Composite Autonomic Symptom Score-31 for the diagnosis of cardiovascular autonomic dysfunction in long-term coronavirus disease 2019

Abstract:

BACKGROUND:
Composite Autonomic Symptom Score-31 (COMPASS-31) is an easy-to-use screening tool for the evaluation of autonomic dysfunction in various diseases affecting neural function but has rarely been used in the assessment of long coronavirus disease 2019 (COVID-19). This study aimed to evaluate the diagnostic accuracy of the COMPASS-31 score in detecting dysfunction of the autonomic nervous system in patients 3 months after COVID-19 infection.


MATERIALS AND METHODS:
Fifty-nine subjects were recruited and grouped into 2: (a) controls (n = 31) who had never had positive polymerase chain reaction results for COVID-19 before and (b) the post-COVID-19 patients (n = 28) who had confirmed COVID-19 infection 3–6 months before recruitment. COMPASS-31 questionnaire was utilized to evaluate subjective symptoms or evidence of autonomic dysfunction. Autonomic dysfunction was assessed objectively by cardiovascular autonomic reflex tests (CARTs) and heart rate variability (HRV). For comparison of quantitative variables between two groups, t-test or Mann-Whitney U test, as appropriate, were used. Sensitivity, specificity, negative predictive value (NPV), positive predictive value (PPV), negative likelihood ratio (LR), and positive LR were used as measures of diagnostic accuracy. Receiver operating characteristic (ROC) curve analysis determined the overall accuracy of COMPASS-31.


RESULTS:
The median COMPASS score was found to be significantly higher in post-COVID-19 participants than controls (15.5 vs. 10, P = 0.021). The median total CART score was also significantly higher in post-COVID-19 participants (0 vs. 1, P < 0.001). Out of 6 domains of the COMPASS score, the median value for orthostatic dysfunction was found to be significantly higher in post-COVID-19 participants than controls (12 vs. 0, P = 0.008). There was significantly fair accuracy of the COMPASS score with an area under the receiver operating curve 0.68 (0.54–0.82) following the total CART score ≥2 as the gold standard in the diagnosis of autonomic dysfunction (P = 0.021). The best cutoff point of the total COMPASS score was 12.5, where the optimal values of sensitivity, specificity, and positive and negative predictive values were achieved. Nonsignificant and weak correlations between CARTs, HRV parameters, and COMPASS score were found.


CONCLUSION:
COMPASS-31 could be used as a user-friendly screening tool to detect autonomic dysfunction in post-COVID-19 cases with acceptable sensitivity and specificity.