Prediction of severity and outcomes of colon ischaemia using a novel prognostic model: a clinical multicenter study

Abstract

Objective To identify risk factors of disease severity and between mild and severe colon ischaemia (CI) patients and to improve clinical outcomes, this study aimed to explore a novel scoring model. Methods Retrospective analyses of hospital records between January 2009 and December 2019 were included. Clinical manifestations, mortality, Oakland score, laboratory tests, colonoscopy, and histopathology were collected. Risk factors of severe CI were determined by univariate and multivariate logistic regression and used for the predicting model. Results A total of 203 patients with CI were included. Serum C-reactive protein (CRP) and albumin ratio (CAR) were much higher in the severe CI group compared with that of the mild CI group (3.33 ± 1.78 versus 0.68 ± 0.97, p < .001). The Oakland score was much higher in the severe CI group (12.00 ± 3.02 versus 8.77 ± 1.63, p < .001). The histopathological finding of fibrin thrombi was an independent risk factor that predicted poor outcomes (20.00% versus. 1.09%, p < .001). Patients present with CAR ≥3.33, Oakland score ≥12, and histopathological fibrin thrombi were independent risk factors. In addition, the final scoring model was 0.042 × Oakland score + 1.040 × CAR + 3.412 × fibrin thrombi, the area under the curve (AUC) was 0.960 (95% confidence interval:0.930–0.990), and the sensitivity and specificity of the novel scoring model were 95% and 92%, respectively. Conclusions The novel prognostic model was established to predict CI severity and clinical outcomes efficiently. Key messages In this article, we discuss the scoring model for clinical outcomes of colon ischaemia patients. In our study, the sensitivity and specificity of a novel scoring model are very high. Thus, laboratory tests (CRP albumin ratio), Oakland score, and histopathological findings (fibrin thrombi) can be assessed efficiently for colon ischaemia outcomes.