Predicting complication risks after sleeve lobectomy for non-small cell lung cancer

Introduction

For centrally located non-small cell lung cancer (NSCLC), sleeve lobectomy is a well-established and preferred alternative to pneumonectomy (1). This operative technique not only facilitates the preservation of lung function in patients but also ensures the complete oncologic resection of tumors (2). Studies have highlighted the favorable outcomes of sleeve resection, including improved overall survival, recurrence rate, and disease-free survival (1,3,4). Sleeve lobectomy can be a technically complex procedure with notable morbidity and mortality. According to previous publications, mortality rates after sleeve lobectomy can be as high as 3%, with morbidity ranging from 21% to 50% (1,2,4-11). Complications may incur a negative impact on short-term outcomes after lung surgery, including prolonged hospitalization, reduced quality of life, and increased readmissions, leading to increased healthcare costs and worsened long-term survival (12-17).

To better understand complication risks, predictive models have been constructed following pneumonectomy, some of which have the potential to support surgical decision-making (18,19). However, to date, there are no effective tools for sleeve lobectomy to estimate anticipated risk of complication. Thus, we sought to develop models to predict the risk of complications after sleeve lobectomy, aiming to support health care teams in making surgical decisions and implementing optimized perioperative management. We present this article in accordance with the TRIPOD reporting checklist (available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-24-325/rc).

Methods

Patients

The records of consecutive patients who underwent sleeve lobectomy for centrally located NSCLC with complete data between July 2016 and December 2019 in Shanghai Pulmonary Hospital were reviewed and retrospectively analyzed. Patients who underwent sleeve lobectomy for small cell lung cancer and metastasis were excluded (Figure 1). All patients received full preoperative examinations, including flexible bronchoscopy, chest X-ray, pulmonary function tests, computed tomography (CT) scan/contrast-enhanced CT, abdominal/brain CT scan, bone scan, and related laboratory examinations. Positron emission tomography (PET) and endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA) were conducted upon suspicion of mediastinal involvement on CT scans. Characteristics of the patients such as comorbidities, demographics, lab results, and pulmonary function test data were obtained from electronic medical records.

Figure 1 Flowchart showing which patients were included and excluded. NSCLC, non-small cell lung cancer.

Postoperative complications were tabulated individually, and patients were classified into two groups based on whether they had developed complications or not. For this study, complications (prolonged air leak, pneumonia, cardiac arrhythmia, postoperative blood transfusion, pulmonary embolism, chylothorax, bronchopleural fistula, pyothorax, hemothorax, heart failure, respiratory failure, and death within 30 days after surgery) were classified using the Clavien-Dindo Classification system (20). Major complications were defined as grade III-V and minor complications were defined as grade I–II. All tumors were reclassified based on the eighth edition of the TNM staging system for lung cancer. (21). Comorbidities were evaluated using the Charlson Comorbidity Index (CCI) (22). A prolonged air leak was considered to be one lasting over five days postsurgery. Follow-up was conducted through telephone calls or outpatient visits, comprising interval medical history, physical examination, and enhanced contrast whole-body CT scans every six months for the initial three years, followed by annual assessments. Overall survival (OS) was measured from the surgery date until death from any cause, or until the final follow-up date. Recurrence-free survival (RFS) was calculated from the surgical resection date to disease progression (relapse or metastasis), death from any cause, or the final follow-up.

The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013). The study was approved by the institutional review board of Shanghai Pulmonary Hospital (No. K22-793; 15 December 2022) and individual consent for this retrospective analysis was waived.

Model development

The models were crafted utilizing the logistic regression method, ensuring that all variables from the structured dataset were incorporated for the multivariable regression models, aimed at predicting the risks of overall, major, or specific minor complications. The variable inclusion was grounded on a thorough univariable analysis conducted between groups, where variables with a significance level of P<0.10 were selected. Furthermore, the derivation cohort was systematically divided into two subsets: a training cohort comprising 60% of the data, dedicated to training the predictive models, and a validation cohort with 40% of the data, used to rigorously test the developed models on unseen data, thereby validating their accuracy and reliability.

