The use of postoperative X-ray to evaluate residual pleural space and predict infectious pleuropulmonary complications after lung resection for infectious disease

Introduction

The postoperative residual pleural space (RPS) after lung resection is always seen with caution by the thoracic surgeon and the respiratory medicine team during the postoperative care period. Its frequency is reported in about 20% of lobectomies and 40% in combined resections such as bilobectomies or segmentectomies associated with lobectomies (1-3). Most of these studies are carried out with a heterogeneous population containing patients undergoing pulmonary resection for different etiologies, which makes it difficult to project the results to a population with specific characteristics, such as those with infectious pulmonary disease (4). Surrounded by peculiarities, resections for such etiologies are related to a higher incidence of postoperative complications, either due to the clinical condition of the patient, sometimes consumed by the infectious process, or due to technical peculiarities of the procedure.

The chronic inflammatory status of the pleura, hilar, and mediastinal tissues make the surgery technically more challenging when compared to the same procedure performed for the treatment of neoplastic diseases in early stages. The presence of enlarged hilar lymph nodes, intense mediastinal fibrosis, retraction of costal arches, and pleuropulmonary adhesions result in an increased surgical time, blood loss, prolonged air leak, need for intensive care unit (ICU) recovery, and a longer time of hospitalization (5). The pathologically colonized lung parenchyma makes more possible the contamination of the pleural space during the procedure and increases the concern about the development of pleuropulmonary infectious complications (5,6).

The incidence of postoperative pleural empyema is well described for neoplastic disease and ranges from 3% to 6% (5). The frequency is higher according to the size of the resection, reaching 12% in pneumonectomies, 3–7% in lobectomies. For pneumonia, it is even higher, reaching 26% for major lung resections (6). As reported by the North American Center for Disease Control (CDC) (7), those conditions are classified as surgical site infection being responsible for the increased postoperative morbidity with a reported mortality of up to 25% for pneumonia and 22% for empyema (8). The already know preoperative risk factors are male gender, diabetes, preoperative chemotherapy and radiotherapy, obesity, malnutrition, and immunosuppression (5-9). Postoperative risk factors are associated with major lung resection, long surgical time, blood loss, prolonged air leak, and presence of residual pleural cavity (5). Most of them frequently observed int patients with RPS. However, even the concept of RPS is poorly elucidated, lacking objective data for the elaboration of more detailed studies that can trace correlation and causality between the presence of the cavity, its volume, duration, and the incidence of pleuropulmonary complications.

Therefore, to answer these questions, we aim to describe a new method of chest X-ray evaluation capable of identifying and measuring the RPS. This makes it possible to identify the frequency of RPS in patients undergoing anatomical lung resection in the treatment of infectious lung diseases, as well as to correlate its presence with the incidence of postoperative infectious pleuropulmonary complications. We present this article in accordance with the STROBE reporting checklist (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2024-2022/rc).

Methods

Study design

This study was conducted as a retrospective cohort study, utilizing a blinded methodology to ensure objectivity in correlating clinical outcomes with the evaluation of radiographic examinations. Specifically, the radiologists tasked with interpreting chest X-rays were blinded to the patients’ clinical data to eliminate potential bias. Clinical and radiological data were retrieved from digital medical records and the institutional surgical database.

The research team played a pivotal role in data management. One group of investigators was responsible for collecting and organizing clinical and radiological data, as well as overseeing study monitoring. Another team, distinct from the clinical evaluators, was tasked with distributing anonymized radiological exams to radiologists who followed a standardized chest X-ray evaluation protocol.

Following the standardized radiographic analysis, the study population was divided into two groups: the study group, comprising patients who developed RPS, and the control group, comprising those who did not. Comparative statistical analyses were performed to assess differences between the two groups, along with a descriptive analysis of each group.

Study population

The study population consisted of patients who underwent lung resection between 2003 and 2020 in a single thoracic surgery service for the treatment of infectious lung disease. Inclusion criteria were patients older than 12 years, undergoing non-pneumonectomy lung resection for infectious disease including bronchiectasis, tuberculosis sequelae, aspergillosis, necrotizing pneumonia, non-tuberculous mycobacteria (NTM), congenital lung malformation. We excluded patients under 12 years of age, with oncological disease as the reason for lung resection, patients who underwent chest wall resection prior to or at the same time as lung resection, and patients who underwent pneumonectomies or wedge resections for diagnostic purpose.

