Background Vitamin A deficiency is a major public health nutrition problem in the developing world. Even subclinical Vitamin A deficiency is associated with increased childhood mortality. Severe maternal vitamin A deficiency may cause increased mortality in the first months of life. There have been a limited number of studies regarding vitamin A status in Turkey. The aim of this study was to assess vitamin A status of healthy children in Manisa, Turkey. Methods Vitamin A status of 100 healthy children aged 36-48 months is evaluated. The children were seen during routine examination. Serum retinol concentrations were measured by high-performance liquid chromatography. Duration of breast feeding, age solid foods introduced, use of supplementary vitamins, weight and height, and intake of specific groups of nutrients on a daily, weekly and monthly basis were collected from a questionnaire completed by the mothers. Height and weight z-scores were calculated according to national standards. Mothers of 20 of the 100 children were known to have normal serum and breast milk retinol concentrations. Children with normal serum retinol concentration were compared with the children with VAD. Student's t-test and Mann-Whitney test were used to compare independent variables. The Pearson correlation analysis test was used to test relation between numeric variables. Results Mean retinol concentration was 0.98 ± 0.32 μmol/L in the whole study group. Serum retinol concentrations were normal (>0.70 μmol/L) in 89% of the children. When children with normal serum retinol concentrations were compared with those with retinol concentrations lower than 0.70 μmol/L, there was no difference in terms of age, gender, weight and height at the time of study, z-scores, birth weight, birth length, duration of breast feeding, time to begin solid food, rate of supplementary vitamin use, and rate of infections (P > 0.05). There was not any relation between vitamin A concentrations and weight and height at the time of study, z-scores, birth weight, birth length, duration of breast feeding, time to begin solid food, vitamin use, and frequency of intake of specific groups of nutrients (P > 0.05). Conclusions This study showed that VAD is a moderate health problem in Manisa.