Yang KZ, Chen L, Xu L, Xu BX, Li MY, Wang Z, Lu Q. Correlation between sarcopenia and esophageal stenosis following endoscopic submucosal dissection and construction of a postoperative stenosis risk model. World J Gastroenterol 2026; 32(5): 114752 [DOI: 10.3748/wjg.v32.i5.114752]
Corresponding Author of This Article
Qin Lu, MD, Associate Chief Physician, Research Fellow, Department of Gastroenterology, Zhongda Hospital Southeast University, No. 87 Dingjiaqiao, Gulou District, Nanjing 210009, Jiangsu Province, China. luqin81287@163.com
Research Domain of This Article
Gastroenterology & Hepatology
Article-Type of This Article
Retrospective Study
Open-Access Policy of This Article
This article is an open-access article which was selected by an in-house editor and fully peer-reviewed by external reviewers. It is distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/
Feb 7, 2026 (publication date) through Jan 28, 2026
Times Cited of This Article
Times Cited (0)
Journal Information of This Article
Publication Name
World Journal of Gastroenterology
ISSN
1007-9327
Publisher of This Article
Baishideng Publishing Group Inc, 7041 Koll Center Parkway, Suite 160, Pleasanton, CA 94566, USA
Share the Article
Yang KZ, Chen L, Xu L, Xu BX, Li MY, Wang Z, Lu Q. Correlation between sarcopenia and esophageal stenosis following endoscopic submucosal dissection and construction of a postoperative stenosis risk model. World J Gastroenterol 2026; 32(5): 114752 [DOI: 10.3748/wjg.v32.i5.114752]
World J Gastroenterol. Feb 7, 2026; 32(5): 114752 Published online Feb 7, 2026. doi: 10.3748/wjg.v32.i5.114752
Correlation between sarcopenia and esophageal stenosis following endoscopic submucosal dissection and construction of a postoperative stenosis risk model
Author contributions: Yang KZ contributed to data curation; methodology; writing - original draft; Xu L contributed to software; Xu BX and Wang Z contributed to validation; Lu Q contributed to conceptualization; Yang KZ, Chen L, Xu L and Li MY contributed to formal analysis; Yang KZ and Lu Q contributed to writing - review & editing; Li MY and Lu Q contributed to funding acquisition; and all contributing authors read and approved the final manuscript.
Supported by the National Natural Science Foundation of China, No. 82300638; the General Program of Jiangsu Provincial Administration of Traditional Chinese Medicine, No. MS2024139; and Zhongda Hospital Affiliated to Southeast University, Jiangsu Province High-Level Hospital Construction Funds, No. YKK24268.
Institutional review board statement: This study was approved by the Ethics Committee of Southeast University, Zhongda Hospital Southeast University (approval No. 2025ZDSYLL373-P01). All study procedures adhered to the Declaration of Helsinki.
Informed consent statement: The ethics committee waived the requirement for obtaining informed consent from the patients owing to the retrospective nature of the study.
Conflict-of-interest statement: The authors declare that they have no conflicts of interest related to the manuscript.
Data sharing statement: The data used and analyzed during the study are available from the lead author and the corresponding authors upon reasonable request.
Open Access: This article is an open-access article that was selected by an in-house editor and fully peer-reviewed by external reviewers. It is distributed in accordance with the Creative Commons Attribution NonCommercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: https://creativecommons.org/Licenses/by-nc/4.0/
Corresponding author: Qin Lu, MD, Associate Chief Physician, Research Fellow, Department of Gastroenterology, Zhongda Hospital Southeast University, No. 87 Dingjiaqiao, Gulou District, Nanjing 210009, Jiangsu Province, China. luqin81287@163.com
Received: September 28, 2025 Revised: November 20, 2025 Accepted: December 29, 2025 Published online: February 7, 2026 Processing time: 123 Days and 14.6 Hours
Abstract
BACKGROUND
Sarcopenia has been indicated to be related to the postoperative outcome of patients with various digestive tract diseases. However, no studies have investigated the association between sarcopenia and esophageal stenosis after endoscopic submucosal dissection (ESD).
AIM
To explore the correlation between sarcopenia and post-ESD esophageal stenosis, and subsequently develop a risk prediction model.
METHODS
Retrospective data from 499 patients who underwent esophageal ESD were collected. After stratification via the L3 skeletal muscle indices (L3-SMIs) into sarcopenia and non-sarcopenia groups, post-ESD stenosis rates were compared. Propensity score matching (PSM) was used for sensitivity analysis. The original cohort was randomly split at a ratio of 7:3 into training (n = 350) and validation (n = 149) groups to construct and validate a risk prediction model for post-ESD stenosis.
