Background: The optimal arterial oxygen saturation in the first weeks of life is unknown for immature newborn infants. Objectives: To determine the effect of targeting high versus low oxygen saturation in the first weeks of life on the outcome of very low and extremely low birth weight infants. Methods: Randomized and observational studies were sought that compared the outcomes in babies with high or low oxygen saturation targeting assessed by pulse oximetry. Results: Ten studies were identified, of which 8 had severe retinopathy of prematurity (n = 3,811) and 8 had bronchopulmonary dysplasia/lung problems (n = 4,612) as outcomes. Two studies also provided survival data. The relative risk (RR) in favor of low SpO2 was 0.42 (95% CI 0.34–0.51) for severe retinopathy of prematurity, 0.73 (95% CI 0.63–0.86) for bronchopulmonary dysplasia/lung problems, and 1.12 (95% CI 0.86–1.45) for mortality. There was 1 randomized trial with retinopathy of prematurity, 3 with bronchopulmonary dysplasia/lung problems, and 1 with mortality as the outcome. When analyzing the randomized trial separately, the RR (95% CI) for severe retinopathy of prematurity was 0.48 (0.34–0.68), for bronchopulmonary dyslasia/lung problems it was 0.79 (0.64–0.97), and for mortality it was 1.27(1.01–1.60). Conclusions: A low oxygen saturation approach reduces severe retinopathy of prematurity by 50%, i.e. from 20.9 to 9.5%, and bronchopulmonary dysplasia/lung problems by 25%, i.e. from 40.8 to 29.7%. Further randomized trials are needed to provide definite conclusions and to assess whether reducing oxygen saturation has an impact on mortality among very and extremely low birth weight infants.