BACKGROUND: Premature discharge from the intensive care unit (ICU) results in ICU readmission and poor outcome. Understanding the clinical features of the readmitted patients may be helpful for intensivists to improve ICU care. We performed this study to determine the causes, outcomes, and risk factors of patients readmitted to the ICU. METHODS: Data was collected from the patients admitted to medical and surgical ICUs of Severance Hospital between January, 1999 and July, 2001 retrospectively. Readmission cause, source, indication, length of ICU stay, Acute Physiology and Chronic Health Evaluation (APACHE) III score, and multiple-organ failure (MOF) score of readmitted patients were evaluated. Non-survivors and survivors after ICU readmission were compared. RESULTS: One hundred and thirty-seven readmitted and 2,412 non-readmitted patients were examined and the readmission rate was 6.3%. Respiratory disease was the major cause of readmission. Readmitted patients had longer initial ICU lengths of stay than non-readmitted patients (13.6 vs 9.4 days, p<0.05). The ICU mortality rate was not significantly higher in the readmitted patients compared with the non-readmitted patients. The MOF score on readmission (5.4 vs 3.1) and APACHE III score on initial discharge (40.7 vs 30.4) and readmission (76.3 vs 44.4), in non-survivors were higher than survivors of the readmitted patients, respectively (p<0.05). CONCLUSIONS: ICU readmission was associated with longer ICU stay and respiratory disease was the major cause of readmission. The MOF score at readmission and APACHE III score at discharge and readmission were significant risk factors of the outcome in readmitted patients.