A C Currie1, D Burling2, E Mainta2, R Ilangovan2, M Moorghen3, P Lung2, O Faiz4,5, R H Kennedy4,5. 1. Department of Surgery, Imperial College London, London, UK. andrew.currie@nhs.net. 2. Department of Intestinal Imaging, Imperial College London, London, UK. 3. Department of Histopathology, St Mark's Hospital and Academic Institute, Harrow, Middlesex, UK. 4. Department of Surgery, Imperial College London, London, UK. 5. Department of Surgery and Cancer, Imperial College London, London, UK.
Abstract
AIM: Full-thickness laparo-endoscopic excision (FLEX) is a new technique developed for the full-thickness excision of colonic adenomas and, potentially, early cancer, avoiding the need for colectomy. FLEX requires accurate preoperative characterization of three key morphological features of the tumour, including its relation to the mesenteric border, its diameter and the circumferential extent of involvement of the bowel wall. This study evaluated the accuracy of CT colonography (CTC) for the assessment of these features in early colonic tumours. METHOD: Consecutive patients undergoing CTC prior to colonic resection for complex benign polyps or UICC Stage 1 cancer were retrospectively analysed by two specialist gastrointestinal radiologists blinded to the subsequent histopathological findings. The location of the tumour in relation to the mesenteric border, its maximum diameter and the circumferential extent of involvement of the colonic wall were correlated with the histopathological examination of the surgical resection specimen. Pearson's correlation coefficient (r) and Kappa agreement (κ) were used to compare the maximum diameter and the circumferential extent of involvement of the colonic wall. RESULTS: Twenty-eight patients with early colonic neoplasia were included. All had had a surgical segmental resection. Four had a benign adenoma and 24 had a TNM Stage 1 cancer. Histopathological assessment of the resected surgical specimen showed that 21 of the 28 lesions were located on the mesenteric border. The median diameter was 35 (interquartile range 28-42) mm; 13 lesions involved less than one-third of the circumference, 11 between one and two-thirds and four more than two-thirds. CTC correctly identified the location of the lesion in relation to the mesenteric border in all 28 cases. Correlation between CTC and histopathology was good for the assessment of the maximum diameter of the lesion (r = 0.81) and the circumferential extent of involvement of the colonic wall (κ = 0.76). CONCLUSION: CTC can accurately assess the key morphological features for the selection of patients with early colonic neoplasia for full-thickness laparo-endoscopic excision. Colorectal Disease
AIM: Full-thickness laparo-endoscopic excision (FLEX) is a new technique developed for the full-thickness excision of colonic adenomas and, potentially, early cancer, avoiding the need for colectomy. FLEX requires accurate preoperative characterization of three key morphological features of the tumour, including its relation to the mesenteric border, its diameter and the circumferential extent of involvement of the bowel wall. This study evaluated the accuracy of CT colonography (CTC) for the assessment of these features in early colonic tumours. METHOD: Consecutive patients undergoing CTC prior to colonic resection for complex benign polyps or UICC Stage 1 cancer were retrospectively analysed by two specialist gastrointestinal radiologists blinded to the subsequent histopathological findings. The location of the tumour in relation to the mesenteric border, its maximum diameter and the circumferential extent of involvement of the colonic wall were correlated with the histopathological examination of the surgical resection specimen. Pearson's correlation coefficient (r) and Kappa agreement (κ) were used to compare the maximum diameter and the circumferential extent of involvement of the colonic wall. RESULTS: Twenty-eight patients with early colonic neoplasia were included. All had had a surgical segmental resection. Four had a benign adenoma and 24 had a TNM Stage 1 cancer. Histopathological assessment of the resected surgical specimen showed that 21 of the 28 lesions were located on the mesenteric border. The median diameter was 35 (interquartile range 28-42) mm; 13 lesions involved less than one-third of the circumference, 11 between one and two-thirds and four more than two-thirds. CTC correctly identified the location of the lesion in relation to the mesenteric border in all 28 cases. Correlation between CTC and histopathology was good for the assessment of the maximum diameter of the lesion (r = 0.81) and the circumferential extent of involvement of the colonic wall (κ = 0.76). CONCLUSION: CTC can accurately assess the key morphological features for the selection of patients with early colonic neoplasia for full-thickness laparo-endoscopic excision. Colorectal Disease