Predictive Value of Preoperative Peak Oxygen Uptake for Postoperative Pulmonary Complications in Lung Cancer Patients with Chronic Obstructive Pulmonary Disease: A Single-Center Retrospective Cohort Study

In recent years, video-assisted thoracoscopic surgery (VATS), which is less invasive than conventional open thoracic surgery, has become popular for lung cancer resection. VATS is less invasive to the chest wall and reportedly offers advantages such as reduced postoperative pain and shorter hospital stay [1, 2].

Postoperative pulmonary complications (PPC), such as pneumonia and atelectasis, are associated with increased mortality and length of hospital stay [3]. A previous study has reported that 54% of patients with lung cancer develop chronic obstructive pulmonary disease (COPD) [4], which is an independent predictor of lung cancer prognosis [5]. COPD is also a risk factor for PPC after lung cancer surgery [6] and can significantly increase the length of hospital stay, duration of ventilator use, and complications after lung cancer resection in patients with COPD [7, 8]. In addition, COPD has been reported to increase the incidence of PPC, regardless of its severity [9].

Although VATS may reduce the incidence of PPC in pneumonectomy compared to that in open thoracotomy [10, 11], accurate prediction of the risk of complications is challenging [12, 13]. The European Respiratory Society (ERS)/European Society of Thoracic Surgeons (ESTS) and American College of Chest Physicians guidelines [14, 15] recommend pulmonary function tests as important objective data for preoperative risk stratification. However, pulmonary function tests can be affected by factors such as patient cooperation and comprehension, and measurements can be affected in patients with moderate to severe COPD [14, 15]. Furthermore, previous research has reported that preoperative pulmonary function tests are not predictive of pulmonary complications in all surgical procedures and that pulmonary function tests alone have limitations in predicting postoperative complications of lung cancer [16].

If both the predicted postoperative % diffusing capacity of the lung for carbon monoxide (PPO-%DLCO) and the predicted postoperative forced expiratory volume % in 1 s (PPO-%FEV₁) are less than 60% of the predicted values in the preoperative pulmonary function test, a cardiopulmonary exercise test (CPET) is recommended [17]. Patients with a preoperative maximal oxygen uptake/weight (peak VO₂/W) of 10–15 mL/kg/min have a high rate of postoperative complications and mortality [18‒20]; hence, patients with VO₂/W <10 mL/kg/min should be considered for reduction surgery or non-surgical treatment [15].

In contrast, a report on peak VO₂/W on preoperative CPET and postoperative complications in patients with lung cancer and COPD highlighted that peak VO₂/W had a weak association with postoperative complications [21]. However, this report only included patients who underwent open thoracotomy and did not include those who underwent VATS. Currently, the relationship between preoperative peak VO₂/W and PPC in lobectomies, including VATS, remains unclear. Furthermore, to the best of our knowledge, no studies have reported on the relationship between the preoperative peak VO₂/W and PPC, including VATS, in patients with lung cancer complicated by COPD. Therefore, this study aimed to clarify the relationship between preoperative peak VO₂/W, PPC, and other factors related to PPC in postoperative lung cancer patients with COPD.

This single-center, retrospective cohort study included lung cancer patients with COPD who underwent preoperative CPET at the Department of Respiratory Surgery, Kindai University Hospital, between January 2017 and March 2024. The diagnosis and classification of COPD were established according to the Global Initiative for Chronic Obstructive Lung Disease guidelines [22]. In this population, the diagnosis was established by a respiratory physician. CPET was performed in patients with preoperative % forced expiratory volume in 1 s (%FEV₁) or % diffusing capacity of the lung for carbon monoxide (%DLCO) <80% of the predicted value, and PPO-%FEV₁ or PPO-%DLCO <60% of the predicted value, or in patients deemed necessary by the physician.

VATS was defined as a thoracoscopic procedure with an incision length of 8 cm or less. The exclusion criteria were as follows: (1) patients who underwent reoperation in the postoperative period, (2) patients who underwent a procedure other than lobectomy, (3) patients with complications that would affect the performance of CPET, and (4) patients who were deemed inappropriate by the principal investigator to participate in the study.

Postoperative complications, such as pneumonia and atelectasis, were defined as PPC, and patients with Clavien-Dindo grade II or higher were followed up for 30 days after surgery [23]. All patients underwent physical therapy before and after surgery.

All participants underwent CPET using a bicycle ergometer according to the ramp 10 W protocol (10 W/min^-1 load increase, 1 W per 6 s). A modified Borg scale was used to measure the degree of subjective symptoms. Dyspnea and leg fatigue were measured at 1-min intervals during exercise and rest periods. Peak VO₂/W and the ventilatory equivalent/ventilatory carbon dioxide (VE/VCO₂) slope were used for analysis.

