Standard preoperative variables are insufficient for individualized risk stratification of RLNP, anastomotic leak, or vocal cord palsy after esophagectomy, and improved prediction will require prospective integration of real-time intraoperative data.
Abstract
Esophageal Cancer: Surgical Treatment of Esophageal Cancer
Preoperative risk stratification for esophagectomy complications relies on clinical prediction models; however, their discriminative performance in multi-institutional settings remains poorly defined. We hypothesized that standard preoperative variables would demonstrate limited predictive validity across heterogeneous surgical cohorts.
We analyzed 2,490 patients undergoing esophagectomy across four institutions in Asia (total n=2,490; individual center n range 75–1,012). Four outcomes were studied: recurrent laryngeal nerve palsy (RLNP), anastomotic leak (AL), pulmonary complications (PC), and vocal cord palsy (VCP). Logistic regression models with bootstrap-validated odds ratios (1,000 iterations) were evaluated by 5-fold cross-validated AUC. SHAP (SHapley Additive exPlanations) via Gradient Boosting Machines quantified variable importance. Decision curve analysis (DCA) assessed net clinical benefit across threshold probabilities 2–70%. Association between tumor location and each complication was assessed using chi-squared tests.
All prediction models demonstrated poor-to-fair discrimination: RLNP AUC 0.533, AL AUC 0.586, PC AUC 0.676, and VCP AUC 0.556. Tumor location was the only statistically significant categorical predictor of RLNP—upper/cervical location was associated with higher RLNP incidence compared to middle thoracic tumors (39.1% vs. 28.0%; OR 1.22, 95%CI 1.05–1.42; p=0.009). No significant association was observed between tumor location and AL, PC, or VCP. DCA demonstrated negligible clinical net benefit for RLNP and AL models; only the PC model provided modest benefit (max net benefit gain +0.057) at threshold probabilities of 5–20%. SHAP analysis identified FEV1%, PNI score, and BMI as the highest-importance variables for RLNP prediction, with tumor location ranking sixth—indicating that location contributes a statistically real but clinically modest signal.
Standard preoperative variables are insufficient for individualized risk stratification of RLNP, anastomotic leak, or vocal cord palsy after esophagectomy. Statistical significance (p=0.009 for location–RLNP association) does not translate to clinically meaningful predictive power (AUC 0.533). Tumor location should be incorporated into RLNP preoperative counseling. Improved prediction will require prospective integration of real-time intraoperative data. Pulmonary complication risk approaches clinically actionable prediction (AUC 0.676) and may guide respiratory prehabilitation targeting.
Esophageal Cancer: Other
Esophagectomy is associated with substantial morbidity, mortality, and resource utilization. Traditional regression-based risk tools may inadequately capture complex nonlinear interactions. Contemporary evidence on machine learning (ML) models predicting postoperative outcomes after esophagectomy was synthesized, focusing on discrimination, validation, and comparison with conventional regression approaches.
A PRISMA-guided systematic review was conducted using Embase, MEDLINE, PubMed, and the Cochrane Library in January 2026. A total of 196 studies were identified. After title and abstract screening, 32 studies underwent full-text review, of which 10 met final inclusion criteria as ML-focused prediction models in esophagectomy populations. Extracted data included study design, cohort size, procedure type, predicted outcome, modeling approach (ML versus regression), validation strategy (internal or external), performance metrics (e.g., area under the receiver operating characteristic curve [AUROC]), and reporting elements such as calibration and decision-curve analysis. ML-focused studies were defined as those applying algorithms including gradient boosting, support vector machines, neural networks, or survival forests to postoperative outcome prediction.
Ten studies applying ML models to esophagectomy outcomes were included (median cohort size 700; range 200–4700). Anastomotic leak was the most frequently predicted outcome (4/10), followed by mortality, major complications, readmission, strictures, and recurrence or survival. Common algorithms included gradient boosting (XGBoost, LightGBM, GBM), support vector machines, neural networks, and survival forests. Reported discrimination ranged from moderate to high (AUROC 0.64 for 90-day mortality and 0.65–0.70 for major complications, increasing to 0.79–0.90 for anastomotic leak prediction; some internally validated models reported AUROC >0.95). Three studies performed independent external validation, and performance generally declined in external cohorts. Comparative analyses demonstrated that ML often matched but did not consistently outperform regression-based models.
