2298
BRIEF CASE REPORTS
abscess or a malignant tumor because of the local bone reaction. X-ray guided drainage, recommended for pelvic abscesses (8, 9), and possible in presacral and spinal extradural abscesses (7), was not performed because of the equivocal radiological appearance. Surgical intervention, with resection of 15 cm of terminal ileum, led to the complete resolution of symptoms. In this case, the underlying cause of the neurological symptoms was most likely an infiltration of the right lumbosacral nerve caused by edema and inflammation of the terminal ileum in the vicinity of the presacral space. The occurrence of unexplained lumbosacral neurological symptoms or signs in a patient with Crohn’s disease necessitates an MRI or CT scan of the lumbosacral spine and pelvis to detect potential neurological compression. Reprint requests and correspondence: Jean-Franc¸ois Demarquay, M.D., Department of Gastroenterology, Hoˆpital Universitaire de l’Archet II, 151 route Saint-Antoine de Ginestie`re, 06202 Nice Cedex 3, France. REFERENCES 1. Keighley MRB, Eastwood D, Ambrose NS, et al. Incidence and microbiology of abdominal and pelvic abscess in Crohn’s disease. Gastroenterology 1982;83:1271–5. 2. Lamport RD, Cheskin LJ, Moscatello SA, et al. Sterile epidural and bilateral psoas abscesses in a patient with Crohn’s disease. Am J Gastroenterol 1994;89:1086 –9. 3. Aitken RJ, Wright JP, Bok A, et al. Crohn’s disease precipitating a spinal extra-dural abscess and paraplegia. Br J Surg 1986;73:1004 –5. 4. Piontek M, Hengels KJ, Hefter H, et al. Spinal abscess and bacterial meningitis in Crohn’s disease. Dig Dis Sci 1992;37:1131–5. 5. Hershkowitz S, Link R, Ravden M, et al. Spinal empyema in Crohn’s disease. J Clin Gastroenterol 1990;12:67–9. 6. Sacher M, Go¨pfrich H, Hochberger O. Crohn’s disease penetrating into the spinal canal. Acta Paediatr Scand 1989;78:647–9. 7. Cassagnou M, Notteghem B, Cotton A, et al. Drainage percutane´ d’un abce`s pre´sacre´ et rachidien extra-dural compliquant une maladie de Crohn. Gastroenterol Clin Biol 1996;20:689 –92. 8. Casola G, van Sonnenberg E, Neff CC, et al. Abscesses in Crohn’s disease: Percutaneous drainage. Radiology 1987;163:19 –22. 9. Safrit HD, Mauro MA, Jaques PF. Percutaneous abscess drainage in Crohn’s disease. AJR 1987;148:859 – 62.
LANSOPRAZOLE: A CAUSE OF SECRETORY DIARRHEA John S. Goff, M.D., F.A.C.G., F.A.C.P. University of Colorado Health Sciences Center, Rocky Mountain Gastroenterology Associates, Department of Gastroenterology, Centura-St. Anthony Central Hospital, Denver, Colorado A case of severe diarrhea and hypergastrinemia after 6 wk of lansoprazole therapy is presented. This represents the only fully evaluated report of severe diarrhea due to lansoprazole and comes to the interesting conclusion that it was a secretory diarrhea likely due to lansoprazole and not a gastrinoma or another cause. (Am J Gastroenterol 1998;93: 2298 –2299. © 1998 by Am. Coll. of Gastroenterology) INTRODUCTION
Lansoprazole (Prevacid, TAP Pharmaceuticals, Deerfield, IL) is a proton pump inhibitor that is frequently used for the treatment of Received Apr. 7, 1998; accepted July 1, 1998.
