Esophageal carcinoma (EC) represents a significant global health burden, ranking as the 11th most common malignancy and the 7th leading cause of cancer-related mortality worldwide1. Endoscopic submucosal dissection (ESD) has emerged as the gold-standard treatment for superficial esophageal neoplasms confined to the mucosa (T1a) or superficial submucosa (T1b-sm1), offering comparable oncologic outcomes to esophagectomy while preserving organ function and minimizing surgical morbidity2,3. Despite its therapeutic advantages, post-ESD esophageal stricture (ES) remains a critical limitation, particularly in extensive resections involving >3/4 of the esophageal circumference where ES rates exceed 90%4, reaching near-universal incidence following circumferential ESD5,6,7 This adverse event prompted the 2017 Japanese Esophageal Cancer Guidelines to exclude lesions involving >2/3 circumference from absolute ESD indications8. Recent advancements in ES prophylaxis have expanded ESD eligibility criteria through multimodal approaches: steroid administration (oral or intralesional)9, biodegradable stents10,11, and tissue engineering strategies (autografts, cell sheets, polyglycolic acid membranes)12,13,14. Systemic prednisolone administration has gained prominence due to its cost-effectiveness and ease of implementation15, enabling even circumferential resections to achieve satisfactory long-term functional outcomes when combined with proactive dilation16,17,18. Nevertheless, approximately 25% of patients develop refractory ES despite optimized steroid regimens19, compounded by corticosteroid-related adverse events including osteoporosis and glucose intolerance. While the efficacy of steroid prophylaxis is established, a critical and understudied question remains: even under standardized prophylactic therapy, which patients are at the highest risk for treatment failure? Current research priorities predominantly focus on prophylaxis optimization, leaving critical gaps in understanding risk factors for steroid-resistant ES. This knowledge deficit hinders risk stratification and targeted interventions for high-risk cohorts. To address this unmet need, we conducted a retrospective cohort study analyzing clinical and procedural predictors of steroid-refractory ES following extensive ESD (>2/3 circumference) in early EC. Through comprehensive regression modeling, we aimed to establish evidence-based criteria for individualized prevention strategies and enhanced postoperative surveillance protocols.

This retrospective dual-center study was conducted at the First Affiliated Hospital of Xi'an Jiaotong University and the First Affiliated Hospital of Zhengzhou University. The study population comprised patients with superficial EC who underwent ESD with immediate postoperative oral steroid prophylaxis between January 2019 and June 2024.

Inclusion criteria were: (1) histopathologically confirmed superficial EC with preoperative contrast-enhanced computed tomography(CT) confirmation of non-metastatic disease; (2) post-ESD mucosal defects encompassing ≥ 2/3 of the esophageal circumference; (3) final pathological diagnosis of high-grade squamous intraepithelial neoplasia (HSIN) or SCC according to WHO classification criteria; (4) availability of complete clinical follow-up data for ≥ 6 months.

Exclusion criteria included: (1) receipt of adjuvant therapies including repeat ESD, surgical resection, chemoradiotherapy, or radiofrequency ablation; (2) absence of standardized ES prevention measures; (3) implementation of alternative ES prevention strategies such as endoscopic stenting or mucosal transplantation. This study adhered to the ethical guidelines of the Helsinki Declaration. The ethics committee that approved this study and waived the consent requirement is the Ethics Committee of The First Affiliated Hospital of Xi'an Jiaotong University (Approval Number: XJTU1AF2021LSK-208).