Machine learning models for postoperative risk prediction after esophagectomy demonstrate promising discrimination, particularly for anastomotic leak. Although external validation remains limited, ML approaches are still in early stages of clinical translation. With prospective data integration and robust multicenter validation, ML has the potential to enhance individualized risk stratification, support shared decision-making, guide perioperative planning, and improve allocation of postoperative resources in esophageal surgery.
T. Wang, Otari Beldishevski-Shotadze, N. Evennett· Diseases of the esophagus· 0 citations
The developed ML models for predicting AL after esophagectomy demonstrated acceptable discriminative ability and identified key risk factors including radiotherapy, hemoglobin drop, bleeding, and lymphocyte count and provide a data-driven foundation for risk stratification.
Chi Zhang, Yongde Liao· Diseases of the esophagus· 0 citations
Esophageal Cancer: Surgical Treatment of Esophageal Cancer
Anastomotic leak remains the most feared complication after esophagectomy, with reported rates of 5–20%. Indocyanine green (ICG) fluorescence angiography enables real-time perfusion assessment, yet no validated scoring system integrates ICG parameters with clinical risk factors. We aimed to develop and internally validate the FLARE (Fluorescence-guided Anastomotic Leak Risk Evaluation) score.
Ambispective analysis of 148 consecutive esophagectomy patients (January 2022–June 2025): prospective-ICG group (n=71) and retrospective non-ICG group (n=77). Primary outcome: anastomotic leak. Univariate analysis used Fisher's Exact Test/Chi-Square Test and Mann-Whitney U test. The 3-variable Clinical-EARS (Esophageal Anastomotic Leak Risk Score) was derived using OR-weighted integer assignment from statistically significant predictors: male sex (3 points), active smoking (2 points), tracheostomy (1 point); maximum 6 points. Variables with ROC-AUC below 0.60 (operative duration, blood loss) were excluded on statistical grounds. Internal validation was performed in the non-ICG cohort using bootstrap resampling (n=2000 iterations) with Hosmer-Lemeshow calibration testing and 5-fold cross-validation. The ICG-FLARE score additionally incorporated ICG perfusion peak time greater than 20 seconds (3 points) and ICG–naked eye assessment mismatch (3 points); maximum 12 points. Secondary outcomes included chyle leak.
Anastomotic leak occurred in 7.0% (ICG) versus 15.6% (non-ICG; OR=0.41, p=0.171). ICG significantly reduced chyle leak (1.4% vs 11.7%; OR=0.11; p=0.018). Male sex (OR=10.53; p=0.006) and smoking (OR=4.74; p=0.018) were the only independent predictors on univariate analysis. Clinical-EARS stratified risk stepwise across all 148 patients: Low (0–1 pts) 2.4%, Medium (2–3 pts) 5.4%, High (4–6 pts) 20.3% (AUC=0.734; 95%CI: 0.640–0.821). Internal validation in the non-ICG cohort confirmed discrimination (AUC=0.685; 95%CI: 0.563–0.798) with minimal optimism (0.049) and excellent calibration (Hosmer-Lemeshow p=0.977). ICG-FLARE substantially improved performance in ICG patients (AUC=0.918; 95%CI: 0.802–0.998; ΔAUC=+0.159). At the optimal cutoff (≥7 points), ICG-FLARE achieved a sensitivity 80.0% and a specificity 90.9%. Perfusion peak time greater than 20 seconds (OR=14.86; p=0.013) and ICG–naked eye mismatch (OR=18.3; p=0.009) were independently predictive.
The FLARE score provides validated, clinically actionable risk stratification for anastomotic leak after esophagectomy. Clinical-EARS achieves robust discrimination (AUC=0.734) with excellent calibration and internal validation, applicable to all patients regardless of ICG availability. ICG-FLARE dramatically enhances performance (AUC=0.918; sensitivity 80%, specificity 90.9%), demonstrating that ICG fluorescence angiography delivers independent prognostic information beyond clinical factors alone. Prospective external validation is warranted.
