Temporal Profile of Retinopathy of Prematurity in Extremely Premature Compared to Premature Infants.

Purpose: To compare the time course of retinopathy of prematurity (ROP) in premature (>= 28 to 34 weeks' gestational age) and extremely premature (< 28 weeks' gestational age) infants. Methods: This study was a retrospective single-center chart review that included 582 eyes (122 premature, 464 extremely premature) of 299 infants (65 premature, 234 extremely premature) diagnosed as having ROP who were born between January 1, 2010, and December 31, 2015. Data collected were analyzed with shared frailty models, chi-square, and Fisher's exact tests. Results: Extremely premature infants were diagnosed as having ROP earlier than premature infants at 33.6 weeks (95% CI [confidence interval]: 33.4 to 33.9) versus 36.0 weeks (95% CI: 34.7 to 36.4, P < .0001). The time from diagnosis to spontaneous regression was more than 4 weeks longer for extremely premature infants than premature infants with stage 1 ROP (82.0 vs 50.0 days, P < .0001), and more than 6 weeks longer for extremely premature infants than premature infants with stage 2 ROP (99 vs 55 days, P < .0001). Extremely premature infants had more bilateral ROP (96.6% vs 87.1%, P < .01) and were three times more likely to require treatment than premature infants (29.7% vs 9.9%, P < .0001). On average, infants who required treatment weighed 137 grams less at birth (P < .0001) and gained 37.7 grams less per week than infants who underwent spontaneous regression (P < .0001). Conclusions: Extremely premature infants developed ROP earlier, had more bilateral ROP, took longer to achieve spontaneous regression, and required treatment more often than premature infants. Infants who required treatment had a lower birth weight and rate of weight gain than infants who underwent spontaneous regression.