Comparison of oxygen saturation measured by pulse oximetry and arterial blood gas analysis in neonates

Abstract

Background Arterial blood gas is usually beneficial to discern thenature of gas exchange disturbances, the effectiveness of com-pensation, and is required for adequate management. AlthoughPaO 2 is the standard measurement of blood oxygenation, oxygensaturation measured by pulse oximetry (SapO 2 ) is now a custom-ary noninvasive assessment of blood oxygenation in newborn in-fants.Objective To compare oxygen saturation measured by pulse oxi-metry (SapO 2 ) and arterial blood gas (SaO 2 ), its correlation withother variables, and to predict arterial partial pressure of oxygen(PaO 2 ) based on SapO 2 values.Methods A cross sectional study was conducted on all neonatesadmitted to Pediatric Intensive Care Unit (PICU) during February2001 to May 2002. Neonates were excluded if they had impairedperipheral perfusion and/or congenital heart defects. Paired t-testwas used to compare SapO 2 with SaO 2 . Correlation between twoquantitative data was performed using Pearson’s correlation. Re-gression analysis was used to predict PaO 2 based on SapO 2 val-ues.Results Thirty neonates were included in this study. The differ-ence between SaO 2 and SapO 2 was significant . There were sig-nificant positive correlations between heart rate /pulse rate andTCO 2 , HCO 3 ; respiratory rate and TCO 2 , HCO 3 , base excess (BE);core temperature and HCO 3 , BE; surface temperature and pH,TCO 2, HCO 3, BE; SapO 2 and pH, PaO 2 ; and significant negativecorrelation between SapO 2 and PaCO 2 ; the correlations were weak.The linear regression equation to predict PaO 2 based on SapO 2values was PaO 2 = -79.828 + 1.912 SapO 2 .Conclusion Pulse oximetry could not be used in place of arterialblood gas analysis available for clinical purpose