RESUMEN
Objective:To evaluate artificial intelligence constructed by deep convolutional neural network (DCNN) for the site identification in upper gastrointestinal endoscopy.Methods:A total of 21 310 images of esophagogastroduodenoscopy from the Cancer Hospital of Chinese Academy of Medical Sciences from January 2019 to June 2021 were collected. A total of 19 191 images of them were used to construct site identification model, and the remaining 2 119 images were used for verification. The performance differences of two models constructed by DCCN in the identification of 30 sites of the upper digestive tract were compared. One model was the traditional ResNetV2 model constructed by Inception-ResNetV2 (ResNetV2), the other was a hybrid neural network RESENet model constructed by Inception-ResNetV2 and Squeeze-Excitation Networks (RESENet). The main indices were the accuracy, the sensitivity, the specificity, positive predictive value (PPV) and negative predictive value (NPV).Results:The accuracy, the sensitivity, the specificity, PPV and NPV of ResNetV2 model in the identification of 30 sites of the upper digestive tract were 94.62%-99.10%, 30.61%-100.00%, 96.07%-99.56%, 42.26%-86.44% and 97.13%-99.75%, respectively. The corresponding values of RESENet model were 98.08%-99.95%, 92.86%-100.00%, 98.51%-100.00%, 74.51%-100.00% and 98.85%-100.00%, respectively. The mean accuracy, mean sensitivity, mean specificity, mean PPV and mean NPV of ResNetV2 model were 97.60%, 75.58%, 98.75%, 63.44% and 98.76%, respectively. The corresponding values of RESENet model were 99.34% ( P<0.001), 99.57% ( P<0.001), 99.66% ( P<0.001), 90.20% ( P<0.001) and 99.66% ( P<0.001). Conclusion:Compared with the traditional ResNetV2 model, the artificial intelligence-assisted site identification model constructed by RESENNet, a hybrid neural network, shows significantly improved performance. This model can be used to monitor the integrity of the esophagogastroduodenoscopic procedures and is expected to become an important assistant for standardizing and improving quality of the procedures, as well as an significant tool for quality control of esophagogastroduodenoscopy.
RESUMEN
Objective:To evaluate the efficacy and safety of endoscopic submucosal dissection (ESD) for early hypopharyngeal carcinoma and precancerous lesions.Methods:Clinical data of 41 patients who received ESD for early hypopharyngeal carcinoma and precancerous lesions from August 2013 to August 2019 in the Department of Endoscopy of Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College were retrospectively analyzed. Main outcome measurements included operation completion rate, operation time, en bloc resection rate, R0 resection rate, complication rate and recurrence.Results:ESD was successfully completed in all 41 cases, with a success rate of 100.0% and a mean time of 49.1 min (ranged 10-110 min). Fifty-four lesions underwent en bloc resection, with an en bloc resection rate of 98.2% (54/55), of which 41 had negative horizontal and vertical margins, and the R0 resection rate was 74.5% (41/55). During the operation of 55 lesions, there was a small amount of blood oozing on the wound surface, and electrocoagulation with thermal biopsy forceps could successfully stop the bleeding. No perforation occurred, and 2 cases (4.3%) had delayed bleeding after ESD, and hemostasis was successful under emergency endoscopy. Postoperative endoscopy showed that 1 case (2.2%) had esophageal entrance stenosis, and the obstruction was relieved after repeated water balloon dilatation. The follow-up period ranged from 3 to 72 months, and the median time was 18 months. One case was found to have mucosal lesions in the same part of the hypopharynx and received ESD treatment again. Follow-up to October 2020, no residual lesions and recurrence were found.Conclusion:ESD is a safe and effective option for the treatment of early hypopharyngeal carcinoma and precancerous lesions, which is worthy of clinical application.
