Determining the optimal duration for premature ventricular contraction monitoring

• Published in: Heart rhythm Vol. 17; no. 12; pp. 2119 - 2125
• Main Authors: Hsia, Brian C., Greige, Nicolas, Patel, Shreyans K., Clark, Rachel M., Ferrick, Kevin J., Fisher, John D., Gross, Jay, Di Biase, Luigi, Krumerman, Andrew
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 01.12.2020
• Subjects: Ambulatory ECG; Arrhythmia; Cardiomyopathy; ECG; Electrocardiogram; Premature ventricular contractions; PVC; Tachyarrhythmia; VPC; Premature ventricular contractions; Arrhythmia; Cardiomyopathy; Electrocardiogram; Tachyarrhythmia; Ambulatory ECG; ECG; VPC; PVC
• ISSN: 1547-5271; 1556-3871
• DOI: 10.1016/j.hrthm.2020.07.013

Premature ventricular contractions (VPC) have hour-to-hour and day-to-day variation. High VPC burden correlates with cardiomyopathy. To determine the optimal duration for ambulatory electrocardiogram monitoring for accurate assessment of VPC burden. Our group performed a retrospective analysis on patch monitors used for any indication with overall VPC burden ≥5.0% between February 1, 2016, and February 1, 2020. We generated cumulative daily VPC averages for each day of wear and performed linear regression analysis between each cumulative daily average and overall burden. Patients were divided into groups based on low or high VPC frequency, and the analysis was repeated. Split-sample validation was used to internally validate the overall prediction model. A total of 116 patches representing 107 patients (mean age: 64.5; female: 48%) were analyzed. Mean overall VPC burden was 13.4% ± 7.5% (range: 5.0%–42.0%). Day 1 R2 was 60%, P < .001, and continued to increase to R2 88%, P < .001 at day 14. Median percent and absolute error decreased from 22.70% (interquartile range [IQR]: 9.73–34.39) and 2.58% (IQR: 1.24–4.59) at day 1 to 5.62% (IQR: 2.82–8.39) and 0.55% (IQR: 0.28–1.05) at day 14. Patients with higher overall VPC frequencies achieved a more rapid rise in R2 relative to those with lower frequencies. Split-sample validation supported the internal validity of our linear regression prediction model. Mobile telemetry for a period of ∼7 days accurately reflects overall VPC burden. Measurement of VPC burden for only 24–48 hours may not accurately reflect total burden. Monitoring for 2 weeks or longer adds little additional VPC information.