Objective: To investigate the effects on oxygenation of targeting the higher versus the lower end of the currently recommended range for pulse oximetry (Spo2).
Design: Nonrandomized crossover trial with the use of within-subject comparisons (two-tailed paired t test).
Setting: Level III neonatal intensive care unit of a university hospital.
Patients: Twenty infants whose lungs were mechanically ventilated (mean +/- SD: birth weight, 1192 +/- 396 gm; gestational age, 28.7 +/- 2.7 weeks; age at time of study, 42 +/- 26 hours).
Interventions: The inspired oxygen concentration was adjusted to achieve Spo2 readings of 93% to 96% versus 89% to 92% (Ohmeda pulse oximeter) or 95% to 98% versus 91% to 94% (Nellcor oximeter).
Measurements: Cardiac output was measured by echocardiography, oxygen content of arterial blood samples by cooximetry, and oxygen consumption by indirect calorimetry.
Results: The inspired oxygen concentrations required to achieve the Spo2 target ranges were 39.8% +/- 8.3% versus 28.7% +/- 6.1% (p < 0.001). The respective arterial oxygen contents were 18.0 +/- 2.6 ml/dl versus 16.9 +/- 2.5 ml/dl (p < 0.001). Oxygen consumption was unchanged. In the lower-oxygen condition no compensatory increase in cardiac output was detected; thus the estimated mixed venous oxygen tension decreased and the oxygen extraction ratio increased. Venous admixture increased from 15% +/- 6% to 31% +/- 9% in the lower-oxygen condition (p < 0.001).
Conclusions: The "low normal" Spo2 target range allowed for less oxygen exposure. No signs of mismatch between systemic oxygen delivery and demand could be detected.