RESUMEN
Objective:To compare the clinical effect of three indwelling methods of plastic biliary stent on relieving obstructive jaundice caused by unresectable hilar cholangiocarcinoma.Methods:A retrospective study was performed on data of 61 patients with obstructive jaundice caused by unresectable hilar cholangiocarcinoma from April 2014 to December 2020 in Cancer Hospital, Chinese Academy of Medical Sciences. Plastic biliary stent placement was used to relieve jaundice, including 18 cases of intragastric indwelling at the end of biliary stent, 31 cases of duodenal papilla indwelling at the end of biliary stent, and 12 cases of horizontal portion of duodenum indwelling at the end of biliary stent. Incidence of fever within 2 weeks, perioperative mortality, 90-day obstruction rate, and median stent patency period were followed up and the results were analyzed.Results:The incidence of fever within 2 weeks of the three groups were significantly different [66.7% (12/18), 58.1% (18/31) and 16.7% (2/12), χ2=7.30, P=0.026]. There were no statistically differences in the perioperative mortality [0 (0/16), 3.2% (1/31) and 0 (0/10), χ2=1.09, P=1.000], 90-day obstruction rate [52.9% (9/17), 48.3% (14/29) and 40.0% (4/10), χ2=1.91, P=0.589], or median stent patency period (66.0 d, 91.5 d and 94.0 d, Z=4.96, P=0.084) among three groups. Conclusion:Patients with biliary plastic stents with ends placed at the horizontal portion of the duodenum show lower incidence of fever within two weeks after implantation, and similar median stent patency period, 90-day obstruction rate and perioperative mortality compared with intragastric indwelling and duodenal papilla indwelling groups. Therefore, biliary plastic stents with ends placed at the horizontal portion of the duodenum should be recommended as the preferred procedure.
RESUMEN
Objective:To determine risk factors for postoperative esophageal refractory stenosis after endoscopic submucosal dissection (ESD) of large-scale early esophageal carcinomas and precancerous lesions.Methods:Two hundred and twelve early esophageal carcinomas or precancerous lesions in 186 patients who underwent ESD larger than 3/4 the total esophageal circumference in Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, between July 2013 and December 2017 were divided into two groups according to session number of endoscopic balloon dilatation (EBD), the refractory stenosis group ( n=69, ≥6 EBD sessions) and non-refractory stenosis group ( n=117, ≤5 EBD sessions). Student′s t-test or Mann-Whitney U test was used for univariate analysis and χ2 test and Fisher exact test were used for comparison of categorical variables. Logistic regression was used for multivariate analysis. Results:Compared with the non-refractory stenosis group, the refractory stenosis group had statistically significant differences in the longitudinal diameter of lesions, the longitudinal diameter of artificial ulcer, lesion location, the circumferential range of lesions and the composition of the muscular layer injury (all P<0.05). After eliminating the factor of the vertical diameter of artificial ulcer (because there was significant correlation between the vertical diameter of artificial ulcer and the longitudinal diameter of lesion in clinical practice), multivariate logistic regression analysis showed that the longitudinal diameter of lesion>5 cm (VS ≤5 cm: P=0.003, OR=3.531, 95% CI:1.547-8.060), the location of lesion in the upper thoracic segment (VS lower thoracic segment: P=0.001, OR=36.720, 95% CI:4.233-318.551), in the cervical segment (VS lower thoracic segment: P=0.003, OR=24.959, 95% CI:2.927-212.795), the whole circumferential lesion (VS ≥3/4 but not the whole circumference: P<0.001, OR=10.082, 95% CI:4.196-24.226) and the presence of muscular layer injury ( P<0.001, OR=7.128, 95% CI:2.748-18.486) were more likely to lead to esophageal refractory stenosis after ESD. Conclusion:The longitudinal lesion diameter of more than 5 cm, the circumferential extent of esophageal ESD, cervical or upper-thoracic esophageal lesions, and muscular layer damage are independent risk factors for postoperative esophageal refractory stenosis after ESD for large-scale esophageal cancer and precancerous lesions.