Feature weight analysis

To determine the major predictors for complication risks in the cohort, we measured the importance of each feature using coefficients derived from the predictive model designed for predicting complication risks. Scaled importance was calculated as the ratio of the relative importance of each feature to the highest relative importance among all features. The feature’s importance weight in the overall complication risk prediction was determined by the ratio of its relative importance to the sum of the relative importance of all features.

Statistical analysis

The discriminative ability of the models for the evaluation of complication risks in centrally located NSCLC was quantified using the area under the receiver operating characteristic (ROC) curve (AUC).

Categorical data are presented as frequency (percentage) and was analyzed using Chi-square analysis. Continuous data with a normal distribution are shown as mean ± standardized deviation and evaluated using the Student’s t-test. For skewed distribution, continuous data are summarized by median and interquartile range (IQR) and analyzed with the Mann-Whitney U test. Survival analyses (OS and RFS) among groups without complications, with minor complications, and with major complications were assessed using the Kaplan-Meier method along with the log-rank test. All statistical analyses and model construction were performed using IBM SPSS 20.0 (IBM Corporation, Armonk, NY, USA) and R version 3.2.4 (https://www.rstudio.com/products/rstudio/).

Results

Patient characteristics

The clinical and therapeutic characteristics of patients are shown in Table 1. A total of 691 patients were included. The median age was 63 (range, 25–83) years, and 89.7% (n=620) of the patients were male. Five hundred and seventeen patients (74.8%) had a smoking history, and 81 patients (11.7%) underwent neoadjuvant therapy (typically with at least 2 cycles of platinum-based 2-drug regimens). Procedures for 150 patients (21.7%) necessitated pulmonary artery reconstruction, while 319 patients (46.2%) underwent video-assisted thoracic surgery (VATS). The median operation time and estimated intraoperative blood loss were 200 (IQR, 165–245) minutes and 100 (IQR, 100–200) mL, respectively. Complications developed among 232 patients (33.6%), including 35 patients (5.1%) suffering major complications and 197 (28.5%) with minor complications. Regarding pathologic stage, 27 patients (3.9%) were classified as stage IA, 245 patients (35.5%) as stage IB, 46 patients (6.7%) as stage IIA, 156 patients (22.6%) as stage IIB, 175 (25.3%) as stage IIIA, and 42 (6.1%) as stage IIIB. Surgical pathologic histology revealed 521 patients (75.4%) with squamous cell carcinoma, 134 (19.4%) with adenocarcinoma, and 36 (5.2%) with other NSCLCs.

Complications and classifications

Postoperative complications and their classification are listed in Figure 2 and Table S1. The overall morbidity was 33.6%. The most frequent complications in this cohort were postoperative blood transfusion (n=130, 18.8%), followed by pneumonia (n=57, 8.3%). Among patients who experienced these two common complications, 14 patients (14/130, 10.8%) with postoperative blood transfusion and six patients (6/57, 10.5%) with pneumonia also developed major complications. Prolonged air leak (n=48, 7.0%) was also a notable adverse event, after which 12.5% (n=6) of patients developed major complications. Five patients in the major complication group experienced bronchopleural fistula (n=5, 0.7%). Additionally, cardiac arrhythmia (n=21, 3.0%), chylothorax (n=8, 1.2%), hemothorax (n=7, 1.0%), respiratory failure (n=6, 0.9%), pyothorax (n=5, 0.7%), and heart failure (n=5, 0.7%) were recorded individually (Table S1).

Figure 2 Classification of complications after sleeve lobectomy. The numerator is the number of specific complications, the denominator is the number of patients with minor complications (n=232, 100% of the patients with complications) or the number of patients with major complications (n=35, 15% of the patients with complications).

Risk factors of complications

Table 1 shows that several variables could serve as predictors for complications after sleeve lobectomy. In the complication group (n=232, 33.6%), patients were predominantly male (93.1% vs. 88.0%, P=0.04) with a higher CCI score (P=0.007). Meanwhile, lower body mass index (BMI) scores (P=0.045), pulmonary artery reconstruction (P<0.001), thoracotomy (P<0.001), longer operation time (P<0.001), more estimated blood loss (P<0.001), and intraoperative blood transfusion (P<0.001) were also significantly associated to postoperative complications.

