Poor agreement between haemoglobin oxygen saturation measured by pulse oximetry and arterialized earlobe blood gas in ambulatory paediatric sickle cell patients

Background: Evidence suggests poor agreement between pulse oximetry SpO 2 and arterial blood gas (ABG) SaO 2 in measuring Hb O 2 saturation in sickle cell disease (SCD) (Blaisdell, Arch Pediatr Adoles Med 2000). Earlobe blood gas co-oximetry (EBG) is an easily performed painless method of measuring SaO 2 , with good agreement with ABG (Zavorsky, Respir Physiol Neurobiol 2007). Aims: To investigate the agreement between SpO 2 and SaO 2 measured by EBG in SCD children. Methods: We reviewed records of SCD patients attending a regional paediatric respiratory clinic, and calculated agreement between paired EBG and SpO 2 measurements. Results: We identified 39 paired SaO 2 and SpO 2 readings from 33 patients (52% male, median age 10y). In 9 cases (23%), difference was u003e±2%. SaO 2 was lower than SpO 2 in one-third and greater in two-thirds. Of 17 hypoxaemic measurements (SpO 2 2 2 92.1%, P=0.02). Figure 1: Bland-Altman plot showing mean EBG and SpO2 O 2 saturations and differences. Unbroken line: mean difference (-0.7%). Broken line: agreement limits (-5.6 to +4.3%). Shaded region: accepted error range (±2%) Conclusion: This study suggests that SpO 2 as a measure of Hb O 2 saturation is inaccurate in a proportion of ambulatory SCD children. EBG should be used more widely to accurately detect changes in arterial Hb O 2 saturation in SCD children.