A novel artificial intelligence-based endoscopic ultrasonography diagnostic system for diagnosing the invasion depth of early gastric cancer.

BACKGROUND: We developed an artificial intelligence (AI)-based endoscopic ultrasonography (EUS) system for diagnosing the invasion depth of early gastric cancer (EGC), and we evaluated the performance of this system. METHODS: A total of 8280 EUS images from 559 EGC cases were collected from 11 institutions. Within this dataset, 3451 images (285 cases) from one institution were used as a development dataset. The AI model consisted of segmentation and classification steps, followed by the CycleGAN method to bridge differences in EUS images captured by different equipment. AI model performance was evaluated using an internal validation dataset collected from the same institution as the development dataset (1726 images, 135 cases). External validation was conducted using images collected from the other 10 institutions (3103 images, 139 cases). RESULTS: The area under the curve (AUC) of the AI model in the internal validation dataset was 0.870 (95% CI: 0.796-0.944). Regarding diagnostic performance, the accuracy/sensitivity/specificity values of the AI model, experts (n = 6), and nonexperts (n = 8) were 82.2/63.4/90.4%, 81.9/66.3/88.7%, and 68.3/60.9/71.5%, respectively. The AUC of the AI model in the external validation dataset was 0.815 (95% CI: 0.743-0.886). The accuracy/sensitivity/specificity values of the AI model (74.1/73.1/75.0%) and the real-time diagnoses of experts (75.5/79.1/72.2%) in the external validation dataset were comparable. CONCLUSIONS: Our AI model demonstrated a diagnostic performance equivalent to that of experts.

High incidence of lung cancer death after curative endoscopic submucosal dissection for superficial esophageal squamous cell carcinoma.

BACKGROUND AND AIM: Following treatment of superficial esophageal squamous cell carcinoma (ESCC), surveillance for a second primary malignancy (SPM) is necessary. However, detailed evidence regarding the timing and prognosis of SPMs is insufficient. We aimed to clarify the details of SPMs and their effects on patient outcomes. METHODS: This retrospective, multicenter study involved 11 hospitals. Patients with superficial ESCC curatively resected using endoscopic submucosal dissection between May 2005 and December 2012, were included in this study. RESULTS: The 5-year survival rate of 187 patients was 92.6% during a median follow-up duration of 96.8 months. Thirty-one patients died, 14 of whom died of SPMs. Compared to patients with SPMs detectable by esophagogastroduodenoscopy (EGD), patients with SPMs detectable only by modalities other than EGD had a significantly higher mortality rate (p < 0.001). Patients with second primary lung cancer (LC) had a high mortality rate (56.3%). Univariate and multivariate analyses showed that multiple Lugol-voiding lesions (LVLs) tended to be associated with SPMs (p = 0.077, hazard ratio [HR] 4.43, 95% confidence interval [CI]: 0.91-6.50), and metachronous ESCC was an independent risk factor for the incidence of second primary LC (p = 0.037, HR 3.51, 95% CI: 1.08-11.41). CONCLUSIONS: SPMs that cannot be detected by EGD, such as LC, must be considered after the curative resection of ESCC. We suggest strict screening by both EGD and computed tomography for patients with multiple LVLs or metachronous ESCC to detect SPMs in their early stages.

A machine learning model for predicting the lymph node metastasis of early gastric cancer not meeting the endoscopic curability criteria.

