Intended for healthcare professionals

Papers

Do patients with suspected heart failure and preserved left ventricular systolic function suffer from “diastolic heart failure” or from misdiagnosis? A prospective descriptive study

BMJ 2000; 321 doi: https://doi.org/10.1136/bmj.321.7255.215 (Published 22 July 2000) Cite this as: BMJ 2000;321:215
  1. Lynn Caruana, senior cardiac technician,
  2. Mark C Petrie, specialist registrar in cardiology,
  3. Andrew P Davie, specialist registrar in cardiology,
  4. John J V McMurray, professor of medical cardiology (j.mcmurray{at}bio.gla.ac.uk)
  1. Department of Cardiology, Western Infirmary, Glasgow G11 6NT
  1. Correspondence to: J McMurray
  • Accepted 25 April 2000

Abstract

Objectives: To characterise the clinical features of patients with suspected heart failure but preserved left ventricular systolic function to determine if they have other potential causes for their symptoms rather than being diagnosed with “diastolic heart failure.”

Design: Prospective descriptive study.

Setting: Outpatient based direct access echocardiography service.

Participants: 159 consecutive patients with suspected heart failure referred by general practitioners.

Main outcome measures: Symptoms (including shortness of breath, ankle oedema, and paroxysmal nocturnal dyspnoea) and history of coronary heart disease and chronic pulmonary disease. Transthoracic echocardiography, body mass index, pulmonary function tests, and electrocardiography.

Results: 109 of 159 participants had suspected heart failure in the absence of left ventricular systolic dysfunction, valvular heart disease, or atrial fibrillation. Of these 109, 40 were either obese or very obese, 54 had a reduction in forced expiratory volume in 1 second to ≤70%, and 97 had a peak expiratory flow rate ≤70% of normal. Thirty one patients had a history of angina, 12 had had a myocardial infarction, and seven had undergone a coronary artery bypass graft. Only seven patients lacked a recognised explanation for their symptoms.

Conclusions: For most patients with a diagnosis of heart failure but preserved left ventricular systolic function there is an alternative explanation for their symptoms—for example, obesity, lung disease, and myocardial ischaemia—and the diagnosis of diastolic heart failure is rarely needed. These alternative diagnoses should be rigorously sought and managed accordingly.

Editorial by Hobbs

Introduction

With the recent wide availability of non-invasive assessments of left ventricular function it has become apparent that many patients diagnosed as having heart failure have preserved left ventricular systolic function. 1 2 It has been proposed that these patients have abnormalities of ventricular filling in diastole, and the term “diastolic heart failure” has been coined.13 It does, however, seem likely that given the non-specificity of the symptoms and signs used to diagnose heart failure at least some of these patients may not have abnormalities of diastolic ventricular function but other causes of their symptoms altogether. We studied consecutive patients who were referred with a diagnosis of suspected heart failure by general practitioners to our direct access transthoracic echocardiography service and were found to have preserved left ventricular systolic function. We defined their clinical characteristics and considered the alternative diagnoses of obesity, respiratory disease, and ischaemic heart disease.

Methods

Patients—We studied patients with preserved left ventricular systolic function who were referred with suspected heart failure to an outpatient based direct access cardiography service. Patients with left ventricular systolic dysfunction (according to qualitative “eyeball” assessment, see below), valvular heart disease, and atrial fibrillation were not studied further. The study was approved by our local committee for medical ethics. Each patient gave written informed consent.

Medical history, drug history, and symptoms—A full clinical history was taken. Symptoms of shortness of breath at rest or on exertion, paroxysmal nocturnal dyspnoea, and ankle swelling were specifically recorded. A past or current history of angina was also specifically elicited. The severity of angina was graded according to the Canadian Cardiovascular Society.4 Current medication was recorded.

Body mass index—Body mass index was calculated in the usual way (weight (kg)/height (m)2). Participants with a body mass index of less than 18.5 were defined as underweight, 18.5-24.9 as normal weight, 25.0-29.9 as overweight, 30.0-39.9 as obese, and ≥40 as extremely obese.5

Respiratory function—All patients had peak expiratory flow rate measured and had spirometry performed—that is, forced expiratory volume in 1 second (FEV1) and forced vital capacity.

Electrocardiography—A standard, resting 12 lead electrocardiogram was recorded in each patient. Pathological Q waves were taken as evidence of previous myocardial infarction. ST/T changes, previous myocardial infarction, or left bundle branch block were considered to be consistent with a diagnosis of coronary heart disease.

