Effect of QRS Morphology and Duration on Clinical Outcomes After Cardiac Resynchronization Therapy　― Analysis of Japanese Multicenter Registry ―

Takafumi Oka; Koichi Inoue; Koji Tanaka; Yuko Hirao; Takaaki Isshiki; Takeshi Kimura; Masakiyo Nobuyoshi; Satoshi Shizuta; Takeshi Arita; Satoki Fujii; Katsuomi Iwakura; Kenshi Fujii; Kenji Ando

doi:10.1253/circj.CJ-17-1383

Abstract

Background: QRS duration (QRSd) and morphology are established response predictors of cardiac resynchronization therapy (CRT). However, evidence in Japanese populations is lacking.

Methods and Results: We retrospectively analyzed the Japanese multicenter CRT database. We divided patients according to their intrinsic QRSd and morphology, and assessed echocardiographic responses and clinical outcomes. The primary endpoint was a composite of all-cause death or hospitalization because of heart failure. A total of 510 patients were enrolled: 200 (39%) had left bundle branch block (LBBB) and QRSd ≥150 ms; 80 (16%) had LBBB (QRSd: 120–149 ms); 61 (12%) had non-LBBB (NLBBB) (QRSd: ≥150 ms); 54 (11%) had NLBBB (QRSd: 120–149 ms); 115 (23%), narrow (<120 ms). The proportion of echocardiographic responders was higher in LBBB (QRSd ≥150 ms) [74% vs. 51% vs. 38% vs. 52% vs. 50%, LBBB (QRSd ≥150 ms) vs. LBBB (QRSd 120–149 ms) vs. NLBBB (QRSd ≥150 ms) vs. NLBBB (QRSd 120–149 ms) vs. narrow, respectively, P<0.001]. During follow-up (3.2±1.5 years), the incidence of the primary endpoint was lowest in the LBBB group (QRSd ≥150) (28.6% vs. 42.3% vs. 45.9% vs. 55.6% vs. 55.3%, respectively, P<0.001). This difference was still significant after adjusting for other baseline characteristics.

Conclusions: In this Japanese patient population, LBBB intrinsic QRS morphology and prolonged QRSd (≥150 ms) exhibited the best response to CRT.

Cardiac resynchronization therapy (CRT) is an established optional therapeutic treatment for patients with heart failure (HF) with reduced ejection fraction to decrease the risk of all-cause death, hospitalization for HF, and ventricular tachyarrhythmia.¹^–⁵ Several randomized control trials (RCTs) have shown that approximately two-thirds of CRT recipients respond to it, but one-third fail to benefit from it. Data concerning predictors of response⁶^,⁷ have resulted in the current guidelines for CRT indication.⁸^,⁹

Of the ECG parameters, intrinsic QRS morphology¹⁰ and QRS duration (QRSd)¹¹ are considered crucial factors in determining CRT response.¹²^–¹⁴ According to consensus, left bundle branch block (LBBB), which often indicates LV mechanical dyssynchrony, is associated with a favorable response to CRT. Prolonged QRSd, which indicates broad conduction disease, also responds favorably to CRT. These 2 parameters have been recognized as important factors in determining the indications of CRT¹⁵^–¹⁷ and are part of the current Western guidelines for CRT implantation.⁸^,⁹

There have been few studies that have validated the effect of intrinsic QRS morphology and QRSd on responses to CRT in Asian populations. There are several epidemiological differences between Western and Asian populations in the baseline characteristics of CRT recipients. Particularly, Asians have less prevalence of morbidity from coronary artery disease (CAD) than in Western populations.¹⁸ Therefore, it is meaningful to validate whether Western evidence is applicable to Asian populations regarding the indications of CRT.

In this study, we divided a cohort derived from a Japanese database of CRT recipients according to intrinsic QRS morphology [LBBB or non-LBBB (NLBBB)] and duration of QRS complex (≥150 ms, 120–149 ms, or <120 ms) to investigate the proportion of echocardiographic responders and the 5-year clinical outcomes.

Methods

CUBIC Registry

This study was a retrospective analysis of cases in the CRT Utilization by Interventional Cardiologists registry (CUBIC).¹⁹ The CUBIC registry is a Japanese multicenter registry of 995 patients who underwent CRT-pacemaker (CRT-P) or CRT-defibrillator (CRT-D) implantation at 11 institutions. Originally, the CUBIC registry was established as part of the CUBIC trial, which investigated the differences in clinical outcomes and responses to CRT between patients with ischemic and those with non-ischemic cardiomyopathy (ICM). All patients who underwent CRT implantation and gave written informed consent for the CUBIC study were included. In principle, the CRT indication of each patient was determined on the basis of the degree of LV dysfunction, New York Heart Association (NYHA) functional class, QRSd and optimal pharmacological therapy. Exclusion for prospectively enrolled cases was at the discretion of the doctors. A total of 307 patients who underwent device implantation between May, 2008 and November, 2010 were prospectively enrolled and followed up for >1.5 years, and 688 patients who underwent device implantation between April, 2004 and September, 2008 were retrospectively enrolled and followed up for up to 5 years. The protocol was approved by the institutional review board at each participating center, and all patients gave informed consent for anonymous use of their data.

