Sex Differences in Clinical Characteristics and Prognosis in a Broad Spectrum of Patients With Heart Failure
Results of the Candesartan in Heart failure: Assessment of Reduction in Mortality and morbidity (CHARM) Program
Background— We wished to test previous hypotheses that sex-related differences in mortality and morbidity may be due to differences in the cause of heart failure or in left ventricular ejection fraction (LVEF) by comparing fatal and nonfatal outcomes in women and men with heart failure and a broad spectrum of left ventricular ejection fraction.
Methods and Results— We compared outcomes in 2400 women and 5199 men randomized in the Candesartan in Heart failure: Assessment of Reduction in Mortality and morbidity (CHARM) program using multivariable regression analyses. A total of 1188 women (50%) had a low LVEF (≤0.40), and 1212 had a preserved LVEF (>0.40). Among the men, 3388 (65%) had a low LVEF, and 1811 had a preserved LVEF. A total of 1216 women (51%) and 3465 men (67%) had an ischemic cause of their heart failure. All-cause mortality was 21.5% in women and 25.3% in men (adjusted hazard ratio [HR], 0.77; 95% CI, 0.69 to 0.86; P<0.001). Fewer women (30.4%) than men (33.3%) experienced cardiovascular death or heart failure hospitalization (adjusted HR, 0.83; 95% CI, 0.76 to 0.91; P<0.001). The risks of sudden death (HR, 0.70; 95% CI, 0.58 to 0.85) and death due to worsening heart failure (HR, 0.72; 95% CI, 0.58 to 0.89) were reduced to a comparable extent. The adjusted risk of cardiovascular hospitalization was also lower in women (HR, 0.88; 95% CI, 0.82 to 0.95), mainly because of a reduced risk of heart failure hospitalization (HR, 0.87; 95% CI, 0.78 to 0.97). Women had a lower risk of death irrespective of cause of heart failure or LVEF.
Conclusions— Among patients with heart failure, women have lower risks of most fatal and nonfatal outcomes that are not explained, as previously suggested, by LVEF or origin of the heart failure.
Received November 1, 2006; accepted April 8, 2007.
Women with heart failure (HF) have a better age-adjusted survival rate than men with the same condition.1,2 There is, however, little consensus about the explanation for this difference.3 Two main hypotheses have been advanced.1–3 Average left ventricular ejection fraction (LVEF) is higher in women than in men with HF, and a higher LVEF is associated with a better survival rate; however, that hypothesis has not been fully explored within any large trial. All of the trials examined so far enrolled patients with a low LVEF (≤0.40) and therefore could only examine the association between sex and outcome over a limited range of LVEF.4–9 Moreover, attributing differences in outcome to differences in LVEF is not straightforward, because sex-related differences in LVEF probably reflect differences in cause of heart failure (ie, hypertension is more common and an ischemic cause is less common in women), age (women with HF are older than men, and preserved LVEF HF is more common in older than in younger patients), and possibly remodeling. Potential interactions between sex, LVEF, and age must be considered to properly interpret these prior data.
Indeed, a second hypothesis explaining the better survival in women relates to the origin of the HF. Two prominent studies suggested that in patients with nonischemic HF, women had a better survival than men (whereas there was no advantage in women with an ischemic cause).4,8 Coronary artery disease is more often identified as the cause of HF in men and is associated with a worse prognosis.1–3
In prior studies, the focus has been on all-cause mortality. There are additional questions about other outcomes that are relevant to this issue. For example, is the difference in mortality between the sexes explained by a difference in any particular mode of death? Are there also differences in nonfatal outcomes, particularly hospital admissions? Those may not be so clearly related to LVEF.
