Letters to the Editor: Droperidol Controversy Mounts
FDA Should Reconsider Droperidol Warning
To the Editor:
The FDA has recently issued a warning concerning droperidol, and this warning was cited in the recent APSF Newsletter (Winter 2001-02). The FDA warning indicated that there was a significant risk of droperidol, even when used in "low" (0.625-1.25 mg) antiemetic doses, inducing serious arrhythmias. The severity of the warning effectively threatens the routine use of the most cost-effective antiemetic administered to millions of patients over several decades. I was skeptical that evidence existed justifying the FDA's warning. I also felt that, to a certain degree, "strong-arm" tactics were being employed. Therefore, I acquired, under the freedom of information act, the FDA's printout of the adverse reports which served as the basis of their conclusions. Interestingly, I needed to inquire 3 times and wait more than 2 months for the FDA to deliver the requested information.
The facts of the adverse droperidol reports were more than interesting. They can be tabulated in many ways. Briefly, there were 273 adverse reports, although several of the cases contained in the report were obvious duplicates or even triplicates. The cases were reported over a 4-year period, from late 1997 to late 2001. I tried to separate from these 279 cases those that reported arrhythmias as an adverse outcome, especially those identified as either prolonged QT interval or Torsades. I also was interested in the dose range of droperidol in those particular cases, and whether or not other drugs were concomitantly administered.
Eight cases involving 0.625 mg were reported, but 2 of these were duplicated, resulting in only 6 such cases in total. Three patients receiving 0.625 mg experienced a tachycardia (presumably sinus), 2 experienced ventricular tachycardia, and 1 experienced Torsades. Of the cases containing a report of prolonged QT, doses of droperidol were 50 mg, 250 mg, and 0.25 mg/kg. Of the 13 cases reporting Torsades, droperidol doses were 0.625 mg (n=1), 2.5 mg (n=2), 3.75 mg (n=1), 25 mg (n=2), 200 mg (n=1), 240 mg (n=3), and an unknown amount in another 3 cases. In many of these cases the route of administration, oral, intravenous, or intramuscular, was not specified. Additional medications were administered in most, but not all of the cases. Many other adverse outcomes were also reported. Frequently, high doses of droperidol and other drugs were involved, and/or cases were complicated by suicide, alcohol intoxication, and so on. For example, of the 79 cases where an arrhythmia was reported, 9 involved a droperidol dose of 10 mg or more, and 33 involved a dose of 20 mg or more. Numerous reported cases involved droperidol doses of 25, 50, and even larger doses.
I fail to see how the data contained in the summary of the adverse reports could be interpreted to mandate the severe warning issued by the FDA, particularly regarding the indication for prophylaxis and treatment of perioperative nausea and vomiting. In addition, many of the serious adverse reports involve very large or excessive doses of droperidol, frequently in patients receiving other psychotropic medication. Finally, the source of many of the adverse case reports was often identified as "Foreign."
I, like many other anesthesiologists, have administered droperidol to hundreds of patients in the dose of 0.625 to 2.5 mg, for many years. The only hemodynamic consequence I ever see is a predictable (and often intended) decrease in blood pressure of 20-30 mm Hg that lasts for 5 to 10 minutes. If one separates out this dose from the adverse report summary, there are a total of 3 cases where a dose of droperidol of 0.625 to 2.5 mg resulted in Torsades. At roughly 1 case per year (3 cases over 4 years), and in light of the likely enormous denominator considering the widespread use of droperidol, this can hardly be a serious indictment.
The literature supports the use of droperidol as a first line antiemetic for postoperative nausea and vomiting (PONV). Many hospitals, in light of the FDA's warning, must now feel obliged to remove droperidol as the first line perioperative antiemetic. Are the alternatives safer? Ondansetron and other similar drugs, if routinely substituted for PONV prophylaxis and treatment, will certainly significantly increase related costs. Of note, quite a few of the patients in the same adverse drug reports also received a 5HT3 antagonist.
The FDA should reconsider how it comes to making such rather dramatic warnings that do not appear to be justified. They should also weigh the consequences of their actions and the alternatives that clinicians will be forced to use. I certainly hope the FDA is not being led by its nose by the pharmaceutical industry, which certainly does not stand to make much money from any of the "older" drugs that we use. Witness the repeated shortages of commonly used, excellent, and necessary drugs such as fentanyl and naloxone. I, for one, continue to administer droperidol to many patients who merit and stand to benefit from prophylactic treatment of PONV.
