5. Yamamoto K, Sargent PA, Fisher MM, et al. Periductal fibrosis and lipocytes (fat-storing cells or Ito cells) during biliary atresia in the Lamprey. Hepatology 1986;6:54 –9. 6. Milani S, Herbst H, Schuppan D, et al. Procollagen expression by nonparenchymal rat liver cells in experimental biliary fibro- sis. Gastroenterology 1990;98:175– 84. Reprint requests and correspondence: Toshiaki Kunimura, M.D., First Department of Pathology, Showa University School of Medicine, 5 Hatanodai-1, Shinagawa-ku, Tokyo, 142-8666, Japan. Received Oct. 10, 2000; accepted Oct. 23, 2000. Re: Brunt et al.—Histological Evaluation of Iron in Liver Biopsies: Relationship to HFE Mutations TO THE EDITOR: Brunt et al. (1) have provided valuable information regarding the interpretation of iron deposition in liver biopsy specimens. Specifically, a “reticuloendothe- lial pattern” was found to provide strong evidence against the presence of C282Y homozygosity, or compound het- erozygosity with respect to the HFE gene. In contrast, the “hepatocellular pattern” was supportive of, but not specific to, these HFE genotypes. However, we were surprised by the absence of information on transferrin saturation levels, an important initial test in the investigation of HFE-associ- ated hereditary hemochromatosis (HH) (2). At our institution, we reviewed the genotypes of individ- uals referred for HFE testing on the basis of liver biopsy histology suggestive of HH. In particular, we reviewed those patients who also had nonphlebotimized transferrin saturation values within the normal or mildly elevated range (56%) to determine if they were part of the HFE-associ- ated spectrum of disease. From our database of 1295 patients, nine were identified who met both of the aforementioned characteristics. Three individuals (two women, one man) had double HFE muta- tions (one C282Y/C282Y and two C282Y/H63D). The cor- responding saturation values ranged from 40% to 55%. Two individuals were heterozygous for the C282Y mutation and were found to also have hepatitis C and alcoholic steato- hepatitis, respectively. The four remaining patients had no C282Y mutations. One of these patients had alcoholic ste- atohepatitis and a transferrin saturation of 42%. The other three had transferrin saturation values of 22%, 22%, and 27%, respectively. These findings indicate that the specificity of a hepato- cellular pattern of iron deposition as an indicator of HFE- associated HH can be increased if interpreted in combina- tion with the patient’s transferrin saturation level. Specifically, patients with a low normal transferrin satura- tion level (30%) are unlikely to have C282Y mutations. We would be interested to know what the transferrin satu- ration values were for the patients from the study by Brunt et al. (1) who lacked the C282Y mutation. Also, how many of the patients with HFE C282Y homozygote or compound heterozygote mutations had transferrin saturation levels 40%? Andre Mattman David Huntsman Gillian Lockitch Genes, Elements and Metabolism Program Children and Women’s Hospital Vancouver, Canada David Owen Department of Anatomic Pathology Vancouver General Hospital Vancouver, Canada REFERENCES 1. Brunt EM, Olynyk JK, Britton RS, et al. Histological evaluation of iron in liver biopsies: Relationship to HFE mutations. Am J Gastroenterol 2000;95:1788 –93. 2. Powell LW, Subramaniam BN, Yapp TR. Haemochromatosis in the new millenium. J Hepatol 2000;32(suppl 1):48 – 62. Reprint requests and correspondence: Andre Mattman, Depart- ment of Medical Biochemistry, St. Paul’s Hospital, 1081 Burrard Street, Vancouver, BC, Canada V6Z 1Y6. Received Oct. 11, 2000; accepted Oct. 23, 2000. Fluticasone in Eosinophilic Corrugated Ringed Esophagus TO THE EDITOR: Idiopathic eosinophilic esophagitis (IEE) often beginning in young children, usually male, frequently manifesting as dysphagia or food impactions, is reported to range endoscopically from being visibly normal to a spectrum of subtle changes (1). Some pediatric series include an occasional multiringed, endoscopically striking, nearly full-length, trachea-like “too small” esophagus that I call in adults “corrugated ringed esophagus” (CRE) (2). Like IEE, CRE generally occurs in highly allergic men, and most demonstrate striking esophageal intramucosal eosino- philia. Experiences in IEE indicate rapid, often within days, major symptom reversal by oral (3, 4), topical corticoste- roids (5), and/or elimination diets (1, 6). I have seen classic eosinophilic CRE with deep longitudinal tears after 42 F (14 mm) dilations developing over 2 yr in a still highly allergic 26-yr-old man (7) with previously visually normal esopha- geal body on three prior videoendoscopies. Not seeing chil- dren, I am not aware of how often immediate endoscopy after dilation in a noncorrugated pediatric IEE population shows classic longitudinal trauma, even into muscle, as in CRE. Impactions in CRE are devastating, so hoping for revers- ibility, I have now treated five with biopsies all showing significant intramucosal eosinophilia with 4 – 6 wk of fluti- 926 Letters to the Editor AJG – Vol. 96, No. 3, 2001