Spectral Features of Heart Rate Variability in Williams Syndrome During Sleep

Background: This study analyzed spectral alterations of heart rate variability (HRV) in Williams syndrome (WS) during sleep, taking into account the multi-fractal properties of RR-interval spectra, including effects of aging and sleep structure. Methods: Using ECG recordings of 20 subjects with WS and matched typically developing (TD) controls, fractal and oscillatory spectral components of RR-intervals were computed. The fractal component was parametrized with a piecewise-linear function, allowing a breakpoint and separate slope and intercept values in the lower- and higher-frequency domains. The dominant peak frequency and prominence were extracted from the LF (0.04-0.15 Hz) and HF (0.15-0.4 Hz) bands. Results: Strong WS/TD group differences were found in the breakpoint frequency, high domain slope, intercept and HF peak prominence. The LF peak frequency showed a slight age-dependent decrease only in TD, and reduced values in WS independent of age. Principal component analysis identified a main fractal component describing typical alterations in the spectrum in WS, which exhibited sleep-structure associations. Conclusions: The broken power-law model successfully characterized the fractal component of RR-interval spectra, capturing altered cardiac regulation in WS, while suggesting the fractal parameters as possible biomarkers of the degree of general autonomic deregulation.