A novel diagnostic tool used to evaluate heart rate patterns controlled by the autonomic nervous system during sleep was found to be valid in assessing patients for potential physiological biomarkers in depression when sleep disturbances, such as insomnia, excessive fatigue, and sleep-related breathing disturbances were reported, according to a study published in BCM Psychiatry.
Researchers assessed the validity of an algorithm-based automated heart rate profiling tool to evaluate patterns of heart rate changes across sleep-wake states to correctly classify patients’ depression status. The algorithm was developed using a training sample of polysomnographic recordings of both patients with diagnosed depression and complaints of sleep disturbances (n=664) and mentally healthy controls (n=529).
The algorithm was verified using a testing sample of 174 distinct recordings of both patients with diagnosed depression (n=87) and healthy controls (n=87). The heart rate profiling algorithm assesses the relationship between heart rate pattern characteristics and mental state using an electrocardiogram and electroencephalogram to measure heart rate, heart rate variability, and sleep stages.
In the training sample, the depression cohort was 72.9% women and had a mean age of 45.0 years old. Additionally, 81.6% were taking psychotropic medications. The control cohort was 55% women and had a mean age of 41.3 years old. In the testing sample, the depression cohort was 59% women, had a mean age of 43.6 years old, and 79.3% were taking psychotropic medications. Meanwhile, the control cohort was 55% women and the mean age was 43.3 years old.
In the testing sample, the depression cohort had a shorter total sleep time (P =.007), longer sleep onset latency (P =.003), longer rapid eye movement onset latency (P <.001), and lower rapid eye movement sleep (P <.001) when compared with the control cohort. The heart rate profiling algorithm had an accuracy of 79.9%, sensitivity of 82.2%, and a specificity of 77%. The concordance between heart rate classification and depression diagnoses agreement across age and sex groups reached at least a moderate level in all subgroups. The sensitivity remained high even among subgroups with potential confounding variables, such as comorbid psychiatric illnesses, cardiovascular disease, and smoking status.
False negatives occurred in 15 patients with depression, and this group had significantly lower absolute nonrapid eye movement sleep 1 (P <.001) when compared with correctly classified patients. False positives occurred in 20 controls, and this group had significantly higher absolute nonrapid eye movement sleep 1 (P =.001) and absolute rapid eye movement sleep (P <.05) when compared with correctly classified controls.
Limitations of this study include data collection occurring at 5 different sites with slightly different acquisition parameters, including patients with a sleep-related breathing problem which could alter heart rate patterns independently, and the inability to assess all sleep variables which could affect sleep recordings due to retrospective nature of the study.
The researchers concluded that “[t]his tool, based on heart rate changes under the influence of autonomic regulation during sleep, was found to be highly generalizable across several potential confounding variables, as well as across differing [electrocardiogram and electroencephalogram] acquisition systems. In addition to providing an improved biological underpinning for the diagnosis of depression, this could possibly offer supplemental information to psychiatric clinical assessment, and objective measures for early screening.”
This study was partially funded by Medibio Limited. Please see the original reference for a full list of authors’ disclosures.
Saad M, Ray LB, Bujaki B, et al. Using heart rate profiles during sleep as a biomarker of depression.BCM Psychiatry. 2019;19(1):168.