Applying spectral analysis to the arterial pulse to discriminate cardiovascular side effects following administration of Moderna's mRNA-1273 vaccine

Abstract

Vaccines against severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) have demonstrated strong efficacy in preventing symptomatic disease, but adverse cardiovascular side effects have been reported. This study investigated whether noninvasive arterial pulse waveform analysis could help identify cardiovascular responses following administration of Moderna's mRNA-1273 vaccine. Radial blood pressure waveforms were recorded for 1 min in 203 individuals who received the vaccine at a medical center. Participants were categorized into four groups: no side effects (n = 18), cardiac side effects (n = 45), vascular side effects (n = 68), and combined cardiac or vascular side effects (n = 72). Spectral analysis was performed on 40 harmonic indices derived from the pulse waveform, including amplitude proportions, phase angles, and variability indices. These features were analyzed using both a self-developed pulse distribution method and a standard machine learning (ML) algorithm. Significant changes in spectral indices were observed post-vaccination. The pulse distribution method achieved an area under the curve (AUC) of 0.76 in detecting cardiovascular side effects, outperforming the ML approach. The performance was not significantly influenced by potential confounders. These findings suggest that vaccine-induced vascular stiffness may cause a mismatch in elastic properties of the vasculature, and that spectral analysis of the arterial pulse may offer a noninvasive approach to assess such changes.