EFFECT OF BOMA CONFINEMENT ON HEMATOLOGIC AND BIOCHEMICAL VALUES IN FREE-RANGING WHITE RHINOCEROS (CERATOTHERIUM SIMUM) IN KRUGER NATIONAL PARK, SOUTH AFRICA

Conservation programs for African rhinoceros depend on capture and translocation to maintain genetic diversity and restore populations (Emslie et al. 2009). With increasing losses due to poaching, it is crucial that animals be moved to safe locations. However, in some cases the novel environment can present new challenges, including availability of preferred vegetation, restricted space, and new social groups. To prepare for these changes and meet requirements such as quarantine for disease screening and treatment of injured animals, free-ranging rhinoceros are often placed into bomas (holding enclosures) as an adaptation step, which can last for weeks to months (Emslie et al. 2009). However, the confinement of free-ranging animals, exposure to humans, novel food, and water sources can result in stress. Although the stress response is adaptive, persistent stress can lead to physiologic disturbances resulting in weight loss, immunocompromise, susceptibility to disease, and decreased reproductive capacity (Dantzer et al. 2014; Fischer and Romero 2019). This may further complicate the success of treating rhinoceros with poaching injuries.

Adaptation to captivity may vary between individuals based on species, sex, age, origin, and other characteristics (Fischer and Romero 2019). As many as 10%–20% of free-ranging white rhinoceros (Ceratotherium simum) are unable to adapt to boma confinement (Emslie et al. 2009; Miller et al. 2016). In a previous study, a boma scoring system, which included daily evaluation of appetite, fecal consistency and volume, and behavior, was developed to provide a method to determine adaptation (Miller et al. 2016). Rhinoceros were considered maladapted if their overall scores remained below 15 out of 20 or showed a downward trend within the first 2 wk of confinement. Early release of maladapted rhinoceros has been recommended to prevent further morbidity and mortality (Rogers 1993; Kruger et al. 1999). Maladaptation results in loss of translocation candidates, along with the time and associated costs and, in some cases, death of the rhinoceros. Therefore, understanding the causes and identifying changes associated with maladaptation are necessary to prevent morbidity and mortality of recently confined white rhinoceros.

Previous studies have reported no obvious association of intrinsic (age, sex, body weight) or capture-related factors on probability of boma adaptation in white rhinoceros (Miller et al. 2022). Measurements of plasma cortisol and fecal glucocorticoid metabolites have shown that white rhinoceros have variable individual responses to capture and captivity (Kruger 2017). However, the physiologic effects of adaptability have not been investigated in boma-confined white rhinoceros. Therefore, our study aimed to evaluate changes in hematologic and serum biochemical values in boma-adapted and maladapted white rhinoceros in Kruger National Park, South Africa. This information will improve understanding of the consequences of confinement and identify changes that may require intervention.

Free-ranging white rhinoceros (n=96) were immobilized, weighed, and transported to temporary holding facilities (bomas) in Kruger National Park (23°49′60″S, 31°30′0″E), South Africa during 2009–2011, prior to translocation, and immobilized for release from bomas as described (Miller et al. 2016, 2022). All rhinoceros were apparently healthy on physical examination and not injured at the time of capture. The study population consisted of 53 females and 43 males, with the majority being subadults (3–7 yr; n=73), along with 22 adults (≥7 yr) and one animal without a recorded age. Animals were caught and cared for according to the South African National Parks (SANParks) Standard Operating Procedures for the Capture, Transportation and Maintenance in Holding Facilities of Wildlife (SANParks 2005; approved by SANParks Animal Use and Care Committee). All rhinoceros received zuclopenthixol acetate (Clopixol-Acuphase, H. Lundbeck Pty. Ltd., North Riding, South Africa); total dose range 50–100 mg intramuscularly (IM) as a long-acting tranquilizer at the time they were off-loaded into the bomas. Rhinoceros were housed in enclosures approximately 25×50 m with ample shade, water, and food offered on cement slabs. Ad libitum feed consisted of a mixture of 50% tef (Eragrostis tef) and 50% high-quality lucerne (Medicargo sativa) hay offered twice daily. Once an animal was eating and defecating normally, it was shifted to a smaller pen (10 m × 20 m). A maximum of two compatible animals were housed together in a boma, but most animals were kept individually. A normal low level of ticks was observed on all rhinoceros at the time of capture. Animals were treated with a topical ectoparasiticide (Flumethrin 0.5% m/v, Drastic Deadline, Bayer (Pty) Ltd. Animal Health Division, Isando, South Africa) at the time of capture as part of the protocol for translocation. Boma adaptation scoring was performed daily based on observations of appetite, fecal consistency/volume, and behavior, as previously reported (Miller et al. 2016). This was performed by an experienced boma manager each day the while rhinoceros were present in the bomas. Rhinoceros with an individual score of less than 15 (out of 20), or a declining score during the first 7–14 d of boma confinement, were considered maladapted. Rhinoceros with boma scores less than 15 were visually examined by a veterinarian, who decided if and when to release this individual. Capture and release weights were recorded using a hanging scale once the rhinoceros was in the crate.

