Summary: | 碩士 === 臺灣大學 === 臨床牙醫學研究所 === 98 === The differential diagnosis of root surface split was difficult even after an exhaustive history taking and comprehensive clinical and radiographic examinations. However, research which had empirically documented in etiology and clinical characteristics of root surface split was scant. Therefore, the objective of the NTU Root Surface Split Study project attempted to assess clinical characteristics, predisposing factors and treatment outcome of the tooth with root surface split. All of the 71 root surface split teeth, 10 were diagnosed by direct inspection on extracted tooth; 61 were diagnosed by histopathological examination. During which descriptive statistic, univariate analysis, bivariable analysis and logistic analysis were performed. Of the 71 teeth studied, 38 (53.5%) were extracted and were excluded. The remaining 33 teeth (46.5%) were examined for treatment outcome of healed, questioned or failed. Evidently, the results from univariate analysis revealed that thirteen variables of clinical characteristics were associated with the root surface split and ten variables of predisposing factors, which thus confirmed the variables, including swelling, periodontal bony destruction, periapical bony destruction, apicocoronal location, fragment shape, and separation mode, as the main clinical characteristics on the tooth with root surface split; as well as the variables, including gender, age, tooth type, and attrition, as the predominant predisposing factors on the tooth with root surface split. The results from bivariate analysis of the 71 study samples revealed statistically significant differences in gender, associated with separated fragment width, and treatment technique; in age, associated with sinus tract, attrition, and apicocoronal location; in tooth type, associated with separated fragment length, apicocoronal location, and initial diagnosis accuracy; in radiopaque image, associated with initial diagnosis accuracy; in apicocoronal location, associated with separated fragment shape. Additive bivariable analysis on the 33 examined subsamples of teeth revealed statistically significant differences in treatment outcome, associated with separated fragment length and apicocoronal location as significance level interpreted at the 0.05; besides, with treatment technique as significance level interpreted at the 0.1. Logistic regression analysis identified significant outcome predictors: (1) apicocoronal location of the separated fragment on apical 1/3 (odds ratio [OR], 0.083; confidence interval [CI], -5.227─0.257) and on middle 1/3 (OR, 1.333; CI, -1.752 – 2.328), and (2) treatment technique (OR, 0.2; CI, -3.41 – 0.192). Treatment outcome was better in tooth which separated root fragment was located on middle 1/3 than on apical 1/3. Although, treatment technique was suggested as an outcome predictor in tooth with root surface split, it should be required confirmation from randomized controlled trials. Consequently, for obtaining enormous amount of empirical evidence to allow assessment with better power, continuation of the project in cumulating data would be necessary.
|