RESUMEN
Objective@#To evaluate the short-term outcomes and safety of submucosal tunneling endoscopic resection (STER) for submucosal tumors (SMT) originating from muscularis propria (MP) layer at esophagogastric junction.@*Methods@#The clinical data of 31 patients with SMT originating from MP layer at esophagogastric junction underwent STER were collected and retrospectively analyzed.@*Results@#The success rate of STER of the thirty-one patients was 100%. The mean tumor size was (2.5±1.3) cm and the average operative time was (95.9±56.7) min. Perforation occurred in 3 patients and was successfully clipped by endo-clips during operation. One patient developed delayed bleeding and the bleeding was stopped by endoscopic hemostasis. Twenty-nine leiomyomas and two stromal tumors (GIST) were finally pathologically diagnosed. No local recurrence and distant metastasis were noted during the mean 15.4 months follow-up of 20 cases. According to the lesion size, 31 patients who received STER were divided into two groups. The operation time of maximum diameter ≥3.5 cm group was (134.0±70.6) min, significantly longer than (80.3±42.6) min of maximum diameter <3.5 cm group (P=0.014). However, the en bloc removal rate, postoperative hospital stay and the complication incidence between the two groups had no obvious differences (P>0.05). Univariate analysis showed that the piecemeal removal group had longer tumor diameter, higher incidence of irregular tumor morphology, and longer operative time than the en bloc removal group (all P<0.05). Stepwise logistic regression analysis showed that irregular shape was a risk factor for failure of en bloc removal (OR=18.000, 95% CI: 1.885~171.88, P=0.012).@*Conclusion@#As a new method of minimally invasive treatment, STER technology appears to be a safe and effective option for patients with SMT originating from MP layer at esophagogastric junction.
RESUMEN
Objective@#To investigate the therapeutic strategy in patients with early gastric cancer after noncurative endoscopic submucosal dissection (ESD).@*Methods@#A total of 107 cases with early gastric cancer receiving noncurative endoscopic submucosal dissection were collected and the patients were classified into an additional gastrectomy group (n=41) and a simple follow-up group (n=66) according to the therapeutic method used after noncurative ESD. The clinicopathological information, short- and long-term clinical outcomes between the two groups were analyzed and compared.@*Results@#The mean age of the patients in the gastrectomy group and follow-up group was(59.2±8.7)years old and(64.7±8.8)years old, respectively. The depth of submucosal invasion was (1445.83±803.12) and (794.71±815.79) μm, respectively. The difference between the two groups was statistically significant (P=0.020 for age and P=0.010 for depth of submucosal invasion). Compared with follow-up group, the patients with undifferentiated histologic type, deep invasion of submucosa (SM2), diffuse type, lymphovascular invasion and neural invasion were more common in the gastrectomy group (P<0.05). The R0 resection rate of ESD in the gastrectomy group was significantly lower than the follow-up group(26.8% vs 65.2%, P<0.001). The positive residual tumor rate and LNM rate of additional gastrectomy group were 31.7%(13/41)and 9.76%(4/41)according to the pathological results after gastrectomy. The gastrectomy group had 2 cases of local recurrence (2/41, 4.9%), while 5(5/66, 7.6%)in the follow-up group(4.9% vs 7.6%, P=0.883). There was no significant difference in overall survival (OS) and disease-free survival (DFS) between the two study groups (P=0.066 and 0.938, respectively).@*Conclusions@#Assessment of LNM risk should be performed in patients with noncurative endoscopic resection. For patients with low risk of LNM who are intolerance of additional gastrectomy due to old age and comorbidities, close follow-up with endoscopy can be considered as an alternative.