Performance of the complication prediction models

Four preoperative variables [sex, BMI, CCI, the first second of forced expiratory volume FEV1% of predicted (FEV1%pre)] and five surgical-related variables (pulmonary artery reconstruction, surgical approach, operation time, estimated blood loss, intraoperative blood transfusion) were identified and included in the model predicting overall complication risks. The model predicting major complication risks selected three variables (intraoperative blood transfusion, number of lymph node stations dissected, and CCI). In the validation set, the models achieved an AUC of 0.853 (95% CI: 0.705, 0.885) for predicting overall complications and an AUC of 0.751 (95% CI: 0.727, 0.762) for predicting major complications (Figure 3A,3B). Intraoperative blood transfusion possessed the highest feature importance weight (72.57%) in the model predicting complications (Table S2). The included variables in each minor complication predictive model and their relative importance are shown in Table S2. The models achieved AUCs of 0.78 (95% CI: 0.77, 0.80) for pneumonia, 0.79 (95% CI: 0.73, 0.79) for prolonged air leak, and 0.89 (95% CI: 0.81, 0.89) for predicting cardiac arrhythmia (Figure S1A-S1C).

Figure 3 Predictive performance of the logistic regression models. Receiver operating characteristic curves of the models predicting overall complications (A), and major complications (B) in the validation cohorts. AUC, area under the curve; CI, confidence interval.

Impact of postoperative complications on prognosis

Median follow-up time was 29 months. The 3-year OS and 3-year RFS were 76.3% and 66.4% respectively. As shown in Figure 4A, different complication levels could significantly stratify patients’ prognoses. Patients with minor complications or major complications presented a significantly decreased tendency in 3-year OS, while the major complication group presented the poorest prognosis. Similar to OS, patients with complications showed a tendency for decreased 3-year RFS (Figure 4B). In the multivariable analysis (Tables 2,3), the presence of postoperative complications was independently associated with adverse RFS [hazard ratio (HR) 1.476; 95% CI: 1.141–3.381; P=0.003] and OS (HR 1.548; 95% CI: 1.149–2.084; P=0.004). The long-term analysis of competing risk showed similar results, indicating that major complications were significantly associated with death due to specific causes (Figure S2).

Figure 4 The relationship between complications and prognosis. Overall survival (A) and recurrence-free survival (B) curves of different complication groups. The green curve refers to the no complication subgroup, the blue curve represents to the minor complication subgroup, and the red curve indicates to the major complication subgroup.

Discussion

Sleeve lobectomy is a widely utilized technique for the resection of centrally located NSCLC (1,2,4). Given the increased complication risks resulting from these procedures, patients who undergo sleeve lobectomy benefit from individualized patient decision-making and healthcare management strategies. In this study, we identified risk factors for complications related to sleeve lobectomy through the presently largest procedure-specific database. Intraoperative blood transfusion, pulmonary artery reconstruction, high CCI, low BMI, thoracotomy, sex male, low FEV1%pre, estimated blood loss, and long operation time were found to be predictors of postoperative complications after sleeve lobectomy.

By modeling these factors, we developed models aimed at predicting both the probability and severity of postoperative complications. These models exhibited robust performance, boasting an AUC of 0.85 for overall complications and 0.75 for major complications respectively, thus demonstrating their efficacy in clinical prognostication.

In general, mortality and morbidity rates of sleeve lobectomy ranged between 0–3% and 21–50%, respectively (1,2,5-11). In our study, the 30-day mortality and complication rates of sleeve lobectomy were 0.4% and 34%, respectively. Minor complications such as pneumonia, and prolonged air leaks could affect the length of stay and patient readmission (16,23). The most prevalent minor complication in our study was postoperative blood transfusion, with 17% morbidity. Major complications, especially anastomotic complications such as bronchopleural fistulas, could seriously affect patients’ postoperative quality of life and even lead to death (24). Previous investigators have shown rates of anastomotic complications following sleeve lobectomy patients of 5.2% (25-27). In our study, the rate of anastomotic complications (bronchopleural fistula) was 0.7%. This difference may be attributed to variations in efficacy of preoperative management (such as correcting malnutrition, treating underlying diseases, and controlling chronic lung infections) and intraoperative protection of the bronchial blood supply.

