Early versus late chest tube removal after surgery for primary spontaneous pneumothorax—a systematic review and meta-analysis

Introduction

Surgical intervention is recommended in patients with recurrent or persistent primary spontaneous pneumothorax (PSP) (1). The preferred surgical strategy is bullectomy (i.e. treating the underlying cause) followed by pleurodesis (i.e., preventing further recurrence) through video-assisted thoracoscopic surgery (VATS) (1,2). Pleurodesis can be achieved chemically and mechanically by pleurectomy or pleural abrasion (1,3,4).

Unfortunately, recommendations regarding postoperative chest tube management are lacking, resulting in practice variation, especially concerning chest tube duration (1,5). This variability may impact recurrence rate, postoperative length of stay (LOS), postoperative pain and patient satisfaction (6,7). Traditionally, postoperative chest tubes are left in place for a fixed period, irrespective of the amount of air leakage or fluid production. A previous review investigated the timing of chest tube removal and highlighted significant discrepancies in chest tube management protocols among studies with LOS directly influenced by chest tube duration (8).

Recent studies showed that chest tube removal on the same day of surgery is safe with a mean postoperative LOS of one day in patients operated for PSP (9,10). In view of a growing interest in early chest tube removal and enhanced recovery after thoracic surgery (ERATS) protocols, in which chest tube management is an important factor (11), a systematic evaluation of the timing of postoperative chest tube removal is warranted. Therefore, the aim of this systematic review and meta-analysis is to evaluate the recurrence rate as safety outcome of early (i.e., direct removal upon cessation of air leakage) versus late (after a fixed number of days) chest tube removal. We present this article in accordance with the PRISMA reporting checklist (available at https://jtd.amegroups.com/article/view/10.21037/jtd-24-1802/rc).

Methods

This study was registered in PROSPERO (ID: CRD42022324988) on 1 March, 2022. The research question guiding this review is: ‘What is the recurrence rate of pneumothorax in patients undergoing VATS for PSP with early versus late chest tube removal?’

The definition of early chest tube removal was direct tube removal upon cessation of air leakage. The criteria for the late removal group were chest tube removal after a fixed minimal time period and/or the provision of pleural fluid production <200 mL/24 hours.

Approval by the institutional board and individual consent was waived due to the nature of the study as a Systematic Review and Meta-analysis.

Eligibility criteria

Clinical (non-)randomized trials and observational studies describing chest tube management in patients undergoing VATS for PSP were considered eligible. Articles referring to chest tube management were only eligible if they provided a clear description of the chest tube protocol, surgical technique and if the recurrence rate was reported with either a minimum follow-up per patient of six months or a mean follow-up of 12 months.

Search strategy

Studies were identified in MEDLINE (PubMed), EMBASE and Cochrane. There was no time limit used in the search. The search strategy contained a combination of index terms and free text words related to pneumothorax and surgical treatment for pneumothorax to find all relevant articles regarding postoperative chest tube management. The complete search strategy is detailed in Table S1. The final search was conducted on January 29, 2024.

Study selection

Two reviewers (Q.C.A.v.S. and L.N.S.) independently screened the titles, abstracts and full-text of all identified studies. Discrepancies were resolved by consulting the senior author (F.J.C.v.d.B.). Studies were excluded in case of ineligible study population (e.g., >5% secondary spontaneous pneumothorax), surgery by thoracotomy, sample size <20 (considered as small case series with bias), ineligible outcome measures, inadequate study design (e.g., case series, reviews and editorials), not written in English or insufficient data presentation, also after scrutinizing supplemental material. The description of chest tube management should at least involve criteria on air leakage. Articles should also use the same surgical treatment per intervention or control group (in case of comparison). If more information was needed from an article, the corresponding authors were contacted by email. In case of no response, the article was excluded from analysis. If articles reported on several intervention and control groups from which only a selection was eligible according to our criteria, only the eligible study groups were included in our analysis.

Outcome measures

Recurrence rate

The primary outcome measure was the percentage of pneumothorax recurrence. Recurrence was defined as an ipsilateral recurrent pneumothorax, also when this occurred early after chest tube removal, requiring intervention and/or hospital admission. The minimum required follow-up period per patient was six months. When studies mentioned patients lost to follow-up, those patients were not included in the analysis.

