REGULAR ARTICLE Dietary vitamin A intake and serum retinol concentrations of preschool children from different socio-economical levels in _ Izmir, Turkey Levent Midyat (midyatlev@yahoo.com), N Zafer Kurugöl, Sadık Aks ¸ it, Mahir Tanrıverdi, Ras ¸ it Vural Yag ˘cı Department of Pediatrics, Ege University Faculty of Medicine, Izmir, Turkey Keywords Children, Malnutrition, Retinol, Vitamin A Correspondence Levent Midyat, Ege Universitesi Cocuk Sagligi ve Hastaliklari AD, 35100, Bornova, Izmir, Turkiye. Tel: +90-532-4059763 | Fax: +90-216-4214110 | Email: midyatlev@yahoo.com Received 26 October 2010; revised 1 December 2010; accepted 4 January 2011. DOI:10.1111/j.1651-2227.2011.02142.x ABSTRACT Aim: Vitamin A deficiency (VAD), especially in its subclinical form, is a world health problem in young children. The aim of this study was to determine the prevalence of VAD among preschool children in various socio-economic groups in _ Izmir, Turkey. Methods: One hundred and one children aged 24–59 months were selected for the study with cluster sampling method. A questionnaire was completed by the parents, and the dietary pattern was assessed by using a 3-day dietary record method. Results: Mean serum retinol concentration of the children was 28.3 ± 10.2 lg ⁄ dL. Serum vitamin A concentration of two children (2%) was below 10 lg ⁄ dL (VAD), whereas 18 (18%) children had vitamin A concentrations between 10 and 20 lg ⁄ dL (subclinical deficiency). We did not find any relationship between mean serum retinol concentrations and the independent variables such as socio-economical levels, age groups, birth weight and gestational age. However, there was a statistically significant positive correlation between serum retinol concentrations and weight-for-age and weight-for-height Z scores of the children (p < 0.05). Conclusion: Because one-fifth of the children under 5 years have VAD in _ Izmir, Turkey, nutrition education must be given to the families during well-child care visits, and routine vitamin supplementation should be considered especially in children with malnutrition. INTRODUCTION Vitamin A has been indicated to be very important for growth, development and maturation of a child. It is a com- ponent of retinal pigments for vision in dim light and is nec- essary for development of bone and tooth, stabilizes the structure and function of mucosal surfaces and is involved with immune response (particularly T-cell function) and mucus production (1). Vitamin A deficiency (VAD) is usu- ally seen as a result of low dietary intake and is associated with malnutrition especially in developing countries (2). VAD is primarily a problem of children aged between 6 months and 6 years and is the leading cause of prevent- able blindness in both children and adults. VAD, even at subclinical levels, may have negative effects on health (3,4), and children with subclinical VAD are at increased risk of impaired immunity, anaemia and increased morbidity and mortality from a range of infectious diseases such as diar- rhoea and measles (5–8). It is estimated that approximately 140 million children of preschool age suffer from VAD worldwide every year, and 4.4 million of these children suffer from xerophthalmia (8– 10). According to WHO (World Health Organisation), when 5% or more of the population have serum retinol con- centrations <10 lg ⁄ dL, VAD is a serious public health prob- lem (11). However, The Pan American Health Organisation (PAHO) considers VAD as a public health problem when 15% or more of the population display serum retinol con- centrations of £20 lg ⁄ dL, which is defined as ‘subclinical vitamin A deficiency’ (12,13). The objective of this study was to determine the preva- lence of VAD among 24–59 months in various socio- economic groups and to evaluate child care characteristics and nutritional status associated with variation in serum ret- inol concentrations in children in _ Izmir province of Turkey. SUBJECTS AND METHODS The study was conducted in _ Izmir, the largest city of the Aegean region of Turkey. The protocol for the study was approved by the ethical committee of Ege University Medi- cal School. One hundred and one children were selected for the study using the cluster sampling method from three urban and three rural areas of _ Izmir. The children were aged between 24 and 59 months in April 2006, when the study was carried out. The families of the children were classified into a lower, middle and upper social group according to their socio-economic conditions and to the total income in that house. The three categories reflect the data of Tu ¨ rkiye _ Is ¸ c ¸ i Sendikaları Konfederasyonu (Turkey Labor Unions Confederation) (14); ‘below the hunger line’, ‘between Acta Pædiatrica ISSN 0803–5253 e24 ª2011 The Author(s)/Acta Pædiatrica ª2011 Foundation Acta Pædiatrica 2011 100, pp. e24–e27