Prediction of vasovagal syncope from heart rate and blood pressure trend and variability: Experience in 1,155 patients

Abstract

Vasovagal syncope (VVS) is a complex fainting disorder commonly triggered by orthostatic stress. We developed an algorithm for VVS prediction based on the joint assessment of RR interval (RR) and systolic blood pressure (SBP). Simultaneous analysis of RR and SBP trends during head-up tilt as well as their variability represented by low-frequency power (LFRR and LFSBP) generated a cumulative risk that was compared with a predetermined VVS risk threshold. When cumulative risk exceeded the threshold, an alert was generated. Prediction time was the duration between the first alert and syncope. In the first 180 sec of head-up tilt, baseline values were established, following which VVS prediction was possible. An analysis was performed using 1,155 patients who had undergone head-up tilt for syncope: 759 tilt-positive and 396 tilt-negative patients. In the tilt-test protocol, at syncope or after 35 min, the patient was returned to supine. In tilt-positive patients, VVS was predicted in 719 of 759 patients (sensitivity 95%), whereas 29 false alarms were generated in 396 tilt-negative patients (specificity 93%). Prediction times varied from 0 to 30 min but were longer than 1 min in 49% of patients. Predicting impending syncope requires use of simultaneous blood pressure and heart rate, which may shorten diagnostic testing time, free patients from experiencing syncope during a diagnostic tilt-test, and have application in risk-guided tilt training and in an implanted device-to-trigger pacing intervention. The prospects for relieving patient discomfort are encouraging.