BACKGROUND: We developed a machine learning (ML) model to predict the risk of lymph node metastasis (LNM) in patients with early gastric cancer (EGC) who did not meet the existing Japanese endoscopic curability criteria and compared its performance with that of the most common clinical risk scoring system, the eCura system. METHODS: We used data from 4,042 consecutive patients with EGC from 21 institutions who underwent endoscopic submucosal dissection (ESD) and/or surgery between 2010 and 2021. All resected EGCs were histologically confirmed not to satisfy the current Japanese endoscopic curability criteria. Of all patients, 3,506 constituted the training cohort to develop the neural network-based ML model, and 536 constituted the validation cohort. The performance of our ML model, as measured by the area under the receiver operating characteristic curve (AUC), was compared with that of the eCura system in the validation cohort. RESULTS: LNM rates were 14% (503/3,506) and 7% (39/536) in the training and validation cohorts, respectively. The ML model identified patients with LNM with an AUC of 0.83 (95% confidence interval, 0.76-0.89) in the validation cohort, while the eCura system identified patients with LNM with an AUC of 0.77 (95% confidence interval, 0.70-0.85) (P = 0.006, DeLong's test). CONCLUSIONS: Our ML model performed better than the eCura system for predicting LNM risk in patients with EGC who did not meet the existing Japanese endoscopic curability criteria. We developed a neural network-based machine learning model that predicts the risk of lymph node metastasis in patients with early gastric cancer who did not meet the endoscopic curability criteria.

Efficacy of prophylactic clip closure in reducing the risk of delayed bleeding after colorectal endoscopic submucosal dissection in patients on anticoagulant therapy: Multicenter prospective study.

OBJECTIVES: The high rate of delayed bleeding after colorectal endoscopic submucosal dissection (ESD) in patients undergoing anticoagulant therapy remains a problem. Whether prophylactic clip closure reduces the rate of delayed bleeding in these patients is unclear. This study aimed to evaluate the efficacy of prophylactic clip closure in patients receiving anticoagulants. METHODS: This multicenter prospective interventional trial was conducted at nine referral centers in Japan. Patients regularly taking anticoagulants, including warfarin potassium or direct oral anticoagulants, and undergoing ESD for colorectal neoplasms were enrolled. The discontinuation of anticoagulants was minimized according to recent guidelines. After the ESD, post-ESD ulcers were prophylactically closed using endoclips. The primary end-point was the incidence of delayed bleeding. The sample size was 45 lesions, and prophylactic clip closure was considered effective when the upper limit of the 90% confidence interval (CI) for delayed bleeding did not exceed 20%. RESULTS: Forty-five lesions were used, and three were excluded. Complete closure was achieved in 41/42 lesions (97.6%). The overall delayed bleeding rate was low, at 4.9% (2/41; 90% [CI] 0.8-14.5), which was significantly lower than that at the prespecified threshold of 20% (P = 0.007). The median closure procedure time was 17 min, and the median number of clips was nine. No massive delayed bleeding requiring transfusion, interventional radiology, or surgery was observed, and no thromboembolic events were observed. CONCLUSION: Prophylactic clip closure may reduce the risk of delayed bleeding following colorectal ESD in patients receiving anticoagulants. TRIAL REGISTRATION: UMIN Clinical Trial Registry (UMIN000036734).

Non-Pure Intestinal Phenotype as an Indicator of Progression in Sporadic Nonampullary Duodenal Adenomas: A Multicenter Retrospective Cohort Study.

INTRODUCTION: We aimed to evaluate the natural course of sporadic nonampullary duodenal adenomas (SNDAs) and determine the risk factors of progression. METHODS: We retrospectively analyzed the follow-up outcomes of patients with biopsy-diagnosed SNDA between April 2010 and March 2016 at 13 institutions. All initial biopsy specimens were centrally evaluated. Only those diagnosed with adenomas were included. Mucinous phenotypes were classified into pure intestinal and non-pure intestinal phenotypes. Cumulative incidence rates of carcinoma and tumor enlargement were evaluated. Tumor enlargement was defined as a ≥25% or 5-mm increase in tumor size. RESULTS: Overall, 121 lesions were analyzed. Within a median observation period of 32.7 months, 5 lesions were diagnosed as carcinomas; the cumulative 5-year incidence of carcinoma was 9.5%. Male sex ( P = 0.046), initial lesion size ≥10 mm ( P = 0.044), and non-pure intestinal phenotype ( P = 0.019) were significantly associated with progression to carcinoma. Tumor enlargement was observed in 22 lesions, with a cumulative 5-year incidence of 33.9%. Initial lesion size ≥10 mm ( P < 0.001), erythematous lesion ( P = 0.002), high-grade adenoma ( P = 0.002), Ki67 negative ( P = 0.007), and non-pure intestinal phenotype ( P = 0.001) were risk factors of tumor enlargement. In a multivariate analysis, an initial lesion size ≥10 mm ( P = 0.010) and non-pure intestinal phenotype ( P = 0.046) were independent and significant risk factors of tumor enlargement. DISCUSSION: Lesion size ≥10 mm and non-pure intestinal phenotype on initial biopsy are risk factors of cancer progression and tumor enlargement in cases with SNDA. Thus, management effectiveness may be improved by focusing on lesion size and the mucinous phenotype.