Patients

We identified 995 patients and included only those with available information of baseline ECG and complete echocardiographic datasets at 6–12 months after implantation (Figure 1). The echocardiographic data were analyzed in a core laboratory. Because of incomplete echocardiography dataset and baseline ECG, 321 patients were excluded. Additionally, 164 patients who underwent CRT following right ventricular pacing were also excluded. Based on the intrinsic QRS morphology, the remaining 510 patients were divided into groups with LBBB, right bundle branch block (RBBB), intraventricular conduction delay (IVCD), and narrow QRS complex (QRSd <120 ms). RBBB and IVCD were combined into NLBBB. Subsequently, based on QRSd, LBBB and NLBBB were further subdivided into LBBB with QRSd ≥150 ms, LBBB with QRSd: 120–149 ms, NLBBB with QRSd ≥150 ms, and NLBBB with QRSd: 120–149 ms.¹⁰ The intrinsic QRS morphology in LBBB was defined by the following criteria: (1) broad, notched R-waves in lateral precordial leads (V5 and V6) and usually leads I and aVL, (2) small or absent initial r waves in the right precordial leads (V1 and V2) followed by deep S waves, (3) absent septal q waves in the left-sided leads, and (4) QRSd ≥120 ms. RBBB was defined with the following criteria: (1) broad, notched R-waves (rsr’, rsR’ or rsR’ patterns) in the right precordial leads (V1 and V2), (2) wide and deep S waves in the left precordial leads (V5 and V6), and (3) QRSd ≥120 ms. IVCD was defined as a conduction disturbance that did not meet the criteria of LBBB or RBBB, and had QRSd ≥120 ms.²⁰

Figure 1.

Flow diagram of the 5 study groups subdivided into the left bundle branch block (LBBB), non-LBBB (NLBBB), and narrow QRS complex, with subsequent subgrouping of the LBBB and non-LBBB groups according to QRS duration (QRSd).

Proportion of Echocardiographic Responders and Clinical Outcomes of Each Group

We defined echocardiographic responders as patients with a relative reduction of left ventricular end-systolic volume (LVESV) ≥15% at 6–12 months after CRT implantation.²¹ Relative reduction of LVESV was estimated as follows: relative reduction of LVESV (%)={LVESV (baseline)−LVESV (6–12 months)}/LVESV (baseline)×100. We calculated the proportion of echocardiographic responders and relative reduction of LVESV in each group. We also analyzed the proportions of super-responders (reduction of LVESV ≥30%) and negative responders [decreased LV ejection fraction (LVEF)].¹⁰ To evaluate an improvement in HF symptoms, we analyzed changes in NYHA functional class 6 months after implantation. We analyzed the baseline patients’ characteristics of each group with univariate analysis. We performed survival analysis of the primary and secondary endpoints among the 5 subgroups. The primary endpoint was a composite of all-cause death or hospitalization because of HF. The secondary endpoints were all-cause death, HF hospitalization, and sustained ventricular tachycardia (VT) and ventricular fibrillation (VF) in patients with CRT-P, and adequate ICD therapy for VT/VF in patients with CRT-D. We compared the clinical outcomes in each group with those of patients with LBBB (≥150 ms), using multivariable Cox proportional hazards models adjusted for several clinical variables that were significant at a prespecified P<0.05 in the univariate analysis. Finally, we divided patients into ICM and non-ICM and groups, and performed a subanalysis of the primary endpoints among the 5 subgroups in the non-ICM and ICM cohorts.

Statistical Analysis

In the univariate analysis of covariates between the groups, continuous variables were analyzed using one-way analysis of variance with Tukey’s post-hoc analysis, and categorical variables were analyzed using the chi-square test. For survival analysis, primary and secondary endpoints were identified using the Kaplan-Meier method with log-rank test. Adjusted multivariable Cox proportional hazards regression analyses were performed to assess the correlation between intrinsic QRS morphology and QRSd, and each outcome. The pool of variables considered were those found to be significant at a prespecified P<0.05 in the univariate analysis. All data are expressed as mean±standard deviation. All P-values reported are 2-sided with a prespecified significance of P<0.05. Analyses were performed using the MedCalc software version 15.8 (MedCalc Software, Ostend, Belgium).

Results

Patients

A majority of the patients underwent implantation with CRT-D devices (n=370, 73%) and the remaining patients underwent implantation with CRT-P devices (n=140, 27%). The overall mean age was 67.8±11.5 years and 71% were men; 31% of the patients had ischemic heart disease as the primary etiology of HF and the severity in 75% was of NYHA functional class III or IV. At baseline, the mean LVEF was 27.7±9.4% and LVESV was 139.3±64.6 mL. In terms of intrinsic QRS morphology, 55% of the study population had LBBB, 11% had RBBB, and 11% had IVCD. The mean QRS duration was 146.4±30.7 ms.

Intrinsic QRS Morphology and QRS Duration

The 510 patients were divided into groups with LBBB with QRSd ≥150 ms (n=200, 39%), LBBB with QRSd 120–149 ms (n=80, 16%), NLBBB with QRSd ≥150 ms (n=61, 12%), NLBBB with QRSd 120–149 ms (n=54, 11%), and narrow QRS complex (n=115, 23%). The baseline characteristics of each group are summarized in Table 1. In the univariate analysis comparing the 5 groups, the following 8 factors were found to be significant with P<0.05: age, percentage of CRT-D, female sex, non-ICM, smoking, prior myocardial infarction (MI), atrioventricular (AV) nodal dysfunction, and moderate-to-severe mitral regurgitation (MR).

