Impact of the clinical use of artificial intelligence-assisted neoplasia detection for colonoscopy: a large-scale prospective, propensity score-matched study (with video).

BACKGROUND AND AIMS: Recently, the use of computer-aided detection (CADe) for colonoscopy has been investigated to improve the adenoma detection rate (ADR). We aimed to assess the efficacy of a regulatory-approved CADe in a large-scale study with high numbers of patients and endoscopists. METHODS: This was a propensity score-matched prospective study that took place at a university hospital between July 2020 and December 2020. We recruited patients aged ≥20 years who were scheduled for colonoscopy. Patients with polyposis, inflammatory bowel disease, or incomplete colonoscopy were excluded. We used a regulatory-approved CADe system and conducted a propensity score matching-based comparison of the ADR between patients examined with and without CADe as the primary outcome. RESULTS: During the study period, 2261 patients underwent colonoscopy with the CADe system or routine colonoscopy, and 172 patients were excluded in accordance with the exclusion criteria. Thirty endoscopists (9 nonexperts and 21 experts) were involved in this study. Propensity score matching was conducted using 5 factors, resulting in 1836 patients included in the analysis (918 patients in each group). The ADR was significantly higher in the CADe group than in the control group (26.4% vs 19.9%, respectively; relative risk, 1.32; 95% confidence interval, 1.12-1.57); however, there was no significant increase in the advanced neoplasia detection rate (3.7% vs 2.9%, respectively). CONCLUSIONS: The use of the CADe system for colonoscopy significantly increased the ADR in a large-scale prospective study including 30 endoscopists (Clinical trial registration number: UMIN000040677.).

Positive detection of exfoliated colon cancer cells on linear stapler cartridges was associated with depth of tumor invasion and preoperative bowel preparation in colon cancer.

BACKGROUND: The aim of this study was to investigate exfoliated cancer cells (ECCs) on linear stapler cartridges used for anastomotic sites in colon cancer. METHODS: We prospectively analyzed ECCs on linear stapler cartridges used for anastomosis in 100 colon cancer patients who underwent colectomy. Having completed the functional end-to-end anastomosis, the linear stapler cartridges were irrigated with saline, which was collected for cytological examination and cytological diagnoses were made by board-certified pathologists based on Papanicolaou staining. RESULTS: The detection rate of ECCs on the linear stapler cartridges was 20 %. Positive detection of ECCs was significantly associated with depth of tumor invasion (p = 0.012) and preoperative bowel preparation (p = 0.003). There were no marked differences between ECC-positive and ECC-negative groups in terms of the operation methods, tumor location, histopathological classification, and surgical margins. CONCLUSIONS: Since ECCs were identified on the cartridge of the linear stapler used for anastomosis, preoperative mechanical bowel preparation using polyethylene glycol solution and cleansing at anastomotic sites using tumoricidal agents before anastomosis may be necessary to decrease ECCs in advanced colon cancer.

[A case of severe enteritis induced by adjuvant chemotherapy for colon cancer].

A 77-year-old man underwent surgery for sigmoid colon cancer. He was diagnosed with Stage IIIa colon cancer; there- fore, we initiated oral administration of adjuvant chemotherapy comprising uracil/tegafur(UFT)plus Leucovorin(LV). However, chemotherapy was stopped after 21 days because of fatigue and diarrhea. He recovered after 3 weeks, and we administered the same regimen with a dose reduction. However, he again experienced fatigue and diarrhea after 20 days; therefore, chemotherapy was discontinued. Subsequently, he was hospitalized 8 times for conditions such as diarrhea, hypoalbuminemia, and fever. Computed tomography revealed thickening of the transverse colonic wall and colonoscopy revealed colitis, which we believe was induced by UFT plus LV. Twelve months after the last chemotherapy session, he was diagnosed with Clostridium difficile colitis. Therefore, we initiated the oral administration of vancomycin, which resulted in rapid recovery from colitis. However, he developed liver metastasis and died 29 months after the initiation of chemotherapy. We believe that this severe case of intractable colitis was caused by UFT plus LV. Therefore, we report this case with a review of the literature on enteritis induced by fluorouracil-based anticancer agents in Japan.

Polyethylene glycol solution (PEG) plus contrast medium vs PEG alone preparation for CT colonography and conventional colonoscopy in preoperative colorectal cancer staging.

PURPOSE: This study evaluated the usefulness of combined polyethylene glycol solution plus contrast medium bowel preparation (PEG-C preparation) followed by dual-contrast computed tomography enema (DCCTE) and conventional colonoscopy. The main purpose of these examinations is the preoperative staging of already known tumors. MATERIALS AND METHODS: One hundred patients with colorectal tumors were alternately allocated to either a polyethylene glycol solution preparation (PEG preparation) group (n=50) or a PEG-C preparation group (n=50) before undergoing conventional colonoscopy and computed tomographic (CT) colonography. After conventional colonoscopy, multidetector row CT scans were performed. Air images were reconstructed for both groups; contrast medium images were additionally reconstructed for the PEG-C preparation group. DCCTE images were a composite of air images and contrast medium images without use of dedicated electronic cleansing software. Quality scores (the presence or absence of blind spots of the colon) were compared between the two groups. RESULTS: Complete tumor images were obtained by DCCTE for all 50 (100%) lesions in the PEG-C preparation group, as compared with only nine of the 50 lesions (18%) in the PEG preparation group (air-contrast CT enema). The overall quality score in the PEG-C preparation group was significantly better than that in the PEG preparation group (P<0.0001). CONCLUSIONS: DCCTE showed the entire colon without blind spots in nearly all patients in the PEG-C preparation group because the areas under residual fluid were reconstructed as contrast medium images. DCCTE and conventional colonoscopy after PEG-C preparation are feasible and safe procedures that can be used for preoperative evaluation in patients with colorectal cancer.

CT air-contrast enema as a preoperative examination for colorectal cancer.

BACKGROUND/AIMS: We conducted a study on three-dimensional computed tomography (CT) images, in particular CT air-contrast enema (CT enema), using multidetector-row CT (MDCT), to see whether CT enema is useful as a preoperative examination for colorectal cancer. We aimed to evaluate the detectability of lesions and the depth of cancer invasion using CT enema. METHODS: 292 patients (328 lesions) with colorectal cancer were enrolled. After an adequate insufflation of the large intestine, MDCT scans were performed. With the data obtained by MDCT, we reconstructed CT enema images. CT enema images were assessed for the detectability of lesions. The depth of invasion was evaluated by the deformity of the lesion on profile images. The deformities were divided into five groups: no deformity, slight deformity, mild deformity, moderate deformity and severe deformity. RESULTS: The detectability of lesions was 97.3%. The reasons for undetectability were due to residual fluid in 8 cases and insufficient colonic distention in 1 case. As the depth of invasion increased, the grade of the deformity became severer (p < 0.0001). CONCLUSIONS: CT enema proved to be an excellent examination tool to detect lesions. The deformity demonstrated by CT enema could be an additional source of information to predict the depth of invasion.