Abstract 4226: Elevated Levels of Lp(a) Predict Coronary Atherosclerosis as Detected by Coronary Artery Calcium Scanning with Computed Tomography
INTRODUCTION Lipoprotein (a) [Lp(a)] has been established as an independent cardiac risk factor. There is considerable variation among different races and the association between Lp(a) and coronary artery calcium (CAC) is unknown in Asians.
METHODS 104 patients (pts) from Jakarta, Indonesia underwent CAC scoring and measurements of various biomarkers including Lp(a) (DAIICHI Pure Chemicals, Japan). Lp(a) >20mg/dl were defined as abnormal; CAC was expressed as Agatston score. Lp(a) values were not normally distributed and Kruskal-Wallis test and median values were used to test Lp(a) levels across CAC ranges (0, 1–100, 101– 400, >400). ROC curve analysis tested the performance of Lp(a) to predict presence of any CAC.
RESULTS 53/104 (51%) pts had an abnormal Lp(a) level. Clinical characteristics were similar between the two groups (60% male, 14% DM, 57% HTN, 30% smoker). Pts with abnormal Lp(a) had higher mean total cholesterol (220±48 vs. 199±147, p=0.01), LDL (147±40 vs. 130±38, p =0.02) and Lp(a) level (10.2±5.2 vs. 40.8±5.2, p=0.001) but similar HDL (46±10 vs. 45±9, p=NS) and triglyceride (154±113 vs. 145±95, p=NS) levels. Lp(a) was a good test in predicting presence of CAC as evidenced by an area under the curve (AUC)=0.7 with the optimal cut-point of >17mg/dl (sensitivity 73%, specificity 73%, p<0.001). Kruskal-Wallis analysis showed median Lp(a) levels to be significantly higher across rising CAC ranges: 0=41.5, 1–100=54.9, 101– 400=62.1, >400=64 mg/dl (p=0.026). After adjusting for the above clinical characteristics and biochemical markers in a multivariate logistic regression model Lp(a) levels remained a significant predictor of CAC (p=0.007).
CONCLUSION This is the first study to demonstrate a significant relationship between Lp(a) levels and presence of CAC in an Asian population.