Triglyceride glucose index and Atherogenic index of plasma for predicting colorectal neoplasms in patients without cardiovascular diseases

Muzhou Han,Hao Wang,Shuyue Yang,Siying Zhu,Guiping Zhao,Haiyun Shi,Peng Li

DOI: https://doi.org/10.3389/fonc.2022.1031259

IF: 4.7

2022-11-14

Frontiers in Oncology

Abstract:Background and aims Colorectal neoplasms (CRN) include colorectal cancer (CRC) and colorectal adenoma (CRA). The relationship between CRN and triglyceride-glucose (TyG) index or between CRN and atherogenic index of plasma (AIP) is unclear. This study aims to investigate the roles of TyG index and AIP in predicting CRN in people without cardiovascular disease (CVD). Methods 2409 patients without CVD underwent colonoscopy were enrolled. Clinical information and relevant laboratory test results of these patients were collected and recorded. According to endoscopic and pathological results, all participants were divided into a neoplasms group and a non-neoplasms group. The TyG index was calculated as ln (TGs×FPG/2), while AIP was calculated as log (TGs/HDL-C). We used uni- and multivariate logistic regression and restricted cubic spline (RCS) to analyze the association between the TyG inedx, AIP and CRN, develop predictive models and construct the nomograms. Receiver operating characteristic (ROC) curves were utilized to evaluate the predictive value for CRN. Results Participants in the neoplasms group were more likely to be older, have higher TyG index, higher AIP and higher rates of fecal occult blood test positivity, and were more likely to be male, smokers and those with the family history of CRC ( P < 0.05). The higher TyG index was related to the higher risk of CRN [OR (95% CI): 1.23 (1.08 - 1.41), P = 0.003]. The higher AIP was related to the higher risk of CRN [OR (95% CI): 1.55 (1.16 - 2.06), P = 0.003]. These two indicators are better for predicting CRN in women than men. The combined use of the TyG index and other independent risk factors (age, sex, smoking status, family history and FOBT) to distinguish CRN was effective, with a sensitivity of 61.0%, a specificity of 65.1% and an AUC of 0.669 (95%CI, 0.639 - 0.698). Likewise, the combined use of the AIP and other independent risk factors to distinguish CRN was also effective, the model had an overall 56.3% sensitivity and 68.7% specificity with an AUC of 0.667 (95%CI, 0.638 - 0.697). Conclusion This study showed that the TyG index and the AIP might be biomarkers that could be used to predict the risk of CRN in patients without CVD.

oncology