This prospective study was done to evaluate the accuracy of pulse oximetry as an indicator of hypoxemia in sickle cell anemia patients presenting to the ED in acute vasoocclusive crisis. Thirteen patients (age older than 18 years) presenting in crisis, with pulse oximetry readings less than 95% had arterial blood gas (ABG) analysis performed. The mean pulse oximetry reading was 91.4% +/- 3.2% (SD); the mean ABG oxygen saturation (ABG O2 sat) was 95.3% +/- 1.4% (SD). The mean O2 sat difference (ABG O2 sat minus pulse oximetry) was 4.5% (P < .001 versus a control group of ED patients receiving ABG and pulse oximetry). Pulse oximetry correlated with ABG O2 sat (r = 0.889, P < .001) and O2 sat difference (r = -0.927, P < .001). No correlation existed between complete blood count, reticulocyte count, historical and physical examination findings, and O2 sat or O2 sat difference. By linear regression, ABG O2 sat = 0.44 (pulse oximetry) + 55.6. This study suggests that pulse oximetry underestimates oxygenation in sickle cell anemia patients presenting in painful crisis.