Blood was collected from the radial vein into ethylenediaminetetraacetic acid (EDTA) and serum Vacutainers (BD Biosciences, Franklin Lakes, New Jersey, USA) during initial immobilization in the field and during immobilization for release from the bomas. Samples were transported on ice packs to the laboratory in Skukuza for processing within 8 h of collection. Manual white blood cell (WBC) and differential counts were performed on blood smears prepared with a commercial eosin-methylene blue stain (Kyro-Quick stain, Kyron Laboratories, Benrose, South Africa). An automated hematology analyzer (Vet ABC, Scil Animal Care Company, Gurnee, Illinois, USA) was used for red blood cell (RBC) counts, hemoglobin concentration, and hematocrit for samples collected in 2011; this was not available in 2009–2010. Serum was separated by centrifugation at 800 × G for 10 min, then 0.2 mL was added to a Large Animal chemistry rotor (ABAXIS, Inc., Union City, California, USA) placed into an ABAXIS VetScan2^® chemistry analyzer (ABAXIS) as previously described (Mathebula et al. 2012).

The length of time between the initial sample and release sample varied due to translocation logistics and reasons for release recorded. Early release (<30 d) occurred when rhinoceros were considered maladapted, based on a low daily boma score (<15; Miller et al. 2016) and inadequate appetite, abnormal defecation or behavior, or significant loss of body condition. Rhinoceros were released in their capture location and usually began eating immediately after release.

Data were assessed for normality visually using a histogram with a density curve and more formally by using the Shapiro-Wilk test for normality. Hematologic and serum biochemical results (mean, SD, median and 25th and 75th percentiles) were calculated for samples collected at the time of capture and release and compared using paired t-tests or a Wilcoxon signed-rank test (according to data distribution). Results were also compared between boma-adapted and maladapted rhinoceros using medians and the Wilcoxon rank sum test. Stata Statistical Software (StataCorp 2009) version 11 was used to perform all data manipulation and statistical analysis. Statistical significance was set at P≤0.05.

White rhinoceros were confined in the bomas for up to 187 d prior to translocation. The mean length of stay for boma-adapted rhinoceros (n=79) was 89.9 d with a range of 39–187 d, which was significantly longer than for maladapted individuals (n=19; 18 early release and one death; P<0.001) with a mean length of stay of 13 d and a range of 8–16 d. Postmortem testing of the rhinoceros that died confirmed Salmonella gastroenteritis as the cause of death. Median daily individual boma adaptation scores for adapted rhinoceros were significantly greater than those of the maladapted rhinoceros starting at day 8 of confinement (13 versus 10, P=0.0012). The adapted rhinoceros median scores continued to increase and plateaued at 16, whereas the scores for maladapted rhinoceros remained at 10 for the duration of confinement period. Although the initial mean (±SD) weights of adapted (1,460±314 kg) and maladapted (1,526±287 kg) rhinoceros were not statistically different (P=0.42), the mean weight loss between capture and release of maladapted rhinoceros (224.0±61.1 kg) was statistically greater (P<0.001) than weight loss (65.9±77.8 kg) in adapted animals. In addition, the weight loss occurred over a shorter time (8–16 d) for the maladapted rhinoceros.

