C disagreement exists on data published on the prevalence of heart failure.1–6 However, studies on both the prevalence and natural history of heart failure generally tend to be retrospective and to have been performed in patients referred to specialized centers (most of whom are in advanced stages of heart failure). Therefore, these patients are not representative of congestive heart failure in the community.6–8 Because of the lack of diagnostic criteria and data on the prevalence of heart failure, we conducted a field study in 1996 to establish its prevalence of this in our community. For these reasons we followed the Framingham clinical criteria of heart failure3 (Table 1). We also included 2 dimensional echo-Doppler examinations to stratify heart failure into systolic or diastolic failure.9–11 • • • In September 1995, the census of Asturias in Northen Spain, with a population of 1,098,725, contained 515,487 persons aged . 40 years. Assuming an epsilon of 6 1, a confidence interval of 95%, and a prevalence of heart failure .1% (1% to 5%), we selected a random sample of 380 subjects, a number statistically representative of this population. The total sample size for the study included 6% more subjects than what was originally estimated to adjust for the possible effect of dropouts; we decided not to replace those who failed to attend appointments. These subjects were stratified according to age and sex. Of the 400 randomized persons, 9 were excluded because of census errors. The remaining 391 persons (100%) underwent medical examination, 367 (93%) underwent electrocardiography, 356 (91%) chest xray examinations; 2 dimensional echo-Doppler studies estimated left ventricular ejection fraction in 351 patients (89%). Two dimensional echo-Doppler studies were recorded with a Hewlett-Packard (Andover, Massachusetts) ultrasound unit (model 1500) and 2.5MHz transducer. Those presenting with a left ventricular ejection fraction .50% (by the Teichholz method) were considered to have normal systolic function.9,12,13 Diastolic function10,11,14 was assessed by left ventricular filling patterns: mitral valve E (m/s)/mitral valve A (m/s) (E/A) ratio, deceleration time, and isovolumetric relaxation time (Table 2). For systolic and diastolic assessment, a mean of 3 measurements was obtained in each patient with sinus rhythm, and a mean of 5 measurements was obtained in those with atrial fibrillation. The echo study results were used for further stratification of heart failure (systolic and diastolic dysfunction). Stratification of the population according to age and sex, presence of heart failure, and echo data results are shown in Table 3. Table 3 shows the stratification according to the age of the population and of the sample aged .40 years. The mean age of the sample was 59.5 6 12.5 years. In terms of gender our sample contained 183 men From the Division of Cardiology, Hospital Central de Asturias, Oviedo University, Asturias, Spain. This work was supported by a Merck, Sharp & Dohme Grant. Dr. Cortina’s address is: Division of Cardiology, Hospital Central de Asturias, C/Julian Claveria s/n, 33006 Oviedo, Asturias, Spain. E mail: ROSARIOCORTINA@teleline.es. Manuscript received October 12, 2000; revised manuscript received and accepted January 17, 2001. TABLE 1 Criteria for Heart Failure
Read full abstract