Abstract
Background
Post-endoscopic submucosal dissection (ESD) electrocoagulation syndrome (PEECS) is a known complication in colorectal laterally spreading tumors (LSTs), but its risk factors in elderly patients remain unclear. This study aimed to develop and validate a risk-stratification scoring system for PEECS in elderly ESD patients.
Methods
A multicenter retrospective study (2020–2025) enrolled 506 elderly patients with colorectal LSTs undergoing ESD, randomly allocated to training (TC, n = 354) and validation (VC, n = 152) cohorts (7:3). Synthetic minority over-sampling technique (SMOTE) was used in the TC to identify risk factors and construct a predictive model, which was validated in the VC.
Results
The incidence of post-ESD PEECS was 8.1% (41 cases). After applying the SMOTE, multivariate analysis identified sex, lesion with fibrosis, and intraoperative bleeding as independent risk factors. The scoring system assigned: 1 point for female sex, 3 points for lesion with fibrosis, and 2 points for intraoperative bleeding. In the VC, the model demonstrated an area under the curve (AUC) of 0.921, with a specificity of 0.949 and an accuracy of 0.908. Following risk stratification, the low-risk group (0–4 points) showed a PEECS incidence of 10.1% in the TC and 5.1% in the VC, while the high-risk group (5–6 points) exhibited 94.5% in the TC and 53.3% in the VC.
Conclusion
In elderly patients with colorectal LSTs undergoing ESD, female sex, fibrotic lesions, and intraoperative bleeding were identified as independent predictors of PEECS. The proposed scoring system demonstrated good discriminatory ability in both cohorts and may be useful for risk stratification in this population.
Supplementary Information
The online version contains supplementary material available at 10.1007/s00384-026-05141-7.
Keywords: Electrocoagulation syndrome, Laterally spreading tumors, Endoscopic submucosal dissection, Elderly, Scoring system
Introduction
Endoscopic submucosal dissection (ESD) has become a widely accepted minimally invasive technique for treating colorectal laterally spreading tumors (LSTs), particularly in elderly patients who are poor candidates for surgery [1, 2]. While ESD offers advantages such as en bloc resection and accurate histopathological evaluation, it is associated with several complications, including post-ESD electrocoagulation syndrome (PEECS) [3]. PEECS is characterized by localized abdominal pain, fever, and inflammatory signs without evident perforation [4, 5]. The reported incidence of PECS following colorectal ESD varies from 1.2% to 10.8% in different studies, with elderly patients potentially being at higher risk due to age-related physiological changes and comorbidities [6–8].
Several risk factors for PEECS have been identified, including large lesion size (40 mm), prolonged procedure time (60 min), and location in the right colon [3, 9, 10]. However, most existing studies have focused on general populations, and specific risk assessment tools for elderly patients with colorectal LSTs remain lacking. This is particularly concerning as the elderly population represents a growing proportion of patients undergoing colorectal ESD procedures worldwide [11]. Current preventive strategies for PEECS include meticulous technique, careful coagulation of exposed vessels, and novel approaches like line-assisted complete clip closure (LACC) for large mucosal defects [12, 13]. Despite these measures, PEECS continues to occur, highlighting the need for better risk stratification tools.
The development of predictive models for PEECS has gained increasing attention. Several scoring systems have been proposed for various ESD-related complications [14–16], but none specifically address PEECS risk in elderly patients with colorectal LSTs. Given the unique physiological characteristics and higher complication risks in this population, a dedicated predictive model could significantly improve clinical decision-making and postoperative management. This study aims to develop and validate a comprehensive predictive scoring model for PEECS specifically tailored for elderly patients undergoing colorectal ESD for LSTs. By incorporating both established and novel risk factors, we hope to provide clinicians with a practical tool for risk stratification and personalized patient management.
Material and methods
Patients
A retrospective analysis was conducted on clinicopathological data from elderly patients (aged ≥ 60 years) who underwent ESD for colorectal LSTs at five medical institutions (Suzhou BenQ Medical Center, Changshu Hospital Affiliated to Soochow University, Kunshan First People's Hospital, Zhangjiagang First People's Hospital, and Nanjing First Hospital) between January 2020 and June 2025. The inclusion criteria were lesions treated with ESD. Exclusion criteria included lesions managed with concurrent surgical intervention for multiplicity, those complicated by intra- or postoperative perforation, and cases with incomplete clinical data. The cohort was randomly divided into a training cohort (TC) and a validation cohort (VC) in a 7:3 ratio. Given the low incidence of post-ESD electrocoagulation syndrome (PEECS), the Synthetic Minority Over-sampling Technique (SMOTE) was utilized in the TC to balance the class distribution of PEECS outcomes. The study protocol, outlined in Fig. 1, was approved by the Clinical Research Ethics Committee of Changshu Hospital Affiliated to Soochow University, and written informed consent was obtained from all participants. All methods were performed in accordance with the relevant guidelines and regulations.
Fig. 1.
Study enrollment process
Endoscopic procedure and equipments
The treatment for colorectal LSTs in this study was ESD, a technique whose detailed protocol has been described elsewhere [17]. Figure 2 provides a visual representation of the ESD process for a lesion in the transverse colon. Prior to the intervention, all patients were placed under general anesthesia via endotracheal intubation. Patient positioning began in the left lateral decubitus position and was dynamically altered to supine or right lateral decubitus to facilitate the operation. The endoscopic setup involved a single-channel endoscope (GIF-Q260J, Olympus or EG-550, Sonoscape) with a distal cap, powered by an ERBE VIO 200D generator. Surgeons utilized electrosurgical knives such as the Kunpeng knife or the Dual Knife, supported by auxiliary tools like metallic clips and a CO₂ insufflator. To ensure patient safety, standard preoperative blood work was performed, and any anticoagulant or antiplatelet medications were withheld for one week. The procedures were conducted by highly experienced endoscopists, each possessing a record of more than 100 colorectal ESDs. To ensure consistency in the intraoperative assessment of key findings such as submucosal fibrosis, all participating endoscopists adhered to a shared predefined definition prior to the commencement of data collection. Assessments were recorded prospectively in a standardized case report form immediately following the procedure.
Fig. 2.
