Association of Calculated Lipid Indices with Coronary Artery Disease severity (assessed by SYNTAX score) and Their Role as Biomarker for Detection of Severe Coronary Artery Disease

Performance of Calculated Lipid Indices as Predictors for Severity

Abstract

Background: Coronary Artery Disease (CAD) is a leading cause of morbidity and mortality globally, with its rising prevalence, especially in South Asia. Dyslipidemia is a well-established risk factor for CAD. Recent research has focused on novel lipid indices like atherogenic Index (AI), atherogenic index of plasma (AIP), lipoprotein combine index (LCI), castelli risk index- I (CRI-I) & castelli risk index-II (CRI-II). These are calculated simply from fasting lipid profile components and they can act as potential predictors of CAD severity. Therefore, these lipid indices offer more comprehensive assessment of atherogenic risk and CAD severity compared to individual lipid components.

Objectives: To evaluate the association of calculated lipid indices with severity of CAD and to assess their role as diagnostic biomarker for detection of coronary artery disease severity (assessed by SYNTAX score).

Methods: This cross-sectional study was conducted at Sir Salimullah Medical College (SSMC) from March 2024 to February 2025. Total 218 patients (age 18- 70 years) with CAD were purposively selected as study subjects from the Cardiology department of Sir Salimullah Medical College Mitford Hospital (SSMCMH) and National Institute of Cardiovascular Diseases (NICVD), Dhaka. The demographic, clinical, and biochemical data were collected. Fasting lipid profile was measured in every patient and lipid indices were calculated using the values of lipid profile components. For all patients coronary angiogram were done, SYNTAX score were calculated to assess CAD severity. All (218) subjects were divided into three groups according to their SYNTAX score; mild CAD (low risk, SYNTAX Score 0 - 22), moderate CAD (intermediate risk; SYNTAX Score 23 - 32) and severe CAD (high risk; SYNTAX Score of > 32). The associations of lipid indices with CAD severity was analyzed by unpaired t-test, spearman rank correlation coefficient test and logistic regression test. To evaluate the diagnostic performance of lipid indices for detection of CAD severity, their cutoff points were determined by Youden Index. Using those cutoff points diagnostic performance of lipid indices were evaluated by ROC (receiver operator characteristic) curve analysis, AUC (area under curve) and performance tests (sensitivity, specificity, positive predictive value, negative predictive value, accuracy). Statistical significance was set at p≤0.05.

Results: Unpaired t-test, Spearman rank correlation coefficient test, logistic regression analysis revealed positive correlation and association of calculated lipid indices with the CAD severity. The CRI-II had the highest (0.647) area under the curve (AUC), and AIP had lowest (0.606) AUC. As predictors of CAD severity AI, and CRI-I showed 68.8% sensitivity, 54.8% specificity, 86.43% PPV, 29.48% NPV, 66.05% accuracy with cutoff point 3.0 and 4.0 respectively. AIP and LCI showed 71.6% sensitivity, 52.4% specificity, 86.3% PPV, 30.55% NPV, 67.89% accuracy with cutoff point 0.13 for both. CRI-II showed 62.5% sensitivity, 59.5% specificity, 86.61% PPV, 27.47% NPV, 61.93% accuracy with cutoff point 4.0.

Conclusion: The calculated lipid indices were found to be positively associated with CAD severity and positively correlated with SYNTAX score. The diagnostic performance of calculated lipid indices for detection of CAD severity was not satisfactory because their AUC, sensitivity and specificity were not promising.

Bangladesh Medical Res Counc Bull 2025;51(3): 110-116