When Bad Things Happen to Good Animals: Christmas Eve 1995 at the Philadelphia Zoo
In the late evening/early morning hours of 23–24 December 1995, there was a fire in the ceiling of a cinderblock holding/exhibit building housing 1.5 western lowland gorillas (Gorilla gorilla gorilla), 1.2 Bornean orangutans (Pongo pygmaeus pygmaeus), 3.1 white-handed gibbons (Hylobates lar), 2.4 ring-tailed lemurs (Lemur catta), 1.1 ruffed lemurs (Varecia variegata variegata), and 1.1 mongoose lemurs (Eulemur mongoz mongoz). Each of the 23 nonhuman primates in that exhibit died of smoke inhalation, before the Philadelphia fire department arrived at the scene. None of the animals was actually burned. No humans were physically injured. Necropsies were performed on each of the animals; smoke inhalation was confirmed as the cause of death, and specimens were saved and sent to numerous researchers. Smoke inhalation poisoning is directly related to blood carbon monoxide levels (%COHb). The zoo engaged a fire toxicology expert to generate a model to estimate the risk to mammalian species of smoke inhalation. Other events surrounding the fire, including the media and the public’s grieving process, insurance and other legal matters occupied a great deal of time and attention from zoo staff members. As we speak, construction is set to begin on a $21 million new primate facility.
Fire is a primal element. It is capable of wreaking devastating destruction in incredibly short time periods. Zoos have moral, legal, and ethical obligations to safeguard their animal collections, their staff, and visitors (including fire prevention and fire safety training), along with the “mundane” health and safety needs of the animals, staff, and public. Since most captive wild animals cannot be evacuated from a burning building in the same manner as able-bodied human beings, a unique set of parameters must be considered during the design phase for animal exhibits/buildings.
The Zoological Society of Philadelphia does not operate with night keepers. The zoo has a 24-hour security force; during night hours, this is a two-person team that patrols the entire facility. In December 1995, there was snow totalling nearly 30 cm on the ground, and the air temp was below 0°C. At approximately 22:00 on 23 December 1995, two guards thought they smelled smoke as they were conducting their patrol in the vicinity of the zoo’s World of Primates exhibit. This facility housed 23 nonhuman primates: 1.5 western lowland gorillas (Gorilla gorilla gorilla), 1.2 Bornean orangutans (Pongo pygmaeus pygmaeus), 3.1 white-handed gibbons (Hylobates lar), 2.4 ring-tailed lemurs (Lemur catta), 1.1 ruffed lemurs (Varecia variegata variegata), and 1.1 mongoose lemurs (Eulemur mongoz mongoz).
Major Amtrack/Conrail rail lines lie just outside the zoo facility, approximately 200 m from the World of Primates. The guards assumed the smoke was coming (as it sometimes had in the past) from a small fire in the rail right-of-way. The guards ignored the smoke and did not return to the area until almost 3 hours later, when they saw flames coming from the roof of the World of Primates facility. The alarm was received by the City of Philadelphia’s Fire Communications Center at 00:44 24 December 1995. Companies responded and had visible fire in the ceiling area of the north section of the World of Primates exhibit. At the same time, zoo staff—including veterinarians, technician, curators, keepers, registrar, maintenance, public relations, guest relations, the zoo president and the senior VP for business affairs—began arriving.
The “fire building” was a series of three one-story masonry buildings (total area 23x70 m) constructed in 1986. The cinderblock buildings were utilized as the indoor sleeping and holding area for the animals, and there were public observation areas that were covered but not enclosed. The fire was confined to the enclosed ceiling areas between the gorilla and orangutan buildings. However, smoke permeated through the entire facility via HVAC ductwork, and thick, black, greasy soot was heavily deposited on all surfaces inside the animal enclosures. A separate animal holding facility (Discovery Center) was connected to the World of Primates by underground ductwork. All 10 primates (bicolor and golden-headed lion tamarins) and one acouchi in this facility survived, although two pregnant female tamarins later miscarried. No humans were physically injured during the fire and its aftermath.
The fire site was cordoned off to the press as the administration building (at the other side of the zoo) became the focus for the media, the site of press conferences, etc. It was several hours before the fire marshal’s office examined the fire scene and allowed the staff to remove the animals’ bodies from the World of Primates. This operation was conducted while all the press were occupied at the administration building. By 07:00 24 December, all of the bodies were identified, tagged, weighed, and in the pathology walk-in cooler awaiting necropsy.
