An Experimental Study of Step Test Index Combined With Heart Rate Variability in Estimating Maximum Oxygen Uptake in Women With Drug Use Disorder.

Background: Maximal oxygen uptake (VO2max), a vital physiological indicator, has been widely used in many fields. In recent years, the measurement method of VO2max has been widely explored in various populations, but few studies have been conducted for women drug abusers. For the importance of VO2max in the formulation of aerobic exercise intensity for drug users, the present study estimated VO2max using the step test index combined with heart rate variability in women with drug use disorder. Methods: Forty women methamphetamine (MA) users without cardiovascular disease and dyskinesia participated in a cardiopulmonary exercise test (CPX) and a 3-min step test. Each of them performed a heart rate variability (HRV) monitoring test after the step test, and VO2max was estimated by step test index and HRV. Results: 1) The step test index had a significant positive correlation with VO2max. The Standard deviation of normal-to-normal interval (SDNN) had a significant positive correlation with VO2max and a significant positive correlation with the step test index; 2) The R-square values of estimated VO2max by step test index and post-SDNN for overall MA users were 0.29 and 0.22, with an accuracy of 93.19% and 92.85%, respectively; 3) The R-square values of the estimated VO2max by step test index and post-SDNN in group I were 0.27 and 0.36, respectively, with an accuracy of 94.04% and 93.99%. The R-square value of the estimated VO2max by step test index in group Ⅱ was 0.44, with an accuracy of 92.65%. However, post-SDNN cannot adequately estimate the VO2max in group Ⅱ; and 4) There was no significant difference in VO2max obtained by CPX, step test index, or post-SDNN regardless of overall or grouping variable analysis. Conclusion: 3-min step test combined with HRV can estimate the VO2max of women MA users to a certain extent, but the size and coverage of the sample size should be further considered. In the future, more methods, such as machine learning or artificial neural networks, should be used.