Endoscopic submucosal dissection procedure for colorectal laterally spreading tumors. A LST located in the transverse colon. B Mucosal incision following submucosal injection and elevation. C Stepwise dissection of the submucosal layer. D Post-resection mucosal defect. E Closure of the defect using hemostatic clips. F Resected specimen pinned to a fixation board for pathological assessment. LST: laterally spreading tumor
Postoperative management and follow-up
Hemostatic clips were used to manage the post-ESD mucosal defect in all patients. Complete closure (defined as clip placement reducing the defect to < 5 mm or approximating the entire margin) was achieved in 78.3% (396/506) of cases, while partial clipping to reduce exposed submucosal area was performed in the remaining 21.7% (110/506) where complete closure was not feasible due to large defect size or unfavorable anatomy. After the tumor is removed, the handling of the specimen depends on the resection type: en bloc lesions are pinned to a foam board, while piecemeal lesions are placed directly into 10% formalin for preservation. All specimens are then prepared for histological assessment via paraffin embedding and sectioning. Patient care after the procedure involves fasting, measures to prevent bleeding, and fluid support. The use of antibiotics is not standard; they are only initiated if a perforation is suspected during the operation or if the patient runs a fever afterwards. Close monitoring for complications like bleeding, perforation, or infection is essential, with treatments applied as necessary. Provided no issues arise, patients can start a liquid diet 24 h after the procedure. A surveillance colonoscopy was scheduled for all patients between 6 months and 1 year post-procedure to evaluate the healing status of the resection site.
Data collection
The variables collected in this investigation covered demographic and clinical characteristics, including sex, body mass index (BMI), major comorbidities, and a history of anticoagulant/antiplatelet medication use. Tumor-related parameters included location, size, depth of invasion, morphology, and histology. Procedural aspects recorded were procedure time, the non-lifting sign, fibrosis degree, intraoperative bleeding, success of en bloc resection, and antibiotic administration. Postoperative outcomes monitored were postoperative bleeding, fasting duration, hospital stay, and PEECS occurrence. All intraoperative findings, including the presence of submucosal fibrosis, were assessed by the operating endoscopist using a shared definition across centers and were recorded prospectively in a standardized manner on the data collection form immediately after the procedure to minimize recall bias and inter-operator variability.
Definitions
PEECS was diagnosed in patients exhibiting concurrent upper abdominal pain and localized tenderness, accompanied by either fever (37.5 C) or systemic inflammatory response (white blood cell count ≥ 10,000 cells/μL or C-reactive protein ≥ 0.5 mg/dL), with no radiographic or endoscopic evidence of perforation [14, 18]. Tumor size and invasion depth were determined primarily from postoperative pathological examination. Complete en bloc resection refers to the removal of the lesion as a single, unfragmented specimen [19]. Laterally spreading tumors were classified into granular (LST-G) and non-granular (LST-NG) types. LST-G lesions present with an aggregated granular or nodular surface resembling a "cobblestone" or "cauliflower" appearance, whereas LST-NG lesions have a smooth, flat surface that may appear subtly depressed, flat, or slightly elevated [20]. Intraoperative findings, including the non-lifting sign and submucosal fibrosis, were recorded based on direct observation during ESD. Submucosal fibrosis was defined based on intraoperative endoscopic findings during submucosal injection and dissection. It was characterized by inadequate or non-lifting of the mucosa after submucosal injection, accompanied by the visualization of web-like, white dense fibrous tissue strands within the submucosal layer. The assessment of fibrosis was performed in real-time by the operating endoscopist using a binary classification (presence or absence), based on previously established endoscopic criteria [21]. Intraoperative bleeding was defined as hemorrhage occurring during ESD that could not be adequately controlled using the endoscopic knife alone and therefore required the use of hemostatic forceps for hemostasis [22]. Minor oozing or bleeding easily controlled by the endoscopic knife with coagulation current was not classified as intraoperative bleeding. Delayed bleeding was identified by clinical signs of hematochezia or melena requiring endoscopic hemostasis within one week post-procedure [23]. Postoperative fever was defined as a maximum body temperature of ≥ 37.5 C occurring within three days after ESD. Finally, total procedure time was calculated from the initiation of submucosal injection to the complete resection of the lesion.
Statistical analysis
Categorical data are presented as counts (percentages) and were compared with the Chi-square test. For continuous variables that were not normally distributed, we report median values with interquartile ranges (IQRs) and employed the Mann–Whitney U test for group comparisons. To address class imbalance in the training cohort, the SMOTE was applied to equilibrate the sample sizes between the PEECS and non-PEECS groups. Potential risk factors for PEECS were first screened through univariate analysis. Variables yielding a P-value < 0.05 were subsequently entered into a multivariate logistic regression model to identify independent predictors. A scoring system was then constructed based on the regression weights (β coefficients) from the multivariate model. The predictor with the smallest absolute β coefficient served as the reference and was assigned 1 point. Scores for other variables were calculated by dividing their respective β coefficients by this reference value and rounding to the nearest integer. The model's discriminative ability was evaluated using the area under the receiver operating characteristic curve (AUC), while its calibration was assessed with a calibration plot. Furthermore, we computed key performance metrics, including sensitivity, specificity, accuracy, positive predictive value (PPV), and negative predictive value (NPV), for a comprehensive evaluation. Throughout the analysis, statistical significance was set at a two-tailed P-value < 0.05. All analyses were conducted using R software (version 4.1.0).
Results
Baseline characteristics of patients
The study cohort comprised 506 patients, distributed as 354 in the TC and 152 in the VC. As demonstrated in Table 1, baseline demographic and clinical characteristics were balanced between the groups, showing no statistically significant differences (P > 0.05). Consistent with this, the intraoperative and postoperative outcomes presented in Table 2 also revealed no significant intergroup differences (P > 0.05). PEECS was observed in 41 patients (8.1%), with group-specific incidences of 7.3% (TC group) and 9.9% (VC group).
Table 1.