The exclusion of patients undergoing pneumonectomy and patients younger than 12 years aims to maintain internal validity, since these cofactors directly influence the presence of residual pleural cavity and standard anatomical measurements of the pulmonary cavity. Cancer patients were excluded as they are not the target population involved in the study. To optimize the external validity and greater power to generalize the results, patients with comorbidities were not excluded. The comorbidities were collected, tabulated, and included in the multivariate statistical analysis.

The sample size calculation was based on an alpha value of 0.05 aiming at a power close to 80% for the proposed statistical tests. Data from the literature and preliminary assessment of the researchers’ clinical practice were used to perform the sample calculation, assuming a difference in the occurrence of pleuropulmonary events of 30% between patients who have a RPS and those who do not. The proportion between groups used for the calculation considered a three times higher incidence of cases in the study group. Based on these parameters, a minimum sample of 88 patients was calculated for comparative analysis of proportions between two independent groups, a minimum sample of 36 patients for comparison of survival curves.

Data collection

Data was stored on an online platform (Google Drive^®) linked to a secure institutional server of our university, with data protection guaranteed.

We collected baseline characteristics such as: sex, age, comorbidities [hypertension, diabetes, immunosuppression, chronic obstructive pulmonary disease (COPD)], smoking history, previous treatment for tuberculosis, type of resection, number of resected lung segments and use of aspiration during postoperative chest drainage.

Surgical procedure

Patients underwent pulmonary resection under general anesthesia with single lung orotracheal intubation. For cases treated by conventional approach, thoracic epidural catheter anesthesia was routinely performed, remaining until the chest tube was removed or adequate pain control achieved. Intraoperative prophylactic antibiotic therapy was routinely performed with cefuroxime. The thoracic access performed was defined by the surgeon according to his analysis of the case. In case of open approach, a lateral muscular-sparing thoracotomy was performed, and in case of minimal invasive approach we used a three-portal video-assisted technique. Cases that required costal resection were excluded from this paper. To perform all the resections, the release of the inferior pulmonary ligament was routinely performed, as well as all the pleuropulmonary adhesions. In some cases, where the surgeon judged the residual cavity to be large intraoperatively, phrenic nerve blockade with 2% lidocaine was performed. Drainage of the cavity was performed with a large-bore tubular chest drain (28-F or greater). In cases with severe pleuropulmonary adhesions with the possibility of bleeding and significant air leakage, two pleural drains were used, one anterior and one posterior. All patients were submitted to postoperative respiratory physiotherapy and rehabilitation exercises during the recovery period on the ward. They were encouraged to walk, perform respiratory exercises, and by indication of the physiotherapist, realized periods of non-invasive positive pressure ventilation.

Method of measurement of postoperative residual pleural cavity

To avoid interference from random radiological findings secondary to the immediate postoperative period, it was decided to select a chest radiograph from the 4^th postoperative day and a second late image performed during the second week of recovery. The images were analyzed in a software (RadiAnt^® DICOM Viewer ver. 2020.0) dedicated to DICOM format. With the aid of this software, two thoracic surgeons, blinded to the patient’s clinical outcome, carried out the evaluation of the radiological exams according to the standardized method. The method was developed in conjunction with a specialized radiology physician team and was based on methods of radiographic evaluation of pneumothorax (10). The researchers dedicated to the evaluation of radiographic exams underwent training with the radiology team to understand in practice the defined standardization and the software use.

The method for measuring RPS involves a systematic visual assessment by the evaluator, followed by precise quantification using the “closed polygon” software tool (Figure 1). Contrast adjustment is employed to enhance the visualization of the pleural cavity. The delineation of the measurement area incorporates the costophrenic sinus as part of the region of interest. Anatomical boundaries are defined as follows: the lateral limit is the inner edge of the costal arches, the medial limit is the ipsilateral edge of the vertebral body, the superior boundary is the inner edge of the first costal arch, and the inferior boundary is marked by the diaphragmatic dome. On the left side, the cardiac silhouette must be meticulously contoured and excluded to ensure accurate measurement of the total pleural cavity area. To reduce biases caused by postural interferences during image acquisition, it was decided to calculate the ratio between the RPS area and the area of the total pleural space in respective hemithorax, since both would be subject to the same rotation and positioning effects during the image acquisition. Therefore, three measurements were performed by the researchers: area corresponding to the postoperative RPS (ARPS), area of the total pleural space from the operated hemithorax (ATPS), and the ratio between ARPS and ATPS. To standardize the measurements, rules and limits were previously defined for the evaluation of both sides, based on the anatomical relationships of the thoracic structures (Figure 1). After performing the measurements, each surgeon computed the results in an anonymized table and returned their analysis to the researchers responsible for data collection and storage.

