Search

Simple Summary: We developed an attention-based whole slide image (WSI)-level classification deep learning model employing surgically and endoscopically resected specimens to predict LNM in T1 CRC. Our AI model with H&E-stained WSIs and without annotations showed good performance power with the validation of an independent cohort in a single center. The area under the curve of our model was 0.781–0.824, higher than that of previous artificial intelligence (AI) studies with only WSIs. Our AI model, which showed the highest sensitivity (92.9%), reduced unnecessary additional surgery by 14.2% more than using the current JSCCR guidelines (68.3% vs. 82.5%). This revealed the feasibility of using an AI model with only H&E-stained WSIs to predict LNM in T1 CRC. According to the current guidelines, additional surgery is performed for endoscopically resected specimens of early colorectal cancer (CRC) with a high risk of lymph node metastasis (LNM). However, the rate of LNM is 2.1–25.0% in cases treated endoscopically followed by surgery, indicating a high rate of unnecessary surgeries. Therefore, this study aimed to develop an artificial intelligence (AI) model using H&E-stained whole slide images (WSIs) without handcrafted features employing surgically and endoscopically resected specimens to predict LNM in T1 CRC. To validate with an independent cohort, we developed a model with four versions comprising various combinations of training and test sets using H&E-stained WSIs from endoscopically (400 patients) and surgically resected specimens (881 patients): Version 1, Train and Test: surgical specimens; Version 2, Train and Test: endoscopic and surgically resected specimens; Version 3, Train: endoscopic and surgical specimens and Test: surgical specimens; Version 4, Train: endoscopic and surgical specimens and Test: endoscopic specimens. The area under the curve (AUC) of the receiver operating characteristic curve was used to determine the accuracy of the AI model for predicting LNM with a 5-fold cross-validation in the training set. Our AI model with H&E-stained WSIs and without annotations showed good performance power with the validation of an independent cohort in a single center. The AUC of our model was 0.758–0.830 in the training set and 0.781–0.824 in the test set, higher than that of previous AI studies with only WSI. Moreover, the AI model with Version 4, which showed the highest sensitivity (92.9%), reduced unnecessary additional surgery by 14.2% more than using the current guidelines (68.3% vs. 82.5%). This revealed the feasibility of using an AI model with only H&E-stained WSIs to predict LNM in T1 CRC. [ABSTRACT FROM AUTHOR]

Background/Aims: Colonoscopy is commonly used for colorectal cancer screening; therefore, the detection of colon subepithelial tumors (SETs) has also increased. Several research studies have been undertaken to diagnose and treat stomach and rectal SETs. The purpose of this study was to determine a diagnostic point for colon SETs by comparing histological findings with the endoscopic characteristics of colon SETs discovered by chance. Methods: A total 194 patients underwent an endoscopic ultrasound (EUS) for suspicious colon SETs during a colonoscopy from May 2014 to December 2021. A total of 105 colon SETs, which were histologically diagnosed, were finally included. Fisher's exact test was used to determine the factors associated with malignant SETs. Results: Colon SETs were predominantly present in the right colon (n = 73, 69.5%), particularly in the transverse colon (n = 32, 30.5%). The majority were smaller than 10 mm (n = 88, 83.8%), and they had hard consistencies (n = 84, 80%) and exhibited no surface changes (n = 96, 91.4%). Most of them were found in the submucosal layers (n = 54, 51.4%) and had a hypoechoic pattern (n = 56, 53.3%) in the EUS. Of the histologically confirmed cases, only three (3/105, 2.9%) were malignant. Most benign lesions were lipomas, suspected parasitic infections, or lesions caused by various inflammatory reactions, including fibrous/fibrocalcific lesions and necrotic nodules. All soft lesions were benign. Two of the three malignant lesions were adenocarcinomas, and the other was lymphoma. For the malignant SETs, there was a statistically significant alteration in the surface of the tumors (p < 0.001), and they were located where the muscularis mucosa layer was included (p = 0.008). The potential malignant SETs, granular cell tumors, and neuroendocrine tumors (NETs) had similar features, such as yellowish hypoechoic masses. Colon NETs were only found in the rectosigmoid junction. Parasitic infections and lesions, resulting in various inflammatory reactions, were observed as pale and hard SETs and mostly revealed as mixed echogenic masses located in the muscularis mucosa, submucosa, or multi-layers in the EUS. Conclusion: This study showed that small colon SETs were mostly benign lesions. Despite its rarity, pathological confirmation is crucial in cases where the SET has surface changes and has been located in a position where the muscularis mucosa layer was included on the EUS, due to the risk of malignancy. [ABSTRACT FROM AUTHOR]

