Differential autonomic contributions to persistent hypotension and bradycardia following carotid artery stenting
Le Cao, Hang Wang, William Robert Kwapong, Yuying Yan, Ruishan Liu, Xuening Zhang, Jincheng Wan, Tianxiang Lan, Fayun Hu, Bo WuBackground
Hemodynamic depression is common after carotid artery stenting (CAS), but whether its two phenotypes, persistent hypotension and bradycardia, reflect shared or distinct autonomic responses remains unclear. We aimed to characterize perioperative autonomic changes associated with these outcomes via continuous Holter monitoring.
Methods
In this prospective, single-center cohort, consecutive CAS patients were enrolled, and a prospective subcohort underwent Holter monitoring from 24 hours before to 24 hours after stenting. Sympathetic activity index (SAI) and parasympathetic activity index (PAI) were derived from heartbeat series to reflect autonomic modulation. Persistent hypotension was defined as systolic blood pressure<90 mmHg or vasopressor requirement for ≥6 hours, and persistent bradycardia as heart rate<60 beats/min for ≥6 hours within 24 hours after CAS. Exploratory mediation-style analysis examined the extent to which autonomic changes accounted for each outcome.
Results
Among 735 enrolled patients, 301 had valid recordings for autonomic analysis. In this monitored subcohort, CAS was associated with a reciprocal autonomic shift, with decreased SAI and increased PAI (both p<0.001). Reduced SAI statistically explained a larger proportion of the modeled association with bradycardia (88.9%; residual association p=0.570), whereas autonomic indices explained only a modest proportion of the modeled association with hypotension (SAI 20.0%; PAI 16.1%; both mediation p<0.001), with a significant residual association remaining (p<0.001).
Conclusions
CAS was associated with sympathetic withdrawal and parasympathetic activation. These exploratory findings suggest that bradycardia was more tightly associated with heart rate-derived autonomic changes than hypotension, which was less fully captured by SAI/PAI and may involve additional factors.