Background: Boerhaave syndrome is defined as a spontaneous perforation of the oesophagus. The mainstay of treatment is resuscitation of the patient, closure of the oesophageal defect, and drainage of perioesophageal and pleural fluid collections. Whether the optimal approach is endoscopic, surgical, or conservative management remains unknown and there are no clear guidelines. The aim of this multicentre retrospective cohort study was to evaluate current treatment strategies and outcomes for Boerhaave syndrome. Methods: A multicentre retrospective analysis of data from 23 participating European tertiary centres was performed. Patients with Boerhaave syndrome treated between January 2019 and December 2023 were eligible for inclusion. The primary endpoint was the length of ICU stay and secondary endpoints included in-hospital mortality, 90-day mortality, and the length of overall hospital stay. Results: In total, 216 patients were included; 151 were men (70%), the median age was 62 (22–95) years, and 81 (40%) were treated >24 h after the start of symptoms. Seventy (32%) patients were managed endoscopically (group I), 73 (34%) were managed surgically (group II), 67 (31%) were managed using a combination of endoscopy and surgery (group III), and 6 (3%) were managed using other methods (group IV). For patients in groups I–III: the median length of ICU stay was 8 (0–67) days, with no differences between the three groups (P = 0.105); the in-hospital mortality rate and 90-day mortality rate were both 12% (P = 0.490 and P = 0.637, respectively); and the median length of overall hospital stay was 27 (range 1–193) days, with the longest stays observed in patients who received combined treatment (P = 0.032). Conclusion: This study provides a comprehensive overview of the current treatment strategies and outcomes for patients with Boerhaave syndrome in Europe.
Background: The aim of this study was to establish a deep learning prediction model for neoadjuvant FLOT chemotherapy response. The neural network utilized clinical data and visual information from whole-slide images (WSIs) of therapy-naïve gastroesophageal cancer biopsies. Methods: This study included 78 patients from the University Hospital of Cologne and 59 patients from the University Hospital of Heidelberg used as external validation. Results: After surgical resection, 33 patients from Cologne (42.3%) were ypN0 and 45 patients (57.7%) were ypN+, while 23 patients from Heidelberg (39.0%) were ypN0 and 36 patients (61.0%) were ypN+ (p = 0.695). The neural network had an accuracy of 92.1% to predict lymph node metastasis and the area under the curve (AUC) was 0.726. A total of 43 patients from Cologne (55.1%) had less than 50% residual vital tumor (RVT) compared to 34 patients from Heidelberg (57.6%, p = 0.955). The model was able to predict tumor regression with an error of ±14.1% and an AUC of 0.648. Conclusions: This study demonstrates that visual features extracted by deep learning from therapy-naïve biopsies of gastroesophageal adenocarcinomas correlate with positive lymph nodes and tumor regression. The results will be confirmed in prospective studies to achieve early allocation of patients to the most promising treatment.
Objective: To determine the impact of severity of esophageal anastomotic leak (AL), standardized by ECCG grading, on long-term survival within a contemporary multicenter cohort. Summary Background data: Historical evidence suggesting AL is associated with increased recurrence and poor survival is limited by heterogeneous definitions and outdated management. This study utilizes the standardized ECCG definition to evaluate the prognostic significance of AL severity under modern perioperative protocols and endoscopic rescue strategies. Methods: An international multicenter cohort across 17 high-volume European centers. Adult patients with esophageal or junctional cancer treated with neoadjuvant CROSS or FLOT, followed by esophagectomy were included (2018–2023). Results: Out of 2905 patients, 425 (14.6%) developed an AL. AL was associated with a nearly two-fold increase in pulmonary complications (46.8% vs. 26.7%; P <0.001). AL requiring surgical reintervention (type III) was associated with an increased 30- and 90-days mortality rate (7.3% vs. 2.7% and 12.3% vs. 4%, P <0.001) and a significant reduction in median overall survival versus the no leak group (33.9 vs. 69.3; months P <0.001). After multivariable adjustment, type III AL was associated with a greater likelihood of death (HR 1.51; 95% CI: 1.04–2.18; P =0.029). A lower rate of adjuvant therapy administration (33.9% vs. 43.3%; P =0.004) was observe for AL patients. No significant survival deficits were observed for leaks treated conservatively or with no-surgical intervention. Conclusions: AL requiring surgical reintervention was associated with impaired survival. This relationship identifies a high-risk clinical profile and represents a critical target for future research into strategies for clinical stabilization and risk mitigation.
17 September 2026 in Journal of Laparoendoscopic & Advanced Surgical Techniques
Background: Accurate preoperative staging of esophageal carcinoma guides the choice between neoadjuvant therapy and primary surgery. Endoscopic ultrasound (EUS) remains the principal modality for locoregional staging, yet its accuracy varies. We evaluated EUS diagnostic performance against final pathology in a homogeneous cohort undergoing primary esophagectomy at a high-volume center and identified factors influencing its reliability. Methods: This retrospective analysis included 295 patients with esophageal adenocarcinoma (78%) or squamous cell carcinoma (22%) who underwent oncological esophagectomy between 2006 and 2017, primarily without neoadjuvant therapy. Preoperative uTNM staging by EUS was compared with pathological pTNM. Predictors of staging concordance were assessed by multivariate logistic regression. Results: Overall T-stage accuracy was 57.2%. EUS correctly identified 85.9% of pT1 tumors but only 17.0% of pT2 and 64.6% of pT3 lesions. Tumors were overstaged in 36.6% and understaged in 6.1% of cases. Nodal staging accuracy was 65.8% (sensitivity 65%, specificity 68%). Younger age, female sex, shorter height, and ASA II status were independent predictors of reduced T-stage accuracy; no significant predictors emerged for N-stage. Conclusion: EUS accurately detects early-stage esophageal cancer but shows limited reliability for intermediate stages, particularly pT2 tumors, in this single-center study. Given the risk of misclassification, EUS findings should be interpreted cautiously and complemented by additional diagnostic tools to improve preoperative assessment and enable individualized treatment decisions.