A previous study considered older age as a risk factor for postoperative complications in centrally located NSCLCs (28). Likewise, in our cohort, older patients were more prone to experiencing complications. The CCI assesses complications based on both the quantity and gravity of diseases. It quantifies complications and can predict the risk of death from diseases (29). In lung cancer patients, a higher CCI score is correlated to a higher risk of 3-year death (30). Blanc et al. discovered that the CCI index serves as an independent risk factor for acute respiratory distress syndrome (ARDS) and tracheal intubation following lung surgery (31). Daffrè et al. discovered that it was linked to perioperative mortality (32). In our study, patients with a higher CCI score experienced more postoperative complications, and the severity of these complications was higher than those with a lower CCI index. This result aligns with the above studies. However, it is important to note that the CCI score is only a quantitative index. In our study, one patient died three days after surgery because of severe ventricular fibrillation. This patient had a 21-year history of hypertension with poor blood pressure control. The postoperative stress state, coupled with hypoxia from intense coughing, triggered the ventricular fibrillation. Interestingly, our predictive models indicated no complication or major complication for this patient. We believe this discrepancy arose as the CCI score did not fully reflect the severity and control of hypertension. Therefore, it is critical to assess the patients’ fundamental states before surgery.

Intraoperative blood transfusion is frequently associated with intraoperative bleeding problems then that have a deep impact on the post-operative course. Among our patients, intraoperative blood transfusion was significantly associated with postoperative complications and possessed the highest feature importance weight in the predictive complication model. A study has also identified intraoperative blood transfusion as a predictor of postoperative complications, such as atrial fibrillation in lung cancer lobectomy (33). Another study has found that blood transfusion is related to unfavorable short-term or long-term outcomes in surgical patients (34). It is reasonable to control the surgical risk and ensure the quality of surgery to reduce the occurrence of intraoperative blood transfusion.

We also developed models to predict minor complications, such as pneumonia, prolonged air leaks, and cardiac arrhythmia. These minor complications could prolong hospitalization, impair the patient’s quality of life, and increase hospital readmissions and medical costs (16). Some prior studies have constructed models to predict these complications in lung cancer patients. Song et al. developed an algorithm to predict postoperative pneumonia with an AUC of 0.71 (35). Kim et al. used intraoperative ventilatory leaks to predict prolonged air leaks after lung resection, and the AUC of the model was 0.80 (36). Another study conducted by Nojiri et al. predicted postoperative atrial fibrillation by analyzing preoperative tissue Doppler echocardiography and achieved an AUC of 0.83 (37). In our study, the models achieved an AUC of 0.78 for predicting pneumonia, an AUC of 0.78 for predicting prolonged air leaks, and an AUC of 0.89 for predicting cardiac arrhythmia. In the validation cohort, one patient died of severe pneumonia and chylothorax after surgery. The models for predicting overall complications or major complications did not assign the right signatures to this patient. However, the occurrence of pneumonia was successfully predicted by the pneumonia model. These minor complication models could be used as a complement to the overall complication and major complication models.

We also conducted a prognosis analysis to investigate the impact of postoperative complications after sleeve lobectomy. The results showed that the complications were associated with adverse outcomes. Pieces of evidence about the relationship between postoperative complications and prognosis in NSCLC are similar to our results. A study conducted by Lugg et al. showed that postoperative complications were an independent predictor of worse 5-year cancer-specific survival in stage I NSCLC (14). Nakada et al. concluded that postoperative complications were significantly related to a poor prognosis in NSCLC patients who underwent lobectomy (15). Lugg et al. and Wang et al. found that a postoperative pulmonary complication after thoracic surgery is associated with a poorer long-term outcome (14,38). We recognize that postoperative complications can have a direct impact on patient outcomes, including recurrence and survival. However, it is also important to consider that these complications may be indicative of an underlying fragility or predisposition of the patient to adverse events. In this context, the predictive models for complications could serve as a valuable tool to help identify and correct preoperative risk factors, thus optimizing patients before surgery. By reducing the risk of postoperative complications, we may be able to improve patient outcomes and reduce the burden of adverse events.

