Statistical analysis
The normal distribution of data was analyzed by the D’Agostino and Pearson omnibus normality test. To remove outliers from normally distributed healthy control individuals, the Rout test was performed. The selection of optimal cut-off values was based on curves constructed by plotting the calculated specificity and sensitivity vs. the corresponding cut-off values. Statistical significance was evaluated using either parametric tests: paired-samples Student t-test and unpaired-samples Student t-test with Welch correction, or non-parametric tests: Wilcoxon matched-pairs signed-rank test or Mann-Whitney U-test for unpaired data, when applicable. Calculations were performed using GraphPad Prism version 6.01 for Windows (GraphPad Software, San Diego California, USA, www.graphpad.com). A p-value < 0.05 was considered statistically significant.