Introduction

Congestive heart failure (CHF) occurs in over 4.5 million people in the USA, and half a million new cases are added each year. In up to 50% of these subjects, systolic function is found to be relatively intact (i.e. a left ventricular ejection fraction [LVEF] > 50%); such cases are considered to have left ventricular diastolic dysfunction (that is, diastolic heart failure). However, there are reports that these patients are often mislabeled, really being due to obesity or pulmonary disease. Two recent studies have examined the features of diastolic heart failure, in an attempt to define the physiological events during the episode, and to suggest strategies for further study and approaches to treatment.

Methods

Cardiovascular Health Study
The first study was a population-based examination of CHF in over-65 year-olds. Participants in the Cardiovascular Health Study (CHS), sponsored by the National Heart, Lung, and Blood Institute, were recruited from 4 different areas in the USA. Those subjects who had echocardiograms in 1994 and 1995 were included. There were 2,920 women and 1,922 men who provided comprehensive data - a careful history, full physical examination, blood chemistry, electrocardiography, carotid wall ultrasonography, and two-dimensional Doppler echocardiography. Assessed LVEF values were classified as normal (>55%) mildly reduced (45% - 54%), and moderately to severely reduced (<44%).

An expert panel reviewed all diagnoses of CHF. Any participant with at least one confirmed episode of CHF before the study was considered prevalent for CHF. Analyses were done of correlates for CHF using logistic regression of demographic and clinical variables, and, lastly, echocardiography variables.

Wake Forest Study
The second study was done at Wake Forest University, Winston Salem, USA, on 38 patients with acute pulmonary edema and systolic blood pressure >160 mm Hg. (Wake Forest was a participating center in the CHS.) Subjects had to be free of evidence of pneumonia, myocardial infarction, or uremia. Two-dimensional transthoracic Doppler echocardiography was done during the acute episode and again one to three days later. These were analyzed to provide the LVEF, the left ventricular regional wall-motion index, and the presence and severity of any mitral valve regurgitation.

Results

Cardiovascular Health Study
The mean age of the 4,842 participants was 77 ± 5 years. There were 425 subjects with a confirmed history of CHF in the CHS study - 8.8%. These were more likely to be men, older, and more often afflicted with a history of myocardial infarction, atrial fibrillation, hypertension, diabetes, COPD (women only), renal dysfunction, and to be smokers, when compared with non-CHF participants. Increased frequency of CHF with age, however, was more pronounced in women than in men.

Those with prevalent CHF had increased left atrial and ventricular dimension on echocardiography. Over half of them (55%) had normal left ventricular systolic function, and in a further 25% it was only mildly reduced. There was a distinct gender difference in this respect - in women with CHF, 67% had normal systolic function, versus 42% in men (p<0.001). Normal systolic ventricular function correlated with lower left ventricular diastolic and systolic dimensions.

Wake Forest Study
The 38 patients in the Wake Forest study had a mean age of 67 ± 13 years; 24 were women, 14 were men. Their mean systolic pressure at entry was 200 ± 26 mm Hg. Mean LVEF values were 0.5 ± 0.15, and mean wall-motion indices 1.6 ± 0.6. In fact, half the patients had a normal LVEF during the episode. Minimal mitral regurgitation was seen in 26 patients, and to a moderate degree in a further 6 patients; no patients had severe regurgitation. Ten patients were receiving a beta-adrenergic blocker. They were treated with furosemide, and nitroglycerin (34 of 38 patients).

At the time of the follow-up examination the mean systolic pressure was 139 ± 17 mm Hg; 29 patients were receiving angiotensin-converting-enzyme inhibitors, 22 beta-adrenergic blockers, and 11 calcium-channel blockers. The LVEFs were virtually the same as during the acute episode - mean values of 0.5 ± 0.13. Wall-motion indices were also unchanged - mean 1.6 ± 0.6. Importantly, the individual LVEFs after treatment correlated directly with those during the acute episode (r=0.83). A similar correlation was found for the wall-motion indices (r=0.98).

Comment

The CHS findings showed that 55% of elderly community-dwelling CHF subjects have a normal LVEF, and a total of 80% have normal or only mildly reduced LVEF. The authors point out that this set of patients has not, to date, been characterized; most clinical trials of heart failure treatments employ an inclusion criterion of LVEF <45. Moreover, women are often under-represented in such research, although they clearly represent a growing subset of patients.

The Wake Forest study was done to examine whether acute pulmonary edema in association with hypertension was due to transient systolic dysfunction. The results showed that this was not the case. Rather, there appeared to be an exacerbation of isolated diastolic dysfunction in these patients.

Taken together, these two studies point up the importance of diastolic heart failure in the elderly (especially elderly women), making a clear argument for further clinical research in this condition. As an editorial in the New England Journal has recently pointed out, treatment of diastolic heart failure today is largely empirical¹. This includes reduced sodium intake, cautious use of diuretics, restoration of sinus rhythm, and the correction of any episode-precipitating factors; obviously, the prevention and control of hypertension is an important target. There are few clinical data to support any particular classes of medications for the treatment of diastolic heart failure, however; more drug trials are urgently needed.