Niacin: The Third or Fourth Option

After the clinical studies, Atherothrombosis Intervention in Metabolic Syndrome with Low HDL/High Triglyceride and Impact on Global Health Outcomes (AIM-HIGH) and Heart Protection Study 2-Treatment of HDL to Reduce the Incidence of Vascular Events (HPS2-THRIVE), the National Lipid Association released a statement on the use of niacin: “We believe that niacin remains a valuable adjunct to statin treatment for LDL-C lowering, and a valuable statin alternative in statin intolerant patients.”1

Niacin has been promoted for decades as a high-density lipoprotein (HDL)-raising drug. It does increase HDL cholesterol (HDL-C) by reducing the number of small, cholesterol- poor HDL particles and increasing the numbers of large, cholesterol-rich particles, but it does not increase the total number of HDL particles (HDL-P).2,3 Studies suggest that changes in HDL subclass distribution are not independently associated with cardiovascular disease (CVD) risk after accounting for HDL-P levels.4 An NLA expert panel in 2011 claimed there was no evidence that a shift in HDL subfractions translated into changes in disease progression or improved outcome.5 So, although niacin increases HDL-C, it has minimal to no effect on increasing HDL-P. Niacin is an LDL-lowering drug, specifically lowering LDL particle number (LDL-P) and apolipoprotein (apoB).3 One proposed mechanism is the inhibition of diacylglycerol O-acyltransferase 2 (DGAT2), which lowers hepatic triglyceride production, leading to decreased apoB and very low-density lipoprotein (VLDL)/LDL secretion.6 Niacin also reduces proprotein convertase subtilisin/kexin type 9 (PCSK9).8 Niacin increases LDL particle size, so LDL-C and LDL-P may not change concordantly.3 In general, niacin drops LDL-C. However, while there’s no published evidence to support it, I’ve personally observed rare occasions where LDL-C may change only minimally despite an effective drop in LDL-P. As such, following apoB or LDL-P may be a more accurate reflection of niacin’s effect than LDL-C.

The IMProved Reduction of Outcomes: Vytorin Efficacy International Trial (IMPROVE-IT) and the meta-analysis by Robinson, et al., make the point that it is not how LDL is reduced that reduces risk, but rather how much it is reduced.9,10

Studies with maximum-dose atorvastatin or rosuvastatin show that 62 to 80 percent of patients did not get below 70mg/dl.11 Based on IMPROVE-IT, ezetimibe can be a second option as an add-on, but what other options exist? Bile acid sequestrants are reasonable, but have gastrointestinal problems and the newest one (colesevelam) is expensive. Niacin also has side effects, but it is now inexpensive with numerous generic formulations. Yet is there any benefit in using niacin?

Imaging studies using carotid intima-media thickness (CIMT) testing, carotid wall volumes by magnetic resonance imaging (MRI), and coronary angiography have shown regression or reduction in progression with niacin added to a statin — ARBITER 2, Oxford Niaspan Study, HATS — or niacin added to a bile acid sequestrant — FATS, AFREGS, CLAS.12-17 In FATS, the niacin- colestipol group did as well at preventing progression and obtaining regression as the lovastatin-colestipol group.

A recent systemic review and meta-regression evaluated 11 clinical trials of niacin, either alone or combined with other lipid-altering therapy, and showed a significant reduction in the composite endpoints of any cardiovascular event (P = 0.007) and any major coronary heart disease event (P = 0.02). There was no association between event reduction and changes in HDL-C.18

Two recent trials, AIM-HIGH and HPS2- THRIVE, were designed to see if adding niacin to an aggressive lipid-lowering regimen (statin and, frequently, ezetimibe) could further reduce residual atherosclerotic risk.

The first, AIM-HIGH, with 3,414 coronary artery disease patients, failed to show any incremental benefit in adding niacin.19 Several limitations of this study have been raised: (1) It was designed to detect a ≥ 25 percent reduction in events and, as such, had too few participants. (2) 90 percent of patients had been on aggressive lipid- lowering efforts for more than a year, which may have led to plaque regression and stabilization prior to entry. (3) The placebo arm had 7 percent more on simvastatin 80mg and 12 percent more on ezetimibe. Niacin is an LDL-lowering drug, so the more aggressive LDL-lowering therapy in the placebo arm may have significantly altered the results.20,21

The second trial, HPS2-THRIVE, demonstrated in 25,673 vascular patients no significant benefit to adding extended-release niacin (combined with laropiprant) to attain further event reduction.22 A major concern about the study was the baseline lipid profile: direct LDL-C 63 mg/dl (calculated LDL-C 59mg/dl), HDL-C 44mg/dl, triglyceride 125mg/dl, and non-HDL 84mg/dl.  Despite this, patients with a higher baseline LDL-C had a “nominally significant trend (P = 0.02) toward a greater reduction in risk” consistent with niacin’s action as an LDL-lowering agent.22 (Figure 1)

HPS2-THRIVE noted the following side effects: Skin — mainly rash or flushing. Gastrointestinal — primarily dyspepsia and diarrhea. Also, an increase in liver function tests (LFTs) >3x ULN (1.4 percent over 3.5 years) but drug-related hepatitis or elevated transaminases plus elevated bilirubin no different than placebo. Myopathy — primarily in the Chinese patients studied and with no significant difference amongst European patients in the trial. Increased bleeding — primarily in the GI tract with no significant difference in intracranial bleeding. Increased infection rate — mainly skin or lower respiratory tract (1.4 percent over four years). Increased glucose levels — disturbances in glucose control, much of which required hospitalization (3.6 percent over four years), were noted. In comparison, the Arterial Disease Multiple Intervention Trial (ADMIT) study of 468 patients on 2,500mg/day niacin showed a glucose rise of 6.3mg/dl in non-diabetics and 8.7mg/dl in diabetics at 60 weeks.24 Another study of 2,000mg/day niacin showed a glucose increase of 7 percent at three months but only 3 percent at six months.25 Increased diagnosis of diabetes — 1.4 percent over four years. Other effects: aggravation of gout.

In summary: Niacin is not an HDL-P-raising drug but does lower LDL-C, apoB, and LDL- P. Weigh the risks and benefits of niacin before using. Periodically monitor LFTs (ALT, AST), glucose, HgAIC, and uric acid. Avoid use in patients with inadequately controlled diabetes. Niacin is reasonable as a third-or fourth-option drug for use in high-risk patients who cannot get LDL to goal or in patients who are statin intolerant 1,22,26,27      

Disclosure statement: Dr. Moran has received speaker honoraria from AstraZeneca, Merck, Health Diagnostic Lab, and LipoScience. He has received consultant fees from Health Diagnostic Lab.

References are listed on page 39 of the PDF.