High-definition (HD) endoscopy is recommended in surveillance colonoscopy for detecting dysplasia in patients with ulcerative colitis (UC). Dye-spray chromoendoscopy (DCE) and narrow-band imaging (NBI) are often used as adjunctive techniques of white-light endoscopy (WLE) in real-world practice. However, the incremental detection ability of DCE and NBI added to HD-WLE for dysplasia and serrated lesions has not yet been evaluated using tandem endoscopy in patients with long-standing extensive UC. We enrolled patients with extensive UC for >8 years who were in clinical remission (partial Mayo score < 2) at the Samsung Medical Center in Seoul, Republic of Korea. HD-WLE was performed first. Subsequently, HD-NBI and HD-DCE with indigo carmine were performed using the segmental tandem colonoscopy technique. A total of 40 patients were eligible, and data obtained from 33 patients were analyzed. The incremental detection rates (IDRs) for dysplasia and serrated lesions were calculated. HD-WLE detected three dysplasia and five sessile serrated adenomas/polyps (SSAs/Ps). HD-NBI and HD-DCE did not detect additional dysplasia (IDR = 0%; 95% confidence interval (CI): 0–56.2%). HD-NBI identified one missed SSA/P (IDR = 7.7%; 95% CI: 1.4–33.3%), and HD-DCE detected seven missed SSAs/Ps (IDR = 53.9%; 95% CI: 29.1–76.8%). Logistic regression found that HD-DCE increased the detection of SSAs/Ps compared to HD-WLE and/or HD-NBI (odds ratio (OR) = 3.16, 95% CI: 0.83–11.92, p = 0.08). DCE in addition to HD-WLE improved the detection of SSAs/Ps, but not dysplasia, in patients with long-standing extensive UC. [ABSTRACT FROM AUTHOR]

Although bowel preparation influences small bowel visibility for small bowel capsule endoscopy (SBCE), the optimal timing for bowel preparation has not been established yet. Thus, the aim of the study was to evaluate the optimal timing of polyethylene glycol (PEG) for small bowel preparation before SBCE. This multicenter prospective observational study was conducted on patients who underwent SBCE following bowel preparation with polyethylene glycol (PEG). Patients were categorized into three groups according to the time used for completing PEG ingestion: group A, within 6 h; group B, 6–12 h; and group C, over 12 h. The percentage of unclean segment in small bowel (unclean image duration / small bowel transit time × 100) and small bowel visibility quality (SBVQ) were evaluated according to the time interval between the last ingestion of PEG and swallowing of small bowel capsule endoscope. A total of 90 patients were enrolled and categorized into group A (n = 40), group B (n = 27), and group C (n = 23). The percentage of unclean segment in the entire small bowel increased gradually from group A to C (6.6 ± 7.6% in group A, 11.3 ± 11.8% in group B, and 16.2 ± 10.7% in group C, p = 0.001), especially in the distal small bowel (11.4 ± 13.6% in group A, 20.7 ± 18.7% in group B, and 29.5 ± 16.4% in group C, p < 0.001). The proportion of patients with adequate SBVQ in group A was significantly (p < 0.001) higher (30/40, 75.0%) than that in group B (17/27, 63.0%) or group C (5/23, 21.7%). In multivariate analysis, group A was associated with an increased likelihood of adequate SBVQ compared with group C (odds ratio [OR]: 13.05; 95% confidence interval [CI]: 3.53–48.30, p < 0.001). Completing PEG ingestion within 6 h prior to SBCE could enhance small bowel visibility. [ABSTRACT FROM AUTHOR]

Video capsule endoscopy (VCE) has become the noninvasive diagnostic standard in the investigation of overt obscure gastrointestinal bleeding (OGIB), with a high positive and negative predictive value. However, the diagnostic yield of the VCE is thought to depend on when it was performed. We evaluate the optimal timing performing VCE relative to overt OGIB to improve the diagnostic yield. A total 271 patients had admitted and underwent VCE for overt OGIB between 2007 and 2016 in Samsung Medical Center, Seoul, Korea. To evaluate the diagnostic yield of VCE for overt OGIB with respect to timing of the intervention, diagnostic yield was analyzed according to the times after latest bleeding. The finding of VCE was classified into P0 or P1 (no potential for bleeding or uncertain hemorrhagic potential) and P2 (high potential for bleeding, such as active bleeding, typical angiodysplasia, large ulcerations or tumors). The P2 lesion was found in 106 patients and diagnostic yield of was 39.1% for overt OGIB. Diagnostic yield of VCE to detect P2 lesion was higher when it is performed closer to the time of latest bleeding (timing of VCE between the VCE and latest bleeding: <24 h, 43/63 (68.3%); 1 days, 16/43 (34.9%); 2 days, 18/52 (34.6%); 3 days, 13/43 (30.2%); 4 days, 7/28 (25.0%); 5–7 days, 6/24 (25.0%), and ≥8 days, 4/18 (22.2%); ptrend < 0.001). The interval between the VCE and latest bleeding were categorized into <24 h (n = 63), 1–2 days (n = 95), 3–7 days (n = 95) and ≥8 days (n = 18). Multivariable analyses showed the odds ratio for P2 lesion detection was 4.99 (95% confidence interval, 1.47–16.89) in <24 h group, compared with ≥8 days group (p < 0.010). The overall re-bleeding rate for those with P2 lesion was higher than for those with P0 or P1 lesion at the end of mean follow up of 2.5 years. The proportion of patients who underwent therapeutic intervention including surgery, endoscopic intervention and embolization was higher when VCE is performed closer to the time of latest bleeding (p = 0.010). Early deployment of VCE within 24 h of the latest GI bleeding results in a higher diagnostic yield for patients with overt OGIB and consequently resulted in a higher therapeutic intervention rate [ABSTRACT FROM AUTHOR]