RESUMEN
Objective@#To investigate the risk factors and survival status of hypopharyngeal carcinoma with synchronous second primary carcinoma of the esophagus.@*Methods@#One hundred and sixty patients with newly diagnosed hypopharyngeal carcinoma from January 2009 to December 2012 were retrospectively reviewed. The clinical data, tumor-related information and follow-up results were collected and analyzed.@*Results@#Forty-three synchronous esophageal carcinomas (27%) were detected in 160 patients with hypopharyngeal carcinoma, and most patients (72%) were at an early stage. On univariate analysis, the median age of less than 55 years old (χ2=4.525, P=0.033), excessive alcohol consumption (χ2=6.942, P=0.008) and invasion site more than 3 anatomical regions (χ2=21.503, P=0.000) had a significant correlation with the occurrence of synchronous esophageal carcinomas. Multivariate analysis showed that excessive alcohol consumption (OR=4.787, P=0.029) and invasion site more than 3 anatomical regions (OR=14.391, P=0.000) were independent risk factors. The median survival time was 26 months in 43 patients with synchronous esophageal carcinomas, which was significantly lower than that (58 months) in patients without secondary primary esophageal carcinomas (χ2=11.981, P=0.001).@*Conclusions@#There is a high incidence of synchronous esophageal carcinoma in hypopharyngeal carcinoma patients, affecting the prognosis of hypopharyngeal carcinoma. Surveillance for esophageal carcinomas in patients with hypopharyngeal carcinoma, especially in excessive alcohol drinkers, is warranted.
RESUMEN
<p><b>OBJECTIVE</b>To investigate the security and efficacy of a new endoscopic closure method of large defects after endoscopic full-thickness resection (EFTR) double purse-string suture using two endoloops and metallic clips via single-channel endoscopy.</p><p><b>METHODS</b>Clinical data of 23 cases with submucosal tumors (SMT) who received endoscopic resection from June 2015 to July 2016 in our National Cancer Center were collected. For gastric and esophageal SMTs or the mucosa layer injured during submucosal tunneling endoscopic resection (STER), double purse-string suture was conducted after EFTR. The key steps of closure were as follows: the endoloop was installed onto the delivery system and inserted into the gastric cavity to the defect location with endoscopy and then opened; the clips were transported into the gastric cavity from the biopsy channel; the endoloop was fixed onto the full thickness of gastric wall along the edge of the defect by clips one by one between the interval of about 5 mm; the endoloop was tightened slowly till the entire circumference of the defect was sutured, thus, one purse-string suture was done; in accordance with the operation above, another endoloop was released, and the second endoloop was fixed at 5-10 mm to the outer edge of the original one, and tied the endoloop gently; this sequence was continued till there was no gap, thus, the double-purse string suture was finished. A total of 23 patients were enrolled in the study, including 18 with gastric tumor and 5 with esophageal tumor, 15 males and 8 females, with the average age of 56 (19 to 76) years.</p><p><b>RESULTS</b>Eighteen cases of gastric SMT were successfully treated by endoscopic EFTR and double purse-string suture. The esophageal mucosa layer of all the 5 cases of esophageal SMT, including tumors of 3 cases located in cervical esophagus at 15-20 cm from the fore-tooth, 1 esophageal leiomyoma case complicated with squamous cell carcinoma in situ, and 1 case of mucosal layer injury during submucosal tunneling endoscopic resection (STER), was successfully repaired by using double purse-string suture. The mean maximum diameter of tumor was 2.3 cm, and the average suture time was 22.8 min. Postoperative pathology showed that 13 cases were gastrointestinal stromal tumors (GIST), 7 cases were leiomyoma, 2 cases were neurilemmoma, and 1 case was leiomyoma complicated with early squamous cell carcinoma in situ. No severe complications occurred during or after the operation.</p><p><b>CONCLUSIONS</b>The double purse-string suture by using metallic clips and endoloops with single channel endoscope is a relatively safe, easy, and reliable technique for repairing large gastric defect after EFTR. For cervical esophageal SMT, or the SMT combined with superficial mucosal lesions, and for the mucosa layer injury during submucosal tunneling endoscopic resection(STER), double purse-string suture is helpful to perform the closure.</p>
