The purpose of this study was to evaluate the treatment outcome of initial endodontic treatment and nonsurgical retreatment performed by an endodontic specialist in his private office. A total of 2,000 teeth were examined clinically and radiographically and the results were analyzed statistically by Pearson or Fisher's Exact test and multivariate logistic regression. The multivariate analysis evaluated joint associations among various factors, using logistic regression models. The dependent variable for this analysis was the dichotomous outcome: healed versus disease. The overall endodontic success rate was 91.45%, and the healed rate was significantly higher for initial endodontic treatments than for nonsurgical retreatments; teeth without lesion than for those with lesions; teeth treated without complications than for those with complications; recall period of 18-24 months than for other periods, and teeth with final coronal restoration than for those without. Of the 1376 teeth treated in the initial endodontic treatment sample, the success rate was 94.0%. Multivariate analysis identified the presence of procedural complications (file breakage, perforation and flare-up), as well as the absence of the restorations at follow-ups as the significant predictors of outcome, showing lower rates of success. Of the 624 teeth in the nonsurgical retreatment sample, 85.9% were successful. Step-wise logistic regression analysis revealed that preoperative radiolucency was a strong statistically significant factor to determine lower rates of success than in its absence. Two additional variables (age and tooth type) were found to have a significant influence on the outcome of the retreatment sample. A higher healed rate was observed for the 50-59 years age groups than others, while multirooted (molars) teeth revealed a significantly lower percentage of success than pre-molars and anterior teeth.331112781282