Although Christmas Eve is one of 5 days in the year that the Philadelphia Zoo is closed to the public, the staff and volunteers and members of the Board of Directors began arriving to the zoo as soon as they heard about the fire and the animals’ deaths. This was apparently a visceral reaction to the news and a need on their part to do something to be helpful. The keepers assigned to work that day went about their duties; several keepers (even though it was a holiday) stayed all day and were extremely helpful during necropsies by recording gross findings and assisting with measurements, collection of specimens, and keeping the prosectors supplied with instruments, etc. Many zoo volunteers and board members were able to help staff the zoo’s main switchboard and field incoming calls. The guest services manager volunteered to ensure that food would be available to staff dealing with the carcasses and to staff in the administration building. Managing staff was an important and delicate part of what happened during this long, long day. Ensuring that people ate and took rest breaks was a very important and necessary function.
Immediately following the fire and for several weeks thereafter, professional counselors were available for staff and volunteers who requested assistance. In the months following the fire, intensive fire safety inspections were conducted, over $500,000 in new fire detection and safety systems were installed throughout the zoo, and training classes in safety and fire safety were conducted for all staff.
The pathology department has a set of three-ring binders in the necropsy suite, which contain basic necropsy procedures, basic and special anatomy drawings, and individual sections on SSP species’ special necropsy instructions (one binder for birds, one for herps, and three for mammals). While the carcasses were being prepared, one keeper volunteered to photocopy 23 copies of all the necessary forms. We were able to set up five necropsy stations and run concurrent gross necropsies (zoo veterinarians and veterinary pathologists conducted the necropsies), and six carcasses were transported to the Wildlife Conservation Center’s pathology department. By 19:00 24 December, a total of 13 necropsies had been completed. By 26 December, all 23 animals had received complete necropsies. Considerable volunteer assistance from numerous staff and colleagues from other institutions made this possible in a relatively short time.
Photographs and (amateur) videotape were shot of the interiors of the buildings and the sites where the animals were found. Not one of the animals was burned. The animals were found in typical resting/sleeping locations in peaceful repose positions. There was no evidence of any struggle in any of the animal areas. Several of the lemurs had apparently evacuated their colons perimortem. Each of the animals’ skin/fur was coated with thick, greasy, black soot, which extended into the mouth and nose, pharynx, trachea, bronchi, and into the lungs. Each of the 23 animals was basically healthy. One adult gorilla was pregnant; the adult orangutan had an unsuspected air sacculitis; and some of the older animals had various degrees of arthritis. Several of the animals had bright pink subcutaneous tissues. Gross findings in each case were compatible with death by smoke (and carbon monoxide) and soot inhalation.
Confirmation of a diagnosis of death by smoke and soot inhalation requires toxicology results indicating elevated carboxyhemoglobin (%COHb) levels in the blood. Analysis of at least 2 ml of whole blood from the affected zoo animals on a co-oximeter resulted in %COHb levels significantly above those considered lethal in humans (G. Purnell, personal communication). Clotted blood will not work in a co-oximeter; however, there is a micro-diffusion technique that can be used to qualitatively measure COHb. This method requires only 0.5–1.0 ml and can be performed on a postmortem clot (G. Purnell, personal communication).
Soot samples from the animals’ skin and fur and from various sites within the building were collected and analyzed for PCBs, as we were concerned about potential risks to staff and clean-up workers. All samples tested negative for PCBs (S. Packham, personal communication).
Since the fire occurred on a holiday, some of the researchers on the SSP special request lists were unavailable to confirm that they wanted tissues. We received numerous phone calls on Christmas Eve itself, while we were conducting the necropsies, from researchers requesting parts. I believe that most of this attention was directly related to the animal taxa involved. The great apes were of interest as sidelights to researchers working on human prostate disease, human bone metabolism, etc., and these inquiries were all opportunistic. I took information and later relayed it to SSP coordinators. In addition, I explained to the researchers (none of whom knew much about zoos) about SSPs and gave them names of people to contact for future specimen requests. At this point, we did not know the composition of the soot and feared it could have toxic components, so I denied all requests for skin, hair, and full body mounts (several museums called with requests for whole bodies).
The Lawyers, the Insurance, the Public, and the Media
A catastrophic event like this fire will soon involve police/fire/insurance investigators, teams of lawyers for all potentially involved parties, and the press (both print and broadcast). It is very important to establish and maintain a chain of custody for all evidence. The principle of a chain of custody is to establish that evidence found in situ was witnessed at the time of discovery and has not been altered in any significant way (e.g., no tampering, substitution, mishandling, mislabeling, or contamination).11 This includes identifying each piece of evidence (a routine component of a complete necropsy) and controlling access to the specimens.
