A retrospective cohort study of operated older patients undergoing lung resection after PFT-based versus CPET-based preoperative stratification

Introduction

Current epidemiology shows that lung cancer remains the most prevalent cancer worldwide (1). Affected patients are typically older with frequent cardiovascular or chronic obstructive pulmonary disease (COPD), often resulting in impaired pulmonary function. While surgery remains a key treatment, it poses significant morbidity/mortality risks in those with severe comorbidities or inadequate cardiopulmonary reserve (2). A reduction in lung ventilation or diffusion function increases the risk of perioperative complications (3). Beyond tumor-stage-appropriate resection planning, preoperative pulmonary function evaluation is crucial for assessing surgical risk. We hope to safely increase the number of surgical patients by examining cardiopulmonary exercise testing (CPET).

Pulmonary function tests (PFTs) are the primary preoperative assessment in thoracic surgery, providing objective data to predict perioperative complication risks and long-term outcomes (4-6), However, surgery imposes significant physiological/metabolic demands requiring dynamic cardiopulmonary evaluation. Static PFTs inadequately reflect actual cardiorespiratory fitness, warranting dynamic assessment. CPET quantifies integrated respiratory, cardiovascular, and metabolic responses during exercise-recovery cycles, objectively evaluating functional reserve and impairment severity. The 2016 European Society of Cardiology (ESC) guidelines recognize CPET’s prognostic value for perioperative risk stratification across surgical specialties. (2,7-9).

Postoperative complications are the primary cause of prolonged hospital stays, increased healthcare costs, and higher mortality rates (10). Numerous studies have identified risk factors associated with postoperative complications, especially pulmonary complications (11,12). This study evaluates CPET’s safety as a preoperative adjunct to pulmonary function by analyzing associations between these assessments and postoperative outcomes. We present this article in accordance with the STROBE reporting checklist (available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2026-1-0180/rc).

Methods

Ethics statement

This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. This retrospective study was approved by the Institutional Review Board of Taizhou Hospital of Zhejiang Province Affiliated to Wenzhou Medical University (No. K20241102), and informed consent from the included patients was waived due to the retrospective nature of the study.

Patients

This retrospective cohort analysis included patients >65 years undergoing video-assisted thoracoscopic surgery (VATS) at Taizhou Hospital (September 2022 to March 2024). Eligibility required preoperative PFTs. Patients whose PFTs suggested they could not tolerate the surgery were further evaluated with CPET. If the results suggested that the surgery could be tolerated, then VATS was performed. Using chest three-dimensional computed tomography (CT) guidance, patients’ postoperative (PPO)-forced expiratory volume in one second (FEV1) and PPO-diffusing lung carbon monoxide (DLCO) were measured. PPO-FEV1 and PPO-DLCO were calculated using the segmental counting method based on three-dimensional CT reconstruction. The number of functional segments to be preserved was estimated, and the PPO value was derived as: PPO (%) = (number of preserved segments / total number of segments) × preoperative value. Suboptimal PFT (defined as either PPO-FEV1 <30% or PPO-DLCO <30%) triggered CPET referral. For patients undergoing CPET assessment, surgical procedures will proceed as planned if their PPO-maximum oxygen uptake (VO₂ peak) exceeds either 10 mL/kg/min or 35% of predicted value; otherwise, the surgery will be canceled. Participants were stratified into PFT group (direct surgical clearance via PFT) and CPET group (surgery permitted after CPET reassessment). The final cohort (n=130) provided complete perioperative demographic, diagnostic, and therapeutic data. Among the 20 patients who did not undergo surgery, 9 cases (45%) were excluded due to multiple cardiovascular risk factors (e.g., coronary artery disease), while the remaining 11 cases (55%) had their surgery canceled because of poor CPET performance.

All patients underwent VATS. The extent of lung resection was classified into three types: (I) lobectomy: removal of an entire lobe of the lung; (II) segmentectomy: anatomical removal of one or more bronchopulmonary segments, preserving the remaining segments of the same lobe; (III) wedge resection: non-anatomical peripheral resection of a small, wedge-shaped portion of the lung. For the purpose of analysis, segmentectomy and wedge resection were collectively considered as sublobar resection.

