← back to the index
RX-006 Thiazolidinedione 2000

Rezulin (troglitazone) — the Diabetes Pill Withdrawn in 2000 After 63 Liver-Failure Deaths

hepatotoxicityiatrogenic-mortalityliver-failurediabetologyendocrinologyhepatology1990s2000s
Patients exposed
~1.9M U.S. patients / >$2.1B in sales
Documented harm
≥63 liver-failure deaths; 90+ acute liver-failure reports; ~10 transplants
On market
1997→2000 (~3.1 years)
Status
Withdrawn

Summary

When the U.S. Food and Drug Administration asked Warner-Lambert to pull Rezulin from the market on 21 March 2000, it ended a three-year arc in which a celebrated first-in-class diabetes pill killed at least 63 people through liver failure while the warning that should have stopped it had been visible from before launch. Rezulin (troglitazone), the first thiazolidinedione — an oral "insulin sensitizer" acting on the PPAR-gamma receptor — was approved on 29 January 1997 and marketed as a breakthrough that let type 2 diabetics use their own insulin more effectively. The gap between that promise and the harm was not subtle: the drug produced idiosyncratic, sometimes fulminant hepatotoxicity, and Britain's Glaxo Wellcome withdrew it from the U.K. market on 1 December 1997 — barely ten months after the U.S. approval — while the FDA kept it on American shelves for another twenty-eight months.

The signal was institutional, not merely clinical. The FDA medical officer originally assigned to the application, Dr. John Gueriguian, had recommended against approval over cardiac and liver concerns; he was removed from the review on 4 November 1996, and his negative analysis was purged from the official record before the drug was cleared on an expedited timeline. Within the first year of marketing, serious hepatotoxicity reports accumulated, and an internal FDA memorandum cited 135 reports of serious liver injury by late 1997. Public Citizen's Health Research Group, led by Dr. Sidney Wolfe, petitioned for withdrawal in July 1998, asking the agency how many more Americans would have to die or need transplants.

The verdict is therefore plain at the outset: a profitable, regulator-blessed, heavily prescribed medicine reached nearly two million Americans while the evidence that condemned it — a foreign withdrawal, a buried internal review, mounting death reports, and an outside petition — sat fully legible in the record. The FDA's response was incrementalism: four label changes between 1997 and 1999 adding ever-stricter liver-monitoring instructions that real-world prescribing could not reliably execute, rather than removal.

What finally forced the withdrawal was arithmetic the labels could not fix. Once two safer thiazolidinediones — rosiglitazone and pioglitazone — reached the market in 1999 without the same fatal liver signal, Rezulin's risk-benefit case collapsed, and the FDA requested its removal on 21 March 2000. Litigation followed: Warner-Lambert, absorbed by Pfizer in 2000, ultimately resolved roughly 35,000 claims for an estimated $750 million, and "Rezulin" became a byword for how a buried safety review and an ignored foreign recall can keep a lethal drug on the market for years.

Timeline

1996-11-04
Reviewer removed before approval
FDA medical officer Dr. John Gueriguian, who had urged against approval citing liver and cardiac concerns, is taken off the review; his negative analysis is later said to have been purged from the file.
1997-01-29
FDA approves Rezulin
Troglitazone is cleared for type 2 diabetes as the first thiazolidinedione, marketed by Warner-Lambert's Parke-Davis division on an expedited review as a breakthrough insulin sensitizer.
1997-07
U.K. approval as Romozin
Glaxo Wellcome licenses and markets troglitazone in Britain, where it had received approval months after the U.S. clearance.
1997-11-26
Internal FDA alarm
An FDA memorandum from Dr. Murray Lumpkin cites 135 reported cases of serious hepatotoxicity associated with the drug within its first year on the U.S. market.
1997-12-01
Britain withdraws the drug
Glaxo Wellcome pulls troglitazone (Romozin) from the U.K. market over liver-toxicity reports — a foreign safety withdrawal ten months after the U.S. launch.
1997-12
U.S. keeps it, adds monitoring
Rather than follow Britain, the FDA and maker respond with label changes recommending periodic liver-enzyme (ALT) monitoring; the drug stays on U.S. shelves.
1998-07
Public Citizen petitions for withdrawal
Dr. Sidney Wolfe and Larry Sasich of Public Citizen's Health Research Group petition the FDA to ban troglitazone, citing accumulating liver-failure deaths.
1999-03-26
FDA advisory committee reviews the drug
The Endocrine and Metabolic Drugs Advisory Committee hears testimony, including Public Citizen's call for withdrawal; the drug survives with still-stricter monitoring labeling.
1999
Safer rivals arrive
Rosiglitazone (Avandia) and pioglitazone (Actos), thiazolidinediones without the same fatal hepatic signal, reach the market, eroding any unique benefit justifying Rezulin's risk.
2000-03-21
FDA requests withdrawal
With at least 63 confirmed liver-failure deaths and safer alternatives available, the FDA asks Warner-Lambert to remove Rezulin; the company complies. Sales had exceeded $2.1 billion.
2000-06
Warner-Lambert absorbed by Pfizer
The merger transfers the mounting Rezulin liability to Pfizer just months after the withdrawal.
2009
Litigation resolved
Pfizer reports having settled all but three of roughly 35,000 Rezulin claims for an estimated total of about $750 million.

