Abstract

This study evaluated the accuracy of pulse oximetry, capnography, and oscillometric blood pressure during general anesthesia in giraffes (Giraffa camelopardalis). Thirty-two giraffes anesthetized for physiologic experiments were monitored using a pulse oximeter probe on the tongue and a capnograph sampling line at the distal end of the endotracheal tube. A blood pressure cuff was placed around the base of the tail, and an indwelling arterial catheter was placed in the auricular artery for blood gas analysis and invasive blood pressure measurements. Giraffes were intermittently ventilated using a Hudson demand valve throughout the procedures. Relationships between oxygen saturation as determined by pulse oximetry (SpO₂) and arterial oxygen saturation (SaO₂), between arterial carbon dioxide partial pressure (PaCO₂) and end-tidal carbon dioxide (P(et)CO₂), and between oscillometric pressure and invasive arterial pressure were assessed, and the accuracy of pulse oximetry, capnography, and oscillometric blood pressure monitoring evaluated using Bland-Altman analysis. All three noninvasive methods provided relatively poor estimates of the reference values. As a novel approach, receiver operating characteristics (ROC) curve fitting was used to determine cut-off values for hypoxia, hypocapnia, hypercapnia, and hypotension for dichotomous decision-making. Applying these cut-off values provided reasonable sensitivity for detection of hypocapnia, hypercapnia, and hypotension, but not for hypoxemia. Noninvasive techniques currently available are not accurate in the anesthetized giraffe and may produce “normal” readings in the face of serious abnormalities. As a consequence, noninvasive anesthetic monitoring results should be interpreted with great caution in giraffes and, ideally, invasive monitoring should be employed.