Purpose: To determine whether lowering oxygen saturation alarm limits for infants at risk for retinopathy of prematurity (ROP) reduces its incidence and/or severity.
Methods: Oximetry alarm limits were lowered to 85% and 93% for all infants with a birth weight 1250 g or less and/or gestational age 28 weeks or less, and maintained until 32 weeks' postmenstrual age or until oxygen saturations were consistently greater than 93% in room air. The new policy was effective for infants born on or after June 1, 2003. ROP data were prospectively collected, and we compared the rate and severity of ROP in the year after the oximeter alarm policy change to the rates in the immediately preceding 3 years.
Results: In the year after the oximeter alarm limit policy change, 4 of 72 infants developed prethreshold ROP compared with 44 of 251 infants in the previous 3-year epoch (17.5% vs 5.6%, p=0.01). Similarly, only 6 of 144 eyes developed prethreshold ROP in the year after the policy change, compared with 84 of 502 in the previous 3 years (16.7% vs 4.2%, p=0.001).
Conclusions: A simple change in oximeter alarm parameters in the first weeks of life for infants with a birth weight 1250 g or less may decrease the incidence of prethreshold ROP.