Severe retinopathy of prematurity and its association with different rates of survival in infants of less than 1251 g birth weight

Abstract

Background: There is controversy over whether improved survival of preterm infants has resulted in a higher incidence of severe (grade 3 or greater) retinopathy of prematurity (ROP).

Aim: To compare survival rates and rates of > or = stage 3 ROP-that is, with a high risk of sequelae-in preterm infants in five English cities where, anecdotally, the incidence of ROP is reported to show considerable variation.

Methods: All infants of birth weight < 1500 g and or gestational age < 32 weeks, born in 1994 in one of the cities or transferred in within 48 hours, were studied. The populations were adjusted for case mix variation using CRIB (clinical risk index for babies, a disease severity scoring system). The incidence of severe ROP, the actual death rate, and that adjusted for disease severity were determined.

Results: The rate of severe ROP per 1000 births was higher in city 1 than in all the other cities. This increase in comparison with city 2 and city 4 was significant (city 1, 167 (95% confidence interval (CI) 96 to 260); city 2, 24 (6 to 59); city 4, 16 (1 to 84)). A significant difference was not seen between city 1 and cities 3 (23 (1 to 120)) and 5 (74 (21 to 79)). The relative risk of developing severe ROP in city 1 compared with all the other cities was 5.5 (2.5 to 11.9). The actual death rate per 1000 births in city 1 was significantly lower than that predicted by modelling death against CRIB score (city 1: actual 270; predicted 385 (95% CI 339 to 431)). In contrast, the other cities had actual death rates as predicted, or worse than predicted, by CRIB.

Interpretation: A significantly higher incidence of severe ROP was identified in one of the five cities studied. Variation in survival rates among high risk infants may explain this observation.