RESUMO
OBJECTIVE: This study assessed the vitamin A nutritional status of preterm infants determined by the vitamin A relative dose-response test (RDR) compared with serum levels of vitamin A, retinol-binding protein (RBP), transthyretin (TTR), and retinol relations with carrier proteins. METHODS: Serum levels of retinol, RBP, and TTR and retinol/RBP, retinol/TTR, and RBP/TTR molar ratios were determined in 120 infants at 7 d and in 92 at 28 d. For the determination of the performance of the tests, the RDR was considered the reference method. The sensitivity and specificity for all possible cutoff values were determined by constructing receiver operator characteristic curves. The areas under the curves were used to estimate the overall accuracy of the tests. The best cutoff values to be used for the calculation of sensitivity and specificity were determined with 95% confidence intervals. RESULTS: RDR indicated vitamin A deficiency in 60% of the infants at 7 d and in 51.1% at 28 d. In the receiver operator characteristic curves, the best area under the curve was 0.710 obtained for serum retinol at 28 d of postnatal age and considered moderately accurate. The least inadequate cutoff level was set at 25 mg/dL, but no value was considered adequate due to low sensitivity and/or low specificity. CONCLUSION: Compared with RDR, the determination of serum levels of retinol, RBP, and TTR and their molar ratios are not adequate to assess nutritional vitamin A status in preterm infants.
