Predictors and outcome of readmission after laparoscopic intestinal surgery.

Journal: World Journal Of Surgery

Background: Previous studies have failed to identify predictors of early readmission after major intestinal operations. The objectives of this study were to determine readmission rates, outcomes, and predictors of readmission for patients undergoing laparoscopic colon and rectal operations.

Methods: Patients readmitted (PR) to the hospital within 30 days of discharge after laparoscopic colon and rectal operations were identified from a prospectively maintained database. The PR group was compared with patients that were not readmitted (NR). Outcomes and variables related to readmission were evaluated.

Results: There were 820 consecutive elective laparoscopic colon and rectal operations performed over a 5-year period, with adequate follow-up data for 787 cases. Seventy-nine (10%) patients were readmitted. There was no difference in the age, sex, surgeon, or type of operation between the PR and NR groups. The most common causes for readmission were bowel obstruction (19%), ileus (18%), intra-abdominal abscess (14%), and anastomotic leak (9%). Overall mean (median) length of stay (LOS) for the index admission was 3.7 +/- 4.3 (3.0) days. Patients in the PR group had a trend toward a longer index admission LOS than the NR group (5.4 +/- 8.8 [3.0] versus 3.5 +/- 3.3 [3.0], p = 0.068). Univariate analysis demonstrated that patients with inflammatory bowel disease, pulmonary comorbidities, and steroid use were more likely to be readmitted. Multivariate analysis confirmed that inflammatory bowel disease and pulmonary comorbidity are independent risk factors for readmission.

Conclusions: Early readmission after laparoscopic colon and rectal operations is not associated with early discharge. Identification of specific patient characteristics indicating risk for early readmission may allow for selective changes in perioperative care or discharge criteria to avoid unexpected readmission. An erratum to this article can be found at