The investigation of outcomes in women with HF has, to date, been limited by the smaller proportion of women (typically around 20% to 25%) than men enrolled in trials.1–3 The main trials in which the relationship between sex and outcome was examined also had a relatively short follow-up; 3 were studies of a β-blocker, and the longest median follow-up was 2 years.4–9 Both of these factors resulted in a small absolute number of events in the studies mentioned.4–9 A higher proportion (32%) and absolute number (2400) of women with HF were enrolled in Candesartan in Heart failure: Assessment of Reduction in Mortality and morbidity (CHARM) than in any previously published placebo-controlled trial (although the proportion of women was still less than in the general population, which probably is a reflection of the lower average age of patients in CHARM and the fact that 2 of the 3 component trials enrolled patients with an LVEF of ≤0.40).10 This, and the longer median follow-up of 38 months, resulted in a large absolute number of clinical events. The other main limitation to the study of sex-related outcomes in HF has been the restricted range of LVEF permitted in previous trials. Patients with a full spectrum of LVEF were enrolled in the CHARM trials. Consequently, CHARM offered a unique opportunity to compare mortality, mode of death, and hospitalizations in men and women with HF.
The CHARM Program
The design, baseline findings, and primary results of CHARM have been reported in detail.10–12 Briefly, it consisted of 3 independent but related trials in which patients with New York Heart Association (NYHA) class II to IV HF were randomized to placebo or candesartan. Patients were enrolled in the individual CHARM trials according to LVEF and treatment with an angiotensin-converting enzyme (ACE) inhibitor. Patients with an LVEF ≤0.40 who were intolerant of an ACE inhibitor were enrolled in CHARM-Alternative (n=2028, 21% women), whereas patients with an LVEF ≤0.40 who were taking an ACE inhibitor were enrolled in CHARM-Added (n=2548, 32% women). Patients with an LVEF >0.40 were randomized in CHARM-Preserved (n=3023, 40% women). In CHARM-Added, patients in NYHA class II had to have had a hospital admission for a cardiac reason in the previous 6 months (this increased the proportion of NYHA class III/IV patients in CHARM-Added). Patients with a serum creatinine ≥3 mg/dL (265 μmol/L) were excluded. CHARM was completed, as planned, 2 years after the last patient was randomized. Because the rate of recruitment varied between the CHARM trials, overall follow-up ranged from a median of 41 months in CHARM-Added to 37 months in CHARM-Preserved and 34 months in CHARM-Alternative (38 months in the overall CHARM program). The primary outcome for each of the 3 component trials was the composite of death due to a cardiovascular cause or unplanned admission to the hospital for the management of worsening HF, and in the overall program, death due to any cause.
Summary statistics of an extensive list of baseline characteristics, including demographics, history and origin of HF, comorbidity, body mass index, vital signs, clinical signs and symptoms of HF, ECG findings, and HF medications, were analyzed by sex. Differences between men and women in those baseline characteristics were compared by individual linear regression for continuous variables (presented as mean differences) and logistic regression analysis for categorical factors (presented as odds ratios).
Cox proportional hazard models were fitted to assess the impact of sex on the main outcomes in the CHARM trials. Those models included the same predictor variables established for the entire CHARM population, ie, diabetic status, body mass index, NYHA class, current smoking status, bundle-branch block, cardiomegaly, prior HF hospitalization, diastolic blood pressure, diagnosis of HF >2 years, previous myocardial infarction, dependent edema, heart rate, pulmonary crackles, pulmonary edema, mitral regurgitation, atrial fibrillation, rest dyspnea, and randomized treatment (candesartan or placebo).13 To establish whether any differences between men and women depended on other characteristics, interactions between sex and LVEF (used as a continuous variable), age, and component trial were assessed for each outcome. Kaplan-Meier survival curves are presented by sex. The survival curves have been standardized to the median level of risk calculated from the coefficients of the predictor variables included in the model. Because the comparison was for women compared with men, an adjustment was in order to ensure that the underlying risk was independent of sex. Kaplan-Meier curves are also presented by primary cause of HF. For these curves, the median level of risk was calculated separately by ischemic and nonischemic origin.
We formally tested whether there was an interaction between sex and ischemic origin using the Cox model. We examined for this possible interaction using ischemic heart disease defined by the investigator as the primary cause of HF. We also performed supportive analyses using various other definitions of ischemic heart disease: (1) ischemic heart disease defined by the investigator as a contributing cause, (2) a history of prior myocardial infarction, and (3) a history of prior myocardial infarction, angina, or coronary revascularization. Finally, we performed additional analyses, adjusting for the treatments for HF recorded at baseline, to examine whether this could explain any observed sex differences.