RESULTS
Sarcopenia was significantly associated with post-ESD esophageal stenosis (48.23% vs 22.35%, P < 0.001). Furthermore, multivariate analysis confirmed its independence as a predictor of this postoperative complication [odds ratio (OR): 3.86; 95% confidence interval: 1.76-8.45; P < 0.001]. This conclusion was consistent across the subgroup analyses and PSM analyses. The risk prediction model incorporating sarcopenia had area under the curve values of 0.848 (training set) and 0.794 (validation set). Calibration curves and Hosmer-Lemeshow tests indicated good calibration of the model. Moreover, decision curve analysis confirmed a positive net clinical benefit for the model.
CONCLUSION
Sarcopenia is an independent risk predictor of post-ESD esophageal stenosis. Our model integrating muscle mass assessment aids in early high-risk identification and intervention.
Core Tip: A novel finding of this study is that a body composition abnormality, quantified by an index associated with computed tomography-defined sarcopenia, is a clinically significant and independent predictor of the development of esophageal stenosis after endoscopic submucosal dissection (ESD). We successfully constructed a risk prediction model for post-ESD stenosis that includes sarcopenia, which achieved good internal validation. This investigation addresses a limitation of prior research, which focused solely on local lesion characteristics for predicting post-ESD esophageal stenosis. Moreover, the model helps in identifying high-risk groups at early time points and formulating individualized treatment strategies.
Citation: Yang KZ, Chen L, Xu L, Xu BX, Li MY, Wang Z, Lu Q. Correlation between sarcopenia and esophageal stenosis following endoscopic submucosal dissection and construction of a postoperative stenosis risk model. World J Gastroenterol 2026; 32(5): 114752
Endoscopic submucosal dissection (ESD) is considered the standard therapeutic approach for superficial esophageal carcinoma. Compared with surgery, ESD has advantages such as minimal trauma, rapid recovery, and fewer complications[1,2]. However, owing to the narrowness of the esophagus, its thin wall, and the absence of a plasma layer[3], complications such as esophageal stenosis are prone to occur after the procedure. Postoperative stenosis can cause dysphagia and other types of discomfort, thus severely affecting patients' quality of life and increasing hospitalization costs. Previous studies have reported esophageal stenosis rates of 11%-20% following ESD. Lesions encompassing more than 3/4 of the circumferential area are among the high-risk factors for stenosis, with an incidence as high as 70% being observed[4]. Moreover, complete circumferential lesion resection results in near-universal stricture formation (approximately 100% incidence)[5].
Therefore, the accurate prediction of high-risk patients and the timely development of measures to prevent postoperative stenosis are crucial for the prognosis of this condition and treatment of these patients. Recent studies on risk factors for postoperative esophageal ESD stenosis have focused on local factors (such as lesion size, extent of infiltration, and operation time) while ignoring systemic factors (such as nutritional status and muscle mass)[6,7]. In clinical practice, body mass index (BMI) does not appear to affect the occurrence of postoperative esophageal ESD stenosis, nor does it comprehensively reflect patients' nutritional status. Sarcopenia is a progressive and generalized skeletal muscle disorder involving the accelerated loss of muscle mass and function (strength or performance). In recent years, this condition has evolved from simple muscle mass loss to a unique muscle disorder linked to the postoperative prognosis of various diseases[8,9].
Hisada et al[10] identified sarcopenia as a clinically significant predictor of both short- and long-term adverse events after ESD for early gastric cancer. Moreover, Su et al[11] further demonstrated its association with a higher incidence of postoperative pneumonia and increased five-year mortality among elderly patients undergoing gastric and colorectal ESD. A meta-analysis encompassing 41 studies revealed that sarcopenia was significantly correlated with poor survival and increased complication risk in esophageal carcinoma cohorts[12]. However, no studies have investigated how sarcopenia affects complications following esophageal ESD[13].
Therefore, this study aimed to examine the link between sarcopenia, as assessed by computed tomography (CT), and the risk of post-ESD esophageal stenosis. In addition, by combining sarcopenia with traditional clinical risk indicators, we developed a predictive model for esophageal post-ESD stenosis. These findings may lead to the utilization of preoperative interventions and individualized treatments for the early identification of patients at high risk for developing esophageal strictures after ESD.
MATERIALS AND METHODS
Study design and patients
This retrospective observational study enrolled 499 individuals who underwent ESD for esophageal mucosal lesions at Zhongda Hospital, Southeast University, between January 2015 and October 2024. This study adhered to the general ethical requirements of the Declaration of Helsinki and was approved by the Ethics Committee of Zhongda Hospital Affiliated with Southeast University (Approval No. 2025ZDSYLL373-P01). Informed consent was waived for all patients because of the retrospective nature of the study.
Inclusion criteria: (1) Age ≥ 18 years; (2) Postoperative pathology-confirmed esophageal high-grade intraepithelial neoplasia or superficial squamous cell carcinoma of the esophagus. The carcinomas were limited to an invasion depth within the upper third of the submucosa (SM1) with no evidence of lymph node metastasis, which met the criteria for ESD; (3) Abdominal CT examinations within 3 months before ESD; and (4) Complete baseline and imaging data.
