Roel Bogie

Impact of endoscopist training on post-colonoscopy colorectal cancer rate

This study was approved by the Institutional Review Boards of the Maastricht University Medical Center+ and is registered in the Netherlands Trial Registry: NTR4844. Approval for the prospective colonoscopy database is assigned by the MEC of the Maastricht University Medical Center+ (MUMC+) with number 14-4-046. The need for informed consent was waived by the Institutional Review Board.

CRC diagnosis within 60 months after colonoscopy?

Yes

Adequate surveillance interval following the index colonoscopy?

Adequate previous colonoscopic examination (e.g. cecal intubation, bowel preparation)?

Yes

CRC in same anatomic segment as previously resected advanced adenoma?

Yes

Negative colonoscopy within 36 months before CRC diagnosis?

No

No

Advanced stage at CRC diagnosis

No

No

Yes

Yes

No

Non-adherence to surveillance intervals

Inadequate bowel examination

Incomplete polyp resection

Missed lesions

Newly developed cancers

Results The post-training cohort of 9353 patients was enrolled from a total of 13,190 colonoscopies in four years (2008-2012). Patients underwent elective colonoscopy for symptoms, surveillance or screening. Patients with IBD, hereditary CRC or a history of CRC were excluded (n=1136). The analyzed post-training population consisted of 8217 patients with a mean age of 59.1 years (SD 15.8) and of which 54% is female. Prior to training (2004-2008), 8,468 colonoscopies were performed. Based on the findings of 2008, we estimated that 5275 unique patients would have had one or more colonoscopies during this period of four years. We assumed that all patients of this study period had 48 months of follow up, since this was the case for the patients included in 2008. With this information, we calculated the total years of follow-up (PYFU) for the pre-training cohort (2004-2008) which was 21,299 person years. The total years of follow-up for patients from the post-training prospective cohort (2008 2012) was measured at 31,993 person years. When comparing the PCCRC rates before and after training, this study shows that the PCCRC rate declined after implementation of training by more than 50% from 2.0/1000 colonoscopies before training to 0.8/1000 colonoscopies after training ( P =0.031). These rates result in an estimated decrease from 0.79/1000 PYFU before training to 0.34/1000 PYFU after training ( P =0.041) ( Figure 9.3 ). Figure 9.2: Modified algorithm by Pabby et al . 5 evaluating the most likely etiology of post colonoscopy colorectal cancers (PCCRCs).

9

177