Oxygen saturation levels and retinopathy of prematurity in extremely preterm infants - a case control study

Purpose To investigate the association of risk factors, including oxygen exposure, for developing retinopathy of prematurity (ROP) in preterm infants at increased risk of ROP. Methods A case-control study was conducted where each infant born at < 28 weeks gestation with ROP was matched with another without ROP over five years (July 2015 - June 2020). Clinical information about the infants was collected from electronic medical records, including method of oxygen delivery, oxygen saturation (SpO(2)), fraction of inspired oxygen (FiO(2)) and mean airway pressure (MAP) measurements. MATLAB was used for a time-averaged analysis. Stata/SE 16.0 was used for statistical analysis. Results 123 ROP/non-ROP pairs were included in this study. The time-averaged SpO(2) analysis showed non-ROP group spent more time in hyperoxia than the ROP group (p < 0.001). The non-ROP group had lower respiratory severity scores and analysis when FiO(2) > 21% showed that were was no difference in SpO(2) between the two groups when the infants were receiving oxygen support. Conditional logistic regressions showed neonatal surgery significantly increased the risk of ROP (OR = 1.4347, p = 0.010), while the influence of birthweight (odds ratio of 0.9965, p = 0.001) and oxygen exposure (OR = 0.9983, p = 0.012) on ROP outcome was found to be negligible as their odds ratios indicated no influence. Conclusions At times when infants were receiving respiratory support (FiO(2) > 21%) the SpO(2) data indicated no difference in SpO(2) between the ROP and non-ROP groups. Analysis of clinical variables found that neonatal surgery increased the odds of developing ROP.