Method for predicting risk of developing intrathoracic anastomotic leak after subtotal esophagectomy by ivor lewis

Abstract

FIELD: thoracic surgery.SUBSTANCE: used to predict the development of cervical anastomotic failure after esophagectomy. Before surgery, the tumor location and the monocytic-lymphocyte ratio (MLR) are determined; on the first day of the postoperative period, the platelet-lymphocyte ratio (PLR) is determined. Then, the position of each predictor is determined on the corresponding horizontal axis of the nomogram in Figure 2, with the data on tumor localization corresponding to 0 points – cancer of the lower third of the esophagus, 12.5 points – cancer of the middle third of the esophagus. The obtained points are summed up, the obtained value is superimposed on the horizontal axis of the “predicted probability” nomogram, and the risk of developing intrathoracic anastomosis failure is determined based on the sum of the points. As the score increases, the risk of intrathoracic anastomosis failure increases.EFFECT: method improves the accuracy of predicting the risk of developing intrathoracic anastomotic leakage in patients with esophageal cancer, which ensures timely prevention of complications and a reduction in hospitalization time due to a comprehensive assessment of tumor localization, MLR, and PLR using a simple nomogram.1 cl, 4 dwg, 2 ex

Description

[0001] The invention relates to medicine, namely to thoracic surgery, and is used to predict the failure of an intrathoracic anastomosis after subtotal resection of the esophagus according to Ivor Lewis.

[0002] In recent years, significant progress has been achieved in the treatment of esophageal cancer due to the introduction of trimodal treatment into routine practice, the effectiveness of which was proven by the study of Shapiro J in 2012 (Shapiro J, van Lanschot JJB, Hulshof MCCM, et al. "Neoadjuvant chemoradiotherapy plus surgery versus surgery alone for esophageal or junctional cancer (CROSS trial)." N Engl J Med. 2012; 366 (22): 2074-2084. DOI: 10.1056 / NEJMoal 112088). Currently, surgical treatment continues to occupy a central place in the multimodal treatment of esophageal cancer, especially when the tumor is localized in the middle and lower third.

[0003] Among radical interventions, the Ivor Lewis operation (thoracoabdominal esophagectomy with the formation of an intrapleural anastomosis) remains the preferred surgical method in most large centers, combining oncological radicalism with a satisfactory functional component.

[0004] However, anastomotic leakage remains a significant problem with the Ivor Lewis procedure. Current data indicate that the incidence of intrapleural anastomotic leakage after Ivor Lewis subtotal esophagectomy remains high. According to a systematic review and meta-analysis by Bras Harriott et al., which included 130 studies and 16,053 patients, the incidence of anastomotic leakage was 6% after open esophagectomy, 8% after hybrid esophagectomy, and 8% after total minimally invasive esophagectomy. The overall postoperative mortality rates for these approaches were 4%, 4%, and 2%, respectively (Bras Harriott C, Angeramo CA, Casas MA, Schlottmann F. Open versus hybrid versus totally minimally invasive Ivor Lewis esophagectomy: Systematic review and meta-analysis. J Thorac Cardiovasc Surg. 2022 Dec;164(6):e 233-e254. doi: 10.1016 / j.jtcvs.2O21.12.051. Epub 2022 Jan 18. PMID: 35164948. https: / / pubmed.ncbi.nlm.nih.gov / 35164948 / ).

[0005] Anastomotic leakage directly leads to the development of pleural empyema and mediastinitis, and indirectly leads to pneumonia and sepsis, especially if treatment is ineffective. All of this leads to increased intensive care and hospital stays, significant treatment costs, increased postoperative mortality, and decreased quality of life.

[0006] The current literature emphasizes the need to identify reliable risk factors for anastomotic leakage, allowing for preoperative patient stratification and the implementation of individualized preventive measures (Kamarajah SK, Lin A, Tharmaraja T, Bharwada Y, Bundred JK, Nepogodiev D, Evans RPT, Singh P, Griffiths EA. Risk factors and outcomes associated with anastomotic leaks following esophagectomy: a systematic review and meta-analysis. Dis Esophagus. 2020 Mar 16;33(3):doz089. doi: 10.1093 / dote / doz089. PMID: 31957798. https: / / pubmed.ncbi.nlm.nih.gov / 31957798 / and Azer M, Miftode S, Bockhom M, El-Sourani N. Evaluation of the use of inflammatory biomarkers in the early detection of anastomotic leakage after esophagectomy: A retrospective analysis. Surg Open Sci. 2022 May 20; 10:12-18. doi: 10.1016 / j.sopen.2022.05.007. PMID: 35800711; PMCID: PMC9253454.