Figure 1 Pre-defined criteria for marking the perimeter of the RPS and the pleural cavity. ARPS, area corresponding to the postoperative RPS; ATPS, area of the total pleural space from the operated hemithorax; RPS, residual pleural space.

At the end of the measurements, an evaluation of agreement between the measurements was carried. In case of divergent measures, a third evaluator, from the radiology team, also unaware of the patient’s clinical outcome, performed a third measurement of the discordant cases, this being considered the definitive one.

Analyzed outcomes

The primary clinical outcome was incidence of empyema and pneumonia in the first 30 days after surgery. The diagnosis of empyema and/or pneumonia was accessed through data from the electronic medical record (leukocytosis, fever, purulent content through the drain), defined by the medical evaluation and infectious screening performed by the team at the time. Secondary clinical outcomes were chest tube duration, air leak, length of hospital stay, reoperation rate, antibiotic therapy time, and hospitalization in the first 30 days postoperative day.

Statistical analysis

The distribution of continuous variables was evaluated by the Shapiro-Wilk test and histogram analysis. Categorical variables were represented as absolute numbers and their frequency displayed as a percentage. Numerical variables are represented by means and standard deviation when with normal distribution or median and interquartile range when with non-normal distribution.

To define the best performance in a classificatory model by the cavity size measurement was used receiver operating characteristic (ROC) curve. The interobserver agreement was evaluated by the kappa concordance coefficient using the Cohen test, to define de concordance rate between both surgeons’ measurement.

Comparison analysis of means or medians between groups was performed using Student’s t-test when distribution was normal and Mann-Whitney test when distribution was not normal. To compare proportions between categorical variables, Fisher’s exact test or Chi-squared was used when the sample exceeds 100 observations.

To identify associated risk factors and the relative risk, univariate analysis and multiple logistic regression were performed. To define overall survival, the Kaplan-Meier curve with log-rank was used to compare groups. In the multivariate analysis, to identify which variables may be correlated to survival and/or event occurrence, the Cox regression model was used.

The tests used were bidirectional. The confidence interval (CI) was set at 95% and statistical significance with P<0.05.

Ethical statement

The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013). The study was approved by the institutional ethics committee of University of Sao Paulo (No. 4.512.453) and individual consent for this retrospective analysis was waived.

Results

A total of 135 patients were included and demographic data are expressed in Table 1. All patients had the chest radiographs available stored in the hospital digital system. After the chest X-ray evaluation, 78 patients (58%) had RPS identified on the 4^th postoperative chest X-ray and the cohort was divided between the two groups by presence or absence of RPS. After a ROC analysis we identified a cut-off point of 20% for the ARPS/ATPS index, with a sensibility and specificity of 75% [area under the curve (AUC) =0.76]. From patients with RPS, 29 (37.2%) presented an ARPS/ATPS index equal or greater than 20%. We observed a high interobserver agreement index of 89.7% of concordance (kappa =0.77). Patients with RPS had significantly more surgery by post-tuberculosis bronchiectasis and invasive aspergillosis (P=0.006). Aspergilloma was present in 39 patients (28.9%) of which 29 (74.4%) presented RPS (P=0.01).

Lobectomy was performed in 92 patients (68.1%), from those 51 (55.4%) were upper lobe resections (Table 1). Patients with RPS were submitted significantly more to upper lobe resections (P=0.01). In upper right lobectomy 69.7% of patients presented RPS on the 4^th postoperative chest X-ray. The median chest tube duration was significantly higher for patients with RPS (7 vs. 4 days, P=0.001) and chest tube aspiration was maintained for more than 1 day in 23 (17.0%) patients. The RPS group had significantly more cases of chest tube aspiration for more than 1 day (24.4% vs. 7.0%, P=0.01), and air leak for more than 5 days (28.2% vs. 10.5% P=0.009). The RPS persisted until the second week chest X-ray in 59 patients (75.6%). The median length of stay was 9 days for RPS group and 6 days for patients without RPS (P=0.001). Antibiotic therapy duration was significantly higher for the RPS group (7 vs. 1 day, P=0.009).

