Test-retest reproducibility of the relative dose response for vitamin A status in Guatemalan adults: issues of diagnostic specificity.

ABSTRACT

The relative dose response (RDR) test was examined with regard to specificity and reproducibility in subjects with adequate dietary intake and normal liver reserves of vitamin A. Twelve subjects were administered an RDR test four times over 22 d, including one placebo test in which the oral dose of vitamin A was omitted. Additionally, static measures of retinol, tocopherol, retinol binding protein (RBP) and prealbumin (PAL) were taken to determine the intra- and inter-individual coefficient of variation for these blood constituents. Intra-individual coefficients of variation were as follows retinol, 8.8%; RBP, 11.5%; PAL, 7.6%; and alpha-tocopherol, 8.8%. The mean RDR in vitamin A-replete subjects was 1.2% and ranged from approximately -25% to 11%. No differences were observed between placebo and true RDR (i.e., with vitamin A) test responses, and there was no difference among the three true RDRs over a period of 22 d. Consistent with the hypothesis upon which the RDR test is based, nascently absorbed vitamin A evidently entered a storage pool in the liver of vitamin A-replete subjects without immediate release to peripheral sites of utilization. Because the RDR test results were normal in all subjects, the procedure appears to offer high test specificity and does not falsely diagnose hypovitaminosis A. Nevertheless, the magnitude and direction of the RDR within an individual over 22 d were highly variable, and this variability may preclude the use of a single measure of the RDR to grade the relative vitamin A nutriture of an individual subject.