Table 1. Baseline Characteristics of Patients With Different Intrinsic QRS Morphologies and QRS Duration

Intrinsic QRS morphology	LBBB*	LBBB	NLBBB^†	NLBBB	Narrow	P value
QRS duration (ms)	≥150	120–149	≥150	120–149	<120
No. of patients	200	80	61	54	115
CRT-D, %	129 (66%)	60 (75%)	44 (72%)	45 (83%)	95 (83%)	0.003
Female sex	75 (38%)	24 (30%)	13 (21%)	11 (20%)	23 (20%)	0.004
Age, years	69.4±11.1	69.7±9.6	67.5±11.2	64.1±11.9	65.7±12.6	0.003
Height, cm	158.9±9.5	157.7±19.9	159.9±22.5	163.3±8.2	161.4±8.3	0.10
Weight, kg	57.4±12.5	58.0±11.0	58.1±12.1	61.5±13.9	58.3±12.0	0.32
BMI, kg/m²	22.5±3.5	22.6±3.5	21.9±3.6	22.9±4.1	22.3±3.7	0.64
Systolic BP, mmHg	120.0±19.6	115.3±18.8	115.1±21.5	117.7±21.6	118.3±21.2	0.37
Diastolic BP, mmHg	68.1±12.0	67.3±13.6	66.6±11.8	70.7±15.5	68.9±13.7	0.45
NYHA functional class III and IV	149 (75%)	62 (78%)	44 (72%)	36 (67%)	90 (78%)	0.53
Non-ICM	154 (77%)	54 (68%)	34 (56%)	36 (67%)	72 (63%)	0.010
Medical comorbidities
Hypertension	101 (51%)	38 (48%)	24 (39%)	30 (56%)	62 (54%)	0.36
Diabetes mellitus	68 (34%)	24 (30%)	26 (43%)	23 (43%)	40 (35%)	0.29
Dyslipidemia	83 (42%)	31 (39%)	24 (39%)	18 (33%)	43 (37%)	0.85
Smoking	29 (15%)	21 (28%)	12 (20%)	12 (22%)	36 (31%)	0.009
CKD	42 (21%)	26 (33%)	20 (33%)	19 (35%)	27 (23%)	0.075
COPD and bronchial asthma	9 (5%)	3 (4%)	5 (8%)	1 (2%)	4 (3%)	0.51
Ischemic stroke	18 (9%)	12 (15%)	6 (10%)	4 (7%)	13 (11%)	0.42
Intracranial bleeding	3 (2%)	1 (1%)	1 (2%)	1 (2%)	0 (0%)	0.76
Prior MI	41 (21%)	25 (31%)	26 (43%)	15 (28%)	37 (32%)	0.010
Prior CABG	18 (9%)	10 (13%)	10 (16%)	6 (11%)	17 (15%)	0.44
Prior PCI	46 (23%)	28 (35%)	22 (36%)	17 (31%)	38 (33%)	0.12
Prior HF hospitalization	157 (79%)	66 (83%)	51 (84%)	48 (89%)	88 (77%)	0.33
Prior valvular disease surgery	16 (8%)	9 (11%)	4 (7%)	1 (2%)	13 (11%)	0.25
Prior ventricular tachyarrhythmia	21 (11%)	10 (13%)	12 (20%)	7 (13%)	20 (17%)	0.29
ECG variables
QRS duration, ms	170.9±16.8	135.7±8.3	171.3±20.4	132.3±7.4	104.7±9.9	<0.001
Sinus nodal dysfunction	193 (97%)	77 (96%)	60 (98%)	52 (96%)	112 (97%)	0.94
AVN dysfunction	38 (19%)	14 (18%)	26 (43%)	12 (22%)	18 (16%)	0.005
Chronic AF	15 (8%)	9 (11%)	6 (10%)	7 (13%)	21 (18%)	0.070
Echocardiographic characteristics
LVEDV, mL	194.0±76.6	191.2±82.1	192.7±59.3	185.1±79.4	175.9±64.5	0.90
LVESV, mL	144.6±69.3	140.2±69.7	140.4±52.4	135.6±65.2	130.5±58.0	0.45
LVEF, %	27.1±9.3	28.9±10.3	28.1±8.8	27.9±7.3	27.5±10.1	0.68
MR ≥moderate	43 (22%)	28 (35%)	11 (18%)	11 (20%)	18 (16%)	0.023
Medications
Aldosterone antagonists	97 (49%)	43 (54%)	36 (59%)	24 (44%)	64 (56%)	0.39
Amiodarone	47 (24%)	19 (24%)	18 (30%)	17 (31%)	33 (29%)	0.18
ACEI	83 (42%)	25 (31%)	27 (44%)	17 (31%)	47 (41%)	0.32
ARB	77 (39%)	35 (44%)	16 (26%)	19 (35%)	43 (37%)	0.30
β-blockers	139 (70%)	55 (69%)	36 (59%)	36 (57%)	77 (67%)	0.66
Digitalis	48 (24%)	15 (19%)	18 (30%)	7 (13%)	28 (24%)	0.23
Diuretics	162 (81%)	65 (81%)	53 (87%)	43 (80%)	91 (79%)	0.79
Statins	64 (32%)	26 (33%)	17 (28%)	12 (22%)	24 (21%)	0.19