AJG – Vol. 93, No. 11, 1998 acid-peptic disease, including the long term treatment of gastroesophageal reflux disease. The reported incidence of diarrhea in a group of 1457 patients treated with lansoprazole was 3.6% (1). The cause of this diarrhea is not established. This report describes a patient with profound diarrhea secondary to lansoprazole, which proved to be secretory in nature. CASE REPORT
The patient is an 81-yr-old woman who presented for evaluation in May 1997. Her complaints at that time were nausea, vomiting, weight loss of about 25 lb, and some constipation. She had moderate chronic obstructive lung disease and was on home oxygen. Her medications included Atrovent, Maxair, and Asthmacort inhalers. She reported no drug allergies. Her past medical history included a hysterectomy in 1954, an appendectomy as a child, a cholecystectomy in 1981, and two limited sigmoid resections for diverticular disease (1980 and 1988). The patient denied any diarrhea. She had not smoked since 1985 and used very little alcohol. There was no family history of medical problems. Esophagogastroduodenoscopy revealed severe erosive esophagitis, which was treated with Prevacid (lansoprazole) 30 mg daily. Biopsies at that time showed normal small bowel histology. In June 1997, the lansoprazole dose was increased to two 30-mg capsules/day. During this time the patient lost a documented 13 lb from May to August 1997. Colonoscopy was done; the only finding was diverticula. In June, the patient was started on pancreatic enzymes (Creon 10) to exclude pancreatic insufficiency. There was no clinical response to the oral pancreatic enzymes. A CT scan disclosed vascular calcifications without other abnormality. Her nausea and vomiting had improved, so her dose of lansoprazole was decreased to 30 mg/day. In early August the patient started having diarrhea. A small bowel series showed only slow transit through an otherwise normal small bowel. Stools were collected for WBC, routine bacterial pathogens, parasites, Yersinia, and cryptosporidia. These studies were all negative. The patient was placed on loperamide (up to 2 mg per day), with no improvement. TSH was normal at 0.43. At this time her WBC was 5500, with a hematocrit of 37 and 400,000 platelets. The WBC differential was normal. Her albumin was 4.1 g/dl. All liver enzymes, bilirubin, electrolytes, BUN, and creatinine were in the normal range. By early September, the patient was having very profuse diarrhea despite stopping all milk products, increasing her pancreatic enzymes, and continuing with loperamide. She ultimately needed to be admitted to the hospital because of severe dehydration and hypokalemia. Her admitting BUN was 139 with a creatinine of 3 and a potassium of 2.9. Her albumin had fallen to 2.6. Her other laboratory tests were unchanged from before. Her renal function returned to normal in a couple of days, after vigorous intravenous hydration, and her potassium was also corrected. The patient was placed on a nothing-by-mouth status except sips of water for her oral medication, which was lansoprazole and diphenoxylate. The lansoprazole was continued because it was mistakenly thought that she had already been tried off this medication without any change in her diarrhea. A central catheter was placed for total parenteral nutrition. The patient continued to have stool outputs of 250 –1000 ml/day, despite no oral intake.
AJG – November 1998
BRIEF CASE REPORTS
Urine collected for 5HIAA was normal at 5.7 mg/dl. A fasting gastrin (on Prevacid, 30 mg/day) was 450 pg/ml. Subsequently, a secretin test was performed. The baseline gastrin was now 200 pg/dl. Gastrin levels after 100 CU (2 CU/kg) of secretin (Ferring Laboratories, Suffern, NY) intravenously were 203, 199, 204, and 195 pg/ml at 1, 5, 10, and 30 min, respectively. A CT scan of the abdomen was performed, which showed no significant abnormalities and was unchanged from one performed several months earlier. An octreoscan was done by the nuclear medicine department, which was negative for focal uptake up to 24 hours after administration of the radiolabeled octreotide. The patient was placed on intravenous famotidine to control her acid reflux problem and the lansoprazole was discontinued. Within 48 h, her rectal output diminished to zero while on no oral intake. She was started back on oral feedings with a slow advancement to a normal diet over the next 2 wk. There was no resumption of diarrhea and she has remained free of diarrhea as an outpatient. DISCUSSION
This patient had about a 6-wk delay in onset of her diarrhea after starting lansoprazole for her acid reflux. Her hospital course confirmed the existence of a secretory diarrhea given the continued significant rectal output despite no oral intake. These circumstances presented an interesting and deceptive cause of secretory diarrhea. With an elevated serum gastrin level and secretory diarrhea, the diagnostic focus became gastrinoma (Zollinger-Ellison syndrome). That diagnosis was excluded in this patient by having a negative secretin test, a negative octreotide scan, and resolution of the diarrhea after discontinuation of lansoprazole. The elevated gastrin levels were undoubtedly due to the effect of lansoprazole, which is a well described phenomenon (2– 4). Diarrhea is reported by the manufacturers of lansoprazole (TAP Holdings, Deerfield, IL) to occur in up to 7.6% of patients on high dose (60 mg) lansoprazole, but it is usually mild and self-limited, and rarely leads to drug stoppage (5). In Freston’s review of the safety of lansoprazole, he reported an incidence of diarrhea of 2.9% in 2729 patients (4). There appear to be no reports in the literature regarding evaluation of these patients for causation. The data presented are strongly suggestive that this patient’s secretory diarrhea was related to lansoprazole, but without rechallenge it cannot be proven. Reprint requests and correspondence: John S. Goff, M.D., 16225 W. 74th Drive, Arvada, CO 80007.
