Does pan‐enteric capsule endoscopy obviate magnetic resonance entero‐colonography in suspected Crohn's disease?

Pan‐enteric capsule (PCE) endoscopy has shown promise in assessing small bowel and colonic mucosa, especially in proximal bowel segments, with greater reach of visualization in the small bowel. ¹ , ² MR entero‐colography (MREC) demonstrates CD related lesions in the small bowel and has shown a moderate to high sensitivity for detecting CD in the colon. ³ Whether these modalities can replace ileocolonoscopy (IC) in the future, as the initial diagnostic modality in selected patients with suspected CD, is unknown. Furthermore, the preferable of the two modalities in this subgroup of patients has yet to be defined.

A recent prospective Danish study by Brodersen and colleagues aimed to compare the diagnostic accuracy and patient compliance between PCE and MREC for the initial diagnosis of ileocolonic CD in 153 patients. ⁴ Both modalities were compared to the gold standard of IC. Sensitivity and specificity for diagnosing CD in the terminal ileum/colon were 67.9%, 76.3% and 87.5%, 87.8% for MREC and PCE, respectively. Terminal ileum sensitivity was numerically higher for PCE, compared to MREC (96.6% vs. 76.9%), and more significantly so for the colon 75% versus 27% (p < 0.001). There was no significant difference in specificity between the modalities (p > 0.08). On a separate analysis, the combined diagnostic yield proximal to the terminal ileum was higher for PCE than for MREC (p < 0.01). There was no difference between MREC and PCE in terms of physical and psychological discomfort, while IC without propofol sedation was considerably less tolerated. The authors conclude that PCE has a high accuracy for diagnosing CD compared to MREC, especially in the colon, and suggested that PCE could be an alternative to IC as first line modality in selected patients with suspected Crohn's disease.

This pivotal study underscores the increased diagnostic yield of PCE compared to MREC with regard to new‐onset CD, especially for colonic and proximal small bowel disease. Moreover, it brings about several clinical questions, which ought to be discussed further. The main strength of the two alternative modalities is that while being less invasive, they add data as to proximal small bowel involvement, which is essential for defining mode of treatment in 15%–20% of CD patients with upper GI involvement. Sensitivity of PCE was superior to MRE for enteric inflammation in the proximal small bowel (97% vs. 71%, p = 0.021). This goes in line with previous studies, some of which show that the correlation between the capsule Lewis score and the MRI MaRIA score is not good in mild mucosal disease as expressed by the Lewis score, and improves significantly as the Lewis score increases. ⁵ , ⁶ , ⁷ An Australian study compared IC and CE in 47 Crohn's patients undergoing both procedures on the same day. ⁸ Correlation (r) between total SES‐CD scores was strongest in the terminal ileum (r = 0.77, P < 0.001). An additional study ⁶ compared the MaRIA score and Lewis index, demonstrating moderate correlation in terminal ileum (r = 0.6, p = 0.002) but not in jejunum (r = −0.3, p = 0.6), or ileum (r = 0.01, p = 0.9).

As PCE was preceded by MREC, in order to exclude patients with small bowel stenosis, patency capsule was not performed in Brodersen's study. Had it been performed, more patients with stenotic elements would have been able to undergo the PCE. In a previous study comparing MREC to PCE, agile patency capsule was performed in 10 patients following MRE findings suggesting stenotic complications. In all cases, the agile patency capsule was excreted intact, and CE was performed thereafter without complications. ⁹ , ¹⁰ In addition, experienced gastroenterologists at each participating center performed IC and CE. Variations between multiple observers were not accounted for in the current study. Although previous reviews have demonstrated relative inter‐observer agreement when using the same standardized scores, ⁶ this ought to be addressed in corroborating studies.

Finally, a notable finding of the current study is that no difference was found in the overall diagnostic performance of PCE when stratifying for bowel cleansing. Overall negative predictive value of a complete PCE was 95%. A reason for this could be extensive inflammation/multiple extensive ulcers in patients with CD detected irrespective of the image quality. This is an important observation and in line with a limited number of studies examining the feasibility of low‐volume bowel cleansing regimens. ¹¹ , ¹²

To conclude, PCE seems preferable to MREC in CD diagnosis, especially with regard to proximal small bowel and colonic early onset inflammation (Figure 1). It can probably be performed using a minimal preparation regimen and in cases of risk for stenosis, it can probably be safely performed following an intact excretion of a patency‐capsule. Corroborating studies would demonstrate whether this less invasive modality with a high NPV and high yield for proximal disease, could obviate diagnostic colonoscopy with histological diagnosis.

FIGURE 1.

Suggested algorithm for patients with suspected Crohn's disease. CRP, C‐reactive protein; IC, ileocolonoscopy; PCE, pan‐enteric capsule

DATA AVAILABILITY STATEMENT

Data sharing is not applicable to this article as no new data were created or analyzed in this study.