Long-term Postprocedural Outcomes of Palliative Emergency Stenting vs Stoma in Malignant Large-Bowel Obstruction.

Importance: Colonic stenting was introduced for palliation of malignant large-bowel obstruction (MLBO) more than 20 years ago but remains controversial. Objective: To compare outcomes after palliative stenting vs stoma creation in patients with MLBO requiring emergency management. Design, Setting, and Participants: This observational cohort study assessed 345 patients from New York State with an urgent or emergency admission to the hospital for obstruction secondary to colorectal cancer and who underwent stenting or stoma creation from October 1, 2009, through December 31, 2013. Patients were excluded if they underwent resection within 1 year of the index admission. Exposures: Palliative stenting vs stoma creation. Main Outcomes and Measures: Primary outcomes included subsequent operation and readmission within 90-day and 1-year follow-up. Secondary outcomes were in-hospital death, major medical and surgical complications, length of stay, total charges, and discharge dispositions. Multivariable hierarchical analyses and propensity score matching were used to compare outcomes between the exposure groups.
Results: The cohort included 345 patients (mean [SD] age, 69.9 [14.4] years in the stoma group and 70.9 [16.8] years in the stent group; 87 men [50.3%] in the stoma group and 90 [52.3%] in the stent group; and 114 non-Hispanic white patients [65.9%] in the stoma group and 90 [52.3%] in the stent group). Most patients undergoing stenting were treated at high-volume (104 [60.5%]) vs medium-volume (42 [24.4%]) or low-volume (26 [15.1%]) hospitals (P < .001). Patients undergoing stenting were significantly less likely to experience prolonged length of stay (odds ratio [OR], 0.50; 95% CI, 0.26-0.97; P = .04), more likely to be discharged to their usual residence (OR, 0.14; 95% CI, 0.07-0.28; P < .001), and tended to have similar or fewer complications (major events: OR, 0.81; 95% CI, 0.30-2.18; P = .68; procedural complications: OR, 0.57; 95% CI, 0.11-1.22; P = .10). There was no significant difference between the groups in terms of 90-day and 1-year readmission to the hospitals (90 days: OR, 0.93; 95% CI, 0.49-1.78; P = .83; 1 year: OR, 0.72; 95% CI, 0.38-1.37; P = .30). Subsequent operation at 90 days was also not different between the groups (OR, 1.34; 95% CI, 0.26-6.89; P = .72), but there was a higher chance of subsequent operation at 1 year after the stenting procedure (OR, 2.93; 95% CI, 1.12-7.68; P = .03), with most subsequent operations being restenting. Conclusions and Relevance: In patients with MLBO and if resection is not part of the treatment plan, stenting is safe and improves the efficiency of care with obvious quality-of-life benefits. It should be offered at experienced centers, and patients should be counseled regarding increased risk of subsequent stenting within 1 year.