A proposal for a comprehensive risk scoring system for predicting postoperative complications in octogenarian patients with medically operable lung cancer: JACS1303

A proposal for a comprehensive risk scoring system for predicting postoperative complications in... Abstract OBJECTIVES Although some retrospective studies have reported clinicopathological scoring systems for predicting postoperative complications and survival outcomes for elderly lung cancer patients, optimized scoring systems remain controversial. METHODS The Japanese Association for Chest Surgery (JACS) conducted a nationwide multicentre prospective cohort and enrolled a total of 1019 octogenarians with medically operable lung cancer. Details of the clinical factors, comorbidities and comprehensive geriatric assessment were recorded for 895 patients to develop a comprehensive risk scoring (RS) system capable of predicting severe complications. RESULTS Operative (30 days) and hospital mortality rates were 1.0% and 1.6%, respectively. Complications were observed in 308 (34%) patients, of whom 81 (8.4%) had Grade 3–4 severe complications. Pneumonia was the most common severe complication, observed in 27 (3.0%) patients. Five predictive factors, gender, comprehensive geriatric assessment75: memory and Simplified Comorbidity Score (SCS): diabetes mellitus, albumin and percentage vital capacity, were identified as independent predictive factors for severe postoperative complications (odds ratio = 2.73, 1.86, 1.54, 1.66 and 1.61, respectively) through univariate and multivariate analyses. A 5-fold cross-validation was performed as an internal validation to reconfirm these 5 predictive factors (average area under the curve 0.70). We developed a simplified RS system as follows: RS = 3 (gender: male) + 2 (comprehensive geriatric assessment 75: memory: yes) + 2 (albumin: <3.8 ng/ml) + 1 (percentage vital capacity: ≤90) + 1 (SCS: diabetes mellitus: yes). CONCLUSIONS The current series shows that octogenarians can be successfully treated for lung cancer with surgical resection with an acceptable rate of severe complications and mortality. We propose a simplified RS system to predict severe complications in octogenarian patients with medically operative lung cancer. Trial Registration Number JACS1303 (UMIN000016756) Lung cancer, Octogenarian, Postoperative complication, Risk scoring system INTRODUCTION Lung cancer is the leading cause of cancer-related deaths worldwide and is a particularly prominent disease in the elderly population [1]. Its incidence in this population is expected to increase as the general population ages, especially in developed countries. Clinicians are increasingly confronted with octogenarians with medically operable non-small-cell lung cancer (NSCLC). The total number of surgical resections for primary lung cancer in Japan reported most recently in 2014 has reached 38 085, showing a steady increase over the years [2]. Among those who underwent surgical resections for this condition, 53.5% of patients were aged 70 years or older and 12.1% of patients were aged 80 years or older. This percentage has continued to increase because the ageing of Japan’s population is becoming more pronounced. The Japanese Joint Committee of Lung Cancer Registry previously conducted and reported a large retrospective cohort study to identify the predictors of long-term survival and the risk factors for complications after surgical resection in octogenarian patients with clinical Stage I NSCLC [3]. Although the operative morbidity and mortality for patients aged 80 years and older were reported to be 8.4% and 1.4%, respectively, and were satisfactorily low and safe compared with 4.5% and 0.8%, respectively, for all patients [4], one-third of the patients died from other causes, such as an illness during the postoperative follow-up period. Comorbidity is a factor associated with both postoperative complications and worse prognosis. However, these results were still limited and based on retrospective cohort studies. Therefore, the Japanese Association for Chest Surgery (JACS) conducted a nationwide multicentre large-scale prospective observational cohort study, JACS1303 (UMIN000016756) as previously described by Saji et al. [5]. Although some retrospective studies have reported clinicopathological scoring systems, such as the Charlson Comorbidity Index (CCI), Simplified Comorbidity Score (SCS) and Glasgow Prognostic Score (GPS) for predicting postoperative complications and survival outcomes for elderly lung cancer patients [6–8], optimized scoring systems and care strategies remain controversial. The first analysis of the JACS1303 study aims to evaluate details of the clinical factors, comorbidities and comprehensive geriatric assessment (CGA) to develop a comprehensive operative risk scoring (RS) system that can predict severe morbidity and mortality for octogenarian patients with medically operative lung cancer. MATERIALS AND METHODS Patient eligibility Octogenarian patients with medically operable lung cancer were enrolled in this cohort. Additional inclusion criteria were as follows: no prior chemotherapy or radiation therapy for lung cancer, sufficient organ function, measurable preoperative comprehensive geriatric assessment and written informed consent. Between April 2015 and December 2016, 1019 octogenarian patients were enrolled from 82 institutions (Fig. 1). Of the 962 patients enrolled, we finally selected 895 patients who underwent surgical resection for primary lung cancer and had sufficient clinical records for comprehensive surgical RS analysis of predictive postoperative complications for octogenarian patients with medically operable lung cancer. Figure 1: View largeDownload slide The CONSORT diagram. CCI: Charlson Comorbidity Index; CGA: comprehensive geriatric assessment; GPS: Glasgow Prognostic Score; SCS: Simplified Comorbidity Score. Figure 1: View largeDownload slide The CONSORT diagram. CCI: Charlson Comorbidity Index; CGA: comprehensive geriatric assessment; GPS: Glasgow Prognostic Score; SCS: Simplified Comorbidity Score. Data collection Data collection and analyses were approved by the institutional review boards. Once informed consent was obtained, recruiting clinicians completed a case report form to record the following clinicopathological factors: age, gender, smoking status, tumour size and location, histological subtypes according to the World Health Organization (WHO) classification [9] and staging according to the seventh edition of the tumour, necrosis and metastasis (TNM) classification of lung and pleural tumours [10]. The status of the patients was recorded as follows: performance status, pulmonary function including vital capacity (VC), %VC, forced expiratory volume in 1 s, and percentage forced expiratory volume in 1 s. The comorbidity index was calculated using the CCI [11] and SCS [12]. The GPS [13] and CGA 7 [14] were also recorded. Surgical-related data were also collected after the completion of surgery. Clinical outcome data related to operative morbidity and mortality, and cancer-related or non-related prognosis were collected 30 days and 90 days after the surgical treatment. Perioperative and postoperative morbidity and adverse events were classified according to both the Common Terminology Criteria for Adverse Events version 4.0. Operative mortality was defined as death at any time during the initial hospitalization or within 30 days after the operation. Statistical analysis The aim of this study was to develop a preoperative risk prediction model for octogenarian patients with medically operative lung cancer as previously described by Saji et al. [5]. This would require evaluation of the risk of Grade 3 or higher adverse events after surgical treatment using logistic regression models with a maximum of 5 predictive risk factors. For this purpose, we needed to observe at least 50 adverse events. The planned sample size of 750 elderly patients aged 80 years and older was determined and based on the assumption that the proportion of adverse events in this cohort would be in the range of 7–8%. Analyses were performed using JMP® Pro version 13.0 (SAS Institute Inc., Cary, NC, USA). RESULTS Postoperative complications and mortalities The clinicopathological and surgical characteristics of the patients are listed in Table 1. The variables of postoperative early complications (30 days) and operative and hospital mortality are listed in Table 2. Complications