PFTs

The PFT is performed with a spirometer (MasterScreen, Jaeger, Germany) and included the following parameters: forced vital capacity (FVC), forced lung volume as a percentage of predicted value (FVC%-pred), FEV1, forced expiratory volume at the first second as a percentage of predicted value (FEV1%-pred), maximal voluntary ventilation (MVV), etc. The test was performed under the supervision of a physician, with the average value recorded. The predicted values followed the European Respiratory Society (ERS)-93 standards (13).

CPETs

CPET was conducted via QUARK PFT ERGO system (COSMED, Italy) with pre-test calibration for gas circulation and ambient humidity. Patients wore a fitted mask connected to the analyzer for symptom-limited maximal exercise testing using individualized Ramp protocols (10–20 W/min) (14). Respiratory gas exchange, 12-lead ECG, and blood pressure were continuously monitored during rest, exercise, and 3-minute unloaded recovery. Cycling was maintained at 60±5 rpm until symptom limitation, defined as inability to sustain revolutions per minute (RPM) due to dyspnea/fatigue; chest discomfort/dizziness; saturation of peripheral oxygen (SpO₂) <88%; or electrocardiogram (ECG) showing ST-segment elevation >0.2 mV or horizontal/downsloping depression >0.2 mV (15). Tests concluded after 3-minute rest if no abnormalities occurred.

Study endpoints

All 130 operated patients completed 30-day follow‑up with no loss to followup. The primary endpoint was postoperative complications, defined as any complications occurring within 30 days of surgery. Complication severity was assessed using the Clavien-Dindo classification (16). Pulmonary infection was prespecified as a secondary endpoint, defined as the presence of new pulmonary infiltrates on chest imaging plus at least two of the following: fever >38 ℃, leukocytosis (>10,000/µL), purulent sputum, or positive microbiological culture. Secondary endpoints included operative time (skin incision to closure), intraoperative blood loss, postoperative hospital stay, chest tube duration, total hospital stay, and intensive care unit (ICU) admission.

Statistical analysis

Propensity score-matched (PSM) analysis was performed to control for baseline characteristics between the 2 groups using the ‘MatchIt’ package in R (version 4.4.0). Each patient’s propensity score was calculated from a multivariate logistic regression model with covariates, including age, body mass index, gender, smoking history, hypertension, diabetes, coronary heart disease, cerebrovascular disease, Chronic Obstructive Pulmonary Disease (COPD), TNM staging and tumour location. Patients in the two groups were matched 1:1 using the nearest-neighbor method with a caliper of 0.2.

Categorical variables were compared using Pearson’s χ² test or Fisher’s exact test. Normally distributed continuous variables were expressed as mean ± standard deviation (SD) and compared using independent t-tests. Non-normally distributed variables were reported as medians [interquartile range (IQR)] and analyzed with the Mann-Whitney U test. Count data were presented as n (%) and compared using χ² or Fisher’s tests. All analyses were performed in R (v4.4.0).

Results

Patients

Between September 2022 and March 2024, a total of 252 patients underwent preoperative PFTs at our hospital, and 130 patients were included in the final study (Figure 1). The baseline characteristics of the patients are shown in Table 1. There were 71 patients (54.62%) in the PFT group [age, 71.76±7.27 years; 45 males (63.38%)] and 59 patients (45.38%) in the CPET group [age, 74.61±5.47 years; 46 males (77.97%)]. The CPET group was significantly older (P=0.02), had a higher proportion of smokers (P=0.01), and showed higher rates of COPD and hypertension, as well as worse PFT results compared with the PFT group. There were more non-smoking patients in the PFT group (64.79% vs. CPET group). The CPET group had higher proportions of patients with hypertension (49.15%) and COPD (64.41%), with a predominance of right-sided tumor location (64.41%). After PSM, 38 sets of matched pairs were generated. Standardized mean difference (SMD) <0.10 indicates that the characteristics of the two groups of patients were well balanced.

