LAUSR

OBJECTIVE Evaluation of the reliability of pulse oximetry at four different attachment sites compared to haemoglobin oxygen saturation measured by a co-oximeter and calculated by a blood gas analyser in immobilized impala. STUDY DESIGN Randomized crossover study. ANIMALS A total of 16 female impala. METHODS Impala were immobilized with etorphine or thiafentanil alone, or etorphine in combination with a novel drug. Once immobilized, arterial blood samples were collected at 5 minute intervals for 30 minutes. Then oxygen was insufflated (5 L minute-1) intranasally at 40 minutes and additional samples were collected. A blood gas analyser was used to measure the arterial partial pressure of oxygen and calculate the oxygen haemoglobin saturation (cSaO2); a co-oximeter was used to measure the oxygen haemoglobin saturation (SaO2) in arterial blood. Pulse oximeter probes were attached: under the tail, to the pinna (ear) and buccal mucosa (cheek) and inside the rectum. Pulse oximeter readings [peripheral oxygen haemoglobin saturation (SpO2) and pulse quality] were recorded at each site and compared with SaO2 and cSaO2 using Bland-Altman and accuracy of the area root mean squares (Arms) methods to determine the efficacy. P value < 0.05 was considered significant. RESULTS Pulse quality was 'good' at each attachment site. SpO2 measured under the tail was accurate and precise but only when SaO2 values were above 90% (bias = 3, precision = 3, Arms = 4). The ear, cheek and rectal probes failed to give accurate or precise readings (ear: bias = -4, precision = 14, Arms = 15; cheek: bias = 12, precision = 11, Arms = 16; and rectum: bias = 5, precision = 12, Arms = 13). CONCLUSIONS AND CLINICAL RELEVANCE In order to obtain accurate and precise pulse oximetry readings in immobilized impala, probes must be placed under the tail and SaO2 must be above 90%. Since SaO2 values are usually low in immobilized impala, pulse oximeter readings should be interpreted with caution.