Evaluating Heart-Rate Variability Metrics As a Measure of Stress During Unsedated Medical Procedures

Abstract

Abstract Heart rate variability (HRV) has been proposed as a measure of psychophysiological stress, however validations often focus on complex stressors that may cause non-stress arousal. We investigate the efficacy of a range of HRV metrics as an objective measure of stress during an unsedated flexible sigmoidoscopy procedure on 47 patients. HRV metrics were calculated at several epochs across the procedure. Acceleration and deceleration capacity, the standard deviation of normal-normal intervals, high-frequency power, and the α1 parameter of detrended fluctuation analysis all showed a significant difference between stressed and unstressed states, to p < 0.0167. Low-frequency power and the low:high frequency ratio both exhibited trends in an unexpected direction, casting doubt on their suitability. The root-mean-square of successive differences and sample entropy failed to differentiate between stressed and unstressed states. Measured effect sizes and standard deviations, combined with population norm values, were used to calculate sample sizes for future studies. Acceleration and deceleration capacities, the standard deviation of normal-normal intervals, and α1 provided the highest power with low sample sizes, allowing well-powered tests a n < 100 for a reduction in stress of 75% or greater. Other metrics which successfully identified stress had large variances that ruled out well-powered trials in small samples.