Endoscopic features and clinical course of colorectal carcinoma resected by cold snare polypectomy.

BACKGROUND AND AIM: Cold snare polypectomy is commonly performed to remove small colorectal polyps. Accidental resection of carcinomas during this procedure has been reported. Herein, we aimed to clarify the clinicopathological features and clinical course of colorectal carcinomas resected by cold snare polypectomy. METHODS: This multicenter retrospective cohort study was conducted at 10 Japanese healthcare centers. Of the colorectal lesions resected by cold snare polypectomy between April 2016 and March 2020, lesions pathologically diagnosed as carcinoma were reviewed. Centralized histology (based on the Vienna classification) and endoscopic reviews were performed. The study endpoints were endoscopic features and clinical outcomes of cold snare polypectomy-resected colorectal carcinomas (Vienna category ≥4.2). RESULTS: We reviewed 74 of the 70 693 lesions resected by cold snare polypectomy. After a central pathological review, 68 lesions were diagnosed as carcinomas. The Japan Narrow-band imaging Expert Team (JNET) classification type 2B, lesion size ≥6 mm, and multinodular morphology were the significant endoscopic predictors of carcinoma resected by cold snare polypectomy. No adverse events related to the procedure occurred. Sixty-three lesions were diagnosed as carcinomas within the mucosal layer, and 34 were curative resections. Of the five carcinoma lesions with submucosal invasion, additional surgery revealed remnant cancer tissues in one lesion. No local or metastatic recurrence was observed during follow-up. CONCLUSIONS: Although most of the carcinomas resected by cold snare polypectomy were within the mucosal layer, few lesions invading the submucosa were identified. Careful pre-procedural endoscopic evaluation, especially focusing on the JNET classification and multinodular morphology, is recommended.

Plasmablastic lymphoma occurring in ulcerative colitis during treatment with immunosuppressive therapy.

A 53-year-old man who had a history of ulcerative colitis (UC) for 2 years underwent colonoscopy as regular follow-up. The results showed an elevated lesion in the descending colon, which was diagnosed as plasmablastic lymphoma (PBL) based on pathological findings. In situ hybridization for the Epstein-Barr virus-encoded RNA probe was positive. Fluorescence in situ hybridization revealed rearrangement of the MYC gene. He had been taking prednisolone, 5-aminosalicylic acid, azathiopurine, and ustekinumab at the diagnosis of PBL and had multiple prior therapies for UC including infliximab, tacrolimus, and tofacitinib due to steroid dependence. PBL is a rare aggressive B cell lymphoma initially described in the oral cavity of human immunodeficiency virus positive patients and it is suspected to have an association with immunocompromised status of patients. The number of cases of PBL in inflammatory bowel disease (IBD) patients is extremely rare. All these patients were administered immunosuppressive therapy including thiopurines or biologics. IBD patients with immunosuppressive therapy have a higher potential for developing lymphoproliferative disorders. Clinicians should be aware of the risk of lymphoma, including PBL.

Diagnostic value of endoscopic ultrasonography for the depth of gastric cancer suspected of submucosal invasion: a multicenter prospective study.

