Original Thoracic
The Impact of Combined Pulmonary Fibrosis and Chronic Obstructive Pulmonary Disease on Long-Term Survival after Lung Cancer Surgery Yasuo Sekine1 Yuichi Sakairi2 Ichiro Yoshino2
Mitsuru Yoshino1
1 Department of Thoracic Surgery, Tokyo Women’s Medical University,
Yachiyo, Chiba, Japan 2 Department of Thoracic Surgery, Chiba University Graduate School of Medicine, Chiba, Japan
Eitetsu Koh1
Atsushi Hata1
Hidemi Suzuki2
Address for correspondence Yasuo Sekine, MD, PhD, Department of Thoracic Surgery, Tokyo Women’s Medical University, Yachio Medical Center, 477-96 Owadashinden, Yachiyo, Chiba 276-8524, Japan (e-mail: [email protected]).
Thorac Cardiovasc Surg 2014;62:332–337.
Abstract
Keywords
► lung cancer ► chronic obstructive pulmonary disease ► pulmonary fibrosis ► combined pulmonary fibrosis and emphysema
Purpose The purpose of this study was to determine the impact of pulmonary fibrosis (PF) on postoperative complications and on long-term survival after surgical resection in lung cancer patients with chronic obstructive pulmonary disease (COPD). Patients and Methods A retrospective chart review was conducted of 380 patients with COPD who had undergone pulmonary resection for lung cancer at the University Hospital between 1990 and 2005. The definition of COPD was a preoperative forced expiratory volume in 1 second/forced vital capacity (FEV1/FVC) ratio of less than 70%; PF was defined as obvious bilateral fibrous change in the lower lung fields, confirmed by computed tomography. Results PF was present in 41 patients (10.8%) with COPD; the remaining 339 patients (89.2%) did not have PF. The preoperative FVC/FEV1 was significantly lower in the group of patients with PF than in the group without (p < 0.05). Acute lung injury and home oxygen therapy were significantly more common in the PF group; however, the 30-day mortality was similar between the groups. The cumulative survival at 3 and 5 years was 53.6 and 36.9%, respectively, in the PF group and 71.4 and 66.1%, respectively, in the non-PF group (p ¼ 0.0009). Increased age, decreased body mass index, advanced pathologic stage, and the existence of PF were identified as independent risk factors for decreased survival. Conclusion PF is a risk factor for decreased survival after surgical treatment in lung cancer patients with COPD.
Introduction It is well known that pulmonary fibrosis (PF) and chronic obstructive pulmonary disease (COPD) are significant risk factors for lung cancer; COPD comprises both pulmonary emphysema and chronic bronchitis.1–3 Cigarette smoking and occupational exposures, among other environmental
received August 19, 2013 accepted after revision November 7, 2013 published online May 30, 2014
hazards, are common risk factors for both PF and COPD. Combined pulmonary fibrosis and emphysema (CPFE) is a clinical and radiological diagnosis that indicates the presence of pulmonary emphysema in the upper lungs and interstitial pneumonia in the lower lungs; various radiological patterns are possible.4 Patients with CPFE may be at significantly increased risk for lung cancer.5 When CPFE patients do indeed
© 2014 Georg Thieme Verlag KG Stuttgart · New York
DOI http://dx.doi.org/ 10.1055/s-0033-1363478. ISSN 0171-6425.
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
332
have lung cancer, survival is significantly worse than in lung cancer patients with emphysema alone.6 The majority of patients with COPD do not have end-stage emphysema, but rather some combination of emphysema and chronic bronchitis, producing varying degrees of airflow obstruction. We previously reported that patients with higher COPD grades have worse long-term survival overall,7 and patients with interstitial lung disease have poor longterm survival after lung cancer surgery.8 We hypothesized that patients with combined PF and COPD may have significantly inferior long-term survival after lung cancer surgery, compared with those who have COPD alone. The majority of patients with CPFE may have advancedstage airway obstruction. As the imaging features of CPFE have not been fully described9 and it is very difficult to define the severity of emphysema radiographically, we focused our investigation on patients with combined PF and the easier-todefine COPD. We determined the clinical features and the short- and long-term outcomes of lung cancer patients with COPD, with and without PF.