We acknowledge that this study is limited by its retrospective nature, and as a result, the incidence of complications in patients may be underestimated. Second, our models were developed based on a single-center database, making it difficult to further verify the robustness and generalization of the models. Therefore, there is a need for future prospective, multi-center studies to validate our findings and further refine the predictive models. Third, because of the small sample size of major complications, our models only offer a broad estimation of major complication risks without specific information on the types of postoperative major complications. Finally, it should be noted that the median follow-up duration of 29 months is slightly shorter than the full three-year period for which OS and RFS were reported. This discrepancy may introduce a degree of bias into our results, and thus, the three-year OS and RFS estimates should be interpreted with caution. However, given the nature of our study and the current data available, we believe that these estimates still provide valuable insights for clinical practice and future research.

Conclusions

We created models to predict the probability and severity of postoperative complications for patients undergoing sleeve lobectomy. These models may help future healthcare teams in making medical decisions, thereby ensuring that patients who are being considered for sleeve lobectomy receive appropriate workup, informed discussions of risk, and individualized recommendations.

Acknowledgments

Funding: This study was supported by the National Natural Science Foundation of China (Nos. 92259205, 82272943, 8210071009), Science and Technology Commission of Shanghai Municipality (Nos. 20XD1403000, 21Y11913400), Special Clinical Research Project of Shanghai Pulmonary Hospital (FKLY20006) and Ningbo Top Medical and Health Research Program (No.2022030208).

Footnote

Reporting Checklist: The authors have completed the TRIPOD reporting checklist. Available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-24-325/rc

Data Sharing Statement: Available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-24-325/dss

Peer Review File: Available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-24-325/prf

