A new model for prediction of bowel gangrene in sigmoid volvulus

BACKGROUND: Sigmoid volvulus is a pathology that can be mortal because it is frequently encountered in elderly patients. In case of bowel gangrene, mortality and morbidity increase further. We planned a retrospective study, in which the effectiveness of the model was evaluated by creating a model that aims to predict whether intestinal gangrene is present in patients with sigmoid volvulus only by blood tests and thus to quickly guide treatment methods. METHODS: In addition to demographic data such as age and gender, laboratory values such as white blood cell, C-reactive protein (CRP), lactate dehydrogenase (LDH), potassium, and colonoscopic findings and whether there was gangrene in the colon during the operation were evaluated retrospectively. In the analysis of the data, independent risk factors were determined by univariate and multivariate logistic regression analyzes as well as Mann–Whitney U and Chi-square tests. Receiver operating characteristic (ROC) analysis was performed for statistically significant continuous numerical data, and cutoff values were determined and Malatya Volvulus Gangrene Model (MVGM) was created. The effectiveness of the created model was again evaluated by ROC analysis. RESULTS: Of the 74 patients included in the study, 59 (79.7%) were male. The median age of the population was 74 (19–88), and gangrene was detected in 21 (28.37%) patients at surgery. In univariate analyzes, leukocytes <4000/mm(3) and >12000/mm(3) (OR: 10.737; CI 95%: 2.797–41.211, p=0.001), CRP ≥0.71 mg/dl (OR: 8.107 CI 95%: 2.520–26.082, p<0.0001), potassium ≥3.85 mmoL/L (OR: 3.889; 95% CI): 1.333–11.345, p=0.013), and LDH ≥288 U/L (OR: 3.889; CI 95%: 1.333–11.345, p=0.013), whereas, in multivariate analyzes, only CRP ≥0.71 mg/dL (OR: 3.965; CI 95%: 1.071–15.462, p=0.047) was found to be an independent risk factor for bowel gangrene. The strength of MVGM was AUC 0.836 (0.737–0.936). In addition, it was observed that the probability of bowel gangrene increased approximately 10 times if MVGM was ≥7 (OR: 9.846; 95% CI: 3.016–32.145, p<0.0001). CONCLUSION: Besides being non-invasive compared to the colonoscopic procedure, MVGM is a useful method for detecting bowel gangrene. In addition, it will guide the clinician in taking the patients with intestinal loop gangrene to emergency surgery without wasting time in the treatment steps, as well as avoiding complications that may occur during colonoscopy. In this way, we think that morbidity and mortality rates can be reduced.