Conditional inference tree models to perceive depth of invasion in T1 colorectal cancer.

BACKGROUND AND AIM: Accurate diagnosis of invasion depth for T1 colorectal cancer is of critical importance as it decides optimal resection technique. Few reports have previously covered the effects of endoscopic morphology on depth assessment. We developed and validated a novel diagnostic algorithm that accurately predicts the depth of early colorectal cancer. METHODS: We examined large pathological and endoscopic databases compiled between Jan 2015 and Dec 2018. Training and validation data cohorts were derived and real-world diagnostic performance of two conditional interference tree algorithms (Models 1 and 2) was evaluated against that of the Japan NBI-Expert Team (JNET) classification used by both expert and non-expert endoscopists. RESULTS: Model 1 had higher sensitivity in deep submucosal invasion than that of JNET alone in both training (45.1% vs. 28.6%, p < 0.01) and validation sets (52.3% vs. 40.0%, p < 0.01). Model 2 demonstrated higher sensitivity than Model 1 (66.2% vs. 52.3%, p < 0.01) in excluding deeper invasion of suspected Tis/T1a lesions. CONCLUSION: We discovered that machine-learning classifiers, including JNET and macroscopic features, provide the best non-invasive screen to exclude deeper invasion for suspected Tis/T1 lesions. Adding this algorithm improves depth diagnosis of T1 colorectal lesions for both expert and non-expert endoscopists.

A novel endoscopic hand-suturing technique for defect closure after colorectal endoscopic submucosal dissection: a pilot study.

BACKGROUND: This study aimed to demonstrate the feasibility of endoscopic hand-suturing (EHS) and attainability of sustained closure after colorectal endoscopic submucosal dissection (ESD). METHODS: EHS was defined as uninterrupted endoscopic suturing of the mucosal defect after colorectal ESD using an absorbable barbed suture and a through-the-scope needle holder. Following individual EHS training using an ex vivo porcine colonic model, two experienced endoscopists performed EHS. Repeat colonoscopy was performed on the third or fourth day after ESD to examine the EHS site. The primary end point was the complete EHS closure rate, and secondary end points were sustained closure and post-ESD bleeding rates. RESULTS: 11 lesions were included. Median size of the mucosal defect was 38 mm (range 25 - 55 mm) and the lesion characteristics were as follows: lower rectum/upper rectum/ascending colon/cecum = 3/3/2/3, and 0-IIa/0-Is + IIa/others = 5/4/2. EHS was not attempted in two patients owing to difficulty in colonoscope reinsertion after ESD and intraoperative perforation, respectively. EHS was performed for nine lesions, and the complete EHS closure rate was 73 %. Median procedure time for suturing was 56 minutes (range 30 - 120 minutes) and median number of stitches was 8 (range 6 - 12). Sustained closure and post-ESD bleeding rates were 64 % and 9 %, respectively. CONCLUSIONS: EHS achieved complete and sustained closure in the colorectum. However, EHS is not currently clinically applicable given the long procedure time. Further modifications of the technique and devices are desirable.

Small intestinal bleeding and stricture caused by Meckel's diverticulum.

A 44-year-old man presented to our hospital with lower gastrointestinal bleeding. We performed balloon-assisted enteroscopy, which revealed diverticulum and stricture at the ileum. The patient underwent segmental small bowel resection and diagnosed with Meckel's diverticulum. We should keep in mind the possibility of intestinal stricture due to Meckel's diverticulum.

Molecular features of colorectal polyps presenting Kudo's type II mucosal crypt pattern: are they based on the same mechanism of tumorigenesis?

BACKGROUND AND STUDY AIMS: The molecular features of serrated polyps (SPs) with hyperplastic crypt pattern, also called Kudo's type II observed by chromoendoscopy, were evaluated. METHODS: The clinicopathological and molecular features of 114 SPs with a hyperplastic pit pattern detected under chromoendoscopy (five dysplastic SPs, 63 sessile serrated adenoma/polyps (SSA/Ps), 36 microvesicular hyperplastic polyps (MVHPs), and 10 goblet cell-rich hyperplastic polyps (GCHPs)) were examined. The frequency of KRAS and BRAF mutations and CpG island methylator phenotype (CIMP) were investigated. RESULTS: Dysplastic SPs and SSA/Ps were frequently located in the proximal colon compared to others (SSA/Ps vs. MVHPs or GCHPs, P < 0.0001). No significant difference was found in the frequency of BRAF mutation among SPs apart from GCHP (60 % for dysplastic SPs, 44 % for SSA/Ps, 47 % for MVHPs, and 0 % for GCHPs). The frequency of CIMP was higher in dysplastic SPs or SSA/Ps than in MVHPs or GCHPs (60 % for dysplastic SPs, 56 % for SSA/Ps, 32 % for MVHPs, and 10 % for GCHPs) (SSA/Ps vs. GCHP, P = 0.0068). When serrated neoplasias (SNs) and MVHPs were classified into proximal and distal lesions, the frequency of CIMP was significantly higher in the proximal compared to the distal SNs (64 % vs. 11 %, P = 0.0032). Finally, multivariate analysis showed that proximal location and BRAF mutation were significantly associated with an increased risk of CIMP. CONCLUSIONS: Distinct molecular features were observed between proximal and distal SPs with hyperplastic crypt pattern. Proximal MVHPs may develop more frequently through SSA/Ps to CIMP cancers than distal MVHPs.