All ESD procedures were conducted by highly experienced endoscopists with comparable proficiency, each having over a decade of clinical practice and having independently performed more than 200 esophageal ESD surgeries. All operations were carried out under general anesthesia with tracheal intubation. Patients were positioned in the left lateral decubitus position, and the procedures utilized an Olympus GIF-Q260J electronic gastroscope (Olympus, Japan). For circumferential esophageal lesions, either the "tunneling technique" or "snare traction method" was employed as auxiliary approaches for submucosal dissection, depending on the specific circumstances. The detailed procedural steps are as follows (Fig. 1):①Lesion Marking: Following staining with 1.25% Lugol's solution, the margin of the esophageal lesion was delineated using a Dual knife (KD-650 L, Olympus, Japan), marking a border 5 mm outside the lesion.②Submucosal Injection: A disposable injection needle (NM-200 L-0523, Olympus, Japan) was used to inject a solution consisting of 0.2 mL methylene blue (with a mass concentration of 10 g/L) mixed with 200 mL of 0.9% sodium chloride solution to elevate the lesion.③Mucosal Incision: Using the Dual knife, an incision was made approximately 5 mm outside the marked border to cut the mucosal edge, exposing the submucosal layer. The lesion was then carefully and progressively dissected from the muscularis propria using electrocautery until complete separation was achieved.④Specimen Handling: The resected specimen was fixed in formalin and sent for pathological examination to determine the nature of the lesion, as well as the involvement of its margins and base.⑤Post-Procedure Inspection: After the specimen was removed, the endoscope was reintroduced to inspect the resection site for any bleeding. Electrocoagulation biopsy forceps (FD-410LR, Olympus, Japan) were used for hemostasis when necessary. Titanium clips were applied to close any significant defects in the muscularis propria to prevent perforation.

Postoperative Management: Patients were kept fasting with water restriction and received parenteral nutrition for 3 days. Acid suppression was achieved with intravenous proton pump inhibitors (PPIs) at a dosage of 40 mg every 12 h. Mucosal protection was provided orally after diet resumption using agents such as sucralfate or rebamipide. An indwelling gastric tube was placed for 3 days to facilitate gastrointestinal decompression. A gradual transition to a normal diet was implemented over the following 3 days, starting with a liquid diet.

Starting from the 3rd day post-ESD, patients were administered oral prednisone acetate tablets at a dosage of 30 mg/day for a duration of 2 weeks. The dosage was then tapered down to 25 mg/day during the 3rd-4th week, followed by a reduction to 20 mg/day in the 5th week, 15 mg/day in the 6th week, 10 mg/day in the 7th week, and finally 5 mg/day in the 8th week. Concurrently, PPIs were administered at a dosage of 20 mg/day, along with oral calcium supplements, which were continued until the discontinuation of steroid therapy. Potassium supplementation was provided as needed based on the patient's condition.

Systematic esophagogastroduodenoscopy (EGD) was performed at 1, 3, 6, and 12 months post-ESD. Patients presenting with dysphagia between scheduled visits underwent immediate endoscopic evaluation. Dysphagia severity was graded according to the Stooler classificationESD. For patients with a grade of 2 or higher, endoscopic bougie/balloon dilation (EBD) is performed. The procedure involves inserting a guidewire, followed by the insertion of a silicone bougie or balloon along the guidewire for dilation. The diameter is expanded to 11-15 mm based on intraoperative conditions (Fig. 2).

ES: Defined as a narrowing with a diameter of less than 9.8 mm, through which a standard endoscope (GIF Q260, Olympus) cannot pass. Clinical Remission of ES: Maintenance of a dysphagia-free state for ≥ 6 months without the need for any intervention. EBD Times: The number of EBD sessions required to achieve ES remission (defined as maintaining an unobstructed esophageal lumen for more than 6 months). If the interval between two dilation sessions is less than 7 days, they are considered part of the same treatment cycle. Refractory benign esophageal strictures (RBES): failure to maintain patency after ≥ 5 EBD sessions to achieve sustained luminal patency for ≥ 6 months. Classification was made at the end of the follow-up period.

Categorical variables are expressed as counts and percentages. Continuous variables are presented as mean ± standard deviation or median and interquartile range (IQR). Categorical variables were compared using χ test, adjusted Chi-square test, or Fisher's exact test. Continuous variables were analyzed with t-test or Mann-Whitney U test. Univariate and multivariate logistic regression identified independent risk factors for ES in steroid-treated patients. Kaplan-Meier curves assessed freedom from EBD rates. Analyses were performed using SPSS 25.0 and GraphPad Prism 10.3.1. Statistical significance was defined as P < 0.05.