Exclusion criteria: (1) Pre-existing esophageal stenosis; (2) Advanced esophageal cancer; (3) Severe cardiopulmonary or other organ dysfunction or other malignant tumors; (4) A history of esophageal ESD or surgical procedures; (5) The absence of abdominal CT examinations within 3 months prior to ESD; and (6) Incomplete baseline data.
Participants were stratified by preoperative CT-assessed body composition into sarcopenic and non-sarcopenic cohorts. Baseline data, including clinicopathological characteristics, postoperative complications of ESD, and adverse events, were contrasted between the two groups. The correlation between sarcopenia and postoperative stenosis after esophageal ESD was analyzed, and a risk prediction model for postoperative stenosis was constructed.
Data collection
Clinical data, including age, sex, BMI, chronic conditions, smoking/drinking history, hospitalization duration, postoperative/intraoperative stenosis prophylaxis (including hormone prophylaxis), ESD wound management (including the use of hemostatic clips or spray glue), muscular injury, prognostic nutritional index (PNI), controlling nutritional status (CONUT) score, white blood cell count, hemoglobin concentration, serum albumin concentration, and the neutrophil-to-lymphocyte ratio, among other parameters, were collected. PNI was calculated as serum albumin concentration (g/L) + 5 × total peripheral lymphocyte count (× 109/L), with a PNI < 50 indicating malnutrition risk. The CONUT score was derived from routine hematological indices (including serum albumin concentration, total lymphocyte count, and total cholesterol concentration), and nutritional status was categorized as normal (0-1 point) or malnutrition (2-12 points)[14]. Endoscopic and histopathological traits included lesion position, histological classification, macroscopic type, lesion length, depth of invasion, and circumferential range.
Definitions of terms
Postoperative delayed bleeding was characterized as a clinically significant hemorrhage that developed 24 hours to 28 days after ESD and necessitated therapeutic endoscopy or blood product administration, accompanied by one or more of the following symptoms: (1) Hematemesis and melena; (2) Hemoglobin decrease > 2 g/dL; or (3) Hemodynamic compromise requiring intervention was identified by systolic pressure decrease > 20 mmHg or heart rate elevation > 20 bpm relative to baseline values. Postoperative delayed perforation was defined as sudden signs of peritoneal irritation without intraoperative perforation, as confirmed by radiographic evidence of pneumoperitoneum on abdominal imaging (CT or plain radiography). Esophageal stenosis was defined as an esophageal lumen diameter < 1 cm or a failure in standard endoscope (diameter approximately 1 cm) passage, accompanied by varying degrees of dysphagia[15].
CT image analysis
Studies comparing the correlation between muscle area at various body levels obtained from CT scans and total body skeletal muscle mass have demonstrated that the muscle area obtained at the third lumbar (L3) level is most closely associated with total skeletal muscle mass[16,17]. Therefore, the L3 skeletal muscle index (L3-SMI) represents the most accurate, commonly used, and stable indicator for quantifying muscle mass and defining sarcopenia. Most retrospective studies have used the L3-SMI as the diagnostic indicator for sarcopenia[10,18-22].
Preoperative abdominal CT images were collected and analyzed using Slice-O-Matic software (version 4.3; Tomovision, Montreal, Canada). The skeletal muscle area (SMA), visceral adipose tissue area (VATA), subcutaneous adipose tissue area (SATA), and intermuscular adipose tissue area (IMAT) at the L3 level were identified and quantified by applying the following Hounsfield unit (HU) thresholds: -29 to +150 for the SMA, -150 to -50 for the VATA, and -190 to -30 for both the SATA and the IMAT (Figure 1).
Figure 1 Assessment of body composition by computed tomography at the third lumbar vertebra (L3) level.
A: Axial computed tomography (CT) image showing the segmentation of different tissue compartments: Skeletal muscle area (SMA) [pink, -29 to +150 Hounsfield units (HU)], visceral adipose tissue area (light blue, -150 to -50 HU), subcutaneous adipose tissue area (dark blue, -190 to -30 HU), and intermuscular adipose tissue area (orange, -190 to -30 HU); B: Sagittal CT image indicating the L3 level for axial analysis. The skeletal muscle index was calculated as the SMA divided by the square of the patient’s height (SMA/height2).
Body composition metrics and sarcopenia definition
The L3-SMI was derived as the SMA divided by height squared (cm2/m2), with visceral obesity defined by VFA ≥ 100 cm2. In accordance with Japanese guidelines, the diagnostic thresholds for sarcopenia are 42.0 cm2/m2 for men and 38.0 cm2/m2 for women[23].
Statistical analysis
Normally distributed quantitative measures were characterized as the mean ± SD, with intergroup comparisons evaluated using independent samples t tests. Nonnormally distributed data were defined by median values and interquartile ranges (IQRs) and were analyzed using Mann-Whitney U tests. Categorical measures are reported as frequency distributions [n (%)], with group contrasts evaluated through Pearson χ2, continuity-adjusted χ2, or Fisher exact tests.
