Prevalence of and risk factors for pulmonary complications after curative resection in otherwise healthy elderly patients with early stage lung cancer

Background and objective The prevalence of lung cancer has been increasing in healthy elderly patients with preserved pulmonary function and without underlying lung diseases. We aimed to determine the prevalence of and risk factors for postoperative pulmonary complications (PPCs) in healthy elderly patients with non-small cell lung cancer (NSCLC) to select optimal candidates for surgical resection in this subpopulation. Methods We included 488 patients older than 70 years with normal spirometry results who underwent curative resection for NSCLC (stage IA-IIB) between 2012 and 2016. Results The median (interquartile range) age of our cohort was 73 (71–76) years. Fifty-two patients (10.7%) had PPCs. Severe PPCs like acute respiratory distress syndrome, pneumonia, and respiratory failure had prevalences of 3.7, 3.7, and 1.4%, respectively. Compared to patients without PPCs, those with PPCs were more likely to be male and current smokers; have a lower body mass index (BMI), higher American Society of Anesthesiologists (ASA) classification, more interstitial lung abnormalities (ILAs), and higher emphysema index on computed tomography (CT); and have undergone pneumonectomy or bilobectomy (all p < 0.05). On multivariate analysis, ASA classification ≥3, lower BMI, ILA, and extent of resection were independently associated with PPC risk. The short-term all-cause mortality was significantly higher in patients with PPCs. Conclusions Curative resection for NSCLC in healthy elderly patients appeared feasible with 10% PPCs. ASA classification ≥3, lower BMI, presence of ILA on CT, and larger extent of resection are predictors of PPC development, which guide treatment decision-making in these patients. Electronic supplementary material The online version of this article (10.1186/s12931-019-1087-x) contains supplementary material, which is available to authorized users.


Introduction
The median age at diagnosis of lung cancer is 70 years, [1] and more than 70% of future lung cancer cases are expected to occur in adults older than 65 years [2]. As a result of early detection of lung cancer by screening [3], and the recent advances in surgical techniques and perioperative management, the need for curative surgical resection of early stage non-small cell lung cancer (NSCLC) in elderly patients has been increasing [4]. Given the current trend, surgeons are encountering larger numbers of elderly patients with surgically resectable NSCLC and have to assess the benefits and risks of surgical treatment in this population.
Postoperative pulmonary complications (PPCs) are the major cause of perioperative morbidity and mortality, occurring in 14-40% of patients after lung resection [5,6]. Old age has been considered as a poor prognostic factor for PPCs due to the tendency for elderly patients to have multiple and inter-related risk factors, such as poor performance, underlying comorbidities, and impaired lung function [7,8]. While decreased lung function is a major determinant for PPCs, and underlying lung diseases are associated with poorer surgical outcomes [9][10][11][12][13], healthy elderly patients with preserved lung function and no underlying lung diseases would be suitable candidates for surgery. In this population, however, the evidence to decide the treatment strategy is still scant, when comparing the risks between postsurgical mortality/morbidity and lung cancer progression [14]. Moreover, although chest computed tomography (CT) is performed in all patients with lung cancer, there is no consensus regarding preoperative chest CT evaluation to detect early structural abnormalities that are not reflected in pulmonary function tests. In this regard, it is imperative that we fully understand the risk factors for PPCs including chest CT findings particularly in healthy elderly patients with lung cancer. Therefore, we aimed to assess the prevalence of PPCs after surgical resection of early stage NSCLC in healthy elderly patients. Furthermore, including the preoperative chest CT findings for thoroughly evaluating the pulmonary condition, we examined to identify independent risk factors for PPCs to select optimal candidates for curative lung resection in this population.

Study population
The present study investigated surgical patients aged over 70 years, as this population is at high risk of PPCs following ACP guidelines [15,16]. Thus, we conducted a retrospective cohort study of 511 patients older than 70 years with normal spirometry results who underwent curative resection for stage I and II NSCLC at Samsung Medical Center between January 2012 and December 2016. We excluded patients who previously underwent lung resection with lobectomy (n = 8), Ivor Lewis operation (n = 5), and those with interstitial lung disease (n = 10). In total, 488 patients were included in the final analysis. This study was approved by the Institutional Review Board of Samsung Medical Center, which exempted the requirement for informed consent as we only used de-identified data retrieved from electronic medical records (IRB no. 2018-06-032-001).