Balloon-assisted enteroscopy (BAE) is an important diagnostic modality for ongoing obscure gastrointestinal bleeding (OGIB). However, it is difficult to determine the optimal insertion route. We retrospectively analyzed the records of patients with OGIB contained in a multicenter enteroscopy database of 1108 balloon-assisted enteroscopy (BAE) procedures (875 patients) to find out factors affecting BAE route selection in patients with OGIB. A total of 603 BAE procedures in 512 patients were investigated: there were 392 (65.0%) bidirectional and 211 (35.0%) unidirectional procedures. Overt OGIB was more frequent in the latter group (p = 0.024). Computed tomography (CT) was more frequently performed in the unidirectional group (p < 0.001). Capsule endoscopy and a small bowel barium study were performed more frequently in the bidirectional group (p < 0.001 and p = 0.039, respectively). Multivariate analysis showed that occult OGIB, capsule endoscopy and a small bowel barium study were independently associated with use of the bidirectional approach (p = 0.011, p = 0.013 and p = 0.046, respectively). Conversely, CT was associated with use of the unidirectional approach (p < 0.001). Conclusion: CT can aid the selection of an optimal insertion route in OGIB patients. However, capsule endoscopy and small bowel barium study are unhelpful. [ABSTRACT FROM AUTHOR]

Little is known about the ability for epithelial regeneration and wound healing in patients with inflammatory bowel diseases. We evaluated the epithelial proliferation and wound healing ability of patients with Crohn's disease (CD) using patient-derived intestinal organoids. Human intestinal organoids were constructed in a three-dimensional intestinal crypt culture of enteroscopic biopsy samples from controls and CD patients. The organoid-forming efficiency of ileal crypts derived from CD patients was reduced compared with those from control subjects (p < 0.001). Long-term cultured organoids (≥6 passages) derived from controls and CD patients showed an indistinguishable microscopic appearance and culturing behavior. Under TNFα-enriched conditions (30 ng/mL), the organoid reconstitution rate and cell viability of CD patient-derived organoids were significantly lower than those of the control organoids (p < 0.05 for each). The number of EdU+ cells was significantly lower in TNFα-treated organoids derived from CD patients than in TNFα-treated control organoids (p < 0.05). In a wound healing assay, the unhealed area in TNFα-treated CD patient-derived organoids was significantly larger than that of TNFα-treated control organoids (p < 0.001). The wound healing ability of CD patient-derived organoids is reduced in TNFα-enriched conditions, due to reduced cell proliferation. Epithelial regeneration ability may be impaired in patients with CD. [ABSTRACT FROM AUTHOR]

This study investigated changes in the intestinal microbiota during 8-week infliximab maintenance therapy in inflammatory bowel disease (IBD) patients in clinical remission. Microbial compositional differences were analyzed according to the trough level of infliximab (TLI) and mucosal healing (MH) status. 16S rRNA gene-based microbiome profiling was performed on 10 and 74 fecal samples from 10 healthy volunteers and 40 adult IBD patients, respectively. Fecal sampling occurred at 1–2 weeks (1W) and 7–8 weeks (7W) after infliximab infusion. TLI was measured by ELISA at 8 weeks, immediately before the subsequent infusion; MH was evaluated by endoscopy within 3 months. There were no significant changes in microbial composition, species richness, or diversity indices between 1W and 7W. However, 7W samples from the patients with TLI ≥ 5 μg/mL showed an increased species richness compared with patients with TLI < 5 μg/mL, and patients with MH showed increased diversity compared with non-MH patients. Beta-diversity analysis showed clustering between samples in the MH and non-MH groups. LEfSe analysis identified differential composition of Faecalibacterium prausnitzii group according to TLI and MH. In conclusion, these results suggest the potential of fecal microbiota as a response indicator. [ABSTRACT FROM AUTHOR]