Prognosis in ischemic heart disease. Can you tell as much at the bedside as in the nuclear laboratory?

BACKGROUND--While the resting left ventricular ejection fraction (LVEF) predicts prognosis in ischemic heart disease, clinical evaluation is also useful. METHODS--To compare the prognostic value of LVEF by resting radionuclide ventriculography with that of clinical signs and symptoms of congestive heart failure (CHF), 170 patients with suspected ischemic heart disease were followed up in this prospective study. Patients had a standardized history and physical examination performed by a study cardiologist immediately before the nuclear scan. Chest roentgenography and radionuclide ventriculography were performed in a standard manner. The diagnosis of CHF was made by validated clinicoradiographic criteria based on the Framingham study. Mortality was determined by means of the National Death Index; median follow-up time was 3 years. RESULTS--There was CHF at baseline in 70 patients, and baseline LVEF was low (< or = 0.4) in 63 patients. Low LVEF was significantly associated with CHF. During follow-up, 55 of the subjects died (overall mortality, 32%). Subjects with CHF had a significantly higher risk of death than those without CHF, and subjects with low LVEF had a higher mortality than those with preserved LVEF. Both CHF and LVEF were independent predictors of mortality. In a Cox model, each percentage increase in LVEF was associated with a 2% decreased mortality, while subjects with CHF had a mortality 2.5 times higher than that of those without CHF. Also, CHF with preserved LVEF had a better prognosis than CHF with depressed LVEF, but this prognosis was worse than that in subjects without CHF. CONCLUSIONS--The clinical diagnosis of CHF, based on clinical evaluation and chest roentgenogram, is a valid predictor of mortality and provides information independent of the radionuclide LVEF in determining prognosis in patients with ischemic heart disease.