Abstract 13375: Quantitative Analysis of Narrow QRS Peaks Predicts Ventricular Arrhythmic Events in Patients With Hypertrophic Cardiomyopathy
Introduction: Localized regions of delayed conduction can promote re-entrant ventricular arrhythmias (VA). We hypothesized that narrow peaks in the QRS complex (QRSp), which may represent regions of conduction delay, would be associated with sustained VA in patients with hypertrophic cardiomyopathy (HCM).
Methods: We prospectively enrolled 100 patients (51±15 yrs, 63 male) with HCM who had received primary (n=96) or secondary prevention ICDs (n=4). Major HCM clinical risk factors (RF) were assessed pre ICD implantation. High resolution (1kHz sampling) digital 12 lead ECGs were recorded continuously for 3 minutes during intrinsic rhythm. Using custom software, QRSp for each precordial lead (Vnp) was defined by automatically detecting the number of positive peaks of ≤25ms duration within the QRS complex. VA was defined as cardiac arrest or appropriate ICD therapy in followup.
Results: Over a median followup of 41 (20-97) months after ICD implantation, VA were observed in 21% of patients. Comparing +VA vs -VA patients, there was no difference in the prevalence of individual RF (nonsustained VT: 58 vs 54%, p=0.8; family history of sudden cardiac death: 32 vs 50%, p=0.2; syncope: 50 vs 41%, p=0.46; septum ≥30mm: 29 vs 18%, p=0.36), the prevalence of ≥1 RF (86 vs. 96%, p=0.11), the prevalence of ≥2 RF (57 vs 60%, p=1.00), or the median number of RF (2 [1-2] vs 2 [1-2], p=0.96). Intrinsic QRS duration (122±48 vs. 112±32 ms, p=0.37) was similar between +VA and -VA. In contrast, QRSp in leads V4, V5 and V6 were greater in +VA vs -VA (V4p: 2 [0.5-3] vs 1 [0-2], p=0.042; V5p: 2 [1-3.5] vs 1 [0-2], p=0.009; V6p: 2 [1-3] vs 1 [0-2], p=0.022). V5p was the only independent predictor of VA in a multivariable model that included the presence of ≥1 RF (odds ratio 1.4, 95% CI 1.1-1.9, p=0.019). ROC curve analysis for the number of RF was not significant (area under curve=0.50, p=0.96). However, ROC curve analysis for V5p (area under curve=0.68, p=0.012) revealed V5p ≥2 to be the optimal cut-point for separating +VA vs -VA (sensitivity 60%, specificity 65%, PPV 31%, NPV 86%).
Conclusions: In patients with HCM, QRSp measured in ECG lead V5 during intrinsic rhythm independently predicts VA. This novel ECG risk marker may reflect localized regions of delayed conduction which increase arrhythmia vulnerability.
- Hypertrophic cardiomyopathy
- Ventricular arrhythmia
- Implantable cardioconvertor defibrillator
- Risk factors
Author Disclosures: A.M. Suszko: None. K. Viswanathan: None. D.A. Spears: None. N.M. Jackson: None. G.E. Jones: None. H. Rakowski: None. A. Woo: None. M. Khurana: None. V.S. Chauhan: None.
- © 2014 by American Heart Association, Inc.