Demographic and clinical characteristics of colorectal LSTs in the training and validation cohorts
| Characteristics | Overall (n = 506) | TC (n = 354) | VC (n = 152) | p-value |
|---|---|---|---|---|
| Sex, n (%) | 0.727 | |||
| Male | 249 (49.2%) | 176 (49.7%) | 73 (48.0%) | |
| Female | 257 (50.8%) | 178 (50.3%) | 79 (52.0%) | |
| BMI, kg/m2, n (%) | 0.985 | |||
| < 18.5 | 81 (16.0%) | 57 (16.1%) | 24 (15.8%) | |
| 18.5–23.9 | 258 (51.0%) | 181 (51.1%) | 77 (50.7%) | |
| ≥ 24.0 | 167 (33.0%) | 116 (32.8%) | 51 (33.6%) | |
| Hypertension, n (%) | 0.837 | |||
| Yes | 143 (28.3%) | 101 (28.5%) | 42 (27.6%) | |
| No | 363 (71.7%) | 253 (71.5%) | 110 (72.4%) | |
| Diabetes, n (%) | 0.756 | |||
| Yes | 148 (29.2%) | 105 (29.7%) | 43 (28.3%) | |
| No | 358 (70.8%) | 249 (70.3%) | 109 (71.7%) | |
| Coronary disease, n (%) | 0.709 | |||
| Yes | 47 (9.3%) | 34 (9.6%) | 13 (8.6%) | |
| No | 459 (90.7%) | 320 (90.4%) | 139 (91.4%) | |
| Anticoagulant and antiplatelet drugs, n (%) | 0.423 | |||
| Yes | 48 (9.5%) | 36 (10.2%) | 12 (7.9%) | |
| No | 458 (90.5%) | 318 (89.8%) | 140 (92.1%) | |
| Tumor location, n (%) | 0.978 | |||
| Right-side colon | 204 (40.3%) | 143 (40.4%) | 61 (40.1%) | |
| Left-side colon | 162 (32.0%) | 114 (32.2%) | 48 (31.6%) | |
| Rectum | 140 (27.7%) | 97 (27.4%) | 43 (28.3%) | |
| Tumor size, mm, n (%) | 0.247 | |||
| < 40 | 317 (62.6%) | 216 (61.0%) | 101 (66.4%) | |
| ≥ 40 | 189 (37.4%) | 138 (39.0%) | 51 (33.6%) | |
| Invasion depth, n (%) | 0.516 | |||
| Mucosal layer | 442 (87.4%) | 307 (86.7%) | 135 (88.8%) | |
| Submucosal layer | 64 (12.6%) | 47 (13.3%) | 17 (11.2%) | |
| Morphology, n (%) | 0.421 | |||
| LST-G | 349 (69.0%) | 248 (70.1%) | 101 (66.4%) | |
| LST-NG | 157 (31.0%) | 106 (29.9%) | 51 (33.6%) |
LSTs lateral spreading tumors, LST-G lateral spreading tumor granular type, LST-NG lateral spreading tumor non-granular type, BMI body mass index, TC training cohort, VC validation cohort
Table 2.
Intraoperative and postoperative outcomes of colorectal LSTs in the training and validation cohorts
| Characteristics | Overall (n = 506) | TC (n = 354) | VC (n = 152) | p-value |
|---|---|---|---|---|
| PEECS, n (%) | 0.340 | |||
| Yes | 41 (8.1%) | 26 (7.3%) | 15 (9.9%) | |
| No | 465 (91.9%) | 328 (92.7%) | 137 (90.1%) | |
| Procedure time, minutes, n (%) | 0.142 | |||
| < 60 | 353 (69.8%) | 240 (67.8%) | 113 (74.3%) | |
| ≥ 60 | 153 (30.2%) | 114 (32.2%) | 39 (25.7%) | |
| Histology, n (%) | 0.323 | |||
| Adenoma (LGIN + HGIN) | 467 (92.3%) | 324 (91.5%) | 143 (94.1%) | |
| Adenocarcinoma | 39 (7.7%) | 30 (8.5%) | 9 (5.9%) | |
| Non lifting sign, n (%) | 0.994 | |||
| Yes | 60 (11.9%) | 42 (11.9%) | 18 (11.8%) | |
| No | 446 (88.1%) | 312 (88.1%) | 134 (88.2%) | |
| Lesion with fibrosis, n (%) | 0.508 | |||
| Fibrosis | 140 (27.7%) | 101 (28.5%) | 39 (25.7%) | |
| Non-fibrosis | 366 (72.3%) | 253 (71.5%) | 113 (74.3%) | |
| Resection, n (%) | 0.576 | |||
| Complete en bloc resection | 420 (83.0%) | 296 (83.6%) | 124 (81.6%) | |
| Piecemeal resection | 86 (17.0%) | 58 (16.4%) | 28 (18.4%) | |
| Intraoperative bleeding, n (%) | 0.837 | |||
| Yes | 143 (28.3%) | 101 (28.5%) | 42 (27.6%) | |
| No | 363 (71.7%) | 253 (71.5%) | 110 (72.4%) | |
| Antibiotics therapy, n (%) | 0.301 | |||
| Yes | 74 (14.6%) | 48 (13.6%) | 26 (17.1%) | |
| No | 432 (85.4%) | 306 (86.4%) | 126 (82.9%) | |
| Postoperative bleeding, n (%) | 0.258 | |||
| Yes | 29 (5.7%) | 23 (6.5%) | 6 (3.9%) | |
| No | 477 (94.3%) | 331 (93.5%) | 146 (96.1%) | |
| Postoperative hospital stay, median (IQR) | 4.0 (3.0, 5.0) | 4.0 (3.0, 5.0) | 4.0 (3.0, 5.0) | 0.672 |
| Postoperative fasting, median (IQR) | 2.0 (1.0, 2.0) | 1.0 (1.0, 2.0) | 2.0 (1.0, 2.0) | 0.207 |
LSTs lateral spreading tumors, PEECS post-endoscopic submucosal dissection electrocoagulation syndrome, LGIN low grade intraepithelial neoplasia, HGIN high grade intraepithelial neoplasia, TC training cohort, VC validation cohort, IQR interquartile range
Comparison between PEECS group and non-PEECS group
Patients were classified into PEECS and non-PEECS groups based on the development of PEECS following ESD. Within the TC, the PEECS group demonstrated a significantly higher proportion of female patients compared to the non-PEECS group (76.9% vs. 48.2%, P = 0.005). Additionally, the prevalence of larger tumors (≥ 40 mm), the presence of fibrosis, and the incidence of intraoperative bleeding were all markedly higher in the PEECS group (69.2% vs. 36.6%, P = 0.001; 80.8% vs. 24.4%, P < 0.001; 73.1% vs. 25.0%, P < 0.001, respectively), as detailed in Table 3. Comparisons between the two groups in the validation cohort are summarized in Supplementary Table 1, which revealed statistically significant differences in patient sex, the presence of fibrosis, and intraoperative bleeding (P < 0.05 for all). Table 4 illustrates the intergroup comparisons after SMOTE adjustment. Significant differences persisted between the PEECS and non-PEECS groups in terms of patient sex, tumor size, fibrosis, and intraoperative bleeding (all P < 0.05).
Table 3.
