You are leaving Medscape Education
Cancel Continue
Log in to save activities Your saved activities will show here so that you can easily access them whenever you're ready. Log in here CME & Education Log in to keep track of your credits.

Table 1.  

Dose, mg
Dose, mg
Dose, mg
Dose, mg
Study 5 10 20 40 80 10 20 40 80 10 20 40 80 10 20 40
HDL __ +7.7 +9.5 +9.6 __ +5.7 +4.8 ‡‡ +4.4 ‡‡§ +2.1 ‡‡§ +5.3 +6.0 +5.2 ‡‡§ +6.8 +3.2 †† +4.4 ‡‡ +5.6 ‡‡§
HDL +7.4 +6.0 +9.1 +12.3 +9.6 +5.0 +7.6 +4.1 § +2.1 ¤ __ __ __ __ __ __ __
ApoA-I +6.1 +5.2 +8.5 +6.2 +9.2 +5.3 +3.6 +1.5 +3.1 ¤ __ __ __ __ __ __ __
Blasetto[20]/12-wk pooled                                
HDL-C +8.2 +8.9 __ __ __ +5.5 *†† __ __ __ __ __ __ __ __ __ __
ApoA-I +6.0 +7.3 __ __ __ +4.1 *†† __ __ __ __ __ __ __ __ __ __
HDL-C +6.9 +9.1 __ __ __ __ __ __ __ __ 6.2 †† __ __ __ 6.2 †† __
ApoA-I +5.4 +5.3 __ __ __ __ __ __ __ __ +4.8 __ __ __ +4.2 __
HDL-C +4.5 7.6 __ __ __ __ __ __ __ __ +6.2 __ __ __ +4.5 __
LDL-C/HDL-C -43.3 -51.1 __ __ __ __ __ __ __ __ -40.8 †† __ __ __ -34.8 *†† __
ApoA-I +6.4 +6.0 __ __ __ __ __ __ __ +7.0 __ __ __ __ +6.9 __
ApoB/ApoA-I -38.2 -41.5 __ __ __ __ __ __ __ __ __ -34.8†† __ __ -29.4 *†† __

Table 1. Effects of Statins on HDL-C and ApoA-I (Boldface = Significant Difference vs Corresponding Rosuvastatin Dose) Change From Baseline, %


* P < .05 vs rosuvastatin 5 mg
†† P < .05 vs rosuvastatin 10 mg
‡‡P < .05 vs rosuvastatin 20 mg
§P < .05 vs rosuvastatin 40 mg
¤P < .05 vs rosuvastatin 80 mg
**Changes are for dose indicated with possible dose titration up 80 mg for rosuvastatin or simvastatin and up to 40 mg for pravastatin.
STELLAR = Statin Therapies for Elevated Lipid Levels Compared Across Doses to Rosuvastatin.


Table 2.  

(Author, Year)
Patients Regimens Duration % Change HDL-C From Baseline Comments
Niacin combinations          
Capuzzi 2003[38] Patients with type IIb or IV hyperlipidemia (HDL< 45 mg/dL) Rosuvastatin 40 mg (n = 46)

Niacin-ER 2 g (n = 70)

Rosuvastatin 40 mg + niacin-ER 1 g (n = 71)

Rosuvastatin 10 mg + niacin ER 2 g (n = 78)
24 weeks +11%



+24% (P < .001 vs rosuvastatin 40 mg)
Rosuvastatin 10 mg-niacin-ER 2 g also raised apoA-I by 11% (P < .017 vs rosuvastatin 40 mg)
Kashyap 2002[39] N = 814 patients with CAD,
diabetes + LDL-C ≥130 mg/dL; > 2 CAD risk factors;
or < 2 risk factors + LDL-C > 190 mg/dL
10 mg/0.5 g (wk 1-4)

20 mg/1.0 g (wk 5-8)

30 mg/1.5 g (wk 9-12)

