Background: The reported prevalence of preserved systolic function (PSF) heart failure (HF) varies widely and has not been well-studied in nonwhite patients. To estimate the prevalence of PSF HF resulting in hospital admission, we studied admissions to a large community hospital serving a racially diverse community.
Methods: The charts of 300 consecutive patients > or =65 years old with a primary discharge diagnosis of HF were reviewed. In patients who met the Framingham criteria for HF diagnosis, an assessment of left ventricular function was obtained from review of chart data. Comparison of baseline characteristics and multivariate analysis of potential predictors of PSF HF was undertaken.
Results: Of the 300 patients, 247 (82%) met the Framingham criteria for diagnosis of HF. Ninety-seven patients (39%) of these had PSF. Twenty (8%) of the diagnosed HF patients had a diagnosis of severe aortic or mitral valvular disease, 9 of whom had preserved systolic functions. Thus 88 (36%) of the HF patients had PSF HF likely resulting from diastolic dysfunction. On multivariate analysis, age, hypertension at presentation, peripheral edema, normal sinus rhythm and a history of coronary artery bypass grafting or coronary angioplasty were significantly associated with PSF HF rather than HF with reduced systolic function. Left bundle branch block or other intraventricular conduction delay was independently associated with reduced systolic function HF. However, because these findings occurred in patients with and without systolic dysfunction, none were pathognomonic of PSF HF.
Conclusion: In this racially diverse, majority nonwhite, cohort of older patients admitted for HF, the Framingham criteria for the diagnosis of HF were met in 82% of the patients. Of these, 39% had PSF HF and 36% had PSF HF in the absence of severe aortic or mitral valve disease, suggesting that diastolic dysfunction was etiologic. Although there was an independent association of PSF HF with a number of clinical characteristics, none of these characteristics was pathognomonic of preserved, versus reduced, systolic function HF.