Figure 1 Flow diagram of the study participants. CPET, cardiopulmonary exercise test; DLCO, diffusing lung carbon monoxide; FEV1, forced expiratory volume in one second; PFT, pulmonary function test; PPO, postoperative.

PFTs

The PFTs results are shown in Table 2. With the exception of FEV1% pred [2.16 (IQR, 1.94–2.40) vs. 2.29 (IQR, 1.99–2.46) L; P=0.14], all other parameters differed significantly between groups, with the CPET group showing worse pulmonary function than the PFT group.

Among patients who underwent CPET, 59 proceeded to surgery, while 20 did not. Of the non-surgical patients, 9 (45%) were excluded due to multiple perioperative risk factors (e.g., coronary artery disease, vertebral artery stenosis), while the remainder were excluded based on poor CPET performance. As shown in Table 3, CPET parameters differed significantly between groups. The surgical group had higher PPO-FEV1 [43.71% (IQR, 33.60–49.12%) vs. 30.27% (IQR, 25.36–42.58); P=0.02], PPO-DLCO [61.00% (IQR, 46.00–73.00%) vs. 54.00% (IQR, 36.50–59.50%); P=0.006] and PPO-VO₂ peak% [56.44% (IQR, 50.92–62.86) vs. 39.38% (IQR, 31.22–46.52); P=0.009] than the non-surgical group. Lower PPO-VO₂ peak% is the main reason why surgeons do not choose surgery.

Perioperative outcomes

Perioperative and pathological outcomes are listed in Table 4. No 30- or 90-day mortality occurred. A total of 32 complications were observed, with 23 cases (71.88%) classified as Clavien-Dindo grade I–II. Complication rates were comparable between groups: 17 patients (23.94%) in the PFT group vs. 15 patients (25.42%) in the CPET group (P=0.76, Table 3). Complication severity (grade I–II vs. III–IV) did not differ significantly. The most frequent complication was pulmonary infection, with no intergroup difference in incidence. Rates of hoarseness, atelectasis, pleural effusion, arrhythmia, and respiratory failure were similar. In the PFT group, one case each of myocardial infarction, cerebral infarction, and active thoracic hemorrhage occurred (none required reoperation). Subgroup analyses did not reveal any significant interaction between group assignment and postoperative complications across different strata, including age, sex, body mass index (BMI), smoking status, hypertension, diabetes, COPD, and tumor location (all P for interaction >0.05). The hazard ratios for most subgroups had confidence intervals crossing 1. Due to limited sample sizes in certain subgroups [e.g., coronary heart disease (CHD), TNM stage IV, benign lesions], these findings are exploratory and should be interpreted with caution (Figure 2).

Figure 2 Subgroup analyses of postoperative complications. The vertical dashed line indicates the overall estimated difference. An estimated difference of less than 0 implies a higher complication rate in the CPET group compared to the PFT group. BMI, body mass index; CHD, coronary heart disease; CI, confidence interval; COPD, chronic obstructive pulmonary disease; CPET, cardiopulmonary exercise test; CVD, cerebrovascular disease; HR, hazard ratio; NA, not applicable; PFT, pulmonary function test; TNM, tumor node metastasis.

The CPET group had longer hospital stays [14 (IQR, 10–18) vs. 11 (IQR, 7–16) days; P=0.04], and drain duration [7 (IQR, 4–9) vs. 4 (IQR, 3–9) days; P=0.045]. Postoperative stays [8 (IQR, 5.5–11.5) vs. 6.5 (IQR, 5–10) days; P=0.08] and operative time [135 (IQR, 96–192.5) vs. 118.5 (IQR, 90–145) min; P=0.09] were numerically higher in the CPET group but not statistically significant. Intraoperative bleeding was similar [20 (IQR, 10–30) vs. 20 (IQR, 10–20) mL; P=0.25]. Sublobar resection was more frequent in the CPET group (62.71% vs. 52.11%; P=0.51). Major complications (Clavien-Dindo grade ≥ III) occurred in 6 patients (8.45%) in the PFT group and 3 patients (5.08%) in the CPET group, with no significant difference between groups (P=0.53). Univariate and multivariable logistic regression analyses were performed to identify factors associated with postoperative complications in Table 5. After adjusting for age, sex, smoking, COPD, and hypertension, the CPET group remained not significantly associated with increased complication risk compared with the PFT group [hazard ratio (HR) =1.28, 95% confidence interval (CI): 0.50–3.29, P=0.61].

