Effects of Early Vitamin A Supplementation on the Risk for Retinopathy of Prematurity in Extremely Preterm Infants

Retinopathy of prematurity (ROP) is a common retinal neovascular disorder and major cause of vision impairment or blindness, despite current treatment of late stage ROP. Because the visual disorders after treatment are often poor, preventive therapy for ROP is still lacking. Although ROP is a multifactorial disease, the altered regulation of vascular endothelial growth factor (VEGF) and insulin-like growth factor (IGF-1) have been implicated in the pathogenesis of ROP. The vascular endothelial growth factor (VEGF) is a hypoxia-inducible cytokine and a vascular endothelial cell mitogen. If VEGF is suppressed, normal vessel growth is inhibited, but if in excess, retinal neovascularization is precipitated. This indicates that VEGF is a critical factor in retinal neovascularization. Inhibition of VEGF at the neovascular phase might prevent destructive neovascularization. However, the choice of any intervention for the inhibition of VEGF should be taken into account very carefully, because VEGF also promotes normal physiological development of blood vessels in many tissues. In addition, this intervention can be applied to all preterm infants when potential side effects are almost minimal.

Vitamin A is one of the most important micronutrients affecting the health of children. Supplementing newborn infants with vitamin A within the first 2 days of life reduced infant mortality by almost 25%, with the greatest benefit to those of low birth weight. Vitamin A has been used in this population prophylactically for chronic lung disease with the large doses and no reported significant adverse effect exists. It is suggested that vitamin A-retinoids and their active metabolite, retinoic acid (RA) have highly potent antiangiogenic activity by inhibiting VEGF expression. Vitamin A (retinol) is converted into retinoic acid in cells. However, the significance of Vitamin A administration has not been investigated to our knowledge in an experimental ROP infant. The aim of this study was to demonstrate the preventive effect of Vitamin A on ROP.

Methods

This prospective, multicenter, randomized study was performed from August 2015 to March 2017 in neonatal intensive care units in China. This study was approved by the Life Science Ethics Committee of Zhengzhou University and the local research ethics committees at the participating centers. Written informed consent was obtained from both parents when an infant was admitted to the NICU.

Patient Population

Patients eligible for the study were extremely preterm infants admitted to the NICU with gestational age <28 weeks, <72 h of age, receiving mechanical ventilation, noninvasive respiratory support or supplemental oxygen (FiO2>0.21) at 24h of age. Infants with genetic metabolic diseases, congenital abnormalities, congenital nonbacterial infection with overt signs at birth, terminal illness as evidenced by PH<7.0 for >2h or persistent bradycardia (heart rate <100 bpm) associated with hypoxia for >2h, or grade III or IV intracranial hemorrhage before randomization were excluded, as were infants for whom parental consent could not be obtained.

Randomization

Subjects were assigned to the oral vitamin A group or the placebo group based on a computer-generated randomization plan. The allocation ratio was 1:1, using variable block sizes. Randomization to the oral vitamin A group or the placebo group was carried out by random number allocation sequence, upon securing the order of admission to the NICU, and within 30 min after written informed consent was obtained.

Vitamin A administration

Subjects will be given the daily dose 1500 IU/day in drop form added to their enteral feeds as soon as minimal feeding is introduced. All extremely preterm infants will be provided with basic administration of an intravenous multivitamin (1ml/kg/day, containing vitamin A 230 IU/kg/day) preparation whilst on parenteral nutrition until fully enterally fed.

Data Collection

All vital signs, including blood pressure, heart rate, oxygen saturation, ventilator settings, and arterial blood gases, were monitored both before and during mechanical ventilation. The primary outcomes were mortality and incidence of BPD, as determined by an oxygen reduction test at 36 weeks of postmenstrual age, and further graded by severity, using criteria adapted from the National Institute of Child Health and Human Development. Mild BPD was defined as the need for supplemental oxygen for <28 days. Moderate BPD was defined as the need for supplemental oxygen at postmenstrual age of 36 weeks without positive pressure support. Severe BPD was defined as the need for mechanical ventilation.

Secondary outcomes were days on mechanical ventilation, hospital stay, surfactant requirement, and the occurrence of retinopathy of prematurity stage 2 or higher, pulmonary hemorrhage, patent ductus arteriosus, necrotizing enterocolitis, or pneumothorax. Doctors were blind as to group allocation during the whole study.

Statistical Analyses

The minimum sample size of 127 in each group (total 254) was estimated based on an expected incidence of ROP (including mild, type 1 and type 2 ROP) in the control group of 62%, a 2-sided significance level of 0.05, and an 80% chance of detecting a relative 30% decrease infrequency. Analyses were performed according to the intention-to-treat principle, and all who could be evaluated were included.

All analyses were performed with statistics software (SPSS 17.0, SPSS Chicago, Illinois). Quantitative data are expressed as mean ± SD. Entry data and outcome differences were compared with the t test and Fisher exact test. Subgroup interaction analyses were performed on the basis of sex, birth weight, single or multiple births, antenatal steroid or postnatal surfactant treatment, and intubation time for mechanical ventilation. All subgroup statistical analyses were evaluated with the Breslow-Day test for interaction. The kappa test was used to examine whether the results were consistent between the 2 centers. The level of statistical significance was set at P <0 .05.