Patients and Methods This study was approved by the Institutional Review Board of University Hospital. The requirement for informed consent was waived because of the study’s retrospective nature. Between January 1990 and April 2005, 1,461 patients with lung cancer underwent pulmonary resection at the University Hospital. Of those, 380 were diagnosed with COPD based on preoperative spirometric data. COPD was defined as a preoperative forced expiratory volume in 1 second (FEV1)/forced vital capacity (FVC) ratio of less than 70%.10 Data were collected from our institutional cancer registry database and from patient follow-up visit records, and consisted of information on preoperative patient characteristics, disease status, operative procedures, postoperative complications, pathologic diagnoses, and follow-up examinations. Prospective follow-up data were collected between April 2005 and August 2011. Preoperative conventional or high-resolution chest computed tomography (CT) was evaluated in all patients to determine not only the extent of each patient’s cancer but also the existence of PF. When abnormalities were seen that were compatible with bilateral lung fibrosis in the lower lobes, such as peripheral reticular opacities, PF was diagnosed. Chest CT was reviewed by independent pulmonologists. Diffusing capacity (DLco) is critically important for identifying the severity of COPD and PF. However, because this study was retrospective in nature, DLco could not be obtained from all patients. Pathological confirmation of PF could not be universally obtained, as some patients had upper-lobe lung cancer and therefore lower-lobe lung tissue with PF was not resected. Patients were divided into two groups: the PF/COPD group and the COPD-alone group. Postoperative morbidities included bacterial pneumonia (confirmed by infiltrative shadows on chest radiography, positive sputum culture, temperature 37.5°C, and a white blood cell count > 10,000/μL), acute lung injury (aggravation
Sekine et al.
of dyspnea on exertion, deterioration of arterial blood gases, and diffuse interstitial abnormalities on radiography compatible with acute interstitial pneumonia), prolonged mechanical ventilation ( 3 days), bronchial stump dehiscence, empyema, tracheostomy, and postoperative home oxygen therapy. Home oxygen therapy was given to patients with a partial pressure of oxygen < 60 torr, with exercise, at the time of hospital discharge.8 After discharge, patients visited our outpatient clinic every 1 to 6 months, unless tumor recurrence or any other problems occurred. The presence of tumor recurrence or metastasis was evaluated every 6 to 12 months by chest CT, brain magnetic resonance imaging, 2-[18F]fluoro-2-deoxy-D-glucose positron emission tomography, bone scintigraphy, and serum tumor markers. When patients returned to their referring doctor or were referred on to other doctors, their status and cause of death, if applicable, were confirmed every 2 years by mail or telephone with the patient, the patient’s family, or the referring doctors. Overall and disease-specific survival was analyzed for each patient group.
Statistical Analysis Data were analyzed using Statistical Package for the Social Sciences, version 20 (IBM Corp., New York, New York, United States). To compare differences between the PF/COPD and COPD-alone groups, Mann–Whitney U-test was used for continuous variables and the chi-square test or Fisher exact test was used for categorical variables. Survival curves were estimated using the Kaplan–Meier method, and the difference in survival times between the groups was calculated by the log-rank test. We estimated the overall survival and the disease-specific survival that indicates the percentage of patients in each group who have not died from a lung cancer in a defined period of time. All preoperative and intraoperative variables including sex, age, body mass index, pathologic type and stage, existence of PF, operative time, and operative blood loss were entered into a logistic regression model in a stepwise manner to identify variables that were independently predictive of long-term survival. A p-value of less than 0.05 was considered to be significant.
Results Patient Characteristics and Postoperative Mortality and Morbidity A total of 41 patients (10.8%) were diagnosed with PF (PF/COPD group); the remaining 339 patients (89.2%) were determined not to have PF (COPD-alone group). The patients’ preoperative characteristics are summarized in ►Table 1. The preoperative FVC and FEV1 were significantly lower in the PF/COPD group than in the COPD-alone group (p < 0.05). Perioperative characteristics are listed in ►Table 2. Although the operative methods were similar between the groups, the maximal tumor size and pathological staging were more advanced in the PF/COPD group. Interestingly, there were more lung cancers in upper lobe than in lower lobe in both groups. Thoracic and Cardiovascular Surgeon
Vol. 62
No. 4/2014
333
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
Impact of Pulmonary Fibrosis on Postoperative Complications
Impact of Pulmonary Fibrosis on Postoperative Complications
Sekine et al.
Table 1 Preoperative patient characteristics COPD-alone (n ¼ 339)
PF/COPD (n ¼ 41)
p value
Male (n, %)
312 (92.0)
38 (92.7)
1.00
Female
27 (8.0)
3 (7.3)
68.0 7.2
66.3 7.8
0.191
21.9 2.9
21.7 3.0
0.589
317 (93.8)
39 (95.1)
1.00
52.6 33.0
49.6 24.4
0.483
FVC (L)
3.08 0.68
2.88 0.64
0.069
FVC (%)
99.2 32.9
87.0 22.0
0.010
Characteristics Sex
Age (y)
a
BMI (kg/m2)a PSH (n, %) a
Smoking index (pack-years) Pulmonary function testsa
FEV1 (L)
1.93 0.49
1.80 0.49
0.110
FEV1 (%)
72.6 16.9
65.3 18.0
0.017
FEV1/FVC (%)
62.5 7.1
62.1 8.2
0.792
PaO2 (torr)
84.7 9.5
83.4 9.2
0.408
PaCO2 (torr)
40.7 3.5
40.1 3.0
0.216
Abbreviations: BMI, body mass index; COPD, chronic obstructive pulmonary disease; FEV1, forced expiratory volume in 1 second; FVC, forced vital capacity; PaCO2, partial pressure of carbon dioxide; PaO2, partial pressure of oxygen; PF, pulmonary fibrosis; PSH, positive smoking history. a Data are presented as mean SD.
