Association of Serum Adiponectin with Coronary Artery Disease and Multi-Vessel Involvement: A Case-Control Study

Background: Adiponectin is an adipocyte-derived protein with anti-atherogenic, anti-inflammatory, and insulin-sensitising properties, and hypoadiponectinaemia has been linked to coronary artery disease (CAD). This study evaluated the association of serum adiponectin with angiographically confirmed CAD compared with healthy individuals, and explored its relationship with the anatomical extent of disease. Methods: In this case-control study conducted between January and May 2026, serum adiponectin was measured by enzyme-linked immunosorbent assay in 60 adults with CAD confirmed by coronary angiography (≥50% luminal stenosis in at least one major coronary artery) and 40 healthy controls. Patients were stratified by the number of diseased vessels (single, double, triple). Group differences were assessed using the Welch t-test and Mann–Whitney U test; the extent relationship was examined by one-way analysis of variance (ANOVA), the Kruskal–Wallis test, and Spearman correlation. Multivariable logistic regression, using Firth's penalized method to accommodate near-complete separation between groups, tested whether adiponectin was independently associated with CAD after adjustment for age, diabetes, body mass index, hypertension, lipid parameters, smoking, and sex. Discrimination was assessed by receiver-operating-characteristic (ROC) analysis. Results: Serum adiponectin was markedly lower in CAD patients than in controls (6.85 ± 2.39 vs 14.45 ± 3.72 ng/mL; p < 0.001). The biomarker showed high discriminatory performance for CAD in this cohort (area under the curve [AUC] 0.955, 95% CI 0.91–0.99); a cut-off of 9.59 ng/mL yielded 90.0% sensitivity and 92.5% specificity, with a positive predictive value of 94.7% and a negative predictive value of 86.0%. Adiponectin remained independently associated with CAD after multivariable adjustment (adjusted odds ratio [OR] 0.51 per 1 ng/mL, 95% CI 0.34–0.74, p = 0.0005) and after direct adjustment for fasting glucose (adjusted OR 0.50, 95% CI 0.33–0.75, p = 0.0009), and adding adiponectin to a clinical-risk-factor-only model increased discrimination from AUC 0.940 to 0.992 (ΔAUC = 0.052, p < 0.0001). Adiponectin decreased progressively across single-, double-, and triple-vessel disease (7.64 ± 2.86, 6.79 ± 2.22, 6.24 ± 1.99 ng/mL), but this trend did not reach statistical significance (ANOVA p = 0.185; Spearman ρ = –0.24, p = 0.070); post-hoc power for this comparison was low (≈ 0.37). Adiponectin correlated inversely with fasting glucose (r = –0.41, p < 0.001) and age (r = –0.24, p = 0.017). Conclusions: Serum adiponectin showed strong discriminatory ability for CAD presence in this cohort, independent of the clinical covariates included in the multivariable model. A weak inverse association with multi-vessel involvement did not reach statistical significance and warrants evaluation in larger, adequately powered cohorts.