BACKGROUND: Although the combination of conventional endoscopy (CE) and endoscopic ultrasonography (EUS) is useful for predicting the depth of early gastric cancer (EGC), the diagnostic value of EUS for submucosal (SM) invasive cancer has not been fully investigated. METHODS: We conducted a multicenter prospective study from May 2017 to January 2021 to evaluate the validity of a diagnostic strategy combining CE and EUS and to clarify the additional value of EUS for EGC suspected of SM invasion. In each case, the diagnosis was first made using CE, followed by EUS, and finally confirmed using a combination algorithm. RESULTS: A total of 180 patients with EGC were enrolled from 10 institutions, of which 175 were analyzed. The histopathological depths were M, SM1, SM2, and ≥ MP in 72, 16, 64, and 23 lesions, respectively. Treatment included 92 endoscopic submucosal dissection cases and 83 surgical cases. The overall diagnostic accuracy classified by M-SM1 or SM2-MP was 58.3% for CE, 75.7% for EUS, and 78.9% for the combination of CE and EUS; the latter two were significantly higher than that of CE alone (P < 0.001). The CE, EUS, and combination accuracy rates in 108 differentiated-type lesions were 51.9%, 77.4%, and 79.6%, respectively; the latter two were significantly higher than CE alone (P < 0.001). A significant additive effect of EUS was observed in CE-SM2 low-confidence lesions but not in CE-M-SM1 lesions or in CE-SM2 high-confidence lesions. Among the nine CE findings, irregular surface, submucosal tumor-like elevation, and non-extension signs were significant independent markers of pSM2-MP. Poorly delineated EUS lesions were misdiagnosed. CONCLUSIONS: EUS provides additional value for differentiated-type and CE-SM2 low-confidence EGCs in diagnosing invasion depth. CLINICAL REGISTRATION NUMBER: UMIN000025862.

Local recurrence after endoscopic resection of sessile serrated lesions: A multicenter prospective study by the Osaka Gut Forum.

BACKGROUND AND AIM: Sessile serrated lesions (SSLs) act as precursors to colorectal cancer, sometimes harbor carcinomas, and are sometimes incompletely resected. We aimed to evaluate local recurrence after endoscopic resection of SSL ≥10 mm. METHODS: This prospective, single-arm, observational study was performed at eight Japanese tertiary institutions. Colorectal lesions ≥10 mm were resected endoscopically, and the pathological diagnosis was either an SSL or hyperplastic polyp (HP). Follow-up colonoscopy was performed 1 year later, and the local recurrence was evaluated by biopsy. RESULTS: From October 2018 to September 2021, 104 cases with 123 lesions were registered. Among the pathologically diagnosed 105 SSLs and 18 HPs, 95 and 7 lesions were diagnosed as SSLs and HPs, respectively, by central pathological review. Among the 104 endoscopically diagnosed SSLs, 86 were diagnosed as SSLs, whereas among the 11 endoscopically diagnosed HPs, two were diagnosed as HPs by central pathological review (the rest were SSLs). Among the 95 patients with 113 lesions who underwent follow-up colonoscopy, resection scars were identified in 95 (84%) lesions. Three (3.1%; 95% confidence interval 0.6-8.7%) local recurrences were diagnosed pathologically among 98 pathologically diagnosed SSLs. Two (6%) local recurrences were diagnosed in patients with SSLs ≥20 mm. CONCLUSIONS: The local recurrence rate after endoscopic resection of SSLs ≥10 mm was 3.1%. Careful follow-up is recommended after endoscopic resection of large SSLs. Endoscopically diagnosed HPs ≥10 mm were sometimes pathologically diagnosed as SSL and should be considered for resection.

Utility of an artificial intelligence system for classification of esophageal lesions when simulating its clinical use.

