Basu S, Khanna P, Srivastava R, Kumar A. Oral vitamin A supplementation in very low birth weight neonates: a randomized controlled trial. Eur J Pediatr 2019;178(8):1255-1265. PMID 31209560.
Dr Aparna Chandrasekaran, DM
Ankura Hospital for women and children,
Dr Srinivas Murki, DM
- Design: Single centre, placebo controlled randomised control trial
- Allocation: Random sequence generation was done by an independent statistician in random permuted blocks of 4,6 and 8
- Blinding: Blinding of the intervention as well as outcome assesment was achieved by preparing vitamin A solution at a concentration of 10000 IU/ml as well as placebo similar in smell and colour. Both preparations were of 20 ml volume and were provided along with a 1 ml dropper. The treating team of doctors and nurses as well as parents were blinded to the intervention.
- Follow-up period: Neonates were followed until hospital discharge or until 36 weeks’ postmenstrual age
- Setting: Tertiary care Level 3 neonatal intensive care unit (NICU)
- Patients: Inborn very low birth weight (VLBW) neonates on any form of respiratory support at 24 hours of age were included in the study. Neonates with major congenital malformations as well as conditions which contraindicated enteral feeding such as perinatal asphyxia and doppler abnormalities were excluded. The study enrolled 196 neonates of which 98 neonates were randomized each group – vitamin A or placebo. The mean gestation ages of the vitamin A and placebo group – 30.9 ± 2.9 and 30.7 ± 2.7 weeks, respectively, were comparable (p=0.49). The groups were also similar with respect to birth weights (1185 ± 194 vs. 1163 ± 181 grams), complete coverage with antenatal steroids (40% vs. 38%), need for surfactant replacement (42% vs. 44%) etc.
- Intervention: Neonates in the intervention and placebo groups received 1 mL of syrup vitamin A or placebo on alternate days starting from 24 hours of age till 28 days of postnatal life (total 14 doses). The mode of administration was via orograstric tube or orally in the mouth based on the mode of feeding.
- Primary outcome: Primary outcome was the composite incidence of all cause mortality and/or oxygen requirement for 28 days.
- Secondary outcomes: Secondary outcomes were the safety and tolerability of vitamin A, serum retinol concentration of neonates at recruitment and on day 28, total duration of oxygen supplementation, duration of respiratory support, incidences of sepsis, hemodynamically significant-PDA, Necrotising enterocolitis (Bell stage IIand beyond), intraventricular hemorrhage (IVH), periventricular leukomalacia (PVL), and Retinopathy of prematurity (ROP).
- Analysis and Sample Size: The sample size was calculated on the premise that existing incidence of BPD (defined as oxygen requirement for ≥ 28 days) in the centre in a similar population of neonates was 64% in the previous year. A total sample size of 178 neonates was required to demonstrate a 20% relative risk reduction with the intervention with a confidence level of 95% and power of 80%. SPSS version 16.0 was used for data entry and analysis. Categorical variables were compared in the two groups using relative risk (RR) with 95% confidence interval (CI), and number neededto treat for benefit (NNTB).
- Patient follow-up: All neonates randomized received the allotted intervention. Eight five neonates (87%) in the vitamin A group and 76 neonates in the placebo group (78%) completed the study. Analysis was by intention-to-treat technique.
PRIMARY OUTCOME: The composite outcome of all-cause mortality and oxygen requirement for 28 days was significantly lower in the vitamin A group (11/98; 11.2%) compared to the placebo group [25/98; 25.3%, relative risk (RR) 0.440; 95% confidence intervals (CI): 0.229–0.844; p < 0.05, NNT (number needed to treat): 7].
SECONDARY OUTCOMES: The incidence of late-onset sepsis and hs-PDA were significantly lower in vitamin A group when compared to the placebo group. The incidence of bronchopulmonary dysplasia (BPD) at 36 weeks’ PMA and the incidence of other morbidities such as NEC were comparable in the two groups (table below).
Compared to placebo group, the requirement and duration of oxygen supplementation and non-invasive respiratory support by continuous positive airway pressure (CPAP) or high flow nasal cannula (HFNC) were significantly lower in vitamin A group. Proportion of neonates requiring mechanical ventilation and the duration of MV were similar in the groups.
A significantly higher proportion (33/98) of neonates had low serum retinol concentrations in the placebo compared with none (0/98) in vitamin A group (p <0.01).
|PARAMETER||VITAMIN A (n=98)||PLACEBO (n=98)||p VALUE|
Composite incidence of all-cause mortality and oxygen requirement for 28 days
|11 (11.2)||25 (25.3)||0.013|
1. Late onset sepsis
BPD – Bronchopulmonary dysplasia
Early postnatal oral vitamin A supplementation was associated with significantly lower incidence of all-cause mortality and oxygen requirement for 28 days in VLBW neonates with respiratory distress.
With increasing survival, bronchopulmonary dysplasia (BPD) is becoming a common morbidity among preterm neonates (1). With the exception of caffeine and systemic postnatal steroids, few pharmacological measures have been shown to be effective in preventing or treating BPD (2). Vitamin A regulates cellular differentiation and maintains respiratory epithelial integrity and hence, appears promising in preventing BPD. Intramuscular (IM) vitamin A has been shown to reduce BPD by 7% (pooled relative risk [RR]: 0.93; 95% CI 0.88 to 0.99) (3). Although effective, IM vitamin A supplementation is painful, carries the risk of infection and may be difficult in extremely preterm neonates with thinned out muscle mass.