To assist the zoo in handling of various aspects of the fire, we retained the services of a nationally known fire toxicologist, Dr. Steven C. Packham. He flew in and consulted with us, our lawyers, and our insurance company. He took soot samples from various sites within the building and from the animals. He coordinated running samples with the City of Philadelphia’s Medical Examiner’s Office (PMEO). He also generated a model that can help estimate the risk to various mammal species of smoke inhalation. His model indicates that the animals in the World of Primates would have died in fewer than 15 minutes from smoke inhalation. These results were presented at the 1997 American Zoo and Aquarium Association (AZA) meeting.
The Philadelphia Zoo is closed to the public on Christmas Eve, Christmas Day, New Year’s Eve, New Year’s Day, and Thanksgiving. After the fire, we chose to remain closed to the public until 2 January 1996. The visiting public had formed attachments to the great apes at the zoo and were probably as emotionally devastated by the fire as the staff. We received hundreds of faxes from around the world; thousands of letters from children and adults; over $1M in contributions from individuals, schools, and corporations to a zoo renewal fund; donations of numerous artworks and sculptures. We created a “memorial gallery” that opened to the public on 2 January 1996. This gallery contained examples of the faxes, letters, and drawings; photographs of the animals; brief histories of each animal; and conservation information on the species involved in the fire. The keepers from the World of Primates were integral participants in the creation of the memorial gallery. The gallery was open until March 1996 and filled needs of both the public and staff to remember and grieve.
We received tremendous support from the AZA, including the support of PR/media specialists. The days immediately following the fire saw incredibly intense media focus on all aspects of the fire. Ill-considered comments made early on were amplified. For example, a mini-controversy over disposition of the bodies was fueled by People for the Ethical Treatment of Animals (PETA) taking a position for burial in a pet cemetery as opposed to any other method of disposition.
I firmly believe that we have an obligation to learn as much as possible from each of the animals in our care. Part of that obligation includes supporting recognized researchers who make requests through the SSP coordinators and such facilities as the U.S. Fish and Wildlife Service (USF&WS) Forensics Laboratory. We sent specimens to as many researchers as we were able. We sent the skeletons of 1.1 of each species (and the young orangutan) to the USF&WS Forensics Laboratory, and the other skeletons were sent to a zooanthropology museum.
We arranged to have life molds and casts made from the hands and feet of 1.1 adult gorilla, orangutan, and gibbon. Molds and casts are being made of the entire skeletons of selected individuals of each species. We see this as a means of ensuring the continuing contribution of these animals to the education of humans worldwide. The zoo’s portion of the proceeds from these sales goes to a restricted account for primate conservation, managed by the Philadelphia Zoo.
The soft tissues were cremated and returned to the zoo for inclusion in a planned memorial in the new primate exhibit, located on the site of the old exhibit, which was completely demolished in March 1997. There has been an accelerated design phase for the new exhibit, and construction is set to begin in fall 1997, with the grand opening set for 1999.
Fires can have four components: flame, heat, smoke, and gases. Heat sufficient to cause harm to lungs is often sufficient to cause fatal obstructive glottal edema in humans.16 Combustion of plastics introduces other gases, including chlorine, phosgene, and hydrochloric acid (HCl). A latent period of 1–6 hours may elapse before pulmonary signs of exposure to HCl manifest. Concentrations of greater than 100 ppm of HCl may result in pulmonary edema and laryngospasm.6,15 Heat-retaining concrete releases HCl for as long as 60 minutes after fire “knock down.”6,15 Although most authors use smoke inhalation and carbon monoxide intoxication interchangeably, there is evidence that immediate respiratory distress may be seen due to the simple mechanical irritation of smoke particles. Carbon particles may also enhance toxic gas uptake by adsorption.6,15 We analyzed soot samples for PCBs. Soot can also be analyzed for chlorine, hydrogen cyanide and dioxins.
Carbon monoxide (CO) is the most common and most serious gaseous hazard during a fire.8 In concentrations of 0.5–1%, it may be lethal to humans in 5 minutes.2 CO poisoning causes a multitude of effects due to tissue hypoxia and possibly cellular poisoning, primarily affecting the central nervous system and the myocardium.2 It is possible to use the %COHb data with known respiratory intoxication rates to generate a “clock” to time the growth of the fire up to the point of death (S. Packham, personal communication).