Binary logistic regression incorporated univariately significant covariates (P < 0.05) into multivariate modeling, with independent predictors identified through sequential forward selection. Subgroup analyses were stratified by age, sex, circumferential range, invasion depth, and histological classification. Differences between the subgroups are presented via forest plots.
Sensitivity analysis using propensity score matching: Sensitivity analysis was performed using propensity score matching (PSM). Propensity scores were calculated via logistic regression, with covariates including age, sex, BMI, and variables related to postoperative stenosis that have been mentioned in previous studies. This study adopted 1:1 nearest-neighbor matching without replacement and conducted dual analyses with caliper values set at 0.2 and 0.1 SD.
Development and validation of the prediction model: To develop a postoperative stenosis prediction model, the original cohort was allocated to a training subset (70%) and a validation subset (30%) via stratified randomization to ensure comparable baseline characteristics. In the training cohort, variables with univariate significance (P < 0.05) were initially screened and then further filtered via LASSO regression (glmnet package) to eliminate multicollinearity for model construction. The model performance was assessed in both cohorts, with discriminative ability assessed via receiver operating characteristic (ROC) analysis (pROC package), accuracy via a calibration curve (rms package), and clinical utility via decision curve analysis (rmda package). A nomogram was constructed for individualized risk prediction using the rms package.
Statistical significance was defined as P < 0.05 (two-tailed). All analyses were performed with SPSS 27.0 (IBM, United States) and R 4.4.3 (R Foundation, Austria).
RESULTS
Patient and lesion characteristics
In total, 1816 patients who underwent esophageal ESD between January 2015 and October 2024 were initially considered, among whom 324 had incomplete baseline data, 588 lacked abdominal CT scans within 3 months before ESD, and 405 had postoperative pathology inconsistent with the indications for esophageal ESD. Ultimately, we included 499 patients for analysis (Figure 2). According to preoperative CT evaluations, the patient cohort was stratified into a sarcopenia group (n = 141) and a non-sarcopenia group (n = 358).
Figure 2 Flowchart of the study participants.
ESD: Endoscopic submucosal dissection.
Table 1 shows the baseline characteristics of both groups, and Table 2 presents their corresponding lesion features. An average age of 69.00 years (IQR: 64.00-74.00) was recorded for the sarcopenia group, whereas 66.00 years (IQR: 61.00-70.75) was the average age among the non-sarcopenic individuals.
Table 1 Patient characteristics, n (%)/median (interquartile range).
In addition to having a greater proportion of female patients, the sarcopenia group had notably lower values for several parameters, including BMI, body weight, serum albumin concentration, SMA, VATA, and SATA, as well as a lower rate of visceral obesity. With respect to the malnutrition status evaluated via the PNI and CONUT parameters, compared with non-sarcopenia patients, sarcopenia patients had higher malnutrition rates, although the difference was not significant (P = 0.116 for the PNI and P = 0.07 for the CONUT score). The anatomic sites of the esophageal lesions significantly differed between the groups (P = 0.023).
Postoperative complications and follow-up
Table 3 presents postoperative complications and short-term prognoses. A notable finding was the significantly higher incidence of esophageal stenosis in sarcopenic patients (48.23% vs 22.35%). The rates of delayed bleeding and perforation, however, showed no intergroup difference. This contributed to a significantly greater overall burden of adverse events in the sarcopenia cohort (P = 0.006), which also underwent additional endoscopic procedures at a higher rate.
Table 3 Postoperative complications and follow-up, n (%).
Table 4 presents the correlation analysis between CT-assessed body composition indices and postoperative stenosis. Preoperative low SMA and low L3-SMI were found to have a significant link to a higher risk of postoperative stenosis (P < 0.001). Conversely, no significant associations were identified between other body composition parameters and postoperative stenosis.
Table 4 Analysis of body composition indices and stenosis, median (interquartile range).
Predictive factors for post-ESD esophageal stenosis
Binary logistic regression was conducted to determine independent predictors of postoperative stenosis after esophageal ESD (Table 5). The table presents indicators with P < 0.05 in both the univariate and multivariate analyses. The results revealed that sarcopenia, a lesion length ≥ 45 mm, muscular injury, and a circumferential range ≥ 3/4 were prominent independent risk factors for postoperative stricture. Notably, a procedure time ≥ 120 minutes reached statistical significance in the univariate analysis (P < 0.05) but demonstrated a borderline P value approaching 0.05 in the multivariate model.
Table 5 Univariate and multivariate analyses of risk factors for stenosis.
Subgroup analysis of postoperative esophageal stenosis incidence
Subgroup analyses revealed significant associations between sarcopenia and postoperative stenosis in most subgroups (all P < 0.05; Figure 3). Notably, no meaningful associations were detected in the subgroups defined by a circumferential lesion range ≥ 3/4 (P = 0.328) or an operation time ≥ 120 minutes (P = 0.075). Interaction tests across all the subgroups revealed P values > 0.05, thus indicating the consistency of the association between sarcopenia and postoperative stenosis across different demographic and clinical features.