Transthoracic echocardiography—Each patient underwent transthoracic echocardiography with an Acuson 128XP10c, with the patient recumbent in the left lateral decubitus position. The same operator (LC) performed all examinations. Qualitative assessment of left ventricular systolic function was made by the “eyeball” technique, with two dimensional images. 6 7 Quantitative assessment was made by measurement of ejection fraction and fractional shortening by using M mode echocardiography. Ejection fraction was also measured using the Simpson's biplane method.8 For diastolic function the E:A ratio was measured as described previously. 1 9 Left ventricular hypertrophy was assessed by M mode echocardiography.

Results

Patients—One hundred and fifty nine patients with suspected heart failure were referred, and 34 had left ventricular systolic dysfunction (18 out of 61 men (30%; 95% confidence interval 19% to 41%) and 16 out of 98 women (16%; 9% to 23%)). Ten patients had atrial fibrillation, two patients had valvular disease, and four patients had both atrial fibrillation and valvular heart disease. One hundred and nine patients had suspected heart failure in the absence of left ventricular systolic function, valvular heart disease, or atrial fibrillation. Details of these 109 patients is shown in table 1. In keeping with findings of epidemiological studies, the patients were elderly and usually female. Most patients had been prescribed diuretics. Thirty three patients were non-smokers, 29 were current smokers, and 47 were former smokers.

Table 1

Characteristics of patients referred for investigation of suspected heart failure. Figures are number (percentage) of patients unless stated otherwise

View this table:

Dyspnoea and ankle swelling—Most men and women reported dyspnoea on exertion, and about a quarter reported either dyspnoea at rest or at night (table 1). Men and women differed in their reporting of ankle swelling, however, with twice as many women (80%) complaining of this symptom.

Angina, myocardial infarction, and coronary artery bypass graft—Thirty one (28%; 20% to 36%) patients had a history of angina though only 11 (10%; 4% to 16%) currently had symptoms (seven were classified as grade I, four as grade II). Twelve (11%; 5% to 17%) patients gave a history of myocardial infarction, and seven had undergone a coronary artery bypass graft. In total, 33 (30%; 21% to 39%) patients either had a history of angina or myocardial infarction or had undergone a coronary artery bypass graft. Eight patients had electrocardiographic evidence consistent with coronary heart disease.

Table 2

Prevalence of abnormalities of body mass index, forced expiratory volume in one second (FEV1), and electrocardiographic abnormalities. Figures are number (percentage) of patients

View this table:

Body mass index—Of the 109 patients with suspected heart failure in the absence of left ventricular systolic dysfunction, valvular heart disease, or atrial fibrillation, 108 had their body mass index calculated. Details are shown in table 2. Thirty five (32%; 23% to 41%) were obese, and five (5%;1% to 9%) were extremely obese.

Respiratory function tests—Of 109 patients, 106 had their respiratory function measured. Table 2 shows the results of these measurements. Fifty four patients (50%; 41% to 59%) had FEV1 less than 70% of that predicted, and 97 (92%; 86% to 98%) had a peak expiratory flow rate ≤70% of normal.

Electrocardiography—Seventy (64%; 55% to 73%) patients had a normal results on electrocardiography. More men than women had abnormalities. These are shown in table 2. As left bundle branch block, myocardial infarction, and ST/T changes are typical findings in coronary heart disease, 24% of men and 16% of women had electrocardiographic evidence of possible coronary heart disease. Taken in conjunction with evidence from the clinical history (33 patients), eight additional patients had either clinical or electrocardiographic evidence of possible coronary heart disease—that is, a total of 41 (38%; 29% to 47%) patients had either clinical or electrocardiographic evidence of possible coronary heart disease.

Transthoracic echocardiography—Table 3 shows measurements of left ventricular systolic function. By using the E:A ratio, 67% (74% men and 64% women) had “diastolic dysfunction.” Left ventricular hypertrophy was detected in about one quarter of patients.

Table 3

Electrocardiographic parameters in patients referred with suspected heart failure

View this table:
Table 4

Overlap between FEV1, left ventricular hypertrophy, angina, and obesity in 106/109 patients with complete data

View this table:

Assessment of the overlap of obesity, respiratory disease, and cardiac abnormalities—Tables 4 and 5 show the overlap between abnormalities of body mass index (overweight, obesity, or extreme obesity), FEV1 (less than 70% of predicted), cardiac structure (left ventricular hypertrophy), and evidence of coronary heart disease (clinical or electrocardiographic). Table 4 compares the overlap of obesity and angina with the presence of FEV1 less than 70% of predicted and left ventricular hypertrophy. Table 5 differs from table 4 in using abnormal body mass index (overweight, obese, and extremely obese) in place of obesity and coronary heart disease (history of angina (past or present), myocardial infarction, coronary artery bypass graft, or ECG changes consistent with coronary heart disease) in place of angina. Of 109 patients for whom complete data were available for all parameters, 106 were included. Only nine (9%; 4% to 14%) patients were of normal weight and had a FEV1 greater than 70% predicted. Of these, two had clinical or electrocardiographic evidence of coronary heart disease. Consequently, only seven patients (7%; 2% to 12%) had no evidence of abnormalities of body mass index, respiratory disease, or coronary heart disease.