Pathological outcomes

Pathological results identified 10 benign lesions (7.75%). Most malignancies (74.62%) were stage I lung cancers; one stage IV case occurred in the CPET group. Pathological staging did not differ between groups (P=0.95). Adenocarcinoma was the most common histologic type in both groups, with no significant intergroup difference (P=0.90).

Discussion

Accurate prediction of perioperative risks is critical for surgical planning and informed consent (17). In this study, we performed a retrospective analysis of patients undergoing lung resection who underwent PFTs alone and those who underwent CPET after PFT. Our results demonstrate that no statistically significant differences were observed between the two groups in terms of postoperative complications. Differences were only noted in length of hospitalization and chest tube duration. Patients in the CPET group had significantly worse pulmonary function and more comorbidities, reflecting the clinical algorithm in which CPET was reserved for borderline cases. Therefore, the two groups are not directly comparable, and our findings should not be interpreted as evidence of CPET’s causal effect on outcomes. Rather, they describe real‑world outcomes in a selected group of higher-risk patients who were cleared for surgery after CPET evaluation.

Postoperative complications can lead to prolonged hospital stays, increased medical costs, and higher mortality. The development of complications is multifactorial and highly variable, usually depending on surgical factors and individual characteristics. Advanced age, extensive surgery, and thoracic procedures are particularly associated with postoperative complications (18,19). For thoracic surgery patients, preoperative pulmonary function assessment is critical. These patients are predominantly elderly, have a history of smoking, and have multiple lung diseases. While PFTs strongly correlate with pulmonary obstruction severity, they do not directly assess gas exchange or cardiovascular reserve (20). In contrast, CPET evaluates the interaction between pulmonary function, cardiovascular health, and peripheral tissue oxygen utilization. Some lung cancer patients initially deemed ineligible for surgery based on FEV1, FVC, or DLCO may qualify after CPET. Accordingly, the 2009 ESTS/ERS guidelines recommend CPET for all patients with FEV1 or DLCO <80% (21). Prior guidelines emphasize VO₂ peak as a key metric for surgical tolerance and cardiorespiratory fitness (22). A VO₂ peak >20 mL/kg/min or >75% of predicted value indicates suitability for all planned surgeries, including pneumonectomy. Postoperative mortality risk also correlates closely with VO₂ peak (23-25); patients with VO₂ peak <10 mL/kg/min face significantly higher mortality.

In our study, some patients initially excluded based on PFTs tolerated surgery after CPET reevaluation. Their perioperative outcomes matched or surpassed those of patients with normal PFTs. In this study, the CPET group had a significantly longer total hospital stay and chest tube duration compared to the PFT group. The reasons for these differences are likely multifactorial. Patients in the CPET group had worse baseline pulmonary function and more comorbidities, which may have independently contributed to a slower postoperative recovery. Additionally, different perioperative management strategies—such as a more conservative approach to chest tube removal or discharge—cannot be excluded. This likely explains the longer hospital stay and extended chest tube duration observed in the CPET group. Notably, the CPET group had a higher rate of sublobar resection (62.71% vs. 52.11%) and longer operative times (135 vs. 118.5 min), suggesting lung function influences (but does not solely determine) postoperative outcomes. It should also be noted that the CPET group underwent a higher proportion of sublobar resections, which suggests that surgeons may have intentionally chosen less extensive resections for these higher‑risk patients. Therefore, the absence of a significant difference in postoperative complications between groups may reflect the adoption of a more conservative surgical strategy rather than being solely attributable to CPET‑based patient selection. Cardiovascular disease and COPD can impair a patient’s cardiopulmonary function, increasing postoperative complication risks. However, our subgroup analysis failed to demonstrate these factors’ effects on complications, likely because most patients had only isolated risk factors. Those with multiple comorbidities were excluded (receiving chemotherapy/targeted therapy instead), resulting in few such cases in our cohort.