Changes in hematologic and serum biochemical values that occurred during boma confinement (39–187 d) in adapted rhinoceros (capture and release data; n=63 with hematologic, n=73 with chemistry results) are shown in Tables 1 and 2, respectively. The mean total WBC counts (16.9×10⁹/L versus 14.9×10⁹/L; P=0.007) and proportions of basophils declined (0.01 versus 0.0003; P=0.01), along with a nearly 50% reduction in the proportions of eosinophils (0.162 versus 0.089; P<0.0001) between capture and release, respectively. An increase in the proportion of neutrophils at release was also statistically significant (0.431 versus 0.512; P<0.0001). There were no significant changes in proportions of lymphocyte or monocyte populations. Mean RBC counts (0.0075±0.0015×10¹²/L versus 0.0080±0.0009×10¹²/L), hemoglobin concentrations (156±37g/L versus 152±17g/L), and hematocrit values (0.443±0.086L/L versus 0.474±0.053L/L) in a subset of rhinoceros from 2011 were not significantly different between capture and release, respectively (P>0.05). Mean total protein, albumin, blood urea nitrogen (BUN), creatine kinase (CK), and aspartate aminotransferase (AST) were significantly increased at the time of release compared to capture values (Table 2; P<0.05). Significantly decreased mean alkaline phosphatase (ALP), gamma-glutamyl transferase (GGT), and magnesium (Mg) values were present at the time of release (P<0.05).

Changes in hematologic and biochemical values that occurred during boma confinement (8–16 d) in maladapted rhinoceros (capture and release data; n=15 with hematologic, n=17 with chemistry results) are shown in Tables 3 and 4, respectively. There were significant increases between capture and release, respectively, in median total WBC count (15.4×10⁹/L versus 21.1×10⁹/L; P=0.045), proportions of neutrophils (0.435 versus 0.550; P=0.0006), and decline in proportions of lymphocytes (0.260 versus 0.230; P=0.029). There was also an increase in band neutrophils, although this was not statistically significant (0 to 0.044±0.077; P=0.12). In 2011 samples, there were decreases in mean RBC counts (0.0085±0.0017×10¹²/L versus 0.0059±0.0015×10¹²/L; P=0.15) and hematocrit values (0.550±0.149L/L versus 0.372±0.102L/L; P=0.07) from capture to release, respectively, although these were not statistically significant (n=4). Median biochemical values in maladapted rhinoceros had statistically significant changes between capture and release, respectively, including increases in BUN and CK and decreases in ALP, GGT, and Mg (Table 4; P<0.05).

Comparisons of median hematologic and biochemistry values at the time of release from boma-adapted and maladapted rhinoceros are shown in Tables 5 and 6. Maladapted rhinoceros had a significantly higher median WBC count, proportion of neutrophils, and a decreased proportion of eosinophils. In addition, for 2011 samples there was a decreased median RBC count (5.9×10⁹/L versus 7.9×10⁹/L; P=0.002) and hematocrit value (0.379L/L versus 0.476L/L; P=0.014) in maladapted compared to adapted animals. There was also a significantly higher median CK and lower serum albumin, AST, BUN, calcium (Ca), phosphorus (P), and Mg values in the maladapted rhinoceros (all comparisons P<0.05).

Confinement of recently captured wild animals is recognized to cause stress and may be complicated by effects of capture, adjustment to unfamiliar dietary items, and potential anorexia associated with long-term tranquilizers (Linklater et al. 2009; Fischer and Romero 2019). In this study, apparently healthy free-ranging white rhinoceros, placed into bomas prior to translocation, showed weight loss and changes in hematologic and serum biochemical values in both adapted and maladapted animals. There was a significant difference in the time between capture and release samples between the adapted and maladapted rhinoceros, with a relatively long period between blood collections in the adapted rhinoceros, which could have influenced these findings. The trends in the adapted rhinoceros were suggestive of adjustment because many of the changes were not clinically relevant or were considered to show a return toward reference values, such as a decrease in WBC count. In contrast, weight loss and changes in blood values in maladapted rhinoceros were significantly greater and were consistent with a decline in condition or catabolic state, despite the shorter duration of confinement.

