Relationship between ultra-short heart rate variability and short-term mortality in hospitalized COVID-19 patients

Objective: To assess the association between ultra -short heart rate variability (US-HRV) and short-term mortality in patients with COVID-19 and develop prognostic prediction models to identify high-risk patients as early as possible. Methods: A retrospective cohort study was performed on 488 patients diagnosed with COVID-19 and hospitalized in the First Affiliated Hospital of Fujian Medical University from December 2022 to January 2023. 10-s electrocardiogram (ECG) data were available for these patients. The US-HRV parameters including standard deviation of all normal -to-normal R -R intervals (SDNN) and root mean square of successive differences between normal -to-normal R -R intervals (rMSSD) were calculated using Nalong ECG software. The endpoint was shortterm mortality, including in-hospital mortality or mortality within 1 week after discharge. Results: Of the 488 patients, 76 (15.6%) died. The SDNN and rMSSD in the death group were significantly lower than those in the survival group ( P < 0.001). The area under the receiver operating characteristic (ROC) curve (AUC) for SDNN and rMSSD to predict mortality was 0.761 and 0.715, respectively. The combined use of SDNN and rMSSD had an AUC of 0.774. The mortality rate in the group with SDNN <= 7.5 ms was higher than that of SDNN >7.5 ms group ( P < 0.05). With the decrease of SDNN, the mortality of patients showed an upward trend, and the mortality of patients with SDNN <= 2 ms was the highest (66.7%). Multivariate logistic regression analysis identified SDNN as an independent predictor of prognosis (odds ratio (OR) = 5.791, 95% confidential interval (CI) 1.615 - 20.765, P = 0.007). The AUC of Model 1 (simple model) was 0.866 (95% CI 0.826 - 0.905). The AUC of Model 2 (comprehensive model) was 0.914 (95% CI 0.881 - 0.947). Conclusion: SDNN was associated with short-term mortality and provided the additional discriminatory power of the risk stratification model for hospitalized COVID-19 patients.