Iron deficiency among incarcerated juvenile delinquents

A population of 163 incarcerated delinquents (126 males and 37 females aged 12-18 years) was studied to determine the prevalence of iron deficiency and to compare hemoglobin (Hgb), mean corpuscular volume (MCV), serum ferritin (SF), and erythrocyte protoparphyrin/hemoglobin (EP/Hgb) as predictors of response to iron therapy. Thirty-two percent of females and 6% of males had SF ≤ 12 ng/ml; 51% of females and 24% of males had SF ≤ 20 ng/ml. The mean SF was 17.7 ng/ml for females and 29.2 ng/ml for males. Of the 163 subjects, 53 were at risk for iron deficiency based on SF, Hgb, EP/Hgb, or MCV criteria. Twenty-one completed treatment with iron, and nine had greater than 1 g rise in Hgb. The following tests identified responders: SF ≤ 12 ng/ml-5/9; SF ≤ 20 ng/ml-9/9; Hgb ≤ third percentile-4/9; Hgb ≤ tenth percentile-7/9; MCV ≤ tenth percentile-2/9; EP/Hgb ≥ 3.0 μg/g Hgb-2/9; EP/Hgb ≥ 2.5 μg/g Hgb-4/9. Serum ferritin ≤ 20 ng/ml had a false positive rate of 57%; Hgb ≤ tenth percentile and EP/Hgb ≥ 2.5 μ g/g Hgb had no false positives. The significance of the high prevalence of iron deficiency among a population of incarcerated adolescents is not clear. Serum ferritin ≤ 20 ng/ml offers the most sensitive screening test to suggest iron-responsive anemia, but it has poor specificity; Hgb ≤ tenth percentile is an alternative screening test that is more available, less expensive, nearly as sensitive, and more specific; EP/Hgb ≥ 2.5 μg/g Hgb, MCV ≤ tenth percentile, SF ≤ 12 ng/ml, and Hgb ≤ third percentile did not prove to be good screening tests because of poor sensitivity.