Roel Bogie

Chapter 12

border, eradication of the residual tissue may reduce recurrence rates. Whether there is sufficient evidence for application of these techniques in common practice, a meta-analysis was performed. Chapter 6 showed an overall 18% reduction in recurrence risk with applying thermal ablation after EMR (95% CI: 11 – 26). Two modalities were studied: snare tip soft coagulation (STSC) and argon plasma coagulation (APC). Subgroup analysis for STSC showed a significant reduction, but for APC significance was not reached, probably because of a small number of studies. When all observational studies about STSC were included, a pooled recurrence risk of only 5% was noted (after 6 - 12 months, 95% CI: 2 – 10). STSC requires no additional materials and can be applied directly after EMR using the same snare. This technique is therefore easy to employ in endoscopy practice and could reduce recurrence of piecemeal resected neoplasms. Chapter 6 concludes that the use of STSC after piecemeal EMR of LNPCPs is recommended. Another simple and apparently effective technique is marking of the lesion prior to EMR. In a study in which soft tip coagulation before lifting is used to mark the borders, EMR was performed in which all mucosa between the markers was removed. This study showed a recurrence risk reduction of 80% (8.1% after marking versus 28.7% without marking, P =0.0004). 43 LST patients at risk LSTs clearly pose a risk for development into CRCs by difficulties in detection and resection, with recurrences. Adequate, complete LST resection should be effective and result in CRC reduction. In Chapter 4 we studied synchronous and metachronous neoplasms found during colonoscopies in patients with large neoplasms (≥10mm). Patients with LSTs had more synchronous neoplasms and also had more flat neoplasms in general than patients with a large polypoid neoplasm. This observation is in line with findings in other studies. 44, 45 During 6-year follow-up, the risk of developing a new high-risk neoplasm (with HDG or SMI) was also increased in LST patients compared to patients with large polypoid neoplasms (hazard ratio of 2.9, 95% CI: 1.8 – 4.6). Although LST patients had their surveillance colonoscopy within a shorter interval and more frequently, correction for this in a Cox regression model did not alter the conclusion. During follow up, LST patients developed more often nonpolypoid neoplasms than patients with large polypoid neoplasms at baseline. The risk of especially LST-NG for developing advanced neoplasia during surveillance, was confirmed in the Japanese Polyp Study. In this study, almost 1500 patients were randomized into surveillance colonoscopy after one and three years and into only after three years, after a double baseline clearing colonoscopy. The presence of LST-NG at baseline was the largest risk factor in univariate analysis for advanced neoplasia within a 3 year colonoscopy surveillance (OR 6.61, 95% CI: 2.11 – 17.61). 46 For non-polypoid neoplasms in general, more frequent advanced neoplasia are found during surveillance when compared to polypoid neoplasms. In the before-mentioned Japanese Polyp Study, baseline presence of small (<10mm) non-polypoid neoplasms, was the second largest risk factor for advanced neoplasms during surveillance (OR 3.04, 95% CI: 1.01 – 12.11). 46 A study comparing the follow-up findings of patients with non-polypoid and polypoid neoplasms found a risk ratio of 1.6 (95%CI: 1.1 – 2.6). 47 Patients with non-polypoid lesions had alsomore often new non polypoid lesions during surveillance, but the risk for developing CRC was similar. 47 So, susceptibility to develop nonpolypoid or polypoid neoplasms is apparently influenced by patient profiles. Since patients with LSTs are known to have a higher risk of synchronous and metachronous neoplasms, themorphology of colorectal neoplasms can be used for risk stratification in surveillance guidelines. Chapter 4 concludes that stricter surveillance intervals should be advised to patients with LSTs. It is not yet standard practice to include morphology in decision making (see Chapter 7 ),

226