Assessment of pulse co-oximetry technology after in vivo adjustment in anaesthetized dogs

To compare values of haemoglobin concentration (SpHb), arterial haemoglobin saturation (SpO2) and calculated arterial oxygen content (SpOC), measured noninvasively with a pulse co-oximeter before and after in vivo adjustment (via calibration of the device using a measured haemoglobin concentration) with those measured invasively using a spectrophotometric-based blood gas analyser in anaesthetized dogs. Prospective observational clinical study. A group of 39 adult dogs. In all dogs after standard instrumentation, the dorsal metatarsal artery was catheterised for blood sampling, and a pulse co-oximeter probe was applied to the tongue for noninvasive measurements. Paired data for SpHb, SpO2 and SpOC from the pulse co-oximeter and haemoglobin arterial oxygen saturation (SaO2) and arterial oxygen content (CaO2) from the blood gas analyser were obtained before and after in vivo adjustment. Bland–Altman analysis for repeated measurements was used to evaluate the bias, precision and agreement between the pulse co-oximeter and the blood gas analyser. Data are presented as mean differences and 95% limits of agreement (LoA). A total of 39 data pairs were obtained before in vivo adjustment. The mean invasively measured haemoglobin–SpHb difference was –2.7 g dL−1 with LoA of –4.9 to –0.5 g dL−1. After in vivo adjustment, 104 data pairs were obtained. The mean invasively measured haemoglobin–SpHb difference was –0.2 g dL−1 with LoA of –1.1 to 0.6 g dL−1. The mean SaO2–SpO2 difference was 0.86% with LoA of –0.8% to 2.5% and that between CaO2–SpOC was 0.66 mL dL–1 with LoA of –2.59 to 3.91 mL dL–1. Before in vivo adjustment, pulse co-oximeter derived values overestimated the spectrophotometric-based blood gas analyser haemoglobin and CaO2 values. After in vivo adjustment, the accuracy, precision and LoA markedly improved. Therefore, in vivo adjustment is recommended when using this device to monitor SpHb in anaesthetised dogs.