Cardiac resynchronization therapy with defibrillator (CRT-D) is an approved treatment for patients with advanced stages of heart failure in the setting of widened QRS, and this therapy is associated with reduction in symptoms, improvement in functional capacity, and decrease in hospitalization and mortality (1). The recently reported randomized MADIT-CRT (Multicenter Automatic Defibrillator Implantation Trial With Cardiac Resynchronization Therapy) trial demonstrated that CRT-D treated patients with New York Heart Association (NYHA) functional class I and II heart failure symptoms, left ventricular ejection fraction (LVEF) ≤0.30 and QRS ≥130 ms had a 34% reduction in the risk of heart failure or death, whichever came first, when compared with patients treated with an implantable cardioverter-defibrillator (ICD) (2). In this substudy from the MADIT-CRT trial, we report the sex-specific outcomes with CRT-D versus ICD therapy and explore the factors associated with the more favorable response to this therapy in women than in men.

The design and primary results of the MADIT-CRT trial were recently published (2). Briefly, the MADIT-CRT study was designed to determine whether CRT-D therapy would reduce the risk of death or heart failure events in patients with mild cardiac symptoms, a reduced ejection fraction, and wide QRS complex when compared to ICD therapy. The patients were randomly assigned in a 3:2 ratio to receive either CRT-D or ICD. From December 22, 2004, through April 23, 2008, a total of 1,820 patients were enrolled at 110 hospital centers. Primary analyses included Cox proportional-hazards regression for heart failure alone and for death at any time and evaluation of 10 prespecified categorical subgroups and treatment interactions. The effects of CRT-D in 7 of these subgroups were presented in the primary analysis of the MADIT-CRT trial (age, sex, NYHA functional class and substrate, QRS duration, LVEF, left ventricular end-diastolic volume, and left ventricular end-systolic volume) and 2 interaction effects between subgroup and treatment were identified. The CRT-D therapy was associated with a greater benefit in women than in men, and in patients with QRS duration ≥150 than <150 ms.

The protocol was approved by the institutional review board at each of the participating centers. Patients of either sex who were at least 21 years of age were enrolled in the study if they had ischemic cardiomyopathy (NYHA functional class I or II) or nonischemic cardiomyopathy (NYHA functional class II only), sinus rhythm, an ejection fraction of ≤0.30, and prolonged intraventricular conduction with a QRS duration of ≥130 ms. All eligible subjects met the guideline indication for ICD therapy. Patients were excluded from enrollment for a variety of reasons as previously reported (2).

Two-dimensional echocardiography (3) was performed at baseline and at the 1-year follow-up in 1,417 patients to assess changes in the left ventricular volumes, left atrial volume, and ejection fraction in the 2 treatment groups. Volumes were estimated by averaging those derived from the 2- and 4-chamber views according to Simpson's method, and the ejection fraction was calculated using a standardized protocol. Volumes were indexed for body surface area.

For categorical variables, the chi-square statistic was used to assess group differences. For continuous variables, comparisons were performed using the t test for independent samples. Cumulative survival curves were determined by Kaplan-Meier analysis (4), with statistical comparison by the log-rank method. The Cox proportional-hazards regression model (5) was used to calculate the risk for specified end points, and selective treatment interactions were evaluated. Outcome analyses were performed according to the intention-to-treat principle. The Wilcoxon rank-sum test was used to evaluate the absolute change in the median difference in echocardiographic parameters between female and male patients who had paired baseline and 12-month recordings. All p values are 2-tailed and have not been adjusted for the stopping rule. Analyses used version 3.0 of the database, which was released on September 30, 2009.

The clinical characteristics of 453 women and the 1,367 men are presented in (Table 1). The female patients were more likely to have nonischemic cardiomyopathy and left bundle branch block conduction pattern than male patients, whereas men were more likely to have ischemic heart disease, prior coronary revascularization, and renal dysfunction than women. The patients were well treated with appropriate cardiac medications, but there were some significant differences in medication utilization between women and men (Table 1).