Univariate analysis identified the risk factors for PEECS in the training cohort
| Characteristics | PEECS (n = 26) | Non-PEECS (n = 328) | p-value |
|---|---|---|---|
| Sex, n (%) | 0.005 | ||
| Male | 6 (23.1%) | 170 (51.8%) | |
| Female | 20 (76.9%) | 158 (48.2%) | |
| BMI, kg/m2, n (%) | 0.646 | ||
| < 18.5 | 5 (19.2%) | 52 (15.9%) | |
| 18.5–23.9 | 11 (42.3%) | 170 (51.8%) | |
| ≥ 24.0 | 10 (38.5%) | 106 (32.3%) | |
| Hypertension, n (%) | 0.522 | ||
| Yes | 6 (23.1%) | 95 (29.0%) | |
| No | 20 (76.9%) | 233 (71.0%) | |
| Diabetes, n (%) | 0.226 | ||
| Yes | 5 (19.2%) | 100 (30.5%) | |
| No | 21 (80.8%) | 228 (69.5%) | |
| Coronary disease, n (%) | > 0.999 | ||
| Yes | 2 (7.7%) | 32 (9.8%) | |
| No | 24 (92.3%) | 296 (90.2%) | |
| Anticoagulant and antiplatelet drugs, n (%) | 0.496 | ||
| Yes | 1 (3.8%) | 35 (10.7%) | |
| No | 25 (96.2%) | 293 (89.3%) | |
| Tumor location, n (%) | 0.756 | ||
| Right-side colon | 10 (38.5%) | 133 (40.5%) | |
| Left-side colon | 10 (38.5%) | 104 (31.7%) | |
| Rectum | 6 (23.1%) | 91 (27.7%) | |
| Tumor size, mm, n (%) | 0.001 | ||
| < 40 | 8 (30.8%) | 208 (63.4%) | |
| ≥ 40 | 18 (69.2%) | 120 (36.6%) | |
| Invasion depth, n (%) | 0.763 | ||
| Mucosal layer | 22 (84.6%) | 285 (86.9%) | |
| Submucosal layer | 4 (15.4%) | 43 (13.1%) | |
| Histology, n (%) | 0.473 | ||
| Adenoma (LGIN + HGIN) | 23 (88.5%) | 301 (91.8%) | |
| Adenocarcinoma | 3 (11.5%) | 27 (8.2%) | |
| Morphology, n (%) | 0.325 | ||
| LST-G | 16 (61.5%) | 232 (70.7%) | |
| LST-NG | 10 (38.5%) | 96 (29.3%) | |
| Non lifting sign, n (%) | 0.530 | ||
| Yes | 4 (15.4%) | 38 (11.6%) | |
| No | 22 (84.6%) | 290 (88.4%) | |
| Lesion with fibrosis, n (%) | < 0.001 | ||
| Fibrosis | 21 (80.8%) | 80 (24.4%) | |
| Non-fibrosis | 5 (19.2%) | 248 (75.6%) | |
| Procedure time, minutes, n (%) | 0.252 | ||
| < 60 | 15 (57.7%) | 225 (68.6%) | |
| ≥ 60 | 11 (42.3%) | 103 (31.4%) | |
| Resection, n (%) | 0.782 | ||
| Complete en bloc resection | 23 (88.5%) | 273 (83.2%) | |
| Piecemeal resection | 3 (11.5%) | 55 (16.8%) | |
| Intraoperative bleeding, n (%) | < 0.001 | ||
| Yes | 19 (73.1%) | 82 (25.0%) | |
| No | 7 (26.9%) | 246 (75.0%) |
PEECS post-endoscopic submucosal dissection electrocoagulation syndrome, LSTs lateral spreading tumors, LST-G lateral spreading tumor granular type, LST-NG lateral spreading tumor non-granular type, BMI body mass index; bold values indicate P < 0.05
Table 4.
Univariate analysis identified the risk factors for PEECS in the training cohort after SMOTE
| Characteristics | Overall (n = 656) | PEECS (n = 328) | Non-PEECS (n = 328) | p-value |
|---|---|---|---|---|
| Sex, n (%) | < 0.001 | |||
| Male | 271 (41.3%) | 101 (30.8%) | 170 (51.8%) | |
| Female | 385 (58.7%) | 227 (69.2%) | 158 (48.2%) | |
| BMI, kg/m2, n (%) | 0.061 | |||
| < 18.5 | 121 (18.4%) | 69 (21.0%) | 52 (15.9%) | |
| 18.5–23.9 | 347 (52.9%) | 177 (54.0%) | 170 (51.8%) | |
| ≥ 24.0 | 188 (28.7%) | 82 (25.0%) | 106 (32.3%) | |
| Hypertension, n (%) | 0.272 | |||
| Yes | 203 (30.9%) | 108 (32.9%) | 95 (29.0%) | |
| No | 453 (69.1%) | 220 (67.1%) | 233 (71.0%) | |
| Diabetes, n (%) | 0.225 | |||
| Yes | 186 (28.4%) | 86 (26.2%) | 100 (30.5%) | |
| No | 470 (71.6%) | 242 (73.8%) | 228 (69.5%) | |
| Coronary disease, n (%) | > 0.999 | |||
| Yes | 64 (9.8%) | 32 (9.8%) | 32 (9.8%) | |
| No | 592 (90.2%) | 296 (90.2%) | 296 (90.2%) | |
| Anticoagulant and antiplatelet drugs, n (%) | 0.286 | |||
| Yes | 62 (9.5%) | 27 (8.2%) | 35 (10.7%) | |
| No | 594 (90.5%) | 301 (91.8%) | 293 (89.3%) | |
| Tumor location, n (%) | 0.616 | |||
| Right-side colon | 271 (41.3%) | 138 (42.1%) | 133 (40.5%) | |
| Left-side colon | 214 (32.6%) | 110 (33.5%) | 104 (31.7%) | |
| Rectum | 171 (26.1%) | 80 (24.4%) | 91 (27.7%) | |
| Tumor size, mm, n (%) | < 0.001 | |||
| < 40 | 339 (51.7%) | 131 (39.9%) | 208 (63.4%) | |
| ≥ 40 | 317 (48.3%) | 197 (60.1%) | 120 (36.6%) | |
| Invasion depth, n (%) | 0.080 | |||
| Mucosal layer | 584 (89.0%) | 299 (91.2%) | 285 (86.9%) | |
| Submucosal layer | 72 (11.0%) | 29 (8.8%) | 43 (13.1%) | |
| Histology, n (%) | 0.886 | |||
| Adenoma (LGIN + HGIN) | 603 (91.9%) | 302 (92.1%) | 301 (91.8%) | |
| Adenocarcinoma | 53 (8.1%) | 26 (7.9%) | 27 (8.2%) | |
| Morphology, n (%) | 0.159 | |||
| LST-G | 480 (73.2%) | 248 (75.6%) | 232 (70.7%) | |
| LST-NG | 176 (26.8%) | 80 (24.4%) | 96 (29.3%) | |
| Non lifting sign, n (%) | 0.112 | |||
| Yes | 90 (13.7%) | 52 (15.9%) | 38 (11.6%) | |
| No | 566 (86.3%) | 276 (84.1%) | 290 (88.4%) | |
| Lesion with fibrosis, n (%) | < 0.001 | |||
| Fibrosis | 394 (60.1%) | 314 (95.7%) | 80 (24.4%) | |
| Non-fibrosis | 262 (39.9%) | 14 (4.3%) | 248 (75.6%) | |
| Procedure time, minutes, n (%) | 0.168 | |||
| < 60 | 466 (71.0%) | 241 (73.5%) | 225 (68.6%) | |
| ≥ 60 | 190 (29.0%) | 87 (26.5%) | 103 (31.4%) | |
| Resection, n (%) | 0.075 | |||
| Complete en bloc resection | 562 (85.7%) | 289 (88.1%) | 273 (83.2%) | |
| Piecemeal resection | 94 (14.3%) | 39 (11.9%) | 55 (16.8%) | |
| Intraoperative bleeding, n (%) | < 0.001 | |||
| Yes | 383 (58.4%) | 301 (91.8%) | 82 (25.0%) | |
| No | 273 (41.6%) | 27 (8.2%) | 246 (75.0%) |
