This series of patients with nonspecific small bowel ulcers manifested specific endoscopic features and benign clinical courses, which is different from previous case reports [2, 3, 10, 18]. Our series was characterized by multiple small ulcers located at the end of the ileum or the ileocecal valve, with mild gastrointestinal symptoms but no bleeding, perforation, or stenosis. The lesions were either self-limiting or persistent. No complications or deterioration were seen in the patients with persistent ulcers, although no special treatment was applied, including surgery.
The average age at the initial diagnosis was 54.7 years, consistent with previous reports . Nonspecific small bowel ulcers affect both sexes, and different reports have shown different representation in the two sexes [2, 18]. In our study, there was a clear predominance of males. Most studies have reported long symptomatic histories before the initial diagnosis. Similarly, in the present study, most patients had suffered from gastrointestinal symptoms for 0.5-11 years. Concomitant lesions, including esophageal leiomyoma, erosions of the stomach, and colon polyps, were present but there was no evidence that they correlated with the nonspecific small bowel ulcers.
Indications for further investigation in this study were diarrhea and/or abdominal pain/discomfort, which are different from the indications reported in the literature. Complications resulting from nonspecific small bowel ulcers have often been reported previously, including bleeding, anemia, intestinal obstruction, and perforation [1, 2, 4, 19]. Interestingly, one patient was accidentally diagnosed by ileoscopy during surveillance for colon polyps, and no gastrointestinal symptoms appeared later. Our case series showed good prognoses. Nonspecific small bowel ulcers were persistent in four of the patients presented here, self-limiting in two, and recurrent in only one, but these recurrent ulcers healed again without intervention. Surgical resection is the most common strategy used to treat nonspecific small bowel ulcers, according to the literature [3, 4, 18]. However, none of our patients showed complications such as perforation, obstruction, bleeding, anemia, etc., so no surgical intervention was required during the 5-9 year observation period. Because no patient underwent surgery, we cannot assess the recurrence rate after surgery in our study. In the literature, some authors describe frequent ulcer recurrence, even after surgery , whereas there are also reports that surgery was generally curative .
In the literature, three distinct syndromes of nonspecific small bowel ulcers have been proposed [3, 18]: 1, isolated nonspecific ulcers [1, 10], which are usually located in the distal ileum and are identified by laparotomy for intestinal obstruction and bleeding, etc.; 2, idiopathic chronic ulcerative enteritis, which manifests with fever, diarrhea, or mucosal atrophy and mimics celiac disease [7, 11, 20] (other terms are also used to describe this condition, such as nongranulomatous chronic idiopathic enterocolitis , chronic ulcerative nongranulomatous jejunoileitis or idiopathic chronic ulcerative enteritis , and chronic nonspecific ulcerative duodenojejunoileitis ); 3, cryptogenetic multifocal ulcerous stenosing enteritis [12, 18], usually presenting with more than 20 ulcers in the small bowel and multiple ulcerative obstructions, which often recur after surgery. All these conditions are considered to be nonspecific and no possible causes have been confirmed, although some possible etiologies have been explored, including vasculitis . When all the clinicopathological features are considered, none of the patients in this study can be referred to any of the syndromes cited above. Our case series is similar to the case reported by Borsch et al. , which was assumed to be IBS before a diagnosis of nonspecific small bowel ulcers was made. As stated above, the symptoms of most of the patients in this study were compatible with the definition of functional bowel disorders and there were no "alarm" symptoms indicating a need for extensive exploration according to the Rome III criteria for functional gastrointestinal disorders . Most of our patients were assumed to have functional bowel disorders, such as IBS, before ileoscopy. As we know, terminal ileum intubations and subsequent ileoscopy are not routinely performed in colonoscopy practice when no inflammatory bowel disease or ileal lesions are suggested before the colonoscopy, Recent data have shown that the terminal ileum intubation rate is low (17%-21%) during colonoscopy in various practice settings . Actually, ileoscopy is technically feasible and adds only a couple of minutes to the duration of the procedure . Some authors have suggested that routine terminal ileum intubation should be applied for patients with abdominal pain, diarrhea, or anemia to improve the diagnostic yield [25–27]. This study seems to further justify the use of ileoscopy during colonoscopy for patients suspected of functional bowel disorders, including IBS.
It has often been stated in the literature that nonspecific small bowel ulcers are difficult to diagnose preoperatively [1–3]. This preconception is changing, as shown in our report and those of others [5, 11], and can be attributed to doctors' increasing awareness of small bowel disease and the substantial progress in visualizing the small intestine by capsule endoscopy and double or single balloon enteroscopy [13, 28, 29], together with the broad application of ileoscopy during colonoscopy [25, 30]. Hence, nonspecific small bowel ulcers can be identified without recourse to surgery.
The causes underlying nonspecific small bowel ulcers remain obscure. Because ileal ulcer is found in various diseases, including infections, neoplasm, inflammatory bowel disease, etc., it is important to explore these possibilities before a diagnosis of nonspecific small bowel ulcers is made. In China, Crohn's disease and intestinal tuberculosis are the most common diseases involving the ileum, and both show granuloma on microscopic evaluation. However, it is always difficult to establish a definite diagnosis of Crohn's disease [31, 32]. Therefore, in clinical practice, much attention has been paid to the differentiation of Crohn's disease and intestinal tuberculosis from ileal ulcers. All the patients reported here underwent extensive investigations to rule out Crohn's disease, intestinal tuberculosis, and other possible causes. No underlying disease appeared during the long-term follow-up, which strengthened the initial diagnosis of nonspecific small bowel ulcers. The negative effects of the long-term use of 5-ASA in two patients may reflect the great differences between nonspecific small bowel ulcers and inflammatory bowel disease.
NSAID enteropathy [10, 33, 34] has been recognized in recent decades as an important cause of small bowel ulcers. Some previously assumed nonspecific small bowel ulcers may have been attributable to chronic NSAID use, especially those with diaphragm changes in the small intestine. In our study, we confirmed that all the patients had not used NSAIDs before their initial diagnosis. Two patients became chronic aspirin users during follow-up. Their final results showed that, in this study, chronic aspirin use did not interfere with the clinical course.
Other than surgery, no effective medication for nonspecific small bowel ulcers has so far been validated . In our study, a variety of empirical medicines were used, including 5-ASA, probiotics, and herbal medicines. Probiotics seemed partly effective in two patients with diarrhea, but they had no direct effect on the ulcer itself. Long-term 5-ASA therapies were prescribed in two patients. However 5-ASA failed in both. Previous data have shown that corticosteroid therapy improves both the symptoms and histology of a certain group of nonspecific small bowel ulcers, called "nongranulomatous chronic idiopathic enterocolitis" , whereas another study showed that prednisolone did not benefit these patients . In our study, no patient received corticosteroid therapy for fear that it might increase the risk of perforation. Generally speaking, no medication was found to be effective for nonspecific small bowel ulcers in this study, although some ulcers healed.