ACEI, angiotensin-converting enzyme inhibitor; AF, atrial fibrillation; ARB, angiotensin-receptor blocker; AVN, atrioventricular node; BMI, body mass index; BP, blood pressure; CABG, coronary artery bypass grafting; CKD, chronic kidney disease; COPD, chronic obstructive pulmonary disease; CRT-D, cardiac resynchronization-defibrillator therapy; HF, heart failure; ICM, ischemic cardiomyopathy; LBBB, left bundle branch block; LVEDV, left ventricular end-diastolic volume; LVEF, left ventricular ejection fraction; LVESV, left ventricular end-systolic volume; MI, myocardial infarction; MR, mitral regurgitation; NLBBB, non-left bundle branch block; NYHA, New York Heart Association; PCI, percutaneous coronary intervention.

Echocardiographic Response and Improvement in HF Symptoms

Table 2 shows the echocardiographic responses 6–12 months after CRT implantation. Postoperative LVEF was the highest in the group with LBBB with QRSd ≥150 ms [LBBB (QRSd ≥150 ms) vs. LBBB (QRSd, 120–149 ms) vs. NLBBB (QRSd ≥150 ms) vs. NLBBB (QRSd, 120–149 ms) vs. narrow (QRSd <120 ms): 37.7±13.5% vs. 34.0±12.1% vs. 31.1±9.5% vs. 32.4±9.3% vs. 33.9±10.5%, P<0.001]. The reduction in LVESV was the largest in the group with LBBB with QRSd ≥150 ms compared with the other groups [33.2±30.8% vs. 16.8±29.7% vs. 12.1±23.3% vs. 12.2±27.5%, 18.8±28.1%, respectively, P<0.001]. The percentage of responders was highest in the LBBB with QRSd ≥150 ms group compared with the other groups (74% vs. 51% vs. 38% vs. 52% vs. 50%, respectively, P<0.001), and the percentage of super-responders was highest in the same group (59% vs. 38% vs. 25% vs. 26% vs. 35%, respectively, P<0.001). The percentage of negative responders was lowest in the group with LBBB with QRSd ≥150 ms compared with the other groups (19% vs. 33% vs. 43% vs. 28% vs. 29%, respectively, P=0.003). The proportion of patients with NYHA functional class III or IV was not different among the 5 groups at baseline (75% vs. 78% vs. 72% vs. 67%, 78%, P=0.53). At 6 months after implantation, the percentage of NYHA III or IV was the lowest in the LBBB with QRSd ≥150 ms group compared with the other groups (11% vs. 25% vs. 33% vs. 28% vs. 21%, respectively, P=0.006) (Table 2).

Table 2. Echocardiographic Response to Cardiac Resynchronization Therapy and Changes in NYHA Functional Class 6 Months After Device Implantation

Intrinsic QRS morphology	LBBB*	LBBB	NLBBB	NLBBB	Narrow	P value
QRS duration (ms)	≥150	120–149	≥150	120–149	<120
No. of patients	200	80	61	54	115
LVEF pre, %	27.1±9.3	28.9±10.3	28.1±8.8	27.9±7.3	27.5±10.1	0.68
LVEF post, %	37.7±13.5	34.0±12.1	31.1±9.5	32.4±9.3	33.9±10.5	<0.001
LVEF change, %	10.6±13.5	5.1±9.9	3.1±9.0	4.5±9.4	4.5±9.4	<0.001
LVESV pre, mL	144.6±69.3	140.2±69.7	140.4±52.4	135.6±65.2	130.5±58.0	0.14
LVESV post, mL	96.8±67.0	118.5±78.8	122.5±57.0	117.6±73.8	102.0±55.8	0.015
LVEDV pre, mL	194.0±76.6	191.2±82.1	192.7±59.3	185.1±79.4	185.1±79.4	0.90
LVEDV post, mL	144.5±74.9	168.3±90.3	173.1±63.6	168.9±89.2	148.1±64.5	0.015
LVESV reduction, %	33.2±30.8	16.8±29.7	12.1±23.3	12.2±27.5	18.8±28.1	<0.001
Responder (LVESV reduction ≥15%)	148 (74%)	41 (51%)	23 (38%)	28 (52%)	58 (50%)	<0.001
Super-responder (LVESV reduction ≥30%)	118 (59%)	30 (38%)	15 (25%)	14 (26%)	40 (35%)	<0.001
Negative responder (LVEF change <0%)	37 (19%)	26 (33%)	26 (43%)	15 (28%)	33 (29%)	0.003
NYHA functional class III and IV pre	149 (75%)	62 (78%)	44 (72%)	36 (67%)	90 (78%)	0.53
NYHA functional class III and IV post	21 (11%)	20 (25%)	20 (33%)	15 (28%)	24 (21%)	0.006

Abbreviations as in Table 1.

