← back to the index
RX-013 Oral NSAID 1998

Duract (bromfenac) — the 10-Day Painkiller Pulled in 1998 for Fatal Liver Failure

hepatotoxicityiatrogenic-mortalityliver-failurehepatologypain-managementrheumatology1990s
Patients exposed
~2.5M prescriptions in ~11 months
Documented harm
≥4 deaths, 8 liver transplants, 12+ serious liver injuries; >50 severe DILI cases reported
On market
1997→1998 (0.94 years)
Status
Withdrawn

Summary

When Wyeth-Ayerst Laboratories pulled Duract from American pharmacies on 22 June 1998, the company cast it as a precaution against rare liver events in patients who had overstayed a clearly printed 10-day limit; the documented record shows the hepatic hazard was visible inside the drug's own approval file nearly a year earlier, and that an FDA medical officer had argued — and lost — for a boxed warning before a single prescription was written. Duract (bromfenac sodium), a short-term non-steroidal anti-inflammatory approved by the U.S. Food and Drug Administration on 15 July 1997 for acute pain lasting ten days or less, instead became one of the fastest market withdrawals of the modern era. In roughly eleven months it generated some 2.5 million prescriptions and was associated with at least four deaths and eight liver transplants from fulminant hepatic failure, plus a further dozen serious liver injuries, with the count of severe drug-induced liver injury cases ultimately climbing past fifty.

The gap between promise and harm was not a post-market surprise. Bromfenac's submission already showed that roughly 15 percent of short-term trial patients developed elevations of the liver enzymes AST or ALT to three times the upper limit of normal — a signal of hepatocellular stress markedly greater than that seen with comparable NSAIDs. The FDA medical reviewer who read that data advocated a black-box warning as a condition of approval and was overruled; the drug shipped instead with a routine label and a duration cap of ten days, on the theory that limiting exposure would contain the risk.

That theory collapsed against ordinary prescribing behavior. Acute pain frequently does not resolve in ten days, refills were written, and patients took bromfenac for weeks. The cases that killed and transplanted clustered almost entirely in this extended-use population — the precise scenario the 10-day rule was meant to prevent but had no mechanism to enforce. The verdict here is therefore plain at the outset: an approved, branded analgesic reached millions while the hepatotoxic signal that would condemn it sat fully legible in the agency's own pre-approval review.

What followed was a rapid, near-total revocation. Wyeth strengthened the warnings in February 1998, the harm continued, and the company withdrew the drug in June; the FDA later formally rescinded the New Drug Application. The molecule survived only by abandoning the bloodstream — reformulated years later as a topical eye drop, where systemic exposure was negligible — while "Duract" became a byword for a withdrawal whose justification had been written before the launch.

Timeline

1997-07-15
FDA approves Duract (NDA 20-535)
Bromfenac sodium is cleared for short-term management of acute pain, ten days or less, despite trial data showing ~15% of patients with AST/ALT elevations to 3x the upper limit of normal.
1997-07
Black-box warning declined
The FDA medical officer reviewing the submission unsuccessfully advocates a boxed hepatotoxicity warning, having concluded bromfenac causes more hepatocellular damage than other NSAIDs; the drug ships with conventional labeling.
1997-07 to 1998-06
Mass exposure
Roughly 2.5 million prescriptions are written in under a year as Duract is promoted for everyday acute pain.
1997-Q4
First serious hepatic cases surface
Post-marketing reports of severe liver injury begin arriving, concentrated in patients who took the drug beyond the 10-day cap.
1998-02
Labeling strengthened
The FDA and Wyeth revise the label to re-emphasize the 10-day limit and warn against prolonged use; prescribing patterns do not meaningfully change.
1998-06
Fatal cases accumulate
Confirmed reports reach at least four deaths and eight liver transplantations from fulminant hepatic failure, with a dozen additional serious injuries.
1998-06-22
Voluntary withdrawal
Wyeth-Ayerst removes Duract from the U.S. market, citing severe hepatic injury resulting in death and transplantation, chiefly after use exceeding ten days.
1999-08
Transplant case formally published
Hunter et al. report in the American Journal of Gastroenterology the first bromfenac-associated fulminant hepatic failure salvaged by liver transplantation.
2005
Bromfenac returns as an eye drop
The molecule is reintroduced topically as Xibrom (ophthalmic), where systemic absorption — and thus hepatic exposure — is minimal.
2006
Systematic review codifies the lesson
Goldkind and Laine review NSAIDs withdrawn for hepatotoxicity, treating bromfenac as the cautionary archetype of a labeled limit that prescribing ignored.
2012-07-09
FDA formally withdraws the NDA
The agency publishes in the Federal Register the withdrawal of approval for Duract capsules, retiring the oral product on paper as well as in practice.

