Effect of decreasing target oxygen saturation on retinopathy of prematurity

Abstract

The authors of previous studies suggest that the oxygenation status of premature infants contributes to the development of retinopathy of prematurity (ROP). In this study we compared the incidence and severity of ROP before and after institution of a new neonatal intensive care unit oxygen protocol. A retrospective chart review was performed of all eligible inborn patients screened for ROP during the 2 years immediately before (Group 1) to and the 2 years after (Group 2) the initiation of a new oxygen protocol. In the new protocol, target oxygen saturation was adjusted from 90%-99% to 85%-93%. Treatment criteria adhered to Early Treatment for Retinopathy of Prematurity guidelines for the study's duration. There were 387 infants in Group 1 and 386 infants in Group 2 (descriptive statistics adjusted for correlation due to multiple births). Mean birth weights (BWs) and gestational ages were 1,194 g and 29.2 weeks (ranges, 525-2,085 g; 23 2/7-39 6/7 weeks) for Group 1 and 1,139 g and 28.9 weeks (ranges, 520-2,500 g; 22 6/7-35 3/7 weeks) for Group 2 (p = 0.02/0.10). ROP developed in 32.7% of infants in Group 1 and 27.8% in Group 2 (p = 0.17). The incidence of ROP requiring treatment was 19.9% in Group 1 and 20.5% in Group 2 (p = 0.91). Subanalysis of infants with BW ≤ 1,000g (Group 1, n = 119; Group 2, n = 141) revealed ROP incidence of 75.1% versus 57.1%, respectively (p < 0.01); treatable disease occurred in 37.5% and 21.9% of affected infants (p = 0.19). Lowering target oxygen saturation for inborn premature infants was associated with decreased incidence of ROP only in infants with BW ≤ 1,000 g. Severity of disease, including need for treatment, was similar in both groups.