Length of hospital stay

The LOS was defined as the number of days the patient remained in the hospital after surgery, with the date of surgery being day 0.

Postoperative prolonged air leakage (PAL)

PAL was defined as persistent air leakage >5 days after surgery. Studies using a different definition for PAL or lacking a definition were excluded from analysis regarding this outcome. PAL was only considered absent if this was specifically mentioned in the article.

Postoperative chest tube duration

Chest tube duration was defined as the number of days the chest tube was in situ with the date of surgery being day 0.

Data collection

Data collection included study design, chest tube management, patient characteristics (age and sex), type of surgery and outcome measures of interest (recurrence rate, LOS, PAL and chest tube duration).

Statistical analysis

All included studies were divided into two groups for comparison, early and late chest tube removal, based on the reported criteria of chest tube management. Continuous data were reported as means with standard deviation (SD) (parametric data) or medians with interquartile range (IQR) (non-parametric data) and categorical data as numbers with percentages. A meta-analysis was performed for the following predetermined outcome measures: recurrence rate, LOS, PAL and chest tube duration. We logit-transformed the outcome and performed univariate random effects meta-analysis using the meta proportion and mean package in R (version 4.3.3). This analysis provided summary estimates which were reported as percentages with 95% confidence intervals (95% CIs) and means with 95% CI. Also, the I² statistics were calculated, presenting the percentage of variability that is attributable to between-study heterogeneity. Significant heterogeneity between studies was indicated when I² was >50% (12). Subgroup differences were tested using the z-test. P<0.05 was considered statistically significant.

Risk of bias (RoB) analysis

The methodological quality of all included studies was independently assessed by two reviewers (Q.C.A.v.S. and L.N.S.) and any disagreement was resolved by discussion with the senior author (F.J.C.v.d.B.). The RoB was evaluated with the Risk of Bias tool (RoB2) for randomized controlled trials (RCTs) and the risk of bias for non-randomized studies for interventions tool (ROBINS-I) for non-RCTs (13,14). The intervention for the evaluation of the RoB concerned early versus late chest tube removal with recurrence rate as outcome measure. (Non-)RCTs with no direct comparison between early and late chest tube removal were assessed as single-arm cohort studies to evaluate RoB precluding the evaluation of bias in the confounding and randomization domain for non-RCTs and RCTs, respectively (Table S2). The Grading Recommendations of Assessment, Development and Evaluation (GRADE) method was used for evaluation of the certainty of the evidence (Tables S3,S4).

Subgroup analysis

Subgroup analysis for recurrence rate was performed for type of surgical pleural intervention (e.g., pleurectomy, abrasion, chemical pleurodesis).

Results

Description of the studies

The search identified 6,442 unique articles, of which 36 were selected for the meta-analysis comprising 9 RCTs and 27 cohort studies [Figure 1, supplementary file (Appendix 1)]. In the included 36 studies, a total of 58 study groups were identified and divided into late (40 groups; 4,837 patients) and early chest tube removal groups (18 groups; 1,329 patients). Study characteristics and RoB assessment are described in Table 1 and Tables S5,S6. Since there were no comparative randomized studies included, all studies were assessed using the ROBINS-1 tool. Our meta-analysis comprised 6,156 patients with PSP and 10 patients with secondary spontaneous pneumothorax, who underwent bullectomy followed by pleurectomy, pleural abrasion or chemical pleurodesis. The mean age was 23.2 years (SD 4.2) in the early removal group and 26.1 years (SD 3.3) in the late removal group (P<0.05). Of all included patients 83% was (5,140/6,166) male with no significant differences between the early and late removal groups. Since 103 of 6,166 patients (1.7%) were lost to follow up, 6,063 patients were analyzed regarding the outcome postoperative recurrence.

Figure 1 PRISMA 2020 flow diagram of study selection. VATS, video-assisted thoracoscopic surgery; PSP, primary spontaneous pneumothorax.