PEECS post-endoscopic submucosal dissection electrocoagulation syndrome, LSTs lateral spreading tumors, LST-G lateral spreading tumor granular type, LST-NG lateral spreading tumor non-granular type, BMI body mass index, SMOTE synthetic minority over-sampling technique
Bold values indicate P < 0.05
Risk factors for PEECS of ESD of colorectal LSTs
Univariate analysis identified sex, tumor size, fibrosis, and intraoperative bleeding as risk factors for PEECS (Table 4). Multivariate analysis confirmed female sex, fibrosis, and intraoperative bleeding as independent risk factors (Table 5). Specifically, the risk was significantly higher in females (OR 5.93, 95% CI 3.11–12.03), patients with fibrosis (OR 91.05, 95% CI 42.97–212.65), and those experiencing intraoperative bleeding (OR 41.46, 95% CI 21.26–87.13) compared to their respective reference groups.
Table 5.
Multivariate analysis identified risk factors for PEECS and developed a scoring system in the SMOTE-adjusted training cohort
| Variable | OR | 95%CI | P-value | Regression coefficient β | Multiple ratio | Score |
|---|---|---|---|---|---|---|
| Sex | ||||||
| Male | 1 | Reference | 1 (Reference) | 0 | ||
| Female | 5.93 | 3.11, 12.03 | < 0.001 | 1.780 | 1.0 | 1 |
| Tumor size, mm | ||||||
| < 40 | 1 | Reference | - | - | - | |
| ≥ 40 | 1.07 | 0.59, 1.92 | 0.830 | - | - | - |
| Lesion with fibrosis | ||||||
| Non-fibrosis | 1 | Reference | 1 (Reference) | 0 | ||
| Fibrosis | 91.05 | 42.97, 212.65 | < 0.001 | 4.611 | 2.6 | 3 |
| Introperative bleeding | ||||||
| No | 1 | Reference | 1 (Reference) | 0 | ||
| Yes | 41.46 | 21.26, 87.13 | < 0.001 | 3.725 | 2.1 | 2 |
PEECS post-endoscopic submucosal dissection electrocoagulation syndrome, OR odds ratios, CI confidence interval
Bold values indicate P < 0.05
Development and validation for PEECS of ESD of colorectal LSTs
Based on the β-coefficients derived from the multivariate analysis of the TC, a point was assigned to each independent risk factor. Specifically, 1 point was given for female sex, 3 points for the presence of fibrosis in the lesion, and 2 points for intraoperative bleeding (Table 5). The association between the total risk score and the incidence of PEECS is shown in Fig. 3. The association between the total risk score and the incidence of PEECS is shown in Fig. 3.
Fig. 3.
Correlation between the total risk score and the incidence of PEECS. PEECS: post-endoscopic submucosal dissection electrocoagulation syndrome
In the TC, the incidence of PEECS progressively increased with higher risk scores: 0.0% (score 0), 3.1% (score 1), 9.4% (score 2), 10.4% (score 3), 43.8% (score 4), 91.0% (score 5), and 96.5% (score 6). The scoring system demonstrated excellent discrimination in the TC, with an AUC of 0.960 (95% CI: 0.945–0.975, Fig. 4A). Calibration was also satisfactory, as the calibration curve closely aligned with the ideal line (P > 0.05; Fig. 5A). The model showed a sensitivity of 0.893, a specificity of 0.948, and an accuracy of 0.921, along with a positive predictive value (PPV) of 0.945 and a negative predictive value (NPV) of 0.899 (Table 6). Based on the total score, patients were stratified into low-risk (scores 0–4) and high-risk (scores 5–6) categories, with corresponding PEECS rates of 10.1% and 94.5%, respectively (Fig. 6).
Fig. 4.
Receiver operating characteristic curves of the predictive model. A ROC curve in the training cohort. B ROC curve in the validation cohort. ROC: receiver operating characteristic
Fig. 5.
Calibration plots of the predictive model. A Calibration plot in the training cohort. B Calibration plot in the validation cohort
Table 6.
The performance of the scoring system in predicting PEECS
| Cohort | AUC | Sensitivity | Specificity | PPV | NPV | Accuracy |
|---|---|---|---|---|---|---|
| Training | 0.960 | 0.893 | 0.948 | 0.945 | 0.899 | 0.921 |
| Validation | 0.921 | 0.533 | 0.949 | 0.533 | 0.949 | 0.908 |
AUC areas under the receiver operating characteristic curves, PPV positive predictive value, NPV negative predictive value
Fig. 6.
Association between different risk categories and the incidence of PEECS. PEECS: post-endoscopic submucosal dissection electrocoagulation syndrome
In the VC, the incidence of PEECS rose progressively with increasing risk scores, with respective rates of 0.0%, 0.0%, 0.0%, 13.6%, 23.5%, 33.3%, and 66.7% for scores 0 through 6. The scoring system maintained strong discriminatory performance in the VC, yielding an AUC of 0.921 (95% CI: 0.872–0.971; Fig. 4B). Good calibration was observed, as the calibration curve showed no significant deviation from the ideal line (P > 0.05; Fig. 5B). The system achieved a sensitivity of 0.533, a specificity of 0.949, and an overall accuracy of 0.908, with corresponding positive and negative predictive values of 0.533 and 0.949, respectively (Table 6). When patients were stratified into low-risk (scores 0–4) and high-risk (scores 5–6) groups, the PEECS rates were 5.1% and 53.3%, respectively (Fig. 6). Additionally, the distribution of patients and PEECS events by score (0–6) is shown in Supplementary Table 2 for the training cohort (after SMOTE) and in Supplementary Table 3 for the validation cohort.
