Journal of the American College of Cardiology Vol. xx, No. x, 2013 © 2013 by the American College of Cardiology Foundation ISSN 0735-1097/$36.00 Published by Elsevier Inc. http://dx.doi.org/10.1016/j.jacc.2012.10.053 CLINICAL RESEARCH Resolution of Left Bundle Branch Block–Induced Cardiomyopathy by Cardiac Resynchronization Therapy Caroline Vaillant, MD,* Raphaël P. Martins, MD,*† Erwan Donal, MD, PHD,*† Christophe Leclercq, MD, PHD,*† Christophe Thébault, MD,* Nathalie Behar, MD,* Philippe Mabo, MD,*† Claude Daubert, MD, FACC*† Rennes, France Objectives The study sought to describe a specific syndrome characterized by isolated left bundle branch block (LBBB) and a history of progressive left ventricular (LV) dysfunction, successfully treated by cardiac resynchronization ther- apy (CRT). Background Isolated LBBB in animals causes cardiac remodeling due to mechanical dyssynchrony, reversible by biventricular stimulation. However, the existence of LBBB-induced cardiomyopathy in humans remains uncertain. Methods Between 2007 and 2010, 375 candidates for CRT were screened and retrospectively included in this study if they met all criteria of a pre-defined syndrome, including: 1) history of typical LBBB for Ͼ5 years; 2) LV ejection fraction (EF) Ͼ50%; 3) decrease in LVEF to Ͻ40% and development of heart failure (HF) to NYHA functional class II to IV over several years; 4) major mechanical dyssynchrony; 5) no known etiology of cardiomyopathy; and 6) super-response to CRT with LVEF Ͼ45% and decrease in NYHA functional class at 1 year. Results The syndrome was identified in 6 patients (1.6%), 50.5 years of age on average at the time of LBBB diagnosis. HF developed over a mean of 11.6 years. At the time of referral, Doppler echocardiograms showed major me- chanical dyssynchrony at left atrioventricular, interventricular, and left intraventricular levels. During CRT, NYHA functional class decreased, LV dimensions normalized and mechanical dyssynchrony was nearly resolved in all patients, and mean LVEF increased from 31 Ϯ 12% to 56 Ϯ 8% (p ϭ 0.027). Conclusion These observations support the existence of a specific LBBB-induced cardiomyopathy resolved by CRT. Its preva- lence, time course, and risk factors need to be prospectively studied. (J Am Coll Cardiol 2013;xx:xxx) © 2013 by the American College of Cardiology Foundation The prevalence of left bundle branch block (LBBB) in the most powerful predictors of a super-response to CRT (9). general population is approximately 1% (1). Occasionally Though the clinical evidence remains weak, these observa- isolated (2), LBBB is more often associated with structural tions support the existence of a LBBB-induced cardiomy- heart disease. It is an independent predictor of cardiovas- opathy (10). In absence of a prospective registry to assess the cular mortality and heart failure (HF) events, suggesting natural history of isolated LBBB, we hypothesized that the that HF might develop as a result of LBBB (3,4). There is most persuasive evidence would be provided by well- firm evidence from clinical and experimental studies (5,6) documented clinical cases, illustrative of an original syn- that LBBB induces abnormal left ventricular (LV) contrac- drome including a history of LBBB and LV dysfunction, tion and impairs global LV function, which might cause reversed by CRT. progressive LV remodeling and HF. In addition, cardiac Methods resynchronization therapy (CRT) is a major means of reverse remodeling and protection against HF events in Patient population. Patients scheduled to undergo CRT patients presenting with typical LBBB (7,8), one of the between 2007 and 2010 were included in this single-center observational study if they presented with all characteristics of a syndrome, including: 1) normal sinus rhythm and Ͼ From the *Department of Cardiology and Vascular Diseases, University Hospital, 5-year history of typical LBBB (11); 2) LV ejection Rennes, France; and †INSERM, CIC-IT 804, Rennes, France. All authors have fraction (EF) Ͼ50% at the time of diagnosis of LBBB; reported that they have no relationships relevant to the contents of this paper to 3) progressive decrease in LVEF to Յ40%; LV end-diastolic disclose. Ն Manuscript received July 13, 2012; revised manuscript received October 23, 2012, diameter 55 mm and development of New York Heart accepted October 28, 2012. Association (NYHA) functional class II to IV; 4) presence Downloaded From: http://content.onlinejacc.org/ on 01/24/2013 2 Vaillant et al. JACC Vol. xx, No. x, 2013 CRT for LBBB-Induced Cardiomyopathy Month 2013:xxx Ͻ Abbreviations of major left heart mechanical for qualitative variables. A p value 0.05 was considered and Acronyms dyssynchrony; 5) no other iden- significant. tifiable cause of cardiomyopathy cardiac ؍ CRT resynchronization therapy (12); and 6) indication for (13), Results and super-response to, CRT, de- -ejection fraction Initial observations. Among 375 recipients of CRT sys ؍ EF fined as a LVEF Ն45% and de- heart failure tems