Methodological and Statistical Comments on T Helper Cytokines Profile and Hepatitis C

I read with much interest paper by Sofian M et al. about serum profile of T helper 1 and 2 cytokines in HCV infected patients. They have presented that “unfortunately, some immunological indicators are not normally distributed, and this can minimally affect the results”. Simply, they could use Mann-Whitney U instead of t test and Kruskal-Wallis H instead of ANOVA so they would be able to have correct results specifically when the sample size is small and also the distribution is not normal. They mentioned“31 matched (age and sex) healthy subjects from the Arak Blood Transfusion Center were enrolled.” If they were paired match, the analysis should be based on the same method (paired).