Use of erythropoietin is associated with threshold retinopathy of prematurity (ROP) in preterm ELBW neonates: a retrospective, cohort study from two large tertiary NICUs in Italy

Early Hum Dev. 2014 Sep:90 Suppl 2:S29-33. doi: 10.1016/S0378-3782(14)50009-6.

Abstract

Background: Retinopathy of prematurity (ROP) is a multifactorial disease with evidence of many associated risk factors. Erythropoietin has been reported to be associated with this disorder in a murine model, as well as in humans in some single-center reports. We reviewed the data from two large tertiary NICUs in Italy to test the hypothesis that the use of erythropoietin may be associated with the development of the most severe stages of ROP in extremely low birth weight (ELBW) neonates.

Design/methods: Retrospective study by review of patient charts and eye examination index cards on infants with birth weight <1000g admitted to two large tertiary NICUs in Northern Italy (Sant'Anna Hospital NICU in Torino, and Ca' Foncello Hospital Neonatology in Treviso) in the years 2005 to 2007. Standard protocol of administration of EPO in the two NICUs consisted of 250 UI/kg three times a week for 6-week courses (4-week in 1001-1500g infants). Univariate analysis was performed to assess whether the use of EPO was associated with severe (threshold) ROP. A control, multivariate statistical analysis was performed by entering into a logistic regression model a number of neonatal and perinatal variables that - in univariate analysis - had been associated with threshold ROP.

Results: During the study period, 211 ELBW infants were born at the two facilities and survived till discharge. Complete data were obtained for 197 of them. Threshold retinopathy of prematurity occurred in 26.9% (29 of 108) of ELBW infants who received erythropoietin therapy, as compared with 13.5% (12 of 89) of those who did not receive erythropoietin (OR 2.35; 95% CI 1.121-4.949; p=0.02 in univariate analysis, and p=0.04 at multivariate logistic regression after controlling for the following variables: birth weight, gestational age, days on supplemental oxygen, systemic fungal infection, vaginal delivery). Use of erythropoietin was not significantly associated with other major sequelae of prematurity (intraventricular hemorrhage, bronchopulmonary dysplasia, necrotizing enterocolitis). © 2014 Elsevier Ireland Ltd. All rights reserved.

Conclusions: Use of erythropoietin is an additional, independent predictor of threshold ROP in ELBW neonates. Larger prospective, population-based studies should further clarify the extent of this association.

Keywords: Erythropoietin; Infection; Prematurity; Preterm neonate; Retinopathy.

MeSH terms

  • Case-Control Studies
  • Cohort Studies
  • Erythropoietin / adverse effects*
  • Female
  • Humans
  • Infant, Extremely Low Birth Weight
  • Infant, Newborn
  • Intensive Care Units, Neonatal
  • Italy
  • Male
  • Retinopathy of Prematurity / diagnosis
  • Retinopathy of Prematurity / etiology*
  • Retrospective Studies

Substances

  • Erythropoietin