of any severity were observed in 308 (34%) patients, of whom 81 (8.4%) had Grade 3–4 severe complications, which required therapeutic intervention. Pneumonia was the most common severe complication observed in 27 (3.0%) patients, followed by prolonged air leakage (1.4%), atelectasis, bronchial fistula, empyema, wound infection, cerebral infarction, subcutaneous emphysema, atrial fibrillation and gastrointestinal disorders. The major common complications, any grade, were prolonged air leakage (8.4%), atrial fibrillation (5.0%), pneumonia (4.1%), subcutaneous emphysema (2.8%), atelectasis (2.1%) and wound infection (1%). Operative (30 days) and hospital mortality rates were 1.0% and 1.6%, respectively. Three patients died due to postoperative pneumonia, including 2 cases of acute exacerbation of interstitial pneumonia and 1 case of aspiration pneumonia. One patient each died due to cerebral infarction, perioperative major haemorrhage, myocardial infarction, liver dysfunction due to liver cirrhosis, myelodysplastic syndromes and cardiopulmonary arrest due to pneumothorax on the 1st day after hospital discharge. Table 1: Characteristics of eligible octogenarian patients (n = 895) Variables    Age, mean (range)  82.7 (80–96)   80–84/85–89/ND, n (%)  716 (80)/171 (19)/8 (1)  Gender     Male/female, n (%)  540 (60)/355 (40)  Performance status     0/1/2/3 or 4/ND, n (%)  661 (74)/207 (23)/17 (2)/4/6  Smoking status     Current smoker/ex-smoker/never smoker, n (%)  47 (5)/503 (56)/345 (38)  Pack-year smoking (smokers, n = 550), mean ± SD  48.4 ± 30.6   ≥20, n (%)  462 (84)  Pulmonary function, mean ± SD     VC (l)/%VC  2.7 ± 0.7/101.1 ± 17.7   FEV1.0 (l)/FEV1.0 (%)  1.8 ± 0.5/73.0 ± 14.7  Comorbidity (yes/no), n (%)  687 (76)/208 (24)   Respiratory disease  340 (38)   History of malignant disease  267 (30)   Diabetes mellitus  173 (19)   Cerebroneural disease  111 (12)   Renal insufficiancy  80 (9)   Cardiovascular disease  42 (5)  Charlson Comorbidity Index, mean ± SD  1.7 ± 1.6  Simplified Comorbidity Score, mean ± SD  8.5 ± 4.9  Glasgow Prognostic Score (0/1/2), n (%)  791 (88)/84 (9)/20 (3)  Comprehensive geriatric assessment 7, mean ± SD  6.4 ± 1.0  Tumour site, n (%)     RUL/RML/RLL  254 (28)/52 (6)/211 (24)   LUL/LLL  232 (26)/146 (16)  Tumour size (cm), mean ± SD  3.0 ± 1.5  Histology, n (%)     Adenocarcinoma/squamous cell carcinoma  607 (68)/194 (22)   Others/ND  75 (8)/19 (2)  Tumour factor, n (%)     pT1a/pT1b/pT2a/pT2b  238 (27)/195 (22)/329 (37)/45 (5)   pT3/pT4/ND  63 (7)/11 (1)/14 (1)  Nodal factor, n (%)     pN0/pN1/pN2/pN3/ND  752 (84)/68 (8)/59 (7)/1/15 (1)  Pathological staging, n (%)     IA/IB/IIA/IIB  400 (45)/267 (30)/80 (9)/49 (5)   IIIA/IIIB/IV/ND  70 (8)/1/13 (1)/15 (1)  Surgical procedures, n (%)     Lobectomy/pneumonectomy  623 (70)/3   Segmental resection/ wedge resection/ND  110 (12)/151 (17)/8 (1)  Mediastinal lymph nodal dissection, n (%)   Systematic/selective  113 (13)/346 (39)   Sampling or none/ND  430 (48)/6  Thoracic approach, n (%)     Open- or minithoracotomy  482 (54)   Video-assisted thoracotomy/ND  400 (45)/13 (1)  Intraoperative factors     Operation time (min), median (range)  175 (33–1233)   Blood loss (ml), median (range)  30 (0–57600)   Blood transfusion, n (%)  26 (3)  Chest tube placement (days), median (range)  3 (0–23)  Hospital days after operation, median (range)  9 (1–91)  Curative resection     R0/R1-2/ND, n (%)  828 (93)/41 (5)/26 (3)  Variables    Age, mean (range)  82.7 (80–96)   80–84/85–89/ND, n (%)  716 (80)/171 (19)/8 (1)  Gender     Male/female, n (%)  540 (60)/355 (40)  Performance status     0/1/2/3 or 4/ND, n (%)  661 (74)/207 (23)/17 (2)/4/6  Smoking status     Current smoker/ex-smoker/never smoker, n (%)  47 (5)/503 (56)/345 (38)  Pack-year smoking (smokers, n = 550), mean ± SD  48.4 ± 30.6   ≥20, n (%)  462 (84)  Pulmonary function, mean ± SD     VC (l)/%VC  2.7 ± 0.7/101.1 ± 17.7   FEV1.0 (l)/FEV1.0 (%)  1.8 ± 0.5/73.0 ± 14.7  Comorbidity (yes/no), n (%)  687 (76)/208 (24)   Respiratory disease  340 (38)   History of malignant disease  267 (30)   Diabetes mellitus  173 (19)   Cerebroneural disease  111 (12)   Renal insufficiancy  80 (9)   Cardiovascular disease  42 (5)  Charlson Comorbidity Index, mean ± SD  1.7 ± 1.6  Simplified Comorbidity Score, mean ± SD  8.5 ± 4.9  Glasgow Prognostic Score (0/1/2), n (%)  791 (88)/84 (9)/20 (3)  Comprehensive geriatric assessment 7, mean ± SD  6.4 ± 1.0  Tumour site, n (%)     RUL/RML/RLL  254 (28)/52 (6)/211 (24)   LUL/LLL  232 (26)/146 (16)  Tumour size (cm), mean ± SD  3.0 ± 1.5  Histology, n (%)     Adenocarcinoma/squamous cell carcinoma  607 (68)/194 (22)   Others/ND  75 (8)/19 (2)  Tumour factor, n (%)     pT1a/pT1b/pT2a/pT2b  238 (27)/195 (22)/329 (37)/45 (5)   pT3/pT4/ND  63 (7)/11 (1)/14 (1)  Nodal factor, n (%)     pN0/pN1/pN2/pN3/ND  752 (84)/68 (8)/59 (7)/1/15 (1)  Pathological staging, n (%)     IA/IB/IIA/IIB  400 (45)/267 (30)/80 (9)/49 (5)   IIIA/IIIB/IV/ND  70 (8)/1/13 (1)/15 (1)  Surgical procedures, n (%)     Lobectomy/pneumonectomy  623 (70)/3   Segmental resection/ wedge resection/ND  110 (12)/151 (17)/8 (1)  Mediastinal lymph nodal dissection, n (%)   Systematic/selective  113 (13)/346 (39)   Sampling or none/ND  430 (48)/6  Thoracic approach, n (%)     Open- or minithoracotomy  482 (54)   Video-assisted thoracotomy/ND  400 (45)/13 (1)  Intraoperative factors     Operation time (min), median (range)  175 (33–1233)   Blood loss (ml), median (range)  30 (0–57600)   Blood transfusion, n (%)  26 (3)  Chest tube placement (days), median (range)  3 (0–23)  Hospital days after operation, median (range)  9 (1–91)  Curative resection     R0/R1-2/ND, n (%)  828 (93)/41 (5)/26 (3)  FEV1.0: forced expiratory volume in 1 s; FEV1.0%: percentage of forced expiratory volume in 1 s; LLL: left lower lobe; LUL: left upper lobe; ND: no data; RLL: right lower lobe; RML: right middle lobe; RUL: right upper lobe; SD: standard deviation; VC: vital capacity. Table 2: Postoperative early complications of octogenarian patients with operable lung cancer (n = 895) Postoperative complications 30 days  Grade 1  Grade 2  Grade 3  Grade 4  Grade 3–4  Grade 5  Pneumonia  0  9  22  5  27 (3.0)  3  Prolonged air leakage  14  49  11  2  13 (1.4)  0  Atelectasis  3  11  4  1  5  0  Bronchial fistula  0  0  3  2  5  0  Empyema  0  3  4  1  5  0  Wound infection  0  5  4  0  4  0  Cerebral infarction  0  4  1  2  3  1  Subcutaneous emphysema  18  6  1  1  2  0  Atrial fibrillation  6  35  2  0  2  0  Pulmonary embolism  1  1  2  0  2  0  Gastrointestinal disorders  1  1  2  0  2  0  Paroxysmal supraventricular tachycardia  5  5  1  0  1  0  Postoperative haemorrhage  3  2  1  0  1  1  Ventricular tachycardia  2  3  0  0  0  0  Recurrent nerve palsy  1  5  0  0  0  0  Myocardial infarction  0  0  0  0  1  1  Ventricular fibrillation  0  4  0  0  0  0  Chylothorax  0  3  0  0  0  0  Others  2  23  4  5  9  3  Any complications (n = 306, 34%)  56  169  62  19  81 (8.4)    Operative mortality (30 days)            9 (1.0)  Hospital mortality            14 (1.6)  Postoperative complications 30 days  Grade 1  Grade 2  Grade 3  Grade 4  Grade 3–4  Grade 5  Pneumonia  0  9  22  5  27 (3.0)  3  Prolonged air leakage  14  49  11  2  13 (1.4)  0  Atelectasis  3  11  4  1  5  0  Bronchial fistula  0  0  3  2  5  0  Empyema  0  3  4  1  5  0  Wound infection  0  5  4  0  4  0  Cerebral infarction  0  4  1  2  3  1  Subcutaneous emphysema  18  6  1  1  2  0  Atrial fibrillation  6  35  2  0  2  0  Pulmonary embolism  1  1  2  0  2  0  Gastrointestinal disorders  1  1  2  0  2  0  Paroxysmal supraventricular tachycardia  5  5  1  0  1  0  Postoperative haemorrhage  3  2  1  0  1  1  Ventricular tachycardia  2  3  0  0  0  0  Recurrent nerve palsy  1  5  0  0  0  0  Myocardial infarction  0  0  0  0  1  1  Ventricular fibrillation  0  4  0  0  0  0  Chylothorax  0  3  0  0  0  0  Others  2  23  4  5  9  3  Any complications (n = 306, 34%)  56  169  62  19  81 (8.4)    Operative mortality (30 days)            9 (1.0)  Hospital mortality            14 (1.6)  Predictive factors for severe postoperative complications Univariate analysis was carried out to identify 17 possible predictive factors (P < 0.25) for severe postoperative complications. We removed 8 factors including total score of SCS, GPS, CGA7, tumour side, tumour location, maximum standardized uptake value, CGA76 and CGA77 from the clinical view and consequently identified gender, smoking history, CGA75: memory, SCS: