The characteristics of premature infants with transient corneal haze.

BACKGROUND: The etiology of transient corneal haze in premature infants is not known and how it relates to clinical outcomes in premature infants is not clear.

OBJECTIVES: To study associated factors of transient corneal haze in premature infants.

METHODS: We performed a retrospective study of 261 premature infants from retinopathy of prematurity (ROP) screening in the neonatal intensive care unit at a tertiary referral hospital. Characteristics of premature infants with and without corneal haze were analyzed by correlation tests, Chi-square tests, and logistic regressions were used for statistical analyses. Associations between corneal haze and birth weight (BW), gestational age at birth (GA), central corneal thickness, intraocular pressure, and other systemic and ophthalmic data were evaluated.

RESULTS: The incidence of corneal haze was 13.4%. Lower BW, lower GA, packed red blood cells (RBC) transfusion, more days on oxygen, older maternal age, bronchopulmonary disease, and stage 3 ROP are associated with corneal haze. The severity of corneal haze decreased with infants' postmenstrual age. Multivariate logistic regression analyses revealed that BW and maternal age are the most important predictors of corneal haze.

CONCLUSION: Low BW and older maternal age are the most important predictors of corneal haze in premature infants. Premature infants with corneal haze could carry more systemic and ocular morbidities. Hence they may require more clinical attention. Corneal haze is unlikely to hinder the treatment of ROP. However, it is possible that corneal haze could hinder the examination of ROP in some infants. If corneal haze does interfere with ROP screening, a closer, more conservative follow-up schedule with a senior ophthalmologist experienced in managing ROP is recommended.