Prognosis in heart failure with preserved left ventricular systolic function: prospective cohort study

BMJ 2003; 327 doi: (Published 10 July 2003)
Cite this as: BMJ 2003;327:78
  1. Philip A MacCarthy, consultant1,
  2. Mark T Kearney, senior lecturer (mark.kearney{at},
  3. James Nolan, consultant2,
  4. Amanda J Lee, statistician3,
  5. Robin J Prescott, statistician3,
  6. Ajay M Shah, British Heart Foundation professor1,
  7. W Paul Brooksby, consultant cardiologist4,
  8. Keith A A Fox, British Heart Foundation professor5
  1. 1Department of Cardiology, Guy's, King's, and St Thomas's School of Medicine, London SE5 9PJ
  2. 2North Staffordshire Cardiac Centre, Stoke on Trent, ST4 6QG
  3. 3Medical Statistics Unit, University of Edinburgh, Edinburgh EH3 9YW
  4. 4Pontefract and Wakefield Hospitals, Pontefract WF8 1PL
  5. 5Department of Cardiology, University of Edinburgh
  1. Correspondence to: M T Kearney
  • Accepted 15 April 2003


Chronic heart failure due to left ventricular systolic impairment is characterised by a poor prognosis and abnormalities of cardiac structure, autonomic and neurohumoral function, and fluid and electrolyte homoeostasis, all of which are thought to contribute to the pathophysiology of this condition. However, some studies have found that 30-50% of all patients with chronic heart failure have preserved left ventricular systolic function.1 Despite this, the natural course of the condition in these patients is controversial, and their pathophysiological characterisation poor. As a result, optimum treatment strategies are unclear. We looked at five year mortality in patients recruited to a large cohort study of chronic heart failure, comparing those having impairment of left ventricular function with those having preserved function.

Participants, methods, and results

We have published details of the United Kingdom heart failure evaluation and assessment of risk trial (UK-HEART) previously.2 3 Five hundred and fifty three unselected ambulant patients were prospectively recruited from April 1993 to December 1995. Patients were enrolled if they had had stable, symptomatic chronic heart failure for at least three months (other primary causes of symptoms were excluded). As well as symptoms of chronic heart failure, all patients had evidence of cardiac dysfunction documented at the index assessment by one or more of the following: systolic left ventricular dysfunction on echocardiography or radionuclide ventriculography; cardiothoracic ratio > 0.55; and pulmonary venous congestion and/or upper lobe venous diversion on chest radiography. Left ventricular hypertrophy was assessed from electrocardiography (on the basis of Sokolow-Lyon criteria). All patients had 24 hour ambulatory monitoring for arrhythmia analysis and assessment of heart rate variability. A global index of total heart rate variability–the standard deviation of all normal R-R intervals (SDNN) (with a low value indicating a disadvantageous neurohumoral profile)–was derived from this recording. All patients were logged on the NHS Central Register (part of the Office of Population Censuses and Surveys), which notified the investigators when patients died.

Most studies confirm that patients with an ejection fraction ≥ 50% can be considered to have preserved left ventricular systolic function. We therefore chose this value to dichotomise patients into those with preserved (≥ 50%) and those with impaired (< 50%) function.4

In all, 522 patients had adequate measurements of ejection fraction, of whom 163 (31%) had values ≥ 50% and 359 (69%) < 50%. Information on deaths was recorded to April 2000, allowing five year survival status to be determined for all patients. The table shows the characteristics of the two groups and statistical methods.

View this table:

Characteristics of 522 patients with chronic heart failure and preserved or impaired left ventricular systolic function. Values are means (standard deviations) unless stated otherwise

Five year mortality was substantial in both groups but significantly greater in patients with impaired left ventricular systolic function (41.5% v 25.2%, P < 0.001). Twenty five per cent of patients with preserved function had non-sustained ventricular tachycardia. Both groups had similar SDNN measurements, which were lower than previously shown in age matched healthy control subjects.5


Mortality is significantly greater in patients with chronic heart failure and impaired left ventricular systolic function than in those with preserved systolic function. However, even the patients with preserved systolic function have a 25% five year mortality. Therefore, clinical heart failure itself has a poor long term prognosis, irrespective of electrocardiographically determined left ventricular systolic function. Autonomic function was abnormal in both groups, and this, allied to the presence of non-sustained ventricular tachycardia and left ventricular hypertrophy, may contribute to the high mortality found in the patients with preserved systolic function.

Our findings add to those of a recent study by Kitzman et al, who found that patients with chronic heart failure and preserved left ventricular systolic function have similar, but not as severe, pathophysiological derangements to those with impaired systolic function.4 These and our data suggest that established treatments for systolic heart failure may also have a role in patients with chronic heart failure and preserved left ventricular systolic function.

Embedded ImageA graph showing survival curves is available on>


  • Contributors PAM and MTK conceived the original idea, did the retrieval, analysis, and interpretation of the data, and wrote the paper. MTK helped to collect the raw data. KAAF, JN, and WPB were responsible for the original concept of the UK-HEART study, designed the database, and collected the data on patients; they also reviewed and revised the current paper. AJL and RJP helped with data acquisition and advised on statistical analysis. AJL did a large part of the statistical analysis. AMS helped with the conception of this project, the study design, the interpretation of data, and the revision of the manuscript. MTK is the guarantor for the paper.

  • Funding PAM was the recipient of an advanced training scholarship from the British Heart Foundation; MTK is a British Heart Foundation (BHF) intermediate research fellow; and AMS and KAAF hold BHF chairs in cardiology.

  • Competing interests None declared

  • Ethical approval Local ethics committees at each participating hospital approved the protocol, and informed written consent was obtained from all patients.