The long-term prognostic influence of left ventricular systolic function (LVSF) in diagnosing patients with congestive heart failure (CHF) has been widely documented. However, the short-term prognostic significance of LVSF on early hospital readmission for CHF is limited. OBJECTIVES: The study purpose was to examine the influence of LVSF on 90-day readmission for CHF. METHODS: Medical records were reviewed for all patients with a principle diagnosis of CHF discharged alive from January 1 through June 30, 1997 at an urban public hospital. LVSF assessments within six months prior to discharge were used to classify patients as normal LVSF or left-ventricular (LV) systolic dysfunction. RESULTS: Among 387 patients, 303 (78.3%) had 6-month assessments of LVSF. Among patients with LVSF assessments, 207 patients, 68.3% had LV systolic dysfunction and 96 patients, 31.7% had normal LV systolic function. A total of 54 (26.1%) of the 207 patients with LV systolic dysfunction were readmitted for CHF within 90 days, versus 13 (13.5%) of the 96 patients with normal LVSF that were readmitted (p<0.01). Multi-variate logistic regression analysis (including: age, gender, prior history of CHF, comorbidities, NYHA class, medication adherence) showed 90-day readmission for CHF was more likely in patients with LV systolic dysfunction than in those with normal LVSF after adjusting for other identified risk factors. (p=0.05) CONCLUSIONS: LV systolic dysfunction appears to increase the likelihood of 90-day readmission among patients with CHF. The findings underscore the ability of LVSF to stratify patients hospitalized for CHF and suggest potential usefulness in aiding management by identifying higher risk patients.