D. Makdani et al., COMPARISON OF METHODS OF ASSESSING VITAMIN-A STATUS IN CHILDREN, Journal of the American College of Nutrition, 15(5), 1996, pp. 439-449
Objective: A study of children (2-8 years; n = 613) in Belize, Central
America, was conducted to determine what proportion of the children m
ight be at risk of vitamin A (vit A) deficiency. The data provide an o
pportunity to compare results of three methods of assessing vit A stat
us in a population which was not severely malnourished. Serum retinyl
ester concentrations were also determined; their relevance to one of t
he tests, the relative dose response (RDR) test is discussed. Methods:
The three methods of assessing vit A status were: RDR test, fasting s
erum retinol concentration, and conjunctival impression cytology (CIC)
. Retinol-binding protein (REP), serum retinyl esters and serum zinc c
oncentrations were also determined. Results: Inadequate vit A status w
as indicated for 17% of subjects by the RDR test (14% cutoff), for 24%
by fasting serum retinol concentration (<0.87 mu mol/L), and for 49%
by ''abnormal'' CIC score. Retinyl esters constituted 24% of serum ret
inoids at the time (5 hours after a retinyl palmitate dose) at which t
he second blood sample is taken for the RDR test. Regression tree anal
yses (CART) indicated ethnicity was a predictor of RDR score; ethnicit
y, stunting and age were predictors of fasting serum retinol concentra
tion; ethnicity and stunting were predictors of 0-hour retinyl ester c
oncentration. Conclusion: The three indices of vit A status did not id
entify the same individuals nor indicate the same percentage of the po
pulation to be at risk for vit A deficiency. Increased concentrations
of retinyl esters at 5 hours compared to those at 0 hours suggest that
insufficient retinol may have been taken up by the liver at 5 hours t
o release all accumulated retinol-binding protein (REP) in deficient i
ndividuals; prevalence of vit A deficiency might therefore be underest
imated by the RDR test The selection of ethnicity as a predictor of RD
R score and of 0-hour retinol and retinyl ester concentrations suggest
s that factors other than vit A status affect vit A metabolism and may
affect the RDR test.