Retinopathy of prematurity in 7 neonatal units in Rio de Janeiro: screening criteria and workload implications

• Published in: Pediatrics (Evanston) Vol. 126; no. 2; p. e410
• Main Authors: Zin, Andrea A, Moreira, Maria Elisabeth Lopes, Bunce, Catey, Darlow, Brian A, Gilbert, Clare E
• Format: Journal Article
• Language: English
• Published: United States 01.08.2010
• Subjects: Brazil - epidemiology; Catchment Area (Health); Humans; Infant, Newborn; Intensive Care Units, Neonatal - statistics & numerical data; Neonatal Screening - methods; Prevalence; Reference Values; Retinopathy of Prematurity - diagnosis; Retinopathy of Prematurity - epidemiology; Severity of Illness Index; Workload - statistics & numerical data; Brazil
• ISSN: 1098-4275
• DOI: 10.1542/peds.2010-0090

The goals were to determine optimal screening criteria for retinopathy of prematurity (ROP) in 7 neonatal units in Rio de Janeiro, Brazil, and to explore the workload implications of applying different criteria. Infants with birth weights of < or = 2000 g or gestational age of <37 weeks were examined by 3 ophthalmologists in 7 of the largest units in Rio de Janeiro, during a 34-month period. ROP was classified by using the international classification, and laser treatment was given to infants developing type 1 ROP. A total of 3437 (87%) of 3953 eligible infants were examined, of whom 124 (3.6% [range: 2.1%-7.8%]) were treated. Eleven infants were treated for aggressive posterior ROP. Appropriate screening criteria for the 2 NICUs with high survival rates (ie, > or =80% among infants with birth weights of <1500 g) would be < or =1500 g or <32 weeks. For NICUs with low survival rates (ie, <80%), appropriate criteria would be < or =1500 g or < or =35 weeks. UK, US, and previous Brazilian criteria would all miss infants needing treatment. ROP programs in Brazil should use the wider criteria of < or =1500 g or < or =35 weeks until further evidence-based criteria become available, although this would mean a slight increase in workload across the city, compared with use of the narrower criteria in the better units. Whether survival rates can be used as a proxy to indicate screening criteria requires further investigation.