Hematologic profiles provide insight into the health status of the individual, with trends suggesting either improvement or deterioration. The decrease in mean WBC count in adapted rhinoceros was consistent with a decreased stress leucocytosis, although both values were within ranges previously reported in white rhinoceros (Miller and Buss 2015). There was no evidence of lymphopenia or anemia associated with boma confinement. The decrease in proportion of eosinophils was most probably due to the parasiticide treatment administered at the time of capture and was consistent with the decreased tick load observed at release. The relative changes in basophil and neutrophil populations were physiologically inconsequential, and values remained within previously published ranges (Miller and Buss 2015). The shifts in hematologic values were similar to those observed in other wild mammals during their adjustment to captivity (St. Auburn et al. 1989; Sánchez-Sarmiento et al. 2015). The minor changes observed at the time of release were not considered to be clinically significant and reflected the level of adjustment achieved during the longer boma confinement period (39–187 d).

Although there were several statistically significant changes in serum biochemical values of adapted rhinoceros between capture and release, many of these changes were relatively minor and reflected a trend toward return to normal reference values. The mild increases in total protein and albumin could be associated with feed quality and quantity in the bomas, or possibly decreased water intake, which was supported by a mild increase in BUN. The increases in AST and CK, along with BUN, also were consistent with a mild catabolic state, which was manifested as weight loss between capture and release. However, overall, changes were relatively small, with most remaining within previously reported serum biochemical values (Mathebula et al. 2012; Miller and Buss 2015). Therefore, although this group of rhinoceros was categorized as adapted, there were measurable physiologic changes which indicated that they had not completely achieved the same condition that they were in at the time of capture, i.e., as a group, they were in a mild-moderate negative energy balance with possible mild dehydration. Adaptation appears to take a minimum of 3 wk, based on scores reaching a plateau (Miller et al. 2016); it is recommended that white rhinoceros put into bomas be given more time (8–12 wk) to regain condition prior to being transported.

The hematologic changes in maladapted rhinoceros between capture and release were consistent with a stress response (Kock et al. 1999; Turner et al. 2002). During the relatively short period of captivity, these rhinoceros developed a stress leukocytosis with left shift, lymphopenia, eosinopenia, and anemia. Changes in WBC counts with a shift in the neutrophil:lymphocyte ratio have been associated with immune consequences of stress in several species and may lead to greater susceptibility to disease (Rogers 1993; Fischer and Romero 2019). Cases of enterocolitis caused by Clostridium and Salmonella spp. have been reported in white rhinoceros put into stressful conditions, and can lead to death, which occurred in one rhinoceros in this study (Kruger et al. 1999; M.A.M. pers. comm.). There was also a trend toward increased proportion of band neutrophils, suggesting the presence of a systemic inflammatory response (Mare et al. 2015). Anemia may be a result of bone marrow suppression due to stress or, more likely, blood loss associated with gastrointestinal ulceration. Stress is a recognized contributor to gastrointestinal tract changes in motility, shifts in microbiome, and ulceration (Caso et al. 2008). Evidence of gastritis as well as punctate ulcers has been observed by endoscopic examination of recently captured white rhinoceros (M.A.M., pers. comm.). Similar changes in hemograms have been observed in translocated black rhinoceros (Kock et al. 1999). Overall, the hematologic changes were consistent with a negative nitrogen balance or catabolic state. This hypothesized catabolic condition associated with inappetence and stress response is supported by the significant mean weight loss between capture and release (224±61.1 kgs). This is an especially important finding since the duration of confinement was short (8–16 d).

The serum biochemical values that significantly changed between capture and release of maladapted rhinoceros included increased BUN and decreased Mg values. Although there was a large increase in mean CK, it was not a statistically significant change, probably due to the small sample size. The decrease in serum Mg concentration was probably due to decreased food intake and, along with increased BUN and significant weight loss, suggests that the maladapted rhinoceros were in a severe catabolic state, which could lead to death if early release did not occur (Kruger et al. 1999; Miller et al. 2016). Although long-term monitoring after release was not performed, immediate observations after release revealed that the rhinoceros started grazing almost immediately, suggesting that they might be able to recover.