Postoperative pulmonary complications and mortality are summarized in ►Table 3. The rate of acute lung injury and of home oxygen therapy introduction was significantly higher in the PF/COPD group. However, 30-day mortality was similar between the groups.
Survival Analyses ►Fig. 1 shows patients’ overall survival following surgery. The cumulative overall survival at 3 and 5 years was 35.8 and 21.4%, respectively, in the PF/COPD group and 64.0 and 55.5%, respectively, in the COPD-alone group (p < 0.0001). The 3and 5-year disease-specific survival was 53.6 and 36.9%, respectively, in the PF/COPD group and 71.4 and 66.0%, respectively, in the COPD-alone group (p ¼ 0.0009) (►Fig. 2). Patients without PF had significantly better survival than patients with PF.
Risk Factors for Long-Term Overall Mortality Multiple logistic regression analysis revealed that increased age, decreased body mass index, advanced pathologic stage, and the existence of PF were independent risk factors for long-term overall mortality (►Table 4).
Discussion We previously reported that lung cancer patients with COPD have poor long-term survival, with survival worsening as the severity of COPD increases.3 In the present study, we focused on the influence of PF on surgical outcomes in patients with COPD and lung cancer. The significant difference between our study population and a population with CPFE is seen in the Thoracic and Cardiovascular Surgeon
Vol. 62
No. 4/2014
results of pulmonary function testing. Pulmonary function testing results can be misleading in patients with CPFE, due to the additional restrictive disorder of PF in patients with airway obstruction. Lung volumes may be preserved because of enlarged emphysematous areas, and patients with CPFE demonstrate preservation of spirometric values and lung volumes,11,12 despite extensive radiographic evidence of lung disease. Patients with PF/COPD may have a greater deterioration in pulmonary function testing than patients with CPFE. The more severe impairment of gas exchange seen in CPFE, manifested as a reduction in the diffusing capacity of the lung for carbon monoxide,13 is likely due to reduced vascular surface area, reduced pulmonary capillary blood volume, and alveolar membrane thickening. Notably, in the present study, patients with PF and emphysema who did not fulfill the FEV1/FVC criterion set for COPD were excluded. Most of our patients showed emphysematous changes on CT in addition to markedly reduced pulmonary function testing results. However, since emphysema should be evaluated by high-resolution CT and this was not performed in all patients, we were not able to definitively diagnose CPFE in the entire study population. In the present study, patients with PF/COPD had more advanced cancer stages than those with COPD alone; the presence of PF may make it difficult to diagnose lung cancer at an early stage because fibrotic change may make a tumor shadow poorly defined. We previously reported that abnormal CT findings and spirometric parameters suggestive of PF and COPD are strong risk factors for lung cancer, even after adjusting for sex, age, and smoking status.14 Patients with CPFE may also have a very high prevalence of lung cancer; the
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
334
Impact of Pulmonary Fibrosis on Postoperative Complications
Sekine et al.