Clinical Outcomes of Each Group

During a mean follow-up of 3.2±1.5 years, the primary endpoints occurred in 211 (8.3% per year) of 510 patients, and the secondary endpoints of all-cause death, HF hospitalization, and VT/VF occurred in 106 (4.2% per year), 173 (6.8% per year), and 108 (5.4% per year) patients, respectively. The incidence of the composite endpoint was significantly different between the groups (28.6% vs. 42.3% vs. 45.9% vs. 55.6% vs. 55.3%, respectively, P<0.001) (Figure 2A). There were significant differences in the incidence of all-cause death (15.5% vs. 23.8% vs. 24.6% vs. 27.8% vs. 22.6%, respectively, P=0.047) and HF hospitalization (22.1% vs. 35.9% vs. 36.1% vs. 48.2% vs. 46.5%, respectively, P<0.001) (Figure 2B,C). The incidence of VT/VF was significantly different between the groups (14.7% vs. 19.0% vs. 15.3% vs. 23.1% vs. 37.7%, respectively, P<0.001) (Figure 2D).

Figure 2.

Kaplan-Meier estimates of cumulative probability of the primary and secondary endpoints in the 5 study groups. (A) Composite endpoint (all-cause death and hospitalization because of heart failure), (B) all-cause death, (C) hospitalization because of heart failure, and (D) VT/VF events. LBBB, left bundle branch block; NLBBB, non-LBBB; VF, ventricular fibrillation; VT, ventricular tachycardia.

Table 3 shows the results of the multivariable Cox hazards models adjusted for clinical variables found to be significant according to a prespecified P<0.05 in the univariate analysis (Table 1), except QRS duration. Models were adjusted with baseline variables such as age at enrollment, sex, CRT-D or CRT-P, ICM or non-ICM etiology, presence of old MI, smoking, moderate-to-severe MR and AV nodal dysfunction. Patients with LBBB with QRSd ≥150 ms were significantly associated with a decreased risk of the composite endpoint and HF hospitalization compared with other groups. Patients with narrow QRSd were exclusively associated with increased risk of VT/VF. As shown in Figure 3A, patients with non-ICM etiology were divided into 5 groups and their clinical outcomes analyzed. There were significant differences in the incidence of the composite endpoint in the same way as in the overall patient group (22.9% vs. 40.4% vs. 44.1% vs. 58.3% vs. 53.5%, P<0.001). In Figure 3B, patients with ICM etiology were analyzed and there were no significant differences in the incidence of the composite endpoint (47.8% vs. 46.2% vs. 48.2% vs. 50.0% vs. 58.1%, P=0.93).

Table 3. Unadjusted and Adjusted HRs of Clinical Outcomes in Overall Patients With Different Intrinsic QRS Morphologies and Duration

Endpoint	No. of events	No. of patients	LBBB QRSd 120–149 ms		NLBBB QRSd ≥150 ms		NLBBB QRSd 120–149 ms		Narrow QRSd <120 ms
Endpoint	No. of events	No. of patients	HR (95% CI)	P value	HR (95% CI)	P value	HR (95% CI)	P value	HR (95% CI)	P value
Composite endpoint
Unadjusted analysis	211	506	1.99 (1.30–3.06)	0.002	2.10 (1.34–3.30)	0.001	2.72 (1.75–4.23)	<0.001	2.42 (1.69–3.46)	<0.001
Adjusted analysis	211	505	1.80 (1.17–2.78)	0.008	1.96 (1.23–3.13)	<0.001	2.77 (1.76–4.36)	<0.001	2.50 (1.72–3.63)	<0.001
All-cause death
Unadjusted analysis	106	510	1.97 (1.11–3.49)	0.020	1.86 (1.01–3.44)	0.049	2.19 (1.19–4.05)	0.013	1.59 (0.94–2.67)	0.083
Adjusted analysis	106	509	1.71 (0.95–3.05)	0.073	1.87 (0.98–3.55)	0.057	2.50 (1.32–4.74)	0.005	1.71 (1.00–2.94)	0.053
HF hospitalization
Unadjusted analysis	173	506	2.15 (1.34–3.46)	0.002	2.11 (1.26–3.51)	0.004	3.01 (1.86–4.88)	<0.001	2.60 (1.74–3.87)	<0.001
Adjusted analysis	173	505	1.96 (1.21–3.15)	0.006	1.97 (1.16–3.34)	0.012	3.07 (1.87–5.05)	<0.001	2.72 (1.79–4.13)	<0.001
VT/VF
Unadjusted analysis	108	502	1.40 (0.75–2.60)	0.29	1.11 (0.53–2.33)	0.78	1.75 (0.90–3.43)	0.10	3.14 (1.97–5.03)	<0.001
Adjusted analysis	108	501	1.30 (0.69–2.43)	0.42	1.02 (0.47–2.18)	0.97	1.34 (0.67–2.66)	0.41	2.90 (1.77–4.75)	<0.001

The group with LBBB and QRSd ≥150 ms is reference. Models were adjusted with baseline variables found to be significant at a prespecified P<0.05 in univariate analysis (Table 1) such as age at enrollment, sex, CRT-D/CRT-P, ICM/non-ICM etiology, presence of old myocardial infarction, smoking, moderate-severe mitral regurgitation and atrioventricular nodal dysfunction. CI, confidence interval; CRT-P, cardiac resynchronization-pacemaker therapy; HR, hazard ratio; VT/VF, ventricular tachycardia/ventricular fibrillation. Other abbreviations as in Table 1.

Figure 3.

