Lung cancer has high incidence and mortality rates globally, and surgical resection remains the primary treatment method.1 Traditional open thoracotomy involves large incisions, strong patient stress responses, and high postoperative pain levels. With the advancement of medical equipment, thoracoscopic surgery has emerged.2,3 Thoracoscopic surgery is a minimally invasive technique that allows complex procedures to be performed through small incisions, using advanced imaging technology and specialized instruments. This approach provides surgeons with enhanced, magnified views of the surgical area, leading to greater precision and typically faster recovery times for patients.4,5 Cough after pulmonary resection (CAP) is one of the complications following thoracoscopic lung resection, with an incidence rate of 25% to 50%. The occurrence of CAP can increase pain, disrupt sleep and eating, and hinder recovery, making it crucial to identify influencing factors and intervene promptly.6,7 The study highlights that factors such as the extent of surgical resection, the number of lymph nodes removed, and smoking history may influence the likelihood of CAP. However, there remains a gap in clinical practice, as reliable models for providing individualized predictions are still lacking.8. A nomogram is a model developed based on the results of multivariate regression analysis, visually representing the contribution of various variables to the outcome. It holds significant potential for clinical application. Therefore, this study collected clinical data from 374 patients who underwent thoracoscopic lung resection and constructed a nomogram prediction model based on the influencing factors of CAP. This study aims to provide a reference for the prevention and treatment of postoperative CAP in lung cancer patients, and the report is as follows.
Study Subjects and MethodsStudy SubjectsA retrospective study was conducted on 374 lung cancer patients who underwent lung resection surgery at our hospital between March 2020 and October 2023.Using a random number table method (modeling group: validation group = 7:3), they were divided into a modeling group (n=262) and a validation group (n=112), as shown i Figure 1. Inclusion criteria: (1) postoperative pathological diagnosis of lung cancer; (2) first-time thoracoscopic lung cancer resection; (3) complete medical records; (4) age ≥18 years; (5) no preoperative radiotherapy or chemotherapy. Exclusion criteria: (1) pre-existing chronic cough or conditions such as bronchiectasis or pneumonia that could cause coughing; (2) conversion from thoracoscopic to open surgery; (3) preoperative use of cough-inducing medications; (4) communication disorders; (5) postoperative complications like pulmonary embolism; (6) follow-up duration of less than one month. See follow Figure 1. This study was approved by the hospital’s medical ethics committee.
Figure 1 Flow chart of case collection.
Abbreviation: VATS, video-assisted thoracic surgery.
MethodsDiagnosis of CAPReferring to the diagnostic criteria for CAP9 (cough lasting ≥2 weeks within one month after surgery, no abnormalities on chest X-ray, excluding cough caused by recurrence or infection), The occurrence of CAP in the modeling group was divided into the CAP group (n=85) and the non-CAP group (n=177).
Clinical Data CollectionData were collected from 374 patients on the following variables: age, gender, smoking history, TNM stage, pathological type, whether preoperative respiratory function training was conducted, the side and scope of surgery, whether peribronchial lymph node dissection was performed, history of underlying conditions (eg, hyperlipidemia, coronary heart disease, diabetes, hypertension), body mass index, surgery duration, and intraoperative blood loss. Smoking history was defined as smoking at least one cigarette per day for a minimum of six consecutive or cumulative months.
Statistical AnalysisData were processed using SPSS 25.0. Count data such as TNM staging, surgical side, and surgical range were expressed as n (%), and the X2 test was used. Measurement data were expressed as () and analyzed using the t-test. Multivariate logistic regression analysis in SPSS was used to assess the factors influencing the development of CAP in lung cancer patients, followed by the Hosmer-Lemeshow test for model evaluation. The rms package in R 4.3.1 was used to construct a nomogram model for predicting the risk of CAP, with internal validation performed using the Bootstrap method, and a calibration curve was plotted. The rmda package in R 4.3.1 was used to draw a clinical decision curve (DCA) to assess the clinical application value of the nomogram. ROC analysis in SPSS was performed to evaluate the predictive discrimination of the nomogram model for CAP and non-CAP. A P-value <0.05 was considered statistically significant.
ResultsComparison of Clinical Data Between the Modeling and Validation GroupsThere were no significant differences in smoking history, TNM staging, surgical range, and other factors between the modeling and validation groups (P > 0.05). See Table 1.