[0007] https: / / pubmed.ncbi.nlm.nih.gOv / 31957798 / ).

[0008] One of the elements of risk stratification is the development and implementation of predictive models into clinical practice, thanks to which patient data is analyzed and risk groups are identified for one or another

[0009] other unfavorable postoperative outcome. For more convenient use of the prognostic model, it is advisable to construct a nomogram.

[0010] A nomogram is a graphical representation of a developed algorithm based on statistically significant predictors, which are presented as continuous variables, and is used to predict the result (outcome) for which it was created.

[0011] There are studies aimed at identifying risk factors and predicting postoperative complications after transthoracic esophagectomy for cancer. In particular, a large multicenter study by Kemeter M. et al (Predictive factors for pneumonia and anastomotic leak after transthoracic esophagectomy / / Diseases of the Esophagus. - 2024. - Supplement 1. - Article: doae057.199.) analyzed prognostic factors for the two most formidable complications - pneumonia and anastomotic leak - in a cohort of 1355 patients. This study identified significant risk factors for anastomotic leak: a history of cardiopulmonary diseases, conversion to thoracotomy, and operative duration. However, the authors note that their attempt to predict both complications and in-hospital mortality using modern machine learning methods was unsuccessful.

[0012] A well-known work by Stam et al. (Stam WT Complications in Gastro-Intestinal Surgery: From Clinical Factors to Al-Driven Management. - VU University Amsterdam, 2024) proposes an approach to predicting complications after gastrointestinal surgery, including anastomotic leakage. The authors used machine learning methods (XGBoost, Support Vector Machines - SVM) to create an integrated prognostic model. The model combines heterogeneous clinical data and biomarkers (including inflammatory and possibly perfusion indicators). However, this model has significant drawbacks:

[0013] 1. Heterogeneity of the training set: The model was developed on a mixed population of patients who underwent various gastrointestinal surgeries, and was not specifically designed for intrathoracic anastomosis after Ivor Lewis esophagectomy. This significantly limits its accuracy and applicability to the target procedure.

[0014] 2. Black-box problem: Machine learning algorithms (XGBoost, SVM) are difficult for clinicians to interpret, as they make it difficult to understand the contribution of specific factors (such as tumor location) to prognosis, thereby reducing the model's credibility in clinical practice.

[0015] 3. Need for validation: Like previous models, it requires rigorous external validation in a specialized cohort of patients after Ivor Lewis esophagectomy.

[0016] Thus, these models are not applicable for reliable prediction of intrathoracic anastomotic failure after Ivor Lewis surgery and subsequent preventive measures in the pre- and intraoperative periods.

[0017] The technical result of the proposed method is the creation of an effective and easy-to-understand nomogram that has high specificity and sensitivity.

[0018] The specified technical and therapeutic results in the implementation of the invention are achieved due to the fact that statistically significant parameters are determined.

[0019] The essence of the claimed invention is as follows.

[0020] A method for predicting the risk of developing intrathoracic anastomotic leak after subtotal esophagectomy according to Ivor Lewis, which includes determining the tumor location and the monocytic-lymphocyte index (MLR) in the patient before surgery, and the platelet-lymphocyte index (PLR) on the first day of the postoperative period, after which the position of each indicator on the corresponding horizontal axis of the nomogram is determined, the corresponding number of points is assigned to each indicator, the obtained points are summed up, and the predicted probability of developing intrathoracic anastomotic leak is determined based on the total score, with the data on tumor location corresponding to 0 points - cancer of the lower third of the esophagus and 12.5 points - cancer of the middle third of the esophagus, while with an increase in the sum of points, the probability of developing intrathoracic anastomotic leak increases.