The significance of the changes during boma confinement was even more apparent when differences in blood results between adapted and maladapted rhinoceros were compared at the time of release. Maladapted rhinoceros had stress hemograms, with significantly decreased mean hematocrit, Ca, P, and Mg levels, and clinical evidence of catabolism supported by the elevated CK levels and rapid weight loss. The mean length of boma confinement of maladapted rhinoceros was 13 d, which agrees with previous reports in which the first 10–14 d are considered the critical time in which white rhinoceros need to be monitored (Rogers 1993; Kruger et al. 1999; Miller et al. 2016). Significant weight loss as well as hematologic and biochemical abnormal values should prompt immediate evaluation of the rhinoceros for intervention, which could include release, supportive treatment with fluid therapy, appetite stimulants, gastroprotection, anti-inflammatory and antimicrobial drugs, or further tranquilization.

There were several limitations in this study that may have influenced the reported results. Blood samples used to assess changes that occurred during boma confinement were only taken at the time of release. Therefore, there was a significantly shorter time period between capture and release blood samples for maladapted compared to adapted rhinoceros (mean 13 d versus 89.9 d, respectively). It is unknown whether the adapted rhinoceros had similar early changes in weight and hematologic and biochemical results, but with the longer duration of confinement, these changes began to return to more-normal values over time. Another limitation is the small sample size of maladapted rhinoceros, which affected power and the ability to declare differences statistically significant. Due to logistic limitations, observations of details in management that might have affected maladaptation were unavailable. More-detailed observations of activity, feeding, and social behavior, evaluation of diet fed, and responses associated with presence of people would be useful for comparing and assessing differences between maladapted and adapted rhinoceros. Interventions such as providing fresh-cut grass, separating incompatible animals, and administration of tranquilizers could be considered in individual rhinoceros showing signs of maladaptation. Because demographic and capture-related factors have been investigated in this cohort of rhinoceros and were not associated with maladaptation (Miller et al. 2022), further studies are needed to explore management and other variables that could predispose white rhinoceros to failed adjustment to boma confinement.

The results of this study suggest that there may be individual differences in the response of white rhinoceros to boma confinement that could be quantified. The tools used (boma scoring system, serial hematologic and serum biochemical analyses, weights) should be applied to other wildlife species that undergo translocation or confinement for treatment. These could be easily adapted to black rhinoceros boma management as well as individual or small groups of ungulates or carnivores that are captured. Identifying species-specific changes in hematologic and biochemical values during confinement would be a useful foundation to improve wildlife capture and to inform whether interventions are required as well as the effect of treatments. This system could also be applied to human-managed wildlife that are undergoing transport, treatment, or other changes in management.

In summary, findings in this study provide information on the effects of boma confinement on selected physiologic variables in recently captured white rhinoceros and identify changes that can be used to understand potential effects on health. As previously reported, the first 1–2 wk were crucial for evaluating individual rhinoceros' adjustment and determining when intervention was required prior to the development of serious consequences (Kruger et al. 1999; Miller et al. 2016). Ideally, serial hematologic and serum biochemical analyses would provide insight into the current health status of each rhinoceros; however, repeated immobilizations may exacerbate the stress associated with boma confinement. Future studies should investigate whether this could be useful information for interventions and decision-making. Although adapted rhinoceros exhibited some negative energy balance during confinement, there was a significant and probably stress- and inappetence-associated catabolic change in maladapted rhinoceros. The boma scoring system previously described (Miller et al. 2016), along with changes in hematologic and serum biochemical values, provide additional evidence for decisions regarding intervention when managing rhinoceros in bomas. Future studies should continue to investigate predisposing factors and effective preventive and therapeutic techniques to improve adaptation and welfare of recently captured white rhinoceros. This is especially important as increasing numbers of compromised rhinoceros are being housed for treatment of poaching injuries.

The authors thank the International Rhino Foundation for their financial support of this study through the International Rhino Foundation grant R2009-4. The support and participation of the Veterinary Wildlife Services team, Kruger National Park, are appreciated for their expertise in capture and care of the rhinoceros in this study. M.A.M is supported by the South African government through the South African Medical Research Council and the National Research Foundation South African Research Chair Initiative (grant 86949). The content is the sole responsibility of the authors and does not necessarily represent the official views of the funders.