REFERENCES 1. Physicians Desk Reference, 50th ed. Montvale, NJ: Medical Economics Co., 1996. 2. Iwao T, Toyonaga A, Kuboyama S, et al. Effects of omeprazole and lansoprazole on the fasting and postprandial serum gastrin and serum pepsinogen A and C. Hepatogastroenterology 1995;42:677– 82. 3. Sontag SJ, Kogut DG, Fleischmann R, et al. Lansoprazole prevents recurrence of erosive reflux esophagitis previously resistant to H2-RA therapy. Am J Gastroenterol 1996;911:758 – 65. 4. Freston JW. Long-term acid control and proton pump inhibitors: Interactions and safety issues in perspective. Am J Gastroenterol 1997;92: 51–5S. 5. Personal communication, TAP Holdings Inc.
2299
PRIMARY AMYLOIDOSIS OF THE MESENTERY AND THE RETROPERITONEUM PRESENTING WITH LYMPHEDEMA Ulrich Halm, M.D., Frieder Berr, M.D., Andrea Tannapfel, M.D., Rainer Klo¨ppel, M.D., Roger Secknus, M.D., and Joachim Mo¨ssner, M.D. Department of Internal Medicine II, Institute of Pathology, and Department of Diagnostic Radiology, University of Leipzig, Leipzig, Germany
We report the case of a 57-yr-old woman presenting with moderate weight loss, abdominal distension, and lymphedema of the legs and vulva. Computed tomography of the abdomen revealed massive thickening of the rectal wall, mesentery, and retroperitoneum. Primary amyloidosis was diagnosed by immunohistochemistry from the rectum and duodenum. To our knowledge, lymphedema due to primary amyloidosis has not yet been reported. The diagnosis should be presumed in the case of retroperitoneal thickening and lymphedema and can be established by immunohistochemistry. (Am J Gastroenterol 1998;93:2299 –2300. © 1998 by Am. Coll. of Gastroenterology) INTRODUCTION
Amyloidosis is caused by a heterogeneous group of diseases, which have extracellular deposition of twisted b-pleated proteins as a common feature (1). In primary amyloidosis or amyloid light chain amyloidosis (AL), the amyloid protein consists of fragments of light chains (l or k) produced by a plasma cell clone (1, 2). We report on a patient with primary amyloidosis (AL) and an extraordinary involvement of the mesentery and the retroperitoneum. CASE REPORT
A 57-yr-old woman was admitted to the hospital with a 1-yr history of fatigue and the feeling of progressive abdominal tension. Until these symptoms began, she had an unremarkable medical history and took no medication. Six months before admission she developed edema of the legs and vulva, which were marked in upright position. She had experienced weight loss of 7 kg over 1 yr but had no further weight reduction thereafter. She had normal appetite and no diarrhea. Physical examination revealed normal vital signs, edema of both legs including the toes, hips, and vulva, and a diffuse, tender induration of the abdomen. The lungs, heart, and rectal examination were normal. Routine laboratory tests showed a white blood count of 10,900/ml with a mild left shift and a slightly elevated serum alkaline phosphatase (217 units/L, normal 60 –190). Other laboratory findings including gGT, ALT, AST, hemoglobin, bilirubin, serum creatinine, and electrolytes were normal, and there was no M-gradient in the serum electrophoresis. A D-Xylose-test was pathological (0.7 mmol/L, normal .2 mmol/L), indicating malabsorption in the duodenum and jejunum. Urinalysis revealed proteinuria of 0.5 g/day but no evidence of light chains in immunofixation electrophoresis. A CT scan of the abdomen and the pelvis showed an extreme, diffuse thickening of the rectal wall and the retroperitoneal and Received Feb. 6, 1998; accepted July 1, 1998.