Determination of hemoglobin saturation in patients with acute sickle chest syndrome: a comparison of arterial blood gases and pulse oximetry.

Abstract

STUDY OBJECTIVES: To evaluate three different methods of measuring oxygen saturation in patients suffering from acute sickle chest syndrome. DESIGN: A prospective, descriptive study of 9 months' duration. SETTING: A tertiary care university hospital. PATIENTS: Adult patients with acute sickle chest syndrome scheduled to undergo RBC exchange transfusion. INTERVENTIONS: None. MEASUREMENTS: Baseline hemoglobin oxygen saturation was determined simultaneously by (1) calculation based on PaO2 and an oxyhemoglobin dissociation curve algorithm, (2) co-oximetry, and (3) pulse oximetry. These same measures were repeated after exchange transfusion. Baseline and postexchange hemoglobin electrophoresis was performed in all patients. RESULTS: Baseline calculated saturation overestimated true saturation (determined by co-oximetry) with a baseline mean bias (co-oximetry minus calculated saturation) of -6.78 +/- 2.63% (95% confidence interval for bias: -8.37% to -5.19%). Pulse oximetry was not different than co-oximetry at baseline with a baseline bias of +1.86 +/- 3.25% (95% confidence interval: -0.1% to 3.82%). After exchange transfusion, there was no bias between either co-oximetry and calculated saturation (mean difference: -0.17 +/- 1.31% [95% confidence interval: -0.95% to 0.61%]), or co-oximetry and pulse oximetry (mean difference: +0.3 +/- 1.53% [95% confidence interval: -0.62% to 1.22%]). CONCLUSIONS: Calculated saturation overestimates true saturation during acute sickle chest syndrome. This discrepancy abates after exchange transfusion. Pulse oximetry more closely follows co-oximetry than does calculated saturation during acute sickle chest syndrome.