The postoperative clinical outcomes demonstrated a total rate of 14.1% of empyema and 11.1% for pneumonia (Table 2). A total of 62 patients (79.5%) with RPS had no pleural infection complication. However, patients with RPS had significantly more empyema (20.5% vs. 5.3%, P=0.01). In that case, the RPS represents a relative risk of 3.89 (95% CI: 1.19–12.74; P=0.02) for empyema development. If the ARPS/ATPS index was equal or higher than 20%, the relative risk increases for 6.26 (95% CI: 2.71–14.46; P=0.001). The empyema risk by prolonged air leak only was 4.28 (95% CI: 1.8–9.8, P=0.001). There was no significant increase in the relative risk for empyema calculated by immunosuppression [risk ratio (RR) =1.5; 95% CI: 0.5–4.5; P=0.48] and diabetes (RR =0.94; 95% CI: 0.24–3.7; P=0.93). The multivariate analysis including ARPS/ATPS <20% on 4^th postoperative chest X-ray; ARPS/ATPS ≥20% and prolonged air leak sowed that ARPS/ATPS ≥ 20% represents an isolated risk factor for development of empyema [odds ratio (OR) =6.12; 95% CI: 1.42–26.6, P=0.01] with a worst prognosis when combined to prolonged air leak (>5 days) (Figure 2).

Figure 2 Multivariate analysis showing the incidence of postoperative empyema during the 30 days. Left: comparative curves considering the RPS size (≥20% or <20% of the total pleural space). Right: comparative curves considering prolonged air leakage (>5 days) and cavity at 4^th PO (≥20%). CI, confidence interval; HR, hazard ratio; PO, postoperative; RPS, residual pleural space.

A total of 14 patients (10.4%) needed pleural reintervention (videothoracoscopy, open window thoracostomy or thoracomioplasty). This number represents 73.7% of the total number of empyema cases. The presence of RPS represented a relative risk of 4.42 (95% CI: 1.03–10.97; P=0.02) for reoperation.

Discussion

Postoperative RPS is a frequent and difficult-to-manage question in the thoracic surgeon’s routine. It is frequently correlated to postoperative complications like pleural infection, prolonged air-leak and is even more expressive in non-neoplastic lung resection population (11-13). In this cohort we ratify with solid data, those concerns by demonstrating an incidence of 58% of RPS with an increased risk on develop postoperative pleural empyema. However, it is also important to define an objective approach for measurement and quantification of the RPSP cavity. Therefore, we describe a simple method of radiographic analysis, that demonstrated a high grade of interobserver agreement. With this method, it was possible to define a cut-off point that represents the highest independent risk for the development of postoperative pleural empyema: a RPS ≥20% of the total pleural space area.

Solak et al. in 2007 published one of the few papers specifically directed to the definition and study of RPS (4). In a population of 140 patients undergoing lung resection, it was observed that 44.1% presented a RPS with a resolution rate of 37.9% by the 7^th postoperative day. Those numbers were close to the observed in our paper with a slightly higher incidence of RPS (58%), that could be justified by differences in study population. Solak’s study population consisted mostly of patients with neoplastic disease or pulmonary emphysema. When evaluating only the group operated on for infectious disease (16 patients), 50% had RPS in the first postoperative day and 37.5% persisted with cavity beyond the 7^th day. In this group, 66.6% had complications, all of them pleuropulmonary infections. Our cohort also showed an expressive incidence of infectious pleural complications in that population (20.5%) ratifying the importance of RPS in the post-resection process by a larger number of patients (n=78). Misthos et al. published another cohort analysis with 966 patients (14). From those 92 (9.5%) developed RPS. The risk factors were upper lobectomies, lower segmentectomies, and the RPS was more frequently identified on malignant disease indication, right side, and apical position (P<0.001). The major incidence on malignant disease etiology could be explained by the lack of detailed information about the benign disease group. There was no discrimination about the exact etiology. Patients operated by tuberculosis and by emphysema represents completely different scenarios of benign disease in with regard on RPS. Since the process of pleural rearrangement depends on mediastinal shift, residual lung hyperinflation, and hemidiaphragm elevation, patients operated by aspergilloma and tuberculosis presents an unfavorable scenario due to the chronic inflammatory mediastinal process, pleural thickening, and remaining lung without a tendency to hyperinflation as seen in patients with pulmonary emphysema.