Outcome measures

Recurrence rate

The recurrence rate was 4.49% (95% CI: 3.33–6.03%; I²=65.6%) in the late and 7.61% (95% CI: 5.44–10.57%; I²=8.2%) in the early chest tube removal group (P=0.02) (Figure 2). The postoperative follow-up period ranged from 12–94 months in the late and 6–96 months in the early removal group. The certainty of the evidence was considered low in both groups (Table S3).

Figure 2 Random effects meta-analysis of the recurrence rate after video-assisted thoracoscopic surgery pleurodesis after early and late chest tube removal. Definitions of study groups 1, 2, 3 can be found in Table S5. CI, confidence interval; I², heterogeneity; Intervention, the researched intervention of the included study; n Rec, number of recurrences; n Total, total number of patients with complete follow-up.

Length of hospital stay

The LOS was analyzed in 48 of the 58 study groups (35 of 40 late and 13 of 18 early tube removal groups). The mean was 4.83 days (95% CI: 4.32–5.39; I²=98.6%) in the late and 4.38 days (95% CI: 4.02–4.78; I²=98.4%) in the early chest tube removal groups (P=0.18) (Table 2; Figure S1). The certainty of the evidence was considered low for the late removal groups and moderate for the early removal groups (Table S3).

PAL

PAL (>5 days) was analyzed in 17 of the 58 study groups (14 of 40 late and 3 of 18 early tube removal groups). The overall incidence of PAL was 6.12% (95% CI: 4.65–8.01%; I²=48%) in the late and 4.35% (95% CI: 1.82–10.02%; I²=0%) in the early chest tube removal groups (P=0.45) (Table 2; Figure S2). The certainty of the evidence was considered moderate for the late removal groups and low for the early removal groups (Table S3).

Chest tube duration

The postoperative chest tube duration was analyzed in 40 of the 58 study groups (33 of 40 late and 7 of 18 early tube removal groups). The mean was 3.42 days (95% CI: 3.08–3.81; I²=98%) in the late and 2.50 days (95% CI: 2.31–2.71; I²=97.6%) in the early chest tube removal groups (P<0.001) (Table 2; Figure S3). The certainty of the evidence was considered low in both groups (Table S3).

Subgroup analysis

Surgical pleural intervention

There was only a significant difference in recurrence rate between early and late chest tube removal after pleurectomy (P=0.003) (Table 3). No differences in recurrence rates were seen after pleural abrasion and chemical pleurodesis (Table 3; Figures S4,S5).

Discussion

Given the growing interest in and clinical impact of early chest tube removal, due to implementation of ERATS strategies, it is important to provide the best available evidence regarding the safety of early chest tube removal in patients surgically treated for PSP. The lack of any study directly comparing early versus late chest tube removal in this patient group precludes the highest level of evidence as assessed by the GRADE method. We attempted to still address this important knowledge gap by including only studies with a very clear description of their chest tube removal criteria and transparent description of recurrence and follow-up. This methodology provides us with the best available evidence, in order to provide surgeons guidance in chest tube management. Although based on low level of evidence, our study indicates that early chest tube removal may be associated with increased postoperative recurrences (P=0.02), underscoring the need for caution when adopting this approach. Although in general ERATS guidelines advocate early chest tube removal, our results warrant a prospective trial comparing early and late chest tube removal before widespread implementation.

A previous descriptive review on timing of chest tube removal also manifested different chest tube management protocols (8). They concluded that chest tube removal as early as postoperative day two appears safe since no postoperative complications occurred and chest radiographs were satisfactory. Different from this narrative review, our systematic review focused on early chest tube removal (i.e., upon cessation of air leak) and its effect on postoperative recurrence rate, which is the most important safety outcome in such evaluation. It is worth noting that factors beyond chest tube management, such as type of surgical pleurodesis and patient factors, may also influence recurrence rates, underscoring the complexity of this analysis.

Although surgical pleurodesis is a necessary step in recurrence prevention, pleurodesis techniques may vary and therewith exert different risks of recurrence, which has previously been explored by two reviews reporting the lowest recurrence rates after VATS chemical pleurodesis (15,16). Therefore, we performed subgroup analyses regarding pleurodesis technique showing similar recurrence rates in the VATS pleural abrasion and VATS chemical pleurodesis group with or without pleural abrasion. Interestingly, after pleurectomy the recurrence rate was 6% lower in the late chest tube removal group compared to the early removal group (P=0.003), suggesting it may be safer leaving the chest tube in place for a few days in these cases. Although the number of pleurectomy cases in the early group was limited, this pleurodesis technique may be considered less effective, potentially requiring a longer chest tube duration. To our knowledge, the safety of chest tube removal being dependent on the type of pleurodesis has not previously been explored and may be an interesting avenue for future research.