Discussion
This multicenter retrospective study developed and validated a simple, clinically applicable risk-scoring system to predict PEECS specifically in elderly patients (≥ 60 years) with colorectal LSTs. Our analysis of 506 patients identified female sex, the presence of submucosal fibrosis, and intraoperative bleeding as independent predictive factors. Based on these, we constructed a scoring model with an excellent discriminatory capacity, achieving an area under the curve of 0.921 in the external validation cohort. This model enables the stratification of patients into distinct risk categories, which can inform tailored postoperative monitoring and potentially preemptive management strategies for this vulnerable, growing patient population.
Fibrosis has been consistently identified as a significant risk factor for PEECS in this study, aligning with previous research that highlights its role in increasing procedural complexity and thermal injury during ESD. The presence of fibrosis likely exacerbates electrocoagulation-induced tissue damage due to impaired submucosal dissection, leading to deeper thermal penetration and localized inflammation, which are key mechanisms underlying PEECS [24, 25]. This is supported by studies showing that fibrotic lesions prolong procedure time and reduce en bloc resection rates, further contributing to post-ESD complications [24]. However, our findings diverge from some earlier reports in two aspects. First, while prior studies primarily focused on fibrosis as a predictor of technical difficulty (e.g., perforation or incomplete resection) [24], this study specifically establishes its direct association with PEECS in elderly colorectal LST patients, a population underrepresented in earlier literature. Second, unlike some esophageal or gastric PEECS studies that emphasized tumor size or muscle layer exposure as dominant risk factors [26, 27], our model highlights fibrosis as a standalone predictor, suggesting site-specific pathophysiology. Notably, the proposed scoring system quantifies fibrosis’s contribution (3 points), underscoring its clinical weight compared to other variables like intraoperative bleeding. These differences may stem from the unique characteristics of colorectal LSTs in elderly patients, where fibrosis is more prevalent due to chronic mucosal injury or prior interventions. Future studies should explore whether fibrosis severity (e.g., mild vs. severe) further stratifies PEECS risk, as preliminary data suggest graded associations. Overall, our results reinforce fibrosis’s critical role in PEECS pathogenesis while contextualizing its relevance within a geriatric population.
The identification of female sex as an independent risk factor for PEECS in elderly colorectal LST patients aligns with prior studies but also introduces nuanced insights. Previous research on superficial colorectal lesions similarly reported female sex as a significant predictor (OR 3.94, P < 0.001), suggesting a consistent sex-based vulnerability across different colorectal ESD cohorts [3]. This consistency may reflect shared biological mechanisms, such as sex-specific differences in inflammatory responses or mucosal healing, as hinted by studies highlighting sex-dimorphic pathways in other gastrointestinal pathologies [28]. However, our study uniquely focuses on the elderly population, where hormonal changes (e.g., postmenopausal estrogen decline) could further exacerbate these differences, a dimension less explored in earlier PEECS literature [29]. Notably, while prior nomograms for PEECS in colorectal lesions incorporated female sex alongside age > 50 years and muscle layer injury [3], our scoring system specifically emphasizes its predictive weight (1 point) in elderly patients, contrasting with broader age-group analyses. This refinement suggests that sex disparities in PEECS risk may be more pronounced in older adults, possibly due to cumulative exposure to fibrotic changes or altered pain perception, as implied by studies linking female sex to prolonged procedure-related complications [30]. Additionally, our model’s high specificity (0.949) in validation underscores the clinical relevance of sex stratification, whereas earlier works primarily focused on composite risk factors without age-specific validation [3, 27].
Intraoperative bleeding was identified as a significant independent predictor of PEECS in elderly patients with colorectal LSTs, contributing 2 points to the predictive scoring model. This aligns with prior studies suggesting that intraoperative bleeding may exacerbate local tissue injury and inflammatory responses, leading to PEECS [29]. The thermal energy applied during hemostasis could induce deeper tissue damage, triggering a systemic inflammatory reaction characterized by fever, leukocytosis, and localized pain—hallmarks of PEECS [29, 31]. Notably, our findings corroborate earlier reports linking procedural complexity (e.g., prolonged ESD duration or extensive resection) to PEECS risk, as bleeding often necessitates additional coagulation, amplifying thermal injury [4, 14]. However, our study diverges from some previous literature in its specific focus on elderly colorectal LSTs. While prior research on gastric or esophageal ESD highlighted factors like muscle layer exposure or lesion size [4, 26], our model uniquely emphasizes intraoperative bleeding as a modifiable risk factor in this population. This distinction may reflect age-related mucosal fragility or altered healing responses in the elderly. Additionally, unlike studies attributing PEECS primarily to antithrombotic use [32, 33], our analysis identified bleeding itself—rather than its predisposing factors—as a direct contributor, suggesting that intraprocedural hemostatic management may mitigate PEECS risk independently of anticoagulant status. Future studies should explore whether proactive bleeding control strategies (e.g., preemptive coagulation or advanced hemostatic techniques) could reduce PEECS incidence in high-risk cohorts.
This study carries substantial clinical significance by providing the first dedicated predictive scoring model for PEECS specifically in elderly patients with colorectal LSTs. The developed three-factor scoring system—incorporating female sex, fibrotic lesions, and intraoperative bleeding—offers a simple, rapid, and evidence-based tool for risk stratification prior to or immediately following ESD. Its high specificity (0.949) and accuracy (0.908) in the validation cohort enable reliable identification of high-risk individuals, who exhibited a PEECS incidence exceeding 50%. In practice, this model can guide tailored postoperative management. Patients stratified into the high-risk category may benefit from enhanced monitoring and closer postoperative observation, allowing for earlier detection of PEECS should it occur. This risk-stratified approach can guide clinical attention and resource allocation, such as more frequent vital sign checks or earlier diagnostic evaluation in high-risk individuals. Conversely, low-risk patients might be considered for accelerated recovery pathways, including earlier diet advancement or shorter hospital stays, thereby optimizing healthcare resource utilization. Importantly, the management of established PEECS follows standardized clinical protocols regardless of risk category; the primary value of this scoring system lies in risk communication, patient counseling, and informing the intensity of postoperative monitoring rather than altering treatment once PEECS develops.