Standard treatment for RPS consists of chest drainage, continuous suction, chest physiotherapy, and bronchoscopic evaluations when necessary. Different tactics to avoid prolonged air fistula and to obliterate the pleural space after resection have already been described, ranging from minimally invasive and reversible techniques such as intermittent phrenic nerve block, to more complex and definitive techniques such as thoracoplasty simultaneously with pulmonary resection (15,16). Thus, the decision-making between maintaining standard clinical treatment and more invasive measures is taken anecdotally and is extremely variable between different specialized centers. Despite Misthos et al. reported a high incidence of pleural infection (40%) and prolonged air-leak (52%) in patients with RPS (14), most of patients (68.5%) had a favorable evolution without the need of invasive procedures. In our cohort, 66% of patients with RPSP had an uneventful recovery sowing that RPS is a benign condition in major part of cases. However, 15.4% of patients with RPS needed a reintervention for pleural complication. Therefore, it is necessary to better divide that group of patients and identify those that are subjected to a higher risk. In our cohort, patients with an ARPS ≥20% than the total pleural space represented this subgroup of higher risk patients. Misthos et al. also stratified the RPS patients into small and large size cavity without define objectively a cut-off point for that definition (14). The patients with large size cavity presented a longer drainage period and a higher rate of empyema. In possession of objective data to define a greater risk group, it is possible to evaluate the early introduction of less invasive methods, such as intermittent phrenic nerve blockade by transcutaneous catheter, or pneumoperitoneum in order to reduce the volume of the cavity and, consequently, the risk of infectious pleural complications (15,17).

The use of chest radiography for this evaluation is a limitation for this study. The chest X-ray is an affordable and easily available exam, but especially subject to variations due to angulation and patient position (18,19). However, performing repeated chest CT scans during the postoperative period is unfeasible and unjustifiable in most cases. To make the method usable in the daily practice of the thoracic surgeon, chest X-ray becomes the most intelligent option. To settle its limitations, we do not use absolute measurement values, but the proportion between two dimensions that are subject to the same positional variations. The measurement protocol designed also aimed to standardize the process, reducing the variability from the analysis process. By following the measurement presets, it was possible to reduce the variability between different observers, reaching a high degree of agreement (kappa =0.77) and allowing its application at the bedside with greater reliability. The development of this method and its precision is critical to provide the surgeon with valuable information as to which RPS dimension is of greatest concern. The generalization of these results applies only to the population composed by infectious etiologies operated patients. Neoplastic patients have a lower incidence of RPS since the absence of known risk factors. However, with the advent of immunotherapy and neoadjuvant treatment for locally advanced cases, the number of patients operated with pleural thickness and mediastinal inflammation probably will increase, and possibly the rate of RPS also (20). Considering the similar pathophysiology regarding the pleural space, they could take benefit of the same results founded in that study. Nevertheless, studies with the neoplastic population are necessary to confirm that similarity.

Conclusions

We describe a new radiographic assessment method capable of being used at daily routine to identify postoperative RPS. The identification of RPS through this method proved to be an important risk factor for infectious pleuropulmonary complications. This information serves as a basis for greater attention in postoperative care and opens the doors for the development of new studies that can help to define control measures of RPS and prevention of postoperative empyema.