40 mg/2.0 g (wk 13-52)
52 weeks



+35%/ +41% at wk 28 and 52, respectively
Baseline HDL-C = 48 mg/dL

LDL/HDL-C ratio ↓58%

10% of patients discontinued because of flushing
Wolfe 2001[40] N = 74 Statins at stable doses + niacin-ER titrated to:
1 g
2 g
≥ 4 wk after titration to 1 g niacin-ER


Retrospective analysis.
Mean incremental changes with
niacin-ER = +35% HDL-C, 24% LDL-C, 39% TG
Guyton 1998[41] N = 269 patients with baseline HDL-C ~44 mg/dL Niacin-ER titrated from
375 mg to 3.0 g daily + statins
at investigators' discretion
48-96 wk +25% to 26% Open-label extension
Niacin-statin reduced TC/HDL ratio by 38%
Fibrate combination Patients Regimens Duration % Change HDL-C from baseline Comments
Vega 2003[42] N = 20 with combined hyperlipidemia;
TG 200-800 mg/dL;
LDL-C >160 mg/dL
Simvastatin 10 mg + fenofibrate 200 mg/day 12 wk +23% from baseline (+16% above simvastatin alone)  

Table 2. HDL and ApoA-I-Raising Effects of Statin-Containing Combination Regimens in Selected Studies


*Niacin doses represent maximum doses in titrations.
ER = extended-release.


Toward a Higher Standard: Raising HDL in Clinical Practice: Clinical Strategies to Elevate HDL


Clinical Strategies to Elevate HDL

Nonpharmacologic Therapy

The following factors may be associated with differing levels of HDL-C,[8] and therapeutic lifestyle counseling may assist in reversing these factors and thus help to elevate HDL-C.

Diet. High-carbohydrate, low-fat diets may be associated with low HDL-C.

Obesity. HDL-C may also decline with obesity, while decreases in body mass index have been associated with increases in HDL-C.[11]

Regular aerobic exercise can raise HDL-C modestly (studies in women have demonstrated almost a 25% rise from baseline following 12 weeks of moderate-to-high aerobic exercise [target heart rate 60% to 80% of maximum]).[12]

Smoking has been associated with low HDL-C as well as increased oxidative stress, endothelial injury, and a range of vascular and other adverse sequelae.[13-17] Conversely, smoking cessation has been associated with significant increases (approximately 4 mg/dL) in HDL-C.[18]

Moderate alcohol use is associated with increased HDL-C in a dose-dependent fashion (where "moderate" is defined as 2-3 drinks a day for men, 1-2 drinks a day for women).

Pharmacologic Therapy

Statins. HMG-CoA reductase inhibitors (statins) modestly raise HDL-C levels. Statins lower LDL-C by inhibiting the rate-limiting enzyme in cholesterol biosynthesis and upregulating hepatic receptor-mediated LDL-C clearance. The modest HDL-C-raising effects of these agents may be attributed in part to inhibition of rho-signaling pathways with activation of PPAR-alpha. Statins also reduce plasma CETP activity, which could help to account for the increases in HDL-C.

In randomized, controlled, head-to-head comparative studies ( Table 1 ), rosuvastatin elevated HDL-C by approximately 6% to 12% over its dose range compared with about 2% to 8% for atorvastatin, 5% to 7% for simvastatin, and 3% to 6% for pravastatin.[19-22]

Statins may confer the strongest cardioprotective effects on patients with low HDL-C to begin with. Low HDL-C was a major CAD risk factor among the 6515 individuals evaluated in the Air Force/Texas Coronary Atherosclerosis Prevention Study (AFCAPS/TexCAPS).[23] The baseline mean HDL-C was 36 mg/dL for men and 40 mg/dL for women; 35% of both the lovastatin and placebo groups had HDL-C < 35 mg/dL. If NCEP ATP-II guidelines had been applied, most participants in AFCAPS/TexCAPS would not have been eligible for pharmacotherapy.