The Ten-Day Fix: How a Known Liver Signal Was Managed With a Label

Bromfenac was not a drug whose danger emerged from the shadows; it arrived with the warning printed on the dossier. In short-term clinical trials, roughly one patient in seven showed serum aminotransferase elevations to three times the upper limit of normal — a recognized marker of hepatocellular injury — and the FDA's reviewing medical officer concluded that bromfenac damaged the liver to a greater degree than other available NSAIDs. The logical response to a dose- and duration-dependent hepatotoxin would have been either to refuse approval or to brand it with the strongest available alarm, a boxed warning, which physicians are trained to read as a near-contraindication. The reviewer argued for exactly that and was overruled. The agency instead approved Duract on 15 July 1997 with a conventional label and a regulatory bet: that confining the drug to ten days or fewer would keep cumulative exposure below the threshold at which livers failed. The hazard was thus acknowledged and then delegated to a sentence of fine print — a control that depended entirely on every prescriber and patient honoring a number.

The Limit That Could Not Be Enforced: Real-World Use Outruns the Rule

The bet failed because acute pain does not respect a ten-day clock. Post-surgical, dental, musculoskeletal, and chronic pain routinely outlast a single short course, and so refills were written and capsules were taken for weeks. With roughly 2.5 million prescriptions filled in under a year, even a small fraction of extended courses produced a steady arrival of catastrophic liver injury. Crucially, the deaths and transplants clustered in precisely the over-ten-day population the label was meant to exclude — confirming that the harm and the rule were correlated, and that the rule had no enforcement beyond hope. Wyeth and the FDA responded in February 1998 the way the system was built to respond: they strengthened the warning, re-emphasizing the duration limit. But a label change cannot retract prescribing habits already in motion, and it cannot make a refill counter exist where none does. Through the first half of 1998 the case reports kept coming — fulminant hepatic failure, emergency transplantation, death — until the accumulating toll made the management-by-label strategy indefensible.

The Fastest Exit: Withdrawal, Reformulation, and a Closed File

Duract's reckoning was swift and nearly complete. On 22 June 1998, less than eleven months after launch, Wyeth-Ayerst voluntarily withdrew the drug, citing severe hepatic injury that had caused death and required transplantation, overwhelmingly in patients who had used it longer than ten days. There was no protracted scientific dispute as with longer-lived withdrawals; the pre-approval signal, the strengthened label, and the body count pointed the same way, and the only remaining question was speed. The published medical literature soon caught up — Hunter and colleagues documented in 1999 the first bromfenac-associated fulminant failure rescued by liver transplant — and by 2006 the episode had hardened into a textbook archetype in Goldkind and Laine's systematic review of hepatotoxic NSAID withdrawals. The molecule itself was not entirely abandoned: bromfenac re-emerged in 2005 as a topical ophthalmic agent (Xibrom, later Bromday and Prolensa), a formulation whose negligible systemic absorption sidestepped the liver entirely. The oral product, however, was finished, and on 9 July 2012 the FDA formally withdrew its New Drug Application — closing on paper a file that prescribing reality had closed fourteen years earlier.

