%0 Journal Article
%T Pica associated with iron deficiency or depletion: clinical and laboratory correlates in 262 non-pregnant adult outpatients
%A James C Barton
%A J Clayborn Barton
%A Luigi F Bertoli
%J BMC Blood Disorders
%D 2010
%I BioMed Central
%R 10.1186/1471-2326-10-9
%X We reviewed charts of 262 non-pregnant adult outpatients (ages ¡Ư18 y) who required treatment with intravenous iron dextran. We tabulated their sex, age, race/ethnicity, body mass index, symptoms and causes of iron deficiency or depletion, serum iron and complete blood count measures, and other conditions at diagnosis before intravenous iron dextran was administered. We excluded patients with serum creatinine >133 ¦̀mol/L or disorders that could affect erythrocyte or iron measures. Iron deficiency was defined as both SF <45 pmol/L and TS <10%. Iron depletion was defined as serum ferritin (SF) <112 pmol/L. We performed univariable comparisons and stepwise forward logistic regression analyses to identify significant correlates of pica.There were 230 women (184 white, 46 black; ages 19-91 y) and 32 men (31 white, 1 black; ages 24-81 y). 118 patients (45.0%) reported pica; of these, 87.3% reported ice pica (pagophagia). In univariable analyses, patients with pica had lower mean age, black race/ethnicity, and higher prevalences of cardiopulmonary and epithelial manifestations. The prevalence of iron deficiency, with or without anemia, did not differ significantly between patients with and without pica reports. Mean hemoglobin and mean corpuscular volume (MCV) were lower and mean red blood cell distribution width (RDW) and platelet count were higher in patients with pica. Thrombocytosis occurred only in women and was more prevalent in those with pica (20.4% vs. 8.3%; p = 0.0050). Mean total iron-binding capacity was higher and mean serum ferritin was lower in patients with pica. Nineteen patients developed a second episode of iron deficiency or depletion; concordance of recurrent pica (or absence of pica) was 95%. Predictors of pica in logistic regression analyses were age and MCV (negative associations; p = 0.0250 and 0.0018, respectively) and RDW and platelet count (positive associations; p = 0.0009 and 0.02215, respectively); the odds ratios of these predictors were low.
%U http://www.biomedcentral.com/1471-2326/10/9