Construction of a diagnostic model for coronary artery stenosis in elderly patients with T2DM based on Lp-PLA2 and hs-cTnI

Objective This study aimed to investigate the associations between lipoprotein-associated phospholipase A2 (Lp-PLA2), high-sensitivity cardiac troponin I (hs-cTnI) and coronary artery stenosis in elderly patients with type 2 diabetes mellitus (T2DM) by a retrospective study design, and to construct a diagnostic model for identifying coronary artery stenosis in this population. Methods Totally 706 elderly patients with T2DM who underwent coronary angiography between February 2024 and February 2026 were enrolled and classified into a coronary artery stenosis group (n = 292) and a group without obstructive coronary artery stenosis (n = 414) based on the coronary angiographic findings. The clinical data and laboratory indicators of the patients were collected, and between-group comparisons were performed by independent-samples t-test or chi-square test. The enrolled 706 patients were randomly assigned in a 7:3 ratio into a training set (n = 494) and a validation set (n = 212). In the training set, multivariate Logistic regression analysis was conducted to identify the independent risk factors, and a diagnostic model was constructed. The model was internally validated in the validation set. The discriminatory ability of the model was assessed by receiver operating characteristic (ROC) curve. Incremental value was assessed by comparing nested Logistic regression models with the DeLong test; model calibration was quantified by the calibration intercept, calibration slope and Brier score; and internal optimism was quantified by bootstrap validation with 1,000 resamples repeating the entire model-building process. Results Age, duration of T2DM, glycated hemoglobin (HbA1c), fasting blood glucose (FBG), lipoprotein(a) [Lp(a)], Lp-PLA2, hs-cTnI and high-sensitivity C-reactive protein (hs-CRP) were significantly higher in the coronary artery stenosis group than those in the non-coronary artery stenosis group (P < 0.001). Multivariate Logistic regression analysis indicated that age (odds ratio [OR] = 1.229, 95% confidence interval [CI]: 1.158–1.303, P < 0.001), duration of T2DM (OR = 1.098, 95%CI: 1.010–1.193, P = 0.028), Lp-PLA2 (OR = 1.024, 95%CI: 1.016–1.033, P < 0.001) and hs-cTnI (OR = 1.377, 95%CI: 1.271–1.492, P < 0.001) were independent risk factors for coronary artery stenosis in elderly patients with T2DM. The area under the ROC curve (AUC) of the diagnostic model based on the above four factors was 0.952 (95% CI: 0.934–0.970) in the training set, with a sensitivity of 0.912 and a specificity of 0.879. In the validation set, the AUC, sensitivity and specificity were 0.959 (95%CI: 0.935–0.984), 0.886 and 0.919 respectively. The full diagnostic model demonstrated significantly higher discriminatory ability than the conventional clinical model incorporating age and duration of T2DM alone (ΔAUC = 0.271, P < 0.001), and the optimism-corrected AUC by bootstrap validation was 0.948. Conclusion Lp-PLA2, hs-cTnI, age and duration of T2DM are independent risk factors influencing coronary artery stenosis in elderly T2DM patients. The diagnostic model constructed based on these factors demonstrates good discriminatory ability and may serve as an auxiliary tool for the clinical identification of obstructive coronary artery stenosis in elderly T2DM patients referred for coronary angiography, pending external validation. Clinical trial number Not applicable.