Niacin-induced hepatotoxicity is dose-related and occurs with both immediate-release and sustained-release formulations. Our retrospective cohort study [1] did not include patients receiving immediate-release niacin, and some cases of elevated hepatic enzyme levels may have been missed. In a prospective study, Lavie and colleagues [2] treated patients with "isolated" low high-density lipoprotein cholesterol levels using a sustained-release niacin preparation similar to the one we used. At a mean daily niacin dose of 2400 mg, 3% of their 36 patients were withdrawn from the study because of threefold elevations in aminotransferase values. Hepatic dysfunction was found in fewer than 3% of nearly 300 other patients [2]. Thus, it is unlikely that we missed a significant number of patients with hepatic dysfunction because of the retrospective nature of our study.

We could identify several factors (hypoglycemic agents, alcohol abuse, and niacin dose) that may predispose patients to niacin-induced hepatotoxicity. By avoiding the use of controlled-release niacin in patients at risk and by treating selected patients with a dose of 1.0 to 2.0 g/d, we have found that controlled-release niacin yields favorable results in terms of tolerance, safety, and efficacy. Both immediate-release and sustained-release niacin were poorly tolerated in the study by McKenney and colleagues [3], which included fewer patients and had withdrawal rates of 39% (9 of 23 patients) and 78% (18 of 23 patients), respectively. Hepatotoxicity, which did not occur in the immediate-release niacin group, was frequent in the sustained-release group, occurring primarily at doses of 2 g/d or higher in 9 of 12 patients. Timed-release niacin capsules, the formulation of sustained-release niacin used by McKenney and colleagues, may not be bioequivalent to other sustained-release niacin formulations [4] and may also be associated with greater hepatoxicity.

Niacin is a broad-spectrum lipid-lowering agent with favorable effects on low-density lipoprotein cholesterol, high-density lipoprotein cholesterol, and triglyceride levels. As indicated in the National Cholesterol Education Program guidelines^[5], niacin can be safely and effectively prescribed with proper patient selection, dosing, and monitoring. Not all niacin formulations are the same. Significant differences in safety and efficacy can exist among different sustained-release preparations of niacin, many of which are available without medical supervision. This issue needs to be explored further, and a regulatory policy should be developed for use of niacin in the management of dyslipoproteinemia.