Previous reports have shown favorable performance of artificial intelligence (AI) systems for diagnosing esophageal squamous cell carcinoma (ESCC) compared with endoscopists. However, these findings don't reflect performance in clinical situations, as endoscopists classify lesions based on both magnified and non-magnified videos, while AI systems often use only a few magnified narrow band imaging (NBI) still images. We evaluated the performance of the AI system in simulated clinical situations. We used 25,048 images from 1433 superficial ESCC and 4746 images from 410 noncancerous esophagi to construct our AI system. For the validation dataset, we took NBI videos of suspected superficial ESCCs. The AI system diagnosis used one magnified still image taken from each video, while 19 endoscopists used whole videos. We used 147 videos and still images including 83 superficial ESCC and 64 non-ESCC lesions. The accuracy, sensitivity and specificity for the classification of ESCC were, respectively, 80.9% [95% CI 73.6-87.0], 85.5% [76.1-92.3], and 75.0% [62.6-85.0] for the AI system and 69.2% [66.4-72.1], 67.5% [61.4-73.6], and 71.5% [61.9-81.0] for the endoscopists. The AI system correctly classified all ESCCs invading the muscularis mucosa or submucosa and 96.8% of lesions ≥ 20 mm, whereas even the experts diagnosed some of them as non-ESCCs. Our AI system showed higher accuracy for classifying ESCC and non-ESCC than endoscopists. It may provide valuable diagnostic support to endoscopists.

Geriatric nutritional risk index as a prognostic indicator in elderly patients with early colorectal cancer undergoing endoscopic submucosal dissection.

OBJECTIVES: Data on the long-term outcomes of endoscopic submucosal dissection (ESD) performed in elderly patients with early colorectal cancer (CRC) are limited. We analyzed the prognosis of elderly CRC patients, not only from the viewpoint of treatment curability but also from the patients' baseline physical condition assessed by several indexes. METHODS: A retrospective analysis of 729 patients aged ≥75 years who underwent ESD for Tis/T1 CRC in 16 institutions was conducted. The patients were classified into three groups based on curability: curative ESD (Group A, n = 582), non-curative ESD with additional surgery (Group B, n = 60), and non-curative ESD without additional surgery (Group C, n = 87). Overall survival (OS) was compared among the groups, and factors associated with reduced OS were investigated. RESULTS: The median follow-up periods in Groups A, B, and C were 41, 49, and 46 months, respectively (P = 0.62), during which 92 patients died. Two patients (0.3%) in Group A, none (0%) in Group B, and three (3.4%) in Group C died of CRC. Three-year OS rates in Groups A, B, and C were 93.9%, 96.1%, and 90.1%, respectively, without a significant difference (P = 0.07). Multivariate analysis indicated low (<96.3) geriatric nutritional risk index (GNRI) as the sole independent predictor for reduced OS (hazard ratio 3.37; 95% confidence interval 2.18-5.22; P < 0.0001). CONCLUSIONS: Low GNRI, but not the curability attained by ESD, was independently associated with reduced OS in patients with early CRC aged ≥75 years.

Renal Dysfunction is an Independent Risk Factor for Rebleeding After Endoscopic Hemostasis in Patients with Peptic Ulcer Bleeding.

BACKGROUND: Despite the progress in endoscopic hemostasis and pharmacological treatment, the mortality rate of peptic ulcer bleeding remains at 5-10%. Rebleeding after peptic ulcer bleeding is believed to be a risk factor for mortality. This study aimed to evaluate whether renal dysfunction is a predictor of rebleeding after endoscopic hemostasis in patients with peptic ulcer bleeding. METHODS: In this retrospective study, consecutive patients with peptic ulcer bleeding who underwent endoscopic hemostasis at our Hospital from January 2010 to December 2018 were enrolled. The relationship between rebleeding within 30 days after endoscopic hemostasis and the patients' admission and endoscopic characteristics were analyzed using univariate and multivariate regression models. RESULTS: Out of 274 patients with peptic ulcer bleeding, 17 (6.2%) patients experienced rebleeding. In the analysis of the patients' admission characteristics, estimated glomerular filtration rate (eGFR) < 30 mL/min/1.73 m2 was an independent risk factor for rebleeding (odds ratio 4.77, 95% confidence interval 1.168-18.211, p = 0.03). Patients with eGFR < 15 mL/min/1.73 m2 with or without hemodialysis had the highest rebleeding rate at 36.8%. With respect to endoscopic characteristics, the rate of rebleeding was associated with combination therapy (p < 0.0001) and active bleeding (p = 0.03). CONCLUSION: Renal dysfunction might be an independent risk factor for rebleeding after endoscopic hemostasis in patients with peptic ulcer bleeding.

