Validation of a qualitative immunochromatographic test for the noninvasive assessment of stress in dogs

Salivary cortisol is regarded as a reliable parameter for the noninvasive assessment of the welfare of animals, because it is strictly related to stress levels. Several methods are available for salivary cortisol measurement in mammals, however rapid diagnostic test for detecting salivary cortisol are confined to humans. The availability of such non invasive diagnostic tools operable in situ would facilitate monitoring of animal welfare. The Cortisol stress test provides a simple and rapid tool to discriminate cortisol levels in canine saliva above or below 4 ng/ml, which has been suggested as the cut-off value for distinguishing unstressed dogs from those experiencing stress. The test is based on a competitive immunochromatographic assay (ICT) using gold nanoparticles as probes, in which the color intensity of the Test line is inversely correlated to the salivary cortisol level. The qualitative result is obtained by the visual observation of the color formed on the Test line compared to that of the Control line. We evaluated the accuracy of the test by determining salivary cortisol in 85 samples of canine saliva belonging to dogs with very variable age, sex, breed, and life history, and comparing the qualitative results to those obtained by a reference ELISA kit. Agreeing results were obtained through the two methods, and the ICT showed high diagnostic sensitivity, specificity and efficiency (100%, 98.4%, and 98.8%, respectively). Furthermore, we evaluated the precision of the test by an experimental design approach, which combines errors due to within-day and between-day variation with the biological variability, and demonstrated that the test could be reliably applied for correctly classifying canine samples, according to their salivary cortisol level. Moreover, we studied the shelf-life of the device in three experimental conditions. We confirmed the stability of the ICT at 4°C and 25°C for at least six months and observed similar results for an accelerated stability study conducted for 7 days at 37°C, which suggest that the stability of ICT device could be estimated by the accelerated experiment alternatively to the real-time study