(C28) Agreement of Resting Respiratory Rate Measurements and Those Obtained by the Maven Collar in Dogs

Abstract: Background – Resting respiratory rate (RRR) monitoring is essential in dogs with advanced cardiac disease to monitor for control of congestive heart failure. The Maven Pet collar system offers a feature measuring RRR in pets. Hypothesis/Objective – To describe the agreement between Maven Pet Smart Collar-obtained and manually counted RRR measurements in dogs. Animals – 18 apparently healthy, student and staff-owned dogs. Methods – Dogs were fitted for a Maven Pet Smart Collar. Owners were trained to video record their dogs at home to document the RRR for later analysis. Owners were asked to collect an aggregate of at least 28, one-minute videos. Videos were collected after the dogs were resting for at least 15 minutes. Videos were used for analysis if they were within one minute from a collar-obtained RRR. Bland-Altman analysis was used to assess agreement between Maven Pet Smart Collar-obtained RRR and RRR measured by the investigators from video recordings. Results – Data was collected from 18 dogs (n = 120 instances). Relative to manually counted RRR, bias for the Maven Pet Smart Collar system was -0.75 breaths/minute (95% limits of agreement -8.28 to 6.79 breaths/min). Conclusions and Clinical Importance – The Maven Pet Smart Collar system closely agrees with manually counted RRR, with relatively narrow limits of agreement. The Maven Pet Smart Collar could be considered for monitoring RRR in dogs, especially in dogs with poorly compliant owners because it is automated. Clinicians will need to consider the limits of agreement in interpretation.