Our novel study suggests non-specific enteritis is a relatively frequent finding in capsule endoscopy practice, in our cohort 13% of capsule studies had inflammatory lesions not meeting a diagnostic threshold. While the majority settled on follow up 49%, almost 1:5 were ultimately diagnosed with Crohn’s disease. Figures which are in keeping with follow up data of patients with acute isolated active ileitis, even though all our subjects had a negative ileo-colonoscopy prior to capsule endoscopy.
An indication of suspected Crohn’s disease for capsule endoscopy was strongly associated with subsequent Crohn’s on follow up, OR 9. Not surprisingly, patients with a subsequent diagnosis of IBS were also more likely to have been referred with suspected Crohn’s disease. However, CD patients alone were more likely to have an abnormal / elevated base line Lewis Score, OR 6. Of note traditional IBD associated biomarkers, CRP and FC did not predict small bowel CD, nor did the extent or distribution of inflammatory lesions. This finding is in keeping with previous longitudinal studies of isolated terminal ileitis on ileo-colonoscopy which demonstrated a positive association with symptoms and subsequent Crohn’s development and a lack of predictive value for either biomarkers, baseline histology or family history [12, 13, 15, 16]. Unfortunately, the Lewis score alone is not a sensitive test for CD. The combination of both indication and Lewis score may be helpful in predicting subsequent CD in NSE patients and warrants further study.
The finding that more subjects without significant gastrointestinal disease on follow up were anemic may simply reflect the fact that the largest indication for CE in our practice is anemia with or without overt gastrointestinal bleeding. None had concomitant vascular small bowel lesions.
Of interest is the significant number of patients in our cohort who were ultimately diagnosed with NSAID related enteritis (14%). While our unit’s patient information and advice leaflet specifically requests patients to avoid NSAID’s for 6 weeks prior to their capsule endoscopy, and current medications are documented on the day of their procedure, the video readers were not always aware of the possibility. Only on subsequent review with a targeted medication history, including over the counter formulations, was the diagnosis established. This highlights the already identified need to interpret inflammatory lesions with a full knowledge of the patient’s clinical history [8]. Of interest is the finding that NSAID related disease on follow up was associated with older age. As such this cohort were more likely to have increased numbers of comorbidities and poly-pharmacy may have contributed to the problem.
Our study has several limitations. Firstly the majority of cases, 52% (169/ 326) with enteritis either Crohn’s or NSE, were performed on patients from outside our institution with no available follow up data. Despite this our cohort represents the largest NSE capsule study (n = 88), with a median follow up of almost 2 years. In addition, with the ready availability of device assisted enteroscopy and capsule endoscopy at our institution, a significant proportion of our population underwent additional endoscopic investigations as well as gastroenterology clinical review adding weight to our findings. On the other hand only a small proportion underwent subsequent dedicated small bowel imaging (20%). It is likely that many, particularly those with suspected Crohn’s disease, already had negative imaging, in keeping with CE referral practice and local guidelines. While there is evidence to link terminal ileal narrowing on CT scans with a subsequent diagnosis of Crohn’s disease on ileocolonoscopy [14], evidence from a recent prospective study of patients with abnormal small bowel MRE findings with a negative initial ileo-colonoscopy suggested few develop significant disease (12%) [10]. As such the added value of subsequent imaging may be low.