Roel Bogie

Chapter 9

A total of 11 patients (mean age [SD] 76.8 years [7.4]; 45% female) were diagnosed with PCCRCs after colonoscopy between February 2008 and February 2012, i.e. in the post-training cohort. Clinical characteristics of these PCCRC cases are shown in Figure 9.4 . From the PCCRC patients, ten patients were older than 70 years. The time to CRC diagnosis varied from 14 to 52 months. In contrast with previous data, PCCRCs were more equally distributed over the proximal and distal colon. Nine patients had diverticular disease and two patients had proven metastatic disease at time of diagnosis. Indication for index colonoscopy were symptoms (54.5%) (i.e. rectal blood loss, anemia) or surveillance (45.5%). Using the Pabby algorithm ( Figure 9.2 ), potentially missed lesions remained the main explanation (45.4%) for the occurrence of PCCRC, while no PCCRC case could be attributed to incomplete resection. Non-adherence to surveillance intervals comprised 18.2% of cases, as well as inadequate bowel examination (18.2%) and newly developed cancers (18.2%). Applying the 60-month follow-up as a sensitivity analysis, the PCCRC rate before training is estimated at 0.64/1000 PYFU. When comparing to the PCCRC rate after training (0.34/1000 PYFU) the difference is less distinctive ( P =0.144).

Start trial

10

4000

8

3000

6

2000

4

Number of PCCRCs

Number of colonoscopies

1000

2

0

0

2004

2005 2006 2007 2008 2009 2010 2011

Year of colonoscopy

Total number of colonoscopies PCCRCs; number of colonoscopies followed by CRC diagnosis after 6-60 months Note: Cumulative PCCRC rate decreased from 2.0 per 1000 colonoscopies before training to 0.8 per 1000 colonoscopies after training

Figure 9.3: Number of yearly post-colonoscopy colorectal cancers (PCCRCs) and number of colonoscopies before and after systematic endoscopists training (February 2004-February 2012).

178