335
Table 2 Perioperative patient characteristics COPD-alone (n ¼ 339)
PF/COPD (n ¼ 41)
25 (7.4%)
2 (4.9%)
Right
10 (40.0)
0 (0)
Left
15 (60)
2 (100)
284 (83.8%)
34 (82.9%)
88 (31.0)
14 (41.2)
Right middle
17 (6.0)
0 (0)
Right lower
44 (15.5)
8 (19.5)
Right upper/middle
7 (2.5)
0 (0)
Right middle/lower
16 (5.6)
2 (4.9)
Left upper
70 (24.6)
8 (19.5)
Left lower
40 (14.1)
2 (4.9)
9 (2.7)
1 (2.4)
21 (6.2)
4 (9.8)
3.5 1.9
4.5 2.3
p value
Operation methods Pneumonectomy
Lobectomy Right upper
Segmentectomy Partial resection Maximal tumor diameter (cm)a
0.794
0.003
Histologic diagnosis Ad
0.746 151 (44.5)
21 (51.2)
Sq
156 (46.0)
17 (41.5)
La
18 (5.3)
1 (2.4)
Others
14 (4.1)
2 (4.9)
1
131 (38.6)
8 (19.5)
2
117 (34.5)
22 (53.7)
3
45 (13.3)
4 (9.8)
4
46 (13.6)
7 (17.1)
Pathologic T status
0.758
Pathologic nodal status
0.097
0
205 (60.5)
1
56 (16.5)
5 (12.2)
2
61 (18.0)
10 (24.4)
3
4 (1.2)
1 (2.4)
13 (3.8)
5 (12.2)
Unknown
20 (48.8)
Pathologic staging IA
0.013 99 (29.2)
5 (12.2)
IB
66 (19.5)
12 (29.3)
IIA
9 (2.7)
0 (0.0)
IIB
47 (13.9)
4 (9.8)
IIIA
54 (15.9)
6 (14.6)
IIIB
40 (11.8)
5 (12.2)
IV
9 (2.7)
2 (4.9)
Unknown
15 (4.4)
7 (17.1)
Abbreviations: Ad, adenocarcinoma; COPD, chronic obstructive pulmonary disease; La, large cell carcinoma; PF, pulmonary fibrosis; Sq, squamous cell carcinoma. Note: Unless otherwise noted, units are given as n (%). a Size was measured on the resected specimen.
Thoracic and Cardiovascular Surgeon
Vol. 62
No. 4/2014
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
Characteristics
Impact of Pulmonary Fibrosis on Postoperative Complications
Sekine et al.
Table 3 Postoperative complications and mortality COPD-alone (n ¼ 339)
PF/COPD (n ¼ 41)
p value
Pneumonia
41 (12.1)
5 (12.2)
1.000
ALI
9 (2.7)
6 (14.6)
0.0026
Bronchial fistula
7 (2.1)
1 (2.4)
0.6024
Characteristics
Empyema
6 (1.8)
1 (2.4)
0.5533
Mechanical ventilation 3 d
13 (3.8)
3 (7.3)
0.3980
HOT
1 (0.3)
3 (7.3)
0.0043
SVT
60 (17.7)
6 (14.6)
0.8272
30-day mortality
4 (1.2)
1 (2.4)
0.4368
Abbreviations: ALI, acute lung injury; COPD, chronic obstructive pulmonary disease; HOT, home oxygen therapy; PF, pulmonary fibrosis; SVT, supraventricular tachycardia. Note: Data are given as n (%).
possibility of malignancy must be retained in the radiological differential diagnosis of pulmonary nodules/masses coexisting with CPFE.15 We found a significantly higher prevalence of acute lung injury in patients with PF/COPD than in those with COPD alone. Saito et al reported that the histopathological finding of interstitial pneumonia is the only predictor of acute respiratory distress syndrome; however, interstitial pneumonia is difficult to identify preoperatively because emphysematous change is also present in the majority of cases, masking the computed-tomography findings.16 We also previously reported that PF confers a higher risk of postoperative complications than does COPD17; the postsurgical outcomes for lung cancer mainly depend on the existence of PF. This suggests that PF, not COPD/emphysema, induces postoperative acute lung injury. There are several limitations and biases in this study. First, all lung cancer patients underwent surgery, which means that patients with markedly advanced COPD and/or PF in whom surgery was contraindicated were not included. Therefore, the participants’ blood gas analyses were not severely compromised when compared with reported values for patients with CPFE.5,18 Second, several reports have shown that the
prevalence of COPD in lung cancer patients varies from 8 to 50%.19,20 Our study revealed that 26% (380/1,461) in surgical patients was diagnosed with COPD. This was relatively lower than expected. This may be because inclusion criteria were based on the pulmonary function test and not on radiological emphysema and pathological airway obstruction and alveolar destruction. Third, operative procedures have progressed throughout the 1990s and the 2000s. In particular, surgical stress to the thoracic wall has been dramatically reduced since the introduction of video-assisted thoracoscopic surgery. This change may have an influence on postoperative complications. In the present study, there was no difference in the operative procedures performed between the groups. In general, limited pulmonary resections are performed in patients with low pulmonary function to reduce the risk of postoperative complications. However, the existence of PF did not influence the operative method in this study. Fourth, we could not confirm some of PF by pathology because resected specimens were different from the areas of radiological fibrosis. Finally, because an inadequate number of patients with PF/COPD were compared with a larger number of patients with COPD alone, some analyses may not have been powered for statistical significance.
