This letter is written to clarify several statements in the article "Update in Endocrinology" [1]. First, the article stated that neutropenia may occur in up to 20% of patients receiving troglitazone. It seems that this statement is based on data from Kumar and coworkers [2]. The actual report referenced states "... there was a decrease in the neutrophil count (to less than 0.9 times the lower limit of normal) in 9-20% of patients receiving troglitazone, with the highest incidences at the highest doses, although similar decreases were seen in 9% of patients on placebo." The data from the U.S. studies (n = 2510 patients), as cited in the Food and Drug Administration (FDA)-approved labeling for troglitazone, indicated that "small decreases in hemoglobin, hematocrit and neutrophil counts (within the normal range) were more common in troglitazone-treated patients than placebo patients and may be related to increased plasma volume seen with troglitazone treatment" [3]. Troglitazone does not appear to lead to true "neutropenia."

Second, the article also erroneously stated that "at least 35 cases of liver failure have occurred among the 600 000 patients receiving troglitazone." The exact wording in the FDA paper of 3 November 1997 from which this statement is drawn is different [4]: "As of October 21, 1997, 35 post-marketing reports of liver injury of various degrees have been received. These reports range from mildly elevated blood levels of the liver transaminase enzymes to liver failure leading to one liver transplant and one death." The inference that all 35 of these reported cases involved liver failure is therefore incorrect. On 1 December 1997, the warning about liver dysfunction was further strengthened, and recommendations for liver function test monitoring were issued.

The final point regards the current approved indications for troglitazone. The article states that "the Food and Drug Administration has approved the drug [troglitazone] for use in patients with type 2 diabetes receiving insulin whose hemoglobin A_1c levels continue to be greater than 0.085." This was, in fact, the initial indicated use for troglitazone, granted in late January 1997. However, on 4 August 1997, the FDA approved troglitazone as an adjunct to diet and exercise and as a combination agent for concomitant use with either sulfonylurea or insulin in patients with type 2 diabetes [3]. The wording about the recommended hemoglobin A_1c level (>0.085) to define the patient population that was appropriate was also dropped with the new indications. This has important implications for the treatment algorithm included in the update. The statement is also made that use with insulin "is currently the major role of troglitazone." In fact, troglitazone has been and continues to be prescribed for all of its indications, including as initial therapy and in combination with other oral agents, as well as in combination with insulin. I believe these clarifications will update the information presented in the article and give the readers the most current facts.