several large cystic areas within the lesion. Kawamoto et al. (7) described it as a lesion located in the second and third echolayers. Hizawa et al. (8) described it as a hyperechoic or isoechoic lesion with various types of internal echoge- nicity, located in the mucosa and submucosa with a minimal degree of cystic formation. Based on the reported endosonographic features of the hamartoma and its histological characteristics [the hamar- toma develops in the deep mucosa and submucosa without involvement of the muscularis propria (9, 10)], we believe that an endosonographic feature of the lesion can be location within the second and third echolayers, not within the fourth echolayer. Mitsunobu Matsushita, M.D. Hiroshi Takakuwa, M.D. Akiyoshi Nishio, M.D. Department of Gastroenterology Tenri Hospital Nara, Japan REFERENCES 1. Block KP, Frick TJ, Warner TFSC. Gastrointestinal bleeding from a Brunner’s gland hamartoma: Characterization by en- doscopy, computed tomography, and endoscopic ultrasound. Am J Gastroenterol 2000;95:1581–3. 2. Matsushita M, Hajiro K, Takakuwa H, et al. Characteristic EUS appearance of Brunner’s gland hamartoma. Gastrointest Endosc 1999;49:670 –1. 3. Boyce GA, Sivak MV, Jr, Ro ¨sch T, et al. Evaluation of submucosal upper gastrointestinal tract lesions by endoscopic ultrasound. Gastrointest Endosc 1991;37:449 –54. 4. Inai M, Sakai M, Kajiyama T, et al. Endosonographic char- acterization of duodenal elevated lesions. Gastrointest Endosc 1996;44:714 –9. 5. Ueno N, Tomiyama T, Tano S, et al. A case of Brunner’s gland hyperplasia: Endoscopic color doppler ultrasonographic find- ings. Endoscopy 1997;29:51. 6. Weisselberg B, Melzer E, Liokumovich P, et al. The endo- scopic ultrasonographic appearance of Brunner’s gland hamar- toma. Gastrointest Endosc 1997;46:176 – 8. 7. Kawamoto K, Yamada Y, Furukawa N, et al. Endoscopic submucosal tumorectomy for gastrointestinal submucosal tu- mors restricted to the submucosa: A new form of endoscopic minimal surgery. Gastrointest Endosc 1997;46:311–7. 8. Hizawa K, Kawasaki M, Kuzuki T, et al. Endosonographic classifications of gastrointestinal submucosal tumors. Dig En- dosc 2000;12:120 –5. 9. Walden DT, Marcon NE. Endoscopic injection and polypec- tomy for bleeding Brunner’s gland hamartoma: Case report and expanded literature review. Gastrointest Endosc 1998;47: 403–7. 10. Levine JA, Burgart LJ, Batts KP, et al. Brunner’s gland hamartomas: Clinical presentation and pathological features of 27 cases. Am J Gastroenterol 1995;90:290 – 4. Reprint requests and correspondence: Mitsunobu Matsushita, M.D., Department of Gastroenterology, Tenri Hospital, 200 Mishima-cho, Tenri, Nara 632-8552, Japan. Received Aug. 14, 2000; accepted Nov. 21, 2000. Dry Cough and Optic Neuritis: Two Rare Complications of Interferon  Treatment in Chronic Viral Hepatitis * TO THE EDITOR: A 54-yr-old female patient with chronic hepatitis C was started on interferon- (IFN--2b, 3 MU t.i.w., s.c.) on August 30, 1999. All of the routine laboratory data were normal, except serum transaminases. Two weeks after institution of IFN treatment, she complained of dry cough, but there were no abnormal findings. Dry cough progressed despite administration of inhaled -2 agonist and corticosteroid. On November 16, she applied to us owing to epistaxis attacks and dry cough, which interfered with her usual daily activities. IFN therapy was stopped. Physical examination and hemorrhagic diathesis tests were normal. Electrocardiogram, chest X-ray, CT, and respiratory func- tion tests were not significant. Subsequently, on November 18 and November 23, the patient experienced epistaxis attacks that were easily controlled by nasal tamponate. Dry cough completely resolved in late December. On the other hand, she complained of headache over the right eye (RE) on December 13. On ophthalmologic examination, the best corrected visual acuity (BCVA) of the RE was 1/10. Right optic nerve head swelling was seen, and a central visual field defect was detected. There was no leakage on fundus flo- rescein angiogram. The left eye examination, neurological findings, electroencephalogram, and cranial and ophthalmic magnetic resonance imaging were all normal. In visual evoked potentials, the wave configuration from the RE was so deformed that it prevented measurement of the waves indicating optic neuritis (Fig. 1A). After 6 wk, the RE optic nerve head swelling decreased and BCVA became 6/10. There was an improvement of the visual field, and visual evoked potentials’ wave configuration of the RE was nearly within normal limits (Fig. 1B). In March and following visits, optic nerve head swelling disappeared and BCVA became 8/10. Dry cough, epistaxis, and optic neuritis during IFN ther- apy, especially with low doses, are extremely rare adverse events (1–5). In our patient, cough developed 2 wk after treatment with IFN- had been initiated. We consider that cough was due to IFN, because it failed to respond to inhaler treatment in the course of IFN; moreover, it increased, but ceased after IFN had been withdrawn. IFN-–related pneu- monia or pneumonitis has been reported (6, 7). In these patients, dyspnea, pulmonary infiltrates, and fever were present. Our patient had no fever, and no pathological find- ings to explain the cough. We supposed that the cause of epistaxis of the patient was excessive cough leading to upper respiratory irritation and congestion. We believe that optic neuritis was due to IFN treatment, because she had recently received no other drugs that may have caused optic neuritis. * This work was accomplished in the University of Suleyman Demirel School of Medicine, Isparta, Turkey. No financial support was received for the study. 1303 AJG – April, 2001 Letters to the Editor