In the months following the Exxon Valdez oil spill (EVOS), 994 sea otters (Enhydra lutris) from oil-spill-affected areas died (Doroff et al. 1993). Carcasses collected from these areas and otters that died in rehabilitation centers are included in this number. The actual number that died was probably much greater.

Within days of the spill, the Exxon Company (USA) funded an effort to rehabilitate oil-contaminated sea otters (Davis 1990). Initially, clinical veterinarians working on the rehabilitation effort performed partial necropsies on some of the sea otters that died. Soon, veterinary pathologists from the University of Alaska and the U.S. Environmental Protection Agency provided assistance. Later, rehabilitation centers were constructed and other veterinarians with special training in pathology were hired by Exxon to provide diagnostic support.

In late April 1989, veterinary pathologists from the U.S. Fish and Wildlife Service (USFWS) assumed responsibility for pathologic evaluation of oil-spill-affected sea otters. The USFWS requested assistance from veterinary pathologists of the Armed Forces Institute of Pathology (AFIP) in June 1989. Eventually, as part of the Natural Resources Damage Assessment program, AFIP veterinary pathologists were asked to carry out histopathological studies of the tissue specimens collected by all parties and to perform necropsies on carcasses that had been collected and frozen. A veterinary clinical pathologist was requested to assess hematology and clinical chemistry findings in otters that had been held in the rehabilitation centers.

In spite of the best efforts of many dedicated people working under extremely difficult conditions, there are significant limitations in the pathological studies. The absence of a detailed necropsy protocol and of full documentation of necropsy findings during the first several weeks after the spill resulted in important data being lost. Often, samples of all major organs were not collected. In some cases, no necropsy report was available. Specimens for toxicologic analysis for petroleum hydrocarbons were not consistently collected. The absence of a detailed toxicology protocol suggests that the samples may not have been collected properly. The lack of a consistent numbering system for identification of specimens caused major problems; some samples were useless because they could not be identified. Many blood samples could not be transported to laboratories quickly enough to prevent significant deterioration of the samples because of inclement weather and the remote locations of the rehabilitation centers. Thus, in a number of cases, data could not be used because significant deterioration of the specimens was considered likely to have occurred. Since more than one laboratory was used to analyze blood samples, problems with comparability of results were encountered. The laboratory tests were performed to aid the clinical veterinarians in diagnosis and treatment of individual animals, not as part of a consistent protocol; thus, there is variation in the amount and type of laboratory data available for each otter. These problems illustrate the need for development of contingency plans that include detailed protocols before disasters occur.