RESUMEN
<p><b>OBJECTIVE</b>To explore the risk factors contributing to the progression-free survival rate of patients undergoing endoscopic non-curative resection.</p><p><b>METHODS</b>Clinicopathological data of patients with early colorectal carcinoma and intraepithelial neoplasia undergoing endoscopic resection in our department from January 2009 to January 2015 were collected. Associated factors affecting the progression-free survival rate of the early colorectal carcinoma after endoscopic non-curative resection were analyzed. Any of the following conditions was defined as endoscopic non-curative resection: (1) positive lateral or vertical cutting margin; (2) submucosa invasion depth ≥1 000 μm; (3) vascular or lymphatic invasion; (4) low differentiation, including signet ring cell carcinoma or mucinous adenocarcinoma; (5) high grade tumor budding.</p><p><b>RESULTS</b>Clinicopathological data of 840 cases were collected. According to China's Endoscopic Screening, Diagnosis and Treatment Guidelines for Early Colorectal Cancer, 56(56/840, 6.7%) cases were defined as the non-curative resection, the metastasis or recurrence rate was 14.3%(8/56), 3-year progression-free survival rate was 85.7%(48/56), and 3-year overall survival rate was 94.6%(53/56). Univariate prognostic analysis showed that 3-year progression-free survival rate in low and moderate-high differentiation adenocarcinoma was 25.0% and 90.4%(χ=6.711, P=0.010), in patients with submucosa invasion depth ≥2 000 μm and <2 000 μm was 75.0% and 93.8%(χ=6.745, P=0.009), and in patients with and without vascular or lymphatic invasion was 60.0% and 88.2%(χ=7.708, P=0.005), whose differences were all significant. Multivariate Cox regression analysis revealed that low differentiation adencarcinoma (P=0.015, HR=8.021, 95%CI: 1.499-42.921), invasion depth ≥2 000 μm (HR=6.823, 95%CI: 1.299-35.848) and vascular or lymphatic invasion (HR=18.143, 95%CI: 2.079-158.358) were independent risk factors for the progression-free survival rate of the early colorectal carcinoma after endoscopic non-curative resection.</p><p><b>CONCLUSION</b>Pathology after endoscopic non-curative resection for early colorectal carcinoma indicates that low differentiation adenocarcinoma, submucosa invasion depth ≥2 000 μm and vascular or lymphatic invasion are independent risk factors of poor prognosis.</p>
RESUMEN
Objective To compare the efficacy, safety, and the life quality of patients with early gastric cancer ( EGC) between endoscopic submucosal dissection ( ESD) and surgical treatment. Methods A total of 460 cases with EGC receiving endoscopic therapy or surgical treatment were collected from October 2009 to January 2015 in the Cancer Hospital, Chinese Academy of Medical Sciences. The clinical efficacy and life quality of ESD and surgical treatment for EGC patients were retrospectively analyzed. Results There were 434 cases collected in the study, including 208 cases ( 229 lesions) in the ESD group and 226 cases in the surgery group. For the short-term clinical outcomes of the ESD group, the hospitalization time ( 7. 85 ± 3. 18 d VS 16. 68±5. 89 d, P<0. 001), hospitalization cost (3782. 30±1898. 84 CNY VS 9685. 60± 3643. 97 CNY, P<0. 001 ) and complications [ 0 ( 0/208 ) VS 6. 2% ( 14/226 ) , P<0. 001 ] were statistically different compared with those of the surgery group. For the long-term clinical outcomes, there was no statistical significance on recurrence rate[0. 4%(1/229) VS 0. 9% (2/226), P=0. 622] between the two groups. The cumulative multiple hazard probability curve showed that the ESD group had a significantly higher risk of multiple primary lesions than the surgery group ( P=0. 004) after the same follow-up period. In order to exclude the influence of confounding factors, COX regression model was used to control the age and other factors, and multiple primary risks of the two groups were also statistically significant ( P=0. 013) . The health score of self-evaluation and life quality between the two groups were statistically significant ( P<0. 001) . Conclusion For the short-term clinical outcomes, the ESD group was better than the surgery group. For the long-term clinical outcomes, multiple primary risks were higher in the ESD group than those in the surgery group, but most of the multiple primary cases were successfully treated with a second ESD. The health score of self-evaluation and life quality were better in the ESD group than those in the surgery group.
