Early predictive value of triglyceride-glucose index combined with controlling nutritional status score for severe acute pancreatitis
10.3760/cma.j.cn115667-20240801-00130
- VernacularTitle:三酰甘油-葡萄糖指数联合控制营养状态评分对重症急性胰腺炎的早期预测价值
- Author:
Wei LI
1
;
Chenyi SHE
;
Yujie CHEN
;
Jun CHENG
;
Song ZHANG
;
Weitian XU
;
Qingming WU
Author Information
1. 武汉科技大学医学部医学院，武汉 430065
- Publication Type:Journal Article
- Keywords:
Severe acute pancreatitis;
Triglyceride-glucose index;
Score of impaired nutritional status;
Metabolic disturbance
- From:
Chinese Journal of Pancreatology
2025;25(3):183-189
- CountryChina
- Language:Chinese
-
Abstract:
Objective:To explore the early predictive value of the triglyceride-glucose (TyG) index and the controlling nutritional status (CONUT) score for severe acute pancreatitis (SAP).Methods:Clinical data from 1 050 hospitalized patients with acute pancreatitis (AP) at the General Hospital of Central Theater Command between January 2019 and December 2023 were retrospectively analyzed. Patients were categorized into mild acute pancreatitis (MAP) group ( n=606), moderately severe acute pancreatitis (MSAP) group ( n=320), and SAP group ( n=124) based on AP severity. General clinical data, laboratory parameters, modified computed tomography severity index (MCTSI), bedside index for severity in acute pancreatitis (BISAP), TyG index, and CONUT score were compared among the three groups. Spearman correlation analysis was used to evaluate the relationship between TyG index, CONUT score and AP severity. Univariate and multivariate logistic regression analyses were performed to identify independent risk factors for AP severity. Receiver operating characteristic curves (ROC) were plotted to calculate the area under the curve (AUC), sensitivity, and specificity for evaluating the predictive efficacy of TyG index, CONUT score, and their combination for SAP. Results:Significant differences on TyG index and CONUT score were observed among AP patients with varying severity (all P value <0.001). Spearman correlation analysis further revealed positive correlations of TyG index ( r=0.174), CONUT score ( r=0.306) with AP severity (both P<0.001). Multivariate logistic regression identified neutrophil count ( OR=1.076, 95% CI 1.027-1.125), MCTSI ( OR=2.565, 95% CI 2.250-2.921), BISAP ( OR=3.522, 95% CI 2.726-4.549), TyG index ( OR=1.859, 95% CI 1.276-2.707), and CONUT score ( OR=1.155, 95% CI 1.035-1.288) as independent risk factors for AP severity. The combined predictive model equation was: -7.342+0.76×TyG+0.439×CONUT. ROC analysis showed that the AUC values of the TyG index, CONUT score, and the combined index (TyG index+CONUT score) were 0.583 (95% CI 0.529-0.637), 0.701 (95% CI 0.652-0.75), and 0.755 (95% CI 0.710-0.801), with sensitivities of 0.706, 0.677, and 0.742, and specificities of 0.884, 0.629, and 0.657, respectively (all P value <0.05). Conclusions:TyG index and CONUT score are positively correlated with AP severity and may serve as reliable predictors for SAP. Their combination could enhance the predictive accuracy for AP.