The extent of emphysema was assessed using computed tomography (CT) scans of 1.0-mm slices obtained within 1 month of lung cancer resection, and the lung percentage of low attenuation area (LAA%) was measured using a Synapse Vincent volume analyzer (Fujifilm Medical, Tokyo, Japan). The software automatically quantified the LAA% in the bilateral lung areas using a threshold of less than −950 Hounsfield units (HU) to distinguish emphysema from other tissues. The LAA% was calculated by subtracting the relevant emphysema volume from the total lung volume and applying the lower threshold of −950 HU to the total lung percentage.

The number of lung areas was calculated to be 18 (right upper lobe, 3; right middle lobe, 2; right lower lobe, 5; left upper lobe, 4; and left lower lobe, 4).

Patients were divided into those with and without PPC (PPC and non-PPC groups, respectively). Continuous variables were compared using an unpaired t test, and categorical variables were compared using Fisher’s exact probability test. The unpaired t test and Fisher’s exact probability test were used to compare patient data with and without PPC. In addition, multivariate logistic regression models were used to analyze the factors associated with PPC. The receiver operating characteristic (ROC) curve was used to analyze whether peak VO₂/W had a higher predictive power for PPC. All statistical analyses were analyzed using the JMP software program (JMP^®, version 14, SAS Institute Inc., Cary, NC, USA). Statistical significance was set at p < 0.05.

Between January 2017 and March 2024, a total of 1,347 patients underwent lobectomy, of whom 564 had COPD and 44 underwent preoperative CPET. Four patients were excluded, as they could not perform accurate CPET measurements (depression, previous stroke, arteriosclerosis obliterans, or home oxygen therapy). Overall, 40 patients were eligible for analysis (Fig. 1). The clinical characteristics of the patients are presented in Table 1. Of the 40 patients, 21 (52.5%) had PPC.

The %DLCO, PPO-%DLCO, and peak VO₂/W were significantly lower in the PPC group than in the non-PPC group (p < 0.01, p < 0.05, and p < 0.001, respectively), whereas the VE/VCO₂ slope was significantly higher in the PPC group than in the non-PPC group (p < 0.05) (Table 1). The differences in other parameters, including preoperative FEV₁, PPO-%FEV₁, PPO-%DLCO, and surgical approach, were not significant between the two groups. Preoperative %FEV₁ was strongly correlated with PPO-%FEV₁ (r = 0.92, p < 0.01, (online suppl. Fig. 1; for all online suppl. material, see https://doi.org/10.1159/000543370), and preoperative %DLCO was strongly correlated with PPO-%DLCO (r = 0.97, p < 0.01; see online suppl. Fig. S2).

We created two models using multivariate logistic analysis: model 1 included preoperative %FEV₁, preoperative %DLCO, peak VO₂/W, and VE/VCO₂ slope as independent variables; model 2 included PPO-%FEV₁, PPO-%DLCO, peak VO₂/W, and VE/VCO₂ slope as independent variables. Both models showed that only peak VO₂/W was a significant independent factor for predicting PPC (Table 2). The area under the ROC curve of peak VO₂/W to predict PPC was 0.93, with a cutoff value for peak VO₂/W of 14.6 mL/min/kg, sensitivity of 78%, and a specificity of 95% (Fig. 2).

This study examined the relationship between preoperative VO₂/W, measured using CPET, and the incidence of PPC in lung cancer patients with COPD undergoing lobectomy, including VATS. Our results indicated that preoperative VO₂/W was the most valuable predictor of PPC in lung cancer patients with COPD. To the best of our knowledge, this is the first study to show the efficacy of the preoperative peak VO₂/W in predicting PPC after lobectomy, including VATS, in lung cancer patients with COPD.

In this study, multivariate logistic analysis revealed that decrease in peak VO₂/W with preoperative CPET was a better predictor of PPC than pulmonary function or surgical approach. Furthermore, preoperative peak VO₂/W had a high specificity of 95% and a good area under the ROC curve with a cutoff value of 14.6 mL/min/kg for predicting PPC.

In the ERS/ESTS clinical guidelines, CPET is recommended for all patients undergoing lung resection if the preoperative %FEV1 or DLCO is less than 80% as a risk assessment for PPC before lung resection in patients with lung cancer [14]. The VO₂/max obtained with CPET has threshold values set at >20 mL/min/kg for low risk, 10–20 mL/min/kg for high risk, and <10 mL/min/kg for very high risk, while 15 mL/min/kg is widely used as a predictive criterion for PPC [25]. However, previous reports did not account for the coexistence of COPD, with the surgical approach being limited to open chest surgery and with no consideration of VATS.

In this study, we included VATS in our analysis but found no significant difference between the surgical approach and PPC, and the risk of PPC was correlated with peak VO₂/W. Furthermore, the cutoff values were similar to those reported in previous studies [25]. The incidence of PPC was relatively high (52.5%) in this study, as we analyzed only patients with COPD who had low pulmonary function (preoperative %FEV1 or %DLCO <80% of the predicted value). As expected, the coexistence of COPD may be correlated with PPC.

