Skin sympathetic nerve activity complexity predicts sudden and nonsudden cardiac mortality in heart failure

Heart failure (HF) is a global health challenge with a significant impact on patients and health care systems. Identifying those at risk of cardiac death (CD) remains difficult but essential. Elevated sympathetic nerve activity (SNA) is linked to ventricular arrhythmias, highlighting its value in diagnosis and risk stratification.

This study aimed to investigate the use of skin sympathetic nerve activity (SKNA) as a marker for stratifying patients at risk of CD beyond established clinical predictors and heart rate variability (HRV).

We analyzed 20-minute, 1-kHz orthogonal electrocardiogram recordings from 588 patients with HF in the MUerte Subita en Insuficiencia Cardiaca dataset after excluding device carriers. SKNA was extracted using high-pass filtering (>300 Hz). We quantified traditional amplitude-based SKNA metrics derived from moving average and root mean square signals and raw SKNA (rSKNA) signal distribution and nonlinear complexity features, including the largest Lyapunov exponent (LE) and multiscale entropy (MSE). Associations with outcomes were assessed using univariate analyses and multivariable Cox models adjusted for age, body mass index, left ventricular ejection fraction, New York Heart Association class, diabetes, and previous myocardial infarction. HRV indices (standard deviation of normal-to-normal interval, low-frequency power to high-frequency power ratio) were evaluated for comparison.

Moving average-/root mean square-based SKNA metrics did not show consistent between-group differences. In contrast, rSKNA complexity measures (LE and MSE) were lower in patients who experienced CD (sudden and nonsudden). In adjusted Cox models, LE and MSE remained independently associated with CD (hazard ratio ≈0.75-0.80; P < .05), whereas the standard deviation of normal-to-normal interval and the low-frequency power to high-frequency power ratio were not independently associated.

rSKNA complexity measures, particularly MSE, provide complementary, noninvasive prognostic information for sudden and nonsudden CD in HF beyond established clinical predictors and HRV.