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-24-325/coif). C.C. serves as an unpaid Associate Editor-in-Chief of Translational Lung Cancer Research from August 2023 to July 2024. M.B.A. received consulting fees for Astra Zeneca, Merck, Bristol Myers Squibb, Ethicon. The other authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013). The study was approved by the institutional review board of Shanghai Pulmonary Hospital (No. K22-793; 15 December 2022) and individual consent for this retrospective analysis was waived.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Park JS, Yang HC, Kim HK, et al. Sleeve lobectomy as an alternative procedure to pneumonectomy for non-small cell lung cancer. J Thorac Oncol 2010;5:517-20. [Crossref] [PubMed]
2. Matsuo T, Imai K, Takashima S, et al. Outcomes and pulmonary function after sleeve lobectomy compared with pneumonectomy in patients with non-small cell lung cancer. Thorac Cancer 2023;14:827-33. [Crossref] [PubMed]
3. Voltolini L, Viggiano D, Gonfiotti A, et al. Complex Sleeve Lobectomy Has Lower Postoperative Major Complications Than Pneumonectomy in Patients with Centrally Located Non-Small-Cell Lung Cancer. Cancers (Basel) 2024;16:261. [Crossref] [PubMed]
4. Yang M, Zhong Y, Deng J, et al. Comparison of Bronchial Sleeve Lobectomy With Pulmonary Arterioplasty Versus Pneumonectomy. Ann Thorac Surg 2022;113:934-41. [Crossref] [PubMed]
5. Li X, Li Q, Yang F, et al. Neoadjuvant therapy does not increase postoperative morbidity of sleeve lobectomy in locally advanced non-small cell lung cancer. J Thorac Cardiovasc Surg 2023;166:1234-1244.e13. [Crossref] [PubMed]
6. Nitsche LJ, Jordan S, Demmy T, et al. Analyzing the impact of minimally invasive surgical approaches on post-operative outcomes of pneumonectomy and sleeve lobectomy patients. J Thorac Dis 2023;15:2497-504. [Crossref] [PubMed]
7. Qiu T, Zhao Y, Xuan Y, et al. Robotic sleeve lobectomy for centrally located non-small cell lung cancer: A propensity score-weighted comparison with thoracoscopic and open surgery. J Thorac Cardiovasc Surg 2020;160:838-846.e2. [Crossref] [PubMed]
8. Gu C, Pan X, Chen Y, et al. Short-term and mid-term survival in bronchial sleeve resection by robotic system versus thoracotomy for centrally located lung cancer. Eur J Cardiothorac Surg 2018;53:648-55. [Crossref] [PubMed]
9. Gonzalez M, Chriqui LE, Décaluwé H, et al. Sleeve lobectomy in patients with non-small-cell lung cancer: a report from the European Society of Thoracic Surgery database 2021. Eur J Cardiothorac Surg 2022;62:ezac502. [Crossref] [PubMed]
10. Zhang C, Ma Y, Yu Z, et al. Comparison of efficacy and safety of hybrid video-assisted thoracoscopic surgery vs. thoracotomy sleeve lobectomy for non-small cell lung cancer: a propensity score matching study. J Thorac Dis 2022;14:2635-44. [Crossref] [PubMed]
11. Campisi A, Dell'Amore A, Faccioli E, et al. A Multicenter Retrospective Case-Control Study on Simple vs Extended Sleeve Lobectomies. 2024. [Epub ahead of print]. doi: 10.1016/j.athoracsur.2024.01.003.
12. Heiden BT, Keller M, Meyers BF, et al. Assessment of short readmissions following elective pulmonary lobectomy. Am J Surg 2023;225:220-5. [Crossref] [PubMed]
13. Rueth NM, Parsons HM, Habermann EB, et al. The long-term impact of surgical complications after resection of stage I nonsmall cell lung cancer: a population-based survival analysis. Ann Surg 2011;254:368-74. [Crossref] [PubMed]
14. Lugg ST, Agostini PJ, Tikka T, et al. Long-term impact of developing a postoperative pulmonary complication after lung surgery. Thorax 2016;71:171-6. [Crossref] [PubMed]
15. Nakada T, Noda Y, Kato D, et al. Risk factors and cancer recurrence associated with postoperative complications after thoracoscopic lobectomy for clinical stage I non-small cell lung cancer. Thorac Cancer 2019;10:1945-52. [Crossref] [PubMed]
16. Lawson EH, Hall BL, Louie R, et al. Association between occurrence of a postoperative complication and readmission: implications for quality improvement and cost savings. Ann Surg 2013;258:10-8. [Crossref] [PubMed]
17. Yamamichi T, Ichinose J, Omura K, et al. Impact of postoperative complications on the long-term outcome in lung cancer surgery. Surg Today 2022;52:1254-61. [Crossref] [PubMed]
