McKenney · JAMA 1994 · Randomized double-blind parallel trial · n=46

A comparison of the efficacy and toxic effects of sustained- vs immediate-release niacin in hypercholesterolemic patients.

Cited 301 times in the scientific literature.

Level 2 - randomized trial

Double-blind randomized controlled trial comparing two active drug formulations.

PubMed 8309029 · record verified 2026-08-29

What was done

A randomized, double-blind, parallel trial compared immediate-release (IR) versus sustained-release (SR) niacin in 46 hypercholesterolemic adults (23 per group) with LDL cholesterol levels exceeding 4.14 mmol/L (160 mg/dL) after 1 month of a Step 1 NCEP diet. Doses were escalated sequentially through 500, 1000, 1500, 2000, and 3000 mg/d, with each dose maintained for 6 weeks. Outcomes included fasting lipid and lipoprotein levels, clinical laboratory tests, symptom questionnaires, and withdrawal rates.

What was found

SR niacin reduced LDL cholesterol significantly more than IR niacin at doses of 1500 mg/d and above, while IR niacin increased HDL cholesterol significantly more than SR niacin at all dose levels. Triglyceride reductions were similar between groups. In the IR group, 9 of 23 patients (39%) withdrew before completing 3000 mg/d (primarily due to vasodilatory symptoms, fatigue, and acanthosis nigricans), with 0 of 23 (0%) developing hepatotoxicity. In the SR group, 18 of 23 patients (78%) withdrew before completing 3000 mg/d (primarily due to gastrointestinal symptoms, fatigue, and aminotransferase elevations), and 12 of 23 (52%) developed hepatotoxic effects.

Why it matters

This trial showed that sustained-release niacin carries a substantial risk of hepatotoxicity compared to immediate-release niacin, providing strong evidence against the unrestricted use of sustained-release formulations.

Limits

The study was small (46 participants) and conducted at a single center. The high attrition rate in both arms (especially 78% in the SR arm) limited the evaluation of the highest doses. Exact numerical lipid changes, confidence intervals, and baseline demographics were not reported in the abstract.

Cited by