To the Editor:—

In December 2001, the Food and Drug Administration (FDA) issued a “black box” warning that droperidol, even at low doses, could cause QT prolongation and/or torsades de pointes. The warning emphasized that significant cardiac morbidity and mortality was associated with droperidol. The intent and/or the effect of the warning will be to markedly restrict droperidol use. We know of several hospitals that have either removed the drug from its formulary or are considering such an action. We queried the FDA to evaluate the data which led the FDA to make this warning.

Under the Freedom of Information Act, we received information about and analyzed the droperidol cases upon which the FDA warning was based. The cases were reported to the FDA over a period of time from November 1, 1997 until January 2, 2002. Some of the data are described in table 1below. There were 273 cases reported, and we identified 127 serious adverse outcomes in which the patient experienced death, prolonged hospitalization, or a life-threatening condition. The source of the case was “foreign” In 94 (74%) of these 127 cases. Problems such as alcohol intoxication, suicide attempts, general anesthesia, multiorgan dysfunction or sepsis often were confounding factors. Several of the cases were entered into the database more than once. There were 89 deaths reported, but the dose of droperidol was 2.5 mg or less in just 2 deaths. The majority of deaths involved droperidol doses that ranged from 25 to 250 mg, but the dose of droperidol was not documented in 14 deaths.

Table 1. Summary of Cases on Which the Droperidol Warning by the Food and Drug Administration Was Based

Cardiac morbidity possibly played a role in 74 of the 127 cases. Most of these cases (57/74) involved droperidol doses that were either excessive or seemingly inappropriate (e.g. , epidural). The droperidol dose was 2.5 mg or less in 17 of these 74 cases. In 12 of these 17 cases, multiple drugs were administered, including other antiemetics and antipsychotics. In 4 of these 12 cases, patients received 10 or more drugs. In only 3 of the 17 cases was droperidol, 2.5 mg or less, the only drug administered, with 1 case resulting in hospitalization and 2 resulting in life-threatening problems. A total of five patients receiving droperidol, 2.5 mg or less, experienced either ventricular tachycardia (n = 2) or torsades (n = 3) but not prolonged QT. Three of these cases were considered life threatening or required prolonged hospitalization, and only one case was fatal (table 1).

If one considers only the cases contained in the FDA database in which the droperidol dose used was 2.5 mg or less and  the outcome was death, the reported incidence of such cases is less than one per year. While this is almost certainly an underestimate (since it is based on anecdotal reporting) and considering the millions of doses given each year, it suggests that droperidol used in low doses is actually quite safe.

Droperidol is a cost-effective antiemetic. Many hospitals, in light of the FDA's warning, may remove droperidol from their formulary. Substituting newer drugs for droperidol will significantly increase costs, and the use of newer drugs might be less safe. For example, consider if a single dose of ondansetron costs about $10.50 and the cost of droperidol is$0.50: Substituting ondansetron for droperidol will increase costs $10,000.00 for every 1,000 doses. Applying this same cost analysis to one third of surgical candidates in the United States (∼10,000,000), the cost impact of the FDA warning could easily be$100 million or more. In addition, there is an unproven assumption that alternatives to droperidol are actually safer.

The FDA warning does not appear justified. Since other FDA warnings have been modified or retracted, perhaps a review of how such warnings are generated is in order.