The authors had full access to and take full responsibility for the integrity of the data. All authors have read and agree to the manuscript as written.
Of the 2400 women enrolled in the CHARM program, 1188 (50%) had a low LVEF (≤0.40, ie, were in CHARM-Alternative or CHARM-Added), and 1212 (50%) had a preserved LVEF. Of the 5199 men enrolled in the CHARM program, 3388 (65%) had a low LVEF, and 1811 had a preserved LVEF.
Sex and Baseline Characteristics
Demographics, Clinical Measurements, ECG Findings, and Medical History
Women were a mean of 3.4 years older than men. Heart rate, systolic blood pressure, and pulse pressure were higher in women. Women had less evidence of ischemic heart disease (angina, previous myocardial infarction, percutaneous coronary intervention, or coronary artery bypass grafting). See Table 1 for more details.
HF Characteristics and Treatment
HF characteristics and treatment are shown in Table 2. The median duration of HF was 23.1 (interquartile range 6.5 to 57.6) months in women and 26.0 (interquartile range 7.5 to 63.2) months in men. Ischemic origin was more common in men than in women (67% versus 51%). Hypertensive origin was more common in women than in men (21% versus 9%), as was a valvular origin (4% versus 2%). Women had more signs and symptoms of HF, although there was no difference in NYHA class. There were differences in the treatments for HF at baseline; a lower proportion of women received a β-blocker, an ACE inhibitor, and digitalis, but a higher proportion received a diuretic. Spironolactone was taken by comparable proportions of women and men. A lower proportion of women than men received aspirin or a lipid-lowering drug. Hormone replacement therapy was taken by 16% of women.
Sex and Clinical Outcomes
Sex and Mortality
Of the 2400 women enrolled, 515 (21.5%) died; of the 5199 men randomized, 1316 (25.3%) died (unadjusted hazard ratio [HR], 0.85; 95% CI, 0.76 to 0.94; P=0.001). Adjustment for other prognostic variables (listed in the footnote to Table 3) accentuated the difference so that women had a 23% lower adjusted risk of death than men (adjusted HR, 0.77; 95% CI, 0.69 to 0.86; P<0.001; Figure 1). A major factor influencing this further reduction in HR was age (Table 3). Because increasing age was strongly associated with increasing risk of mortality, and women were a mean of 3.4 years older than men, adjustment for age lowered the HR further from 1 (Figure 2). The impact of sex on all-cause mortality was similar in all 3 component trials, after adjustment for other prognostic variables (adjusted HR in CHARM-Alternative, 0.73 [95% CI, 0.61 to 0.88]; adjusted HR in CHARM-Added, 0.79 [95% CI, 0.66 to 0.95]; and CHARM-Preserved adjusted HR, 0.77 [95% CI, 0.64 to 0.93]; P for interaction between sex and study=0.84).
Cardiovascular Death or HF Hospitalization
Of the 2400 women enrolled, 730 (30.4%) experienced the primary outcome of cardiovascular death or HF hospitalization; of the 5199 men enrolled, 1730 (33.3%) experienced that outcome (unadjusted HR, 0.91; 95% CI, 0.83 to 0.99; P=0.032). As with all-cause mortality, adjustment for other prognostic variables accentuated the difference so that women had a 17% lower adjusted risk of this composite outcome than men (adjusted HR, 0.83; 95% CI, 0.76 to 0.91; P<0.001; Table 3). The influence of sex on this outcome was similar in each CHARM component trial (adjusted HR in CHARM-Alternative, 0.78 [95% CI, 0.67 to 0.92]; in CHARM-Added, 0.82 [95% CI, 0.70 to 0.95]; and in CHARM-Preserved, 0.86 [95% CI, 0.74 to 1.00]; P for interaction between sex and component trial=0.70).
