The Effects of Niacin on Lipoprotein Subclass Distribution

John M. Morgan, MD; Christina M. Carey, PA-C; Anne Lincoff, MD; David M. Capuzzi, MD, PhD

In This Article

Efficacy of Niacin on Lipoprotein Subclasses

Niacin therapy has differential effects on HDL, producing favorable changes in subclasses and apo A-I. In 41 patients with dyslipidemia who received crystalline niacin at 4000 mg/d for 6 weeks, mean serum levels of HDL2cholesterol increased by 135% as assessed using ultracentrifugation techniques (p<0.001).[35] A similar evaluation was performed in 23 patients with dyslipidemia using gradient gel electrophoresis. After 6 weeks of niacin 4000 mg/d,[7] the 45% increase in total HDL cholesterol level was predominantly attributable to a 126% rise (p<0.001) in HDL2 cholesterol. The contribution of HDL2b and HDL2a to total HDL mass increased by 144% (p<0.001) and 17% (p<0.01), respectively. These were paralleled by a 34% decrease in the HDL3b (p<0.001) and 26% decrease (p<0.01) in the HDL3c subparticle proportions.

Our group performed a post-hoc evaluation of baseline and 12-week plasma samples from 60 patients with dyslipidemia who had been randomized to receive once-daily extended-release niacin (niacin ER) at either 1000 mg or 2000 mg, or to placebo.[8] Using a proton nuclear magnetic resonance assay, in the 1000 mg/d niacin group, concentrations of HDL H5 and H4, the two largest of five measured HDL subparticles and the two that make up the more cardioprotective HDL2ab subfraction, were increased by 70% and 39%, respectively (p<0.001 vs. placebo for both). In the 2000 mg/d group, increases were more pronounced at 89% and 75%, respectively (p<0.001 vs. placebo for both). No significant changes were observed in levels of HDL H3, H2, or H1, which collectively correspond to the HDL3abc subfraction (Figure 2).

Effects of niacin on high-density lipoprotein subclasses. ER=extended-release. Reprinted with permission from Am J Cardiol. 2003;91:1432-1436.[8]

Studies have shown that different niacin formulations can have variable effects on HDL subparticle distribution. In a randomized comparison of crystalline niacin with a sustained-release niacin preparation, 71 patients with dyslipidemia were treated with 3000 mg/d for 6 months.[36] Mean HDL2 cholesterol levels increased by 36% on crystalline niacin but decreased by 5% on sustained-release niacin (p<0.05). In contrast, a study comparing crystalline niacin with niacin ER in 223 patients with dyslipidemia showed equivalent efficacy.[37] Mean HDL2 and HDL3 cholesterol levels increased by 37% and 17%, respectively, on niacin ER compared with 33% and 16%, respectively, on crystalline niacin (Figure 3). In addition, apo A-I levels increased by 8% with niacin ER and 6% with crystalline niacin. A similar increase in apo A-I levels (7%), as well as a decrease in apo A-II (-21%), was reported in an early study of crystalline niacin involving five healthy adults.[38]

Effects of niacin on high-density lipoprotein (HDL) subclasses. ER=extended-release; IR=immediate-release or crystalline; *p<0.05 vs. niacin ER and niacin IR 1500 mg/d. Adapted with permission from Metabolism. 1998;47:1097-1104.[37]

Niacin shifts the distribution of LDL subclass from smaller toward larger particles and seems to have a more pronounced effect in patients with pattern B dyslipidemia than in those expressing the pattern A phenotype. One study evaluated 26 patients with CHD and dyslipidemia for changes in gradient gel electrophoresis-measured LDL subclass distribution 3-6 months after starting on crystalline niacin 3000 mg/d.[6] Compared with baseline, mean peak LDL particle diameter increased by 0.74 nm (p<0.0001) overall, and by 0.94 nm among the 17 pattern B patients compared with 0.36 nm among the nine patients with the pattern A phenotype (p=0.022). In addition, mean HDL cholesterol level increased by 13.4 mg/dL (p<0.01) among pattern B patients but by only 0.7 mg/dL (p=NS) in pattern A patients.

In our post-hoc evaluation of 60 patients randomized to niacin ER or placebo, levels of the smaller LDL subparticles L1 and L2 tended to decrease while the larger L3 subparticles tended to increase at both niacin dosages, although these changes were not significant vs. placebo.[8] On the other hand, niacin decreased the number of LDL particles by 15% at 1000 mg/d and by 23% at 2000 mg/d (p=0.002 for both). Both dosages were also associated with significant increases in LDL particle size (p=0.027).

Niacin's effects on LDL subparticle size and distribution have also been evaluated in diabetic populations. In the open-label study of 42 diabetics, 6-8 weeks of crystalline niacin, mean dose 2765 mg/d, was associated with an 0.11 nm increase in mean peak LDL particle diameter (p<0.0001).[39] Mean small, dense LDL mass decreased by 44%, from 27 mg/dL to 15 mg/dL (p<0.0001). Similarly, in the evaluation of 23 persons with diabetes using niacin ER, mean peak LDL particle diameter increased a significant 0.9 nm and mean small, dense LDL mass decreased by 43% (p<0.0001 for both).[40] In two studies comparing niacin preparations with statins in patients with diabetes, niacin had a greater effect on LDL particle size, statins had a greater effect on particle number, and the combination conferred complementary benefits that corrected the entire atherogenic profile.[41,42]

Large VLDL subparticles are frequently found in the presence of other lipoprotein disorders, so their independence as a risk factor is not established. However, studies that show a favorable treatment effect on VLDL-associated TG are suggestive, given the status of elevated TG as an emerging risk factor in the Adult Treatment Panel III guidelines.[1] In our study of 60 patients with dyslipidemia randomized to niacin ER or placebo, total VLDL-associated TG decreased 24% and 37% at 1000 mg/d and 2000 mg/d, respectively.[8] Of the VLDL subparticles V1 through V6 (numbered smallest to largest), the V2, V4, and V5 subfractions accounted for the overwhelming proportion of total VLDL. The lower-dose niacin regimen was associated with 25% and 34% decreases in the V4 and V5 subfractions, respectively, compared with baseline values. The corresponding decreases with higher-dose niacin reached 49% and 57%, respectively. In a diabetic study, levels of V2 through V6 were reduced by 25%-34% on crystalline niacin 3000 mg/d.[41]

Niacin therapy, therefore, has diverse beneficial effects on all lipoprotein subfractions that are consistent with its proven ability to lower the risk of angiographic and clinical CHD. Moreover, the combination of niacin and a statin may be among the best available treatment options for many patients with complex forms of dyslipidemia. Although the cutaneous side effects may discourage some patients, newer formulations may help reduce their impact.[37] Niacin can be used safely in the majority of patients, including those with type 2 diabetes, with appropriate medical care.[43]