Table 5

Overlap between FEV1, left ventricular hypertrophy, coronary heart disease, and abnormal body mass index (BMI ≥25) in 106/109 patients with complete data

View this table:

Discussion

Along with others, we have found that many patients presenting with heart failure have preserved left ventricular systolic function. 13 1015 While these patients may have “diastolic dysfunction” it is also possible that there are other explanations for their symptoms. The problem about making a diagnosis of “diastolic” heart failure non-invasively is that there is no agreement on how this should be done, and different criteria for diastolic dysfunction give enormously differing prevalences. 2 16 One of the most commonly used criterion, an E:A ratio of <1, showed that most of our patients with heart failure and preserved left ventricular systolic function could be said to have diastolic dysfunction. Rather than examine the vexed issue of how one defines diastolic dysfunction with echocardiography we have examined an alternative—that is, could there be another explanation for these patients' symptoms?

The most obvious alternative diagnoses are obesity, respiratory disease, and myocardial ischaemia. We found that the first two of these were common. A third of patients were either obese or very obese. Half of the patients had a considerable reduction in FEV1 (to 70% or less) and 89% had a peak expiratory flow rate less than or equal to 70% of normal. Remarkably, only nine patients were of normal weight and had FEV1 greater than 70% predicted. Though we sought to identify myocardial ischaemia only by recording a history of angina, 31 patients admitted to this symptom, 12 had a history of myocardial infarction, and seven had undergone coronary artery bypass surgery. Furthermore, at least 20 patients had electrocardiographic changes consistent with myocardial ischaemia or infarction. If we had undertaken exercise stress electrocardiography even more patients with myocardial ischaemia would probably have been identified. Of the nine patients with normal weight and FEV1 greater than 70%, a further two had clinical or electrocardiographic evidence of coronary heart disease. In other words, only seven patients in this study with a diagnosis of heart failure but preserved left ventricular systolic function lacked a recognised explanation for their symptoms—that is, a diagnosis of diastolic heart failure was inappropriate. The important message for clinicians is that an echocardiogram suggesting diastolic dysfunction on the basis of an abnormal E:A ratio is not diagnostic and represents insufficient investigation.

What is already known on this topic

Patients with suspected heart failure but preserved left ventricular systolic function are commonly said to have “diastolic heart failure”

What this study adds

Most of these patients have an alternative explanation for their symptoms, such as obesity, pulmonary disease, and myocardial ischaemia

Complete investigation of these patients requires more than an echocardiogram

Improved patient care should result from recognition of the true cause of a patient's symptoms as there are appropriate management strategies for these alternative diagnoses; this is preferable to ascribing symptoms to diastolic heart failure for which there is no evidence based treatment

We clearly need to improve differentiation of breathlessness due to isolated diastolic dysfunction from that with other causes. It would seem that a rigorous search for non-cardiac causes of breathlessness must be pursued, with pulmonary function testing, calculation of body mass index, resting and exercise electrocardiography, and, probably, chest radiography. Even if these other causes are excluded it may still be difficult to be sure that a patient's breathlessness is definitely cardiac in origin. One possibility is that measurement of plasma natriuretic peptide concentrations might further refine the diagnostic process. It might be expected that increased left ventricular mass, wall stress, or filling pressures would increase secretion of atrial or brain natriuretic peptides. This possibility needs to be tested further.

Of course, it is also possible that patients may have more than one cause of their dyspnoea. This real diagnostic dilemma reinforces the need for better means of determining whether or not there really is a non-systolic cardiac contribution in such cases. We believe that improved patient care should result from recognition of the true cause of a patient's breathlessness as appropriate management of the correct alternative diagnoses can improve presenting symptoms.

In summary, we have shown that in most patients with a diagnosis of heart failure but preserved left ventricular systolic function there are alternative explanations for their symptoms—for example, obesity, lung disease, and myocardial ischaemia. For that reason the diagnosis of diastolic heart failure is probably unnecessary, even though a high proportion of these patients will have echocardiographic evidence of diastolic dysfunction.16

Acknowledgments

Contributors: JJVM had the original idea for the study. JJVM, APD, and LC were involved in the study design. LC collected the data and performed the echocardiograms. MCP and JJVM analysed and interpreted the data. The paper was written jointly by JJVM and MCP. JJVM is the study guarantor.

Footnotes

  • Funding None.

  • Competing interests None declared.

References

  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
View Abstract