Addition of medications (β-blockers, ACE inhibitors, amiodarone, spironolactone, digitalis, diuretics, lipid-lowering therapy, and aspirin) to the multivariable model had little effect on the relationship between sex and mortality (the HR for female sex decreased from 0.77 to 0.76; data not shown) or on the relationship between sex and cardiovascular death or HF hospitalization (the HR decreased from 0.83 to 0.81). The adjusted HRs for women taking hormone replacement therapy versus those who were not were 0.83 (95% CI, 0.63 to 1.08; P=0.17) for all-cause mortality and 0.99 (95% CI, 0.80 to 1.22; P=0.94) for cardiovascular death or HF hospitalization:.
Clinical Outcomes: Interaction Between Sex and Cause of HF
Figure 3 shows mortality in men and women with an ischemic (Figure 3A) and nonischemic (Figure 3B) cause of HF. Mortality was higher in men than in women in both etiologic groups (P for interaction=0.36). The same was also the case for cardiovascular death or HF hospitalization (P=0.55). Our supportive analyses gave similar results.
Clinical Outcomes: Interaction Between Sex and LVEF
LVEF in women was a mean of 6.3% higher than in men (Table 2). As shown in Figure 4, however, the better survival in women persisted after adjustment for LVEF and other important predictors of mortality, particularly age (the mean ages in the lowest and highest LVEF approximate quintiles were 65.6 and 69.6 years in women and 63.8 and 65.2 years in men, respectively). The better outcome in women was also seen for cardiovascular death or HF hospitalization (Table 3; Figure 4B). There was no evidence of an interaction between sex and LVEF (P for interaction for death=0.65; P for cardiovascular death or hospitalization for HF=0.73; Table 3). There was also no evidence of an interaction between sex and LVEF for HF hospitalization alone (see below).
Sex and Mode of Death in HF
Of all deaths, the proportion adjudicated as having a cardiovascular cause was 81% (1060/1316) in men and 78% (400/515) in women. Women had a lower overall adjusted risk of cardiovascular death (HR, 0.78; 95% CI, 0.69 to 0.88; P<0.001; Table 4). Sudden death accounted for the largest proportion of cardiovascular deaths in both men (489/1060, 46%) and women (154/400, 39%), and women had a lower risk than men (adjusted HR, 0.70; 95% CI, 0.58 to 0.85; P<0.001). The proportion of cardiovascular deaths due to progressive HF was 32% (336/1060) in men and 33% (133/400) in women, and women had a lower risk than men (adjusted HR, 0.72; 95% CI, 0.58 to 0.89; P=0.002).
Sex and Hospital Admissions
Rates of first hospital admission per 1000 patient-years of follow-up are shown in Table 4. Women were significantly less likely to experience a cardiovascular admission than men (rate 58.4 versus 71.7 per 1000 patient-years; adjusted HR, 0.78; 95% CI, 0.69 to 0.88; P<0.001), mainly because of a lower risk of HF hospitalization. Overall, however, the adjusted risk for all-cause hospitalization differed little between women and men (220.5 versus 238.0 per 1000 patient-years; adjusted HR, 0.88; 95% CI, 0.82 to 0.95; P=0.001) because there was no difference between men and women in rates of noncardiovascular hospitalization. There was no evidence of an interaction between sex and LVEF for HF hospitalization (interaction P=0.54; footnote to Table 3).
Effect of Candesartan According to Sex
The reduction in cardiovascular death or HF hospitalization associated with the use of candesartan was identical in men and women in CHARM overall (P for interaction=0.89), as was the effect of candesartan on all-cause mortality (P for interaction=0.98).
Given the uniquely broad spectrum of patients randomized in CHARM compared with prior trials, we were able to examine the effect of sex on clinical outcomes in HF, taking account of age, cause of HF, and LVEF. We found that there were important differences in several patient characteristics between men and women, especially age and LVEF. After adjustment for these and other prognostic variables (of which age was of major importance), women had a better outcome than men. This could not be explained by differences in either cause of HF or LVEF, as previously suggested.