The primary end point of heart failure or death (whichever came first) occurred in 376 patients: 29 of 275 (11%) in women with CRT-D, 51 of 178 (29%) in women with ICD, 159 of 814 (20%) in men with CRT-D, and 137 of 553 (25%) in men with ICD. These end points included 54 deaths and 322 heart failure events. Kaplan-Meier estimates of the probability of the primary end point in women and men with CRT-D and ICD therapy are shown in (Figure 1). Overall, women receiving CRT-D therapy had a significantly better outcome than women receiving ICD therapy and men receiving ICD or CRT-D therapy during an average follow-up duration of 2.4 years.

The hazard ratios for CRT-D to ICD therapy in women and men for heart failure or death (whichever came first), heart failure only, or death at any time in the overall study population as well as by disease etiology, QRS duration, and left bundle branch block (LBBB) conduction disturbance are presented in (Table 2). For all 3 outcomes, women achieved better results from CRT-D therapy than men, with significant differences between the sexes (interaction p < 0.05) for all 3 outcomes in the total population and for 2 of the 3 outcomes in those with nonischemic cardiomyopathy, QRS <150 ms, and LBBB (Table 2). The CRT-D therapy was associated with significantly reduced mortality in women (p = 0.02) but not in men in the total population ((Table 2) and Figure 2), in women with QRS ≥150 ms, and in women with LBBB conduction disturbance (Table 2). In these 3 mortality analyses, the sex-by-treatment interactions were significant at p < 0.05 (Table 2).

We explored for treatment interactions in prespecified sex-related subgroups. In patients with nonischemic cardiomyopathy and QRS ≥150 ms, the hazard ratio for heart failure or death was 0.28 (p = 0.001) for women and 0.72 (p = 0.28) for men, with a significant p = 0.047 interaction. In patients with nonischemic cardiomyopathy and LBBB, the hazard ratio for heart failure or death was 0.22 (p < 0.001) for women and 0.73 (p = 0.24) for men, with a significant p < 0.005 interaction. The interaction analyses for the end point of death in these at-risk subgroups were not significant.

The changes in echocardiographic volumes and ejection fraction parameters between baseline and 1-year follow-up for women and men in the total population and by disease etiology, QRS duration, and LBBB are presented in (Table 3). All echocardiography parameters improved to a significantly greater degree with CRT-D therapy than with ICD therapy within both the female group and the male group (p < 0.001). Women had consistently greater improvements in reverse cardiac remodeling with CRT-D therapy than did men, with the most significant differences evident in the total population and in patients with QRS ≥150 ms or LBBB (Table 3).

Adverse events were more frequent among the CRT-D–treated patients (11%) than in the ICD-treated patients (4.5%). Women had an overall higher likelihood of all device-related adverse events than men, 10.5% versus 7.9%, respectively (p = 0.001). Women were more likely to have pneumothorax (3% in women vs. 0.73% in men), but men were more likely to have lead dislodgement (1.7% in women vs. 3.2% in men).

In this substudy from the MADIT-CRT trial, 25% of the study population was female. Although men received significant benefit from CRT-D therapy, women had significantly better results with CRT-D therapy than men for death or heart failure (whichever came first), for heart failure only, and for death at any time. Women had a significant 72% reduction in all-cause mortality in the total population, with even greater reductions in mortality for those with QRS ≥150 ms or with LBBB, with significant sex-by-treatment interactions in the 3 patient groups for the mortality end point.

There were significant differences in baseline characteristics between women and men that could have contributed in part to the observed findings. A greater proportion of the female cohort had a substrate of nonischemic cardiomyopathy and an underlying LBBB pattern, but the percentage of patients with QRS ≥150 ms and the mean QRS durations were not significantly different between women and men. Women had a higher utilization of beta-blockers than men. Men had a greater proportion of ischemic cardiomyopathy and a history of atrial fibrillation and renal dysfunction compared with women. The latter 2 clinical characteristics have been associated with poor prognosis and higher risk of death in the MADIT-2 study population (6). A higher proportion of men had a right bundle branch block pattern, and an association between the presence of right bundle branch block and poor outcome with CRT has been noted (7).