Despite its strengths, several limitations of this study must be acknowledged. Firstly, its retrospective and multicenter design, while providing a robust sample size, inherently carries the risk of selection bias and unmeasured confounding variables. Prospective validation in a dedicated cohort is essential to confirm the generalizability and real-world performance of the scoring system. Secondly, while SMOTE was employed to address class imbalance, this synthetic oversampling technique may not fully replicate the complexities of real-world data distribution, potentially affecting model generalizability. Thirdly, the external validation cohort, although independent, was drawn from similar geographic and clinical settings within China, limiting the model's applicability to other ethnic or healthcare populations without further validation. Fourthly, although all procedures were performed by experienced endoscopists (each with > 100 colorectal ESDs) using standardized definitions and prospective data recording to minimize inter-operator variability, individual endoscopist skill was not directly quantified or included as a covariate in the analysis. Given that technical proficiency in managing submucosal fibrosis and intraoperative bleeding is likely to influence the risk of PEECS, the absence of operator-level adjustment represents a potential source of residual confounding. Future prospective studies should consider incorporating objective metrics of operator experience or technique (e.g., case volume, procedure time-adjusted difficulty, or skill scores) to better isolate the impact of lesion-related factors from operator-dependent effects. Lastly, the study did not investigate dynamic or modifiable perioperative factors, such as specific electrosurgical generator settings, detailed hemostatic techniques, or postoperative pharmacological interventions, which could influence PEECS risk and offer avenues for prevention. Future prospective, multinational studies incorporating objective biomarkers and standardized technical parameters are needed to refine and validate this predictive tool.
In conclusion, this multicenter study developed and validated a simple scoring system for predicting PEECS in elderly patients undergoing ESD for colorectal LSTs. The model, which incorporates female sex, fibrotic lesions, and intraoperative bleeding, demonstrated strong discriminatory ability and good calibration in both the training and validation cohorts. This tool may aid clinicians in identifying individuals at elevated risk for this complication, thereby supporting more informed, risk-stratified postoperative management. Further prospective studies are warranted to validate its clinical utility and to explore whether its application can lead to improved patient outcomes in routine practice.
Supplementary Information
Below is the link to the electronic supplementary material.
Author contributions
Conception and design: Qian Zhang, Luojie Liu, and Wenwen Diao. Administrative support: Xiaodan Xu and Jian Chen. Provision of study materials or patients: Fengcheng Zang, Bin He, Yunfu Feng, Zhibing Wang, and Luojie Liu. Collection and assembly of data: Fengcheng Zang, Bin He, Yunfu Feng, Zhibing Wang, and Wenwen Diao. Data analysis and interpretation: Jian Chen and Luojie Liu. Manuscript writing: All authors. Final approval of manuscript: All authors.
Funding
This study was supported by the Changshu Science and Technology Program (CY202339), the Suzhou Youth Science and Technology Project for the Advancement of Science, Education, and Health (KJXW2023067), and the Suzhou 23rd Batch of Science and Technology Development Plan (Clinical Trial Institution Capability Enhancement) Project (SLT 2023006), and Changshu Municipal Key Medical Disciplines (CSZDXK202302).
Data availability
Data availability can be requested from the corresponding author with a reasonable justification.
Declarations
Consent for publication
Not applicable.
Disclosures
Qian Zhang, Fengcheng Zang, Bin He, Yunfu Feng, Zhibing Wang, Xiaodan Xu, Jian Chen, Luojie Liu and Wenwen Diao disclosed no conflict of interest for this article.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Qian Zhang and Fengcheng Zang contributed equally to this work.
References
- 1.Alric H, Barret M, Becar A et al (2024) Endoscopic submucosal dissection versus endoscopic mucosal resection for laterally spreading rectal tumours. Colorectal Dis. 10.1111/codi.17268 [DOI] [PubMed] [Google Scholar]
- 2.Tashima T, Ohata K, Nonaka K et al (2017) Endoscopic submucosal dissection for laterally spreading tumors involving the appendiceal orifice. Surg Endosc 31(12):5444–5450. 10.1007/s00464-017-5598-6 [DOI] [PubMed] [Google Scholar]
- 3.Qiu J, Zhang Y, Ouyang Q et al (2024) Risk factors and a nomogram for prediction of post-endoscopic submucosal dissection electrocoagulation syndrome for superficial colorectal lesions. Surg Endosc 38(7):3716–3727. 10.1007/s00464-024-10886-9 [DOI] [PubMed] [Google Scholar]
- 4.Liao F, Shen Z, Rong J et al (2024) Nomogram for predicting electrocoagulation syndrome after endoscopic submucosal dissection of esophageal tumors. Surg Endosc 38(4):1877–1883. 10.1007/s00464-024-10685-2 [DOI] [PubMed] [Google Scholar]
- 5.Shimura T, Iwasaki H, Ozeki T et al (2024) A prospective cohort study of a new electrosurgical unit for preventing colorectal post-endoscopic submucosal dissection coagulation syndrome. J Gastroenterol Hepatol 39(3):473–479. 10.1111/jgh.16444 [DOI] [PubMed] [Google Scholar]
- 6.Kim M, Choi CW, Kim ER et al (2024) Risk prediction of post-endoscopic submucosal dissection coagulation syndrome. Dig Dis 42(4):309–318. 10.1159/000539007 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Katano T, Shimura T, Nomura S et al (2021) Optimal definition of coagulation syndrome after colorectal endoscopic submucosal dissection: A post hoc analysis of randomized controlled trial. Int J Colorectal Dis 36(7):1479–1485. 10.1007/s00384-021-03921-x [DOI] [PubMed] [Google Scholar]
- 8.Seo M, Song EM, Cho JW et al (2019) A risk-scoring model for the prediction of delayed bleeding after colorectal endoscopic submucosal dissection. Gastrointest Endosc 89(5):990-998.e2. 10.1016/j.gie.2018.11.029 [DOI] [PubMed] [Google Scholar]
- 9.Shichijo S, Takeuchi Y, Shimodate Y, Kansai Endoscopic Device Selection Conference in Kansai Research Group et al (2022) Performance of perioperative antibiotics against post-endoscopic submucosal dissection coagulation syndrome: a multicenter randomized controlled trial. Gastrointest Endosc 95(2):349–359. 10.1016/j.gie.2021.08.025 [DOI] [PubMed] [Google Scholar]
