Erythropoietin and retinopathy of prematurity: a meta-analysis

We performed a meta-analysis to study the association between erythropoietin (EPO) and the development of retinopathy of prematurity (ROP) in preterm newborn infants. Studies were identified through PubMed (1966-) and ISI databases (1965-) literature searches. Results and effect sizes are expressed as odds ratio (OR) with 95 % confidence intervals (CI). Fourteen studies identified to the meta-analysis, including 3,484 preterm newborn infants. A total of 563 of 1,221 babies treated with EPO had ROP (46.1 %) vs. 420 of 1,134 babies without EPO (37.0 %). No significant difference was found in the ROP risk between the two groups, with the OR 1.592 (95 % CI 0.901-2.812). A total of 192 of 1,298 babies treated with EPO had severe ROP (stage 3-4) (14.8 %) vs. 166 of 1,199 babies without EPO (13.8 %). The OR was 1.203 (95 % CI 0.763-1.896). No significant publication bias was found. Sensitivity analyses showed the results were robust.

Our meta-analysis indicates that EPO treatment is not associated with the development of ROP in preterm infants. But this conclusion should be confirmed by further high-quality researches.