EPIDEMIOLOGY OF HEART FAILURE

Studies of the epidemiology of heart failure have been complicated by the lack of universal agreement on a definition of heart failure, which is primarily a clinical diagnosis. National and international comparisons have therefore been difficult, and mortality data, postmortem studies, and hospital admission rates are not easily translated into incidence and prevalence.

Several different systems have been used in large population studies, with the use of scores for clinical features determined from history and examination, and in most cases chest radiography, to define heart failure.

The Task Force on Heart Failure of the European Society of Cardiology has recently published guidelines on the diagnosis of heart failure, which require the presence of symptoms and objective evidence of cardiac dysfunction. Reversibility of symptoms on appropriate treatment is also desirable.

Echocardiography is recommended as the most practicable way of assessing cardiac function, and this investigation has been used in more recent studies. In the Framingham heart study a cohort of 5209 subjects has been assessed biennially since 1948, with a further cohort (their offspring) added in 1971. This uniquely large dataset has been used to determine the incidence and prevalence of heart failure, defined with consistent clinical and radiographic criteria.

Several recent British studies of the epidemiology of heart failure and left ventricular dysfunction have been conducted, including a study of the incidence of heart failure in one west London district (Hillingdon heart failure study) and large prevalence studies in Glasgow (north Glasgow MONICA study) and the West Midlands ECHOES (echocardiographic heart of England screening) study. It is important to note that epidemiological studies of heart failure have used different levels of ejection fraction to define systolic dysfunction.

The Glasgow study, for example, used an ejection fraction of 30% as their criteria, whereas most other epidemiological surveys have used levels of 40-45%. Indeed, prevalence of heart failure seems similar in many different surveys, despite variation in the levels of ejection fraction, and this observation is not entirely explained.