Table 1 Comparison of Clinical Data Between Modeling Group and Validation Group[(), n (%)]
Univariate Analysis of CAP Occurrence in the Modeling Group PatientsUnivariate analysis showed that, compared to the non-CAP group, the CAP group had a significantly higher proportion of patients with a smoking history (X2=4.963), no preoperative respiratory function training (X2=5.959), lobectomy as the surgical range (X2=6.198), and a higher percentage of bronchial tree perilymphatic dissection (X2=8.648) (P<0.05). There were no significant differences in gender, surgical side, and surgery duration between the two groups (P > 0.05). See Table 2.
Table 2 Univariate Analysis of CAP Occurrence in the Modeling Group[(), n (%)]
Multifactorial Logistic Regression Analysis of CAP Occurrence in the Modeling Group PatientsTaking the occurrence or non-occurrence of CAP in the modeling group as the dependent variable (occurrence=1, non-occurrence=0) and the factors with statistically significant differences in Table 2 as independent variables, multifactorial logistic regression analysis showed that smoking history (OR=6.285, 95% CI: 3.03113.036), lack of preoperative respiratory function training (OR=20.293, 95% CI: 7.51854.779), surgical range (OR=20.667, 95% CI: 7.73455.228), and bronchial tree perilymphatic dissection (OR=5.883, 95% CI: 2.82912.235) were all influencing factors for the occurrence of CAP of in lung cancer patients (P < 0.05). See Table 3.
Table 3 Multivariate Logistic Regression Analysis of CAP Occurrence in the Modeling Group
Nomogram Model for Predicting the Risk of CAP in Lung Cancer PatientsA nomogram model for predicting the risk of postoperative CAP in lung cancer patients was developed using R software, incorporating the four influencing factors identified through multivariate analysis. The results indicated that smoking history contributed 61 points, lack of preoperative respiratory function training contributed 99 points, lobectomy as the surgical range contributed 100 points, and bronchial perilymphatic dissection contributed 59 points. The total score for the four factors is summed to obtain the overall score, and by drawing a vertical line from the total score axis to the prediction probability axis, the corresponding probability of CAP occurrence can be determined. For example, a total score of 223 corresponds to a 0.7 risk of postoperative CAP, while a score of 267 corresponds to a 0.9 risk. See Figure 2.
Figure 2 A Nomogram model for predicting the risk of CAP in lung cancer patients.
Abbreviation: CAP, cough after pulmonary resection.
Evaluation of the Nomogram Prediction ModelIn the Hosmer-Lemeshow (H-L) goodness-of-fit test, the modeling group had a X2=4.125, P=0.660, and the validation group had a X2=11.980, P=0.101. The calibration curves (Figure 3A and B) showed slopes close to 1 for both the modeling and validation groups. The ROC curves (Figure 3C and D) showed that the AUC for the modeling group was 0.794 (95% CI: 0.737–0.851), while the AUC for the validation group was 0.751 (95% CI: 0.652–0.851). In the DCA curve (Figure 4), the horizontal line along the x-axis indicates that none of the patients developed CAP or received intervention, resulting in a net benefit rate of 0. The gray diagonal line represents the scenario where all patients developed CAP and received intervention. The further the red curve is from these two lines, the higher the net benefit rate. From the graph, it is evident that the nomogram model shows high clinical application value when the predicted high-risk threshold is between 0.08 and 0.98.
Figure 3 Evaluation of Nomogram prediction model for predicting the risk of CAP after VATS in lung cancer patients (A)Calibration curve of model group; (B)ROC curve of model group; (C) Calibration curve of verify group; (D) ROC curve of verify group.
Abbreviations: CAP, cough after pulmonary resection; VATS, video-assisted thoracic surgery; ROC, receiver operating characteristic.
Figure 4 DCA curve of the predictive model for the risk of CAP after VATS in lung cancer patients.
Abbreviations: DCA, decision curve analysis; CAP, cough after pulmonary resection; VATS, video-assisted thoracic surgery.
DiscussionClinically, it is believed that moderate coughing after lung cancer surgery is beneficial for expectoration and promoting lung re-expansion, reducing the chances of lung infection. However, prolonged coughing is detrimental to recovery and may develop into chronic cough, affecting the quality of life later on.10,11 In our study, 32.44% of the 262 patients in the modeling group developed CAP postoperatively. Lu et al12 reported that 36.6% of 112 patients who underwent lobectomy experienced refractory cough after surgery. In a study by Xie et al13 39.77% (68/171) of lobectomy patients developed chronic cough postoperatively. These high rates of postoperative cough among lung cancer patients highlight the need for clinical attention, and understanding the factors influencing postoperative CAP is crucial for reducing its incidence.
