Background: The presence of type 2 diabetes mellitus (T2DM) in cardiovascular disease (CVD) is associated with poorer clinical outcomes. To identify biomarkers is essential to predict T2DM onset and progression in the context of secondary prevention. Telomere length (TL), a marker of cellular aging, has been linked to oxidative stress, inflammation, and metabolic dysfunction, including diabetes-related complications. This study aims to evaluate TL as a potential biomarker for T2DM in patients with coronary heart disease (CHD).
Methods: We included 956 patients from the CORDIOPREV study with available TL data. Participants were classified according to their diabetes status at the end of the dietary intervention: 407 were free of T2DM and 549 were classified as having T2DM. TL was measured by quantitative PCR. Patients with TL values below the 20th percentile of the study population were categorized as having short telomeres.
Results: TL was significantly shorter in T2DM patients compared to non-T2DM (1.26±0.74 vs. 1.38±0.84; p=0.026). Each 1-SD increase in TL was associated with a 17% lower risk of T2DM (OR 0.83, 95% CI 0.69-0.98). Patients at risk of short TL (<20th percentile) showed a higher prevalence of T2DM (20% vs. 17%; p=0.035) and a 42% increased risk of its presence (OR 1.42, 95% CI 0.99-2.02). In stepwise regression models, TL, age, BMI, insulin, triglycerides, and HbA1c emerged as independent predictors of T2DM. After adjusting for fasting glucose, TL lost significance as a continuous variable, but short TL risk remained independently associated with T2DM (OR 1.57, p=0.049).
Conclusions: Shorter TL is associated with the presence of T2DM in patients with CHD, supporting its potential role as an independent biomarker for T2DM risk in secondary prevention.
Introduction: Low-density lipoprotein cholesterol (LDL-C) is a significant cardiovascular risk factor, as direct measurement is expensive and often unavailable in most clinical laboratories. The Friedewald formula (FD), despite its widespread use since 1972, has notable limitations, especially at high triglyceride levels and low LDL-C concentrations. Machine learning (ML) techniques offer promising alternatives for accurate LDL-C estimation, potentially overcoming traditional formula limitations by leveraging complex pattern recognition in lipid profile data.
Material and methods: This retrospective study analyzed 34,678 lipid profiles from patients over 18 years attending Hospital Virgen Macarena, Seville (January 2021-December 2022). The study was approved by the Ethics Committee (CEI HVM-VR_03/2024). All lipid parameters (total cholesterol, triglycerides, HDL-C, LDL-C) were measured using Cobas 6000 analyzer. Twenty-two machine learning models were developed using Python's PyCaret library with 80/20 train-test split. Models included Linear Regression, Random Forest, XGBoost, LightGBM, and Gradient Boosting among others. Performance was evaluated using coefficient of determination (R2), mean absolute error (MAE), and root mean square error (RMSE). Four triglyceride subgroups were analyzed: <150, 150-250, 250-400, and >400mg/dL.
Results: The dataset comprised 34,678 individuals with mean values: total cholesterol 204.6±73.36mg/dL, triglycerides 203.95±143.94mg/dL, HDL-C 51.83±18.45mg/dL, and LDL-C 120.38±62.29mg/dL. LightGBM achieved the highest performance (R2=0.965, RMSE=11.35, MAE=7.99), followed by Gradient Boosting (R2=0.962, RMSE=11.89, MAE=7.87) and XGBoost (R2=0.958, RMSE=12.49, MAE=8.3). Traditional formulas showed inferior performance: Martin-Hopkins (R2=0.951, RMSE=13.82, MAE=9.3) and Friedewald (R2=0.926, RMSE=16.92, MAE=11.97). Performance differences were more pronounced at triglyceride levels≥250mg/dL, with ML models maintaining R2>0.92 while classical formulas deteriorated significantly, particularly Friedewald (R2=0.34) at triglycerides>400mg/dL.
Conclusions: Machine learning models, particularly boosting algorithms (LightGBM, Gradient Boosting, XGBoost), significantly outperformed traditional LDL-C calculation formulas across all triglyceride ranges. These AI-based approaches yielded superior accuracy and robustness, especially in challenging clinical scenarios with elevated triglycerides where conventional formulas fail. Implementation of ML models in clinical laboratories could provide more reliable LDL-C estimations, contributing to improved cardiovascular risk stratification and patient management. This technological advancement represents a promising transformation in laboratory medicine methodology.
Background: Telomere length (TL) has emerged as a recognized biomarker of biological aging and cardiovascular risk. Several studies have associated short TL with increased incidence of cardiovascular disease and adverse outcomes. However, its value as a predictor of arterial injury remains unclear, especially in secondary prevention. This study examined the baseline association between TL and common carotid intima-media thickness (IMT-CC) in patients with coronary heart disease (CHD).
Methods: Of the 1002 CHD patients enrolled in the CORDIOPREV study, 903 completed baseline assessments of IMT-CC and TL. IMT-CC was measured bilaterally using high-resolution B-mode Doppler ultrasonography and categorized, according to the European Society of Cardiology guidelines, into three groups reflecting arterial injury: <0.7mm, 0.7-0.9mm, and ≥0.9mm. TL was determined using the quantitative PCR method. Patients were classified as at risk of short TL if their value was below the 20th percentile, and as non-risk if above this threshold.
Results: Patients with the highest IMT-CC values (≥0.9mm) had significantly shorter TL compared to those in the lowest group (<0.7mm, p=0.005), with a progressive decrease across IMT-CC categories (p for trend=0.02). The proportion of participants at risk of short telomeres increased significantly from 15% to 30% across these categories (p<0.001). In addition, the highest IMT-CC group showed a twofold greater risk of short telomeres compared to the reference group (<0.7mm; OR 2.373, 95%CI 1.4962-3.764), while no significant association was observed for the intermediate IMT-CC group (0.7-0.9mm; OR 1.352, 95%CI 0.896-2.039).
Conclusions: In CHD patients, shorter TL is associated to advanced subclinical arterial damage, supporting its potential role as a complementary biomarker for vascular aging and improved risk stratification in secondary prevention.

