Non-steroidal anti-inflammatory drug-induced enteropathy as a major risk factor for small bowel bleeding

Background: Small bowel (SB) bleeding has been known to account for 5% of all gastrointestinal (GI) bleeding cases and 80% of obscure GI bleeding cases. Although angioectasia is the common etiology of SB bleeding, non-steroidal anti-inflammatory drug (NSAID)-induced SB lesion is also reported as a major cause in previous studies from the Eastern countries. Herein, we assessed the frequency of NSAID-induced SB lesion in Korean patients with obscure GI bleeding. Methods: We retrospectively analyzed medical records of all consecutive patients aged ≥18 years who underwent capsule endoscopy from March 2018 to February 2019 at Ulsan University Hospital and Kosin University Gospel Hospital. Results: Of the 83 subjects (age, mean ± standard deviation: 59 ± 18 years; age range: 18–84 years; men, n=52; women, n=31), 55 (66.2%) had a clear bloody stool and 28 (33.8%) had a normal stool, but all had iron deficiency anemia. A significantly higher frequency (40 of 51) of ulcerative/erosive lesions was observed in patients with inactive bleeding but visible SB lesions than other causes, and as a result, NSAID-induced enteropathy accounted for 41.7% (25 of 60) of all SB bleeding cases. Conclusions: Contrary to findings of previous studies from the Western countries, ulcerative/erosive lesions were found to have higher occurrence than angioectasia in this study, with an implication of NSAIDs in the etiology of obscure GI bleeding. Aggressive small intestine examination is required for patients with unexplained GI bleeding.

inflammatory drug (NSAID)-induced SB lesion is the major cause reported in Japanese studies. 9,10 NSAIDs are widely prescribed in most clinical conditions, but these drugs are well known to cause GI complications. 11 Although various studies have reported an association between NSAID intake and GI adverse effects, 12,13 the role of NSAIDs in small intestinal bleeding remains to be elucidated. Therefore, we aimed to assess the frequency of NSAID-induced SB lesions in Korean patients with obscure GI bleeding who underwent capsule endoscopy.

Study population
We retrospectively evaluated medical records of all consecutive patients aged ≥18 years with obscure GI bleeding who underwent capsule endoscopy from March 2018 to February 2019 at our two medical institutions, Ulsan University Hospital and Kosin University Gospel Hospital. Indication for the test included bloody stool in patients with unidentified bleeding lesions during diagnostic evaluations, including the upper GI endoscopy, colonoscopy, and abdominal computed tomography (CT) in the last 3 months. 5 We also included patients who had no visible bloody stool, showed persistent or repeated exacerbation of iron deficiency anemia despite iron supplement for >6 months, and no hemorrhagic lesions identified during the above screening methods in the last 3 months. 14 Rebleeding was defined as recurrent anemia (≥2 g/dL decrease in hemoglobin level), overt melena/ hematochezia or occult GI bleeding during the follow-up period. 15 This study was approved by the Institutional Review Board received 40 mg oral simethicone before the procedure to prevent air-bubble formation. 16 Moreover, we used 2L polyethylene glycol solution before examination for the improved image quality of SB. 17 Five gastroenterologists with extensive GI endoscopic experience reviewed all capsule video images.
All images were extensively discussed to reach a diagnosis.
Ulcerative lesions were defined as mucosal penetrating lesions with diameters > 5 mm. 15 An erosion was a roundish area of mucosal disruption smaller than 5 mm of diameters. 18 The etiology of ulcerated lesions was determined based on clinical information and endoscopic findings. We

Statistical analysis
Continuous variables were compared using Student's t-test, and categorical variables were analyzed with chi-square or Fisher's exact test. A two-tailed p-value <0.05 was considered statistically significant. All statistical analyses were performed using the SPSS statistical package for Windows, Version 24.0 (SPSS Inc., Chicago, IL, USA).

Sample analysis
We assessed 83 subjects who underwent capsule endoscopy during the study period (age, mean ± standard deviation: 59 ± 18 years; age range: 18-84 years; men, n = 52; women, n = 31; Table 1). Of these, 55 (66.2%) patients had a clear bloody stool and 28 (33.8%) had a normal stool, all with iron deficiency anemia ( Figure 1). Among 37 (44.6%) patients with a history of low-dose aspirin or NSAID medication, only three treated with selective cyclooxygenase (COX)-2 inhibitors showed a normal stool. The majority of patients with active bleeding during endoscopy had angioectasia (8 of 9 patients). Fecal occult blood test, performed in patients with normal stools (16 of 28), was negative in ulcerative/erosive lesions was observed in patients with inactive bleeding but visible SB lesions than other causes, among whom 62.5% (25 of 40) had a history of low-dose aspirin or NSAID medication.
As a result, NSAID-induced enteropathy accounted for 41.7% (25 of 60) of all SB bleeding cases. Five cases of previously undiagnosed Crohn's disease were identified. Other uncommon etiologies associated with unknown GI bleeding included radiation ileitis (n = 3) and SB polyps (n = 3).

