Artificial intelligence-based diagnosis of upper gastrointestinal subepithelial lesions on endoscopic ultrasonography images.

BACKGROUND: Endoscopic ultrasonography (EUS) is useful for the differential diagnosis of subepithelial lesions (SELs); however, not all of them are easy to distinguish. Gastrointestinal stromal tumors (GISTs) are the commonest SELs, are considered potentially malignant, and differentiating them from benign SELs is important. Artificial intelligence (AI) using deep learning has developed remarkably in the medical field. This study aimed to investigate the efficacy of an AI system for classifying SELs on EUS images. METHODS: EUS images of pathologically confirmed upper gastrointestinal SELs (GIST, leiomyoma, schwannoma, neuroendocrine tumor [NET], and ectopic pancreas) were collected from 12 hospitals. These images were divided into development and test datasets in the ratio of 4:1 using random sampling; the development dataset was divided into training and validation datasets. The same test dataset was diagnosed by two experts and two non-experts. RESULTS: A total of 16,110 images were collected from 631 cases for the development and test datasets. The accuracy of the AI system for the five-category classification (GIST, leiomyoma, schwannoma, NET, and ectopic pancreas) was 86.1%, which was significantly higher than that of all endoscopists. The sensitivity, specificity, and accuracy of the AI system for differentiating GISTs from non-GISTs were 98.8%, 67.6%, and 89.3%, respectively. Its sensitivity and accuracy were significantly higher than those of all the endoscopists. CONCLUSION: The AI system, classifying SELs, showed higher diagnostic performance than that of the experts and may assist in improving the diagnosis of SELs in clinical practice.

Randomized Phase II Study of Consecutive-Day versus Alternate-Day Treatment with S-1 as Second-Line Chemotherapy in Advanced Pancreatic Cancer.

OBJECTIVE: To evaluate the efficacy and safety of alternate-day administration of S-1 as second-line chemotherapy for unresectable pancreatic cancer in a multicenter, randomized, phase II study. METHODS: Patients with histologically proven, unresectable pancreatic cancer treated with chemotherapy not including S-1 as first-line therapy were randomly assigned to receive either daily or alternate-day treatment with S-1. The primary end point was overall survival (OS), and the secondary end points were progression-free survival (PFS), time to treatment failure (TTF), response rate, and adverse events. RESULTS: A total of 77 patients were enrolled, of which 75 were included in the final analysis. The median OS was 4.5 months in the daily group and 4.4 months in the alternate-day group (HR 1.178; 95% CI 0.741-1.875), with no significance in PFS and TTF. The response rate was 2.8% in the daily group and 0% in the alternate-day group. Grade 3 or higher adverse events occurred with significantly higher incidence in the daily group (47.2 vs. 25.6%, p = 0.044). CONCLUSION: As a second-line chemotherapy for unresectable pancreatic cancer, although the efficacy in both groups was comparable and we can expect fewer toxicities with alternate-day administration of S-1, the noninferiority of alternate-day treatment to daily treatment with S-1 was not verified.

Natural history of pancreatic cystic lesions: A multicenter prospective observational study for evaluating the risk of pancreatic cancer.