[0021] The invention is explained by a detailed description and illustrations, which depict:

[0022] Fig. 1 - Calibration curve of the predictive model during internal validation.

[0023] Fig. 2 - Nomogram for predicting anastomotic leakage after the Ivor Lewis operation.

[0024] Fig. 3 - clinical example of assessing the probability of failure of the intrathoracic anastomosis after subtotal resection of the esophagus No. 1.

[0025] Fig. 4 - clinical example of assessing the probability of failure of the intrathoracic anastomosis after subtotal resection of the esophagus No. 2.

[0026] The method is carried out as follows.

[0027] Postoperative data are analyzed. A score is then calculated, the expected incidence of cervical anastomotic leak is determined, and appropriate measures are taken to minimize the possibility of this complication occurring preoperatively.

[0028] Clinical example 1.

[0029] Patient S., 65 years old, was admitted to the surgical department of thoracic oncology for surgical treatment. Diagnosis on admission: lower thoracic esophageal cancer, stage CT3N0M0 Pet. In the preoperative period, the values ​​of the monocytic-lymphocyte index were calculated - 36 points, the platelet-lymphocyte index on the first day of the postoperative period - 56 points. The total score was 92 points. According to the nomogram (Fig. 3), the total score of less than 124 corresponds to a low risk of developing intrathoracic anastomotic leak. The patient underwent subtotal esophageal resection according to Ivor Lewis. In the postoperative period, daily clinical observation, laboratory monitoring, monitoring of the level of inflammatory markers, computed tomography of the chest organs on the 6th day and radiography of the esophagus with water-soluble contrast on the 9th day after surgery were carried out.The examination revealed no signs of intrathoracic anastomotic leak. The patient was discharged on the 12th day in satisfactory condition. The absence of complications confirms the accuracy of the previously calculated prognosis.

[0030] Clinical example 2.

[0031] Patient K., 46 years old, was hospitalized in the surgical department of thoracic oncology with a diagnosis of mid-thoracic esophageal cancer, stage CT2N0M0 Pet. In the preoperative period, the values ​​​​of the monocyte-lymphocyte index were calculated - 78 points,

[0032] platelet-lymphocyte index on the first postoperative day was 38 points. The total score was 128 points. According to the nomogram (Fig. 4), a value of more than 124 points corresponds to a high risk of developing cervical anastomotic leakage. The patient underwent subtotal esophageal resection according to Ivor Lewis. On the 10th day, the patient developed vomiting with the release of approximately 50 ml of turbid fluid with a pungent odor. Endoscopic examination revealed anastomotic leakage with the formation of a paraanastomotic "cavity". Multispiral computed tomography with intravenous and oral contrast enhancement was performed on the 10th postoperative day, revealing leakage of contrast agent into the pleural cavity. A failure of the intrathoracic anastomosis was diagnosed, which confirmed the high risk calculated using the stated method.

[0033] Using the proposed method in clinical practice allows for the construction of an effective and easy-to-understand nomogram with high specificity and sensitivity (79% and 82%, respectively). This method enables accurate prediction of complications and intrathoracic anastomotic leaks, optimizes pre- and intraoperative anastomotic leak prevention, thereby reducing patient hospital stays and the burden on the healthcare system by lowering treatment costs.

Claims

A method for predicting the risk of developing intrathoracic anastomotic leak after subtotal esophagectomy according to Ivor Lewis, which includes determining pre- and postoperative patient data with subsequent risk calculation using a nomogram, characterized in that the tumor localization and monocytic-lymphocyte index (MLR) are determined before the operation, the platelet-lymphocyte index (PLR) is determined on the first day of the postoperative period; then the position of each predictor on the corresponding horizontal axis of the nomogram in Figure 2 is determined, wherein the data on tumor localization correspond to 0 points - cancer of the lower third of the esophagus, 12.5 points - cancer of the middle third of the esophagus; the obtained points are summed up, the obtained value is superimposed on the horizontal axis of the nomogram "predicted probability" and the risk of developing intrathoracic anastomotic leak is determined based on the sum of points, wherein the risk of intrathoracic anastomotic leak increases with an increase in the sum of points.

Images