Kaplan-Meier estimates of cumulative probability of the primary endpoints in the 5 study groups in (A) the non-ICM, and (B) the ICM subgroups. ICM, ischemic cardiomyopathy.

Discussion

Major Findings

We sought to investigate the effect of intrinsic QRS morphology and duration on echocardiographic response and clinical outcomes after CRT implantation in a Japanese population based on a multicenter database. Our major findings were: (1) patients with intrinsic LBBB QRS morphology and prolonged QRSd (≥150 ms) were most likely to be echocardiographic responders and super-responders, and the least likely to be negative responders, and (2) patients with intrinsic LBBB QRS morphology and wide QRS duration (≥150 ms) were consistently associated with the most favorable clinical outcomes (free from all-cause death or HF hospitalization).

Background of CRT Recipients in Japan

Few studies are available concerning the response to CRT stratified by intrinsic QRS morphology and QRSd in Asian patients. For the interpretation of the results of the current study, we would like to point out the epidemiological difference between Japanese and Western populations in the characteristics of CRT recipients. Compared with Western populations, East Asian populations generally have lower morbidity from CAD. In the Western world, morbidity of CAD has been reported to be 4–5-fold higher than in Japan.¹⁸ In our study population from a Japanese registry, only 30% of patients had an ICM etiology, whereas 70% of patients in prior Western studies had an ICM etiology. Thus, the majority of CRT recipients in the Japanese population had non-ICM etiology, which has been reported as a positive predictor for response to CRT.⁶^,⁷ In the current study, the proportion of echocardiographic responders might be slightly higher than previously reported for Western countries. To investigate whether the etiology of LV dysfunction affected clinical outcome, we added a subanalysis of the composite endpoint in non-ICM and ICM populations separately (Figure 3). The result for the non-ICM cohort was similar to that for the overall patients. Among the non-ICM population, patients with LBBB and QRSd ≥150 ms were the most likely to benefit from CRT. However, we did not find differences in the clinical outcomes of the ICM cohort. In our analysis, the number of the patients with ICM might have been too small to investigate the prognosis after CRT implantation. We need to enroll more patients to conclude this issue.

Effect of Intrinsic QRS Morphology and QRS Duration on CRT Effectiveness

Intrinsic QRS morphology refers to the areas of delayed activation resulting in mechanical dyssynchrony, and duration of QRS refers to the severity of the activation delay. Typically, LBBB is associated with delayed activation of the lateral and posterolateral portions of the LV, so it is the target of LV lateral wall pacing. Prolonged QRSd implies a greater delay in activation, which needs to be corrected using biventricular pacing. Therefore, LBBB intrinsic QRS morphology and prolonged QRSd are of considerable importance in predicting the response to CRT.¹⁵^,¹⁷ In the current study, Japanese patients with LBBB intrinsic QRS morphology and QRSd ≥150 ms were the most likely of all the study patients to become echocardiographic responders at 6–12-month follow-up and they exhibited the most favorable clinical outcomes even after adjustment for several covariates. In contrast, patients with NLBBB morphology and QRSd between 120 and 149 ms, and narrow QRS complex exhibited the poorest long-term prognoses.¹⁶^,²²^,²³ Although there are epidemiological differences in the basic characteristics between patients from Japan and the Western world, LBBB intrinsic QRS morphology and wide duration of QRS appear to universally indicate a beneficial response to CRT.

A debate has been ongoing regarding which is the more important regarding CRT response: intrinsic QRS morphology or QRSd.²⁴ Dupont et al stratified patients according to both QRS morphology and QRSd in the same way as we did, and concluded that QRS morphology was the more important baseline ECG determinant of CRT response.¹⁰ In our data, however, the clinical outcomes of patients with LBBB with QRSd 120–149 ms and those with non-LBBB with QRSd ≥150 ms were almost similar. Neither of these QRS criteria should be considered in isolation, as they both reflect the ability of CRT to address LV mechanical dyssynchrony only when considered together. At the moment, LBBB intrinsic QRS morphology is not involved in the Class I indication for CRT in the current JCS guidelines.²⁵ Both of these QRS criteria should be considered in future guidelines on CRT recommendations.

CRT Response in Patients With Narrow QRS Complex

In our study, 115 patients with narrow QRS complex were enrolled and approximately 50% of them were echocardiographic responders. The proportion of super-responders in the narrow QRS group was higher than we expected. We speculate that correction of AV dyssynchrony and bradycardia by CRT, and optimal medication therapy might improve the response in this group. Correction of bradycardia also possibly allowed physicians to increase the dose of β-blocker. Although the echocardiographic response of the narrow QRSd group was not poor, the risk of HF hospitalization was the highest in those with narrow QRSd and NLBBB with QRSd 120–149 ms. Intriguingly, whereas CRT did not affect the risk of VT/VF in the 4 groups with QRSd ≥120 ms, only patients with narrow QRS complex exhibited an increased risk of VT/VF. The pro-arrhythmic effects of CRT have been argued in several prior studies.²⁶^–²⁸ Biventricular pacing could increase the dispersion of repolarization, and induce VT/VF in some patients. Before pacing, a narrow QRS duration could mean smaller activation delay compared with wide QRS duration. In patients with a narrow QRS complex, biventricular pacing might increase the activation delay and dispersion of repolarization. Several RCTs have failed to demonstrate the effectiveness of CRT in patients with a narrow QRS complex, regardless of the presence of LV dyssynchrony.²⁹^–³¹ Notably, the EchoCRT trial was terminated because of higher mortality rate in the CRT arm compared with the control arm.²⁹ Therefore, CRT should not be recommended in patients with a narrow QRS complex.

