ANALYTICAL APPROACH The Case of the Tainted Richard H. Eckerlin, Joseph G. Ebel, Jr., Jack D. Henion Equine Drug Testing and Toxicology Diagnostic Laboratory New York State College of Veterinary Medicine Cornell University 925 Warren Dr. Ithaca, NY 14850 Thomas R. Covey Sciex, Inc. 55 Glencameron Rd., #202 Thornhill, Ontario, Canada L3T 1P2 Recently, accidental and malicious contamination of pharmaceuticals has been reported in the United States (1). Fortunately, through observations by alert medical professionals, the signs of poisoning have often led to a timely diagnosis and catastrophic conse- quences have been avoided. Unfortu- nately, there is much less control and monitoring of similar problems in vet- erinary medicine. Therefore adverse drug reactions in animals caused by contaminated pharmaceuticals and feeds may occur. In many instances these problems are not recognized. A general lack of control over an animal's environment clouds the clinical picture and does not provide clues to the possi- ble source of the problem. We recently encountered a situation in which a grossly contaminated vial of injectable dexamethasone caused the death of four animals. The suspect dexamethasone was rigorously ana- lyzed using standard chemical tech- niques, to no avail. Thus new analytical technology had to be applied to identi- fy the toxic component in the tainted injectable dexamethasone formulation. History A referring veterinarian contacted our laboratory staff because of a drug- related incident. A bottle of generic in- jectable dexamethasone with an as- sumed formulation similar to that of a brand-name injectable dexamethasone was used in a clinically acceptable Dexamethasone manner. However, the loss of three ani- mals and the near loss of a fourth ani- mal (all of whom were injected with dexamethasone from the same vial) alerted the veterinarian to a potential toxicology problem. A normal dog ac- quired from a client for euthanasia was then injected with the suspect dexa- methasone and died within 10 min. The suspect vial and a similar vial with the same lot number from the animal clinic were given to our laboratory for chemical analysis. Initial methodologies The two vials of suspect injectable dexamethasone were compared with a vial of control dexamethasone using ex- tensive thin-layer chromatography (TLC) and gas chromatography/mass spectrometry (GC/MS). However, no differences were found among the three vials. The samples were also ana- lyzed for the presence of cyanide and fluoride as well as for gross contamina- tion by inorganic materials via energy- dispersive X-ray fluorescence spectros- copy; the results showed no differ- ences. Inductively coupled plasma emission spectroscopy also did not re- veal any significant differences among the three vials. Alternate methodologies Because the TLC and GC/MS results were inconclusive, further biological testing was undertaken to verify whether the toxic component was still present and active in the suspect dexa- methasone sample. The goal of these experiments was to reproduce the syn- drome with a similar dosage in a clini- cally healthy dog as well as in another species (mice). If this effect could be demonstrated, we would have a biologi- cal means of monitoring the fate of the toxic component following efforts to isolate it by selective extraction from the aqueous medium. If the unknown toxic component could be separated in 0003-2700/89/0361-053A/$01.50/0 © 1988 American Chemical Society ANALYTICAL CHEMISTRY, VOL. 61, NO. 1, JANUARY 1, 1989 · 53 A