Search

Chronic wasting disease (CWD) is a highly infectious, fatal prion disease that affects cervid species. One promising method for CWD surveillance is the use of detection dog–handler teams wherein dogs are trained on the volatile organic compound signature of CWD fecal matter. However, using fecal matter from CWD-positive deer poses a biohazard risk; CWD prions can bind to soil particles and remain infectious in contaminated areas for extended periods of time, and it is very difficult to decontaminate the affected areas. One solution is to use noninfectious training aids that can replicate the odor of fecal matter from CWD-positive and CWD-negative deer and are safe to use in the environment. Trained CWD detection dogs’ sensitivity and specificity for different training aid materials (cotton, GetXent tubes, and polydimethylsiloxane, or PDMS) incubated with fecal matter from CWD-positive and CWD-negative deer at two different temperatures (21 °C and 37 °C) for three different lengths of time (6 h, 24 h, and 48 h) were evaluated. Cotton incubated at 21 °C for 24 h was identified as the best aid for CWD based on the dogs’ performance and practical needs for training aid creation. Implications for CWD detection training and for training aid selection in general are discussed.

Working dogs are at a high risk of canine non-pyrogenic hyperthermia, a life-threatening condition that can occur due to physical exertion or environmental factors that inhibit dogs’ ability to cool themselves. Two frequently recommended cooling methods to reduce body temperature are water immersion and the application of isopropyl alcohol to paw pads. This cross-over study compared the relative efficacy of these methods in 12 working-dogs-in-training with post-exertional heat stress. On each study day, dogs had a physical exam and performed a warm-up exercise followed by sequential recalls in which dogs ran approximately 25 m between two designated handlers for 10 min until they showed multiple signs of heat stress or their core temperature reached 105 °F (40.6 °C). Dogs’ temperature and heart rate were collected after each recall. Dogs completed three study days, and each day, randomly received one of three interventions: passive cooling (no intervention), partial water immersion, or isopropyl alcohol. Post-intervention dogs rested for 20 min. Partial water immersion and isopropyl alcohol both cooled dogs more than no intervention, and water immersion cooled dogs more efficiently than isopropyl alcohol. Additionally, the application of isopropyl alcohol raised dogs’ heart rates more than water immersion or no intervention, suggesting that the process of applying isopropyl alcohol is potentially stressful to dogs. Thus, partial water immersion is preferred to cool dogs post-exertion due to its more efficient cooling and better tolerance of use.

The spread of SARS-CoV-2, which causes the disease COVID-19, is difficult to control as some positive individuals, capable of transmitting the disease, can be asymptomatic. Thus, it remains critical to generate noninvasive, inexpensive COVID-19 screening systems. Two such methods include detection canines and analytical instrumentation, both of which detect volatile organic compounds associated with SARS-CoV-2. In this study, the performance of trained detection dogs is compared to a noninvasive headspace-solid phase microextraction-gas chromatography-mass spectrometry (HS-SPME-GC-MS) approach to identifying COVID-19 positive individuals. Five dogs were trained to detect the odor signature associated with COVID-19. They varied in performance, with the two highest-performing dogs averaging 88% sensitivity and 95% specificity over five double-blind tests. The three lowest-performing dogs averaged 46% sensitivity and 87% specificity. The optimized linear discriminant analysis (LDA) model, developed using HS-SPME-GC-MS, displayed a 100% true positive rate and a 100% true negative rate using leave-one-out cross-validation. However, the non-optimized LDA model displayed difficulty in categorizing animal hair-contaminated samples, while animal hair did not impact the dogs’ performance. In conclusion, the HS-SPME-GC-MS approach for noninvasive COVID-19 detection more accurately discriminated between COVID-19 positive and COVID-19 negative samples; however, dogs performed better than the computational model when non-ideal samples were presented.