ทำออสโตมหรอตอลำไส? ปจจยทำนายอสระของการทำออสโตมในการผาตดลำไสใหญและทวารหนกแบบฉกเฉน ณ โรงพยาบาลทวไประดบจงหวดทมทรพยากรจำกด: การศกษายอนหลง 10 ป (พ.ศ.2558-2568)

Objective: The intraoperative decision to restore bowel continuity or create a stoma during emergency colorectal resection is complex, and the factors that drive this decision remain poorly characterised, particularly in resource-limited settings. We aimed to identify independent predictors of stoma creation among adults undergoing emergency colorectal resection at a Thai provincial referral hospital. Materials and Methods: We conducted a retrospective cohort study of consecutive adult patients undergoing emergency colorectal resection at Phatthalung Hospital, a provincial referral hospital in southern Thailand, between January 2015 and December 2025. Candidate predictors spanning eight pre-specified domains—demographic characteristics, body mass index, physiological reserve, physiological compromise, immunonutritional status, disease characteristics, surgeon factors, and operative context—were evaluated using univariable and Firth penalised multivariable logistic . A full model incorporating all covariates and a physiologic model excluding disease location (to isolate non-anatomical predictors) were compared. Results: Of 371 patients, 190 (51.2%) underwent stoma creation. Disease location was the dominant predictor: only 1 of 93 patients (1.1%) with right colonic disease received a stoma, compared with 43.9% with left colonic and 75.5% with sigmoid or rectal disease. After adjustment, ASA class III (aOR 2.53), high-risk MPI (aOR 5.00), and surgeon experience 5–10 vs <5 years (aOR 2.83) were associated with higher odds of stoma creation, while morbid obesity was associated with lower odds (aOR 0.19). The full model demonstrated excellent discrimination (area under the curve [AUC] 0.889); the physiologic model excluding disease location retained acceptable discrimination (AUC 0.697). Thirty-day mortality was similar between the stoma and anastomosis groups (8.4% vs 9.4%), but only 15 of 190 stomas (7.9%) were subsequently reversed, predominantly because of cancer progression or medical unfitness for further surgery. Conclusion: In this resource-limited provincial setting, half of all patients undergoing emergency colorectal resection received a stoma, driven primarily by disease location, peritonitis severity, and physiological reserve rather than by surgeon or timing factors alone. Given the very low reversal rate, the decision to divert should be regarded as a de facto decision for a permanent stoma in most patients, reinforcing the need for objective decision-support tools and realistic preoperative counselling in resource-limited surgical settings.