In contrast, Shafiek et al. [21] reported that peak VO₂/W was not an independent factor for postoperative complications and suggested the VE/VCO₂ slope as a predictor of PPC. However, in their report, the area of lung cancer resection was not constant, and VATS was not considered. In our study, the VE/VCO₂ slope was significantly different between the two groups; however, logistic regression analysis showed that peak VO₂/W was the only predictor of PPC. We only included patients who had undergone lobectomy; therefore, this discrepancy may be due to differences in the surgical approach (VATS or open surgery) and excision range.

Although the mechanism underlying the association between preoperative peak VO₂/W and PPC remains unclear, the involvement of decreased resistance to oxidative stress and weakened respiratory muscle function has been suggested [26, 27]. Patients with low exercise tolerance may have difficulty in appropriate regulation of oxygen supply after surgery and may not be able to cope with the increased metabolic demands of the various stresses associated with surgery. Impairment in cardiac pump function, respiratory function, and oxygen utilization within skeletal muscles may result in early postoperative deterioration of comorbidities due to inappropriate postoperative pain management, residual anesthesia, fluid overload, and ventilator-induced lung injury [20]. CPET can assess the overall exercise response of patients, including cardiovascular, respiratory, and skeletal muscle functions, which together defines peak VO₂/W [27].

Diaphragmatic dysfunction has also been reported after thoracic, abdominal, and cardiac surgeries [28]. Patients with diaphragmatic dysfunction detected using an ultrasound imaging device in the first 24 h after surgery have a significantly higher risk of developing PPC, such as atelectasis, within 7 days after surgery [29]. The risk of diaphragmatic dysfunction is higher with open chest surgery than with VATS [29]. The association between diaphragmatic dysfunction and PPC can be explained by the reduced ventilatory function and expiratory capacity of patients [30, 31]. Peak VO₂/W in patients with COPD is closely related to diaphragm function [32, 33]. In patients with lung cancer complicated by COPD, diaphragmatic function may be impaired even before surgery, and a low peak VO₂/W may affect the development of PPC. Therefore, diaphragmatic function may need to be considered in risk assessment, in addition to CPET.

This study has some limitations. First, this was a single-center study with a small sample size, and the comorbidities included not only COPD but also orthopedic, cardiac, and cardiovascular diseases. However, these comorbidities may not have affected the CPET results, given the inclusion and exclusion criteria for CPET and the performance status of the population being maintained within 0–1, making the present results relatively reliable. Second, nutritional status may also be related to PPC; however, the characteristics of this population did not include an index of nutritional status. Nevertheless, as no cases of sarcopenia were observed in the study population, the influence of nutritional status was considered low. Third, compared with pulmonary function tests, peak VO₂/W requires equipment to perform CPET, which might be difficult to perform in all hospitals due to the high cost of the equipment. However, preoperative CPET is recommended according to the ERS/ESTS and American College of Chest Physicians guidelines. Furthermore, since peak VO₂/W is associated with PPC, duration of ventilator use, and mortality, we actively performed it as a risk assessment in our institution.

In conclusion, the findings highlight the limitations of traditional pulmonary function tests in this context and underscore the value of CPET in providing a more accurate risk assessment. Incorporating CPET into preoperative evaluations can enhance risk stratification and guide perioperative management strategies, thereby potentially reducing the incidence of PPC in this high-risk population. Future research should focus on larger, multicenter studies to validate these findings and explore the integration of CPET with other predictive tools to further refine risk assessment protocols for lung cancer surgery patients with COPD.

We would like to thank Editage (www.editage.jp) for English language editing.

This study protocol was reviewed and approved by the Committee for Ethics at the Kindai University School of Medicine, Approval No. R05-053, and was conducted in accordance with the ethical standards established in the 1964 Declaration of Helsinki and subsequent amendments. Opt-out informed consent protocol was used for use of participant data for research purposes. This consent procedure was reviewed and approved by the Committee for Ethics at the Kindai University School of Medicine, Approval No. R05-053, date of decision: July 21, 2023.

The authors have no conflicts of interest to declare.

This work was supported by Grant-in-Aid for Scientific Research (21K11325).

Masaya Noguchi: conceptualization, data curation, formal analysis, investigation, methodology, visualization, and writing – original draft. Toshiki Takemoto: investigation, resources, and writing – review and editing. Masashi Shiraishi: conceptualization, investigation, resources, and supervision. Ryuji Sugiya: investigation. Hiroki Mizusawa: investigation and resources. Tamotsu Kimura: investigation and project administration. Akira Tamaki: supervision and writing – review and editing. Yasuhiro Tsutani: investigation, project administration, resources, and writing – review and editing. Yuji Higashimoto: conceptualization, funding acquisition, methodology, project administration, supervision, and writing – review and editing.

The data that support the findings of this study are not publicly available due to their containing information that could compromise the privacy of research participants but are available from the corresponding author (M.N.) upon reasonable request.