This study has three main limitations. Firstly, CPET’s underutilization in China (due to cost and technical complexity) limited statistical power; given the relatively limited sample size, the present findings should be interpreted with caution, and larger studies are necessary to further validate the role of CPET in this setting. Secondly, postoperative complication diagnoses may vary subjectively between physicians and radiologists. As a retrospective design, human error in medical record review is possible. Thirdly, being a single-center study, our results may lack generalizability to other populations. Nonetheless, these findings provide clinically relevant insights for other institutions.

Conclusions

In our study, lung cancer patients stratified by preoperative PFT levels showed comparable perioperative outcomes whether undergoing CPET-guided surgery or not. This confirms CPET as a feasible supplementary tool that does not result in a statistically significant increase in postoperative complications in selected patients who proceed to surgery.

Acknowledgments

The authors thank Taizhou Hospital of Zhejiang Province Affiliated to Wenzhou Medical University for providing the research environment.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2026-1-0180/rc

Data Sharing Statement: Available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2026-1-0180/dss

Peer Review File: Available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2026-1-0180/prf

Funding: The present study was supported by Taizhou Science and Technology Project (No. 22ywa36), Medical and Health Science and Technology Project of Zhejiang Province (No. WKJ-ZJ-26101), and Scientific Research Foundation of Taizhou Enze Medical Center (Group) (Nos. 2025EZLY02 and 2024EZLY04).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2026-1-0180/coif). M.K. reports financial support from Taizhou Science and Technology Project. J.S. reports financial support from Medical and Health Science and Technology Project of Zhejiang Province and Scientific Research Foundation of Taizhou Enze Medical Center (Group). 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. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Institutional Review Board of Taizhou Hospital of Zhejiang Province Affiliated to Wenzhou Medical University (No. K20241102). Informed consent from the included patients was waived due to the retrospective nature of the study.

