Gastroduodenal Crohn's disease and pancreatitis.

Isolated gastroduodenal Crohn's disease (GCD) with concomitant acute pancreatitis as the major presenting symptom is very rare. As mentioned in the clinical case study by Rao and colleagues,1 the incidence of Crohn's disease in the foregut, and even more so in an isolated gastric localization, is extremely low.2,3 The presence of granulomas is suggestive of Crohn's disease, though only 9% of patients with gastric involvement of Crohn's disease had granulomas in a study by Wright and colleagues.4 On the other hand, in a series of 72 patients diagnosed with granulomatous gastritis, described by Ectors and associates,5 37 (52%) had Crohn's disease, 18 (25%) were found to have isolated granulomatous gastritis, and only 1 case of granulomatous gastritis corresponded to the diagnosis of sarcoidosis. Helicobacter pylori infection was detected in 92% of the patients. The relationship between granulomatous gastritis and H. pylori infection has been widely discussed in the literature. Shapiro and associates6 postulated that an association between H. pylori and granulomatous gastritis cannot be ruled out if known cases of Crohn's disease and sarcoidosis are excluded. Notably, in the case reported by Rao and coworkers,1 there was no evidence of H. pylori infection. The diagnosis of gastroduodenal Crohn's disease was made on the basis of clinical presentation as well as findings from endoscopy (though duodenal involvement is not specifically mentioned or illustrated), endoscopic ultrasound, histology, and response to treatment. The patient is 50 years old, which is not a typical age of presentation for Crohn's disease. Furthermore, he is African American, and the incidence of Crohn's disease in African Americans is much lower than in whites.7 In contrast, the incidence of sarcoidosis is 3–4 times higher in the African-American population compared to that of the white population.8 Furthermore, the presence of aphthoid ulcers has been described in granulomatous gastritis.9 Considering that steroids and 6-mercaptopurine (6-MP) are also successfully used in the treatment of sarcoidosis, the diagnosis of sarcoidosis with gastric and pancreatic involvement in the patient treated by Rao and colleagues1 cannot be completely excluded, particularly as the stomach is the most frequently affected part of the gastrointestinal tract in sarcoidosis. Interestingly, a clinical case study on infliximab (Remicade, Centocor) therapy for hepatic and intestinal sarcoidosis has recently been published in this journal.10 Tumor necrosis factor-alfa (TNF-α) appears to be of importance to T helper cells, which play a crucial role in the formation of granulomas in sarcoidosis. Furthermore, a good response to anti-TNF-α therapy with infliximab in acute pancreatitis complicating active Crohn's disease has been reported.11 However, acute exacerbation of chronic relapsing pancreatitis accompanying Crohn's disease after infliximab infusion has been described as well.12 To our knowledge, there are no previously published reports on isolated gastroduodenal Crohn's disease with acute pancreatitis as a presenting manifestation. Crohn's disease presenting with pancreatitis has been described, however, in patients without preexisting symptoms of intestinal inflammation. Kugathasan and colleagues13 described 3 cases of pediatric Crohn's disease presenting with pancreatitis. In 1 of these cases, there was ileal and gastric involvement but no duodenal involvement. The other 2 cases had only distal intestinal localization of the disease. There was no appropriate explanation for the pancreatitis in any of the cases. Two similar cases, of a 17-year-old patient and a 20-year-old patient, were reported by Triantafillidis and coworkers.14 Gschwantler and associates15 described a patient presenting with extrahepatic cholestasis and an enlarged head of the pancreas. A Whipple procedure was performed because of a suspected pancreatic head carcinoma. The histologic specimen showed noncaseating granulomatous inflammation in the stomach, duodenum, and the head of the pancreas, leading to a diagnosis of Crohn's disease with gastroduodenopancreatic localization. Moreover, several cases of inflammatory bowel disease (mainly Crohn's disease) complicated by acute pancreatitis have been described in the literature.16–18 Most of these cases have been considered to be the results of adverse effects from medical treatment of the inflammatory bowel disease. Focusing on Crohn's disease of the foregut, the etiologic factors of pancreatitis that should be primarily considered include biliary lithiasis, duodenal involvement of Crohn's disease, primary sclerosing cholangitis, and drug intake (namely purines, metronidazole, mesalamine, and steroids).19–21 A substantial percentage of patients with inflammatory bowel disease show nonspecific high levels of serum lipase and amylase without clinical symptoms of pancreatitis.22,23 The height of these serum levels is related to the severity of disease. In the case presented by Rao and associates,1 there were also radiologic signs of acute pancreatitis. A study by Oishi and colleagues24 showed 6.3% abnormal pancreatic imaging in a group of 255 Crohn's disease patients. Only 13% of these patients had complaints related to pancreatitis. Assuming that there was actual acute pancreatitis in the patient treated by Rao and colleagues,1 the etiology remains obscure. Yet, acute pancreatitis as an extraintestinal manifestation of Crohn's disease has been suggested previously.15 There have also been reports on the presence of autoantibodies against exocrine pancreas in Crohn's disease suggesting an autoimmune etiology of pancreatitis in some Crohn's disease cases.25 Rao and coworkers1 initially treated their patient with prednisone, mesalamine, and omeprazole, but after the prednisone was tapered, some of the patient's symptoms recurred. Omeprazole and mesalamine were omitted, and prednisone was restarted at the original dose of 60 mg daily, tapered later on, and then given concomitantly with 100 mg daily of 6-MP, a dose that has maintained positive long-term clinical response. This is in accordance with the assumption that the approach to medical therapy of Crohn's disease of the proximal intestine is similar to that of more distal localization of the disease.3,26 The use of maintenance therapy with proton pump inhibitors appears to be advisable but will not affect the immunologic background of the disease. In summary, the case presented by Rao and colleagues1 appears to be the first published case of isolated GCD presenting with acute pancreatitis. In a case of pancreatitis presenting with Crohn's disease, the most common causes of pancreatitis mentioned above should be ruled out. If the etiology remains unclear, a pancreatic localization of Crohn's disease or pancreatitis as an extraintestinal manifestation can be considered. In both cases, the therapeutic approach should be focused on treating intestinal disease. There may be a role for anti-TNF-α therapy, but randomized controlled studies are necessary for further investigation. In the case reported by Rao and coworkers,1 sarcoidosis cannot be ruled out completely as an alternative diagnosis, particularly as the patient has remained clinically asymptomatic for more than 9 years. Although the therapeutic approaches in sarcoidosis and Crohn's disease are similar, the courses of the diseases are ultimately very different. Therefore, a definite distinction is of clinical relevance.