Statistical analysis
Statistical analysis was performed by SAS 9.4. All data were tested by two-side test and the probability value (P) less than 0.05 was considered statistically significant.AUC and Cmax were logarithmically transformed and analyzed by linear mixed effect model. Sequence, period and formulation were fixed effects, and subject within sequence was included as a random effect. Analysis of variance (ANOVA) of cross-over design was performed on the log-transformed variables. The geometric mean ratios (GMRs) of the primary pharmacokinetic parameters and their 90% confidence intervals (CIs) were calculated, and the test formulation was judged as bioequivalence if it fell within the equivalent range (80-125%). Bioequivalence was assessed separately in both the fasting and fed groups.