In our study, by comparing clinical data of patients in the CAP and non-CAP groups, we found that a history of smoking, preoperative respiratory function training, surgical range, and bronchial tree perilymphatic dissection are influencing factors for the development of CAP in lung cancer patients. Mu et al14 reported that the occurrence of postoperative CAP in lung cancer patients was related to resection of the right upper lobe. In a study by Wu et al15 postoperative acid reflux and female gender were also identified as independent risk factors for CAP after lung resection. However, these findings were not observed in our study and require further investigation.
The development of predictive models has become a key focus of clinical research in recent years, but there are still few reports on CAP risk prediction models for lung cancer patients. Based on the above influencing factors, a nomogram prediction model for assessing the risk of CAP occurrence in lung cancer patients was developed. This model, being a visual graph, offers better readability and enables individualized predictions for patients. The model shows that a smoking history contributes 61 points. However, some studies suggest that preoperative smoking may reduce cough sensitivity, potentially acting as a protective factor against postoperative CAP, which requires further investigation. Lack of preoperative respiratory function training adds 99 points, as such training can improve respiratory muscle strength, enhance pulmonary ventilation and gas exchange, and help prevent postoperative hypoxia and ease respiratory difficulties.16 Lobectomy as the surgical procedure contributes 100 points, while bronchial perilymphatic dissection adds 59 points. The impact of surgical range and bronchial tree lymph node dissection on CAP is mainly due to surgical trauma. Lobectomy involves the removal of a larger portion of lung tissue, which reduces intrathoracic pressure, alters the thoracic structure, causes bronchial distortion, and changes airflow dynamics.17 Vagal nerve C-fibers, the primary cough receptors crucial for the cough reflex, are distributed in the larynx, trachea, and bronchi within the lungs.18 Perilymphatic dissection around the bronchial tree can cause trauma to the tracheal wall, trigger the release of neurogenic inflammatory factors, and damage the vagus nerve, leading to increased cough sensitivity.19 Clinically, based on the individual patient’s condition with regard to the four factors, the corresponding score and prediction probability can be calculated, allowing for preoperative prevention and postoperative care to be planned in advance. It is recommended that patients undergo respiratory function training before surgery. For those with a high predicted risk of CAP, patients should be informed prior to surgery and advised to avoid exposure to irritants such as gases, dust, and pollen after surgery to reduce the risk of CAP. Additionally, they should be advised to stay warm and avoid catching colds to prevent worsening CAP symptoms.
To assess the predictive efficacy of this nomogram model, this study verifies the model. The results showed that both passed the Hosmer-Lemeshow goodness-of-fit test, indicating consistency with actual conditions. In ROC analysis, the AUCs for the modeling and validation groups were 0.794 and 0.751. It is suggested that the prediction differentiation between CAP and non-CAP is reasonable. Additionally, the DCA curve demonstrated that the nomogram model provided a higher net clinical benefit when predicting risk values between 0.08 and 0.98, indicating its high value for clinical application.
In summary, smoking history, lack of preoperative respiratory function training, surgical scope, and bronchial perilymphatic dissection are key factors influencing the development of CAP in lung cancer patients. The nomogram model based on these four factors demonstrates strong predictive accuracy and clinical relevance. However, as this is a retrospective, single-center study, there are certain limitations, and the findings need to be validated through multi-center research.
Research Involving Human ParticipantsThe study was approved by Lishui Central Hospital ethics review board and with the 1964 helsinki Declaration. Written informed consent to participate in this study was provided by the participants.
Data Sharing StatementThe original contributions presented in the study are included in the article.
Consent for PublicationAll authors give consent for publication.
FundingThere is no funding to report.
DisclosureAuthors declared no conflict of interest.