Treatment with lovastatin 20-40 mg/day (titrated to LDL-C < 110 mg/dL) increased HDL-C by 6% while lowering LDL-C by 25%, whereas increases in both parameters were modest with placebo. Over a median follow-up of 5.2 years, lovastatin treatment lowered coronary events (fatal or nonfatal MI, sudden death, unstable angina) by 37% compared with placebo (P < .001). The clinical benefits of the statin in preventing coronary events were most marked in patients in the lower tertile of baseline HDL-C (≤ 34 mg/dL).[23]

This finding in a primary prevention study was echoed by data in a subgroup analysis of an angiographic trial. In the Lipid and Coronary Atherosclerosis Study (LCAS) of CAD patients with mild to moderate elevations of LDL-C, patients with low HDL-C (< 35 mg/dL; n = 68) at baseline had the greatest degree of CAD progression.[24] Thus, although their HDL-raising effects are modest, statins reduce cardiovascular risk in patients with low HDL-C.


In the AFCAPS/TexCAPS and LCAS studies, patients with low HDL-C derived the greatest benefit from statin treatment.


Fibrates. Fibrates are ligands for PPAR-alpha. Binding and activation of PPAR-alpha by fibrates upregulate expression of apoA-I and apoA-II, the 2 main constituents of HDL.[25] In addition, treatment with fibrates has been associated with reduced expression of apoC-III and increased expression of lipoprotein lipase (LPL), both of which result in reduced triglycerides and lower-density LDL particles.

Treatment with fibrates has been shown to significantly reduce coronary events in patients irrespective of MI history. The significant cardioprotective benefits of gemfibrozil (vs placebo) were first demonstrated in the Helsinki Heart Study involving middle-aged men with dyslipidemia but no evidence of CAD.[26] A total of 4081 asymptomatic men with mean total cholesterol of 289 mg/dL and HDL-C 47 mg/dL were randomized to gemfibrozil 1200 mg/day or placebo and followed for up to 6 years. Gemfibrozil treatment increased HDL-C by about 10% within the first 25 weeks. Total cholesterol and LDL-C declined by about 10%. Placebo had minimal effects on the lipid panel. Therapy with gemfibrozil reduced the relative risk of coronary events by 34% (P < .02) compared with placebo.[26]

Other trials have demonstrated significant cardioprotective benefits of fibrates in patients with CAD. In the aforementioned VA-HIT,[3] gemfibrozil 1200 mg/day for about 5 years reduced the risk of CAD death, nonfatal MI, or stroke by 24% (P < .001). In subgroup analyses, gemfibrozil reduced the risk of nonfatal MI, CAD death, and confirmed stroke by 30% (vs placebo) in patients with HDL< 31.5 mg/dL compared with 24% in those with HDL-C ≥ 31.5 mg/dL.[3] This finding of potentially strong cardioprotective benefits in patients with low HDL-C is consistent with the results of AFCAPS/TexCAPS and LCAS. In the Bezafibrate Infarction Prevention (BIP) trial, treatment with bezafibrate reduced cardiovascular risk in a subgroup of subjects with high triglyceride and low HDL-C levels.[27,28]

Other benefits of raising HDL-C with fibrates have been demonstrated in angiographic studies of patients with CAD:

a. Bezafibrate Coronary Atherosclerosis Intervention Trial (BECAIT).[29] A total of 92 men with a history of MI (75% with 1- or 2-vessel disease) and dyslipidemia were randomized to treatment with bezafibrate 600 mg/day or placebo and followed for 2 to 5 years. Compared with placebo, bezafibrate raised HDL-C by > 9% and apoA-I by > 4% while lowering VLDL by 35% and modestly reducing LDL-C. Patients randomized to bezafibrate exhibited significantly less angiographic progression, including changes in mean segment diameters and percent stenosis, as well as a significant decline in coronary events, compared with placebo.

b. Lopid Coronary Angiography Trial (LOCAT). A total of 372 evaluable men with a history of coronary artery bypass grafting were treated with gemfibrozil 1200 mg/day (n = 185) or received placebo (n = 187). Treatment with the fibrate increased HDL-C by 21% compared with an increase of 7% with placebo. The placebo-adjusted decline in triglycerides with gemfibrozil was about 31%, while fibrate-induced changes in total cholesterol and LDL-C were minimal compared with placebo. Men randomized to gemfibrozil had significantly less angiographic progression compared with controls at treatment month 32. Development of new lesions was also significantly less likely in men who received gemfibrozil.

