Roel Bogie

Optimizing post-polypectomy surveillance: A practical guide for the endoscopist

Polyp resection Next to diagnosis, completeness of polyp resection affects the recommendation on surveillance interval. Incomplete polyp resection contributes to 9-19% of the postcolonoscopy CRCs (PCCRCs). 20, 45, 58-60 A study by Pohl et al. including experienced gastroenterologists found that 10.1% (95% CI: 6.9 – 13.3) of polyps 5-20 mm in size were incompletely resected. 61 In particular, large size (RR 2.1) and sessile serrated adenomas/polyps (SSA/Ps) (RR 3.7) were independent risk factors for incomplete resection. Endoscopists with high polypectomy rates also have a lower risk of PCCRCs in their patients (OR 0.61, 95% CI: 0.42 – 0.89). 62 A recent meta-analysis assessed the effectiveness of endoscopic resection, using the need for subsequent surgery as primary endpoint. Overall, 8% (95% CI: 7 – 10) of all patients needed surgery, of whom 7% (95% CI: 6 – 9) were for incomplete resection of lesions ≥20mm, highlighting room for potential improvement. 63 In cases of suspected incomplete polyp resection, tattooing of the location is advisable to facilitate identification at next colonoscopy or surgery. 64 With increasingadoptionof endoscopic resection techniques (endoscopicmucosal resectionand endoscopic submucosal dissection) in the West, the need for surgical treatment of large colorectal polyps and early CRC will decline. 65-67 Such shift in management may also increase colonoscopy follow-up. Two meta-analyses on post-EMR recurrence rates (mean follow-up: 23 months) showed similar rates: 13.1% versus 15%. Piecemeal resection had a significant higher recurrence rate than en-bloc resection (OR 4.4, 95% CI: 2.1 – 9.4): 68 20% (95% CI: 16 – 25) for piecemeal versus 3% (95% CI: 2 – 5) for en-bloc resection. 69 Pooling analyses including follow-up data showed that 91 to 96% of all polyp recurrences are detected within six months, confirming the need for follow-up at 6 months to assess radicality. 69 Lesions with high-grade dysplasia or CRC had a greater rate of recurrence than those with low-grade dysplasia. 69 Endoscopic resection of laterally spreading tumors irrespective of size is now safe and the recurrence rates dramatically decreased. 67, 70-73 Whether 1-year follow up should be recommended after piecemeal resection of laterally spreading tumors is presently unknown and needs clarification. The number of repeat colonoscopies to safeguard radicality of resection will likely decrease with training improvements and sustained practice. Serrated polyps A number of revised post-polypectomy guidelines recommend inclusion of serrated polyps to assess surveillance intervals after polypectomy, although evidence is weak. 1, 2, 5-7 Hyperplastic polyps (which are the most common form of serrated polyps) are benign lesions, without any risk of malignant transformation. 74, 75 Several observational studies found that hyperplastic polyps alone are associated with slightly increased risk of adenomas, but not of advanced adenomas. 76, 77 Compared with the presence of adenoma alone, simultaneous presence of hyperplastic polyps and adenomas at baseline colonoscopy is not associated with increased risk of (advanced) adenomas at surveillance. 78 Some studies showed that presence of serrated polyps is associatedwith synchronous advanced neoplasia. 79-82 In contrast to hyperplastic polyps, the increasingly recognized sessile serrated adenomas/polyps (SSA/Ps) and traditional serrated adenomas are associated with risk of progression to CRC. Several observational studies found that SSA/Ps are risk factors for having synchronous advanced adenomas and CRC, especially when large (≥10mm) or dysplastic. 79-81, 83 87 The presence of large serrated polyps correlated with both synchronous distal and proximal advanced adenomas and CRC, although the correlation was stronger for proximal neoplasms. 79 While awaiting more data, presence of large or dysplastic serrated polyps is now considered a risk

7

137