Nomogram-based risk prediction model employing serum biomarkers to assess intestinal injury risk in patients with metabolic syndrome

Yongqing Chen^1,2Guangxu Wen³Hongmei Wang^1,2Qiyu Yang⁴Zilang Luo^1,2Xin Wang¹Jing Ouyang^5*Jiadan Yang^1*

• ¹Department of Pharmacy, The First Affiliated Hospital of Chongqing Medical University, Chongqing, China
• ²College of Pharmacy, Chongqing Medical University, Chongqing, China
• ³Department of Gastrointestinal Surgery, The First Affiliated Hospital of Chongqing Medical University, Chongqing, China
• ⁴Department of Radiation Oncology, Chongqing University Cancer Hospital & Chongqing Cancer Institute & Chongqing Cancer Hospital, Chongqing, China
• ⁵Clinical Research Center, Chongqing Public Health Medical Center, Chongqing, China

Objective: Patients with metabolic syndrome (MetS) are more likely to have intestinal injury that may accelerate the disease process. We developed a risk prediction model for the non-invasive, rapid, and accurate assessment of intestinal injury in patients with MetS based on serum biomarkers.Methods: Patients with MetS who underwent colonoscopy were enrolled in this study. Based on the results of the colonoscopy, the participants were divided into the intestinal injury and non-intestinal injury groups. Blood samples were collected to detect laboratory indicators and quantify serum biomarkers. Univariate and multivariate logistic regression analyses were employed to identify predictors of intestinal injury in patients with MetS and to construct a nomogram-based risk prediction model. We employed bootstrapping and 5-fold cross-validation to validate the model internally, with the area under the curve (AUC) used to assess the predictive efficacy, the calibration curve utilized to evaluate the calibration degree, and decision curve analysis (DCA) used to evaluate the clinical practicability of the model.Results: The study included 263 participants. Our multivariate logistic regression analysis indicated that clinical features such as age, body mass index, neutrophil percentage, as well as serum biomarkers including diamine oxidase and lipopolysaccharide, were predictive factors for intestinal injury in patients with MetS. The model had strong repeatability (bootstrap method: precision: 0.873, 5-fold cross-validation: AUC: 0.948±0.012), differentiation (AUC: 0.957), and accuracy (Hosmer-Lemeshow χ2 = 3.985, P = 0.858), while DCA results confirmed the clinical utility of the nomogram.Conclusions: Serum biomarkers are effective variables to assess intestinal injury in patients with MetS via our nomogram-based risk prediction model. Clinical trial registration: registration number: ChiCTR2400088476 (retrospectively registered).