during the study period, 6 patients (1.6 %) ranging in ؍ HF crease in NYHA functional class -age between 36 and 60 years had the pre-defined compo ؍ LBBB left bundle branch within Ն12 months. block nents of the syndrome including super-responses to CRT Data collection. The onset of left ventricular (Fig. 1). The main characteristics of the 3 women and 3 ؍ LV LBBB and history of LV func- .men at the time of LBBB diagnosis are shown in Table 1 ؍ NYHA New York Heart tion were retrospectively ascer- Association The indication for the qualifying electrocardiogram was tained by review of medical chest pain in 2, and undetermined in 4 asymptomatic records. patients. LBBB was intermittent and rate-dependent in 2 ELECTROCARDIOGRAM. The 12-lead surface electrocar- patients, before becoming permanent within 1 and 3 years, diograms were recorded at 25 mm/s during spontaneous respectively. The mean intrinsic QRS duration was 137 Ϯ rhythm, before implantation and after 12 months of 21 ms. The baseline LVEF was Ն50% in all patients. biventricular pacing, and analyzed by 2 independent Symptoms consistent with HF developed within a mean observers. The QRS duration was measured as previously of 11.6 years (range 5 to 21 years) after LBBB diagnosis. All reported (8). patients reported chest pain on exertion. Combined treat- ment with an angiotensin-converting enzyme inhibitor and ECHOCARDIOGRAPHY AND CARDIAC DYSSYNCHRONY. Tr- beta-adrenergic blocker was initiated in 5 patients, and a ansthoracic echocardiograms were recorded before (baseline) beta-adrenergic blocker alone in 1 patient. and after Ն12 months of CRT, using a ViVid Seven system (GE Healthcare, Milwaukee, Wisconsin). The LV dimensions were recorded using 2-dimensional echocardiogram-directed M-mode, and analyzed by 2 independent observers. LVEF was estimated by LV biplane Simpson’s method. LV contrac- tile function was analyzed by speckle tracking, using global longitudinal strain (14). Left atrioventricular dyssynchrony was defined as a LV filling time/R-R interval ratio Ͻ40%, and interventricular dyssynchrony as a Ͼ45 ms interval between pre-aortic and pre-pulmonary ejection times (15). Intra-LV dyssynchrony was first ascertained by M-mode echocardiography, in search of “septal flash,” defined as a systolic stretch occur- ring after initial shortening of the early activated septum (16). Using the B-mode speckle tracking software (17), dyssynchrony was ascertained as the delay between the earliest and latest peaks of longitudinal strain recorded in the mid-segment of the lateral and septal walls in the apical 4-chamber view, and radial strain recorded in the antero- septal and posterior walls in the parasternal short axis view. Intra-LV dyssynchrony was defined as a Ն130 ms delay, a peak strain occurring after the aortic valve closure, or both. Implantation and programming of the CRT systems. The CRT devices were implanted transvenously. The right ventricular lead was placed preferentially in the mid-septum. The LV lead was placed in a lateral or posterolateral tributary of the coronary sinus. Systematic efforts were made to obtain the narrowest biventricular paced QRS during the implant procedure and initial programming. Figure 1 Patient Screening Process Statistical analyses. Individual data are reported because of the small sample size. Paired measurements before and after Flow of patients from the screening to the selection of the final patient. CRT ϭ cardiac resynchronization therapy; HF ϭ heart failure; LBBB ϭ left bundle branch 12 months of CRT were compared, using nonparametric block; LVEF ϭ left ventricular ejection fraction; NYHA ϭ New York Heart Association. Wilcoxon test for quantitative variables and McNemar test Downloaded From: http://content.onlinejacc.org/ on 01/24/2013 JACC Vol. xx, No. x, 2013 Vaillant et al. 3 Month 2013:xxx CRT for LBBB-Induced Cardiomyopathy MainTable Characteristics 1 Main Characteristics of Individual of Patients Individual Patients At Time of LBBB Diagnosis Follow-Up Chest Pain as Intrinsic QRS Time Between Diagnosis of Time Between Diagnosis Intrinsic QRS Patient First Rate-Related Duration QRS LBBB and Onset of of LBBB and CRT Duration Implanted No. Sex/Age Symptom LBBB (ms) Axis Heart Failure Symptoms (yrs) System Implant (yrs) (ms)* System M/36 Ϫ ؉ 160 Ϫ65 7 19 175 CRT-D 1 M/41 ϩ ؉ 150 61 21 22 150 CRT-D 2 3 F/51 ϪϪ120† 15 17 18 155 CRT-P 4 F/51 ϪϪ140 Ϫ30 9 9 160 CRT-P 5 M/56 ϪϪ100† 15 11 11 160 CRT-D F/60 ؉ Ϫ 155 Ϫ14 5 5 155 CRT-P 6 *Before cardiac resynchronization therapy (CRT) implantation. †Patients with intermittent left bundle branch block (LBBB). D ϭ defibrillator; P ϭ pacemaker. Time to referral for CRT. The persistence, despite optimal Table 2 and Figure 3. Mean QRS duration decreased by 29 medical therapy, of symptoms consistent with NYHA func- ms with CRT (p ϭ 0.027). Mean LVEF increased from tional class II in 2, and class III in 4 patients, prompted the 31 Ϯ 12% before CRT to 56 Ϯ 8% after 12 months of CRT decision to proceed with CRT.