Although patients with nonischemic HF had a better survival than those with ischemic HF, women had better outcomes than men, irrespective of the cause of HF. This finding is important in the light of conflicting observations from prior studies. The first clinical trial to suggest a sex-etiology interaction was the Flolan International Randomized Survival Trial (FIRST).4 In FIRST, 471 patients (112 women) with severe HF were randomized, and a stratified analysis was performed according to cause of HF. In patients with a nonischemic cause, the relative risk of death for men compared with women was 3.08 (95% CI, 1.56 to 6.09; P=0.001). In patients with an ischemic cause (67 men and 44 women), the relative risk was 1.64 (95% CI, 0.87 to 3.09; P=0.127). A formal interaction test of the relationship between cause of HF and the association between sex and outcome, however, was not significant (P=0.275). The much larger Beta-blocker Evaluation of Survival Trial (BEST) randomized 2708 patients (593 women) with mainly NYHA class III HF.8 In BEST, the relationship between sex and survival was modified by cause as in FIRST. In the nonischemic group, men (n=756) had a significantly worse survival than women (n=365; P<0.01). In the ischemic group, there was a trend for a better survival in men (n=1359) than in women (n=228; P=0.152). A significant coronary heart disease–sex interaction was identified (P=0.011). Although those findings are supported by an observational study,14 a sex-etiology interaction was not found in the Metoprolol Extended-Release Randomized Intervention Trial in Heart Failure (MERIT-HF),7 and in the second Cardiac Insuffi6ciency Bisoprolol Study trial (CIBIS-2),5 there was a trend to a sex-etiology interaction in the opposite direction, ie, the survival advantage of women over men was more marked in patients with an ischemic compared with a nonischemic cause of HF. We believe that CHARM (with 5199 men [3465 with an ischemic cause of HF] and 2400 women [1216 with an ischemic cause of HF]) offers greater statistical certainty about whether there is a sex-etiology interaction in relation to outcome given the much larger number of female deaths (n=515) than in the prior trials (BEST, n=163; MERIT-HF, n=64; and CIBIS-2, n=53; the number in FIRST was not reported but was small).
We also addressed the hypothesis that women have a better survival than men because of a higher LVEF. Although women did have a higher mean LVEF than men in CHARM, they were also, on average, older. Their adjusted risk of death and of the composite of cardiovascular death or HF hospitalization was lower across the spectrum of LVEF. Furthermore, no interaction between sex and LVEF was apparent on formal statistical testing.
We also examined mode of death according to sex in CHARM. Although this was also addressed in CIBIS-2, there were only 53 female deaths in total in that trial, of which 14 were unclassified, which precludes any definitive conclusions.5 Women in CHARM had a lower adjusted risk of both cardiovascular and noncardiovascular death, with both reduced to a similar extent. Women’s risk of the 2 main modes of cardiovascular death in HF (pump failure and sudden death) was also similarly lower than in men.
We also examined a range of other nonfatal outcomes that were not analyzed in detail in prior trials. Women were less likely to experience an admission for a cardiovascular reason than men, irrespective of LVEF, because of a lower risk of HF hospitalization. Two prior trials did describe unadjusted rates of HF hospitalization for men and women separately. In the Digitalis Investigators Group trial (DIG), 34.4% of women in the placebo group experienced HF hospitalization compared with 34.7% of men, but these proportions were not adjusted for other variables that might influence hospitalization.6 In the Valsartan Heart Failure Trial (Val-HeFT), 19% of women and 15.5% of men (P=0.017) experienced HF hospitalization, although when the longer survival of women was taken into account, the rate of HF hospitalization did not differ between the sexes, and adjustment for other predictive variables was not performed.9 We found that women overall had a 13% lower adjusted risk of HF hospitalization than men, and this risk was also less in women with low LVEF (Table 3). The crude rate of HF hospitalization was higher in women than in men with preserved LVEF, but this difference disappeared after adjustment for other factors affecting this outcome.
