A Bedside Prediction Score for Strangulation in Adults With Acute Mechanical Intestinal Obstruction: A Prospective...

Background Early recognition of bowel strangulation in acute mechanical intestinal obstruction (AMIO) is critical but remains difficult when computed tomography (CT) is unavailable, delayed, or not immediately actionable. This study aimed to develop and internally validate a CT-independent bedside prediction score for estimating the risk of bowel strangulation in adults with AMIO in Taiz, Yemen. Materials and methods This 12-month prospective observational cohort study, with diagnostic and prediction-model components, enrolled consecutive adults with suspected AMIO at Al-Thawra Hospital and Al-Jumhuri General Teaching Hospital in Taiz, Yemen. Clinical, laboratory, imaging, operative, and outcome data were recorded prospectively. Final strangulation was defined by operative evidence of bowel ischemia, gangrene, necrosis, nonviability, or resection for compromised bowel; successful conservative resolution without later evidence of bowel compromise was classified as non-strangulated. A CT-independent Strangulated Intestinal Obstruction (SIO) score was derived from a five-predictor logistic model and internally validated with 1,000 bootstrap resamples. Results Of 230 adults, 212/230 (92.2%) underwent surgery, and 109/212 (51.4%) had intraoperative strangulation. The complete-case prediction cohort included 226 patients. The final score retained pain changing to persistent, abdominal distension, tenderness grade, neutrophilia greater than 75%, and systemic inflammatory response syndrome (SIRS) status. The apparent area under the receiver operating characteristic curve (AUC) was 0.953, and the optimism-corrected area under the curve was 0.947. At a score of 7 or higher, sensitivity was 100/109 (91.7%), specificity was 92/117 (78.6%), positive predictive value was 100/125 (80.0%), negative predictive value was 92/101 (91.1%), and accuracy was 192/226 (85.0%). Observed strangulation risk was 1/76 (1.3%) in the low-risk category, 25/63 (39.7%) in the intermediate-risk category, and 83/87 (95.4%) in the high-risk category. Conclusion The score showed promising internal performance as an early triage and escalation aid in this resource-limited pathway. It should support, rather than replace, senior surgical judgment and requires external validation before routine use.