Preoperative evaluation
Clinical characteristics, including age at surgery, smoking status, body mass index (BMI), comorbidity, tumor histology, pathologic stage, and type of surgery, were obtained from electronic medical records. Tumors were staged according to the 7th edition of the TNM classification [17].
The routine preoperative evaluation included spirometry and CT of the chest, positron emission tomography/CT, brain magnetic resonance imaging, and flexible bronchoscopy. Distant metastasis was also assessed by chest CT as radiographic study of the upper abdomen (i.e., liver, spleen, stomach and adrenal gland) was included in the chest CT examination for all patients. Spirometry and diffusing capacity of the lung for carbon monoxide (DLco) measurements were performed using Vmax 22 spirometers (SensorMedics, Yorba Linda, CA, USA) according to the American Thoracic Society/ European Respiratory Society recommendations [18,19]. Preserved pulmonary function was defined as a ratio of pre-bronchodilator forced expiratory volume in 1 s (FEV 1 ) to forced vital capacity (FVC) > 0.70 and FVC ≥ 80% of the predicted value [20].

CT findings
We analyzed and quantified the size and lobar location of the tumor, interstitial lung abnormalities (ILAs), emphysema index, superimposed infection, mosaic attenuation, airway abnormality, and ratio of main pulmonary artery-to-ascending aorta diameter on chest CT, which was performed closest to the operation, by using available software (Extended Brilliance Workspace v3.0; Philips Medical Systems, Cleveland, OH, USA).
ILAs were defined as particular patterns of increased lung density, including ground-glass or reticular abnormalities, diffuse centrilobular nodularity, nonemphysematous cysts, and honeycombing or traction bronchiectasis, affecting more than 5% of any lung zone [21]. ILAs were assessed subjectively by adding and scoring the overall extent of fibrosis-related lung parenchymal abnormalities, which were estimated from the nearest 5% of parenchymal involvement to the whole lung volume (100%) [22,23]. We used an in-house computer software to assess and automatically calculate the emphysema index, which was defined as the volume fraction of the lung below − 950 HU [23,24].

Postoperative pulmonary complications
To prevent the occurrence of PPCs, every patient received postoperative respiratory physiotherapy program as well as early ambulation and lung expansion maneuvers, including deep breathing exercises, incentive spirometry, and airway clearance techniques such as active coughing and sputum expectoration. If the respiratory care was not effective, physiotherapist (specialized nurse for respiratory care) assessed the patient twice or three times per day, and actively conducted postural drainage and vibrator.
We considered the following PPCs that occurred during hospitalization or readmission for 60 days postoperatively: pneumonia, acute respiratory distress syndrome (ARDS), respiratory failure, significant atelectasis requiring bronchoscopy or reintubation, bronchopleural fistula/empyema, prolonged air leakage lasting for more than 5 days, and pneumothorax [25]. Severe PPCs were defined as ARDS, pneumonia, or respiratory failure. The definitions of PPCs are listed in Additional file 1: Table S1.

Statistical analysis
Data are presented as n, median (interquartile range), or n (%). Categorical variables were compared using Pearson's χ 2 test, and continuous variables were compared using Student's t-test. To identify risk factors independently associated with the occurrence of PPCs, we conducted a multivariate logistic regression, which included clinically relevant variables regardless of the p-value in the univariate analyses. Overall survival rate was estimated using the Kaplan-Meier method, and survival distributions between different groups were compared using the log-rank test.
All tests were two-sided, and a p-value of less than 0.05 was considered to indicate statistical significance. All statistical analyses were performed using IBM SPSS Statistics for Windows Version 24.0 (IBM Corp., Armonk, NY, USA).

Factors associated with PPC development
When the 52 patients with PPCs were compared to the 436 patients without PPCs, those with PPCs were more likely to be male (p = 0.003), current or exsmokers (p = 0.013), have a lower BMI (p = 0.007), have an ASA classification ≥3 (p = 0.021), have more ILAs (p = 0.004), and have a higher emphysema index (p = 0.006). However, no significant differences were observed in age; comorbidities; serum hemoglobin and albumin levels; spirometry and DLco results; and other CT findings ( Table 1). As shown in Table 2, while the proportion of patients with adenocarcinoma was higher among those without PPCs, the proportion of patients with squamous cell carcinoma was higher among those with PPCs (30.8% [16/52] vs. 14.9% [65/437]; p = 0.003). However, no significant difference was observed in tumor location or tumor stage between the two groups. Regarding surgical procedures, the patients with PPCs were more likely to have undergone pneumonectomy or bilobectomy than were those without PPCs (11.5% [6/52] vs. 2.8% [12/ 437]; p = 0.005). However, no significant difference was observed in the proportion of video-assisted thoracic surgery or extent of lymph node dissection between the two groups (  (Table 5). There were 470 (96.3%) patients who underwent lobectomy, segmentectomy or wedge resection in our study population. Among them, lower BMI (adjusted OR, 0.86; 95% CI, 0.78-0.95; p = 0.005), and the presence of ILA (adjusted OR, 1.07; 95% CI, 1.02-1.13; p = 0.008) were independently associated with PPC occurrence, analyzed by multivariate logistic regression (Additional file 2: Table S2).

