Author + information
- Henry Krum, MBBS, PhD∗ ( and )
- Ingrid Hopper, MBBS
- Centre of Cardiovascular Research and Education in Therapeutics, School of Public Health and Preventive Medicine, Monash University, Melbourne, Australia
- ↵∗Reprint requests and correspondence:
Prof. Henry Krum, Centre of Cardiovascular Research and Education in Therapeutics, School of Public Health and Preventive Medicine, Monash University, 99 Commercial Road, Melbourne, Victoria 3004, Australia.
Advances in heart failure (HF) drug and device therapies over the past 3 decades have made major inroads into the lethality of this disease. Angiotensin-converting enzyme (ACE) inhibitors and angiotensin receptor blockers (ARBs), beta-blockers, mineralocorticoid receptor antagonists (MRAs), as well as device-based approaches including cardiac resynchronization therapy and implantable cardioverter defibrillator therapy, have resulted in substantial mortality, morbidity and quality of life (QoL) benefits to such patients, particularly those with HF and reduced ejection fraction (HFREF).
Beta-blockers are arguably the most potent therapy in reducing mortality in HF. Mortality reductions in mild to moderate as well as advanced HFREF in the pivotal trials additional to background ACE inhibitors and diuretics were consistently of the order of 30% (1–3). This was accompanied by improvements in cardiac remodeling parameters, HF symptoms, and QoL measures.
In this issue of JACC: Heart Failure, Rush et al. (4) have reviewed the totality of HF randomized controlled trials in the past 3 decades, evaluating cardiovascular (CV) mortality according to beta-blocker usage in the trials. The authors have been meticulous in including all of the major, predominantly HFREF, randomized control trials over this period. The analysis included 66 trials, including 136,182 participants and 32,140 deaths, with participants mostly with New York Heart Association (NYHA) functional class II and III symptoms and a weighted average left ventricular ejection fraction of 27%. The trials were divided into 3 groups according to the proportion of patients treated with a beta-blocker. The proportion of CV deaths decreased from 87% with low beta-blocker use to 80% with high beta-blocker use. Non-CV deaths rose from 11.4% to 19.1% with high beta-blocker therapy, representing a proportional increase of two-thirds in non-CV deaths. The reduction in CV mortality was associated with a rise in non-CV deaths, which was due mostly to malignancy.
This analysis confirms what we have known for some time, which is that mortality rates are falling with modern HF treatment; these data allow us to go some way in quantifying the major therapeutic advances that have been made in this field. This analysis examines background beta-blocker use within these trials, and as the authors acknowledge, concomitant with increased beta-blocker use over recent years has been increased use of ACE inhibitor/ARB, rapid uptake of MRAs following the RALES study (Randomized Aldactone Evaluation Study) (5) and the EMPHASIS-HF study (Eplerenone in Mild Patients Hospitalization and Survival Study in Heart Failure) (6) as well as the advent of devices, and it is therefore impossible to attribute improved CV mortality (and accompanying relative increase in non-CV mortality) entirely to beta-blocker therapy alone. However, a sensitivity analysis adjusting for ACE inhibitor/ARB use and adjusting for MRA use, and also using a meta-analytic approach, demonstrated that beta-blockers contributed most to the reduction in CV deaths.
This analysis includes the individual study outcomes of the placebo group and the intervention group. Many of these interventions are of novel agents that turned out to result in neutral or even adverse clinical outcomes. Examples include the BEST The Beta-Blocker Evaluation of Survival Trial study (7), in which bucindolol failed to improve survival in NYHA functional class III and IV HF, and the GISSI-HF (Gruppo Italiano per lo Studio della Sopravvivenza nell'Infarto miocardico study) (8), in which rosuvastatin failed to improve survival in NYHA functional class II to IV HF. By definition, none of these interventions constitute standard therapy and thus their inclusion in the analysis may potentially skew the results. An analysis examining just the placebo group would be informative.
The falls in CV mortality have had consequences for the design of HF clinical trials. As the authors note, all-cause mortality has become an increasingly insensitive endpoint in HF clinical trials. Although this is an easily definable and confirmable endpoint, it has the obvious weakness that causes of death such as trauma and malignancy that are unrelated to HF (or its treatment) are included. More recent trials have required a more refined approach to capturing outcomes of most interest and over the past 15 years, HF trials have used CV mortality rather than all-cause mortality as at least a component of the primary endpoint. All-cause mortality should not be abandoned as an outcomes measure, because no reduction in all-cause mortality in the presence of a reduction in CV mortality may be a safety signal that the treatment is shifting the cause of death. This was seen in The Digitalis Investigation Group trial (9), in which symptomatic HF decreased with reduced HF hospitalizations, but no reduction in mortality was seen due to a presumed rise in arrhythmic deaths.
CV mortality allows greater precision with regard to evaluation of the benefits or otherwise of a HF therapeutic agent, but practically, it is expensive because independent adjudication committees are required to review the medical records and determine cause of death. There is also residual uncertainty about the cause of death in the absence of post-mortems and when the study participant is found deceased at home with an unwitnessed death. There is uncertainty about whether these deaths should be included as CV mortality or excluded from the analysis, but as the authors found, this accounted for only 5.7% of all deaths.
The inclusion of HF hospitalizations in outcomes has received increasing interest. One of the advantages of this outcome is that it reflects costs of HF treatment, which are rapidly increasing. Multiple admissions for HF reflects poor prognosis, and this outcome can capture that aspect, and the costs associated with admissions are meaningful for health care payers. Again there are major issues with this as a clinical outcome. HF hospitalization may reflect regional practices and preferences, with significant regional variation in bed days and need for admission. Some centers use short-stay units or give outpatient intravenous diuretics in an effort to avoid “formal” admissions, which further blurs the definition of a HF hospitalization. It can also be difficult to determine the relative contribution of HF to an admission when multiple organ dysfunction exists; for example, with primary pneumonia resulting in HF decompensation.
All of these outcomes do not necessarily reflect what is important to the patient. This has been termed the “patient journey.” QoL measures have traditionally been viewed as soft science, because they are somewhat subjective and often have not been particularly well-correlated with harder mortality outcomes (10). However, for the patient, these measures may be far more meaningful than the blunt instruments described previously. Alternatives to describe the patient journey have been suggested in the literature (11–13); however, these have not been widely accepted nor stringently validated, and regulators are somewhat uncertain of their clinical utility.
HF mortality is laudably falling due in large part to the addition of beta-blockers to our clinical armamentarium. However, these major gains cannot be seen in isolation from the contribution of other therapies to improvements in HF mortality. Beta-blockers continue to be underprescribed in HF, and this continues to be a major challenge going forward. Additionally, this analysis demonstrates that improvements in HF mortality forecast increasing difficulty demonstrating improved outcomes in future trials, and work is needed to develop appropriate clinical endpoints in contemporaneous HF trials. This endpoint evaluation work is urgently needed to optimize evaluation of new treatments to reduce the still unacceptably high mortality and morbidity associated with the condition.
↵∗ Editorials published in JACC: Heart Failure reflect the views of the authors and do not necessarily represent the views of JACC: Heart Failure or the American College of Cardiology.
Drs. Krum and Hopper have reported that they have no relationships relevant to the contents of this paper to disclose.
- American College of Cardiology Foundation
- Packer M.,
- Fowler M.B.,
- Roecker E.B.,
- et al.
- Rush C.J.,
- Campbell R.T.,
- Jhund P.S.,
- et al.
- Anker S.D.,
- Agewall S.,
- Borggrefe M.,
- et al.