Agreement between arterial partial pressure of carbon dioxide and saturation of hemoglobin with oxygen values obtained by direct arterial blood measurements versus noninvasive methods in conscious healthy and ill foals

Abstract

To determine agreement between indirect measurements of end-tidal partial pressure of carbon dioxide (PetCO(2)) and saturation of hemoglobin with oxygen as measured by pulse oximetry (SpO(2)) with direct measurements of PaCO(2) and calculated saturation of hemoglobin with oxygen in arterial blood (SaO(2)) in conscious healthy and ill foals. Validation study. 10 healthy and 21 ill neonatal foals. Arterial blood gas analysis was performed on healthy and ill foals examined at a veterinary teaching hospital to determine direct measurements of PaCO(2) and PaO(2) along with SaO(2). Concurrently, PetCO(2) was measured with a capnograph inserted into a naris, and SpO(2) was measured with a reflectance probe placed at the base of the tail. Paired values were compared by use of Pearson correlation coefficients, and level of agreement was assessed with the Bland-Altman method. Mean ± SD difference between PaCO(2) and PetCO(2) was 0.1 ± 5.0 mm Hg. There was significant strong correlation (r = 0.779) and good agreement between PaCO(2) and PetCO(2). Mean ± SD difference between SaO(2) and SpO(2) was 2.5 ± 3.5%. There was significant moderate correlation (r = 0.499) and acceptable agreement between SaO(2) and SpO(2). Both PetCO(2) obtained by use of nasal capnography and SpO(2) obtained with a reflectance probe are clinically applicable and accurate indirect methods of estimating and monitoring PaCO(2) and SaO(2) in neonatal foals. Indirect methods should not replace periodic direct measurement of corresponding parameters.