Peter L. Bailey, MD
Professor, Anesthesiology
Director,
Cardiac Anesthesia
Strong Memorial Hospital
University of Rochester
Rochester, NY
Providers Need to Take Warning Seriously
To the Editor:
While the data linking droperidol to serious ventricular arrhythmias undergo appropriate scientific sifting and winnowing, we believe anesthesia providers should weigh seriously the FDA "black-box" warning regarding the use of even low dose (0.625-1.25 mg) droperidol as perioperative therapy for nausea and vomiting. Our increasing recognition of the electrophysiologic, pharmacological, and even genomic basis for polymorphic ventricular tachycardia, along with awareness of adverse events associated with butyrophenones in the critical care arena, will place the action by the FDA in appropriate context.
QT Interval and Torsades
Drugs that evoke K channel blockade prolong cardiac repolarization, and thus, prolong the ECG QT interval. When excessive, this may evoke triggered activity in the ventricle, causing a polymorphic ventricular tachycardia known as Torsades de Pointes. Low serum K, slow heart rates, and pre-existing genetic factors that reduce K channel function may additively (or even synergistically) predispose patients to QT prolongation, and thus, to drug-induced Torsades. Torsades occurs in 1-8% of patients who receive QT-prolonging drugs, and although "idiosyncratic," may also be viewed as an acquired form of the rare congenital long-QT syndrome. With rapid advances in sequencing the human genome, ion channel mutations have been identified that provoke arrhythmias only when patients are exposed to K-channel blocking drugs.1 These "silent" mutations provide a genetic rationale and explanation for the untoward response of patients to a myriad of drugs that prolong the QT interval. Although we cannot yet identify these "acquired" long QT patients through efficient genetic screens, these technologies are developing.
An ever expanding number of therapeutic agents are associated with QT prolongation and Torsades, including many antiarrhythmics, phenothiazines, butyrophenones, antihistamines, antifungals, antibiotics, and tricyclic antidepressants. The common thread of proarrhythmic risk and QT prologation in many of these compounds is potent blockade of the rapid delayed rectifier cardiac potassium channel (IKr), encoded by the gene known as HERG.2 Specifically, droperidol blocks IKr at therapeutic concentrations of 10-400 nmol/L.3,4 Furthermore, these effects can be compounded by metabolic interactions at the cytochrome p450 system, where drugs such as cyclobenzaprine, fluoxetine, and others increase droperidol toxicity.5 At present, clinical awareness of the patient's family history, recognition of potential arrhythmia triggering agents, and judicious use of QT interval monitoring are clinical practices that should be considered.
ICU Experience with Butyrophenones
Interactions of butyrophenones and long QT syndrome have been noted for many years in the ICU.6-9 While the documented incidence of life-threatening cardiac arrhythmias is small, 72% of patients who develop ventricular tachycardia associated with butyrophenones have a history of cardiovascular disease. For instance, one patient developed Torsades de Pointes secondary to intravenous haloperidol even after revascularization (CABG) surgery.10
Is the FDA Action Warranted?
We, like others, await further study and understanding of the actual perioperative cardiac risk associated with butyrophenones. Clearly, some agents prolong the QT interval, but rarely cause Torsades (e.g., amiodarone). However, unlike the decision to use amiodarone for life-threatening arrhythmias, a wide number of alternative strategies and drugs already exist for perioperative antiemetic therapy. Hence, until further clinical data are available to ensure the safety of droperidol, clinicians should carefully evaluate the warning expressed by the FDA. A fundamental anesthesia safety principle may also be applicable to this situation: where insufficient data exist to make an evidence-based decision or comparison, a "fail-safe" approach should apply. In the absence of sufficient data with droperidol, the apparently safer approach should prevail. Hence, we would urge practitioners to thoughtfully review the many factors in the perioperative period which can lower the threshold for drug-induced Torsades (electrolyte imbalance, bradycardia), and in that context reconsider the benefit/risk ratio of droperidol administration for postoperative nausea.
Richard C. Prielipp, MD
Professor and Section Head
Critical Care Medicine
Department of Anesthesiology
Wake Forest University School of Medicine
Winston-Salem, North Carolina 27157-1009
Jeffrey R. Balser, MD, PhD
Chairman, Department of Anesthesiology
Associate Dean, Physician-Scientist Development
The James Tayloe Gwathmey Professor of Anesthesiology and Pharmacology
Vanderbilt University School of Medicine
Nashville, Tennessee
Disclosure: Drs. Prielipp and Balser are consultants and advisory board members of Wyeth-Ayerst; Dr. Prielipp is also a consultant for both GlaxoSmithKline and Abbott Laboratories.