to the editor: Drs. Florence and Yeager¹ provided a current and comprehensive summary of the screening, diagnosis and treatment of patients with type 2 diabetes (formerly non–insulin-dependent diabetes). However, I want to point out that their discussion about the alpha-glucosidase inhibitors, miglitol (Glyset) and acarbose (Precose) tablets, is inaccurate with respect to hepatotoxicity.

The authors make the following statements: “Dose-dependent hepatotoxicity is associated with this drug class, so liver function tests should be carefully monitored in patients receiving higher dosages of these medications (i.e., more than 50 mg three times daily). Transaminase elevations are reversible with discontinuation of the drug and are often asymptomatic. Serum transaminase levels should be checked every three months for the first year patients take the medication and periodically thereafter.”¹ These statements are misleading with respect to miglitol because one of the pivotal U.S. clinical trials revealed a similar and very low incidence of elevated liver enzymes when the drug was compared with placebo.²

As a result of these and other clinical trial safety data on miglitol, the U.S. package insert for miglitol does not contain a precaution statement about hepatotoxicity; monitoring of transaminase levels is unnecessary, and the drug is not contraindicated in patients with cirrhosis.³

In contrast, the U.S. package insert for acarbose includes the following statement in the precautions section: “In long-term studies (up to 12 months, and including Precose doses up to 300 mg t.i.d.) conducted in the United States, treatment-emergent elevations of serum transaminases (AST and/or ALT) above the upper limit of normal (ULN), greater than 1.8 times the ULN and greater than 3 times the ULN occurred in 14 percent, 6 percent and 3 percent, respectively, of Precose-treated patients as compared to 7 percent, 2 percent, and 1 percent of placebo-treated patients.”⁴ According to the product label, acarbose is contraindicated in patients with cirrhosis.

A synopsis of clinical trial safety data for acarbose concluded that sustained treatment-emergent increases in serum transaminase levels of at least 1.8 times the ULN were evident in 3.8 percent of patients receiving acarbose versus 0.9 percent on placebo in U.S. clinical trials.⁵ A recent, thorough review of clinical trial safety data for miglitol confirmed that “Unlike acarbose, treatment with miglitol was not associated with elevated serum transaminase (alanine or aspartate transferase) levels.”⁶ In conclusion, I want to emphasize that physicians should be aware that the alphaglucosidase inhibitor, miglitol, has not been linked to hepatotoxicity.

editor's note: This letter was sent to the authors of “Treatment of Type 2 Diabetes Mellitus,” who declined to reply.