- 10.Qian J, Tao YL, Zheng SS (2025) Risk modeling of delayed postoperative bleeding after endoscopic submucosal dissection for early colorectal cancer and precancerous lesions. World J Gastrointest Surg 17(9):107326. 10.4240/wjgs.v17.i9.107326 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Draganov PV, Wang AY, Othman MO et al (2019) AGA institute clinical practice update: Endoscopic submucosal dissection in the United States. Clin Gastroenterol Hepatol 17(1):16-25.e1. 10.1016/j.cgh.2018.07.041 [DOI] [PubMed] [Google Scholar]
- 12.Lee SP, Sung IK, Kim JH et al (2017) A randomized controlled trial of prophylactic antibiotics in the prevention of electrocoagulation syndrome after colorectal endoscopic submucosal dissection. Gastrointest Endosc 86(2):349-357.e2. 10.1016/j.gie.2016.11.022 [DOI] [PubMed] [Google Scholar]
- 13.Bahin FF, Rasouli KN, Byth K et al (2016) Prediction of clinically significant bleeding following wide-field endoscopic resection of large sessile and laterally spreading colorectal lesions: A clinical risk score. Am J Gastroenterol 111(8):1115–1122. 10.1038/ajg.2016.235 [DOI] [PubMed] [Google Scholar]
- 14.Feng Y, Zheng S, He B et al (2025) Innovative development and validation of a scoring system for predicting electrocoagulation syndrome post-endoscopic submucosal dissection of gastric gastrointestinal stromal tumors. Surg Endosc. 10.1007/s00464-025-12360-6 [DOI] [PubMed] [Google Scholar]
- 15.Zhang M, Ma J, Tian W et al (2024) Prediction of post-ESD esophageal stricture by a nomogram and risk factor analysis of ineffective oral steroids prophylaxis. BMC Gastroenterol 24(1):360. 10.1186/s12876-024-03448-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Park K, Kim NY, Kim KJ et al (2022) A simple risk scoring system for predicting the occurrence of aspiration pneumonia after gastric endoscopic submucosal dissection. Anesth Analg 134(1):114–122. 10.1213/ANE.0000000000005779 [DOI] [PubMed] [Google Scholar]
- 17.Zou J, Chai N, Linghu E et al (2021) Efficacy and safety of endoscopic submucosal tunnel dissection for rectal laterally spreading tumors. Surg Endosc 35(8):4356–4362. 10.1007/s00464-020-07927-4 [DOI] [PubMed] [Google Scholar]
- 18.Kimura H, Yabuuchi Y, Notsu A et al (2021) Features of post-endoscopic submucosal dissection electrocoagulation syndrome for early gastric neoplasm. J Gastroenterol Hepatol 36(11):3164–3169. 10.1111/jgh.15583 [DOI] [PubMed] [Google Scholar]
- 19.Tanaka S, Kashida H, Saito Y et al (2020) Japan Gastroenterological Endoscopy Society guidelines for colorectal endoscopic submucosal dissection/endoscopic mucosal resection. Dig Endosc 32(2):219–239. 10.1111/den.13545 [DOI] [PubMed] [Google Scholar]
- 20.Dixon MF (2002) Gastrointestinal epithelial neoplasia: Vienna revisited. Gut 51(1):130–131. 10.1136/gut.51.1.130 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Matsumoto A, Tanaka S, Oba S et al (2010) Outcome of endoscopic submucosal dissection for colorectal tumors accompanied by fibrosis. Scand J Gastroenterol 45(11):1329–1337. 10.3109/00365521.2010.495416 [DOI] [PubMed] [Google Scholar]
- 22.Song HY, Kim SJ, Lee J et al (2024) Risk factors for post-endoscopic submucosal dissection electrocoagulation syndrome in patients with colorectal neoplasms: A multicenter, large-scale, retrospective cohort study by the Honam Association for the Study of Intestinal Disease (HASID). J Clin Med 13(13):3932. 10.3390/jcm13133932 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Iwatsubo T, Takeuchi Y, Yamasaki Y et al (2019) Differences in clinical course of intraprocedural and delayed perforation caused by endoscopic submucosal dissection for colorectal neoplasms: a retrospective study. Dig Dis 37(1):53–62. 10.1159/000492868 [DOI] [PubMed] [Google Scholar]
- 24.Chiba H, Tachikawa J, Arimoto J et al (2020) Predictive factors of mild and severe fibrosis in colorectal endoscopic submucosal dissection. Dig Dis Sci 65(1):232–242. 10.1007/s10620-019-05735-y [DOI] [PubMed] [Google Scholar]
- 25.Kuroha M, Shiga H, Kanazawa Y et al (2021) Factors associated with fibrosis during colorectal endoscopic submucosal dissection: Does pretreatment biopsy potentially elicit submucosal fibrosis and affect endoscopic submucosal dissection outcomes? Digestion 102(4):590–598. 10.1159/000510145 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Ma DW, Youn YH, Jung DH et al (2018) Risk factors of electrocoagulation syndrome after esophageal endoscopic submucosal dissection. World J Gastroenterol 24(10):1144–1151. 10.3748/wjg.v24.i10.1144 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Kim Y, Na HK, Ahn JY et al (2024) Association of local steroid injection as a risk factor for electrocoagulation syndrome after esophageal endoscopic submucosal dissection. Surg Endosc 38(7):3858–3865. 10.1007/s00464-024-10927-3 [DOI] [PubMed] [Google Scholar]
- 28.Medzikovic L, Aryan L, Eghbali M (2019) Connecting sex differences, estrogen signaling, and microRNAs in cardiac fibrosis. J Mol Med (Berl) 97(10):1385–1398. 10.1007/s00109-019-01833-6 [DOI] [PubMed] [Google Scholar]
- 29.Qiu J, Ouyang Q, Zhang Y et al (2022) Post-endoscopic submucosal dissection electrocoagulation syndrome: a clinical overview. Expert Rev Gastroenterol Hepatol 16(11–12):1079–1087. 10.1080/17474124.2022.2156858 [DOI] [PubMed] [Google Scholar]
- 30.Chen W, Zhu XN, Wang J et al (2019) Risk factors for Mallory-Weiss tear during endoscopic submucosal dissection of superficial esophageal neoplasms. World J Gastroenterol 25(34):5174–5184. 10.3748/wjg.v25.i34.5174 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Yamaguchi H, Fukuzawa M, Kawai T et al (2020) Predictive factors of postendoscopic submucosal dissection electrocoagulation syndrome and the utility of computed tomography scan after esophageal endoscopic submucosal dissection. Digestion 101(5):579–589. 10.1159/000501478 [DOI] [PubMed] [Google Scholar]
- 32.Sugimoto M, Murata M, Kawai T (2023) Assessment of delayed bleeding after endoscopic submucosal dissection of early-stage gastrointestinal tumors in patients receiving direct oral anticoagulants. World J Gastroenterol 29(19):2916–2931. 10.3748/wjg.v29.i19.2916 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Sanomura Y, Oka S, Tanaka S et al (2018) Taking warfarin with heparin replacement and direct oral anticoagulant is a risk factor for bleeding after endoscopic submucosal dissection for early gastric cancer. Digestion 97(3):240–249. 10.1159/000485026 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Data availability can be requested from the corresponding author with a reasonable justification.