A Multicenter Study Analyzing Weight Gain to Predict Retinopathy of Prematurity

Abstract

 

Purpose:

 

Retinopathy of prematurity (ROP) is a leading cause of preventable childhood blindness. However, <8% of infants examined by current criteria require treatment and diagnosis only occurs when ROP is established. We need to differentiate early those infants at higher risk for severe ROP. WINROP (Weight, IGF, Neonatal ROP) is a free online surveillance algorithm using longitudinal postnatal weight measurements to better predict early, infants’ risk for developing severe ROP. The purpose of this study was to validate WINROP as a tool to help predict ROP in geographically and racially diverse populations of premature infants in 10 neonatal intensive care units in the U.S. and Canada.

 

Methods:

 

WINROP analysis was performed on 1706 premature infants selected by conventional criteria for ROP screening, with median gestational age 28 weeks at birth and median birth weight of 1016 g. ROP exams consisted of standard binocular indirect ophthalmoscopy with scleral depression and ROP was classified according to international standards. Weekly postnatal weight measurements were entered into WINROP, which calculated deviations between observed and reference values of expected weight gain velocity in infants without ROP. An alarm signaled when the rate of weight gain decreased compared with healthy GA matched control subjects, indicating high risk of ROP.

 

Results:

 

An alarm was signaled in 1101 infants: 78 alarmed at week 0 and 1023 alarmed for poor postnatal weight gain at a median time of 3 weeks. Type 1 ROP developed in 144 of these infants. Conversely, no alarm signaled in 605 infants. Of these, 603 did not develop type 1 ROP. The overall sensitivity of WINROP to detect type 1 ROP was 98.6% (144/146 infants) and the specificity was 38.7% (603/1560 infants). The positive predictive value was 13.1% (144/1101 infants) and the negative predictive value was 99.7% (603/605 infants).

 

Conclusions:

 

This multicenter study validated the use of postnatal weight gain analysis by WINROP to help predict very early, infants at high risk for type 1 ROP. With early and sensitive detection of those at higher risk and 99.7% detection of those at no or low risk, WINROP is a valuable tool to optimize patient ROP screening and care.

 

Keywords: clinical (human) or epidemiologic studies: risk factor assessment • retinopathy of prematurity • clinical (human) or epidemiologic studies: health care delivery/economics/manpower