Oral vitamin A supplementation has been evaluated as an alternative to IM vitamin A. Despite being equally effective as IM vitamin A in replenishing vitamin A stores, oral vitamin A has been shown to be ineffective in preventing BPD or death (4,5).
This RCT by Basu and colleagues demonstrates that oral vitamin A supplementation among very low birth weight neonates at 10000 IU/dose until 28 days of life is associated with a significantly lower incidence of “death or BPD” (6). The study is conducted with appropriate block randomization, allocation concealment and blinding with high internal validity. The authors have demonstrated the effectiveness of the intervention by showing a significant decrease in low vitamin A (<20 μg/dL) status in the supplemented group.
Although the intervention is simple, cost effective and promises to reduce the burden of a disease deemed difficult to prevent or treat, there are a few questions to ponder. Firstly, there is a possibility of unequal distribution of confounding factors such as gestational age, invasive ventilation, sickness severity, sepsis and a hemodynamically significant PDA (hs-PDA), which may influence the outcome, more so, with a multifactorial outcome like BPD. Other than gestational age, the authors have not used stratified randomization or performed regression analysis for other important prognostic variables. Authors are unclear whether that hs-PDA is a baseline variable or an outcome. Considering that hs-PDA is an early morbidity usually presenting within the first week of life, it would have been appropriate to consider it is a baseline variable.
Secondly, only 13% of the enrolled neonates were born before 28 weeks’ gestation, a group which is at highest risk of BPD and hence, most likely to benefit from the intervention (7). Nearly 18% of the entire study population (35/196) either died or left against medical advice before completing the allotted intervention. Also, the new consensus definition of BPD – which assesses oxygen requirement at 36 weeks’ postconceptional age – would have been more appropriate as the primary outcome as it more accurately identifies the neonates at risk of long term respiratory and neurodevelopmental sequelae (8). These 3 limitations may restrict the generalizability of the findings.
Third – the choice of “death or BPD” as the primary outcome warrants discussion. Most prematurity related deaths (up to 63%) occur within 72 hours (9) and hence, unlikely to be influenced by an intervention administered over 4 weeks. Hence, BPD alone or “BPD and late neonatal death” would have been a better primary outcome.
Finally, the sample size has been calculated hypothesizing that the intervention would result in a decline in the primary outcome from 64% to 52%, whereas the observed incidence of the primary outcome was only 25% in the placebo group, which may indicate a significant change in other variables such as delivery room or initial respiratory management, use of surfactant, non-invasive ventilation etc.
Given that BPD is multifactorial in origin and influenced by multiple interventions such as resuscitation practices, ventilatory strategies, nutritional policies and fluid management, the effect of supplementing one oral vitamin in modulating its risk is attractive and also intriguing. More data on the duration and safest dose for supplementation as well as the choice of a population which is likely to benefit the most is required.
- Bhunwal S, Mukhopadhyay K, Bhattacharya S, Dey P, Dhaliwal LK. Bronchopulmonary Dysplasia in Preterm Neonates in a Level III Neonatal Unit in India. Indian Pediatr 2018;55(3):211-215.
- McPherson C. Pharmacotherapy for the Prevention of Bronchopulmonary Dysplasia: Can Anything Compete with Caffeine and Corticosteroids? Neonatal Netw 2019;38(4):242-24.
- Darlow BA, Graham PJ, Rojas-Reyes MX.Vitamin A supplementation to prevent mortality and short- and long-term morbidityin very low birth weight infants. Cochrane Database of Syst Rev 2016;8:CD000501.
- Landman J, Sive A, Heese HD, Van der Elst C, Sacks R. Comparison of enteral and intramuscular vitamin A supplementationin preterm infants. Early Hum Dev 1992;30(2):163–170.
- Wardle SP, Hughes A, Chen S, Shaw NJ.Randomised controlledtrial of oral vitamin Asupplementation in preterm infants toprevent chronic lung disease. Arch Dis Child Fetal Neonatal Ed 2001;84(1):F9–F13
- Basu S, Khanna P, Srivastava R, Ashok K. Oral vitamin Asupplementation in very low birth weight neonates: a randomizedcontrolled trial. Eur J Pediatr 2019;178(8):1255–1265.
- Stoll BJ, Hansen NI, Bell EF, Shankaran S, Laptook R, Walsh MC, et al. Neonatal outcomes of extremely preterm infants from the NICHD neonatal research network. Pediatrics 2010;126(3):443–456.
- Ehrenkranz RA, Walsh MC, Vohr BR, Jobe AH, Wright LL, Fanaroff AA, et al. Validation of the National Institutes of Health consensus definition of bronchopulmonary dysplasia. Pediatrics 2005;116(6):1353-60.
- Sankar MJ, Natarajan CK, Das RR, Agarwal R, Chandrasekaran A, Paul VK. When do newborns die? A systematic review of timing of overall and cause-specific neonatal deaths in developing countries. J Perinatol 2016;36 Suppl 1:S1-S11