Journal Logo

Toxicology Rounds

Toxicology Rounds

A Fatal Drug Interaction You Need to Know

Gussow, Leon MD

Author Information
doi: 10.1097/

    The Libby Zion case is probably the best-known example of a purported fatal drug interaction. Ms. Zion, an 18-year-old Bennington College student, presented to the emergency department of a Manhattan hospital complaining of fever and ear pain on the evening of March 4, 1984. She gave a history of taking the monoamine oxidase inhibitor phenelzine (Nardil) for depression.

    Ms. Zion was admitted to a medical floor for antibiotics and observation. She was given meperidine (Demerol) apparently because of pain and the fact that she exhibited unusual muscular jerking. She became more agitated after that, and was placed in physical restraints.

    Her temperature had risen to 108°F by morning, and she suffered a fatal cardiac arrest before adequate cooling methods could be instituted. Blamed for this tragic outcome were resident and intern overwork and fatigue, inadequate attending supervision of interns and residents, failure to admit to an intensive care unit, failure to re-evaluate a deteriorating febrile patient in timely fashion, and other alleged deficiencies in cognitive thinking and clinical oversight.

    A commonly accepted explanation for Ms. Zion's death is that the meperidine given in the hospital interacted with her regular prescribed dose of phenelzine to produce serotonin syndrome, a diagnosis that can present with severe hyperthermia, mental status changes, and neuromuscular abnormalities. This potential drug-drug interaction had been the subject of several medical publications before 1984, but was certainly not widely appreciated by medical practitioners. Edward Boyer, MD, PhD, and Michael Shannon, MD, MPH, in fact, noted in their review that “more than 85 percent of physicians are unaware of the serotonin syndrome as a clinical diagnosis.” (N Engl J Med 2005;352[11]:1112.)

    I've never quite accepted that the interaction between phenelzine and meperidine explained the clinical course in this case, at least not completely. My reading of the evidence is that Ms. Zion was deteriorating and exhibiting bizarre symptoms even before receiving meperidine. But serotonin syndrome certainly could have contributed to the severe hyperthermia and agitation, and it is possible that a fatal outcome could have been avoided without the interaction.

    Fatal drug interactions are rare, but are devastating when they occur because they could have been avoided in most cases. Fortunately, neither meperidine nor phenelzine is used much anymore, making this interaction somewhat of a historical curiosity. Emergency physicians should be aware of a number of important drug interactions, however, some involving common or even over-the-counter drugs like acetaminophen and warfarin.

    Increased INR

    Aspirin and nonsteroidal anti-inflammatory drugs are generally contraindicated in patients taking warfarin because both drugs impair platelet function and can cause significant upper gastrointestinal bleeding. Acetaminophen is often considered the analgesic of choice for these patients, and many clinicians assume that it is a safe choice in these situations.

    This assumption is not completely justified, however. Evidence has accumulated since 1968 that acetaminophen can increase the international normalized ratio (INR) in patients also taking warfarin. The precise mechanism of this drug-to-drug interaction is not completely clear, but several hypotheses are plausible. The toxic metabolite of acetaminophen — N-acetyl-p-benzoquinone imine (NAPQI) — inhibits several vitamin K-dependent enzymes crucial to maintaining coagulation function.

    Warfarin is also administered as a racemic mixture, with the S-enantiomer having five times the anticoagulation effect of the R-enantiomer. Acetaminophen may selectively affect the metabolism of these two different forms, tipping the balance toward the more potent S-enantiomer.

    Authors of a retrospective study using a postmortem computerized toxicology database found that patients on acetaminophen and warfarin had 2.7 times the risk of fatal hemorrhage compared with those on warfarin alone. (Eur J Clin Pharmacol 2010;66[1]:97.)

    Some clinicians recommend warning patients on warfarin about the potential interaction with acetaminophen and that an INR should be checked three to five days after any patient on warfarin takes at least 2 g of acetaminophen per day for more than three days. No studies or evidence, however, demonstrate that more frequent blood tests would improve outcomes or decrease the incidence of fatal hemorrhage in these patients. A patient on warfarin who presents with an unexplained increase of a previously stable INR at the very least should be questioned about recent acetaminophen use.

    The hope was that electronic medical records would eliminate adverse events by alerting physicians when a patient was taking drugs that could interact adversely. Almost any drug can potentially interact with any other drug, however, and the profusion of drug interaction warnings soon produced alert fatigue and desensitization. Practitioners still have to be aware of these important interactions and not rely on the computer to save them.

    Share this article on Twitter and Facebook.

    Access the links in EMN by reading this on our website or in our free iPad app, both available at

    Comments? Write to us at [email protected].

    Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.