AUTHOR=Brown Alexander A. , Ferguson Bradley J. , Jones Vovanti , Green Bruce E. , Pearre Justin D. , Anunoby Ifeoma A. , Beversdorf David Q. , Barohn Richard J. , Cirstea Carmen M. 

TITLE=Pilot Study of Real-World Monitoring of the Heart Rate Variability in Amyotrophic Lateral Sclerosis

JOURNAL=Frontiers in Artificial Intelligence

VOLUME=5

YEAR=2022

URL=https://www.frontiersin.org/journals/artificial-intelligence/articles/10.3389/frai.2022.910049

DOI=10.3389/frai.2022.910049

ISSN=2624-8212

ABSTRACT=<sec><title>Aims</title><p>Cardiovascular dysautonomia may impact the quality of life and survival in amyotrophic lateral sclerosis (ALS). Such dysfunction is not systematically assessed in these patients. Wearable devices could help. The feasibility of a wearable biosensor to detect heart rate variability (HRV), a physiological marker of sympathovagal balance, was studied for the first time in real-world settings in ALS.</p></sec><sec><title>Methods</title><p>Five ALS patients (two early/three late; one bulbar-onset; mildly-to-moderately disabled) and five age/sex/BMI/comorbidities-matched controls underwent assessment of 3-day HRV <italic>via</italic> VitalConnect biosensor (worn on the left thorax). De-identified data captured by the biosensor were transferred to a secure cloud server <italic>via</italic> a relay Bluetooth device. Baseline ALS severity/anxiety and physical activity during testing were documented/quantified. Time-domain HRV measures (i.e., pNN50) were analyzed.</p></sec><sec><title>Results</title><p>An overall 3-day abnormal HRV (pNN50 &lt; 3%), was found in three out of five patients (mean ± SD for the group, 2.49 ± 1.51). Similar changes were reported in controls (12.32 ± 21.14%). There were no statistically significant relationships between pNN50 values and baseline anxiety or physical activity during the tested days (<italic>p</italic> &gt; 0.05 for both groups). A negative correlation was found between pNN50 values and age in patients (<italic>p</italic> = 0.01) and controls (<italic>p</italic> = 0.09), which is similar with what is found in the general population. In line with prior studies, pNN50 values were independent of disease stage (<italic>p</italic> = 0.6) and disability (<italic>p</italic> = 0.4).</p></sec><sec><title>Conclusions</title><p>These preliminary results suggest that remote HRV measures using the VitalConnect is feasible and may constitute an improved strategy to provide insights into sympathovagal balance in ALS. Further work with larger sample sizes is warranted.</p></sec>