The relationship between iron deficiency and cognitive test scores among 5398 children, 6 to 16 years old, was studied at the University of Rochester School of Medicine, NY, by analysis of data obtained from the National Health and Nutrition Examination Survey III 1988-1994. Iron deficiency was based on measures of serum ferritin, transferrin saturation, and free erythrocyte protoporphyrin. Standardized test scores were compared for children with normal iron values, iron deficiency without anemia, and iron deficiency with anemia. Among this nationally representative sample of school-aged children and adolescents, 3% were iron-deficient. Among adolescent girls, the prevalence of iron deficiency was 8.7% (only 1.5% with anemia). Average math scores for iron-deficient children with and without anemia were 86.4 and 87.4, respectively, compared to 93.7 for children with normal iron status (P<.05). The block design test score was also significantly lower in iron-deficient children with anemia (8.0 vs 9.5; P<.05), and other tests showed a trend toward lower scores. The risk of scoring below average in math for iron deficient children was greater than twice that in children with normal iron values. This increased risk was present in both iron-deficient children with or without anemia. 
COMMENT. Iron-deficiency, with or without anemia, in school-aged children and adolescents is associated with lower scores on standardized math tests. The prevalence of iron deficiency is highest in adolescent girls, reaching 8.7% in a nationally US representative sample. Screening for iron deficiency should be considered in children with learning and behavior problems, even in those without anemia. A few previous studies have demonstrated adverse effects of anemia on cognitive performance, and variable effects of iron therapy on learning and memory. In future prospective studies, measures of iron status in relation to cognitive test scores appear to be more significant than hemoglobin values.
Reversible focal neurologic deficits were reported in a 14-year-old black female adolescent treated at Duke University Medical Center (see Progress in Pediatric Neurology I, PNB Publishers, 1991;pp397-8) . After transfusion and treatment with ferrous sulfate, a facial palsy resolved in 12 hours and a VI nerve palsy and somnolence resolved by the fifth day. Despite continued bleeding from hereditary telangiectasia, the neurologic exam remained normal while supplemental iron and a normal hemoglobin were maintained. See Ped Neur Briefs (May 1997;11:33-34) for further reference to iron deficiency and learning disorders.