loading  Checking for direct PDF access through Ovid


Dear Sir,I read with great interest the article entitled ‘Comparison of fracture resistance in root canals of immature sheep teeth after filling with calcium hydroxide or MTA’ by Andreasen et al. (1) published in your esteemed journal. I want to share few of my thoughts regarding this study. I appreciate the great work of the writers; however, in my opinion, this article has some technical errors:As mentioned in the discussion, the reduction in fracture strength of teeth by long-term use of calcium hydroxide has been shown in the previous study of the authors (2). Therefore, the primary purpose of the present study was to evaluate the effect of MTA on fracture strength of teeth in comparison with calcium hydroxide. According to Table 1, the teeth in MTA group have greater fracture resistance; however, the small sample size of the study makes the differences statistically insignificant. With proper sample size, MTA may show reinforcing effect on the tooth (3, 4). This study may serve as a pilot for another study with sufficient sample size.I have also a question about statistical analysis. The data of the study have been analyzed by the multiple t-tests; however, for comparing four groups, one-way analysis of variance (anova) followed by a proper post hoc test such as Tukey test is recommended assuming the normality of the data and homogeneity of variances. Why did the authors select multiple t-tests instead of anova? Comparing multiple groups by a series of independent sample t-tests is a type of ‘multiple testing’ (5). In this case, for comparing four groups of the study, a series of six independent samples t-tests is needed wherein the alpha (type I error) of 0.05 increases to 0.19 (1−(0.95)4). Therefore, the family error rate is much more than the alpha level defined in the statistical analysis (i.e., 0.05) (5). Contrary to the results of the study, anova shows that even the difference between calcium hydroxide and control group is not significant.I feel that this study can be more refined using sufficient sample size and proper statistical analysis.

    loading  Loading Related Articles