Detection of patients with methamphetamine dependence with cue-elicited heart rate variability in a virtual social environment

• Published in: Psychiatry research Vol. 270; pp. 382 - 388
• Main Authors: Wang, Yong-Guang, Shen, Zhi-Hua, Wu, Xuan-Chen
• Format: Journal Article
• Language: English
• Published: Ireland Elsevier B.V 01.12.2018
• Subjects: Adult; Amphetamine-Related Disorders - diagnosis; Amphetamine-Related Disorders - physiopathology; Amphetamine-Related Disorders - psychology; Craving - drug effects; Craving - physiology; Cue-induced craving; Cues; Electrocardiography - drug effects; Electrocardiography - methods; Heart Rate - drug effects; Heart Rate - physiology; Heart rate variability; Humans; Male; Methamphetamine; Methamphetamine - adverse effects; Middle Aged; Photic Stimulation - methods; Social Environment; Supervised machine learning; Virtual Reality; Cue-induced craving; Supervised machine learning; Heart rate variability; Methamphetamine; Virtual reality
• ISSN: 0165-1781; 1872-7123; 1872-7123
• DOI: 10.1016/j.psychres.2018.10.009

•A methamphetamine-related virtual social environment (METH-VSE) was developed.•Patients with METH dependence showed a greater HRV under METH-VSE.•Normal controls showed a smaller HRV under METH-VSE.•HRV features predicted patients with METH dependence accurately. In the present study, we developed a methamphetamine (METH)-related virtual social environment to elicit subjective craving and physiological reactivity. Sixty-one male patients who were abstinent from METH use and 45 age-matched healthy males (i.e., normal controls) were recruited. The physiological electrocardiogram (ECG) signals were recorded before (resting-state condition) and during viewing of a METH-cue video in the virtual environment (cue-induced condition). The cue-induced subjective craving was measured with a visual analogue scale (VAS) for patients with METH dependence. The results indicated that the cue-induced condition elicited significant differences in heart rate variability (HRV) between patients with METH dependence and normal controls. The changes of HRV indexes on time domain and non-linear domain from the resting-state condition to the cue-induced condition were positively correlated with the score on VAS of METH craving. Using a supervised machine learning algorithm with the features extracted from HRV changes, our results showed that the discriminant model provided a high predictive power for distinguishing patients with METH dependence from normal controls. Our findings support that immersing subjects with METH dependence in a METH-related virtual social environment can successfully induce physiological reactivity, and cue-induced physiological signal changes may have a potential implication in clinical practice.