Use of a Portable Analyzer for Venous Blood Gas and Biochemistry Analysis in Free-Ranging Indian Flying Foxes (Pteropus giganteus) in Myanmar

Obtaining reliable hematologic and biochemical parameters for wildlife in the field can be challenging, but portable blood analyzers can facilitate monitoring of health and physiology of free-ranging species. We used an i-STAT 200 blood gas analyzer with CG8+ and CHEM8 cartridges (Abbott Point of Care Inc., Princeton, New Jersey, USA) to measure acid–base status, blood gas, and biochemistry values in free-ranging Indian flying foxes (Pteropus giganteus).

Okkan, Myanmar, was selected for its close wildlife–domestic animal–human interface. Our study population was sourced from a colony of approximately 250 P. giganteus known to roost above human residences. Colony size fluctuates seasonally, with a peak population of about 2,000 individuals during the wet season. Ambient temperatures during the study period ranged from 27.8 to 35.0 C, and humidity averaged 66%.

As part of a larger viral surveillance and movement tracking project, 13 apparently healthy males, 3 juveniles and 10 adults, were captured over three nights in April 2018 using a size 11 nylon mist-net suspended between bamboo masts. All fieldwork was conducted in accordance with the Institutional Animal Care and Use Committees of the University of California at Davis (protocol 19300) and Smithsonian's National Zoological Park (protocol 16–05), and with approvals of Myanmar's Ministry of Agriculture, Livestock and Irrigation and Ministry of Natural Resources and Environmental Conservation. Each night, the nets were placed in a different location within 100 m of the same roost. Bats were captured in the evenings at the time of roost departure, then disentangled from the nets and placed within pillowcases within 5 min of capture. They remained in the pillowcases for up to 40 min before handling. Animals were examined, sexed, and aged based on the presence of penile barbs. Body condition was scored based on pectoral muscle mass and sternum prominence (McLaughlin et al. 2007; Hossain et al. 2013a, b); hydration was assessed through skin turgor, salivation, and position of the eye in the socket. All animals were deemed to be in good body condition, with appropriate hydration status. Because bats were also fitted with GPS collars for a separate study, recaptures were avoided. Species was later confirmed by DNA barcoding (Townzen et al. 2008).

Physical examination and sample acquisition occurred within a 5-min time span. For each bat, 2 mL of blood was drawn from the brachial vein at the distal humerus and transferred to a lithium heparin vacutainer (BD Vacutainer^™;, Becton, Dickinson and Company, Franklin Lakes, New Jersey, USA). Blood gas measurements were obtained immediately, using CG8+ cartridges, from whole blood directly from the syringe. Within 20 min of sample acquisition, biochemistry values were measured, using CHEM8 cartridges, on the blood in the lithium heparin vacutainers. The i-STAT unit and cartridges were kept inside a cooler with ice packs; cartridges were removed immediately before blood collection, and the i-STAT was brought out for cartridge placement but returned to the cooler during sample analysis. All animals received 10–20 mL of fruit juice orally for hydration and were released near the trap site, with flight monitored for normality.

Table 1 displays descriptive statistics for body weight, blood gas analysis, and biochemical analysis. Hematologic and biochemical values have previously been reported in P. giganteus (Heard and Whittier 1997; McLaughlin et al. 2007; Hossain et al. 2013a, b; McMichael et al. 2015). Hematocrit, blood urea nitrogen (BUN), bicarbonate, sodium, potassium, and chloride in this study appeared similar to other pteropids; as expected for frugivorous bats with low-protein diets, BUN was relatively low (Heard and Whittier 1997; McLaughlin et al. 2007; Hossain et al. 2013a, b; McMichael et al. 2015). Our sample size was small, and further evaluation would be beneficial; a minimum of 40 individuals is recommended for reference range determination and comparison between species, sexes, and seasons (Solberg 1987; McMichael et al. 2016).

Compared to other reports from P. giganteus (McLaughlin et al. 2007; Hossain et al. 2013a), a broad range of blood glucose concentrations was observed in our study (Table 1) from both the CG8+ (mean 224.50±SD 69.720, range 87–337 mg/dL) and CHEM8 (mean 238.10±SD 111.07, range 69–396 mg/dL) panels. High glucose levels probably reflect acute or prolonged physiological stress in response to capture and handing, mediated by a glucocorticoid surge. As reported previously in Pteropus spp. bats, capture, restraint, and handling-induced stress can induce elevations in plasma cortisol even at 90 min postcapture (Smythe et al. 1989; Widmaier and Kunz 1993; Reeder et al. 2004; McMichael et al. 2014). Another possible explanation is fruit juice administration before sampling of some individuals, in response to ambient heat and observed panting. Conversely, lower values might be attributable to artifact from delayed sample analysis (manufacturer instructions recommend within 10 min for CHEM8), or a fasting period (Widmaier and Kunz 1993; Day et al. 2001; Heard et al. 2006).

It is noteworthy that not all sample analyses yielded viable results. Manufacturer guidelines recommend cartridge storage at 2–8 C and analyzer operation at 16–30 C and below 90% humidity; however, ambient field temperatures reached 37.8 C. Unfortunately, temperatures within the cooler were not recorded, and it is uncertain how heat may have affected the results.

Inhalant anesthesia and short-term physical restraint have been associated with measurable hematologic and biochemical changes in Old World fruit bats (Heard and Huft 1998). In our study, we decided that the benefits of quick restraint and short handling times outweighed anesthetic risks. Potential consequences include hyperthermia and stress, which surely affected these results and impede interpretation. Additionally, some bats were observed panting, which can mask acid–base disturbances. Interpretation is also complicated by venous sampling. This precludes interpretation of oxygenation status but it is a more feasible option for wildlife due to potential hazards of arterial sampling in field conditions.

Despite logistic challenges, field blood gas and biochemical analysis may still have value; physiologic data collection can contribute to projects linking physiologic states with disease status. When considered with physical examination and hematocrit, inferences may be made regarding renal health and function and hydration status from BUN, creatinine, and metabolic acid–base disturbances. Additionally, electrolyte derangements can be corrected in the field with selection of appropriate fluids.

This project was supported by Smithsonian Women's Committee (SWC grant 40); the Morris Animal Foundation (MAF) and Dennis and Connie Keller through a training partnership; and the Judy and John W. McCarter Global Health Internship.