RESUMEN
BACKGROUND: Vitamin A deficiency (VAD) is endemic in Brazil and health professionals have difficulty in recognizing its subclinical form. In addition, serum retinol concentrations do not always represent vitamin A status in the organism. OBJECTIVE: To identify VAD in preschool children by the serum 30-day dose-response test (+S30DR) and to examine its potential as a tool for the assessment of vitamin A status in the community. DESIGN: A prospective transverse study in which blood samples were obtained from 188 preschool children for the determination of serum retinol concentrations and the children were submitted to ocular inspection and anthropometric measurements. Information about the presence of diarrhea and/or fever during the 15 days preceding the study was also obtained. The children received an oral dose of 200,000 IU vitamin A immediately after the first blood collection. A second blood sample was obtained 30-45 days after supplementation in order to determine the +S30DR. RESULTS: In all, 74.5% (140/188; 95% confidence interval: 68.3-80.7%) of the children presented +S30DR values indicative of low hepatic reserves. The mean serum retinol concentration was significantly lower before supplementation (0.92 and 1.65 micromol/l, respectively; P < 0.0001). No child presented xerophthalmia; 3.7% (7/188) of the children were malnourished. The presence of fever and/or diarrhea during the 15 days preceding the first blood collection did not affect the +S30DR value. CONCLUSIONS: The prevalence of VAD in the study group was elevated. +S30DR proved to be a good indicator of subclinical VAD in children from an underdeveloped country.