Introduction

Recent advances in treatment approaches for lesions around 20 mm in size including wide-field endoscopic mucosal resection (EMR), endoscopic submucosal dissection (ESD), or endoscopic full-thickness resection (EFTR), increase the importance of the accuracy in tumor size measurement due to higher risks of perforation in en bloc EMR for lesions ≥25 mm⁽¹⁾, unexpectedly prolonged operation time in ESD, or incomplete resection in EFTR for lesions ≥20 mm⁽²⁾. However, available data on the accuracy of tumor size measurement for lesions ≥20 mm is scarce.

Aims & Methods

We aimed to assess the accuracy and the presence of biases in tumor size measurement for lesions ≥20 mm, and factors related to underestimated lesion size as 20 mm using large-scale, prospectively-collected data from en bloc resected specimens. In this post-hoc analysis of an ESD cohort study, pathologically-proven colorectal epithelial neoplasms resected in en bloc fashion were included from the data of the CREATE-J study^{(3, 4)}. Exclusion criteria were neuroendocrine tumor, residual or recurrent lesions after previous treatments, colitis-associated dysplasia, or insufficient data. The reference standard was pathologically-measured lesion size. Study measurements were endoscopic measurement accuracy defined within a 5 mm margin of error, the proportion of lesions with a size discrepancy ≥10 mm or ≥20 mm, and terminal digits preferences in endoscopically assessed lesion size. In addition, we examined characteristics and predictors of endoscopically 20 mm lesions with pathologically ≥ 25 mm size.

Results

Of 1889 en bloc resected lesions in 1809 patients, a total of 61 lesions (14 neuroendocrine tumors, 34 residual lesions, 4 colitis-related lesions, and 9 lesions with insufficient data) in 60 patients were excluded. Finally, 1828 lesions in 1749 patients were analyzed. The accuracy of endoscopic measurement in total, 20-29 mm, 30-39 mm, and 40- mm size categories was 53.4%, 63.7%, 47.5%, and 40.3%, respectively. Discrepancies of ≥10 mm or ≥20 mm occurred in 19.1% and 4.5% of lesions, respectively. Size measurement inaccuracies tended to be more frequent with increasing endoscopic lesion size (Table). In terminal digits of an endoscopic size measurement, endoscopists strongly preferred to select 0 (65.2%, 1192/1828) or 5 (30.0%, 548/1828). Of 366 endoscopically 20 mm lesions, 26.7% were found to be pathological ≥ 25 mm lesions. Multivariate analysis revealed that endoscopic 20-mm sized lesions with pathological ≥ 25-mm were significantly associated with polypoid lesions [odds ratio (OR) 2.8, 95% confidence interval (CI) 1.1-6.8] and laterally spreading tumors granular type (OR 2.0, 95% CI: 1.1-3.5).

Table. Lesions with the difference between endoscopic and pathologic lesion sizes ≥ 10 mm and ≥ 20 mm

Conclusion

Colonoscopists may be inaccurate and biased in the measurement of lesion size in large colorectal lesions. These findings suggest potential risks associated with size-based therapeutic decisions for large colorectal lesions ≥20 mm.

References

1. Burgess NG, Bassan MS, McLeod D, Williams SJ, Byth K, Bourke MJ. Deep mural injury and perforation after colonic endoscopic mucosal resection: a new classification and analysis of risk factors. Gut. 2017;66(10):1779-89.
2. Didden P, van Eijck van Heslinga RAH, Schwartz MP, Arensman LR, Vleggaar FP, de Graaf W, et al. Relevance of polyp size for primary endoscopic full-thickness resection of suspected T1 colorectal cancers. Endoscopy. 2022;54(11):1062-70.
3. Kobayashi N, Takeuchi Y, Ohata K, Igarashi M, Yamada M, Kodashima S, et al. Outcomes of endoscopic submucosal dissection for colorectal neoplasms: Prospective, multicenter, cohort trial. Dig Endosc. 2022;34(5):1042-51.
4. Ohata K, Kobayashi N, Sakai E, Takeuchi Y, Chino A, Takamaru H, et al. Long-term Outcomes After Endoscopic Submucosal Dissection for Large Colorectal Epithelial Neoplasms: A Prospective, Multicenter, Cohort Trial From Japan. Gastroenterology. 2022;163(5):1423-34.e2.