Although early chest tube removal was expected to result in a shortened LOS, this meta-analysis found a non-significant (P=0.18) reduction in LOS of only half a day. Factors that potentially obscured a reduction in LOS by early tube removal are the retrospective nature of most included studies, high heterogeneity, variation in discharge criteria and different study aims. Next to chest tube duration, which was a day shorter in the early group (P<0.001), the ERATS guideline outlines 44 additional crucial items for enhanced recovery, all of which impact LOS (11,17). Since ERATS items are scarcely reported in studies and it was not our aim to correct for all ERATS items in our review, any potential difference in LOS due to early or late chest tube removal could have been missed. Furthermore, some studies specifically evaluating early chest tube removal and its effect on LOS were excluded from our review, since their outcomes did not adhere to our inclusion criteria. Furuya for example investigated early chest tube removal showing a mean LOS of only 1 day (9). In this study, chest tube removal was delayed in only 15% of cases due to persistent air leak or bloody drainage, resulting in a mean LOS of two days, which is still lower than the mean LOS found in our review. Another small prospective study even showed safe chest tube removal in the operating room in case of absence of air leakage, with a mean LOS of 1 day (10). Based on these two studies, a shorter LOS after early chest tube removal was expected, whereas our review could not confirm these findings.

Next to an effect on recurrence and LOS, there may be an interaction between chest tube duration and PAL. Although the presence of a chest tube in itself is not a risk factor for PAL, chest tube duration and/or the level of suction on the drainage system may influence its persistence. We demonstrated a comparable postoperative incidence of 4–6% after early and late chest tube removal (P=0.45). This is in accordance with previous studies with reported incidences of PAL up to 8% (6,8). Between studies the amount of applied suction varied from −5 to −25 cmH₂O based on in-house protocols and incomplete lung expansion or active air leakage (18-20). Interestingly, a retrospective study showed that water seal following VATS pleurodesis was associated with reduced chest tube duration and LOS, suggesting that the use of less or no suction could be beneficial (21). This is in line with results from a RCT in lobectomy cases, demonstrating significantly decreased chest tube duration after suction with −2 cmH₂O compared to −10 cmH₂O (22). The underlying physiology puts forward that excessive suction may maintain the source of air leakage.

Although our findings advocate caution regarding early chest tube removal, we must acknowledge several limitations. First, the absence of studies directly comparing early with late tube removal in patients with PSP necessitated inclusion of studies solely based on the accurate description of our outcome measures of interest and tube management criteria, which does not automatically imply that these criteria were strictly adhered to. For instance, only one study explicitly specified the method and device used to determine the cessation of air leakage (23). Seven additional studies provided clear descriptions or visualizations of the devices employed, of which 2 were digital and 5 analog (20,24-29). While digital drainage systems enable objective assessment of air leakage, the use of analog devices may result in a more subjective appraisal, potentially influencing chest tube duration. Moreover, variations in the reported removal criteria as well as in perioperative care, such as surgical approach and the number of chest tubes used, introduced heterogeneity across studies. These differences cannot be fully accounted for, thereby limiting the robustness of our recommendations. Secondly, the quality of all studies, assessed with the ROBINS-1 tool, was scored in the majority as having moderate RoB. Consequently, owing to the heterogeneous nature across studies and the moderate to high RoB, the overall certainty of the evidence according to the GRADE method was considered low. Finally, potential benefits of early chest tube removal lie within other outcome measures such as improved patient satisfaction and decreased postoperative pain, which are scarcely reported and hence must be confirmed by high-quality research (7,30). These limitations prevent drawing firm conclusions regarding postoperative chest tube management, and therefore a newly designed RCT is established focusing on the safety of postoperative chest tube and pain management in patients operated for PSP; the Pneumotrial (clinicaltrials.gov; NCT06053476).