BACKGROUND AND AIM: The aim of this study is to elucidate the natural history of pancreatic cystic lesions (PCLs), including branch duct-type intraductal papillary mucinous neoplasm (BD-IPMN), via midterm follow-up analysis of a multicenter prospective observational study (NSPINAL study). METHODS: From July 2011 to October 2016, 881 patients with PCLs were enrolled in NSPINAL study, and 664 patients with > 12 months of follow up were analyzed. Every patient was asymptomatic, and endoscopic ultrasound was performed at the initial diagnosis to exclude high-risk individuals. Follow up included endoscopic ultrasound, computed tomography, or magnetic resonance imaging at least once a year. Serial morphological changes and the pancreatic cancer (PC) incidence, including malignant progression of PCLs, were evaluated. RESULTS: The 664 patients (358 men) were followed for a median of 33.5 months (interquartile range 29). The cyst and main pancreatic duct sizes were 16.6 ± 9.3 and 2.3 ± 1.0 mm, respectively. Morphologically, 518 cases were multilocular, 137 were unilocular, and 9 had a honeycomb pattern; 269 cases involved multifocal lesions. Ninety-six patients (14.5%) showed worsening progression on imaging. There were two resectable and four unresectable cases of pancreatic ductal adenocarcinoma and three cases of malignant BD-IPMN. The 3-year risk of developing PC was 1.2%. The standardized incidence ratio for PC among PCLs was 10.0 (95% confidence interval 3.5-16.5), and the standardized incidence ratio among BD-IPMN was 16.6 (95% confidence interval 5.1-28.1). Multivariate analysis showed that development of symptoms and worsening progression were significant predictors of PC. CONCLUSIONS: Malignant progression of PCLs, including PC development, is not uncommon. Patients with PCLs should be carefully monitored to detect pancreatic ductal adenocarcinoma at early stages.

Risk factors for rebleeding in gastroduodenal ulcers.

BACKGROUND: Rebleeding after hemostasis of the gastroduodenal ulcer (GDU) is one of the indicators associated with death among GDU patients. However, there are few studies on risk score that contribute to rebleeding after endoscopic hemostasis of bleeding peptic ulcers. AIMS: The aim of this study was to identify factors associated with rebleeding, including patient factors, after endoscopic hemostasis of bleeding gastroduodenal ulcers and to stratify the risk of rebleeding. METHODS: We retrospectively enrolled 587 consecutive patients who were treated for Forrest Ia to IIa bleeding gastroduodenal ulcers with endoscopic hemostasis at three institutions. Risk factors associated with rebleeding were assessed using univariate and multivariate logistic regression analyses. The Rebleeding Nagoya University (Rebleeding-N) scoring system was developed based on the extracted factors. The Rebleeding-N score was internally validated using bootstrap resampling methods. RESULTS: Sixty-four patients (11%) had rebleeding after hemostasis of gastroduodenal ulcers. Multivariate logistic regression analysis revealed four independent rebleeding risk factors: blood transfusion, albumin <2.5, duodenal ulcer, and diameter of the exposed vessel ≧2 mm. Patients with 4 risk factors in the Rebleeding-N score had a 54% rebleeding rate, and patients with 3 risk factors had 44% and 25% rebleeding rates. In the internal validation, the mean area under the curve of the Rebleeding-N score was 0.830 (95% CI = 0.786-0.870). CONCLUSIONS: Rebleeding after clip hemostasis of bleeding gastroduodenal ulcers was associated with blood transfusion, albumin <2.5, diameter of the exposed vessel ≧2 mm, and duodenal ulcer. The Rebleeding-N score was able to stratify the risk of rebleeding.

[Complete necrosis of a giant tumor in liver by transcatheter arterial embolization and percutaneous transhepatic portal embolization before liver resection].

A 67-year-old man admitted for abdominal pain. Abdominal CT showed a liver tumor 14 cm in diameter in the right lobe, and a low density area spreading along the left branch of the Glisson's sheath. Serum protein induced by vitamin K absence or antagonist-II (PIVKA-II) level was found extremely high. Transcatheter arterial embolization (TAE) was performed three times, resulting the lesion reduced to 7cm in diameter. As the last TAE showed little effectiveness, we planed to percutaneous transhepatic portal embolization (PTPE) followed by extended right hepatectomy. After PTPE, PIVKA-II was normalized. Histologically, complete necrosis of the tumor lesions was observed in the resected liver.

Management of Pancreatolithiasis: A Nationwide Survey in Japan.

