Patients readmitted to the ICU during the same hospitalization: a 2-year study

Objective

To determine risk factors and outcomes in critically ill patients who were readmitted to the ICU during their hospital stay.

Setting

A general ICU of a tertiary community hospital.

Patients and methods

We retrospectively analysed ICU readmissions between 1 January 2003 and 31 December 2004. The data analysed included patients' clinical characteristics, APACHE II, SOFA score, TISS-28, length of ICU stay, ICU and hospital mortality.

Results

During the study period, 735 patients were admitted to the ICU. Among patients who survived (467 patients), 42 patients (27 male, 15 female) (9.03%) were readmitted. The mean age was 53 ± 17.7 years. The prevalent cause of readmission was respiratory and cardiovascular complications (63.61%) followed by sepsis (12.22%), surgical problems (11.74%), neurological disorders (7.34%) and miscellaneous (5.09%). Patients whose neurological and respiratory disorders were the main admission reasons in the ICU had the highest readmission rate. APACHE II on first admission was estimated for non-readmitted patients (NREAD) (21 ± 7.5) and for readmitted patients (READ) (24 ± 6.6) (predicted mortality was 37%, and 47%, respectively), while initially admitted READ patients required less need of organ support than at the time of readmission (SOFA score was 8.0 ± 3.3 and 9.5 ± 3.5 respectively). Patients needed more therapeutic procedure at readmission than at their first ICU admission (TISS-28 on first ICU admission was 31 ± 5.4 and at readmission was 35.3 ± 5.6). The ICU mortality in READ patients was 42.7% and their hospital mortality was 65.7% (NREAD hospital mortality was 52.8%). The time between extubation to ICU discharge was 1.42 days (± 0.79), the median interval between first ICU discharge and readmission was 3.97 days (12 hours–14 days), while 11 patients (26.19%) were in need of the ICU less than 48 hours after discharge.

Conclusion

Patients with neurological and respiratory disorders were at greatest risk of requiring ICU readmission. Respiratory and cardiovascular complications were the major reasons for readmission. The readmitted patients appeared to be sicker and they had a higher risk of hospital death than non-readmitted patients. Probably, if an intermediate ICU was available in our hospital, the readmission rate in ICU would be visibly lower.