Published reports from France and Norway suggest a frequent incidence of anaphylaxis to rocuronium and have raised concerns about its safety. We hypothesized that the Food and Drug Administration Adverse Event Reporting System could be used to confirm whether there has been an unusual incidence of anaphylactic events for rocuronium in the United States (U.S.) and whether the reporting patterns differ within and outside of the U.S.. We queried the Food and Drug Administration Adverse Event Reporting System for 1999 through the first quarter of 2002 for all adverse events for the drugs rocuronium and vecuronium and then searched on the terms considered to represent possible anaphylaxis using proprietary software. We compared the frequency of these terms in data both for rocuronium and vecuronium. We then assessed the occurrence of reports of anaphylaxis-related terms in reports from the U.S. compared with reports originating outside of the U.S.. For rocuronium, the database contained 311 reports, 166 domestic and 145 from foreign sources. Fifty percent of the foreign reports contained an anaphylaxis term versus 20% of the domestic reports (P < 0.001). For vecuronium, the comparable figures were 17% and 19% (not significant) and the total number of reports was 243. The incidence of the reports containing anaphylaxis terms did not differ between vecuronium and rocuronium in the U.S. but were significantly different for foreign reports (P < 0.001). These data confirm that U.S. anesthesia providers have not observed a significant difference in anaphylactic reactions between the two commonly used intermediate-acting muscle relaxants and suggest that frequency of reports of anaphylaxis may be significantly influenced by the area from which the reports originate.