results For lesion type, 83 cases of endodontic-periodontal combined lesion group were matched to 166 cases of isolated endodontic lesion group. A Cox propotional harzard regression model with the matched data identified endodontic-periodontal combined lesion as a negative effect factor of the outcome (p=.001). For surgery type, 67 cases of re-surgery group were matched to 268 cases of surgery group. . There was no significant difference in prognosis between the surgery and re-surgery group (p>0.05). For through-and-through lesion, 18 cases of treatment group were matched to 72 cases of control group. There was no significant difference in prognosis between the cases with and without through-and through lesions (p>0.05). For sinus perforation, 17 cases with perforations were matched to 68 cases with no perforations. There was no significant difference in prognosis between the cases with and without perforations (p>005). For Endocem, 28 cases of Endocem group were matched to 112 cases of MTA group. There was no significant difference in prognosis between Endocem group and MTA group (p>0.05).

4. Conclusions Lesion type was significant predictors of the outcome of endodontic microsurgery. Isolated endodontic lesion was significantly associated factors with better prognosis of endodontic micrusrgery. Surgery type, through-and-through lesion, sinus perforation and Endocem were not significantly associated factors on the outcome of the surgery. Various factors can be related to the prognosis of the endodontic microsurgery based on an analysis that utilized PSM, a useful statistical matching method for observational studies.