Clinical Implications

LBBB intrinsic QRS morphology and wide QRS duration (≥150 ms) had a close association with a beneficial response after CRT implantation in this Japanese cohort. These 2 ECG parameters should be taken into consideration together when discussing the indications for CRT in the Japanese population.

Study Limitations

This study should be interpreted after taking into account its methodological limitations. First, it was a post-hoc analysis and 69% of patients were retrospectively enrolled. Second, 32% of patients were excluded from the analysis because of incomplete data. Third, we compared clinical outcomes among 5 groups that were subdivided according to intrinsic QRS morphology and duration. Some of the baseline patient characteristics were different between groups. Fourth, because of the lack of a control group without CRT, we had to set the group with LBBB (QRSd ≥150 ms) as the reference group in our multivariate Cox regression analysis.

Conclusions

In this Japanese population, LBBB intrinsic QRS morphology and wide QRS duration (≥150 ms) had the strongest association with a beneficial response after CRT implantation.

Funding

The CUBIC trial was sponsored by Medtronic Japan, Inc.

Acknowledgement of Grant Support

None.

Conflict of Interest

K. Inoue received honoraria for lecturing from Medtronic Japan, Inc. K.A. received honoraria for lecturing from Medtronic Japan, Inc.

References

1. Abraham WT, Fisher WG, Smith AL, Delurgio DB, Leon AR, Loh E, et al. Cardiac resynchronization in chronic heart failure. N Engl J Med 2002; 346: 1845–1853.
2. Bristow MR, Saxon LA, Boehmer J, Krueger S, Kass DA, De Marco T, et al. Cardiac-resynchronization therapy with or without an implantable defibrillator in advanced chronic heart failure. N Engl J Med 2004; 350: 2140–2150.
3. Cazeau S, Leclercq C, Lavergne T, Walker S, Varma C, Linde C, et al. Effects of multisite biventricular pacing in patients with heart failure and intraventricular conduction delay. N Engl J Med 2001; 344: 873–880.
4. Cleland JG, Daubert JC, Erdmann E, Freemantle N, Gras D, Kappenberger L, et al. The effect of cardiac resynchronization on morbidity and mortality in heart failure. N Engl J Med 2005; 352: 1539–1549.
5. Moss AJ, Hall WJ, Cannom DS, Klein H, Brown MW, Daubert JP, et al. Cardiac-resynchronization therapy for the prevention of heart-failure events. N Engl J Med 2009; 361: 1329–1338.
6. Goldenberg I, Moss AJ, Hall WJ, Foster E, Goldberger JJ, Santucci P, et al. Predictors of response to cardiac resynchronization therapy in the Multicenter Automatic Defibrillator Implantation Trial with Cardiac Resynchronization Therapy (MADIT-CRT). Circulation 2011; 124: 1527–1536.
7. Hsu JC, Solomon SD, Bourgoun M, McNitt S, Goldenberg I, Klein H, et al. Predictors of super-response to cardiac resynchronization therapy and associated improvement in clinical outcome: The MADIT-CRT (multicenter automatic defibrillator implantation trial with cardiac resynchronization therapy) study. J Am Coll Cardiol 2012; 59: 2366–2373.
8. Brignole M, Auricchio A, Baron-Esquivias G, Bordachar P, Boriani G, Breithardt OA, et al. 2013 ESC Guidelines on cardiac pacing and cardiac resynchronization therapy: The Task Force on cardiac pacing and resynchronization therapy of the European Society of Cardiology (ESC). Developed in collaboration with the European Heart Rhythm Association (EHRA). Eur Heart J 2013; 34: 2281–2329.
9. Epstein AE, DiMarco JP, Ellenbogen KA, Estes NA 3rd, Freedman RA, Gettes LS, et al. 2012 ACCF/AHA/HRS focused update incorporated into the ACCF/AHA/HRS 2008 guidelines for device-based therapy of cardiac rhythm abnormalities: A report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines and the Heart Rhythm Society. Circulation 2013; 127: e283–e352.
10. Dupont M, Rickard J, Baranowski B, Varma N, Dresing T, Gabi A, et al. Differential response to cardiac resynchronization therapy and clinical outcomes according to QRS morphology and QRS duration. J Am Coll Cardiol 2012; 60: 592–598.
11. Biton Y, Kutyifa V, Cygankiewicz I, Goldenberg I, Klein H, McNitt S, et al. Relation of QRS duration to clinical benefit of cardiac resynchronization therapy in mild heart failure patients without left bundle branch block: The Multicenter Automatic Defibrillator Implantation Trial with Cardiac Resynchronization Therapy Substudy. Circ Heart Fail 2016; 9: e002667.
12. Birnie DH, Ha A, Higginson L, Sidhu K, Green M, Philippon F, et al. Impact of QRS morphology and duration on outcomes after cardiac resynchronization therapy: Results from the Resynchronization-Defibrillation for Ambulatory Heart Failure Trial (RAFT). Circ Heart Fail 2013; 6: 1190–1198.