In heart failure studies, women, especially those with nonischemic heart disease, have been shown to have an overall survival advantage (8), and other groups have shown that women achieve greater reduction in left ventricular volumes and improvement in LVEF than men after CRT therapy (9). No prior study has demonstrated a significantly greater benefit from device therapy for women than men regarding mortality or cardiac-related outcomes in an overall study population or by disease etiology. In the current MADIT-CRT analysis, CRT-D:ICD hazard ratios were significantly better for women than for men for all 3 end points in the total population and for 2 of the 3 end points in those with nonischemic cardiomyopathy (see sex-by-treatment interaction p values in Table 2). Women with nonischemic cardiomyopathy were uniquely responsive to CRT relative to men, and the reason for this sex-related beneficial effect is unclear. It is possible that among patients with heart disease, the risk of heart failure is greater for women than for men, resulting in a greater benefit from preventive CRT-D therapy in women.

It is generally appreciated that in subjects without heart disease, women have, on average, approximately 10 ms shorter QRS durations than men (10). In subjects with heart failure, prolongation of QRS ≥120 ms occurs in 14% to 47% of patients overall (11). In the MADIT-CRT trial, we used the same entry criteria of QRS ≥130 ms for both women and men. Thus, for any given QRS duration ≥130 ms, women might have, on a relative basis, more conduction disturbance and greater cardiac dyssynchrony than men, and this might explain why women were more responsive to cardiac resynchronization therapy than men in this trial. It is interesting that LBBB was present in 70% of the MADIT-CRT patients and in 87% of female patients. Although both sexes with LBBB benefited from CRT-D therapy, women with LBBB achieved a significantly better result with this therapy than did men with LBBB (Table 2).

Several major CRT trials involved patients with NYHA functional class III and IV heart failure (9,12- 14). In the MIRACLE (Multicenter InSync Randomized Clinical Evaluation) study (14), 32% of the study group were women; and among women receiving CRT, there was a reduction in heart failure hospitalization or death compared with the control group (hazard ratio: 0.157, p = 0.002). No differences were seen among men for either end point with CRT, even when accounting for baseline demographics and heart failure etiology (14). In the COMPANION (Comparison of Medical Therapy, Pacing, and Defibrillation in Heart Failure) trial, women made up 33% of the study population, and these patients had a 56% reduction in the risk of sudden cardiac death with CRT-D compared with optimal pharmacologic therapy, with female sex associated with reduced risk in conjunction with CRT-D therapy (hazard ratio: 0.56, p = 0.003) (9). When CRT or CRT-D therapy was compared to optimal pharmacologic therapy in COMPANION, these therapies were similarly beneficial in reducing mortality in women and men, and there were no significant sex-by-treatment interactions (15). In the CARE-HF (Cardiac Resynchronization in Heart Failure) study, about 25% of the study group were women, but the CRT to medical therapy hazard ratios for women and men for the primary end point (death or hospitalization) were similar, in the 0.62 to 0.64 range (12). In the recently reported European cohort of the REVERSE (Resynchronization Reverses Remodeling in Systolic Left Ventricular Dysfunction) trial involving 162 patients with NYHA functional class I and II heart failure, LVEF ≤0.35, and QRS ≥120 ms, approximately 20% of the subjects were female, and the clinical composite end point of worsened heart failure was reduced to a similar degree with CRT therapy in women and men (16). This REVERSE trial had higher LVEF and less prolonged QRS duration criteria for enrollment than the MADIT-CRT study did, and there was limited power to find sex differences in outcome in view of the small number of patients in the trial.

Women in the MADIT-CRT trial obtained significantly greater reductions in death or heart failure (whichever came first), heart failure alone, and all-cause mortality with CRT-D therapy than men. These more favorable results for women were associated with consistently greater echocardiographic evidence of reverse cardiac remodeling in women than in men.