Prospective evaluation of patients readmitted after cardiac surgery: Analysis of outcomes and identification of risk factors

ObjectiveReducing hospital readmissions after adult cardiac surgery is necessary as part of the solution to achieving improved efficiency in health care. Patients who had undergone cardiac surgery were studied to develop strategies that may diminish the need for hospital readmission.MethodsOver a 25-month period, 2096 patients underwent cardiac surgical procedures; 102 of these patients required readmission within 30 days of discharge. Time-matched patients (n = 249), not readmitted, served as a control group. Patient demographics and perioperative variables were analyzed by univariate analyses. Logistic regression analysis identified independent risk factors for readmission.ResultsThe most common diagnoses given for readmission were congestive heart failure (26 of 102, 25%), infection (23 of 102, 23%), and arrhythmias (15 of 102, 15%). The comorbidities more prevalent among readmitted patients were diminished ejection fraction (44% ± 17% vs 56% ± 13%; P < .0001), chronic obstructive pulmonary disease (23 of 102, 23% vs 23 of 249, 9%; P = .0008) and chronic renal insufficiency (26 of 102, 26% vs 24 of 249, 10%; P = .0001). Multivariate logistic regression identified chronic obstructive pulmonary disease (odds ratio [OR], 2.0; P = .05), diminished ejection fraction (OR, 0.8; P < .0001), a lower education level (OR, 0.5; P = .0001), and a prolonged length of stay (OR, 1.6; P = .009) as predictive of readmission. Failure to see a physician early in the postoperative period was associated with a 6-fold increase in the risk of readmission (P < .0001).ConclusionsPatients readmitted after cardiac surgery have specific comorbidities and are of lower socioeconomic status. They are admitted most commonly for exacerbation of congestive heart failure or infectious reasons. This study suggests that seeing a physician early after discharge may have an impact on reducing readmissions after cardiac surgery.