Introduction Capsule endoscopy (CE) relies on adequate bowel preparation to ensure good luminal views. Studies have given variable results on the efficacy of bowel preparation1,2. Our department moved from a regimen of 24 hour of clear fluids only (NP) to 24 hour of clear fluids and the addition of one sachet of polyethylene glycol (PEG1) after a retrospective audit of results3. As our results indicated PEG1 improved lesion detection compared to NP, we changed preparation to 24 hour of clear fluid combined with 2 sachets of PEG (PEG2). We present a retrospective look at our findings.
Method Patients who underwent CE from January 2016 till July 2016 were identified via Rapid capsule endoscopy software. Data collected included indication for test, total lesions detected (TL) and lesions excluding aphthous ulcers (EL). 3 independent assessors reviewed images at 20 min intervals to determine adequacy of bowel preparation. Overall adequacy was determined by majority opinion. Data from our previous analysis using the same methods was added and data comparison of lesion detection in each arm was compared by Fisher’s test chi squared analysis.
Results A total of 31 patients were identified in this timeframe but only 30 patients were included due to failure of one procedure. All patients received PEG2. Our results are shown in figure 1.
The only result which showed statistical significance was between PEG1 and PEG2 when comparing all lesions identified (p=0.045) although it appears that PEG1 was better than PEG2; and that PEG2 detected less lesions overall than NP. However, when normal examinations were discounted from all three arms and the subsequent results compared, there were statistically significant results between NP(36%) and PEG2(70%) showing PEG2 was better at detecting significant lesions.(x2 5.790,p=0.0161).
Conclusion In our study, it appears that bowel preparation with PEG is better than NP in detecting significant lesions, but that PEG2 is inferior to PEG1. This goes against recent larger scale prospective studies4,5 which suggest that clear fluid preparation is not inferior to full bowel preparation; and it is unusual that half dose preparation seems more effective than full dose. However, our numbers are small and it is a retrospective analysis. Additionally, in our dataset, there were more normal studies in the PEG2 arm than the other two, which may have skewed results. Further large scale prospective studies need to be carried out to clarify this difficult question.
. Nivet al. World J Gastroenterol.2013Feb 28;19(8):1264–70.
. Rosaet al. World J Gastrointest Endosc.2013Feb 16;5(2):67–73.
. Omeret al. Poster - United European Gastroenterology Week,Barcelona 2015 Oct.
. Hookey et al. Gastrointest Endosc.2017Jan;85(1):187–193.
. Kleinet al. Ann Gastroenterol.2016Apr-Jun;29(2):196–200.
Disclosure of Interest None Declared