Abstract
This study examines the accuracy of oxygen saturation measured by Nellcor N200 pulse oximetry (SpO2) compared with arterial oxygen saturation (SaO2) measured through a three-wavelength fiberoptic umbilical catheter in 10 neonates who needed mechanical ventilation. Real-time SaO2 was validated with a reference method every 4 hours. Oxygen saturation readings (SaO2 and SpO2), along with pulse rate and pulsation, were recorded continuously every second through a computer. Concurrent care events and neonatal responses were recorded. Data were completed on 10 neonates who had an umbilical arterial catheter. Desaturation events (<90%) as measured by both SaO2 and SpO2 were described and compared. A total of 959 desaturation events occurred during an average of 51 hours of monitoring per subject. Of these events, 63% were associated with frozen SpO2 readings, and 18% of frozen readings occurred when SaO2 was <90%. Bias for SpO2 compared with SaO2 was +5.03%, with 5.6% of the readings outside the range of two standard deviations. However, 67% of the readings exceeded the 4% difference criterion between measurements. Future studies need to examine the desaturation events in relation to oxygenation status as measured by different methods.