18. Thomas PA, Berbis J, Baste JM, et al. Pneumonectomy for lung cancer: contemporary national early morbidity and mortality outcomes. J Thorac Cardiovasc Surg 2015;149:73-82. [Crossref] [PubMed]
19. Yu X, Gao S, Xue Q, et al. Development of a nomogram for predicting the operative mortality of patients who underwent pneumonectomy for lung cancer: a population-based analysis. Transl Lung Cancer Res 2021;10:381-91. [Crossref] [PubMed]
20. Dindo D, Demartines N, Clavien PA. Classification of surgical complications: a new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg 2004;240:205-13. [Crossref] [PubMed]
21. Goldstraw P, Chansky K, Crowley J, et al. The IASLC Lung Cancer Staging Project: Proposals for Revision of the TNM Stage Groupings in the Forthcoming (Eighth) Edition of the TNM Classification for Lung Cancer. J Thorac Oncol 2016;11:39-51. [Crossref] [PubMed]
22. Charlson ME, Pompei P, Ales KL, et al. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis 1987;40:373-83. [Crossref] [PubMed]
23. Yang R, Zhou Y, Gao S, et al. Effects of 24-hour postoperative intravenous fluid on postoperative outcomes after lobectomy: a retrospective observational study. J Thorac Dis 2022;14:2602-10. [Crossref] [PubMed]
24. Hu XF, Duan L, Jiang GN, et al. A clinical risk model for the evaluation of bronchopleural fistula in non-small cell lung cancer after pneumonectomy. Ann Thorac Surg 2013;96:419-24. [Crossref] [PubMed]
25. Comacchio GM, Schiavon M, Azzolina D, et al. Does Induction Therapy Increase Anastomotic Complications in Bronchial Sleeve Resections? World J Surg 2019;43:1385-92. [Crossref] [PubMed]
26. Koryllos A, Lopez-Pastorini A, Zalepugas D, et al. Bronchus Anastomosis Healing Depending on Type of Neoadjuvant Therapy. Ann Thorac Surg 2020;109:879-86. [Crossref] [PubMed]
27. Gómez-Caro A, Boada M, Reguart N, et al. Sleeve lobectomy after induction chemoradiotherapy. Eur J Cardiothorac Surg 2012;41:1052-8. [Crossref] [PubMed]
28. Pricopi C, Mordant P, Rivera C, et al. Postoperative morbidity and mortality after pneumonectomy: a 30-year experience of 2064 consecutive patients. Interact Cardiovasc Thorac Surg 2015;20:316-21. [Crossref] [PubMed]
29. Birim O, Kappetein AP, Bogers AJ. Charlson comorbidity index as a predictor of long-term outcome after surgery for nonsmall cell lung cancer. Eur J Cardiothorac Surg 2005;28:759-62. [Crossref] [PubMed]
30. Yang CC, Fong Y, Lin LC, et al. The age-adjusted Charlson comorbidity index is a better predictor of survival in operated lung cancer patients than the Charlson and Elixhauser comorbidity indices. Eur J Cardiothorac Surg 2018;53:235-40. [Crossref] [PubMed]
31. Blanc K, Dechartres A, Zaimi R, et al. Patients experiencing early acute respiratory failure have high postoperative mortality after pneumonectomy. J Thorac Cardiovasc Surg 2018;156:2368-76. [Crossref] [PubMed]
32. Daffrè E, Prieto M, Huang H, et al. Normalized Pulmonary Artery Diameter Predicts Occurrence of Postpneumonectomy Respiratory Failure, ARDS, and Mortality. Cancers (Basel) 2020;12:1515. [Crossref] [PubMed]
33. Ishibashi H, Wakejima R, Asakawa A, et al. Postoperative Atrial Fibrillation in Lung Cancer Lobectomy-Analysis of Risk Factors and Prognosis. World J Surg 2020;44:3952-9. [Crossref] [PubMed]
34. Lu Q, Zhang J, Gao WM, et al. Intraoperative Blood Transfusion and Postoperative Morbidity Following Liver Resection. Med Sci Monit 2018;24:8469-80. [Crossref] [PubMed]
35. Song Y, Liu J, Lei M, et al. An External-Validated Algorithm to Predict Postoperative Pneumonia Among Elderly Patients With Lung Cancer After Video-Assisted Thoracoscopic Surgery. Front Oncol 2021;11:777564. [Crossref] [PubMed]
36. Kim WH, Lee HC, Ryu HG, et al. Intraoperative ventilatory leak predicts prolonged air leak after lung resection: A retrospective observational study. PLoS One 2017;12:e0187598. [Crossref] [PubMed]
37. Nojiri T, Maeda H, Takeuchi Y, et al. Predictive value of preoperative tissue Doppler echocardiographic analysis for postoperative atrial fibrillation after pulmonary resection for lung cancer. J Thorac Cardiovasc Surg 2010;140:764-8. [Crossref] [PubMed]
38. Wang S, Li X, Li Y, et al. The long-term impact of postoperative pulmonary complications after video-assisted thoracic surgery lobectomy for lung cancer. J Thorac Dis 2017;9:5143-52. [Crossref] [PubMed]