Contributing Factors

01
A pre-approval hepatotoxicity signal accepted instead of acted on
Bromfenac entered the market with trial data showing ~15% of patients sustaining 3x-ULN liver-enzyme elevations and a reviewer's finding that it injured the liver more than peer NSAIDs. The agency approved it anyway, converting a documented hazard into a managed risk. When the strongest warning the system offers is requested and refused at the threshold, the drug is launched with its own indictment already written.
02
A safety control delegated to a printed limit with no enforcement
The entire safety case rested on confining use to ten days — a number on a label, not a mechanism. No refill cap, no mandatory monitoring, no dispensing lock backstopped it. Risk management that depends on universal voluntary compliance with a duration rule is not a control; it is an assumption, and one that ordinary clinical practice was always going to violate.
03
A mismatch between the indication and human pain
Acute pain frequently persists beyond ten days, so the drug's approved use and its dangerous use were separated only by patient endurance and prescriber discretion. An indication that invites the very behavior its safety limit forbids builds the failure mode into the product. The harm concentrated exactly where the label said not to go, because that is where patients in pain naturally went.
04
A label change mistaken for a remedy
Faced with mounting cases in early 1998, the response was to re-emphasize the existing warning rather than to withdraw. Strengthening fine print does nothing to recall prescriptions in flight or to alter the structural absence of a refill limit. Treating a labeling edit as a corrective action, when the original labeling was already the failed control, lost months of additional exposure.
05
A blockbuster launch velocity that amortized rare harm into real bodies
A 1-in-roughly-10,000 acute-liver-failure rate sounds tolerable until ~2.5 million prescriptions pass through it in under a year. Marketing an everyday analgesic at scale multiplies a small per-patient probability into a deterministic stream of catastrophic outcomes. Speed of uptake, not rarity of the event, set the death and transplant counts.

Aftermath

The material toll — at least four deaths, eight transplanted livers, and more than a dozen further serious injuries, with severe drug-induced liver injury reports eventually exceeding fifty — was small beside contemporaneous mass-tort drugs but disproportionate to a product on the market less than a year. Its durable ripple was regulatory and pedagogical. Bromfenac became a fixture in pharmacovigilance teaching and in formal reviews such as Goldkind and Laine's, cited whenever a hepatotoxic drug is proposed for approval under a duration restriction, as the proof that a 10-day limit on a label does not survive contact with prescribing. The episode strengthened the institutional argument — vindicated in retrospect — for honoring a reviewing medical officer's call for a boxed warning when the underlying liver data justify it, and for designing exposure controls that do not rely on perfect adherence. The molecule's reincarnation as an eye drop underscored the verdict: bromfenac was tolerable only where it never reached the liver in quantity. What remains is the byword. "Duract" is invoked whenever a drug is approved with a known toxicity that a printed usage limit is supposed to contain — shorthand for the lesson that a safety rule with no enforcement mechanism is merely a prediction of where the harm will fall.

Lessons

  1. Treat a reviewing officer's rejected boxed-warning request as a recorded prediction, not a settled dispute: when the person closest to the safety data asks for the strongest alarm and is overruled, the harm they foresaw is the most likely thing to happen next.
  2. Never accept a usage limit as a safety control unless something enforces it — a refill cap, mandatory monitoring, or a dispensing constraint; a number on a label is an assumption about behavior, and behavior under pain is the one variable you cannot regulate by printing.
  3. Match the indication to how the condition actually behaves: if the approved use forbids exactly what real patients will do, the dangerous use is not misuse, it is the foreseeable default, and the product owns its consequences.
  4. Read a strengthened warning as evidence the original control already failed, not as the fix: if a printed limit was the safety case and the limit is being re-emphasized because people are harmed, the correct response is usually withdrawal, not louder fine print.
  5. Multiply every per-patient risk by your launch curve before you scale: a rare event at blockbuster volume is a guaranteed body count, so size the toll in absolute lives at the rate you intend to ship, not at the rate you hope to dispense.

References