Correlates of major complications or death in patients admitted to the hospital with congestive heart failure

Background: When triaging a patient who has heart failure, the physician must estimate the patient's short-term risk of a major complication or death. Methods: Prospective cohort study of 435 patients admitted nonelectively to an urban university hospital between February 2, 1993, and February 2, 1994, with a complaint of shortness of breath or fatigue and evidence of congestive heart failure on admission chest radiograph. Results: Major adverse events occurred in 18% of patients who had ejection fractions less than 0.50, 16% of those with ejection fractions of 0.50 or more, and 19% of those with previous heart failure, ejection fractions of 0.50 or more, and no significant valvular disease. In multivariate analyses of all patients, independent correlates (P less than or equal to .01) of major complications or death during hospitalization were initial systolic blood pressure of 90 mm Hg or less (adjusted odds ratio [OR], 5.5; 95% confidence interval [CI], 1.7-17.1), respiratory rate more than 30 breaths per minute on admission to the hospital (OR, 4.6; 95% CI, 2.4-8.8), serum sodium level of 135 mmol/L or less (OR, 2.2; 95% CI, 1.3-4.0), and ST-T wave changes on initial electrocardiogram neither known to be old nor attributable to digoxin (OR, 5.1; 95% CI, 2.9-8.9). However, even patients with none of these 4 risk factors had a 6% rate of a major complication or death. Conclusions: No truly low-risk group existed. Patients without hypotension, tachypnea, hyponatremia, or electrocardiographic changes of ischemia represent the best candidates for triage to less intensely monitored settings, but clinical judgment is essential.