Helicobacter pylori eradication prevents secondary gastric cancer in patients with mild-to-moderate atrophic gastritis.

BACKGROUND AND AIM: Whether Helicobacter pylori eradication prevents metachronous recurrence after endoscopic resection (ER) of early gastric cancer remains controversial. This multicenter retrospective study aimed to evaluate the long-term (> 5 years) effects of H. pylori eradication by stratifying patients' baseline degrees of atrophic gastritis. METHODS: A total of 483 H. pylori-positive patients who had undergone ER for early gastric cancer were divided into two groups-(i) those having undergone successful H. pylori eradication within 1 year after ER (eradicated group, n = 294) and (ii) those with failed or not attempted H. pylori eradication (non-eradicated group, n = 189). The cumulative incidences of metachronous gastric cancer between the two groups were compared for all patients, for patients with mild-to-moderate atrophic gastritis (n = 182), and for patients with severe atrophic gastritis (n = 301). RESULTS: During a median follow-up of 5.2 years (range 1.1-14.8), metachronous cancer developed in 52 (17.7%) patients in the eradicated group and in 35 (18.5%) patients in the non-eradicated group (P = 0.11, log-rank test). In patients with mild-to-moderate atrophic gastritis (111 and 71 in the eradicated and non-eradicated groups, respectively), the cumulative incidence of metachronous cancer was significantly lower in the eradicated group than that in the non-eradicated group (P = 0.03, log-rank test). However, no significant intergroup difference was observed in patients with severe atrophic gastritis (P = 0.69, log-rank test). CONCLUSIONS: Helicobacter pylori eradication had a preventive effect on the development of metachronous gastric cancer in patients with mild-to-moderate atrophic gastritis.

Prophylactic clipping not effective in preventing post-polypectomy bleeding for < 20-mm colon polyps: A multicenter, open-label, randomized controlled trial.

BACKGROUND AND AIM: Prophylactic clipping (PC) after polypectomy has the potential to prevent post-polypectomy bleeding (PPB). We aimed to evaluate the effectiveness of PC in preventing PPB for < 20-mm polyps. METHODS: This multicenter, open-label, randomized controlled trial conducted from December 2013 to June 2017 at 10 institutions randomly assigned 1080 patients with < 20-mm colon polyps to the non-PC and PC groups. Allocation factors were institution, antiplatelet drug use, and polyp number. The primary endpoint was differences in PPB rates between the groups. The severity of PPB and post-procedural abdominal symptoms were also investigated. These endpoints in intention-to-treat and per-protocol (PP) analyses were evaluated. RESULTS: We investigated 1039 patients with 2960 lesions. There was no significant difference between the groups in characteristics including age, sex, hypertension, diabetes, hyperlipidemia, antiplatelet drug use, and lesion characteristics such as type and size. Excluding the clip used in the non-PC group, intraoperative bleeding, and deviation of protocol, 903 patients were investigated in PP analysis. There was no significant difference in the PPB rate between the non-PC and PC groups (2.7% vs 2.3%, P = 0.6973 [intention-to-treat analysis]; 3.0 vs 2.4%, P = 0.7353 [PP analysis]). Severe PPB (≥ grade 3) was similar between the groups. Total procedure time was significantly shorter in the non-PC group than in the PC group (31 vs 36 min, P = 0.0002). Post-procedural abdominal fullness was less common in the non-PC group than in the PC group (20.8% vs 25.6%, P = 0.0833). CONCLUSION: Prophylactic clipping is not effective in preventing PBB for < 20-mm colon polyps (UMIN000012163).

Effect of anticoagulants on the risk of delayed bleeding after colorectal endoscopic submucosal dissection.

