Risk factor algorithm used to predict frequent premature ventricular contraction-induced cardiomyopathy

• Published in: International journal of cardiology Vol. 233; pp. 37 - 42
• Main Authors: Park, Kyoung-Min, Im, Sung Il, Park, Seung-Jung, Kim, June Soo, On, Young Keun
• Format: Journal Article
• Language: English
• Published: Netherlands Elsevier B.V 15.04.2017
• Subjects: Algorithms; Cardiomyopathies - diagnosis; Cardiomyopathies - etiology; Cardiomyopathies - physiopathology; Cardiomyopathy; Cardiovascular; Electrocardiogram; Electrocardiography, Ambulatory; Female; Follow-Up Studies; Humans; Incidence; Male; Middle Aged; Predictive Value of Tests; Premature ventricular contraction; Republic of Korea - epidemiology; Retrospective Studies; Risk Assessment - methods; Risk Factors; Stroke Volume; Symptom; Ventricular Function, Left; Ventricular Premature Complexes - complications; Ventricular Premature Complexes - epidemiology; Ventricular Premature Complexes - physiopathology; Premature ventricular contraction; Cardiomyopathy; Symptom; Electrocardiogram
• ISSN: 0167-5273; 1874-1754
• DOI: 10.1016/j.ijcard.2017.02.007

Premature ventricular contraction (PVC) QRS duration (QRSd) and high PVCs burden are known as a risk factor of PVC-induced cardiomyopathy (CMP). The aim of this study is to find useful algorithm to predict PVC-induced CMP. 180 patients (99 males, 51±14years) with frequent PVCs (>10%/24h), who underwent successful PVC ablation, were studied. Typical PVC-related symptoms were defined as the presence of palpitations or dropped beats during PVC. Group A (n=144) was symptomatic and Group B (n=36) was asymptomatic. The incidence of CMP was significantly higher in group B (group A=19%, group B=66%, p<0.001). In group A, there were significant differences, between the patients with normal EF and CMP, in terms of sex (p=0.005), daily PVC burden (p=0.012), distribution of PVCs with a LV site (p<0.009), and PVC QRSd (p<0.001). In group B, the PVC QRSd was significantly wider in patients with CMP. Multivariate analysis showed that PVC QRSd (p<0.001), PVC burden (p=0.022), and LV site (p=0.043) were risk factors for CMP. Using our scoring algorithm for this patient sample, we are able to predict the development of PVC-induced CMP with 80% sensitivity, 81% specificity, 64% positive predictive value, and 91% negative predictive value.