Small bowel obstruction: a population-based appraisal.

Abstract

BACKGROUND: Small bowel obstruction (SBO) is a common reason for surgical consultation, but little is known about the natural history of SBO. We performed a population-based analysis to evaluate SBO frequency, type of operation, and longterm outcomes. STUDY DESIGN: Using the California Inpatient File, we identified all patients admitted in 1997 with a diagnosis of SBO. Patients were excluded if they had a diagnosis of bowel obstruction in the previous 6 years (1991 to 1996). Of the remaining cohort, the natural history of SBO over the subsequent 5 years (1998 to 2002) was analyzed. Index hospitalization outcomes (eg, surgical versus nonsurgical management, length of stay, in-hospital mortality), and longterm outcomes, including SBO readmissions and 1-year mortality, were evaluated. RESULTS: We identified 32,583 patients with an index admission for SBO in 1997; 24% had surgery during the index admission. The distribution of surgical procedures was: 38% lysis of adhesions, 38% hernia repair, 18% small bowel resection with lysis of adhesions, and 6% small bowel resection with hernia repair. Patients who underwent operations during index admission had longer lengths of stay, lower mortality, fewer SBO readmissions, and longer time to readmission than patients treated nonsurgically. Regardless of treatment during the index admission, 81% of surviving patients had no additional SBO readmissions over the subsequent 5 years. CONCLUSIONS: Most of the 32,583 patients requiring admission for index SBO in 1997 were treated nonsurgically, and few of these patients were readmitted. This is the first longitudinal population-based analysis of SBO evaluating surgical versus nonsurgical management and outcomes, including mortality and readmissions.