Serum Cortisol in Captive Asian Elephants (Elephas maximus) in Different Management Systems at Busch Gardens Tampa Bay
American Association of Zoo Veterinarians Conference 2006
Ray L. Ball, DVM; Otto Fad
Busch Gardens Tampa Bay, Tampa, FL, USA

Abstract Introduction

Cortisol is a widely accepted measure of stress in wild and captive animals. In the past, captive elephant management systems have been criticized as potential stress inducers. The analysis of fecal cortisol is non-invasive and has been used to give long-term evaluations of social and ecologic pressures in elephants and other species.2 Salivary cortisol has also been used as a minimally invasive technique to measure social stress in captive elephants.1 The herd of Asian elephants at Busch Gardens Tampa Bay (BGT) changed from a traditional contact management (free contact, FC) to a protected contact (PC) system utilizing positive-reinforcement-based operant conditioning in 2004. Serum cortisol was measured after the change and evaluated along with banked samples from before. Long-term sampling will be utilized to measure this transition but evaluating a single process will hopefully reflect the overall changes that can be expected with this change in management. While the individual variations are notable and other issues potentially confound the issue, it appears that this transition has lowered the serum cortisol in this herd. In addition to serum cortisol measurements, the actual process of collecting the samples appears to be less stressful behaviorally. Pathologic processes should not be discounted when considering cortisol levels in evaluating stress in captive elephants.

Methods and Materials

Six female Asian elephants (Studbook numbers 30, 32, 304, 34, 35, 3) had been managed in a free-contact system for many years. Studbook number 304 was captive born and the others were wild born. Serum was collected intermittently during this management system to bank and for reproductive hormone analysis. The elephants were placed in lateral recumbency by the handlers and blood collected from the ear vein on the caudal aspect of the down ear. Reproductively sound animals were bled more frequently than the others. Serum was frozen at -80°C until analyzed.

In August 2004 the first group of three animals was moved to the new barn and started the new positive-reinforcement, PC management system. Within five weeks, all animals had been moved over. All animals had been trunk washed and were culture negative for Mycobacterium tuberculosis and negative on the newly developed multiantigen print immunoassay (MAPIA) and lateral-flow technology (rapid test) developed to detected antigen to M. tuberculosis.

As the caudal aspect of the ear was used for sampling, each elephant was asked to station in a static chute designed to allow training of voluntary ear-presentation for manipulation and blood collection. Handler safety and creating an effective learning environment for the elephants required training each to proceed to the chute solo and station there calmly. General desensitization techniques were applied as session durations were increased. Within the chute, individual elephants had significant room to maneuver. Since no physical restraint or sedation was utilized, animals were trained to cooperate fully and voluntarily, allowing for blood sampling and other husbandry procedures. By May 2005 training for voluntary bloods draws was firmly established on all six animals.

The first approximately 20 samples collected under this new system were matched against the samples collected in the previous system. Samples were selected against the animal having an active problem or being on therapy for any reason. Several animals had undergone a drug trial and these samples were selected against as well. Serum was again stored in -80°C freezer until analyzed at Conservation and Research Center (CRC) Endocrine Research Laboratory, Smithsonian Institution, National Zoological Park, Front Royal, VA. T-tests were utilized to discern any statistically significant results in the mean serum cortisol collected from animals before and after the implementation of the new husbandry systems. Results were considered significant at alpha levels <0.05.


The results and simple means of serum cortisol are listed in Table 1. Elephant number 34 had essentially the same level of cortisol in both systems. Elephant number 32 had a reduction in the mean cortisol level of approximately 32% (20.84 versus 14.28 ng/ml) from the FC to the PC system. Elephant number 304 had a similar reduction of 37% in the mean cortisol (22.59 versus 14.29 ng/ml). Statistical analyses results are reported here (means, standard deviations, t-test results).

Table 1
Serum cortisol concentration of Asian elephants at Busch Gardens Tampa Bay in free contact management (FC) and protected contact management (PC).a [CLKB6]


Serum was chosen over salivary and fecal sampling to measure cortisol for several reasons. While fecal and salivary cortisol changes can reflect stresses within a reasonable period after the stressor (approximately 24 hours),1,6 serum cortisol is more likely to be reflective of the stressors closer to the moment of sampling. The methodology is straightforward and less subject to the hazards for sample storage. Timeliness of the sample result is also a benefit to serum sampling. Blood sampling is a required husbandry practice in all elephant holding facilities belonging to the American Zoo and Aquarium Association (AZA). While fecal cortisol samples may be useful to look at over a long-term period to evaluate the transition from FC to PC, we choose to additionally look at how one specific task, blood collection, was affected by making this transition. Fecal cortisol has been used to measure stress in transportation and environmental stress in some species but are not thought to be reflective of the stress in a diagnostic procedure itself.4 For this evaluation, the lag time between the potential stressor (blood collection) and the means to measure the stressor are same. Elephants number 304 and 32 both had significant reductions in the mean serum cortisol levels. Both are in good health and had no apparent inflammatory problems. The logical deduction here is that the sampling process itself is less stressful in the PC management than the FC management. Elephant 34 and 30 had essentially the same level of serum cortisol as measured by the mean in the different management systems. Elephant 34 has developed significant uterine leiomyomas during the period measured. Elephant 30 has recently had clinical bouts of anterior enteritis and is suspected of having a dietary hypersensitivity to wheat. Even with these two pathologic processes, the serum cortisol did not rise. Elevations in cortisol are quite often explained as resulting from social, behavioral, or environmental causes and little attention is paid to inflammatory causes. Associations between infections and elevated cortisol3,5 have been noted in wild animals. It is reasonable to assume that if these two processes did not exist, these levels would indeed be lower. Based on the other two elephants, a reduction of approximately 30% could be expected. Overall, it appears that collecting blood from the elephants at BGT in the PC system is less stressful that the FC system. As this is an example of how the routine husbandry and medical husbandry is now conducted, it can be expected that the overall net effect is going to be lowered stress in the elephants at BGT.


We would like to thank the Pachyderm Palace (elephant management) team and the veterinary staff at Busch Gardens Tampa Bay, but especially the six elephants for their patience with us.

Literature Cited

1.  Dathe HH, Kuckelkorn B, Minnemann D. Salivary cortisol assessment for stress detection in the Asian elephant (Elaphas maximus): a pilot study. Zoo Biol. 1992;11:285–289.

2.  Foley CAH, Papagegeorge S, Wasser SK. Noninvasive stress and reproductive measures of social and ecological pressures in free-ranging African elephants. Conserv Biol. 2001;15(4):1134–1142.

3.  Hoby S, Schwarzenberger F, Doeherr MG, et al. Steroid hormone related male biased parasitism in chamois, Rupicapra rupicapra rupicapra. Vet Parasitol. 2006;Feb 21.

4.  Mostl E, Maggs JL, Schrotter G, et al. Measurement of cortisol metabolites in faeces of ruminants. Vet Res Commun. 2002;26(2):127–139.

5.  Muehlenbein. Intestinal parasite infections and fecal steroid levels in wild chimpanzees. Am J Phys Anthropol. 2006.

6.  Stead SK, Meltzer DG, Palmer R. The measurement of glucocorticoid concentrations in the serum and faeces of captive African elephants (Loxodnda africana) after ACTH stimulation. J S Afr Vet Assoc. 2000;71(3):192–196.


Speaker Information
(click the speaker's name to view other papers and abstracts submitted by this speaker)

Ray L. Ball, DVM
Busch Gardens Tampa Bay
Tampa, FL, USA

MAIN : All : Cortisol in Elephants in Different Management Systems
Powered By VIN