diabetes mellitus, SCS: respiratory disease, pack-year, albumin (Alb), C-reactive protein and %VC as candidate predictive factors. To develop a logistic regression model with a maximum of 5 risk factors, we applied forward selection with the corrected Akaike information criteria [15] and identified the following 5 predictive factors: gender, CGA75: memory, SCS: diabetes mellitus, Alb and %VC (Table 3). Additionally, 5-fold cross-validation was performed as an internal validation to reconfirm these 5 predictive factors. The discriminative ability of the model was evaluated using the area under the receiver operator curve (AUC, concordance statistic). The AUCs for 5-fold cross-validation were 0.64, 0.75, 0.71, 0.67 and 0.64, and the average AUC was 0.68. Table 3: Univariate and multivariate analysis of predicting for severe operative complications Variables  Univariate analysis   Multivariate analysis   P-value  Odds ratio  95% CI  P-value  Odds ratio  95% CI  Male (vs female)  <0.001  3.23  1.74–5.98  <0.001  2.73  1.50–5.31  Smoking (vs no smoking)  0.001  2.54  1.42–4.57  Not included in multivariate analysis  CGA75: memory (vs positive)  0.002  2.27  1.38–3.72  0.021  1.86  1.10–3.09  SCS: diabetes mellitus (vs none)  0.042  1.82  1.06–3.11  0.13  1.54  0.87–2.64  SCS: respiratory disease (vs none)  0.15  1.47  0.90–2.40  Not included in multivariate analysis  Pack-year (vs < 20)  0.001  2.29  1.36–3.85  Not included in multivariate analysis  Alb (vs ≥ 3.8)  0.001  2.07  1.22–3.53  0.079  1.66  0.94–2.85  CRP (vs < 1.0)  0.11  1.82  0.86–3.83  Not included in multivariate analysis  %VC (vs > 90)  0.007  2.01  1.22–3.31  0.074  1.61  0.96–2.69  Lobectomy (vs limited resection)  0.59  1.21  0.69–2.10  Not included in multivariate analysis  Mediastinal lymph nodal dissection (vs none)  0.11  1.90  0.89–4.04  Not included in multivariate analysis  Video-assisted thoracic surgery (vs other approaches)  0.065  0.54  0.29–0.98  Not included in multivariate analysis  Variables  Univariate analysis   Multivariate analysis   P-value  Odds ratio  95% CI  P-value  Odds ratio  95% CI  Male (vs female)  <0.001  3.23  1.74–5.98  <0.001  2.73  1.50–5.31  Smoking (vs no smoking)  0.001  2.54  1.42–4.57  Not included in multivariate analysis  CGA75: memory (vs positive)  0.002  2.27  1.38–3.72  0.021  1.86  1.10–3.09  SCS: diabetes mellitus (vs none)  0.042  1.82  1.06–3.11  0.13  1.54  0.87–2.64  SCS: respiratory disease (vs none)  0.15  1.47  0.90–2.40  Not included in multivariate analysis  Pack-year (vs < 20)  0.001  2.29  1.36–3.85  Not included in multivariate analysis  Alb (vs ≥ 3.8)  0.001  2.07  1.22–3.53  0.079  1.66  0.94–2.85  CRP (vs < 1.0)  0.11  1.82  0.86–3.83  Not included in multivariate analysis  %VC (vs > 90)  0.007  2.01  1.22–3.31  0.074  1.61  0.96–2.69  Lobectomy (vs limited resection)  0.59  1.21  0.69–2.10  Not included in multivariate analysis  Mediastinal lymph nodal dissection (vs none)  0.11  1.90  0.89–4.04  Not included in multivariate analysis  Video-assisted thoracic surgery (vs other approaches)  0.065  0.54  0.29–0.98  Not included in multivariate analysis  Alb: albumin; CGA: comprehensive geriatric assessment; CI: confidence interval; CRP: C-reactive protein; SCS: Simplified Comorbidity Score; VC: vital capacity. Comprehensive operative risk scoring system The estimated odds ratios and their confidence intervals according to the 5 risk predicting factors for severe postoperative complications are shown in multivariate analysis (Table 3). The linear function based on the estimated regression coefficients was as follows: 0.50 (gender: male) + 0.31 (CGA75: memory: yes) + 0.25 (Alb: <3.8 ng/ml) + 0.24 (%VC: ≤90) + 0.22 (SCS: diabetes mellitus: yes). On the basis of the weight of variables in the function, we developed a simplified RS system as follows: RS = 3 (gender: male) + 2 (CGA75: memory: yes) + 2 (Alb: <3.8 ng/ml) + 1 (%VC: ≤90) + 1 (SCS: diabetes mellitus: yes). The receiver operating characteristic curve is shown in Fig. 2. The RS was moderately discriminatory with an AUC of 0.70. The RS of patients ranged between 0 and 9, and the relationship between RS and predicted probability is shown in Table 4. On the basis of the predicted probabilities for severe postoperative complications, patients were classified into 3 risk groups, i.e. low-risk group (RS: 0–2; predicted probability: <5%; n = 308), intermediate-risk group (RS: 3–7; predicted probability: 5–24%; n = 560) and high-risk group (RS: 8–9; predicted probability: ≥25%; n = 27). Table 4: Distribution of the patients and their predicted risk for severe postoperative complications Score  Number of patients  Risk observed % (predicted)  Frequency observed (predicted)  Patient’s risk  0  176  1.7 (2.4)  3 (4)  Low risk <5%  1  63  4.8 (3.3)  3 (2)  2  69  5.8 (4.6)  4 (3)    3  223  6.3 (6.3)  14 (14)  Intermediate risk 5–24%  4  109  6.4 (8.6)  7 (9)  5  122  13.9 (11.6)  17 (14)  6  74  17.6 (15.5)  13 (11)  7  32  12.5 (20.3)  4 (7)  8  22  22.7 (26.2)  5 (6)  High risk ≥25%  9  5  40.0 (33.2)  2 (2)  Score  Number of patients  Risk observed % (predicted)  Frequency observed (predicted)  Patient’s risk  0  176  1.7 (2.4)  3 (4)  Low risk <5%  1  63  4.8 (3.3)  3 (2)  2  69  5.8 (4.6)  4 (3)    3  223  6.3 (6.3)  14 (14)  Intermediate risk 5–24%  4  109  6.4 (8.6)  7 (9)  5  122  13.9 (11.6)  17 (14)  6  74  17.6 (15.5)  13 (11)  7  32  12.5 (20.3)  4 (7)  8  22  22.7 (26.2)  5 (6)  High risk ≥25%  9  5  40.0 (33.2)  2 (2)  Risk scoring system = 3 (gender: male) + 2 (CGA75: memory: yes) + 2 (Alb: < 3.8 ng/ml) + 1 (%VC: ≤ 90) + 1 (SCS: diabetes mellitus: yes). Figure 2: View largeDownload slide The receiver operator characteristics curve for predictive value of the risk score, with an area under the receiver operator curve of 0.70. Figure 2: View largeDownload slide The receiver operator characteristics curve for predictive value of the risk score, with an area under the receiver operator curve of 0.70. DISCUSSION Clinicians are increasingly confronted with octogenarians with medically operable NSCLC, who accounted for approximately 5000 patients and comprised 12% of patients who underwent surgical resection for primary lung cancer in Japan in 2014 [4]. The average life expectancy for an 80-year-old individual is increasing and has reached an impressive 8.89 years for men and 11.79 years for women according to the 2015 annual report from the Ministry of Health, Labour and Welfare in Japan [16]. Therefore, the latest, third edition of the American College of Chest Physicians evidence-based guidelines have recommended that patients with lung cancer should not be denied resection on the grounds of age alone [17]. Postoperative complications and mortality after lung resection in octogenarians range from 17.5% to 69.4% and from 0% to 21%, respectively, depending on the type of surgery and selection of patients, with long-term (5-year) survival rate ranging from 24% to 66% [18]. According to the Japanese large cohort previously described, operative morbidity and mortality in octogenarian patients were reported to be 8.4% and 1.4%, respectively, and were satisfactorily low and safe compared with 4.5% and 0.8%, respectively, for all patients [4]. This is also supported by our present series, in which complications of any severity were observed in 308 (34%) patients, of whom 81 (8.4%) had Grade 3–4 severe complications that required therapeutic intervention. Operative (30 days) and hospital mortality were 1.0% and 1.6%, respectively. Postoperative complications may influence long-term cancer survival. Rueth et al. [19] first reported that the occurrence of in-hospital postoperative complications was an independent predictor of worse 5-year cancer-specific survival (hazard ratio 1.38, 95% confidence interval 1.17–1.64) in elderly patients who underwent lobectomy for Stage I NSCLC using a population-based survival analysis. Recently, Lugg et al. [20] reported the long-term impact of developing a postoperative pulmonary complication after lung surgery and demonstrated that those who developed a postoperative pulmonary complication had a higher rate of non-cancer-related deaths (11% vs 5%; P = 0.006) and that postoperative complication was a significant independent risk factor for late death in NSCLC patients (hazard ratio 2.0, 95% confidence interval 1.9–3.2; P = 0.006) in their cohort. In our present series, pneumonia was the most common severe complication, observed in 27 (3.0%) patients, followed