Bias, sensitivity, and specificity of Doppler ultrasonic flow detector measurement of blood pressure for detecting and monitoring hypotension in anesthetized dogs.

OBJECTIVE To determine the bias, sensitivity, and specificity of Doppler ultrasonic flow detector measurement of blood pressure (DBPM) to detect hypotension in dogs with various disease states and to determine whether patient characteristics could affect accuracy of DBPM in dogs. DESIGN Prospective cross-sectional study. ANIMALS 146 client-owned dogs undergoing general anesthesia at a veterinary teaching hospital between April 2007 and August 2010. PROCEDURES Data collected for each dog were breed, limb conformation, sex, American Society of Anesthesiologists physical status classification, anesthetic protocol, surgical procedure, arterial catheter size and location, and DBPM location. Doppler and invasive blood pressure measurements (IBPMs; criterion standard) were simultaneously recorded every 5 minutes throughout anesthesia. Hypotension was defined as mean arterial blood pressure < 60 mm Hg or DBPM < 90 mm Hg. Repeated-measures Bland-Altman analysis was performed to determine bias between DBPMs and IBPMs. Overall sensitivity and specificity of DBPM to detect hypotension were calculated with 2 methods, and values were recalculated for specific patient groups and compared. RESULTS Bias of DBPM was 2.8 mm Hg with wide 95% limits of agreement (-46.4 to 51.9 mm Hg). For the 2 calculation methods, sensitivity of DBPM to detect hypotension was 69.2% and 66.7% and specificity was 82.2% and 86.8%. No significant differences in sensitivity or specificity were identified regarding limb conformation, gauge of catheter (20 vs 22) used for IBPM, or side (ipsilateral or contralateral) of paired measurements. CONCLUSIONS AND CLINICAL RELEVANCE Results indicated that in dogs of the present study, DBPM was unreliable for detecting hypotension.