The Breakthrough Mechanism: How a Novel Receptor Built a Blockbuster

Rezulin answered a genuine unmet need. Type 2 diabetes is driven by insulin resistance, and before the thiazolidinediones no oral agent directly addressed it; existing drugs either squeezed more insulin from the pancreas or slowed glucose absorption. Troglitazone, developed in Japan by Sankyo and licensed to Warner-Lambert, activated the PPAR-gamma nuclear receptor and made peripheral tissues more responsive to a patient's own insulin — a novel, scientifically elegant mechanism that lowered blood sugar without forcing insulin secretion. That story was compelling enough to win an expedited FDA review and rapid uptake: within three years roughly 1.9 million Americans had taken the drug and sales passed $2.1 billion. The mechanism was real and the benefit measurable. What the marketing narrative obscured was that the molecule also carried an idiosyncratic, unpredictable hepatic toxicity — liver injury that struck a small fraction of users without warning, typically one to six months into therapy, and could progress to fulminant failure. A breakthrough on the metabolic axis was, on the hepatic axis, a loaded chamber.

The Ignored Recall: Britain Withdraws, the United States Adds a Label

The decisive warning came early and from abroad. On 1 December 1997 — ten months after the American launch — Glaxo Wellcome withdrew troglitazone from the British market over reports of liver toxicity. A peer regulator in a developed nation had judged the same molecule too dangerous to sell. The FDA did not follow. Instead, the agency and Warner-Lambert chose the path of monitoring: a sequence of label revisions across 1997 to 1999 recommending periodic measurement of liver enzymes so that injury might be caught before it turned fatal. The strategy assumed a level of real-world surveillance that primary-care prescribing for a chronic disease could not deliver — patients miss tests, results lag, and idiosyncratic injury can outrun a quarterly blood draw. Meanwhile the internal record was already damning: a November 1997 FDA memo counted 135 serious hepatotoxicity reports, and the reviewer who had flagged the hazard before approval, John Gueriguian, had been removed from the file. When Public Citizen petitioned for withdrawal in July 1998 and testified before the FDA's advisory committee in March 1999, the agency again declined to remove the drug, choosing tighter labels over revocation while the death toll climbed.

The Reckoning: Safer Rivals, a Forced Withdrawal, and a $750 Million Bill

Rezulin was ultimately undone not by a new revelation but by the disappearance of its excuse. So long as troglitazone was the only insulin sensitizer, the FDA could weigh its fatal liver risk against a unique benefit. That calculus broke in 1999 when two successors in the same class — rosiglitazone and pioglitazone — reached the U.S. market offering the PPAR-gamma mechanism without the same lethal hepatic signal. With at least 63 confirmed liver-failure deaths, more than 90 reports of acute liver failure, and roughly ten patients needing transplants, the risk now bought nothing that safer drugs did not provide. On 21 March 2000 the FDA requested withdrawal, and Warner-Lambert complied. The reckoning then moved to the courts. Warner-Lambert was absorbed by Pfizer in mid-2000, transferring the liability; by 2009 Pfizer had resolved roughly 35,000 claims for an estimated $750 million. The breakthrough narrative was retired, and the case entered the regulatory canon as the drug that should have been pulled the moment Britain pulled it.

