Sideropenic anemia and celiac disease: one study, two points of view.

Recent studies have pointed to the relationship between iron deficiency anemia and celiac disease, although data on the prevalence of celiac disease in anemic patients have been conflicting, and there is no agreement on the best screening procedure for CD in these patients. Our aims were to evaluate the relationship between anemia and celiac disease (CD) from two different points of view--the hematology clinic and the pediatric gastroenterology department--and to evaluate the utility of anti-endomysial antibody determination in screening anemic patients for CD using human umbilical cord as substrate. We studied 130 patients with CD (58 males, 72 females; median age 18 months) diagnosed at a department of Pediatric Gastroenterology, and 85 patients with iron deficiency anemia (38 males, 47 females; median age 48 years) observed at a hematology outpatient clinic. From the 85 adult patients with iron deficiency anemia, we selected a subgroup of 25 subjects with no improvement in Hb after two months of iron therapy (80 mg/day orally). Routine hematochemical tests were performed in all 215 patients. All pediatric and adult subjects underwent immunological screening for celiac disease (AGA and EmA assay); intestinal biopsy was also performed on patients testing positive. In the adult anemic patients a serum sample was stored at -20 degrees C on first observation, and after 6-18 months EmA on human umbilical cord were assayed. In the pediatric patients with CD, anemia was observed in 91/130 patients (70% of cases, the most frequent symptom after poor growth); however, this was the only presenting symptom of CD in 2/130 patients (1.5% of cases). Anemia was sideropenic in 41/91 patients (iron <45 microg/dl, ferritin <15 microg/liter). In the adult patients with iron deficiency anemia, immunological screening (AGA and EmA) showed suspected CD in 5/85 cases (5.8%), with diagnosis confirmed on intestinal biopsy. These five patients were in the subgroup of iron supplementation therapy nonresponders. CD prevalence in the refractory anemia subgroup was, therefore, 5/25 (20%). On diagnosis the hematological indices of the anemia + CD patients were not different than those of the refractory anemia patients without CD. The median age of the CD + anemia patients was significantly lower than that of the whole group of anemic subjects, and there was also a prevalence of females (4/5 cases). The results of the EmA determination on human umbilical cord in the adult anemic patients showed a perfect concordance with those using a traditional kit that uses monkey esophagus as substrate. In the pediatric age group many cases of CD with anemia as the only sign of the disease are probably not diagnosed. In our adult patients with sideropenic anemia, CD prevalence was 5-6%; however, the observation of anemic patients not responding to oral iron therapy makes a diagnosis of CD much more probable. EmA determination on human umbilical cord is the most logical approach to screen anemic patients for suspected CD.