Introduction Video capsule endoscopy (VCE) is now a routine procedure for investigating the small bowel. It is not apparent whether the yield and utility of the procedure is the same when used in routine practise, compared to earlier studies1–3. The yield of the procedure is important when planning local provision of enteroscopy services.
Methods We retrospectively audited 162 consecutive procedures in a single UK centre. In our analysis we defined diagnostic yield as confirming or establishing a diagnosis or definitively refuting a diagnosis previously suggested by other investigations. Patients were fasted for at least 8 hours and not given bowel preparation. Equipment was from Olympus UK. Studies were initially read by a non-medically qualified healthcare worker before approval by the senior author, after a learning phase when the initial 50 examinations were dual read. Patency capsules were used in patients with obstructive type symptoms or known Crohn’s disease.
Results The overall diagnostic yield was 35% which is slightly lower but comparable to early studies3. The yield for overt GI bleeding (OGIB) was 43%, lower than the often reported yield of approximately 60%. Of the 162 patients 14 were referred on for enteroscopy (8.6%) which is lower than many early reports. Of the 14 referred, 10 were for anaemia and 4 for OGIB. 11 had successful therapy with balloon enteroscopy and 1 was diagnosed and a tattoo placed for lesion resection. In 9/162 studies (5.5%) pathology was found within reach of gastroscopy or colonoscopy and had been missed on previous examination. One retained capsule required surgical removal (0.6%). Four capsules were retained in the stomach (2.4%) and 30 were seen to enter the small bowel but did not reach the caecum (18.5%). 16 studies were deemed inadequate due to food debris (9.8%). 2 failed due to equipment malfunction. Total incomplete examination rate was therefore 30%. Average small bowel transit time was 281 minutes (range 90–710). Anaemia (n = 87) was the most common indication with a diagnostic yield of 33%. OGIB (n = 37) was the next most common indication yielding 43%. Query Crohn’s (n = 22) had a yield of 14%. Malabsorption query Coeliac (n = 5) yielded 40%. Unsurprisingly known Coeliac (n = 2) gave a complete yield of 100%. Abnormal Imaging (n = 4) returned a 75% yield. Weight loss (n = 1) gave 0% yield. Abdo pain (n = 4) had a 25% yield.
Conclusion In routine practise VCE is a useful technique to diagnose patients with small bowel disorders. Yield seems slightly lower than previous studies but is still significant. Incomplete examination rate was high, mostly due to inadequate battery life. Only 8.6% of patients required enteroscopy which has implications for service provision.
Disclosure of Interest None Declared.
Triester et al. 2005
Marmo et al. 2005
Hartmann et al.2005