Acknowledgments

This article was presented at the 30th European Conference on General Thoracic Surgery of the European Society of Thoracic Surgeons, The Hague, The Netherlands, 19 June–21 June 2022.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2024-2022/rc

Data Sharing Statement: Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2024-2022/dss

Peer Review File: Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2024-2022/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2024-2022/coif). The 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 ethics committee of University of Sao Paulo (No. 4.512.453) 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. Barker WL. Natural history of residual air spaces after pulmonary resection. Chest Surg Clin N Am 1996;6:585-613.
2. Wareham EE, Barber H, McGoey JS, et al. The persistent pleural space following partial pulmonary resection. J Thorac Surg 1956;31:593-600.
3. Bell JW. Management of the postresection space in tuberculosis. III. Role of pre- and postresection thoracoplasty. J Thorac Surg 1956;31:580-92.
4. Solak O, Sayar A, Metin M, et al. Definition of postresectional residual pleural space. Can J Surg 2007;50:39-42.
5. Özkan S, Yazıcı U, Ertan Aydın E, et al. Risk Factors in Development of Postoperative Empyema. J Clin Anal Med 2014;5:19-24.
6. Radu DM, Jauréguy F, Seguin A, et al. Postoperative pneumonia after major pulmonary resections: an unsolved problem in thoracic surgery. Ann Thorac Surg 2007;84:1669-73. [Crossref] [PubMed]
7. Horan TC, Gaynes RP, Martone WJ, et al. CDC definitions of nosocomial surgical site infections, 1992: a modification of CDC definitions of surgical wound infections. Infect Control Hosp Epidemiol 1992;13:606-8.
8. Nagasaki F, Flehinger BJ, Martini N. Complications of surgery in the treatment of carcinoma of the lung. Chest 1982;82:25-9. [Crossref] [PubMed]
9. Imperatori A, Nardecchia E, Dominioni L, et al. Surgical site infections after lung resection: a prospective study of risk factors in 1,091 consecutive patients. J Thorac Dis 2017;9:3222-31. [Crossref] [PubMed]
10. Rhea JT, DeLuca SA, Greene RE. Determining the size of pneumothorax in the upright patient. Radiology 1982;144:733-6. [Crossref] [PubMed]
11. Shapiro M, Swanson SJ, Wright CD, et al. Predictors of major morbidity and mortality after pneumonectomy utilizing the Society for Thoracic Surgeons General Thoracic Surgery Database. Ann Thorac Surg 2010;90:927-34; discussion 934-5. [Crossref] [PubMed]
12. Mariani AW, Vallilo CC, de Albuquerque ALP, et al. Preoperative evaluation for lung resection in patients with bronchiectasis: should we rely on standard lung function evaluation? Eur J Cardiothorac Surg 2021;59:1272-8. [Crossref] [PubMed]
13. Petrella F, Rizzo S, Bertolaccini L, et al. The “Balloon-Like” Sign: Differential Diagnosis between Postoperative Air Leak and Residual Pleural Space: Radiological Findings and Clinical Implications of the Young-Laplace Equation. Cancers (Basel) 2022;14:3533. [Crossref] [PubMed]
14. Misthos P, Kokotsakis J, Konstantinou M, et al. Postoperative residual pleural spaces: characteristics and natural history. Asian Cardiovasc Thorac Ann 2007;15:54-8. [Crossref] [PubMed]
15. Patella M, Saporito A, Mongelli F, et al. Management of residual pleural space after lung resection: fully controllable paralysis of the diaphragm through continuous phrenic nerve block. J Thorac Dis 2018;10:4883-90. [Crossref] [PubMed]
16. Krasnov DV, Skluev SV, Petrova YK, et al. Modern Collapse Therapy for Pulmonary Tuberculosis. Thorac Surg Clin 2019;29:47-58. [Crossref] [PubMed]
17. Pecoraro A, Garbarino GM, Peritore V, et al. Early Induction of Bedside Pneumoperitoneum in the Management of Residual Pleural Space and Air Leaks After Pulmonary Resection. World J Surg 2021;45:624-30. [Crossref] [PubMed]
18. Galetin T, Defosse J, Schieren M, et al. Sensitivity of chest ultrasound for postoperative pneumothorax in comparison to chest X-ray after lung resecting surgery. Eur J Cardiothorac Surg 2020;57:846-53. [Crossref] [PubMed]
19. Porter ED, Fay KA, Hasson RM, et al. Routine Chest X-Rays After Thoracic Surgery Are Unnecessary. J Surg Res 2020;250:188-92. [Crossref] [PubMed]
20. Forde PM, Spicer J, Lu S, et al. Neoadjuvant Nivolumab plus Chemotherapy in Resectable Lung Cancer. N Engl J Med 2022;386:1973-85. [Crossref] [PubMed]