by prolonged air leakage (1.4%), atelectasis, bronchial fistula and empyema, indicating that postoperative pulmonary complications are significant in elderly surgical patients with lung cancer. Although the long-term effects of developing a severe postoperative complication in hospitals following lung cancer surgery have not been elucidated, in our present series, there were 18 patients with non-cancer-related deaths, including death due to cerebral infarction, aspiration pneumonia, unknown cardiopulmonary arrest and so on, observed even within the short median follow-up time of 6.5 months. We hypothesize that postoperative complications have a role in these non-cancer-related deaths observed later. Therefore, to optimize the safety and quality of surgical treatments for elderly NSCLC patients, we believe that accurate prediction of the risk of postoperative complications is important before deciding on surgical resection for octogenarian patients with medically operative lung cancer. There have been reports of several scoring systems involving weighting comorbid conditions that are useful for evaluating patients’ physiological function and predicting postoperative morbidity and mortality. The most widely used clinical scoring system is the CCI reported by Charlson et al. in 1987 [21], which is widely recommended as a survival predictor for patients with NSCLC [22]. Not all comorbidities have the same impact on survival, and careful selection of patients is more important than the patient’s score. Recently, the SCS was developed and validated by modification of the CCI [12] for patients with NSCLC. This tool scores 7 comorbidities based on the risk of death and have found to be useful in predicting outcomes for patients with lung cancer [6, 23]. Recently, inflammation-based scoring systems such as the GPS have been reported as simple, useful and objective prognostic predictors for a variety of cancers [7]. Moreover, growing support favours using the CGA, particularly for elderly patients. We selected clinical factors, CCI and SCS as comorbidities, GPS as general conditions and CGA7 as CGA in this study. Then, in the first analysis of this study, we developed a comprehensive operative RS system that can predict severe complications in octogenarian patients with medically operative lung cancer. Consequently, the 5 predictive factors, gender, CGA75: memory, SCS: diabetes mellitus, Alb and %VC, were identified for predicting severe complications through multivariate analysis. We further reconfirmed these 5 predictive factors using 5-fold cross-validation as an internal validation. Moreover, we developed a simplified RS system as follows: RS = 3 (gender: male) + 2 (CGA75: memory: yes) + 2 (Alb: <3.8 ng/ml) + 1 (%VC: ≤90) + 1 (SCS: diabetes mellitus: yes) on the basis of the weight of variables in the function and proposed an AUC as 0.70 using our prospective nationwide large series. To our best knowledge, this is the first comprehensive operative RS system that can predict severe complications in octogenarian patients with medically operative lung cancer. Limitations There are several limitations in this study. Comorbidity has been correlated with survival [24] in addition to postoperative complications [22]. In this first analysis, we had to focus on identifying predictive factors for severe operative complications because of the short median follow-up period for our series at the time of this first analysis. In 2019, we will perform the final analysis for this study including survival outcome comparisons with CCI, SCS, GPS, CGA7 and our proposed system. Additionally, we failed to demonstrate that a specific mode of operation such as limited resection, mediastinal lymph nodal sampling or video-associated thoracic surgery was significantly associated with decreasing the rate of surgical morbidity and mortality in our present series. During the last 2 decades, advances in perioperative care and in surgical techniques have been encouraged, and numerous studies have provided useful knowledge to thoracic surgeons. These our wise experience are likely to be selection bias from a less aggressive approach applied to the selected octogenarians. Hopefully, the final analysis of this study will provide some useful information regarding these issues. CONCLUSION This is the first large observational cohort study to date in this age demographics of octogenarian patients with medically operative lung cancer using a prospective nationwide well-organized setting. We propose a simplified RS system as follows: RS = 3 (gender: male) + 2 (CGA75: memory: yes) + 2 (Alb: <3.8 ng/ml) + 1 (%VC: ≤90) + 1 (SCS: diabetes mellitus: yes) (AUC = 0.70) as the first comprehensive operative RS system that can predict severe complications in octogenarian patients with medically operative lung cancer. ACKNOWLEDGEMENTS Medical English writing assistance was provided by Crimson Interactive Pvt. Ltd. Funding This study was supported by a Grant-in-Aid for Committee for Scientific Affairs, The Japanese Association for Thoracic Surgery and a Grant of Japanese Respiratory Foudation. Conflict of interest: none declared. REFERENCES 1 Torre LA, Bray F, Siegel RL, Ferlay J, Lortet-Tieulent J, Jemal A. Global cancer statistics, 2012. CA Cancer J Clin  2015; 65: 87– 108. Google Scholar CrossRef Search ADS PubMed  2 Committee for Scientific Affairs, The Japanese Association for Thoracic Surgery, Masuda M, Okumura M, Doki Y, Endo S, Hirata Y, Kobayashi J et al.   Thoracic and cardiovascular surgery in Japan during 2014: annual report by The Japanese Association for Thoracic Surgery. Gen Thorac Cardiovasc Surg  2016; 64: 665– 97. Google Scholar CrossRef Search ADS PubMed  3 Okami J, Higashiyama M, Asamura H, Goya T, Koshiishi Y, Sohara Y et al.   Pulmonary resection in patients aged 80 years or over with clinical stage I non-small cell lung cancer: prognostic factors for overall survival and risk factors for postoperative complications. J Thorac Oncol  2009; 4: 1247– 53. Google Scholar CrossRef Search ADS PubMed  4 Sawabata N, Miyaoka E, Asamura H, Nakanishi Y, Eguchi K, Mori M et al.   Japanese lung cancer registry study of 11,663 surgical cases in 2004: demographic and prognosis changes over decade. J Thorac Oncol  2011; 6: 1229– 35. Google Scholar CrossRef Search ADS PubMed  5 Saji H, Ueno T, Nakamura H. Committee for Scientific Affairs TJAfCSTJ: prospective observational cohort study of postoperative risk and prognosis scoring for elderly patients with medically operable lung cancer (JACS1303). Gen Thorac Cardiovasc Surg  2016; 64: 634– 5. Google Scholar CrossRef Search ADS PubMed  6 Girones R, Torregrosa D, Gomez-Codina J, Maestu I, Tenias JM, Rosell R. Prognostic impact of comorbidity in elderly lung cancer patients: use and comparison of two scores. Lung Cancer  2011; 72: 108– 13. Google Scholar CrossRef Search ADS PubMed  7 Miyazaki T, Yamasaki N, Tsuchiya T, Matsumoto K, Kunizaki M, Taniguchi D et al.   Inflammation-based scoring is a useful prognostic predictor of pulmonary resection for elderly patients with clinical stage I non-small-cell lung cancer. Eur J Cardiothorac Surg  2015; 47: e140– 5. Google Scholar CrossRef Search ADS PubMed  8 Haruki T, Yurugi Y, Wakahara M, Matsuoka Y, Miwa K, Araki K et al.   Simplified comorbidity score for elderly patients undergoing thoracoscopic surgery for lung cancer. Surg Today  2017; 47: 718– 25. Google Scholar CrossRef Search ADS PubMed  9 Travis WD, Brambilla E, Burke AP, Marx A, Nicholson AG. WHO Classification of Tumours of the Lung, Pleura, Thymus and Heart , 4th edn. Lyon, France: International Agency for Research on Cancer, 2015. 10 TNM Classification of Malignant Tumours , 7th edn. Oxford, UK: Wiley-Blackwell, 2009. 11 Charlson ME, Sax FL, MacKenzie CR, Fields SD, Braham RL, Douglas RGJr. Resuscitation: how do we decide? A prospective study of physicians' preferences and the clinical course of hospitalized patients. JAMA  1986; 255: 1316– 22. Google Scholar CrossRef Search ADS PubMed  12 Colinet B, Jacot W, Bertrand D, Lacombe S, Bozonnat MC, Daures JP et al.  ; oncoLR health network. A new simplified comorbidity score as a prognostic factor in non-small-cell lung cancer patients: description and comparison with the Charlson’s index. Br J Cancer  2005; 93: 1098– 105. Google Scholar CrossRef Search ADS PubMed  13 Kawashima M, Murakawa T, Shinozaki T, Ichinose J, Hino H, Konoeda C et al.   