RESUMEN
Objective To compare the efficacy, safety, and the life quality of patients with early gastric cancer ( EGC) between endoscopic submucosal dissection ( ESD) and surgical treatment. Methods A total of 460 cases with EGC receiving endoscopic therapy or surgical treatment were collected from October 2009 to January 2015 in the Cancer Hospital, Chinese Academy of Medical Sciences. The clinical efficacy and life quality of ESD and surgical treatment for EGC patients were retrospectively analyzed. Results There were 434 cases collected in the study, including 208 cases ( 229 lesions) in the ESD group and 226 cases in the surgery group. For the short-term clinical outcomes of the ESD group, the hospitalization time ( 7. 85 ± 3. 18 d VS 16. 68±5. 89 d, P<0. 001), hospitalization cost (3782. 30±1898. 84 CNY VS 9685. 60± 3643. 97 CNY, P<0. 001 ) and complications [ 0 ( 0/208 ) VS 6. 2% ( 14/226 ) , P<0. 001 ] were statistically different compared with those of the surgery group. For the long-term clinical outcomes, there was no statistical significance on recurrence rate[0. 4%(1/229) VS 0. 9% (2/226), P=0. 622] between the two groups. The cumulative multiple hazard probability curve showed that the ESD group had a significantly higher risk of multiple primary lesions than the surgery group ( P=0. 004) after the same follow-up period. In order to exclude the influence of confounding factors, COX regression model was used to control the age and other factors, and multiple primary risks of the two groups were also statistically significant ( P=0. 013) . The health score of self-evaluation and life quality between the two groups were statistically significant ( P<0. 001) . Conclusion For the short-term clinical outcomes, the ESD group was better than the surgery group. For the long-term clinical outcomes, multiple primary risks were higher in the ESD group than those in the surgery group, but most of the multiple primary cases were successfully treated with a second ESD. The health score of self-evaluation and life quality were better in the ESD group than those in the surgery group.
RESUMEN
<p><b>OBJECTIVE</b>To evaluate the efficacy of radiofrequency ablation(RFA) combined with endoscopic resection(ER) for eradicating widespread early non-flat type esophageal squamous cell carcinoma (ESCC) and precancerous lesions.</p><p><b>METHODS</b>Retrospective analysis was performed on the clinical data of 4 patients with early non-flat type ESCC and precancerous lesions in January 2010 at the Cancer Institute and Hospital, Chinese Academy of Medical Sciences. Proportion of patients with histological complete response (CR) 3 months, 12 months to 5 years after operation and adverse events were observed.</p><p><b>RESULTS</b>These 4 patients were all male, aged from 47 to 71 (mean age 62) years, including 2 of ESCC, 1 of HGIN, 1 of MGIN confirmed by pathology. USL length was 6-12 (mean 8.5) cm. Treatment area (TA) length was 8-14 (mean 10.5) cm. Three cases were 0-II a (mean length 2 cm), and 1 case 0-II c (mean length 4 cm). Lesions of 2 cases were complete cycle, and other 2 cases occupied 3/4 circumference. Four patients completed their operations successfully. Total operation time was 42-105 (mean 66.8) min, RFA time was 3-12 (mean 8.25) min, and ER time was 6-20 (10.25) min, without bleeding and perforation. The mean hospital stay was 3 days. Pathology examination showed that 2 cases were ESCC G2 (lesion length 12, 8 cm; non-flat type lesion length 3, 4 cm), 1 was HGIN (lesion length 12 cm; non-flat type lesion length 1 cm) and 1 was MGIN (lesion length 6 cm; non-flat type lesion length 2 cm). Three cases were CR 3 months, 1 to 5 years after operation. One case had HGIN at 3-month and MGIN at 1-year and 3-year during follow up, and was CR after treatment with HALO. Postoperative esophageal stenosis occurred in 4 cases. Among them, 2 cases were mild without treatment, and 2 were severe, who were relieved by endoscopic water sac dilation for 5-8 (mean 6.5) times.</p><p><b>CONCLUSION</b>RFA combined with ER is effective and safe in the treatment of patients with early non-flat esophageal squamous cell carcinoma and precancerous lesions.</p>