Carbon monoxide exposure during pregnancy is teratogenic14 and has been reported to cause stillbirth and congenital defects.1,3,5,7,12,17 It was interesting that, of the surviving animals in Discovery House, the only two pregnant animals did not have successful births. Both of these mothers later conceived and bore healthy, live offspring.
Cherry-red skin and mucous membranes are not commonly found.4,10,13 Analysis of blood samples for %COHb is proof of carbon monoxide poisoning. Blood samples collected during necropsy (even hours after death) and properly handled will yield valid results.9 However, blood samples collected hours after a necropsy may have artificially low %COHb (G. Purnell, personal communication).
We attempt to learn as much as possible from each of our captive wild animals. In this case, we tried to ensure that, even after death, these animals can continue serving as eloquent ambassadors for their species.
I had a tremendous amount of help from a large number of people during a time that was difficult for everyone. I am indebted to Drs. K. Hinshaw, D. Ialeggio, K. Wright, M. Goldschmidt, D. Dambach, D. McAloose, A. Osborne, M. Lin and to S. Skeba, K. Kranz, Dr. S. Packham, Dr. T. Grand, G. Purnell, P. Farina and a terrific group of keepers.
1. Barlow, S.M. and F.M. Sullivan. 1982. Reproductive Hazards of Industrial Chemicals. An Evaluation of Animal and Human Data. Academic Press, London.
2. Beasley, V.R. 1990. Smoke inhalation. Vet. Clin. N. Am: Small Anim Pract. 20(2):545–556.
3. Brown, D.B., G.L. Meuller, and F.C. Golich. 1992. Hyperbaric oxygen treatment for carbon monoxide poisoning in pregnancy: a case report. Aviat Space Environ Med. 63:1011–1014.
4. Burney, R.E., S. Wu, M.J. Nemiroff. 1982. Mass carbon monoxide poisoning: clinical effects and results of treatment in 184 victims. Ann. Emerg. Med. 11:394–399.
5. Caravati, E.M., C.J. Adams, and S.M. Joyce. 1988. Fetal toxicity associated with maternal carbon monoxide poisoning. Ann. Emerg. Med. 17:714–717.
6. Dyer, R.F. and V.H. Esch. 1976. Polyvinyl chloride toxicity in fires. J Am Med Assoc. 235:393–397.
7. Farrow, J.R., G.J. Davis, and T.M. Roy. 1990. Fetal death due to nonlethal maternal carbon monoxide poisoning. J. Forensic Sci. 35:1448–1452.
8. Fein, A.M. 1989. Toxic gas inhalation. Emerg. Med. 21:53–63.
9. Kojima, T. 1986. Production of carbon monoxide in cadavers. Forensic Sci. Int. 32:67–77.
10. Larcan, A. and H. Lambert. 1981. Current epidemiological, clinicobiological and therapeutic aspects of acute carbon monoxide poisoning. Bull. Acad. Natl. Med. (Paris). 165:471–478.
11. Melbye, J. and S.B. Jimenez. 1997. Chain of custody from the field to the courtroom. In: Forensic Taphonomy. W.D. Haglund and M.H. Sorg, eds. CRC Press, NY, NY. Pp. 65–74.
12. Morris, G.L. 1985. Perinatal piglets under sublethal concentrations of atmospheric carbon monoxide. J. Anim. Sci. 61:1070–1079.
13. Norkool, D.M. and J.N. Kirkpatrick. 1985. Treatment of acute carbon monoxide poisoning with hyperbaric oxygen: a review of 115 cases. Ann. Emerg. Med. 14:1168–1171.
14. Norman, C.A. and Halton, D.M. 1990. Is carbon monoxide a workplace teratogen? A review and evaluation of the literature. Ann. Occup. Hyg. 34:335–347.
15. Pasternak, M., B.T. Binn, and R.F. Browner. 1982. Studies of the chemical mechanism of smoke particulates formation during the combustion of chlorinated polymers. Combus. Sci. Technol. 28:263.
16. Watanabe, K. and K. Makino. 1985. The role of carbon monoxide poisoning in the production of inhalation burns. Ann. Plastic Surg. 14:23.
17. Woody, R.C. and M.A. Brewster. 1990. Telencephalic dysgenesis associated with presumptive maternal carbon monoxide intoxication in the first trimester of pregnancy. J. Toxicol. Clin. Toxicol. 28:467–475.