S1068 Safety and Feasibility of Same-Day Discharge After Endoscopic Submucosal Dissection: A Western Multicenter Prospective Cohort Study

Abstract

Introduction: Patients undergoing endoscopic submucosal dissection (ESD) in Asia are traditionally admitted for routine post-procedural observation. Data on the feasibility of same-day discharge (SDD) following ESD remains scarce. We aimed to evaluate the safety and feasibility of SDD after ESD and to identify factors associated with admission. Methods: Retrospective post-hoc analysis of multicenter prospective cohort study of adults who underwent ESD at 10 participating centers from 2016 to 2021. The primary endpoint was safety and feasibility of SDD after ESD and to identify factors associated with admission. Endoscopists with < 50 cases were defined as low-volume ESD endoscopists. Lesions with Paris 0-IIc or III were considered to have a depressed morphology component. Results: 831 patients (57% male, median age 67 [IQR 57-74]) underwent esophageal (n=240), gastric (n=126), and colorectal (n=465) ESD, of which 588 (71%) underwent SDD and 243 (29%) were admitted. 13 (2%) in the SDD group had delayed bleeding, 9 requiring repeat endoscopy and 2 needing blood transfusion. Four patients (1%) had delayed perforation, of which only one required surgery. Out of the 243 admissions, 223 (92%) were discharged within 24 hours after routine post-procedural observation. Nineteen patients (8%) were admitted for the management of adverse events: 3 for bleeding and 16 for intraprocedural perforation. All these complications were managed endoscopically, and none required surgery. Mean hospital length of stay was 1.3±0.8 days. Interestingly, on multivariate analysis, lesion size ≥45 mm inversely correlated with admission (OR: 0.5; 95% CI: 0.3-0.8; P=0.001). All of the following factors were independently associated with a higher likelihood of admission: ESD in esophagus/stomach vs colorectal (OR: 1.7; 95% CI: 1.1-2.6; P=0.011), low endoscopist volume (OR: 2.1; 95% CI: 1.3-3.3; P=0.001), invasive cancer histology at baseline (OR: 1.9; 95% CI: 1.2-3.1; P=0.010), adverse events (OR: 2.7; 95% CI: 1.5-4.8; P=0.001), and procedure time >75 minutes (OR: 13.5; 95% CI 8.5-21.3; P< 0.001). (Table) Conclusion: SDD is feasible and safe, with less than 3% of patients developing delayed adverse events following SDD. Factors including ESD location, lesion size, baseline histology, endoscopist volume, and procedural time appear to correlate with need for admission. Additional studies are needed to identify predictors for SDD after ESD. Table 1. - Baseline characteristics of patients hospitalized and discharged following endoscopic submucosal dissection for esophageal lesions Baseline characteristic All (831) Admitted (243) Discharged (588) P-value Age (median, IQR) 67 (57 – 74) 68 (59 – 73) 66 (57 – 74) 0.185 Male sex 477 (57%) 138 (57%) 339 (58%) 0.823 ASA classIIIIIIIV 82 (10%)374 (45%)355 (43%)20 (2%) 12 (5%)103 (42%)112 (46%)16 (7%) 70 (12%)271 (46%)243 (41%)4 (1%) 0.005 Upper GI LocationEsophagusDistal esophagus/GEJEsophageal bodyStomachColorectal 240 (29%)161 (19%)79 (10%)126 (15%)465 (56%) 83 (35%)62 (26%)21 (9%)41 (17%)119 (49%) 157 (27%)99 (17%)58 (10%)85 (14%)346 (59%) 0.009 Endoscopist volume < 50 176 (21%) 80 (33%) 96 (16%) < 0.0001 Prior intervention at ESD site 197 (24%) 35 (14%) 162 (28%) < 0.0001 No fibrosisSubmucosal fibrosisMildModerateSevere 467 (56%)362 (44%)110 (13%)92 (11%)160 (19%) 158 (65%)85 (35%)14 (6%)16 (7%)55 (23%) 309 (53%)279 (47%)96 (16%)78 (13%)105 (18%) 0.0001 Lesion size, mm (median, IQR) 44 (32 – 60) 41 (30 – 60) 45 (33 – 59) < 0.0001 Depressed morphologyNon-depressed morphologyFlatElevatedMixedOther (i.e. submucosal) 111 (13%)720 (87%)375 (45%)208 (25%)90 (11%)47 (6%) 48 (20%)195 (80%)75 (31%)81 (33%)29 (12%)10 (4%) 63 (11%)525 (89%)300 (50%)127 (21%)61 (10%)37 (6%) 0.0004 Invasive cancerNon-invasiveIntramucosal cancerHigh grade dysplasiaLow grade dysplasiaSessile serrated adenomaHyperplastic polypNon-dysplastic lesionCarcinoid tumorOther 140 (17%)691 (83%)122 (15%)175 (21%)214 (26%)31 (4%)15 (2%)40 (5%)15 (2%)79 (10%) 63 (26%)180 (74%)40 (16%)39 (16%)46 (19%)16 (7%)5 (2%)9 (4%)3 (1%)22 (9%) 77 (13%)511 (87%)82 (14%)136 (23%)168 (29%)15 (3%)10 (2%)31 (5%)12 (2%)57 (10%) < 0.0001 Adverse eventsBleedingPerforationInfectionStrictureSedation-related 15 (2%)20 (2%)0 (0%)27 (3%)0 (0%) 3 (1%)16 (7%)0 (0%)13 (5%)0 (0%) 12 (2%)4 (1%)0 (0%)14 (2%)0 (0%) < 0.0001 Total procedure time, min (median, IQR) 76 (56 – 120) 136 (91 – 198) 69 (51 – 89) < 0.0001 En bloc resection rate 790 (94%) 219 (86%) 571 (96%) < 0.0001 R0 resection rate 713 (85%) 180 (73%) 533 (89%) < 0.0001 Curative resection rate 679 (81%) 172 (70%) 507 (85%) < 0.0001 IQR=Interquartile range. ASA=American Society of Anesthesiologists. GEJ=Gastroesophageal junction. HGD=High grade dysplasia. LGD=Low grade dysplasia.