Retinopathy of prematurity in practice. I: screening for threshold disease

Eye (Lond). 2003 Mar;17(2):183-8. doi: 10.1038/sj.eye.6700296.


Aims: To review an 11-year period of screening for retinopathy of prematurity (ROP) in the north of England by a single ophthalmologist. To identify the gestational ages and birth weights of babies reaching different stages of ROP. To investigate the workload involved in screening to detect threshold ROP, and that the practical outcomes had narrower inclusion criteria for screening. To identify babies treated for threshold disease.

Methods: During the period August 1987-October 1998, babies were screened according to the national guidelines and the results were prospectively entered onto a computerised database. These data were then systematically reviewed.

Results: Data were available for 484 babies: 203 (41.9%) developed any ROP, 46 (9.5%) reached stage 3 ROP, and 25 (5.2%) reached threshold and were treated. Data on 425 babies showed them to require an average of 2.3 screening examinations per baby. It took an average of 39 screening examinations to detect one case of threshold ROP. The more premature and lighter birth weight babies required the most examinations. Therefore, restricting the inclusion criteria for screening would only have reduced the total number of screenings modestly and could have allowed us to miss two of our threshold cases who were both of 30 weeks gestational age and >1400 g birth weight.

Conclusions: Screening is time consuming but worthwhile in view of the benefits of treatment. As applied to babies in the north of England, the current national screening criteria are satisfactory. The results of treatment of the babies identified in this study are presented in the accompanying paper.

MeSH terms

  • Birth Weight
  • England
  • Gestational Age
  • Humans
  • Infant, Newborn
  • Infant, Premature
  • Neonatal Screening / methods*
  • Neonatal Screening / standards
  • Ophthalmology
  • Prospective Studies
  • Retinopathy of Prematurity / diagnosis*
  • Retrospective Studies
  • Workload