Conclusions

This meta-analysis indicates that early chest tube removal following VATS pleurodesis for PSP may be associated with a slightly higher recurrence rate, emphasizing the need for caution when considering this approach. Although the quality of evidence is considered low due to lack of direct comparative studies, this review provides the best available evidence to guide clinicians in chest tube decision-making. Our findings furthermore indicate to maintain a conservative approach regarding chest tube removal after VATS pleurodesis by pleurectomy. A prospective RCT should be conducted to obtain reliable and definite insight into the safety and benefits of early chest tube removal.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the PRISMA reporting checklist. Available at https://jtd.amegroups.com/article/view/10.21037/jtd-24-1802/rc

Peer Review File: Available at https://jtd.amegroups.com/article/view/10.21037/jtd-24-1802/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-24-1802/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.

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. MacDuff A, Arnold A, Harvey J, et al. Management of spontaneous pneumothorax: British Thoracic Society Pleural Disease Guideline 2010. Thorax 2010;65:ii18-31. [Crossref] [PubMed]
2. van Steenwijk QCA, Spaans LN, Heineman DJ, et al. Variation in perioperative care for recurrent primary spontaneous pneumothorax: a Dutch survey. Ann Laparosc Endosc Surg 2023;8:9. [Crossref]
3. Tschopp JM, Bintcliffe O, Astoul P, et al. ERS task force statement: diagnosis and treatment of primary spontaneous pneumothorax. Eur Respir J 2015;46:321-35. [Crossref] [PubMed]
4. Henry M, Arnold T, Harvey J, et al. BTS guidelines for the management of spontaneous pneumothorax. Thorax 2003;58:ii39-52. [Crossref] [PubMed]
5. Baumann MH, Strange C, Heffner JE, et al. Management of spontaneous pneumothorax: an American College of Chest Physicians Delphi consensus statement. Chest 2001;119:590-602. [Crossref] [PubMed]
6. van Steenwijk QCA, Spaans LN, Heineman DJ, et al. Population-based study on surgical care for primary spontaneous pneumothorax. Eur J Cardiothorac Surg 2024;65:ezae104. [Crossref] [PubMed]
7. Refai M, Brunelli A, Salati M, et al. The impact of chest tube removal on pain and pulmonary function after pulmonary resection. Eur J Cardiothorac Surg 2012;41:820-2; discussion 823. [Crossref] [PubMed]
8. Dearden AS, Sammon PM, Matthew EF. In patients undergoing video-assisted thoracic surgery for pleurodesis in primary spontaneous pneumothorax, how long should chest drains remain in place prior to safe removal and subsequent discharge from hospital? Interact Cardiovasc Thorac Surg 2013;16:686-91. [Crossref] [PubMed]
9. Furuya T, Ii T, Yanada M, et al. Early chest tube removal after surgery for primary spontaneous pneumothorax. Gen Thorac Cardiovasc Surg 2019;67:794-9. [Crossref] [PubMed]
10. Igai H, Matsuura N, Numajiri K, et al. Feasibility of tubeless thoracoscopic bullectomy in primary spontaneous pneumothorax patients. Gen Thorac Cardiovasc Surg 2023;71:138-44. [Crossref] [PubMed]
11. Draeger TB, Gibson VR, Fernandes G, et al. Enhanced Recovery After Thoracic Surgery (ERATS). Heart Lung Circ 2021;30:1251-5. [Crossref] [PubMed]
12. Higgins JP, Thompson SG. Quantifying heterogeneity in a meta-analysis. Stat Med 2002;21:1539-58. [Crossref] [PubMed]
13. Sterne JA, Hernán MA, Reeves BC, et al. ROBINS-I: a tool for assessing risk of bias in non-randomised studies of interventions. BMJ 2016;355:i4919. [Crossref] [PubMed]
