Introduction Capsule endoscopy (CE) is the first line of investigation for examining the small bowel (SB) mucosa. While standard preparation (SP) is more convenient for patients, mucosal visibility may deteriorate within the distal SB. Recent meta-analyses suggest that bowel-cleansing agents (BCA) can improve small bowel image quality (IQ). The influence of BCA compared to SP on reader confidence levels (RCL) when excluding clinically significant findings (CSF) has not been examined.
Aim To compare RCL when excluding CSF and assessing IQ during reading following SP or BCA prior to CE.
Methodology We performed a retrospective analysis of SB capsule images of 100 consecutive patients who underwent a complete CE examination at our institution from Oct 2012 - Mar 2013. Patients had SP (intake of clear liquids for 18 h and 12 h fasting prior to the procedure without BCA) or BCA (2l of polyethylene glycol (PEG) or magnesium citrate (MC) in addition to SP). The participants’ demographic and clinical data were collected and SB transit time (SBTT) calculated. A four point scale was used to assess IQ (grade 1 = <80% of mucosa visible ± excessive debris ± severely reduced brightness to grade 4 = ≥ 90% of mucosa visible ± mild debris± mildly reduced brightness). The SBTT was divided into quartiles (Q1-Q4) by time and the IQ score, RCL and number of CSF for each quartile were determined by a gastroenterologist experienced in CE, blinded to the preparation. Procedures were examined in randomised order.
Results 49 (49%) patients had SP (group A) while 51 (51%) had one of the BCA (39% had PEG and 61% had MC, group B). There was no significant difference in age (p = 0.87), sex (p = 0.57), indication (p = 0.25) and SBTT (group A: 264 ± 112 mins vs. group B: 233 ± 100 mins, p = 0.14) between groups. For each quartile, IQ scores were significantly higher for group B than A except in Q1 (Q1: 3.7 ± 0.7 vs. 3.5 ± 0.6, p = 0.06; Q2: 3.6 ± 0.5 vs. 3.1 ± 0.6, p < 0.0001; Q3: 3.2 ± 0.6 vs. 2.3 ± 0.7, p < 0.0001; Q4: 2.8 ± 0.5 vs. 1.9 ± 0.8, p < 0.0001). There was no difference in detection of CSF between group A and B (41% vs. 51%, p = 0.33, respectively). For each quartile, RCL for excluding significant findings were significantly higher for group B than A except in Q1 (Q1: 100% vs. 96%, p = 0.06; Q2: 96% vs. 73%, p < 0.0001; Q3: 88% vs. 33%, p < 0.0001; Q4: 77% vs. 20%, p < 0.0001). There was no significant difference in IQ or RCL between PEG or MC. 3 procedures (all SP) were considered unsatisfactory for IQ with recommendation to repeat these after BCA.
Conclusion BCA pre-capsule endoscopy significantly improve small bowel IQ and RCL when excluding CSF. Bowel cleansing appears to be an important parameter for optimising the qualitive aspects of CE reading.
Disclosure of Interest None Declared.