Significance of the Glasgow Prognostic Score as a prognostic indicator for lung cancer surgery. Interact CardioVasc Thorac Surg  2015; 21: 637– 43. Google Scholar CrossRef Search ADS PubMed  14 Fukuse T, Satoda N, Hijiya K, Fujinaga T. Importance of a comprehensive geriatric assessment in prediction of complications following thoracic surgery in elderly patients. Chest  2005; 127: 886– 91. Google Scholar CrossRef Search ADS PubMed  15 Hurvich CM, Tsai CL. Regression and time series model selection in small samples. Biometrika  1989; 76: 297– 307. Google Scholar CrossRef Search ADS   16 The 2015 annual report from the Ministry of Health, Labour and Welfare in Japan, http://www.mhlw.go.jp/toukei/saikin/hw/life/life15/. 17 Brunelli A, Kim AW, Berger KI, Addrizzo-Harris DJ. 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. Google Scholar CrossRef Search ADS PubMed  18 Guerra M, Neves P, Miranda J. Surgical treatment of non-small-cell lung cancer in octogenarians. Interact CardioVasc Thorac Surg  2013; 16: 673– 80. Google Scholar CrossRef Search ADS PubMed  19 Rueth NM, Parsons HM, Habermann EB, Groth SS, Virnig BA, Tuttle TM et al.   The long-term impact of surgical complications after resection of stage I nonsmall cell lung cancer: a population-based survival analysis. Ann Surg  2011; 254: 368– 74. Google Scholar CrossRef Search ADS PubMed  20 Lugg ST, Agostini PJ, Tikka T, Kerr A, Adams K, Bishay E et al.   Long-term impact of developing a postoperative pulmonary complication after lung surgery. Thorax  2016; 71: 171– 6. Google Scholar CrossRef Search ADS PubMed  21 Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis  1987; 40: 373– 83. Google Scholar CrossRef Search ADS PubMed  22 Birim O, Kappetein AP, Bogers AJ. Charlson comorbidity index as a predictor of long-term outcome after surgery for nonsmall cell lung cancer. Eur J Cardiothorac Surg  2005; 28: 759– 62. Google Scholar CrossRef Search ADS PubMed  23 Jacot W, Colinet B, Bertrand D, Lacombe S, Bozonnat MC, Daures JP et al.   Quality of life and comorbidity score as prognostic determinants in non-small-cell lung cancer patients. Ann Oncol  2008; 19: 1458– 64. Google Scholar CrossRef Search ADS PubMed  24 Battafarano RJ, Piccirillo JF, Meyers BF, Hsu HS, Guthrie TJ, Cooper JD et al.   Impact of comorbidity on survival after surgical resection in patients with stage I non-small cell lung cancer. J Thorac Cardiovasc Surg  2002; 123: 280– 7. Google Scholar CrossRef Search ADS PubMed  © The Author 2017. Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. All rights reserved. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png European Journal of Cardio-Thoracic Surgery Oxford University Press

Loading next page...
 
/lp/ou_press/a-proposal-for-a-comprehensive-risk-scoring-system-for-predicting-YWFhUplzoB
Publisher
Oxford University Press
ISSN
1010-7940
eISSN
1873-734X
D.O.I.
10.1093/ejcts/ezx415
Publisher site
See Article on Publisher Site

Abstract

Abstract OBJECTIVES Although some retrospective studies have reported clinicopathological scoring systems for predicting postoperative complications and survival outcomes for elderly lung cancer patients, optimized scoring systems remain controversial. METHODS The Japanese Association for Chest Surgery (JACS) conducted a nationwide multicentre prospective cohort and enrolled a total of 1019 octogenarians with medically operable lung cancer. Details of the clinical factors, comorbidities and comprehensive geriatric assessment were recorded for 895 patients to develop a comprehensive risk scoring (RS) system capable of predicting severe complications. RESULTS Operative (30 days) and hospital mortality rates were 1.0% and 1.6%, respectively. Complications were observed in 308 (34%) patients, of whom 81 (8.4%) had Grade 3–4 severe complications. Pneumonia was the most common severe complication, observed in 27 (3.0%) patients. Five predictive factors, gender, comprehensive geriatric assessment75: memory and Simplified Comorbidity Score (SCS): diabetes mellitus, albumin and percentage vital capacity, were identified as independent predictive factors for severe postoperative complications (odds ratio = 2.73, 1.86, 1.54, 1.66 and 1.61, respectively) through univariate and multivariate analyses. A 5-fold cross-validation was performed as an internal validation to reconfirm these 5 predictive factors (average area under the curve 0.70). We developed a simplified RS system as follows: RS = 3 (gender: male) + 2 (comprehensive geriatric assessment 75: memory: yes) + 2 (albumin: <3.8 ng/ml) + 1 (percentage vital capacity: ≤90) + 1 (SCS: diabetes mellitus: yes). CONCLUSIONS The current series shows that octogenarians can be successfully treated for lung cancer with surgical resection with an acceptable rate of severe complications and mortality. We propose a simplified RS system to predict severe complications in octogenarian patients with medically operative lung cancer. Trial Registration Number JACS1303 (UMIN000016756) Lung cancer, Octogenarian, Postoperative complication, Risk scoring system INTRODUCTION Lung cancer is the leading cause of cancer-related deaths worldwide and is a particularly prominent disease in the elderly population [1]. Its incidence in this population is expected to increase as the general population ages, especially in developed countries. Clinicians are increasingly confronted with octogenarians with medically operable non-small-cell lung cancer (NSCLC). The total number of surgical resections for primary lung cancer in Japan reported most recently in 2014 has reached 38 085, showing a steady increase over the years [2]. Among those who underwent surgical resections for this condition, 53.5% of patients were aged 70 years or older and 12.1% of patients were aged 80 years or older. This percentage has continued to increase because the ageing of Japan’s population is becoming more pronounced. The Japanese Joint Committee of Lung Cancer Registry previously conducted and reported a large retrospective cohort study to identify the predictors of long-term survival and the risk factors for complications after surgical resection in octogenarian patients with clinical Stage I NSCLC [3]. Although the operative morbidity and mortality for patients aged 80 years and older were reported to be 8.4% and 1.4%, respectively, and were satisfactorily low and safe compared with 4.5% and 0.8%, respectively, for all patients [4], one-third of the patients died from other causes, such as an illness during the postoperative follow-up period. Comorbidity is a factor associated with both postoperative complications and worse prognosis. However, these results were still limited and based on retrospective cohort studies. Therefore, the Japanese Association for Chest Surgery (JACS) conducted a nationwide multicentre large-scale prospective observational cohort study, JACS1303 (UMIN000016756) as previously described by Saji et al. [5]. Although some retrospective studies have reported clinicopathological scoring systems, such as the Charlson Comorbidity Index (CCI), Simplified Comorbidity Score (SCS) and Glasgow Prognostic Score (GPS) for predicting postoperative complications and survival outcomes for elderly lung cancer patients [6–8], optimized scoring systems and care strategies remain controversial. The first analysis of the JACS1303 study aims to evaluate details of the clinical factors, comorbidities and comprehensive geriatric assessment (CGA) to develop a comprehensive operative risk scoring (RS) system that can predict severe morbidity and mortality for octogenarian patients with medically operative lung cancer. MATERIALS AND METHODS Patient eligibility Octogenarian patients with medically operable lung cancer were enrolled in this cohort. Additional inclusion criteria were as follows: no prior chemotherapy or radiation therapy for lung cancer, sufficient organ function, measurable preoperative comprehensive geriatric assessment and written informed consent. Between April 2015 and December 2016, 1019 octogenarian patients were enrolled from 82 institutions (Fig. 1). Of the 962 patients enrolled, we finally selected 895 patients who underwent surgical resection for primary lung cancer and had sufficient clinical records for comprehensive surgical RS analysis of predictive postoperative complications for octogenarian patients with medically operable lung cancer. Figure 1: View largeDownload slide The CONSORT diagram. CCI: Charlson Comorbidity Index; CGA: comprehensive geriatric assessment; GPS: Glasgow Prognostic Score; SCS: Simplified Comorbidity Score. Figure 1: View largeDownload slide The CONSORT diagram. CCI: Charlson Comorbidity Index; CGA: comprehensive geriatric assessment; GPS: Glasgow Prognostic Score; SCS: Simplified Comorbidity Score. Data collection Data collection and analyses were approved by the institutional review boards. Once informed consent was obtained, recruiting clinicians completed a case report form to record the following clinicopathological factors: age, gender, smoking status, tumour size and location, histological subtypes according to the World Health Organization (WHO) classification [9] and staging according to the seventh edition of the tumour, necrosis and metastasis (TNM) classification of lung and pleural tumours [10]. The status of the patients was recorded as follows: performance status, pulmonary function including vital capacity (VC), %VC, forced expiratory volume in 1 s, and percentage forced expiratory volume in 1 s. The comorbidity index was calculated using the CCI [11] and SCS [12]. The GPS [13] and CGA 7 [14] were also recorded. Surgical-related data were also collected after the completion of surgery. Clinical outcome data related to operative morbidity and mortality, and cancer-related or non-related prognosis were collected 30 days and 90 days after the surgical treatment. Perioperative and postoperative morbidity and adverse events were classified according to both the Common Terminology Criteria for Adverse Events version 4.0. Operative mortality was defined as death at any time during the initial hospitalization or within 30 days after the operation. Statistical analysis The aim of this study was to develop a preoperative risk prediction model for octogenarian patients with medically operative lung cancer as previously described by Saji et al. [5]. This would require evaluation of the risk of Grade 3 or higher adverse events after surgical treatment using logistic regression models with a maximum of 5 predictive risk factors. For this purpose, we needed to observe at least 50 adverse events. The planned sample size of 750 elderly patients aged 80 years and older was determined and based on the assumption that the proportion of adverse events in this cohort would be in the range of 7–8%. Analyses were performed using JMP® Pro version 13.0 (SAS Institute Inc., Cary, NC, USA). RESULTS Postoperative complications and mortalities The clinicopathological and surgical characteristics of the patients are listed in Table 1. The variables of postoperative early complications (30 days) and operative and hospital mortality are listed in Table 2. Complications of any severity were observed in 308 (34%) patients, of whom 81 (8.4%) had Grade 3–4 severe complications, which required therapeutic intervention. Pneumonia was the most common severe complication observed in 27 (3.0%) patients, followed by prolonged air leakage (1.4%), atelectasis, bronchial fistula, empyema, wound infection, cerebral infarction, subcutaneous emphysema, atrial fibrillation and gastrointestinal disorders. The major common complications, any grade, were prolonged air leakage (8.4%), atrial fibrillation (5.0%), pneumonia (4.1%), subcutaneous emphysema (2.8%), atelectasis (2.1%) and wound infection (1%). Operative (30 days) and hospital mortality rates were 1.0% and 1.6%, respectively. Three patients died due to postoperative pneumonia, including 2 cases of acute exacerbation of interstitial pneumonia and 1 case of aspiration pneumonia. One patient each died due to cerebral infarction, perioperative major haemorrhage, myocardial infarction, liver dysfunction due to liver cirrhosis, myelodysplastic syndromes and cardiopulmonary arrest due to pneumothorax on the 1st day after hospital discharge. Table 1: Characteristics of eligible octogenarian patients (n = 895) Variables    Age, mean (range)  82.7 (80–96)   80–84/85–89/ND, n (%)  716 (80)/171 (19)/8 (1)  Gender     Male/female, n (%)  540 (60)/355 (40)  Performance status     0/1/2/3 or 4/ND, n (%)  661 (74)/207 (23)/17 (2)/4/6  Smoking status     Current smoker/ex-smoker/never smoker, n (%)  47 (5)/503 (56)/345 (38)  Pack-year smoking (smokers, n = 550), mean ± SD  48.4 ± 30.6   ≥20, n (%)  462 (84)  Pulmonary function, mean ± SD     VC (l)/%VC  2.7 ± 0.7/101.1 ± 17.7   FEV1.0 (l)/FEV1.0 (%)  1.8 ± 0.5/73.0 ± 14.7  Comorbidity (yes/no), n (%)  687 (76)/208 (24)   Respiratory disease  340 (38)   History of malignant disease  267 (30)   Diabetes mellitus  173 (19)   Cerebroneural disease  111 (12)   Renal insufficiancy  80 (9)   Cardiovascular disease  42 (5)  Charlson Comorbidity Index, mean ± SD  1.7 ± 1.6  Simplified Comorbidity Score, mean ± SD  8.5 ± 4.9  Glasgow Prognostic Score (0/1/2), n (%)  791 (88)/84 (9)/20 (3)  Comprehensive geriatric assessment 7, mean ± SD  6.4 ± 1.0  Tumour site, n (%)     RUL/RML/RLL  254 (28)/52 (6)/211 (24)   LUL/LLL  232 (26)/146 (16)  Tumour size (cm), mean ± SD  3.0 ± 1.5  Histology, n (%)     Adenocarcinoma/squamous cell carcinoma  607 (68)/194 (22)   Others/ND  75 (8)/19 (2)  Tumour factor, n (%)     pT1a/pT1b/pT2a/pT2b  238 (27)/195 (22)/329 (37)/45 (5)   pT3/pT4/ND  63 (7)/11 (1)/14 (1)  Nodal factor, n (%)     pN0/pN1/pN2/pN3/ND  752 (84)/68 (8)/59 (7)/1/15 (1)  Pathological staging, n (%)     IA/IB/IIA/IIB  400 (45)/267 (30)/80 (9)/49 (5)   IIIA/IIIB/IV/ND  70 (8)/1/13 (1)/15 (1)  Surgical procedures, n (%)     Lobectomy/pneumonectomy  623 (70)/3   Segmental resection/ wedge resection/ND  110 (12)/151 (17)/8 (1)  Mediastinal lymph nodal dissection, n (%)   Systematic/selective  113 (13)/346 (39)   Sampling or none/ND  430 (48)/6  Thoracic approach, n (%)     Open- or minithoracotomy  482 (54)   Video-assisted thoracotomy/ND  400 (45)/13 (1)  Intraoperative factors     Operation time (min), median (range)  175 (33–1233)   Blood loss (ml), median (range)  30 (0–57600)   Blood transfusion, n (%)  26 (3)  Chest tube placement (days), median (range)  3 (0–23)  Hospital days after operation, median (range)  9 (1–91)  Curative resection     R0/R1-2/ND, n (%)  828 (93)/41 (5)/26 (3)  Variables    Age, mean (range)  82.7 (80–96)   80–84/85–89/ND, n (%)  716 (80)/171 (19)/8 (1)  Gender     Male/female, n (%)  540 (60)/355 (40)  Performance status     0/1/2/3 or 4/ND, n (%)  661 (74)/207 (23)/17 (2)/4/6  Smoking status     Current smoker/ex-smoker/never smoker, n (%)  47 (5)/503 (56)/345 (38)  Pack-year smoking (smokers, n = 550), mean ± SD  48.4 ± 30.6   ≥20, n (%)  462 (84)  Pulmonary function, mean ± SD     VC (l)/%VC  2.7 ± 0.7/101.1 ± 17.7   FEV1.0 (l)/FEV1.0 (%)  1.8 ± 0.5/73.0 ± 14.7  Comorbidity (yes/no), n (%)  687 (76)/208 (24)   Respiratory disease  340 (38)   History of malignant disease  267 (30)   Diabetes mellitus  173 (19)   Cerebroneural disease  111 (12)   Renal insufficiancy  80 (9)   Cardiovascular disease  42 (5)  Charlson Comorbidity Index, mean ± SD  1.7 ± 1.6  Simplified Comorbidity Score, mean ± SD  8.5 ± 4.9  Glasgow Prognostic Score (0/1/2), n (%)  791 (88)/84 (9)/20 (3)  Comprehensive geriatric assessment 7, mean ± SD  6.4 ± 1.0  Tumour site, n (%)     RUL/RML/RLL  254 (28)/52 (6)/211 (24)   LUL/LLL  232 (26)/146 (16)  Tumour size (cm), mean ± SD  3.0 ± 1.5  Histology, n (%)     Adenocarcinoma/squamous cell carcinoma  607 (68)/194 (22)   Others/ND  75 (8)/19 (2)  Tumour factor, n (%)     pT1a/pT1b/pT2a/pT2b  238 (27)/195 (22)/329 (37)/45 (5)   pT3/pT4/ND  63 (7)/11 (1)/14 (1)  Nodal factor, n (%)     pN0/pN1/pN2/pN3/ND  752 (84)/68 (8)/59 (7)/1/15 (1)  Pathological staging, n (%)     IA/IB/IIA/IIB  400 (45)/267 (30)/80 (9)/49 (5)   IIIA/IIIB/IV/ND  70 (8)/1/13 (1)/15 (1)  Surgical procedures, n (%)     Lobectomy/pneumonectomy  623 (70)/3   Segmental resection/ wedge resection/ND  110 (12)/151 (17)/8 (1)  Mediastinal lymph nodal dissection, n (%)   Systematic/selective  113 (13)/346 (39)   Sampling or none/ND  430 (48)/6  Thoracic approach, n (%)     Open- or minithoracotomy  482 (54)   Video-assisted thoracotomy/ND  400 (45)/13 (1)  Intraoperative factors     Operation time (min), median (range)  175 (33–1233)   Blood loss (ml), median (range)  30 (0–57600)   Blood transfusion, n (%)  26 (3)  Chest tube placement (days), median (range)  3 (0–23)  Hospital days after operation, median (range)  9 (1–91)  Curative resection     R0/R1-2/ND, n (%)  828 (93)/41 (5)/26 (3)  FEV1.0: forced expiratory volume in 1 s; FEV1.0%: percentage of forced expiratory volume in 1 s; LLL: left lower lobe; LUL: left upper lobe; ND: no data; RLL: right lower lobe; RML: right middle lobe; RUL: right upper lobe; SD: standard deviation; VC: vital capacity. Table 2: Postoperative early complications of octogenarian patients with operable lung cancer (n = 895) Postoperative complications 30 days  Grade 1  Grade 2  Grade 3  Grade 4  Grade 3–4  Grade 5  Pneumonia  0  9  22  5  27 (3.0)  3  Prolonged air leakage  14  49  11  2  13 (1.4)  0  Atelectasis  3  11  4  1  5  0  Bronchial fistula  0  0  3  2  5  0  Empyema  0  3  4  1  5  0  Wound infection  0  5  4  0  4  0  Cerebral infarction  0  4  1  2  3  1  Subcutaneous emphysema  18  6  1  1  2  0  Atrial fibrillation  6  35  2  0  2  0  Pulmonary embolism  1  1  2  0  2  0  Gastrointestinal disorders  1  1  2  0  2  0  Paroxysmal supraventricular tachycardia  5  5  1  0  1  0  Postoperative haemorrhage  3  2  1  0  1  1  Ventricular tachycardia  2  3  0  0  0  0  Recurrent nerve palsy  1  5  0  0  0  0  Myocardial infarction  0  0  0  0  1  1  Ventricular fibrillation  0  4  0  0  0  0  Chylothorax  0  3  0  0  0  0  Others  2  23  4  5  9  3  Any complications (n = 306, 34%)  56  169  62  19  81 (8.4)    Operative mortality (30 days)            9 (1.0)  Hospital mortality            14 (1.6)  Postoperative complications 30 days  Grade 1  Grade 2  Grade 3  Grade 4  Grade 3–4  Grade 5  Pneumonia  0  9  22  5  27 (3.0)  3  Prolonged air leakage  14  49  11  2  13 (1.4)  0  Atelectasis  3  11  4  1  5  0  Bronchial fistula  0  0  3  2  5  0  Empyema  0  3  4  1  5  0  Wound infection  0  5  4  0  4  0  Cerebral infarction  0  4  1  2  3  1  Subcutaneous emphysema  18  6  1  1  2  0  Atrial fibrillation  6  35  2  0  2  0  Pulmonary embolism  1  1  2  0  2  0  Gastrointestinal disorders  1  1  2  0  2  0  Paroxysmal supraventricular tachycardia  5  5  1  0  1  0  Postoperative haemorrhage  3  2  1  0  1  1  Ventricular tachycardia  2  3  0  0  0  0  Recurrent nerve palsy  1  5  0  0  0  0  Myocardial infarction  0  0  0  0  1  1  Ventricular fibrillation  0  4  0  0  0  0  Chylothorax  0  3  0  0  0  0  Others  2  23  4  5  9  3  Any complications (n = 306, 34%)  56  169  62  19  81 (8.4)    Operative mortality (30 days)            9 (1.0)  Hospital mortality            14 (1.6)  Predictive factors for severe postoperative complications Univariate analysis was carried out to identify 17 possible predictive factors (P < 0.25) for severe postoperative complications. We removed 8 factors including total score of SCS, GPS, CGA7, tumour side, tumour location, maximum standardized uptake value, CGA76 and CGA77 from the clinical view and consequently identified gender, smoking history, CGA75: memory, SCS: diabetes mellitus, SCS: respiratory disease, pack-year, albumin (Alb), C-reactive protein and %VC as candidate predictive factors. To develop a logistic regression model with a maximum of 5 risk factors, we applied forward selection with the corrected Akaike information criteria [15] and identified the following 5 predictive factors: gender, CGA75: memory, SCS: diabetes mellitus, Alb and %VC (Table 3). Additionally, 5-fold cross-validation was performed as an internal validation to reconfirm these 5 predictive factors. The discriminative ability of the model was evaluated using the area under the receiver operator curve (AUC, concordance statistic). The AUCs for 5-fold cross-validation were 0.64, 0.75, 0.71, 0.67 and 0.64, and the average AUC was 0.68. Table 3: Univariate and multivariate analysis of predicting for severe operative complications Variables  Univariate analysis   Multivariate analysis   P-value  Odds ratio  95% CI  P-value  Odds ratio  95% CI  Male (vs female)  <0.001  3.23  1.74–5.98  <0.001  2.73  1.50–5.31  Smoking (vs no smoking)  0.001  2.54  1.42–4.57  Not included in multivariate analysis  CGA75: memory (vs positive)  0.002  2.27  1.38–3.72  0.021  1.86  1.10–3.09  SCS: diabetes mellitus (vs none)  0.042  1.82  1.06–3.11  0.13  1.54  0.87–2.64  SCS: respiratory disease (vs none)  0.15  1.47  0.90–2.40  Not included in multivariate analysis  Pack-year (vs < 20)  0.001  2.29  1.36–3.85  Not included in multivariate analysis  Alb (vs ≥ 3.8)  0.001  2.07  1.22–3.53  0.079  1.66  0.94–2.85  CRP (vs < 1.0)  0.11  1.82  0.86–3.83  Not included in multivariate analysis  %VC (vs > 90)  0.007  2.01  1.22–3.31  0.074  1.61  0.96–2.69  Lobectomy (vs limited resection)  0.59  1.21  0.69–2.10  Not included in multivariate analysis  Mediastinal lymph nodal dissection (vs none)  0.11  1.90  0.89–4.04  Not included in multivariate analysis  Video-assisted thoracic surgery (vs other approaches)  0.065  0.54  0.29–0.98  Not included in multivariate analysis  Variables  Univariate analysis   Multivariate analysis   P-value  Odds ratio  95% CI  P-value  Odds ratio  95% CI  Male (vs female)  <0.001  3.23  1.74–5.98  <0.001  2.73  1.50–5.31  Smoking (vs no smoking)  0.001  2.54  1.42–4.57  Not included in multivariate analysis  CGA75: memory (vs positive)  0.002  2.27  1.38–3.72  0.021  1.86  1.10–3.09  SCS: diabetes mellitus (vs none)  0.042  1.82  1.06–3.11  0.13  1.54  0.87–2.64  SCS: respiratory disease (vs none)  0.15  1.47  0.90–2.40  Not included in multivariate analysis  Pack-year (vs < 20)  0.001  2.29  1.36–3.85  Not included in multivariate analysis  Alb (vs ≥ 3.8)  0.001  2.07  1.22–3.53  0.079  1.66  0.94–2.85  CRP (vs < 1.0)  0.11  1.82  0.86–3.83  Not included in multivariate analysis  %VC (vs > 90)  0.007  2.01  1.22–3.31  0.074  1.61  0.96–2.69  Lobectomy (vs limited resection)  0.59  1.21  0.69–2.10  Not included in multivariate analysis  Mediastinal lymph nodal dissection (vs none)  0.11  1.90  0.89–4.04  Not included in multivariate analysis  Video-assisted thoracic surgery (vs other approaches)  0.065  0.54  0.29–0.98  Not included in multivariate analysis  Alb: albumin; CGA: comprehensive geriatric assessment; CI: confidence interval; CRP: C-reactive protein; SCS: Simplified Comorbidity Score; VC: vital capacity. Comprehensive operative risk scoring system The estimated odds ratios and their confidence intervals according to the 5 risk predicting factors for severe postoperative complications are shown in multivariate analysis (Table 3). The linear function based on the estimated regression coefficients was as follows: 0.50 (gender: male) + 0.31 (CGA75: memory: yes) + 0.25 (Alb: <3.8 ng/ml) + 0.24 (%VC: ≤90) + 0.22 (SCS: diabetes mellitus: yes). On the basis of the weight of variables in the function, we developed a simplified RS system as follows: RS = 3 (gender: male) + 2 (CGA75: memory: yes) + 2 (Alb: <3.8 ng/ml) + 1 (%VC: ≤90) + 1 (SCS: diabetes mellitus: yes). The receiver operating characteristic curve is shown in Fig. 2. The RS was moderately discriminatory with an AUC of 0.70. The RS of patients ranged between 0 and 9, and the relationship between RS and predicted probability is shown in Table 4. On the basis of the predicted probabilities for severe postoperative complications, patients were classified into 3 risk groups, i.e. low-risk group (RS: 0–2; predicted probability: <5%; n = 308), intermediate-risk group (RS: 3–7; predicted probability: 5–24%; n = 560) and high-risk group (RS: 8–9; predicted probability: ≥25%; n = 27). Table 4: Distribution of the patients and their predicted risk for severe postoperative complications Score  Number of patients  Risk observed % (predicted)  Frequency observed (predicted)  Patient’s risk  0  176  1.7 (2.4)  3 (4)  Low risk <5%  1  63  4.8 (3.3)  3 (2)  2  69  5.8 (4.6)  4 (3)    3  223  6.3 (6.3)  14 (14)  Intermediate risk 5–24%  4  109  6.4 (8.6)  7 (9)  5  122  13.9 (11.6)  17 (14)  6  74  17.6 (15.5)  13 (11)  7  32  12.5 (20.3)  4 (7)  8  22  22.7 (26.2)  5 (6)  High risk ≥25%  9  5  40.0 (33.2)  2 (2)  Score  Number of patients  Risk observed % (predicted)  Frequency observed (predicted)  Patient’s risk  0  176  1.7 (2.4)  3 (4)  Low risk <5%  1  63  4.8 (3.3)  3 (2)  2  69  5.8 (4.6)  4 (3)    3  223  6.3 (6.3)  14 (14)  Intermediate risk 5–24%  4  109  6.4 (8.6)  7 (9)  5  122  13.9 (11.6)  17 (14)  6  74  17.6 (15.5)  13 (11)  7  32  12.5 (20.3)  4 (7)  8  22  22.7 (26.2)  5 (6)  High risk ≥25%  9  5  40.0 (33.2)  2 (2)  Risk scoring system = 3 (gender: male) + 2 (CGA75: memory: yes) + 2 (Alb: < 3.8 ng/ml) + 1 (%VC: ≤ 90) + 1 (SCS: diabetes mellitus: yes). Figure 2: View largeDownload slide The receiver operator characteristics curve for predictive value of the risk score, with an area under the receiver operator curve of 0.70. Figure 2: View largeDownload slide The receiver