13. Gold MR, Thebault C, Linde C, Abraham WT, Gerritse B, Ghio S, et al. Effect of QRS duration and morphology on cardiac resynchronization therapy outcomes in mild heart failure: Results from the Resynchronization Reverses Remodeling in Systolic Left Ventricular Dysfunction (REVERSE) study. Circulation 2012; 126: 822–829.
14. Peterson PN, Greiner MA, Qualls LG, Al-Khatib SM, Curtis JP, Fonarow GC, et al. QRS duration, bundle-branch block morphology, and outcomes among older patients with heart failure receiving cardiac resynchronization therapy. Jama 2013; 310: 617–626.
15. Bilchick KC, Kamath S, DiMarco JP, Stukenborg GJ. Bundle-branch block morphology and other predictors of outcome after cardiac resynchronization therapy in Medicare patients. Circulation 2010; 122: 2022–2030.
16. Sipahi I, Chou JC, Hyden M, Rowland DY, Simon DI, Fang JC. Effect of QRS morphology on clinical event reduction with cardiac resynchronization therapy: Meta-analysis of randomized controlled trials. Am Heart J 2012; 163: 260–267.e263.
17. Zareba W, Klein H, Cygankiewicz I, Hall WJ, McNitt S, Brown M, et al. Effectiveness of cardiac resynchronization therapy by QRS morphology in the Multicenter Automatic Defibrillator Implantation Trial-Cardiac Resynchronization Therapy (MADIT-CRT). Circulation 2011; 123: 1061–1072.
18. JCS Joint Working Group. Guidelines for secondary prevention of myocardial infarction (JCS 2011): Digest version. Circ J 2013; 77: 231–248.
19. Oka T, Inoue K, Tanaka K, Toyoshima Y, Isshiki T, Kimura T, et al. Duration of reverse remodeling response to cardiac resynchronization therapy: Rates, predictors, and clinical outcomes. Int J Cardiol 2017; 243: 340–346.
20. Hara H, Oyenuga OA, Tanaka H, Adelstein EC, Onishi T, McNamara DM, et al. The relationship of QRS morphology and mechanical dyssynchrony to long-term outcome following cardiac resynchronization therapy. Eur Heart J 2012; 33: 2680–2691.
21. Ypenburg C, van Bommel RJ, Borleffs CJ, Bleeker GB, Boersma E, Schalij MJ, et al. Long-term prognosis after cardiac resynchronization therapy is related to the extent of left ventricular reverse remodeling at midterm follow-up. J Am Coll Cardiol 2009; 53: 483–490.
22. Cunnington C, Kwok CS, Satchithananda DK, Patwala A, Khan MA, Zaidi A, et al. Cardiac resynchronisation therapy is not associated with a reduction in mortality or heart failure hospitalisation in patients with non-left bundle branch block QRS morphology: Meta-analysis of randomised controlled trials. Heart 2015; 101: 1456–1462.
23. Nery PB, Ha AC, Keren A, Birnie DH. Cardiac resynchronization therapy in patients with left ventricular systolic dysfunction and right bundle branch block: A systematic review. Heart Rhythm 2011; 8: 1083–1087.
24. Poole JE, Singh JP, Birgersdotter-Green U. QRS duration or QRS morphology: What really matters in cardiac resynchronization therapy? J Am Coll Cardiol 2016; 67: 1104–1117.
25. JCS Joint Working Group. Guidelines for non-pharmacotherapy of cardiac arrhythmias (JCS 2011): Digest version. Circ J 2013; 77: 249–274.
26. Basu Ray I, Fendelander L, Singh JP. Cardiac resynchronization therapy and its potential proarrhythmic effect. Clin Cardiol 2007; 30: 498–502.
27. Blaschke F, Knaus T, Celebi O, Krebs A, Nitardy A, Habedank D, et al. Ventricular tachycardia or ventricular fibrillation occurs less often in patients with left bundle branch block and combined resynchronization and defibrillators than in patients with narrow QRS and conventional defibrillators. Europace 2012; 14: 224–229.
28. Bradfield JS, Shivkumar K. Cardiac resynchronization therapy-induced proarrhythmia: Understanding preferential conduction within myocardial scars. Circ Arrhythm Electrophysiol 2014; 7: 1000–1002.
29. Ruschitzka F, Abraham WT, Singh JP, Bax JJ, Borer JS, Brugada J, et al. Cardiac-resynchronization therapy in heart failure with a narrow QRS complex. N Engl J Med 2013; 369: 1395–1405.
30. Thibault B, Harel F, Ducharme A, White M, Ellenbogen KA, Frasure-Smith N, et al. Cardiac resynchronization therapy in patients with heart failure and a QRS complex <120 milliseconds: The Evaluation of Resynchronization Therapy for Heart Failure (LESSER-EARTH) trial. Circulation 2013; 127: 873–881.
31. Yu CM, Chan YS, Zhang Q, Yip GW, Chan CK, Kum LC, et al. Benefits of cardiac resynchronization therapy for heart failure patients with narrow QRS complexes and coexisting systolic asynchrony by echocardiography. J Am Coll Cardiol 2006; 48: 2251–2257.

Corresponding author

Register with J-STAGE for free!