Validation of the diagnostic accuracy of the mirocam small bowel capsule endoscopy with the application of express play mode analysis of the miroview software

A Székely; G Balogh; I Hritz; L Madácsy

doi:10.1055/s-0031-1278513

Zeitschrift für Gastroenterologie, Table of Contents

Validation of the diagnostic accuracy of the mirocam small bowel capsule endoscopy with the application of express play mode analysis of the miroview software

A Székely ¹, G Balogh ², I Hritz ¹, L Madácsy ¹

¹1st Department of Internal Medicine, Fejér Megyei Szent György Hospital, Székesfehérvár
²Department of Surgery, Kaposi Mór Teaching Hospital, Kaposvár

Abstract

Introduction: Capsule endoscopy is a widely accepted, non-inasive diagnostic procedure for the small bowel (SB) diseases and obscure gastrointestinal bleeding. However, interpretation of the whole video recording is a time-consuming and costly procedure that can take up to 2 hours. MiroCam Intromedic capsule endoscope system offers a new MiroView image processing and analysis software with the possibility of quick view mode and express play mode (EPM), with which the evaluation time of the complete video record could be significantly reduced to 20min. The aim of our current study were to compare the diagnostic accuracy of capsule endoscopy analysis between EPM and original picture-to-picture analysis (PPA), and to measure the analysis time difference between the two methods. Method: We selected 13 patients with positive results and obvious pathology in the small bowel detected by Mirocam SB CE. The detected lesions were as follows: angiodysplasia 2 pts; NSAID enteropathy 4 pts; inflammation and erosions 2 pts; Crohn ulcers 2 pts; focal lymphangiectasia 3 pts. Thereafter all CE recordings were re-evaluated by two independent CE experts without the knowledge of the previously marked findings and randomly applied EPM versus PPA analysis. The diagnostic accuracy of the two different methods then were compared with the success of indentifying all previously described SB pathology and the total number of frames contained the particular small bowel lesions. The mean evaluation time of the EMP and PPA mode of analysis were also calculated and compared. Results: All pathological lesions were depicted by the investigators regardless of the selected evaluation mode of the software (EPM vs. PPA). Nevertheless, significantly greater average number of pictures depicted of the SB pathology were captured by the PPA than the EPM mode, 44.6 vs. 11.6 (P=0.015). In contrast the average evaluation time of whole capsule recording was significantly shorter with the application of EPM versus PPA analysis mode, 16min vs. 55min., respectively (P=2.5×10E-8).

Conclusions: The newly developed EPM of the MiroView software could significantly reduce the evaluation time of the investigation, while it has no significant effect to the diagnostic accuracy of the SB CE investigation.