In the United States, it was introduced and manufactured by Parke-Davis in the late 1990s but turned out to be associated with an idiosyncratic reaction leading to drug-induced hepatitis.

The medical officer assigned to the review, Dr. John L. Gueriguian, cited Rezulin's potential to harm the liver and the heart, and he questioned its viability in lowering blood sugar for patients with adult-onset diabetes, recommending against the drug's approval.

[10] Parke-Davis said at the advisory committee that the risk of liver toxicity was comparable to placebo and that additional data of other studies confirmed this.

[11] According to Peter Gøtzsche, when the company provided these additional data one week after approval, they showed a substantially greater risk for liver toxicity.

[13] Glaxo Wellcome received approval from the British Medicines Control Agency (MCA) to market troglitazone, as Romozin, in July 1997.

[9][20] Dr. David J. Graham, an FDA epidemiologist charged with evaluating the drug, warned on March 26, 1999 of the dangers of using it and concluded that patient monitoring was not effective in protecting against liver failure.

[9][21] Dr. Janet B. McGill, an endocrinologist who had assisted in the Warner–Lambert's early clinical testing of Rezulin, wrote in a March 1, 2000 letter to Sen. Edward M. Kennedy (D-Mass.

): "I believe that the company... deliberately omitted reports of liver toxicity and misrepresented serious adverse events experienced by patients in their clinical studies.

[23] Dr. Robert I. Misbin, an FDA medical officer, wrote in a March 3, 2000 letter to Senator John Ashcroft of strong evidence that Rezulin could not be used safely.

[25] Since the withdrawal in 2000, mechanisms of troglitazone hepatotoxicity have been extensively studied using a variety of in vivo,[26] in vitro,[27] and computational methods.

Initially, the metabolic toxicity was largely associated with reactive metabolite formation from the thiazolidinedione and chromane rings of troglitazone.