The representative chromatograms of ATV, o- ATV, p- ATV, d5-ATV, and RSV were shown in Figure 1. The retention times of ATV,o- ATV, p- ATV, d5-ATV, and RSV were 3.06 min, 2.78 min, 1.81 min, 3.05 min, and 1.83 min, respectively. In addition, the standard curves of ATV, o- ATV, and p- ATV all showed good linearity over the concentration range of 0.082–50 ng/mL. The regression equation was listed in Table1. Meanwhile, the correlation coefficient (r²) was all above 0.99. Moreover, the lower limit of quantification (LLOQ) of ATV, o- ATV, andp- ATV was 0.082 ng/mL.

The results of intra- and inter-batch precision and the accuracy of ATV,o- ATV, and p- ATV were listed in Table 2. The precision and accuracy variation were all less than 15%, which indicated that the precision and accuracy were satisfactory.

Herein, matrix effect (MF) values (%) of ATV at high-quality control (HQC) and low-quality control (LQC) concentrations were 106.6 ± 6.5 and 106.5 ± 6.1, respectively; o- ATV at HQC and LQC concentrations were 95.9 ± 11.3, 105.3 ± 7.4, respectively; and p- ATV at HQC and LQC concentrations were 89.8 ± 8.9, 90.1 ± 6.8, respectively. The MF variations were all less than 15%, which suggested that there were few matrix effects. The HQC, middle-quality control (MQC), and LQC extraction recoveries of the ATV were 91.7%, 100.8%, and 93.20%, respectively; o- ATV were 71.2%, 70.01%, and 69.2%; andp- ATV were 87.1%, 77.48%, and 75.1%. MF values and extraction recovery values were listed in Tables 3 and 4.

The results of the analytes’ stability were displayed in Table 5. ATV,o- ATV, and p- ATV were all stable and had no significant degradation observed in rat plasma under different conditions (4 h at room temperature, 24 h in autosampler, 24 h in 4 ℃ refrigerator, and 36 days in a -20 ℃ refrigerator).