Changes in Vitamin A Levels and the Effect of Early Vitamin A Supplementation on Vitamin A Levels in Infants Throughout the First 6 Months of Life: A Prospective Cohort Study in Chongqing, China

Objectives: This study aimed to explore the changes in infant vitamin A (VA) status and the effect of early VA supplementation on VA level throughout the first 6 months of life. Methods: A prospective cohort study was conducted in Chongqing, China. A total of 1,016 healthy infants were enrolled at birth. Then, 930, 882, 854 and 822 healthy infants were followed up at postnatal day 7 and postnatal months 1, 3, and 6, respectively. Blood samples and dietary survey and physical development data were collected. Serum VA was measured by chromatography tandem-mass spectrometry and was classified according to the VA deficiency (VAD) criteria for older children aged 6–70 months (<0.70, 0.70–1.05, ≥1.05 μmol/L). Normally distributed continuous variables are presented as the mean ± standard deviation. The categorical variables are described by the frequency and percentage (%). The reference interval for the VA level was the 2.5th−97.5th percentile. Changes in VA status with age and the relationship of VA supplementation with VA level were investigated by generalized estimating equations followed by Bonferroni post hoc test, controlling for the effects of feeding pattern and sex. Results: Infant VA levels increased significantly from 0.499 ± 0.146 to 1.061 ± 0.414 μmol/L with age at 6 months, even without VA supplementation (P < 0.05). From birth to 6 months, the percentage of infants with a VA level <0.70 μmol/L decreased from 88.6 to 19.5%. During follow-up, no infant demonstrated clinical VAD conditions, such as night blindness, conjunctival xerosis or Bitot's spots. Less than 7.0% of infants were underdeveloped in terms of weight, length and head circumference. The VA status of infants with VA≥0.588 μmol/L at birth gradually increased to adequate VA (VA ≥ 1.05 μmol/L) at 6 months. For these infants, there was no significant difference in VA level between the VA supplementation and non-supplementation groups (P > 0.05). Infants with VA <0.430 μmol/L at birth still had VA <0.70 μmol/L at 6 months; in this group, VA levels increased by 0.08 μmol/L more among supplemented infants than among non-supplemented infants (P < 0.05). Conclusions: A low VA level among neonates at birth may be a normal physiological state and may increase with age; thus, not all neonates may need early VA supplementation. More multicenter studies are needed to determine a new cutoff point for the diagnosis of neonatal VAD and the administration of nutritional intervention.