Hypotheses: The Atherothrombosis Intervention in Metabolic Syndrome with Low HDL/High Triglycerides and Impact on Global Health Outcomes (AIM HIGH) tested whether 1500-2000 mg of extended-release niacin (ERN) would reduce cardiovascular disease (CVD) risk in 3,414 patients with CVD, low high density lipoprotein cholesterol (HDL-C < 40 mg/dL), and moderate triglyceride elevation (TG, 150 - 400 mg/dL) already on statin treatment. No benefit of ERN versus placebo on the primary endpoints was noted, except in the most dyslipidemic subgroup with HDL-C < 33 mg/dL and TG > 198 mg/dL (hazard ratio 0.74, p=0.07, JACC 2013; 62:1580-4). In the HDL Atherosclerosis Treatment Study (HATS) we documented that about 2300 mg/day of niacin and simvastatin increased apoA-I in very large α-1 HDL by 115%, and this increase was related to lack of CVD progression (ATVB 2003;23:847-52). Our hypothesis was that the lack of benefit in in AIM HIGH might be related to a lack of changes in HDL particles, adiponectin, myeloperoxidase (MPO), and serum amyloid A (SAA).
Methods: In an AIM-HIGH subset (n=2,333) we measured plasma apolipoprotein A-I (apoA-I) levels in HDL particles (preβ-1, α-4, α-3, α-2, and α-1) by gel electrophoresis and immunoblotting, and plasma adiponectin, MPO, and SAA by immunoassay at baseline and 1 year.
Results: As compared to the placebo group, the ERN group had significantly (p<0.001) greater mean % changes from baseline for HDL-C (+16%), apoA-I in very large α-1 HDL (+19%), medium α-3 HDL (-9.5%), α-4 HDL (-10.0%) and preβ-1 HDL (-7.1%), and in adiponectin (+67%), with no significant differences in changes in MPO or SAA. Similar changes were seen in the most dyslipidemic subgroup. At baseline for the entire group or on trial for the entire group, the ERN group, or the placebo group, none of these parameters or changes in these parameters were significantly related to new CVD events. The mean increase in apoA-I in α-1 HDL at 21% with ERN versus baseline was much less than what we observed in HATS.
Conclusions: Our data indicate that ERN in AIM HIGH induced substantially lower increases in very large HDL particles than was observed in HATS with niacin and simvastatin, and this difference may have accounted for some of the lack of benefit on CVD risk reduction in AIM HIGH.