(OR08-07-23) Sensitivity and Specificity of Breast Milk To Determine Vitamin A Status in Reference to the Retinol Isotope Dilution Test Among Lactating Thai Women

Objectives: This study determined the sensitivity and specificity of breast milk retinol concentrations in estimating vitamin A deficiency (VAD) among lactating women from rural and urban populations in Thailand, using total liver vitamin A reserves (TLRs) as the reference standard.

Methods: This analysis used data from a study of 94 lactating women, 6-8 wk postpartum, from rural (Ayutthaya) and urban (Bangkok) areas. Breastmilk (fore and hindmilk combined) retinol content was determined by HPLC. Total body stores (TBSs) of vitamin A were measured using the ¹³C-retinol isotope dilution (RID) technique; TLRs were estimated using assumptions for lactation status. Sensitivity and specificity of breastmilk retinol were assessed using TLRs as the reference standard, using cutoffs for VAD of ≤1.0 µmol retinol/L breastmilk and ≤0.10 µmol retinol/g liver.

Results: TLR and breastmilk retinol values were available for 88 women. TLR values revealed 13 cases of VAD (14.8%). Using breastmilk retinol to determine VAD compared with TLRs resulted in 8 true positive VAD cases, 65 true negatives, 10 false positives and 5 false negatives. Sensitivity and specificity were thus 61.5% and 86.7%, respectively, with a total of 85% of women correctly classified as deficient or adequate according to breastmilk retinol.

Conclusions: Breastmilk retinol concentrations had good sensitivity and specificity to determine VAD in lactating Thai women when compared with RID-estimated TLRs. In low-resource community studies aimed at improving vitamin A status of lactating women, breast milk may be a more accessible body pool. More research is needed to determine best sampling times.

Funding Sources: The fieldwork was supported by the International Atomic Energy Agency and the Thailand National Research Council.