The primary end point of this study is the mini-mental state examination in the Korean version of the consortium to establish a registry for Alzheimer's disease assessment packet (MMSE-KC) score obtained 7 days after living donor liver transplantation (LDLT). A sample size of 78 patients was calculated to detect a difference of 2.5 in the MMSE-KC score 7 days after LDLT between the temporary portocaval shunt (TPCS) and non-TPCS groups, achieving a statistical power of 80% at a two-tailed alpha error of 5% under a standard deviation of 3, an unequal allocation ratio between the groups (non-TPCS:TPCS = 83:33), which is the ratio of both types of surgery performed between May 2011 and October 2013 in our institution, and a drop-out rate of 20%, on the assumption that the Mann-Whitney U test would be performed due to the non-normal distribution of MMSE-KC scores in the TPCS group. A small number of patients was expected to be assigned to the TPCS group based on the number of previous cases in our institution and the stringent exclusion criteria.
The data are presented as the means ± standard deviations for normally distributed data, medians (1st quartile - 3rd quartile) for non-normally distributed data, and numbers of patients (percentage) for qualitative data. The assumption of normality is tested using the Kolmogorov-Smirnov and Shapiro-Wilk tests. If at least one null hypothesis of the two tests is not rejected, the data are determined to be normally distributed. Univariate comparisons between the two groups are performed using independent samples Student's t-test for normally distributed variables, the Mann-Whitney U test for non-normally distributed variables, and Fisher's exact test for categorical variables. If Student's t-test is used, the 95% confidence interval of the mean difference and the corresponding P-value is presented. Repeated measures analysis of variance with post hoc paired Student's t-test using Bonferroni correction is performed for normally distributed variables to determine the significance of between- and within-groups effects and interaction between the variables (group versus phase). To determine the significant interval change in non-normally distributed variables during surgery, Friedman's test with post hoc Wilcoxon signed-rank test under Bonferroni correction is used. For the comparison of within-group changes in the total sum scores and subscores of the MMSE-KC, Wilcoxon signed rank test is used. Rank analysis of covariance is used to compare the MMSE-KC score obtained 7 days after surgery between the groups by controlling for the covariates that are found to correlate with the score based on nonparametric Spearman's correlation analysis. The statistical analysis is performed using IBM SPSS Statistics software, version 19.0.0 (IBM Corp., Armonk, NY, United States). A two-tailed P \< 0.05 is considered statistically significant.