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. Sung H, Ferlay J, Siegel RL, et al. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J Clin 2021;71:209-49. [Crossref] [PubMed]
2. Chao WH, Tuan SH, Tang EK, et al. Effectiveness of Perioperative Cardiopulmonary Rehabilitation in Patients With Lung Cancer Undergoing Video-Assisted Thoracic Surgery. Front Med (Lausanne) 2022;9:900165. [Crossref] [PubMed]
3. Puente-Maestú L, Villar F, González-Casurrán G, et al. Early and long-term validation of an algorithm assessing fitness for surgery in patients with postoperative FEV₁ and diffusing capacity of the lung for carbon monoxide < 40%. Chest 2011;139:1430-8. [Crossref] [PubMed]
4. Brunelli A, Kim AW, Berger KI, et al. Physiologic evaluation of the patient with lung cancer being considered for resectional surgery: Diagnosis and management of lung cancer, 3rd ed: American College of Chest Physicians evidence-based clinical practice guidelines. Chest 2013;143:e166S-90S.
5. Sankar A, Thorpe KE, Gershon AS, et al. Association of preoperative spirometry with cardiopulmonary fitness and postoperative outcomes in surgical patients: A multicentre prospective cohort study. EClinicalMedicine 2020;23:100396. [Crossref] [PubMed]
6. Dankert A, Dohrmann T, Löser B, et al. Pulmonary Function Tests for the Prediction of Postoperative Pulmonary Complications. Dtsch Arztebl Int 2022;119:99-106. [Crossref] [PubMed]
7. Older PO, Levett DZH. Cardiopulmonary Exercise Testing and Surgery. Ann Am Thorac Soc 2017;14:S74-83. [Crossref] [PubMed]
8. Rushwan A, Stefanou D, Tariq J, et al. Increased minute ventilation-to-carbon dioxide slope during cardiopulmonary exercise test is associated with poor postoperative outcome following lung cancer resection. Eur J Cardiothorac Surg 2024;65:ezad337. [Crossref] [PubMed]
9. Wittekind SG, Redington A. Functional outcomes after pulmonary valve replacement: how can we expect patients to rehabilitate if we do not help them? Eur J Cardiothorac Surg 2021;61:73-4. [Crossref] [PubMed]
10. Xu M, Yang X, Guo L. Effectiveness of preoperative and perioperative pulmonary rehabilitation nursing program for the management of patients undergoing thoracic surgery: A systematic review and meta-analysis. Pak J Med Sci 2024;40:1280-6. [Crossref] [PubMed]
11. Shelley BG, McCall PJ, Glass A, et al. Association between anaesthetic technique and unplanned admission to intensive care after thoracic lung resection surgery: the second Association of Cardiothoracic Anaesthesia and Critical Care (ACTACC) National Audit. Anaesthesia 2019;74:1121-9. [Crossref] [PubMed]
12. Baar W, Semmelmann A, Knoerlein J, et al. Risk Factors for Postoperative Pulmonary Complications Leading to Increased In-Hospital Mortality in Patients Undergoing Thoracotomy for Primary Lung Cancer Resection: A Multicentre Retrospective Cohort Study of the German Thorax Registry. J Clin Med 2022;11:5774. [Crossref] [PubMed]
13. Pellegrino R, Viegi G, Brusasco V, et al. Interpretative strategies for lung function tests. Eur Respir J 2005;26:948-68. [Crossref] [PubMed]
14. Balady GJ, Arena R, Sietsema K, et al. Clinician's Guide to cardiopulmonary exercise testing in adults: a scientific statement from the American Heart Association. Circulation 2010;122:191-225. [Crossref] [PubMed]
15. Spruit MA, Singh SJ, Garvey C, et al. An official American Thoracic Society/European Respiratory Society statement: key concepts and advances in pulmonary rehabilitation. Am J Respir Crit Care Med 2013;188:e13-64. [Crossref] [PubMed]
16. 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]
17. Teh E, Sinha S, Joshi N, et al. Cardiopulmonary exercise testing (CPET) and the prediction of perioperative events in patients undergoing lung resection in the modern era: A comparison of clinical, CPET and combined assessment. J Clin Anesth 2020;62:109749. [Crossref] [PubMed]
18. Odor PM, Bampoe S, Gilhooly D, et al. Perioperative interventions for prevention of postoperative pulmonary complications: systematic review and meta-analysis. BMJ 2020;368:m540. [Crossref] [PubMed]
19. Huang Q, Rauniyar R, Yang J, et al. Risk stratification of postoperative pulmonary complications in elderly patients undergoing lung cancer resection: a propensity score-matched study. J Thorac Dis 2023;15:3908-18. [Crossref] [PubMed]
20. Kallianos A, Rapti A, Tsimpoukis S, et al. Cardiopulmonary exercise testing (CPET) as preoperative test before lung resection. In Vivo 2014;28:1013-20.
21. Brunelli A, Charloux A, Bolliger CT, et al. ERS/ESTS clinical guidelines on fitness for radical therapy in lung cancer patients (surgery and chemo-radiotherapy). Eur Respir J 2009;34:17-41. [Crossref] [PubMed]
22. Lim E, Baldwin D, Beckles M, et al. Guidelines on the radical management of patients with lung cancer. Thorax 2010;65:iii1-27. [Crossref] [PubMed]
23. Kasikcioglu E, Toker A, Tanju S, et al. Oxygen uptake kinetics during cardiopulmonary exercise testing and postoperative complications in patients with lung cancer. Lung Cancer 2009;66:85-8. [Crossref] [PubMed]
24. Brunelli A, Belardinelli R, Refai M, et al. Peak oxygen consumption during cardiopulmonary exercise test improves risk stratification in candidates to major lung resection. Chest 2009;135:1260-7. [Crossref] [PubMed]
25. Fang Y, Ma G, Lou N, et al. Preoperative Maximal Oxygen Uptake and Exercise-induced Changes in Pulse Oximetry Predict Early Postoperative Respiratory Complications in Lung Cancer Patients. Scand J Surg 2014;103:201-8. [Crossref] [PubMed]