Comparison of Mercury, Digital Rectal, and Non-Contact Infrared Thermometry for Body Temperature Measurement in Dogs

Objectives: In canine veterinary medicine, determining body temperature is essential for clinical diagnosis, patient monitoring, and evaluating animal welfare. Although digital rectal and non-contact infrared (IR) thermometers are frequently used as alternatives to traditional mercury-in-glass thermometers, their clinical agreement remains controversial. This study aimed to evaluate the agreement of digital rectal and non-contact IR thermometers with mercury-in-glass rectal thermometry in dogs and to explore the relationships between measurement methods using regression analysis.Materials and Methods: A total of 151 client-owned dogs (63 males and 88 females), including both clinically healthy dogs and dogs with various health conditions, were included in this comparative study. The animals were not stratified according to their clinical health status. Body temperature was measured sequentially using a mercury-in-glass rectal thermometer (reference method; 3-minute measurement duration), a digital rectal thermometer, and a non-contact IR thermometer positioned 5 cm from the hairless inguinal region. Data were evaluated using linear mixed-effects models, Pearson's correlation, linear regression, and Bland–Altman agreement analysis.Results: Mean body temperatures were 38.71 ± 0.05 °C for mercury-in-glass, 38.58 ± 0.05 °C for digital rectal, and 37.94 ± 0.05 °C for IR thermometers (p < 0.0001). A strong positive correlation was observed between digital rectal and mercury-in-glass measurements (r = 0.84; p < 0.0001; 95% CI: 0.79–0.89), whereas a weak positive correlation was found between IR and mercury-in-glass measurements (r = 0.35; p < 0.0001; 95% CI: 0.20–0.48). Bland–Altman analysis showed a mean difference of 0.13 °C (95% limits of agreement: −0.51 to 0.77 °C) between mercury-in-glass and digital rectal thermometers, compared with 0.77 °C (95% limits of agreement: −0.88 to 2.42 °C) between mercury-in-glass and non-contact IR thermometers.Conclusions: Digital rectal thermometry demonstrated a stronger correlation with mercury-in-glass rectal thermometry than non-contact IR thermometry. However, clinical interchangeability should be interpreted cautiously and evaluated in relation to the observed limits of agreement. Non-contact IR measurements obtained from the inguinal region showed greater variability, suggesting that this method should not be considered a direct substitute for rectal temperature measurement under the conditions investigated.