Fig. 1 Overall survival after pulmonary resection for lung cancer. The cumulative overall survival at 3 and 5 years was 35.8 and 21.4%, respectively, in the PF/COPD group and 64.0 and 55.5%, respectively, in the COPD-alone group (p
The Impact of Combined Pulmonary Fibrosis and Chronic Obstructive Pulmonary Disease on Long-Term Survival after Lung Cancer Surgery Yasuo Sekine1 Yuichi Sakairi2 Ichiro Yoshino2
Mitsuru Yoshino1
1 Department of Thoracic Surgery, Tokyo Women’s Medical University,
Yachiyo, Chiba, Japan 2 Department of Thoracic Surgery, Chiba University Graduate School of Medicine, Chiba, Japan
Eitetsu Koh1
Atsushi Hata1
Hidemi Suzuki2
Address for correspondence Yasuo Sekine, MD, PhD, Department of Thoracic Surgery, Tokyo Women’s Medical University, Yachio Medical Center, 477-96 Owadashinden, Yachiyo, Chiba 276-8524, Japan (e-mail: [email protected]).
Thorac Cardiovasc Surg 2014;62:332–337.
Abstract
Keywords
► lung cancer ► chronic obstructive pulmonary disease ► pulmonary fibrosis ► combined pulmonary fibrosis and emphysema
Purpose The purpose of this study was to determine the impact of pulmonary fibrosis (PF) on postoperative complications and on long-term survival after surgical resection in lung cancer patients with chronic obstructive pulmonary disease (COPD). Patients and Methods A retrospective chart review was conducted of 380 patients with COPD who had undergone pulmonary resection for lung cancer at the University Hospital between 1990 and 2005. The definition of COPD was a preoperative forced expiratory volume in 1 second/forced vital capacity (FEV1/FVC) ratio of less than 70%; PF was defined as obvious bilateral fibrous change in the lower lung fields, confirmed by computed tomography. Results PF was present in 41 patients (10.8%) with COPD; the remaining 339 patients (89.2%) did not have PF. The preoperative FVC/FEV1 was significantly lower in the group of patients with PF than in the group without (p < 0.05). Acute lung injury and home oxygen therapy were significantly more common in the PF group; however, the 30-day mortality was similar between the groups. The cumulative survival at 3 and 5 years was 53.6 and 36.9%, respectively, in the PF group and 71.4 and 66.1%, respectively, in the non-PF group (p ¼ 0.0009). Increased age, decreased body mass index, advanced pathologic stage, and the existence of PF were identified as independent risk factors for decreased survival. Conclusion PF is a risk factor for decreased survival after surgical treatment in lung cancer patients with COPD.
Introduction It is well known that pulmonary fibrosis (PF) and chronic obstructive pulmonary disease (COPD) are significant risk factors for lung cancer; COPD comprises both pulmonary emphysema and chronic bronchitis.1–3 Cigarette smoking and occupational exposures, among other environmental
received August 19, 2013 accepted after revision November 7, 2013 published online May 30, 2014
hazards, are common risk factors for both PF and COPD. Combined pulmonary fibrosis and emphysema (CPFE) is a clinical and radiological diagnosis that indicates the presence of pulmonary emphysema in the upper lungs and interstitial pneumonia in the lower lungs; various radiological patterns are possible.4 Patients with CPFE may be at significantly increased risk for lung cancer.5 When CPFE patients do indeed
© 2014 Georg Thieme Verlag KG Stuttgart · New York
DOI http://dx.doi.org/ 10.1055/s-0033-1363478. ISSN 0171-6425.
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
332
have lung cancer, survival is significantly worse than in lung cancer patients with emphysema alone.6 The majority of patients with COPD do not have end-stage emphysema, but rather some combination of emphysema and chronic bronchitis, producing varying degrees of airflow obstruction. We previously reported that patients with higher COPD grades have worse long-term survival overall,7 and patients with interstitial lung disease have poor longterm survival after lung cancer surgery.8 We hypothesized that patients with combined PF and COPD may have significantly inferior long-term survival after lung cancer surgery, compared with those who have COPD alone. The majority of patients with CPFE may have advancedstage airway obstruction. As the imaging features of CPFE have not been fully described9 and it is very difficult to define the severity of emphysema radiographically, we focused our investigation on patients with combined PF and the easier-todefine COPD. We determined the clinical features and the short- and long-term outcomes of lung cancer patients with COPD, with and without PF.
