Conjunctival impression cytology with transfer as a field-applicable indicator of vitamin A status for mass screening.

The increasing importance of vitamin A deficiency in even its mild subclinical form underlines the need for a mass screening test. Clinical, biochemical and cytological methods of assessing vitamin A deficiency in a public health setting have been described and widely used. The cytological method shows promise because it enables early detection of vitamin A deficiency. However interpretation is problematic since histopathological changes are gradual with the progressive disappearance of goblet cells and appearance of enlarged epithelial cells. The reliability and validity of the impression cytology with transfer (ICT) test were assessed in order to produce a meaningful standard for this cytological method. The ICT test was performed in Senegal on 1451 children, in the course of two surveys conducted in 1989 and 1990 in rural areas. Reliability, estimated by Cohen's kappa test for evaluating intra-reader variability, and sensitivity were highest for the abnormal-normal classification (kappa = 0.91; 95% confidence interval (CI): 0.89-0.93; and sensitivity = 74%; 95% CI 66-82%). The ICT method is a cheap, noninvasive and easy test to perform in the field. This method is also reproducible and fairly sensitive according to the abnormal-normal classification. As illustrated by our proposed 50% cutoff of abnormal cytology calculated in relation to 5% of serum retinol values below 0.35 mumol/L criterion, ICT only requires a small sample for the assessment of the overall health of a community in contrast to xerophthalmia and blood vitamin A deficiency tests.