Evaluation of rebleeding and treatment
During capsule endoscopy, 58 (69.9%) patients had anemia with hemoglobin <10 g/dL. SB bleeding was predominant in >50% (17 of 28) of patients with normal stool. All patients with SB bleeding due to angioectasia showed bloody stools, while ulcerative lesions were prevalent in patients with normal stool. Rebleeding occurred in 9 patients during the 6-month follow-up period. All patients with Crohn's disease (n = 5) who presented with persistent iron deficiency anemia required specific biologic treatments. Among two patients with radiation ileitis, one underwent surgical treatment due to recurrent bloody stool, whereas the other patient with persistent iron deficiency anemia refused an SB resection. A patient with a rebleeding episode from angioectasia received angiographic embolization. In our patient samples, a 65-year-old woman, who received NSAIDs for treating fibromyalgia for >2 years, presented with an ulcerative lesion. The patient had unresolved anemia despite adequate iron administration and showed recurrent abdominal pain, which lasted for >6 months. An abdominal CT revealed multiple strictures of the SB. During the procedure, when the capsule failed to pass through the site, the retained capsule was retrieved through a surgical resection, which identified five sites of stenosis in the distal ileum. Postoperative 3-month follow-up revealed no recurrence of anemia and abdominal pain (Figure 2).

Discussion
A diagnostic yield of 57-62% have been reported with capsule endoscopy for unexplained GI bleeding, 19 with the most common diagnosis of angioectasia (50%), followed by ulcers (26.8%) and tumors (8.8%), as suggested by a recent systematic literature review. 14 The present study revealed that the detection rate of SB bleeding and lesion in capsule endoscopy was 72.3% (60 of 83 patients) for obscure GI bleeding, in which, NSAID-induced enteropathy accounted for 41.7% (25 of 60) of all SB bleeding cases. Contrary to previous reports from the Western countries, we observed a higher occurrence rate of ulcerative/erosive lesions than angioectasia in our Korean patients with obscure GI bleeding, showing the implication of low-dose aspirin or NSAID medications in the disease etiology.
NSAIDs are frequently used anti-inflammatory analgesic agents that represent 7.7% of worldwide prescriptions, of which 90% are prescribed to elderly (>65 years) patients. 20 The mechanism of NSAID-induced enteropathy is supposed to be mediated through COX inhibition. 21 Administration of low-dose aspirin (irreversible nonselective COX inhibitor) is also associated with a SB mucosal injuries; taking of low-dose enteric-coated aspirin at 100 mg in healthy volunteers reported large erosions or ulcers in 60% of these participants. 22,23 In the present study, a history of low-dose aspirin or NSAID medications was common in patients with obscure GI bleeding (44.6%), showing a higher frequency of SB ulcerative lesions (68%) than that of other sources of lesions (angioectasia, n = 6; upper or lower GI bleeding, n = 6).
Prostaglandins (PG) play an important role in regulating GI blood flow and mucus production; therefore, NSAID-induced suppression of PG production has been implicated in small intestinal damage. 24,25 Previously, COX-1 inhibition was regarded to be dominantly related with GI mucosal injuries. However, in a recent animal model study, small intestinal damage developed only when both COX-1 and COX-2 were prohibited. 26 This result indicates that COX-2-derived PGs also play an important role in the maintenance of tissue integrity and repairing of mucosal injury. However, clinical research has shown conflicting results. Several studies showed an improved GI safety profile with selective COX-2 inhibitors compared to non-selective NSAIDs, 27, 28 while others studies indicated no significant differences in SB injuries between these NSAIDs. 29,30 In this study, among 37 patients with a history of low-dose aspirin or NSAID medications, all three patients treated with the selective COX-2 inhibitors showed a normal stool, suggesting favorable GI safety outcomes of selective COX-2 inhibitor therapy. Considering selective COX-2 inhibitors are not completely safe for the SB, further long-term studies with a larger sample size are warranted to establish the safety profile of the drug in the SB.
Furthermore, the impact of capsule endoscopy on clinical outcomes remains controversial despite reports of SB mucosal damage in 70% of patients taking NSAIDs, 31,32 because it remains unclear whether SB mucosal injuries contribute to significant bleeding. 33 Although patients with NSAIDinduced SB injury show low frequency of severe bleeding in the SB, 34 rebleeding rates of 21-35% has been reported in patients with SB ulcerations during a mean follow-up period of 17.1-29.7 months. 15,35 These reports suggest the clinical implication of SB ulcers, which cannot be ignored.
The most effective method of preventing NSAID-induced enteropathy is discontinuation of NSAIDs if it is possible. 12 Until recently, there are no definite strategy to prevent NSAID-induced enteropathy. 13,   Categorical and continuous variables are presented as number (%) and mean ± SD, respectively.