Abstract 823: C-reactive Protein and Amyloid A as Predictors of Cardiovascular Events in Type 2 Diabetes in CARDS
Background: Inflammatory markers such as C-reactive protein (CRP) and serum amyloid A (sAA) have been proposed as biomarkers of cardiovascular disease (CVD), but information in diabetes is limited.
Methods: We measured CRP and sAA (both mg/L) at baseline and after 12 months of treatment in the randomised Collaborative Atorvastatin Diabetes Study (CARDS), which demonstrated efficacy of 10 mg atorvastatin in the primary prevention of CVD in patients with type 2 diabetes. We used results for CRP and sAA available in 87% of the CARDS patients (placebo [n=1215] and atorvastatin [n=1250]) in whom a total of 181 primary events occurred. The median duration of follow-up was 3.9 years. CRP and sAA were measured in plasma using latex particle-enhanced immunonephelometric assays with the BN Prospec® system (Dade-Behring UK Ltd). Time to first CVD event was evaluated by Cox proportional hazards models using quartiles of baseline biomarker concentrations with adjustment for treatment allocation and relevant covariates.
Results: At baseline, patients on placebo and atorvastatin had similar concentrations of CRP (median [interquartile range] of 1.5 [0.6 –3.6] vs. 1.3 [0.6 –3.1]) and sAA (3.0 [1.7–5.2] vs. 2.9 [1.6 – 4.9]). CRP correlated with sAA both at baseline and at 12 months (r=0.66 and r=0.65; p<0.0001). At 1 year follow up, there was an increase in within-person CRP in the placebo group of 0.2 [−0.8 –1.5], that was significantly different (p<0.0001) from the decrease in the atorvastatin group, −0.1[−1.0 – 0.7]. The within-person change in sAA concentrations at 1 year did not differ significantly between treatment groups. Baseline CRP was not associated with CVD during follow up (HR varied from 0.9 to 1.2 for comparisons between quartiles; p>0.2). Patients with CVD events had a slight increase in CRP (0.1, [−0.8 –1.3]), not significantly different (p=0.13) from patients without events (0, [−0.9 –1.1]). Similar results were observed for sAA.
Conclusion: CRP is reduced by atorvastatin relative to placebo. In these diabetic patients with relatively low baseline CRP, reduction in CVD events from atorvastatin treatment was seen in patients with and without elevated CRP levels.