Patients and Methods This study was approved by the Institutional Review Board of University Hospital. The requirement for informed consent was waived because of the study’s retrospective nature. Between January 1990 and April 2005, 1,461 patients with lung cancer underwent pulmonary resection at the University Hospital. Of those, 380 were diagnosed with COPD based on preoperative spirometric data. COPD was defined as a preoperative forced expiratory volume in 1 second (FEV1)/forced vital capacity (FVC) ratio of less than 70%.10 Data were collected from our institutional cancer registry database and from patient follow-up visit records, and consisted of information on preoperative patient characteristics, disease status, operative procedures, postoperative complications, pathologic diagnoses, and follow-up examinations. Prospective follow-up data were collected between April 2005 and August 2011. Preoperative conventional or high-resolution chest computed tomography (CT) was evaluated in all patients to determine not only the extent of each patient’s cancer but also the existence of PF. When abnormalities were seen that were compatible with bilateral lung fibrosis in the lower lobes, such as peripheral reticular opacities, PF was diagnosed. Chest CT was reviewed by independent pulmonologists. Diffusing capacity (DLco) is critically important for identifying the severity of COPD and PF. However, because this study was retrospective in nature, DLco could not be obtained from all patients. Pathological confirmation of PF could not be universally obtained, as some patients had upper-lobe lung cancer and therefore lower-lobe lung tissue with PF was not resected. Patients were divided into two groups: the PF/COPD group and the COPD-alone group. Postoperative morbidities included bacterial pneumonia (confirmed by infiltrative shadows on chest radiography, positive sputum culture, temperature 37.5°C, and a white blood cell count > 10,000/μL), acute lung injury (aggravation
Sekine et al.
of dyspnea on exertion, deterioration of arterial blood gases, and diffuse interstitial abnormalities on radiography compatible with acute interstitial pneumonia), prolonged mechanical ventilation ( 3 days), bronchial stump dehiscence, empyema, tracheostomy, and postoperative home oxygen therapy. Home oxygen therapy was given to patients with a partial pressure of oxygen < 60 torr, with exercise, at the time of hospital discharge.8 After discharge, patients visited our outpatient clinic every 1 to 6 months, unless tumor recurrence or any other problems occurred. The presence of tumor recurrence or metastasis was evaluated every 6 to 12 months by chest CT, brain magnetic resonance imaging, 2-[18F]fluoro-2-deoxy-D-glucose positron emission tomography, bone scintigraphy, and serum tumor markers. When patients returned to their referring doctor or were referred on to other doctors, their status and cause of death, if applicable, were confirmed every 2 years by mail or telephone with the patient, the patient’s family, or the referring doctors. Overall and disease-specific survival was analyzed for each patient group.
Statistical Analysis Data were analyzed using Statistical Package for the Social Sciences, version 20 (IBM Corp., New York, New York, United States). To compare differences between the PF/COPD and COPD-alone groups, Mann–Whitney U-test was used for continuous variables and the chi-square test or Fisher exact test was used for categorical variables. Survival curves were estimated using the Kaplan–Meier method, and the difference in survival times between the groups was calculated by the log-rank test. We estimated the overall survival and the disease-specific survival that indicates the percentage of patients in each group who have not died from a lung cancer in a defined period of time. All preoperative and intraoperative variables including sex, age, body mass index, pathologic type and stage, existence of PF, operative time, and operative blood loss were entered into a logistic regression model in a stepwise manner to identify variables that were independently predictive of long-term survival. A p-value of less than 0.05 was considered to be significant.
Results Patient Characteristics and Postoperative Mortality and Morbidity A total of 41 patients (10.8%) were diagnosed with PF (PF/COPD group); the remaining 339 patients (89.2%) were determined not to have PF (COPD-alone group). The patients’ preoperative characteristics are summarized in ►Table 1. The preoperative FVC and FEV1 were significantly lower in the PF/COPD group than in the COPD-alone group (p < 0.05). Perioperative characteristics are listed in ►Table 2. Although the operative methods were similar between the groups, the maximal tumor size and pathological staging were more advanced in the PF/COPD group. Interestingly, there were more lung cancers in upper lobe than in lower lobe in both groups. Thoracic and Cardiovascular Surgeon
Vol. 62
No. 4/2014
333
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
Impact of Pulmonary Fibrosis on Postoperative Complications
Impact of Pulmonary Fibrosis on Postoperative Complications
Sekine et al.
Table 1 Preoperative patient characteristics COPD-alone (n ¼ 339)
PF/COPD (n ¼ 41)
p value
Male (n, %)
312 (92.0)
38 (92.7)
1.00
Female
27 (8.0)
3 (7.3)
68.0 7.2
66.3 7.8
0.191
21.9 2.9
21.7 3.0
0.589
317 (93.8)
39 (95.1)
1.00
52.6 33.0
49.6 24.4
0.483
FVC (L)
3.08 0.68
2.88 0.64
0.069
FVC (%)
99.2 32.9
87.0 22.0
0.010
Characteristics Sex
Age (y)
a
BMI (kg/m2)a PSH (n, %) a
Smoking index (pack-years) Pulmonary function testsa
FEV1 (L)
1.93 0.49
1.80 0.49
0.110
FEV1 (%)
72.6 16.9
65.3 18.0
0.017
FEV1/FVC (%)
62.5 7.1
62.1 8.2
0.792
PaO2 (torr)
84.7 9.5
83.4 9.2
0.408
PaCO2 (torr)
40.7 3.5
40.1 3.0
0.216
Abbreviations: BMI, body mass index; COPD, chronic obstructive pulmonary disease; FEV1, forced expiratory volume in 1 second; FVC, forced vital capacity; PaCO2, partial pressure of carbon dioxide; PaO2, partial pressure of oxygen; PF, pulmonary fibrosis; PSH, positive smoking history. a Data are presented as mean SD.
