Background: Malignant bowel obstruction (MBO) secondary to peritoneal carcinomatosis carries a grave prognosis. We evaluated clinicopathologic factors that predict outcomes after palliative operations for MBO.
Methods: Data on patients undergoing laparotomy for palliation of gastrointestinal MBO at City of Hope between 1995 and 2000 were retrospectively collected. Successful palliation was defined as the ability to tolerate solid food (TSF).
Results: Sixty-three patients underwent operative treatment. In 20 patients, MBO was the first presentation of disease; for others, the median disease-free interval was 15 months. The complication rate was 44%, and postoperative mortality was 15%. The median length of stay was 12 days. Twenty-nine patients (45%) were discharged from the hospital on a regular diet; 22 (76%) continued to eat until their last follow-up. Median survival was 90 days. Univariate factors for longer survival were TSF on discharge, colorectal primary, and nonmetastatic status at first diagnosis. Patients with ascites and whose cancer first presented with MBO had an inferior survival. Noncolorectal primary remained a multivariate predictor for decreased survival. TSF was predicted by the absence of ascites, an obstruction not involving the small bowel, and a preoperative albumin of >3.0 mg/dl. Multiple logistic regression analysis yielded presence of ascites and small-bowel obstruction as predictors of inability to TSF.
Conclusions: Only one third of patients with MBO from peritoneal carcinomatosis will have prolonged postoperative palliation with significant, but acceptable, treatment-related morbidity. TSF at discharge is a useful predictor of continued palliation for most patients. Patients with colorectal cancer may have superior survival outcome and better palliation; others are at risk for poor outcomes, especially in the presence of ascites and MBO of small bowel. In these patients, highly selective use of laparotomy is recommended.