Abhijit Talukdar, Mohit Malhotra, Dibyajyoti Deka et al.· Diseases of the esophagus· 0 citations
Background Esophageal cancer remains one of the leading causes of cancer-related mortality worldwide. Anastomotic leakage (AL) following esophagectomy is a major postoperative complication that significantly impacts patient outcomes, including mortality, morbidity, prolonged hospital stays, and increased healthcare costs. Despite advances in surgical techniques and adjuvant therapies, predicting the risk of AL remains a challenge. Objective This study aims to develop and validate a predictive model for assessing the risk of AL in esophageal cancer patients undergoing esophagectomy, based on comprehensive clinical and laboratory variables. Methods This retrospective cohort study included 650 esophageal cancer patients who underwent esophagectomy between January 2015 and May 2025, divided into a training set (n = 455) and a validation set (n = 195) at 7:3 ratio. Baseline demographic, clinicopathological, and laboratory data were collected, with AL as the primary outcome, defined according to the Esophagectomy Complications Consensus Group (ECCG). Univariable and multivariable logistic regression, restricted cubic splines (RCS), and nomogram development to identify predictors, with model performance assessed using receiver operating characteristic (ROC) curve, calibration plots, and decision curve analysis (DCA). Results Seven significant predictors of AL were identified in the training set: age, neoadjuvant radiotherapy, C-reactive protein-albumin-lymphocyte (CALLY) index, hypertension, neutrophil-to-lymphocyte ratio (NLR), neutrophil-to-monocyte ratio (NMR), and platelet-to-lymphocyte ratio (PLR). A nomogram model was developed, showing good discrimination (AUC = 0.813) and calibration in the training set. The validation cohort demonstrated moderate predictive accuracy (AUC = 0.763), with consistent net benefits observed across different risk thresholds in DCA. Conclusions In conclusion, this study established a potentially useful predictive model for AL risk, which may facilitate individualized risk stratification, guide perioperative decision-making, and ultimately contribute to reducing AL incidence and improving postoperative recovery.
Ruonan Tan, Lili Guo, Saitian Li et al.· Frontiers in Oncology· 0 citations
This LASSO-derived nomogram provides transparent, bedside-applicable risk stratification for postoperative composite complications in CRC surgery, identifying nutritional status as the dominant modifiable predictor and supporting targeted perioperative optimization.
Lu Wang, Shunshun Wang, Liqin Deng· Frontiers in Nutrition· 0 citations
Esophageal Cancer: Surgical Treatment of Esophageal Cancer
Esophagectomy for esophageal cancer carries substantial morbidity, yet the extent to which institutional surgical strategy—encompassing operative approach, lymphadenectomy extent, and neoadjuvant treatment protocol—determines postoperative complication profiles remains poorly characterized across different geographic practice environments.
We performed a retrospective analysis of a prospectively maintained international multicenter database (ISDE) comprising 2,490 patients who underwent curative-intent esophagectomy across four centers (Centers A–D; n=825, 499, 1,012, and 154 respectively). Primary outcomes were postoperative complication rates, in-hospital mortality, and length of stay. Given substantial heterogeneity in neoadjuvant therapy rates (Center A 81.1% versus Center D 11.0%), pathological stage distribution, and lymphadenectomy extent, all cross-institutional comparisons were performed descriptively with formal confounding adjustment deferred to propensity-matched subgroup analyses.
Marked inter-institutional variation in surgical strategy was observed. Center A was characterized by near-universal thoracoscopic-laparoscopic esophagectomy (93.7% thoracoscopy), three-field lymphadenectomy, and the highest neoadjuvant treatment rate (81.1%). Center B predominantly underwent open Ivor-Lewis esophagectomy (80.6%), with robotic assistance in 13.6% and concurrent chemoradiotherapy in 46.3%. Center C demonstrated approximately 70% minimally invasive adoption with two-field dissection, while Center D uniformly employed McKeown esophagectomy with selected cervical dissection. Center A exhibited the highest rate of recurrent laryngeal nerve palsy (28.6%) and chylothorax (7.8%), reflecting systematic radical dissection. Despite this, in-hospital mortality was 0.7%, ICU stay was a median of 3 days (IQR 2–3), and postoperative length of stay was 14 days (IQR 12–19). Anastomotic leakage rates were comparable across centers with available data (Center A 9.3%; Center C 10.7%; Center D 11.0%). Recurrent laryngeal nerve injury was lower in centers employing two-field dissection (Center C 8.3%; Center D 15.6%).
Operative strategy is the primary driver of postoperative complication phenotype in esophageal cancer surgery. Three-field lymphadenectomy uniquely confers high recurrent laryngeal nerve palsy and chylothorax rates despite low mortality. Valid cross-institutional outcome benchmarking requires rigorous adjustment for neoadjuvant treatment exposure, pathological stage composition, and dissection field extent.
Si-miao Lu, Yi Zhu, Yong-tao Han et al.· Diseases of the esophagus· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