OBJECTIVES: The aim of this study was to assess prevailing treatment of pancreatolithiasis in Japan. METHODS: We surveyed clinical data from 1834 patients (1479 men and 355 women) at 125 hospitals. RESULTS: Extracorporeal shock-wave lithotripsy (ESWL) was performed alone in 103 patients (5.6%), ESWL plus an endoscopic procedure in 446 (24.3%), endoscopic treatment alone in 261 (14.2%), and surgery in 167 (9.1%). Other treatments were given to 358 (19.5%), whereas 499 (27.2%) received no treatment. Symptoms were relieved in 85.7% after ESWL, 80.8% after endoscopic treatment alone, and 92.8% after surgery. Early complication rates within 3 months after ESWL, endoscopic treatment alone, and surgery were 8%, 4.5%, and 27.1%, respectively. Late complications after ESWL, endoscopic procedures alone, and surgery were 1.7%, 2.5%, and 8.2%, respectively. Symptom relief but also early and late complications were greater after surgery than after ESWL and endoscopic treatment. Among 417 patients undergoing ESWL, 61 (14.6%) required surgery, as did 32 (16%) of 200 patients treated endoscopically. Surgery was required less frequently following initial operative treatment (11/164 patients [6.7%]). Nonsurgical initial treatments were chosen more frequently. CONCLUSIONS: First-line treatment of pancreatolithiasis should be ESWL with or without endoscopy because of minimal invasiveness and fewer complications.

Phenotypes and Chronic Organ Damage May Be Different among Siblings with Wilson's Disease.

Background and Aims: Cloning of ATP7B provided evidence that Wilson's disease is a hepatic copper toxicosis with a variety of extrahepatic complications. Affected siblings with the same genetic background and exposure to similar environmental factors may be a good model for the study of genotype-phenotype correlation. Methods: Twenty-three affected siblings in 11 families were selected from a database. The first phenotypes were determined according to the international proposal. The final types of chronic organ damage were re-evaluated for life-long management. Results: Phenotypes were identical in 5 of the families and different in 6 of the families. The acute hepatic phenotype H1 was found in 3 younger siblings and 1 older sibling. All survived an acute episode of hemolysis with underlying chronic liver disease. One also presented complication with neurological disease. The neurological phenotype N1 with neuropsychiatric symptoms and hepatic disease was found in 2 aged siblings of 1 family, in an older sibling in 3 families and in the oldest sibling in 1 family. Phenotypes in siblings were mainly split by either H1 occurring in random order or age-dependent N1. Types of chronic organ damage were identical in 8 of the families and different in 3 of the families. The same combination of chronic liver disease was found in 6 families and chronic liver disease complicated with neurological disease in 2 families. Split organ damage in siblings was found when an older sibling was complicated by neurological disease. There was no reverse combination of a younger sibling being complicated by neurological disease in any of the families. Conclusion: Phenotype combinations of siblings were mainly modified by externally-induced hemolytic episodes, while chronic organ damage in siblings was split by age-dependent neurological complications.

[A case of intestinal intussusception caused by multiple lipoma diagnosed preoperatively].

A 45-year-old man presented with intermittent abdominal pain for a month. Intestinal intussusception was diagnosed by the findings of abdominal ultrasonography and computed tomography. Intussusception was reduced by barium enema. Based on magnetic resonance imaging, we diagnosed intestinal intussusception due to multiple lipoma. At laparotomy, after successful reduction of the intussusception with Hutchinson's maneuver, an ileoceal resection was performed. Intussusception in adults is relatively rare and may be difficult to diagnose preoperatively. In the diagnosis of this disease, abdominal echo, CT, and MRI are useful.

Efficacy of STI571 for a patient with metastatic gastrointestinal stromal tumor.

A 65-year-old man was referred to our hospital for evaluation of his huge abdominal tumor. He was diagnosed as having a gastrointestinal stromal tumor arising from the stomach. Seven months after surgery, multiple liver metastases and mesenteric dissemination occurred. He was medicated with STI571, which works by blocking proliferation of malignant cells with expression of c-kit. The tumors shrank and serum lactate dehydrogenase and alkaline phosphatase concentrations fell to below the normal limit three months later. STI571 was effective medicine for the metastatic gastrointestinal stromal tumor for six months in this case.