Background and study aims In patients receiving antithrombotic therapy, the risks of delayed bleeding after endoscopic procedures for gastrointestinal neoplasms become a major problem. Few reports have shown the effects of delayed bleeding in patients taking anticoagulants after colorectal endoscopic submucosal dissection (ESD). This study aimed to evaluate the delayed bleeding events after colorectal ESD in patients receiving anticoagulant therapy. Patients and methods We retrospectively analyzed 87 patients taking anticoagulants who underwent colorectal ESD from April 2012 to December 2017 at 13 Japanese institutions participating in the Osaka Gut Forum. Among these patients, warfarin users were managed with heparin bridge therapy (HBT), continued use of warfarin, a temporary switch to direct oral anticoagulation (DOAC), or withdrawal of warfarin, and DOAC users were managed with DOAC discontinuation with or without HBT. We investigated the occurrence rate of delayed bleeding and compared the rates between warfarin and DOAC users. Results The delayed bleeding rate was 17.2 % among all patients. The delayed bleeding rate was higher in DOAC users than in warfarin users (23.3 % vs. 11.4 %, P  = 0.14), although no statistically significant difference was observed. In DOAC users, the delayed bleeding rates for dabigatran, rivaroxaban, apixaban, and edoxaban users appeared similar (30 %, 18.2 %, 22.2 %, and 25 %, respectively). The onset of delayed bleeding in both warfarin and DOAC users was late, averaging 6.9 and 9.4 days, respectively. Conclusions Among patients taking anticoagulants, the risk of delayed bleeding after colorectal ESD was relatively high and the onset of delayed bleeding was late.

Comparison of performances of artificial intelligence versus expert endoscopists for real-time assisted diagnosis of esophageal squamous cell carcinoma (with video).

BACKGROUND AND AIMS: Narrow-band imaging (NBI) is currently regarded as the standard modality for diagnosing esophageal squamous cell carcinoma (SCC). We developed a computerized image-analysis system for diagnosing esophageal SCC by NBI and estimated its performance with video images. METHODS: Altogether, 23,746 images from 1544 pathologically proven superficial esophageal SCCs and 4587 images from 458 noncancerous and normal tissue were used to construct an artificial intelligence (AI) system. Five- to 9-second video clips from 144 patients captured by NBI or blue-light imaging were used as the validation dataset. These video images were diagnosed by the AI system and 13 board-certified specialists (experts). RESULTS: The diagnostic process was divided into 2 parts: detection (identify suspicious lesions) and characterization (differentiate cancer from noncancer). The sensitivities, specificities, and accuracies for the detection of SCC were, respectively, 91%, 51%, and 63% for the AI system and 79%, 72%, and 75% for the experts. The sensitivity of the AI system was significantly higher than that of the experts, but its specificity was significantly lower. Sensitivities, specificities, and accuracy for the characterization of SCC were, respectively, 86%, 89%, and 88% for the AI system and 74%, 76%, and 75% for the experts. The receiver operating characteristic curve showed that the AI system had significantly better diagnostic performance than the experts. CONCLUSIONS: Our AI system showed significantly higher sensitivity for detecting SCC and higher accuracy for characterizing SCC from noncancerous tissue than endoscopic experts.

Classification for invasion depth of esophageal squamous cell carcinoma using a deep neural network compared with experienced endoscopists.

BACKGROUND AND AIMS: Cancer invasion depth is a critical factor affecting the choice of treatment in patients with superficial squamous cell carcinoma (SCC). However, the diagnosis of invasion depth is currently subjective and liable to interobserver variability. METHODS: We developed a deep learning-based artificial intelligence (AI) system based on Single Shot MultiBox Detector architecture for the assessment of superficial esophageal SCC. We obtained endoscopic images from patients with superficial esophageal SCC at our facility between December 2005 and December 2016. RESULTS: After excluding poor-quality images, 8660 non-magnified endoscopic (non-ME) and 5678 ME images from 804 superficial esophageal SCCs with pathologic proof of cancer invasion depth were used as the training dataset, and 405 non-ME images and 509 ME images from 155 patients were selected for the validation set. Our system showed a sensitivity of 90.1%, specificity of 95.8%, positive predictive value of 99.2%, negative predictive value of 63.9%, and an accuracy of 91.0% for differentiating pathologic mucosal and submucosal microinvasive (SM1) cancers from submucosal deep invasive (SM2/3) cancers. Cancer invasion depth was diagnosed by 16 experienced endoscopists using the same validation set, with an overall sensitivity of 89.8%, specificity of 88.3%, positive predictive value of 97.9%, negative predictive value of 65.5%, and an accuracy of 89.6%. CONCLUSIONS: This newly developed AI system showed favorable performance for diagnosing invasion depth in patients with superficial esophageal SCC, with comparable performance to experienced endoscopists.