Postoperative pulmonary complications and mortality are summarized in ►Table 3. The rate of acute lung injury and of home oxygen therapy introduction was significantly higher in the PF/COPD group. However, 30-day mortality was similar between the groups.
Survival Analyses ►Fig. 1 shows patients’ overall survival following surgery. The cumulative overall survival at 3 and 5 years was 35.8 and 21.4%, respectively, in the PF/COPD group and 64.0 and 55.5%, respectively, in the COPD-alone group (p < 0.0001). The 3and 5-year disease-specific survival was 53.6 and 36.9%, respectively, in the PF/COPD group and 71.4 and 66.0%, respectively, in the COPD-alone group (p ¼ 0.0009) (►Fig. 2). Patients without PF had significantly better survival than patients with PF.
Risk Factors for Long-Term Overall Mortality Multiple logistic regression analysis revealed that increased age, decreased body mass index, advanced pathologic stage, and the existence of PF were independent risk factors for long-term overall mortality (►Table 4).
Discussion We previously reported that lung cancer patients with COPD have poor long-term survival, with survival worsening as the severity of COPD increases.3 In the present study, we focused on the influence of PF on surgical outcomes in patients with COPD and lung cancer. The significant difference between our study population and a population with CPFE is seen in the Thoracic and Cardiovascular Surgeon
Vol. 62
No. 4/2014
results of pulmonary function testing. Pulmonary function testing results can be misleading in patients with CPFE, due to the additional restrictive disorder of PF in patients with airway obstruction. Lung volumes may be preserved because of enlarged emphysematous areas, and patients with CPFE demonstrate preservation of spirometric values and lung volumes,11,12 despite extensive radiographic evidence of lung disease. Patients with PF/COPD may have a greater deterioration in pulmonary function testing than patients with CPFE. The more severe impairment of gas exchange seen in CPFE, manifested as a reduction in the diffusing capacity of the lung for carbon monoxide,13 is likely due to reduced vascular surface area, reduced pulmonary capillary blood volume, and alveolar membrane thickening. Notably, in the present study, patients with PF and emphysema who did not fulfill the FEV1/FVC criterion set for COPD were excluded. Most of our patients showed emphysematous changes on CT in addition to markedly reduced pulmonary function testing results. However, since emphysema should be evaluated by high-resolution CT and this was not performed in all patients, we were not able to definitively diagnose CPFE in the entire study population. In the present study, patients with PF/COPD had more advanced cancer stages than those with COPD alone; the presence of PF may make it difficult to diagnose lung cancer at an early stage because fibrotic change may make a tumor shadow poorly defined. We previously reported that abnormal CT findings and spirometric parameters suggestive of PF and COPD are strong risk factors for lung cancer, even after adjusting for sex, age, and smoking status.14 Patients with CPFE may also have a very high prevalence of lung cancer; the
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
334
Impact of Pulmonary Fibrosis on Postoperative Complications
Sekine et al.