References1. Jakobsen E, Olsen KE, Bliddal M, Hornbak M, Persson GF, Green A. Forecasting lung cancer incidence, mortality, and prevalence to year 2030. BMC Cancer. 2021;21(1):985. doi:10.1186/s12885-021-08696-6
2. Hallet J, Rousseau M, Gupta V, et al. Long-term functional outcomes among older adults undergoing video-assisted versus open surgery for lung cancer: a population-based cohort study. Ann Surg. 2023;277(6):e1348–e1354. doi:10.1097/SLA.0000000000005387
3. Lim E, Harris RA, McKeon HE, et al. Impact of video-assisted thoracoscopic lobectomy versus open lobectomy for lung cancer on recovery assessed using self-reported physical function: VIOLET RCT. Health Technol Assess. 2022;26(48):1–162. doi:10.3310/THBQ1793
4. Nakagawa K, Yoshida Y, Yotsukura M, Watanabe S-I. Minimally invasive open surgery (MIOS) for clinical stage I lung cancer: diversity in minimally invasive procedures. Jpn J Clin Oncol. 2021;51(11):1649–1655. doi:10.1093/jjco/hyab128
5. Jeon YJ, Choi YS, Cho JH, et al. Thoracoscopic vs open surgery following neoadjuvant chemoradiation for clinical n2 lung cancer. Semin Thorac Cardiovasc Surg. 2022;34(1):300–308. doi:10.1053/j.semtcvs.2021.01.002
6. Zhu Y-F, S-B W, Zhou M-Q, et al. Increased expression of TRPV1 in patients with acute or chronic cough after lung cancer surgery. Thorac Cancer. 2019;10(4):988–991. doi:10.1111/1759-7714.13042
7. Sun X, Lan Z, Li S, et al. Trajectories and risk factors of persistent cough after pulmonary resection: a prospective two-center study. Thorac Cancer. 2023;14(36):3503–3510. doi:10.1111/1759-7714.15147
8. Li X, Li X, Zhang W, et al. Factors and potential treatments of cough after pulmonary resection: a systematic review. Asian J Surg. 2021;44(8):1029–1036. doi:10.1016/j.asjsur.2021.01.001
9. Pan L-Y, Peng L-P, Xu C, et al. Predictive factors of cough after uniportal video-assisted thoracoscopic pulmonary resection. J Thorac Dis. 2020;12(10):5958–5969. doi:10.21037/jtd-20-2652
10. Lin R, Chen W, Zhu L, Pan X. Comparison of postoperative cough-related quality of life and recovery between sublobectomy and lobectomy for early-stage non-small cell lung cancer patients: a longitudinal study. BMC Pulm Med. 2022;22(1):154. doi:10.1186/s12890-022-01954-8
11. Lin R, Zhu L, Pan X. Validation of the simplified cough symptom score in non-small cell lung cancer patients after surgery. Gen Thorac Cardiovasc Surg. 2022;70(8):735–739. doi:10.1007/s11748-022-01791-w
12. X-F L, Min X-P, Lu B, Fan G-H, Zhu T-Y. Bronchial morphological changes are associated with postoperative intractable cough after right upper lobectomy in lung cancer patients. Quant Imaging Med Surg. 2022;12(1):196–206. doi:10.21037/qims-21-368
13. Xie M-R, Zhu Y-F, Zhou M-Q, et al. Analysis of factors related to chronic cough after lung cancer surgery. Thorac Cancer. 2019;10(4):898–903. doi:10.1111/1759-7714.13021
14. Mu T, Li J, Huang Q, Li X. Characteristics and risk factors for persistent cough after pulmonary resection. Ann Thorac Surg. 2023;115(6):1337–1343. doi:10.1016/j.athoracsur.2022.05.054
15. Wu Y, Song W, Zhu D, Wang Y, Che G. Risk factors for cough after pulmonary resection. World J Surg Oncol. 2023;21(1):348. doi:10.1186/s12957-023-03235-y
16. Qin -P-P, Jin J-Y, Wang W-J, Min S. Perioperative breathing training to prevent postoperative pulmonary complications in patients undergoing laparoscopic colorectal surgery: a randomized controlled trial. Clin Rehabil. 2021;35(5):692–702. doi:10.1177/0269215520972648
17. Pan L, Zhou X, Xu C, et al. Effects of preserving the bronchial artery on cough after thoracoscopic lobectomy. Thorac Cancer. 2021;12(15):2151–2160. doi:10.1111/1759-7714.14012
18. Gu S, Wang W, Wang X, et al. Effects of preserving the pulmonary vagus nerve branches on cough after pneumonectomy during video-assisted thoracic surgery. Front Oncol. 2022;12:837413. doi:10.3389/fonc.2022.837413
19. Wu X, Xing H, Chen P, et al. Lymph node dissection is a risk factor for short-term cough after pulmonary resection. Curr Oncol. 2022;29(1):294–307. doi:10.3390/curroncol29010027
留言 (0)