Although we found that the better outcome in women than men with HF is not explained by variables suggested previously, we have not discovered an alternative explanation for this differential survival. Indeed, the difference may not specifically or wholly reflect a cardiovascular factor, because we, notably, found that women had a significantly lower noncardiovascular mortality, as well as cardiovascular mortality. Similarly, the comparable proportional reductions in the risk of sudden death and death from worsening HF in women do not point to a specific protective cardiac mechanism, although pump function and electrical stability are clearly linked. In that respect, investigator-reported LVEF represents only a crude summary measure of cardiac function that does not take account of other aspects of cardiac remodeling (such as differential changes in gene expression, muscle hypertrophy, extracellular matrix deposition, and chamber geometry), diastolic ventricular function, and other functional changes of prognostic importance, such as the presence of mitral regurgitation. It may be useful, therefore, for future studies of sex-related differences in HF to make a detailed characterization of cardiac structure, function, and, where possible, cellular and molecular biology. Another limitation of the present study was the absence of neurohumoral and other biomarkers, which may also differ between men and women with HF, although existing studies have given conflicting findings in this respect.8,9
Laboratory variables that reflect known markers of prognosis in HF may differ between men and women, eg, estimated glomerular filtration rate (which may be lower in women than in men, even though estimated glomerular filtration rate equations account for lower muscular mass in women) and hemoglobin.15–17 These were not included in the multivariable model in the present study because they were only available in a subset of patients. Differential compliance does not explain the female survival advantage, because women in CHARM showed worse adherence to therapy than men.18 Differences between men and women in mental health (eg, depression or self-reported quality of life, which was different in a substudy of men and women in CHARM19) and life circumstances (eg, employment status and domestic circumstances [partner, family, home versus long-term care]) may also be important but are rarely studied in detail.
A further limitation of the present report is that it describes a clinical trial cohort. Our basic findings, however, agree with those from large observational cohorts and add to them because of the more extensive baseline characterization, reporting of nonfatal outcomes, and adjudication of mode of death in CHARM.20–22
In conclusion, we found that among patients with HF, women had lower risks of most fatal and nonfatal outcomes, which were not explained by either cause of the HF or LVEF. Specifically, the risk of both noncardiovascular and cardiovascular death was lower in women. The risks of the 2 main types of cardiovascular death were also lower in women. Although the rate of all-cause hospitalization was comparable between sexes, the risk of cardiovascular hospitalization was lower in women, mainly because of a lesser risk of HF hospitalization. Why women have a better outcome in HF than men is unknown and is not explained by variables suggested previously.
Source of Funding
The CHARM program was funded by AstraZeneca which was responsible for data collection.
The data analysis for this manuscript was performed independently by Tim Clayton and Dr Pocock. The Executive Committee (Drs Pfeffer, Swedberg, Granger, McMurray, and Yusuf) supervised the management of the study and were primarily responsible for the interpretation of the data and review and approval of the manuscript. Dr Michelson is an employee of AstraZeneca. All other authors have received research grants, honoraria for lectures and/or consulting fees from AstraZeneca.
Guest Editor for this article was Edgardo Escobar, MD.
Petrie MC, Dawson NF, Murdoch DR, Davie AP, McMurray JJ. Failure of women’s hearts. Circulation. 1999; 99: 2334–2341.
Adams KF Jr, Sueta CA, Gheorghiade M, O’Connor CM, Schwartz TA, Koch GG, Uretsky B, Swedberg K, McKenna W, Soler-Soler J, Califf RM. Gender differences in survival in advanced heart failure: insights from the FIRST study. Circulation. 1999; 99: 1816–1821.
Simon T, Mary-Krause M, Funck-Brentano C, Jaillon P. Sex differences in the prognosis of congestive heart failure: results from the Cardiac Insufficiency Bisoprolol Study (CIBIS II). Circulation. 2001; 103: 375–380.
Ghali JK, Pina IL, Gottlieb SS, Deedwania PC, Wikstrand JC; MERIT-HF Study Group. Metoprolol CR/XL in female patients with heart failure: analysis of the experience in Metoprolol Extended-Release Randomized Intervention Trial in Heart Failure (MERIT-HF). Circulation. 2002; 105: 1585–1591.