335
Table 2 Perioperative patient characteristics COPD-alone (n ¼ 339)
PF/COPD (n ¼ 41)
25 (7.4%)
2 (4.9%)
Right
10 (40.0)
0 (0)
Left
15 (60)
2 (100)
284 (83.8%)
34 (82.9%)
88 (31.0)
14 (41.2)
Right middle
17 (6.0)
0 (0)
Right lower
44 (15.5)
8 (19.5)
Right upper/middle
7 (2.5)
0 (0)
Right middle/lower
16 (5.6)
2 (4.9)
Left upper
70 (24.6)
8 (19.5)
Left lower
40 (14.1)
2 (4.9)
9 (2.7)
1 (2.4)
21 (6.2)
4 (9.8)
3.5 1.9
4.5 2.3
p value
Operation methods Pneumonectomy
Lobectomy Right upper
Segmentectomy Partial resection Maximal tumor diameter (cm)a
0.794
0.003
Histologic diagnosis Ad
0.746 151 (44.5)
21 (51.2)
Sq
156 (46.0)
17 (41.5)
La
18 (5.3)
1 (2.4)
Others
14 (4.1)
2 (4.9)
1
131 (38.6)
8 (19.5)
2
117 (34.5)
22 (53.7)
3
45 (13.3)
4 (9.8)
4
46 (13.6)
7 (17.1)
Pathologic T status
0.758
Pathologic nodal status
0.097
0
205 (60.5)
1
56 (16.5)
5 (12.2)
2
61 (18.0)
10 (24.4)
3
4 (1.2)
1 (2.4)
13 (3.8)
5 (12.2)
Unknown
20 (48.8)
Pathologic staging IA
0.013 99 (29.2)
5 (12.2)
IB
66 (19.5)
12 (29.3)
IIA
9 (2.7)
0 (0.0)
IIB
47 (13.9)
4 (9.8)
IIIA
54 (15.9)
6 (14.6)
IIIB
40 (11.8)
5 (12.2)
IV
9 (2.7)
2 (4.9)
Unknown
15 (4.4)
7 (17.1)
Abbreviations: Ad, adenocarcinoma; COPD, chronic obstructive pulmonary disease; La, large cell carcinoma; PF, pulmonary fibrosis; Sq, squamous cell carcinoma. Note: Unless otherwise noted, units are given as n (%). a Size was measured on the resected specimen.
Thoracic and Cardiovascular Surgeon
Vol. 62
No. 4/2014
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
Characteristics
Impact of Pulmonary Fibrosis on Postoperative Complications
Sekine et al.
Table 3 Postoperative complications and mortality COPD-alone (n ¼ 339)
PF/COPD (n ¼ 41)
p value
Pneumonia
41 (12.1)
5 (12.2)
1.000
ALI
9 (2.7)
6 (14.6)
0.0026
Bronchial fistula
7 (2.1)
1 (2.4)
0.6024
Characteristics
Empyema
6 (1.8)
1 (2.4)
0.5533
Mechanical ventilation 3 d
13 (3.8)
3 (7.3)
0.3980
HOT
1 (0.3)
3 (7.3)
0.0043
SVT
60 (17.7)
6 (14.6)
0.8272
30-day mortality
4 (1.2)
1 (2.4)
0.4368
Abbreviations: ALI, acute lung injury; COPD, chronic obstructive pulmonary disease; HOT, home oxygen therapy; PF, pulmonary fibrosis; SVT, supraventricular tachycardia. Note: Data are given as n (%).
possibility of malignancy must be retained in the radiological differential diagnosis of pulmonary nodules/masses coexisting with CPFE.15 We found a significantly higher prevalence of acute lung injury in patients with PF/COPD than in those with COPD alone. Saito et al reported that the histopathological finding of interstitial pneumonia is the only predictor of acute respiratory distress syndrome; however, interstitial pneumonia is difficult to identify preoperatively because emphysematous change is also present in the majority of cases, masking the computed-tomography findings.16 We also previously reported that PF confers a higher risk of postoperative complications than does COPD17; the postsurgical outcomes for lung cancer mainly depend on the existence of PF. This suggests that PF, not COPD/emphysema, induces postoperative acute lung injury. There are several limitations and biases in this study. First, all lung cancer patients underwent surgery, which means that patients with markedly advanced COPD and/or PF in whom surgery was contraindicated were not included. Therefore, the participants’ blood gas analyses were not severely compromised when compared with reported values for patients with CPFE.5,18 Second, several reports have shown that the
prevalence of COPD in lung cancer patients varies from 8 to 50%.19,20 Our study revealed that 26% (380/1,461) in surgical patients was diagnosed with COPD. This was relatively lower than expected. This may be because inclusion criteria were based on the pulmonary function test and not on radiological emphysema and pathological airway obstruction and alveolar destruction. Third, operative procedures have progressed throughout the 1990s and the 2000s. In particular, surgical stress to the thoracic wall has been dramatically reduced since the introduction of video-assisted thoracoscopic surgery. This change may have an influence on postoperative complications. In the present study, there was no difference in the operative procedures performed between the groups. In general, limited pulmonary resections are performed in patients with low pulmonary function to reduce the risk of postoperative complications. However, the existence of PF did not influence the operative method in this study. Fourth, we could not confirm some of PF by pathology because resected specimens were different from the areas of radiological fibrosis. Finally, because an inadequate number of patients with PF/COPD were compared with a larger number of patients with COPD alone, some analyses may not have been powered for statistical significance.
Fig. 1 Overall survival after pulmonary resection for lung cancer. The cumulative overall survival at 3 and 5 years was 35.8 and 21.4%, respectively, in the PF/COPD group and 64.0 and 55.5%, respectively, in the COPD-alone group (p