Figure 3 Subgroup analysis of the association between sarcopenia and postoperative stenosis risk.
This forest plot displays the odds ratios (ORs) and 95% confidence intervals for the associations across various patient subgroups. An OR greater than 1 indicates a higher risk of stenosis in the sarcopenia group. The 'Interaction P' value tests whether the effect of sarcopenia on stenosis risk differs significantly between the subgroups. Circumferential: Circumferential resection range; ≤ M2: Invasion depth limited to the lamina propria mucosae or shallower; > M2: Invasion depth beyond the lamina propria mucosae. OR: Odds ratio; BMI: Body mass index; HGA: High-grade adenoma; ESCC: Esophageal squamous cell carcinoma.
PSM
PSM (utilizing 1:1 nearest-neighbor matching under a caliper width of 0.2) successfully matched 166 pairs of patients. After matching, most of the standardized mean differences were < 0.1, thus indicating good intergroup balance (Figure 4 and Supplementary Table 1). Analysis of the matched cohort confirmed that sarcopenia remained a significant and independent predictor of post-ESD stenosis, even after adjusting for confounders [odds ratio (OR): 3.65, 95% confidence interval (CI): 1.61-8.76; Table 6]. A double sensitivity analysis was further conducted with a caliper value of 0.1, yielding 150 well-matched patient pairs (Supplementary material). Multivariate logistic regression analysis yielded consistent results (OR: 3.69, 95%CI: 1.55-9.41), thereby validating the robustness of the findings (Table 6).
Figure 4 Assessment of covariate balance via standardized mean differences.
The plot compares the standardized mean differences (SMDs) of the baseline variables before and after propensity score matching (caliper = 0.2). The results demonstrate that matching effectively improved the balance between the patient groups, as most post-matching SMDs fell below the 01 threshold. SMD: Standardized mean difference; BMI: Body mass index; CONUT: Controlling Nutritional Status score; PNI: Prognostic Nutritional Index; NLR: Neutrophil-to-lymphocyte ratio; WBC: White blood cell count; HGB: Hemoglobin concentration; ALB: Serum albumin concentration.
Table 6 Univariate and multivariate logistic regression analyses after propensity score matching.
Construction and validation of a risk prediction model
The original cohort was randomly allocated to a training set (n = 350) and a validation set (n = 149) in a 7:3 ratio, employing a stratified sampling approach to maintain balance between groups for the development and visualization of the post-ESD stenosis prediction model (Supplementary Table 3). On the basis of the training set data, 15 potential predictors were first filtered through univariate logistic regression (P < 0.05), followed by variable selection with LASSO regression (λ.lse = 0.067, 10-fold cross-validation) and their integration into the multivariate regression analysis (Figure 5). Five key variables were ultimately identified: Sarcopenia (OR: 2.89, 95%CI: 1.55-5.41), a circumferential range ≥ 3/4 (OR: 6.84, 95%CI: 3.50-13.70), a lesion length ≥ 45 mm (OR: 2.37, 95%CI: 1.25-4.47), muscular injury (OR: 4.97, 95%CI: 2.32-10.90), and a procedure time ≥ 120 minutes (OR: 2.42, 95%CI: 1.21-4.88) (Table 7). A nomogram was constructed using the regression coefficients of these variables (Figure 6A).
Figure 5 Variable selection using the least absolute shrinkage and selection operator regression model.
A: The least absolute shrinkage and selection operator coefficient profile plot shows the trajectory of each variable's coefficient as the penalty parameter (lambda) increases. Each colored line represents a variable; B: The 10-fold cross-validation plot for tuning parameter (lambda) selection. The left vertical dashed line indicates the lambda value at which the model achieves minimum binomial deviance (λ.min), whereas the right dashed line represents the most regularized model within one standard error of the minimum (λ.1 se). The final predictors were selected at the λ.1 se value to obtain a more parsimonious model.
Figure 6 Development and validation of a predictive nomogram for post-endoscopic submucosal dissection stenosis.
A: The nomogram is used to calculate an individual patient's risk of postoperative stenosis. For each predictor (including sarcopenia, lesion length, procedure time, and circumferential range), a vertical line is drawn upward to the 'Points' axis to determine the corresponding score. The sum of these scores is located on the 'Total Points' axis, and a line drawn downward to the 'Risk Probability' axis indicates the personalized stenosis risk; B: Receiver operating characteristic curves demonstrate the discriminative ability of the model. The area under the curve was 0.848 (95% confidence interval: 0.802-0.894) in the training cohort and 0.794 (95% confidence interval: 0.708-0.88) in the validation cohort, indicating good and consistent predictive performance. ROC: Receiver operating characteristic; AUC: Area under the curve.
Table 7 Variables selected from the training set data for constructing the risk prediction model.