Majahalme SK, Baruch L, Aknay N, Goedel-Meinen L, Hofmann M, Hester A, Prescott MF, Feliciano N; Val-HeFT Study Investigators. Comparison of treatment benefit and outcome in women versus men with chronic heart failure (from the Valsartan Heart Failure Trial). Am J Cardiol. 2005; 95: 529–532.
Pfeffer MA, Swedberg K, Granger CB, Held P, McMurray JJ, Michelson EL, Olofsson B, Ostergren J, Yusuf S, Pocock S; CHARM Investigators and Committees. Effects of candesartan on mortality and morbidity in patients with chronic heart failure: the CHARM-Overall programme. Lancet. 2003; 362: 759–766.
McMurray J, Ostergren J, Pfeffer M, Swedberg K, Granger C, Yusuf S, Held P, Michelson E, Olofsson B; CHARM committees and investigators. Clinical features and contemporary management of patients with low and preserved ejection fraction heart failure: baseline characteristics of patients in the Candesartan in Heart failure-Assessment of Reduction in Mortality and morbidity (CHARM) programme. Eur J Heart Fail. 2003; 5: 261–270.
Pocock SJ, Wang D, Pfeffer MA, Yusuf S, McMurray JJ, Swedberg KB, Ostergren J, Michelson EL, Pieper KS, Granger CB. Predictors of mortality and morbidity in patients with chronic heart failure. Eur Heart J. 2006; 27: 65–75.
Hillege HL, Nitsch D, Pfeffer MA, Swedberg K, McMurray JJ, Yusuf S, Granger CB, Michelson EL, Ostergren J, Cornel JH, de Zeeuw D, Pocock S, van Veldhuisen DJ; Candesartan in Heart Failure: Assessment of Reduction in Mortality and Morbidity (CHARM) Investigators. Renal function as a predictor of outcome in a broad spectrum of patients with heart failure. Circulation. 2006; 113: 671–678.
O’Meara E, Clayton T, McEntegart MB, McMurray JJ, Lang CC, Roger SD, Young JB, Solomon SD, Granger CB, Ostergren J, Olofsson B, Michelson EL, Pocock S, Yusuf S, Swedberg K, Pfeffer MA; CHARM Committees and Investigators. Clinical correlates and consequences of anemia in a broad spectrum of patients with heart failure: results of the Candesartan in Heart Failure: Assessment of Reduction in Mortality and Morbidity (CHARM) Program. Circulation. 2006; 113: 986–994.
Granger BB, Swedberg K, Ekman I, Granger CB, Olofsson B, McMurray JJ, Yusuf S, Michelson EL, Pfeffer MA; CHARM Investigators. Adherence to candesartan and placebo and outcomes in chronic heart failure in the CHARM programme: double-blind, randomised, controlled clinical trial. Lancet. 2005; 366: 2005–2011.
Lewis EF, Lamas GA, O’Meara E, Granger CB, Dunlap ME, McKelvie RS, Probstfield JL, Young JB, Michelson EL, Halling K, Carlsson J, Olofsson B, McMurray JJ, Yusuf S, Swedberg K, Pfeffer MA; CHARM Investigators. Characterization of health-related quality of life in heart failure patients with preserved versus low ejection fraction in CHARM. Eur J Heart Fail. 2007; 9: 83–91.
Gustafsson F, Torp-Pedersen C, Burchardt H, Buch P, Seibaek M, Kjoller E, Gustafsson I, Kober L; DIAMOND Study Group. Female sex is associated with a better long-term survival in patients hospitalized with congestive heart failure. Eur Heart J. 2004; 25: 129–135.
Galvao M, Kalman J, DeMarco T, Fonarow GC, Galvin C, Ghali JK, Moskowitz RM. Gender differences in in-hospital management and outcomes in patients with decompensated heart failure: analysis from the Acute Decompensated Heart Failure National Registry (ADHERE). J Card Fail. 2006; 12: 100–107.