New Diagnostic Model Predicts Coronary Artery Stenosis in Seniors with Diabetes

A retrospective study has unveiled a promising diagnostic model for detecting obstructive coronary artery stenosis in elderly patients with type 2 diabetes mellitus (T2DM). This model integrates two key biomarkers—lipoprotein-associated phospholipase A2 (Lp-PLA2) and high-sensitivity cardiac troponin I (hs-cTnI)—alongside demographic and clinical data.

Study Overview and Methodology

Conducted between February 2024 and February 2026, the research analyzed a sample of 706 older adults diagnosed with T2DM who underwent diagnostic coronary angiography. Participants were divided into two groups based on their angiographic results: one group consisting of 292 individuals with coronary artery stenosis and the other comprising 414 patients without significant arterial narrowing.

Key Findings

Initial comparisons between the two groups revealed significant differences in various clinical parameters and biomarker levels. Specifically, those with coronary artery stenosis were generally older, had a longer history of diabetes, and showed elevated levels of glycated hemoglobin (HbA1c), fasting blood glucose (FBG), Lp(a), Lp-PLA2, hs-cTnI, and hs-CRP. These differences were statistically significant, with p-values below 0.001.

Creating the Predictive Model

To develop an accurate predictive model, researchers randomly assigned the participants in a 7:3 ratio into a training group of 494 patients and a validation group of 212 patients. The analysis utilized multivariate logistic regression to identify independent risk factors for coronary artery stenosis. Four notable predictors emerged: advanced age (odds ratio [OR] = 1.229), duration of diabetes (OR = 1.098), elevated Lp-PLA2 (OR = 1.024), and increased hs-cTnI (OR = 1.377).

Model Performance

When combined into a single diagnostic model, these four factors achieved an impressive area under the receiver operating characteristic curve (AUC) of 0.952, demonstrating a sensitivity of 0.912 and a specificity of 0.879. The model's reliability was further confirmed through tests in the internal validation cohort, where it yielded an AUC of 0.959, maintaining strong sensitivity (0.886) and specificity (0.919).

Incremental Value of Biomarkers

The study also assessed the additional diagnostic benefit of including Lp-PLA2 and hs-cTnI compared to a traditional model that considered only age and diabetes duration. Utilizing the DeLong test, researchers found a significant enhancement in predictive capability, with an increase in discriminatory capacity by ΔAUC = 0.271 (p < 0.001).

Robustness of the Model

To address potential overfitting, the authors performed internal bootstrap validation with 1,000 resamples, which confirmed a stability in the model’s performance, yielding an optimism-corrected AUC of 0.948.

Conclusion and Future Directions

The study concludes that while this new diagnostic model shows great potential as a clinical tool for identifying obstructive coronary artery disease in older adults with diabetes, it is important to acknowledge the limitations of the retrospective nature of the research. Further prospective studies with diverse populations will be necessary to validate these findings and potentially implement the model into clinical practice.