Systolic and Nonsystolic Heart Failure: Title and subTitle BreakEqually Serious Threats

Heart failure is a common disease with increasing prevalence, accounting for an estimated 1.1 million hospitalizations annually in the United States.¹ Despite development of effective treatments, including β-blockers, angiotensin-converting enzyme inhibitors, spironolactone, defibrillators, and cardiac resynchronization, heart failure remains a serious disease and is associated with a poor prognosis.²

Clinical heart failure can occur with reduced systolic function, typically defined as a left ventricular ejection fraction (LVEF) of less than 40% and not more than 55% or with preserved LVEF (or nonsystolic heart failure). With modern imaging techniques, LVEF is relatively easy to measure but clinical heart failure is much more difficult to define precisely. This is an important issue because use of an imprecise definition of clinical heart failure may result in many patients with preserved LVEF being incorrectly diagnosed with heart failure.³ Although heart failure with preserved LVEF is generally considered to be primarily due to diastolic dysfunction, heart failure with reduced systolic function also can include a component of diastolic dysfunction.⁴

The term diastolic dysfunction implies clinically important impairment of the left ventricular filling. Classic descriptions of this disorder derive from invasive hemodynamic measures with high-fidelity catheters.^{4 - 5} A widely accepted noninvasive gold standard for assessing left ventricular diastolic function does not exist. Doppler echocardiography has emerged as a potentially valuable tool for identifying left ventricular diastolic dysfunction with a variety of measures and patterns described as markers of delayed myocardial relaxation and impaired filling. For example, in a community-based study, Redfield et al⁶ defined diastolic dysfunction according to velocities of blood flow across the mitral valve and pulmonary vein ostia during different phases of diastole. While these measures have been shown to have prognostic value,⁶ the absence of a widely accepted noninvasive gold standard has confused the discussions.

One problem is that the general perception of heart failure most often is based on the vast literature about heart failure with reduced LVEF and about systolic dysfunction. In contrast, less is known about the prevalence, prognosis, and treatment of heart failure with preserved LVEF and of diastolic heart failure. Studies on the prevalence of heart failure with preserved LVEF report proportions ranging from 20% to more than 60%, primarily due to differences in the populations studied and the definition of preserved LVEF (ranging from ≥55% to ≥40%). The data on prognosis also vary. For instance, in the Framingham study with a rather small cohort of patients with heart failure, the reported annual mortality rate was 8.7%; this rate is higher than that for persons without heart failure (3%-4%) but lower than that for patients with heart failure and reduced LVEF (18.9% per year).⁷

Unlike heart failure with reduced LVEF, there is not a database of large randomized trials supporting clear evidence-based therapeutic strategies. Most agree that patients with reduced LVEF should be candidates for angiotensin-converting enzyme inhibitors and β-blocker treatment but no such consensus exists for treatment of diastolic dysfunction among patients with preserved LVEF. At this point, perhaps the best one can argue is that angiotensin-converting enzyme inhibitors or angiotensin II receptor blockers should be part of the treatment regimen for patients with heart failure and preserved LVEF,^{4 ,8} which is reasonable considering that the majority of these patients also have hypertension.⁸

The study by Bursi and colleagues⁹ in this issue of JAMA provides substantial new evidence about the prevalence and prognosis of heart failure with reduced and preserved LVEF and diastolic heart failure in the community. The authors used Framingham criteria for diagnosing clinical heart failure, used established population-based methods for identifying patients with heart failure in Olmsted County, and used some of the best available echocardiographic methods for evaluating left ventricular systolic and diastolic function. The validity of the methods used for diastolic dysfunction is strengthened by the previous use of these methods in the general population study in Olmsted County,⁶ which showed substantially fewer persons with diastolic dysfunction compared with the present study in patients with heart failure. Furthermore, the validity of the patient selection criteria is supported by the substantially higher mortality rate of patients in this heart failure cohort, compared with the expected mortality rate, as well as by an increased levels of B-type natriuretic peptide in patients with heart failure and preserved LVEF, those with heart failure and reduced LVEF, and even those with diastolic heart failure.

The major findings in the prospective study of Bursi et al⁹ are that of the 556 patients in a fairly unselected heart failure population, 308 (55%) had preserved LVEF. The higher proportion compared with many previous studies could be due to the different echocardiographic methods used, or as suggested by the authors, by a temporal increase in heart failure with preserved LVEF. Diastolic dysfunction was present in approximately 80% of the patients, similar among patients with preserved LVEF as well as those with reduced LVEF, although patients with reduced LVEF had more severe diastolic dysfunction. Heart failure with preserved LVEF and diastolic heart failure are often linked to aging, hypertension, obesity, and diabetes,^{10 - 11} all of which are becoming increasingly common.^{12 - 13} These conditions may induce left ventricular hypertrophy and fibrosis, leading to myocardial stiffening, impaired filling, and clinical heart failure.¹⁴

Mortality in patients with preserved LVEF was remarkably similar to those with reduced LVEF (16% at 6 months for both groups). This finding contrasts with that reported in other studies that most often demonstrate a mortality rate of patients with heart failure and preserved LVEF to be in between the mortality rate in the general population and the rate in patients with heart failure and reduced LVEF.^{7 ,15 - 16} The difference could be due to the fact that only a few of these prior studies have been conducted in unselected populations.⁷ The prognosis of patients with varying degrees of diastolic heart failure, with or without reduced LVEF, was not specifically reported and should be the subject of further investigation.

Similar observations regarding the high frequency and poor prognosis of heart failure with preserved LVEF are reported in this issue of JAMA in a study by Gheorghiade and colleagues.¹⁷ Using the Organized Program to Initiate Lifesaving Treatment in Hospitalized Patients with Heart Failure (OPTIMIZE-HF) registry, which includes data on 48 612 patients hospitalized with a diagnosis of heart failure and follow-up data on a sample of 5791 patients, the investigators demonstrated that patients with higher systolic blood pressure at admission had a better prognosis in terms of in-hospital mortality and postdischarge mortality at 60 to 90 days than those with lower systolic blood pressure at admission. The authors also report no significant difference for in-hospitality mortality and postdischarge mortality rates between patients with preserved vs reduced left ventricular systolic function. Of note, for any given level of admission systolic blood pressure, mortality was similar irrespective of whether LVEF was preserved. However, the registry did not specifically report data on diastolic dysfunction with or without preserved LVEF.

The findings by Bursi et al⁹ and Gheorghiade et al¹⁷ that approximately half of patients with heart failure, whether observed in the community or in the hospital, have preserved systolic function and that mortality in these patients is similar to that for patients with heart failure and reduced systolic function have important implications. Just as heart failure with reduced LVEF has long been recognized as a common and serious disease and has been the subject of a number large-scale clinical trials, the entities of heart failure with preserved LVEF and diastolic dysfunction equally deserve attention. Deciphering the mechanisms and developing evidence-based treatments for these major public threats deserve the highest priority.