The ROC curve evaluation revealed excellent model discrimination. The area under the curve (AUC) values were 0.848 (95%CI: 0.80-0.89) for the model derivation group and 0.794 (95%CI: 0.71-0.88) for the internal validation group (Figure 6B). Calibration curve analysis demonstrated good model performance, with slopes close to 1 detected in both the training cohort (Brier score = 0.135; slope = 0.96) and the validation cohort (Brier score = 0.148; slope = 0.98), thereby indicating optimal calibration (Figure 7A and B). The Hosmer-Lemeshow test demonstrated adequate fit in both groups (P = 0.89 for model development; P = 0.79 for model testing). Decision curve analysis confirmed the clinical utility of the model across threshold probability ranges of 12%-79% (training) and 16%-84% (validation) (Figure 7C and D).
Figure 7 Calibration and decision curve analysis of the post-endoscopic submucosal dissection stenosis prediction model.
A and B: Calibration curves plot the predicted probability against the observed frequency of stenosis. The dashed diagonal line represents perfect calibration. The close fit of the solid line (our model) to the diagonal, along with low Brier scores (training: 0.135; validation: 0.148), indicates excellent agreement between predictions and actual outcomes; C and D: Decision curve analysis evaluates the clinical net benefit of the model across various threshold probabilities. The “prediction model” line shows that using the nomogram for clinical decision-making provides greater net benefit than the strategies of “treating all” or “treating none” of the patients across a wide range of risk thresholds do, demonstrating its potential clinical utility.
DISCUSSION
To the best of our knowledge, this study represents an inaugural investigation to examine the relationship between CT-defined sarcopenia and complications following esophageal ESD. Moreover, our analysis revealed a strong association between sarcopenia and the development of post-ESD stenosis, indicating an elevated risk for this complication. In our research, the overall occurrence of postoperative stenosis after esophageal ESD was 29.7% (148/499); moreover, among lesions with a circumferential range ≥ 3/4, the incidence of postoperative stenosis was 73.4% (80/109), which aligns with the findings of previous studies[24,25].
Sarcopenia is widely recognized as being associated with poor prognosis of various advanced malignancies[21,26-28]. Recent studies have also revealed that sarcopenia increases the risk of postoperative complications after endoscopic treatment for early-stage gastrointestinal tumors. Systematic evidence synthesis demonstrated that sarcopenia was an independent risk factor for clinically significant post-ESD complications, particularly pneumonia development, in early-stage gastric cancer cohorts[13]. A retrospective study involving 497 patients revealed that sarcopenia independently predicted post-ESD complications in older adults with early-stage colorectal cancer, although overall survival remained unaffected[29]. However, these studies on sarcopenia have predominantly focused on elderly patients. Reports have indicated that hospitalized patients younger than 65 years have significantly higher health care expenditures than their older counterparts (≥ 65 years) do[30]; moreover, more than 10% of young adults have sarcopenia[31]. Our study revealed that 19.4% (37/191) of patients younger than 65 years old had sarcopenia, which represents a high prevalence that is potentially linked to eating disturbances caused by their esophageal disorders. Therefore, the nutritional status of this patient subgroup should not be overlooked.
Correlation analysis between body composition and postoperative stenosis revealed that patients with postoperative stenosis exhibited lower muscle mass. Multivariate logistic regression analysis further indicated that sarcopenia was a standalone predictor of risk for postoperative stenosis after esophageal ESD, along with a lesion circumferential range ≥ 3/4, a lesion length ≥ 45 mm, and muscular injury, with these findings aligning with those of earlier research[32-34]. The results from PSM (controlling for intergroup covariate imbalance) also revealed that sarcopenia increased the risk of postoperative stenosis. One plausible explanation for this result is that sarcopenia is associated with a systemic inflammatory status mediated by inflammatory mediators and a redox imbalance[35]. Oxidative stress promotes protein degradation and the production of proinflammatory cytokines (including interleukin-6, C-reactive protein, and tumor necrosis factor-α) by regulating transcription factors (such as nuclear factor κB)[36]. Yang et al[37] reported that chronic inflammation in esophageal diseases can lead to progressive or excessive fibrosis, with extracellular matrix over deposition ultimately causing stenosis. Additionally, sarcopenic patients may exhibit insufficient protein synthesis, thereby delaying mucosal wound healing, promoting fibrosis, and increasing stenosis risk, which is supported by our finding that sarcopenia patients demonstrated lower albumin levels (P < 0.05).
Subgroup analyses demonstrated that sarcopenia remained a significant risk factor for postoperative stenosis across all of the stratified subgroups, which was consistent with the results of the primary analysis. No notable difference was observed among the groups in the subgroup with a lesion circumferential range ≥ 3/4, which was possibly due to the strong independent association of extensive circumferential involvement with postoperative stenosis. Similarly, no significant associations were observed in the subgroup with a procedure time of ≥ 120 minutes, which might have stemmed from the limited sample size (93/499, 18.6%) and thus led to inadequate statistical power. Notably, all of the subgroup interaction P values were > 0.05, thus indicating that stratified factors did not affect the intensity of the link between sarcopenia and postoperative stenosis. These findings highlight the need to monitor muscle status in patients aged < 65 years with a lesion circumferential range < 3/4 or a procedure time < 120 minutes, representing an easily overlooked population in clinical practice.