[Anaphylactic shock secondary to the use of macrogol 4000 as a bowel cleanser:a case report].

A 44-year-old man was administered Niflec® containing macrogol 4000 as a bowel cleanser for colonoscopic examination. Immediately after ingestion, he experienced oral cavity discomfort and nasal congestion, followed by acute urticaria and presyncope. His systolic blood pressure and peripheral capillary oxygen saturation dropped to 66mmHg and 89%, respectively. Fluid infusion, as well as steroid and epinephrine administration, improved his symptoms. Skin prick tests were then performed using Niflec®, macrogol 4000, and Actosin® ointment (containing macrogol 4000), all of which were positive. Therefore, the patient was diagnosed with anaphylactic shock caused by macrogol 4000 included in Niflec®. Macrogol present in bowel cleansers used for colonoscopy rarely causes anaphylactic shock. However, clinicians need to be mindful of this risk. Prompt and appropriate treatment is needed should this condition occur.

[Good Controlled Chylothorax in Gastric Carcinoma].

Chylothorax has been reported to be caused by accidental injuries in half of all cases in Japan, and < 10% of these cases have been associated with malignant tumors, including lymphoma. Chylothorax is a rare complication of gastric carcinoma. We successfully treated a 58-year-old man with gastric carcinoma, chylothorax, and ascites using a combination of talc pleurodesis and a lipid-limited diet. Case: A 58-year-old man with advanced stage of poorly differentiated gastric adenocarcinoma presented to our hospital with complaints of shortness of breath. Whole-body computerized tomographic images suggested massive pleural effusion and ascites. Examination of pleural fluid and ascites revealed elevated serum triacylglycerol levels of up to 913mg/dL with numerous free-floating cancer cells. Malignant chylothorax was diagnosed. A lipid-limited diet and octreotide were started, followed by talc pleurodesis for pleural effusion. The patient with controlled pleurisy died of gastric cancer on day 55 after pleurodesis.

Prophylactic clip closure may reduce the risk of delayed bleeding after colorectal endoscopic submucosal dissection.

BACKGROUND AND STUDY AIMS: Endoscopic submucosal dissection (ESD) has a high en bloc resection rate and is widely performed for large superficial colorectal tumors, but delayed bleeding remains one of the most common complications of colorectal ESD. The aim of the present study was to evaluate the clinical efficacy of prophylactic clip closure of mucosal defects for the prevention of delayed bleeding after colorectal ESD. PATIENTS AND METHODS: We enrolled consecutive patients with colorectal lesions between January 2012 and May 2017 in this retrospective study. In the early part of this period, post-ESD mucosal defects were not closed (non-closure group); however, from January 2014, post-ESD mucosal defects were prophylactically closed with clips when possible (closure group). The main outcome measured was delayed bleeding. Variables were analyzed using the chi-squared test, Fisher's exact test, or Student's t-test. RESULTS: Of 156 lesions analyzed, 61 were in the non-closure group and 95 in the closure group. Overall, delayed bleeding occurred in 5 cases (3.2 %). The delayed bleeding rate was 0 % (0/95) in the closure group and 8.2 % (5/61) in the non-closure group ( P  = 0.008). The mean procedure time for closure was 10.4 ±â€Š4.6 min (range 3 - 26 min). CONCLUSIONS: We demonstrated that prophylactic clip closure of mucosal defects might reduce the risk of delayed bleeding after colorectal ESD.