Discussion
Our study showed that the overall prevalence of PPCs following lung resection was 10% for patients with early stage NSCLC aged over 70 years old with preserved lung function. Among the factors associated with PPC occurrence, we found several independent predictors of PPCs: ASA classification ≥3, lower BMI, ILAs on chest CT, and greater surgical extent.
PPC incidence after pulmonary resection ranged from 13 to 16% in prospective observational studies that included patients of all ages and with poor lung function [5,26,27]. Our study included only elderly patients with preserved lung function and showed a 10% overall incidence of PPCs after surgery for early stage NSCLC. This result is comparable to that of a previous study showing  [27]. In addition, age over 70 years was not identified as a risk factor for PPCs in our study, indicating that elderly patients should not be precluded from lung cancer surgery solely on the basis of their age. We focused on exploiting detailed radiological parameters and hypothesized that particular radiologic findings would reflect the pulmonary physiologic state even under preserved pulmonary function. As a result, although we excluded clinical interstitial lung disease and abnormal lung function, including chronic obstructive pulmonary disease, ILAs and emphysema were independent factors that predicted PPCs in elderly patients. Previous studies have well documented that abnormal CT findings, such as ILA and emphysema, can present in the elderly without any respiratory symptoms or spirometric abnormalities. While the clinical implications of these findings have not been fully elucidated, recent studies have shown that these abnormal CT findings are associated with rapid lung function decline [28,29] and increased all-cause mortality [21,29,30]. Our study also provided prognostic implications of detecting subclinical radiologic abnormalities, which are associated with the development of PPCs after lung resection surgery. These results implied that radiological abnormalities such as ILA and emphysema could be important determinants that increase the incidence of PPCs in elderly patients even when their lung function is normal. Among ILA and emphysema, ILA remained an independent factor that increased the incidence of PPCs in the present study. Given that emphysematous lung destruction is strongly associated with impaired lung function [31], the predictive role of ILA in PPCs might be stronger than that of emphysema in elderly patients with preserved lung function. Our study suggested that radiologic assessment would assist treatment decision-making even in healthy elderly patients who are candidates for lung resection. We believe that there are real opportunities to find out the possibility of development of PPCs even in this healthy elderly population through preoperative CT. Nevertheless, further study is needed to determine the role of CT biomarkers in predicting PPCs by using longitudinal data.
The extent of lung resection (pneumonectomy or bilobectomy) are also well-known predictive risk factors of PPCs [5]. Pneumonectomy is associated with not only a significant decrease in pulmonary function but also numerous potential complications that involve the pulmonary and cardiovascular systems [32]. Moreover, a previous study reported that the physical quality of life in patients who underwent pneumonectomy was significantly lower than that of the general population [33]. In particular, patients older than 70 years and those with a low preoperative quality of life appear more likely to have an unsatisfactory quality of life at 6 months after pneumonectomy than do other patients [34], and our previous data showed that patients older than 70 years with early stage lung cancer had a worse long-term survival rate after pneumonectomy than younger patients   [35]. Thus, elderly patients scheduled to undergo pneumonectomy would require intensive preoperative cardiopulmonary evaluation as well as an assessment of preoperative physical quality of life. Among the risk factors we identified, higher ASA classification and lower BMI are well-known risk factors that predict PPC occurrence after lung cancer surgery [36,37], as a poor physical status, including low muscle mass, are strongly related to ineffective respiration and decreased physical activity after surgery.
The major limitation of our study was its retrospective, single-center design. Thus, patient-centered outcomes, such as the quality of life during the perioperative period, could not be investigated in our study. Further prospective studies are required to assess the impact of the quality of life on PPCs. In addition, as this study was conducted only in patients at a tertiary hospital, the results might not be generalizable to different settings. However, for a single institution study, this is a relatively large cohort of elderly patients.
In conclusion, pulmonary resection for early stage NSCLC seems feasible and should thus be provided to healthy elderly patients after risk factors for PPCs, such as poor performance, lower BMI, ILA on chest CT, and the extent of surgery, are adequately addressed preoperatively. Future prospective studies with long-term follow-up are needed to verify the prognostic significance of these risk factors.