14. Sterne JAC, Savović J, Page MJ, et al. RoB 2: a revised tool for assessing risk of bias in randomised trials. BMJ 2019;366:l4898. [Crossref] [PubMed]
15. Sudduth CL, Shinnick JK, Geng Z, et al. Optimal surgical technique in spontaneous pneumothorax: a systematic review and meta-analysis. J Surg Res 2017;210:32-46. [Crossref] [PubMed]
16. Sim SKR, Nah SA, Loh AHP, et al. Mechanical versus Chemical Pleurodesis after Bullectomy for Primary Spontaneous Pneumothorax: A Systemic Review and Meta-Analysis. Eur J Pediatr Surg 2020;30:490-6. [Crossref] [PubMed]
17. Batchelor TJP, Rasburn NJ, Abdelnour-Berchtold E, et al. Guidelines for enhanced recovery after lung surgery: recommendations of the Enhanced Recovery After Surgery (ERAS®) Society and the European Society of Thoracic Surgeons (ESTS). Eur J Cardiothorac Surg 2019;55:91-115. [Crossref] [PubMed]
18. Chen JS, Hsu HH, Chen RJ, et al. Additional minocycline pleurodesis after thoracoscopic surgery for primary spontaneous pneumothorax. Am J Respir Crit Care Med 2006;173:548-54. [Crossref] [PubMed]
19. Chen JS, Hsu HH, Huang PM, et al. Thoracoscopic pleurodesis for primary spontaneous pneumothorax with high recurrence risk: a prospective randomized trial. Ann Surg 2012;255:440-5. [Crossref] [PubMed]
20. Min X, Huang Y, Yang Y, et al. Mechanical pleurodesis does not reduce recurrence of spontaneous pneumothorax: a randomized trial. Ann Thorac Surg 2014;98:1790-6; discussion 1796. [Crossref] [PubMed]
21. Wagner G, Asban A, Xie R, et al. Early Water Seal of Chest Tubes Following Video-Assisted Thoracic Surgery Pleurodesis. J Surg Res 2023;283:1033-7. [Crossref] [PubMed]
22. Holbek BL, Christensen M, Hansen HJ, et al. The effects of low suction on digital drainage devices after lobectomy using video-assisted thoracoscopic surgery: a randomized controlled trial†. Eur J Cardiothorac Surg 2019;55:673-81. [Crossref] [PubMed]
23. Zhang D, Miao J, Hu X, et al. A clinical study of efficacy of polyglycolic acid sleeve after video-assisted thoracoscopic bullectomy for primary spontaneous pneumothorax. J Thorac Dis 2017;9:1093-9. [Crossref] [PubMed]
24. Fung S, Ashmawy H, Safi SA, et al. Effectiveness of Video-Assisted Thoracoscopic Surgery with Bullectomy and Partial Pleurectomy in the Treatment of Primary Spontaneous Pneumothorax-A Retrospective Long-Term Single-Center Analysis. Healthcare (Basel) 2022;10:410. [Crossref] [PubMed]
25. Mithiran H, Leow L, Ong K, et al. Video-Assisted Thoracic Surgery (VATS) Talc Pleurodesis Versus Pleurectomy for Primary Spontaneous Pneumothorax: A Large Single-Centre Study with No Conversion. World J Surg 2019;43:2099-105. [Crossref] [PubMed]
26. Gossot D, Galetta D, Stern JB, et al. Results of thoracoscopic pleural abrasion for primary spontaneous pneumothorax. Surg Endosc 2004;18:466-71. [Crossref] [PubMed]
27. Marcheix B, Brouchet L, Renaud C, et al. Videothoracoscopic silver nitrate pleurodesis for primary spontaneous pneumothorax: an alternative to pleurectomy and pleural abrasion? Eur J Cardiothorac Surg 2007;31:1106-9. [Crossref] [PubMed]
28. Lin JB, Chen JF, Lai FC, et al. Transareolar pulmonary bullectomy for primary spontaneous pneumothorax. J Thorac Cardiovasc Surg 2016;152:999-1005. [Crossref] [PubMed]
29. Jung HS, Kim HJ. Simultaneous Viscum pleurodesis and video-assisted thoracic surgery (VATS) bullectomy in patients with primary spontaneous pneumothorax. Sci Rep 2021;11:22934. [Crossref] [PubMed]
30. Batchelor TJP. Enhanced recovery after surgery and chest tube management. J Thorac Dis 2023;15:901-8. [Crossref] [PubMed]