According to the Global Leadership Initiative on Malnutrition consensus, reduced skeletal muscle mass is established as a key phenotypic diagnostic criterion for malnutrition[38]. As an emerging nutritional assessment metric, the L3-SMI is acknowledged as being the most prevalently used and robust indicator for quantifying muscle mass and defining sarcopenia. It has gained broad acceptance in the prognostic stratification of chronic hepatic disorders and diverse oncologic malignancies[19,22,39]. However, prior predictive models for postoperative stenosis following esophageal ESD have focused primarily on intraoperative lesion attributes, thereby neglecting patients' nutritional status, including muscle mass parameters[5,40,41]. A growing body of evidence links sarcopenia-related muscle loss to poor overall survival following esophageal cancer surgery[42,43]. In line with this, our study identifies sarcopenia as a significant independent risk factor of stenosis following ESD among individuals diagnosed with early-stage esophageal cancer. Accordingly, we identified 5 key factors associated with post-ESD stenosis through robust analyses: Sarcopenia, a lesion circumferential range ≥ 3/4, a lesion length ≥ 45 mm, muscular injury, and a procedure time ≥ 120 minutes. To facilitate risk prediction, we created a model for post-ESD esophageal stenosis using these variables, along with a corresponding nomogram. This model integrates markers of nutritional status, lesion features, and surgical details, thereby addressing gaps in previous studies. We evaluated the predictive performance of this model via ROC curve analysis, the Hosmer-Lemeshow test, calibration curve analysis, and decision curve analysis, all of which demonstrated good risk prediction ability.
Limitations
Several study limitations should be acknowledged when these findings are interpreted. First, its nonrandomized, retrospective, single-center design inevitably introduces risks of selection bias and unmeasured confounding. Although we employed PSM to mitigate the impact of observed variables, residual confounding from factors such as nutritional status (e.g., albumin and prealbumin levels), operator experience, and detailed medication history cannot be fully excluded. Second, there are no standardized criteria for the normal reference range of the L3-SMI that was used to assess sarcopenia, which, combined with potential variations in image quality and segmentation, could introduce measurement bias and affect the generalizability of our sarcopenia assessment. Moreover, owing to its retrospective design, this study evaluated only muscle mass rather than muscle quality (e.g., grip strength). Third, the robustness of the internal validation of the model requires confirmation through external testing. Consequently, future multicenter prospective studies with expanded cohorts are needed to standardize the definitions of sarcopenia, incorporate muscle function parameters, and ultimately validate and refine this predictive tool.
CONCLUSION
In summary, this study revealed that CT-quantified sarcopenia was significantly correlated with post-ESD stenosis and identified it as an independent predictor of this complication. A risk prediction model incorporating sarcopenia for post-ESD stenosis was constructed and validated, which demonstrated good predictive efficacy.
ACKNOWLEDGEMENTS
We would like to thank Ms. Jin-Fang Sun from Southeast University for her valuable assistance with the biostatistical analysis in this study.
Footnotes
Provenance and peer review: Unsolicited article; Externally peer reviewed.
Peer-review model: Single blind
Specialty type: Gastroenterology and hepatology
Country of origin: China
Peer-review report’s classification
Scientific Quality: Grade A, Grade B
Novelty: Grade B, Grade B
Creativity or Innovation: Grade B, Grade C
Scientific Significance: Grade A, Grade B
P-Reviewer: Shahid Y, MD, Postdoctoral Fellow, Pakistan; Sonmez G, MD, PhD, Türkiye S-Editor: Li L L-Editor: A P-Editor: Wang CH
Pimentel-Nunes P, Libânio D, Bastiaansen BAJ, Bhandari P, Bisschops R, Bourke MJ, Esposito G, Lemmers A, Maselli R, Messmann H, Pech O, Pioche M, Vieth M, Weusten BLAM, van Hooft JE, Deprez PH, Dinis-Ribeiro M. Endoscopic submucosal dissection for superficial gastrointestinal lesions: European Society of Gastrointestinal Endoscopy (ESGE) Guideline - Update 2022.Endoscopy. 2022;54:591-622.
[RCA] [PubMed] [DOI] [Full Text][Cited by in Crossref: 410][Cited by in RCA: 460][Article Influence: 115.0][Reference Citation Analysis (0)]
Kitagawa Y, Ishihara R, Ishikawa H, Ito Y, Oyama T, Oyama T, Kato K, Kato H, Kawakubo H, Kawachi H, Kuribayashi S, Kono K, Kojima T, Takeuchi H, Tsushima T, Toh Y, Nemoto K, Booka E, Makino T, Matsuda S, Matsubara H, Mano M, Minashi K, Miyazaki T, Muto M, Yamaji T, Yamatsuji T, Yoshida M. Esophageal cancer practice guidelines 2022 edited by the Japan esophageal society: part 1.Esophagus. 2023;20:343-372.
[RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)][Cited by in RCA: 268][Reference Citation Analysis (0)]
Kirk B, Cawthon PM, Arai H, Ávila-Funes JA, Barazzoni R, Bhasin S, Binder EF, Bruyere O, Cederholm T, Chen LK, Cooper C, Duque G, Fielding RA, Guralnik J, Kiel DP, Landi F, Reginster JY, Sayer AA, Visser M, von Haehling S, Woo J, Cruz-Jentoft AJ; Global Leadership Initiative in Sarcopenia (GLIS) group. The Conceptual Definition of Sarcopenia: Delphi Consensus from the Global Leadership Initiative in Sarcopenia (GLIS).Age Ageing. 2024;53:afae052.
[RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)][Cited by in Crossref: 324][Cited by in RCA: 273][Article Influence: 136.5][Reference Citation Analysis (0)]
Hisada H, Tsuji Y, Obata M, Cho R, Nagao S, Miura Y, Mizutani H, Ohki D, Yakabi S, Takahashi Y, Sakaguchi Y, Kakushima N, Yamamichi N, Fujishiro M. The impact of sarcopenia on short- and long-term outcomes of endoscopic submucosal dissection for early gastric cancer.J Gastroenterol. 2022;57:952-961.
[RCA] [PubMed] [DOI] [Full Text][Cited by in Crossref: 15][Cited by in RCA: 20][Article Influence: 5.0][Reference Citation Analysis (0)]
Wang C, Tang M, Chen D, Zhou Y, Liang G, Shen R, Chen T. Enhancing Patient Outcomes: A Novel Nomogram Prediction Model Based on Systemic Immune-Inflammation Index for Esophageal Stricture After Endoscopic Submucosal Dissection.Cancer Med. 2024;13:e70264.
[RCA] [PubMed] [DOI] [Full Text][Cited by in RCA: 3][Reference Citation Analysis (0)]
Hacker UT, Hasenclever D, Baber R, Linder N, Busse H, Obermannova R, Zdrazilova-Dubska L, Valik D, Lordick F. Modified Glasgow prognostic score (mGPS) is correlated with sarcopenia and dominates the prognostic role of baseline body composition parameters in advanced gastric and esophagogastric junction cancer patients undergoing first-line treatment from the phase III EXPAND trial.Ann Oncol. 2022;33:685-692.
[RCA] [PubMed] [DOI] [Full Text][Cited by in Crossref: 4][Cited by in RCA: 53][Article Influence: 13.3][Reference Citation Analysis (0)]
Geng ZH, Zhu Y, Li QL, Fu PY, Xiang AY, Pan HT, Xu MD, Chen SY, Zhong YS, Zhang YQ, Ma LL, Hu JW, Cai MY, Qin WZ, Chen WF, Zhou PH. Muscular injury as an independent risk factor for esophageal stenosis after endoscopic submucosal dissection of esophageal squamous cell cancer.Gastrointest Endosc. 2023;98:534-542.e7.
[RCA] [PubMed] [DOI] [Full Text][Cited by in RCA: 16][Reference Citation Analysis (0)]
Jensen GL, Cederholm T, Correia MITD, Gonzalez MC, Fukushima R, Pisprasert V, Blaauw R, Braz DC, Carrasco F, Cruz Jentoft AJ, Cuerda C, Evans DC, Fuchs-Tarlovsky V, Gramlich L, Shi HP, Hasse JM, Hiesmayr M, Hiki N, Jager-Wittenaar H, Jahit S, Jáquez A, Keller H, Klek S, Malone A, Mogensen KM, Mori N, Mundi M, Muscaritoli M, Ng D, Nyulasi I, Pirlich M, Schneider S, de van der Schueren M, Siltharm S, Singer P, Steiber A, Tappenden KA, Yu J, van Gossum A, Wang JY, Winkler MF, Compher C, Barazzoni R. GLIM consensus approach to diagnosis of malnutrition: A 5-year update.JPEN J Parenter Enteral Nutr. 2025;49:414-427.
[RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)][Cited by in RCA: 15][Reference Citation Analysis (0)]
Di Cola S, D'Amico G, Caraceni P, Schepis F, Loredana S, Lampertico P, Toniutto P, Martini S, Maimone S, Colecchia A, Svegliati Barone G, Alessandria C, Aghemo A, Crocè SL, Adinolfi LE, Rendina M, Lapenna L, Pompili E, Zaccherini G, Saltini D, Iavarone M, Tosetti G, Martelletti C, Nassisi V, Ferrarese A, Giovo I, Masetti C, Pugliese N, Campigotto M, Nevola R, Merli M; EpatoSarco working group. Myosteatosis is closely associated with sarcopenia and significantly worse outcomes in patients with cirrhosis.J Hepatol. 2024;81:641-650.
[RCA] [PubMed] [DOI] [Full Text][Cited by in Crossref: 19][Cited by in RCA: 33][Article Influence: 16.5][Reference Citation Analysis (0)]