Contributing Factors

01
A buried internal review and a removed reviewer
The FDA medical officer assigned to troglitazone, John Gueriguian, recommended against approval over liver and cardiac concerns and was removed from the review on 4 November 1996, his analysis later said to have been purged from the file before the drug was cleared. When the one institutional voice trained to catch the hazard is taken off the case before approval, the system loses its designed safeguard at the exact point it is most needed.
02
An ignored foreign withdrawal
Glaxo Wellcome pulled the same molecule from the British market on 1 December 1997 over liver toxicity. A peer regulator's safety withdrawal is among the strongest possible signals, yet the FDA treated it as a reason to add monitoring rather than to reconsider approval — keeping the drug on U.S. shelves for twenty-eight more months. Discounting another competent regulator's "no" is a recurring path to preventable harm.
03
Monitoring as a substitute for removal
The agency's response to idiosyncratic, fast-moving hepatotoxicity was a series of label changes mandating periodic liver-enzyme testing. Such monitoring presumes a surveillance discipline that routine outpatient prescribing cannot sustain, and it cannot reliably catch injury that progresses between tests. A label that shifts the burden of catching a lethal reaction onto overworked clinicians and patients is a management strategy, not a safety solution.
04
Expedited approval of a first-in-class mechanism
Troglitazone was the first thiazolidinedione, cleared on an expedited review for its novel benefit before its idiosyncratic toxicity could be characterized at scale. Genuine mechanistic novelty creates pressure to approve fast and a blind spot for rare, delayed harms that only mass exposure reveals — the benefit is legible early, the body count only later.
05
Risk tolerated until a substitute existed
Rezulin's fatal hepatic risk was tolerated as the price of a unique drug, and removal came only once rosiglitazone and pioglitazone offered the same mechanism more safely. Anchoring a withdrawal decision to the arrival of competitors, rather than to the harm itself, means patients absorb the danger for as long as the maker enjoys a monopoly.

Aftermath

The material toll was concentrated and lethal: at least 63 deaths from liver failure, more than 90 reported cases of acute liver failure, and roughly ten transplants, against nearly 1.9 million exposed Americans and more than $2.1 billion in sales. Pfizer, having inherited the liability through its 2000 acquisition of Warner-Lambert, resolved approximately 35,000 claims for an estimated $750 million by 2009. The durable ripple ran through both regulation and clinical practice. The episode became a standing exhibit in congressional and journalistic scrutiny of the FDA's expedited-approval and post-market processes — a 2000 Los Angeles Times investigation by David Willman, which won a Pulitzer Prize, documented the buried review and the ignored British withdrawal in detail. The thiazolidinedione class survived in its safer successors, but those, too, accrued their own warnings; rosiglitazone would later draw boxed cardiovascular cautions, underscoring how a class can be rehabilitated without being absolved. What remains is the name as shorthand. "Rezulin" is invoked whenever a regulator keeps a profitable, novel drug on the market by stacking liver-monitoring labels onto a known fatal signal — and especially when a foreign regulator's withdrawal has already shown the way. It is the canonical case of a recall that arrived years after the evidence demanded it.

Lessons

  1. Treat a peer regulator's withdrawal as a verdict, not a data point: when a competent foreign authority pulls a drug for safety, the default should be to remove it and prove it safe — not to add a label and keep selling.
  2. When you find yourself managing a fatal risk with monitoring, ask whether the monitoring can actually be performed: if catching the harm depends on perfect, timely testing across millions of routine prescriptions, you are managing the paperwork, not the danger.
  3. Protect the dissenting reviewer: when the one expert assigned to find the hazard recommends against approval, removing that person and purging the analysis does not eliminate the risk — it only delays its discovery until the death reports arrive.
  4. Do not let a monopoly justify a tolerated harm: if a drug's lethal risk is acceptable only because nothing safer exists, treat that as a countdown, and revisit the decision the moment an alternative appears — or before.
  5. Read accumulating spontaneous reports as a trend, not noise: 135 serious hepatotoxicity reports in the first year were a signal, and counting them while declining to act is how a preventable toll becomes a final one.

References