Differential associations of angiographic extent and severity of coronary artery disease with asymmetric dimethylarginine but not insulin resistance in non-diabetic men with stable angina: a cross-sectional study

BACKGROUND Asymmetric dimethylarginine (ADMA), an endogenous nitric oxide synthesis inhibitor, and insulin resistance (IR) have been implicated in atherogenesis. Our aim was to estimate relations between ADMA, the magnitude of IR and angiographic indices of extent and severity of coronary atherosclerosis in non-diabetic men with stable coronary artery disease (CAD). METHODS We studied 151 non-diabetic men (mean age 57 ± 11 years) with stable angina, obstructive CAD (at least 1 luminal diameter stenosis of ≥70% in major coronary segments) and without heart failure, and 34 age-matched controls free of ≥50% coronary narrowings. The following CAD indices were computed: the number of major epicardial vessels with ≥70% stenosis, Sullivan extent score representing a proportion of the visible coronary tree with vessel wall irregularities, and Gensini score which reflects both CAD severity and extent, yet assigning a heavier weight to proximal segments and to the more severe narrowings by a non-linear point system. An estimate of IR was derived by homeostasis model assessment (HOMA-IR) from fasting insulin and glucose. RESULTS Among the CAD patients, the proportions of subjects with 1-vessel, 2- vessel and 3-vessel CAD were 26%, 25% and 49%, respectively. ADMA levels were higher in patients with obstructive CAD compared to the controls (0.51 ± 0.10 vs. 0.46 ± 0.09 μmol/L [SD], P = 0.01), whereas HOMA-IR was similar (median, 3.2 [interquartile range: 2.4-4.9] vs. 2.9 [2.3-4.7], P = 0.2). Within the CAD group, ADMA increased across ascending quartiles of Sullivan score (Spearman's rho = 0.23, P = 0.004), but not with Gensini score (rho = 0.12, P = 0.15) or the number of vessels involved (rho = 0.08, P = 0.3). ADMA correlated to log-transformed Sullivan score (Pearson's r = 0.21, P = 0.008), which was only slightly attenuated upon multivariate adjustment (β = 0.19 ± 0.08 [SEM], P = 0.015). HOMA-IR did not differ according to any measure of angiographic CAD (P ≥ 0.2). ADMA and log (HOMA-IR) were mutually unrelated (r = 0.07, P = 0.4). CONCLUSIONS ADMA is associated with diffuse but not focal coronary atherosclerosis in non-diabetic men with stable CAD irrespectively of the degree of IR. The independent relationship between ADMA and coronary atherosclerotic burden may contribute to the well-recognized prognostic effect of ADMA in CAD.