Niacin. Niacin is the most effective HDL-raising agent currently on the market. The beneficial lipid effects of niacin are mediated at least in part by binding to adipose nicotinic acid receptors with downregulation of free fatty acid mobilization, as well as reduction of hepatic VLDL output, thus indirectly raising HDL-C levels. Niacin also promotes hepatic apoA-I production and slows hepatic clearance of apoA-I and HDL through as-yet unknown mechanisms.[30]

Niacin has been a dyslipidemia treatment standby for nearly 4 decades. Many of the earliest angiographic regression studies performed by David Blankenhorn and colleagues involved treatment with niacin, mainly in combination with a bile acid resin (colestipol).[31-33] Long-term follow-up from the Coronary Drug Project demonstrated that treatment with niacin reduced all-cause mortality by 11% (P = .0004) compared with placebo in 8341 men with CAD ages 30-64 at baseline.[34] These late longevity benefits were observed nearly a decade after trial termination and niacin discontinuation.

Niacin increases HDL-C to the greatest extent of all available monotherapies: by 15% to 35% at higher daily doses on the order of 3 grams.[2,35] Recent interest in niacin centers largely on its use as an adjunct to statins and other therapies (see below, "Combination Regimens"). Fixed-dose combinations of extended-release niacin with lovastatin are commercially available in the United States.

The Arterial Disease Multiple Intervention Trial (ADMIT) suggested that niacin may be administered successfully to many patients with diabetes and peripheral arterial disease who do not tolerate statins or fibrates.[36] Patients receiving immediate-release niacin up to 3 g daily for 48 weeks had increases of 29% in HDL-C. Glucose levels rose moderately but significantly, by 8.7 mg/dL in men with diabetes and 6.3 mg/dL in those without diabetes. Hemoglobin A1C levels were unchanged from baseline in patients with diabetes who received niacin treatment.[36]

Fish oils (omega-3 fatty acids). Dietary modification to increase the consumption of cold-water fish (eg, salmon) rich in polyunsaturated fats may help to raise HDL-C. Capsules containing omega-3 fatty acids (1.48 g of docosahexaenoic acid + 1.88 g of eicosapentaenoic acid) have been formulated and are commercially available in some parts of the world (as Omacor). In a recent small study in patients with familial combined hyperlipidemia, treatment with this formulation for 8 weeks increased HDL-C by 8%, particularly the more buoyant HDL2 subfraction. Levels of the antioxidant HDL-associated enzyme paraoxonase were also increased by 10%.[37]

Combination Regimens

Combinations of statins with either niacin or fibrates represent potentially complementary regimens. Effects of statin-niacin and statin-fibrate combinations in selected studies are presented in Table 2 .[38-42]

One concern with statin combination regimens is the development of myopathy and/or rhabdomyolysis when statins are administered together with certain medications, including agents with shared metabolic routes that increase systemic exposure of either agent. In postmarketing surveillance studies, 55% of 601 cases of statin-induced rhabdomyolysis were associated with agents influencing statin metabolism, including fibrates (13% of cases), cyclosporine (8%), macrolide antibiotics (7%), warfarin (5%), digoxin (4%), and azole antifungals (2%).[43,44] Many of these agents are cytochrome P450 (CYP) 3A4 inhibitors. On the other hand, concomitant administration of rosuvastatin 10 mg together with fenofibrate 67 mg 3 times daily did not increase the systemic exposure of either agent, and the combination regimen was well tolerated.[45]

Prescribers contemplating a combination of the above agents are urged to consider their relative benefits and risks. Consensus guidelines concerning the assessment and management of patients with possible statin-induced myopathy have been issued.[46] One approach to minimize statin-fibrate interactions involves progressively titrating drug doses from low initial doses and administering the statin at night and fibrate in the morning.[25] Regular monitoring of muscle and liver enzymes is also important.