Vitamin A status in the institutionalized elderly. Critical analysis of four evaluation criteria: dietary vitamin A intake, serum retinol, relative dose-response test (RDR) and impression cytology with transfer (ICT).

Infraclinical vitamin A deficiencies may be health-threatening for elderly people, yet they are difficult to assess unequivocally in this population. In this study, we evaluate the vitamin A status of an elderly institutionalized population (49 subjects, 83.6 +/- 6.1 years of age), by examining four different criteria: the dietary vitamin A intake, the retinol concentration in serum, the relative dose-response test and the impression cytology with transfer. The incidence of infra-clinical deficiencies was estimated to be 55% by examining dietary vitamin A intake, 21% by using the RDR test, 6% by the ICT and 2% from serum retinol values. These variations are not due to the choice of threshold values for each of the methods, yet rather to poor correlations between the results given by these methods. Canonical correlation analyses indicate that some parameters related to retinol secretion from the liver, including Zn, prealbumin and retinol-binding protein, can affect individual patient response towards the different methods. Validation of the RDR test in this elderly population was not successful because of poor reproducibility and moderate correction of RDR-detected vitamin A deficiencies by vitamin A supplementation. The method chosen for the determination of vitamin A status in elderly people must be carefully evaluated to account for possible age-related changes in the patient response to the method employed. In the institutionalized elderly population examined in this study, we observed a low vitamin A intake, whereas serum retinol and ICT are within normal ranges and while RDR test's responses appear too variable to draw any conclusion.