operator characteristics curve for predictive value of the risk score, with an area under the receiver operator curve of 0.70. DISCUSSION Clinicians are increasingly confronted with octogenarians with medically operable NSCLC, who accounted for approximately 5000 patients and comprised 12% of patients who underwent surgical resection for primary lung cancer in Japan in 2014 [4]. The average life expectancy for an 80-year-old individual is increasing and has reached an impressive 8.89 years for men and 11.79 years for women according to the 2015 annual report from the Ministry of Health, Labour and Welfare in Japan [16]. Therefore, the latest, third edition of the American College of Chest Physicians evidence-based guidelines have recommended that patients with lung cancer should not be denied resection on the grounds of age alone [17]. Postoperative complications and mortality after lung resection in octogenarians range from 17.5% to 69.4% and from 0% to 21%, respectively, depending on the type of surgery and selection of patients, with long-term (5-year) survival rate ranging from 24% to 66% [18]. According to the Japanese large cohort previously described, operative morbidity and mortality in octogenarian patients were reported to be 8.4% and 1.4%, respectively, and were satisfactorily low and safe compared with 4.5% and 0.8%, respectively, for all patients [4]. This is also supported by our present series, in which complications of any severity were observed in 308 (34%) patients, of whom 81 (8.4%) had Grade 3–4 severe complications that required therapeutic intervention. Operative (30 days) and hospital mortality were 1.0% and 1.6%, respectively. Postoperative complications may influence long-term cancer survival. Rueth et al. [19] first reported that the occurrence of in-hospital postoperative complications was an independent predictor of worse 5-year cancer-specific survival (hazard ratio 1.38, 95% confidence interval 1.17–1.64) in elderly patients who underwent lobectomy for Stage I NSCLC using a population-based survival analysis. Recently, Lugg et al. [20] reported the long-term impact of developing a postoperative pulmonary complication after lung surgery and demonstrated that those who developed a postoperative pulmonary complication had a higher rate of non-cancer-related deaths (11% vs 5%; P = 0.006) and that postoperative complication was a significant independent risk factor for late death in NSCLC patients (hazard ratio 2.0, 95% confidence interval 1.9–3.2; P = 0.006) in their cohort. In our present series, pneumonia was the most common severe complication, observed in 27 (3.0%) patients, followed by prolonged air leakage (1.4%), atelectasis, bronchial fistula and empyema, indicating that postoperative pulmonary complications are significant in elderly surgical patients with lung cancer. Although the long-term effects of developing a severe postoperative complication in hospitals following lung cancer surgery have not been elucidated, in our present series, there were 18 patients with non-cancer-related deaths, including death due to cerebral infarction, aspiration pneumonia, unknown cardiopulmonary arrest and so on, observed even within the short median follow-up time of 6.5 months. We hypothesize that postoperative complications have a role in these non-cancer-related deaths observed later. Therefore, to optimize the safety and quality of surgical treatments for elderly NSCLC patients, we believe that accurate prediction of the risk of postoperative complications is important before deciding on surgical resection for octogenarian patients with medically operative lung cancer. There have been reports of several scoring systems involving weighting comorbid conditions that are useful for evaluating patients’ physiological function and predicting postoperative morbidity and mortality. The most widely used clinical scoring system is the CCI reported by Charlson et al. in 1987 [21], which is widely recommended as a survival predictor for patients with NSCLC [22]. Not all comorbidities have the same impact on survival, and careful selection of patients is more important than the patient’s score. Recently, the SCS was developed and validated by modification of the CCI [12] for patients with NSCLC. This tool scores 7 comorbidities based on the risk of death and have found to be useful in predicting outcomes for patients with lung cancer [6, 23]. Recently, inflammation-based scoring systems such as the GPS have been reported as simple, useful and objective prognostic predictors for a variety of cancers [7]. Moreover, growing support favours using the CGA, particularly for elderly patients. We selected clinical factors, CCI and SCS as comorbidities, GPS as general conditions and CGA7 as CGA in this study. Then, in the first analysis of this study, we developed a comprehensive operative RS system that can predict severe complications in octogenarian patients with medically operative lung cancer. Consequently, the 5 predictive factors, gender, CGA75: memory, SCS: diabetes mellitus, Alb and %VC, were identified for predicting severe complications through multivariate analysis. We further reconfirmed these 5 predictive factors using 5-fold cross-validation as an internal validation. Moreover, we developed a simplified RS system as follows: RS = 3 (gender: male) + 2 (CGA75: memory: yes) + 2 (Alb: <3.8 ng/ml) + 1 (%VC: ≤90) + 1 (SCS: diabetes mellitus: yes) on the basis of the weight of variables in the function and proposed an AUC as 0.70 using our prospective nationwide large series. To our best knowledge, this is the first comprehensive operative RS system that can predict severe complications in octogenarian patients with medically operative lung cancer. Limitations There are several limitations in this study. Comorbidity has been correlated with survival [24] in addition to postoperative complications [22]. In this first analysis, we had to focus on identifying predictive factors for severe operative complications because of the short median follow-up period for our series at the time of this first analysis. In 2019, we will perform the final analysis for this study including survival outcome comparisons with CCI, SCS, GPS, CGA7 and our proposed system. Additionally, we failed to demonstrate that a specific mode of operation such as limited resection, mediastinal lymph nodal sampling or video-associated thoracic surgery was significantly associated with decreasing the rate of surgical morbidity and mortality in our present series. During the last 2 decades, advances in perioperative care and in surgical techniques have been encouraged, and numerous studies have provided useful knowledge to thoracic surgeons. These our wise experience are likely to be selection bias from a less aggressive approach applied to the selected octogenarians. Hopefully, the final analysis of this study will provide some useful information regarding these issues. CONCLUSION This is the first large observational cohort study to date in this age demographics of octogenarian patients with medically operative lung cancer using a prospective nationwide well-organized setting. We propose a simplified RS system as follows: RS = 3 (gender: male) + 2 (CGA75: memory: yes) + 2 (Alb: <3.8 ng/ml) + 1 (%VC: ≤90) + 1 (SCS: diabetes mellitus: yes) (AUC = 0.70) as the first comprehensive operative RS system that can predict severe complications in octogenarian patients with medically operative lung cancer. ACKNOWLEDGEMENTS Medical English writing assistance was provided by Crimson Interactive Pvt. Ltd. Funding This study was supported by a Grant-in-Aid for Committee for Scientific Affairs, The Japanese Association for Thoracic Surgery and a Grant of Japanese Respiratory Foudation. Conflict of interest: none declared. REFERENCES 1 Torre LA, Bray F, Siegel RL, Ferlay J, Lortet-Tieulent J, Jemal A. Global cancer statistics, 2012. CA Cancer J Clin  2015; 65: 87– 108. Google Scholar CrossRef Search ADS PubMed  2 Committee for Scientific Affairs, The Japanese Association for Thoracic Surgery, Masuda M, Okumura M, Doki Y, Endo S, Hirata Y, Kobayashi J et al.   Thoracic and cardiovascular surgery in Japan during 2014: annual report by The Japanese Association for Thoracic Surgery. Gen Thorac Cardiovasc Surg  2016; 64: 665– 97. Google Scholar CrossRef Search ADS PubMed  3 Okami J, Higashiyama M, Asamura H, Goya T, Koshiishi Y, Sohara Y et al.   Pulmonary resection in patients aged 80 years or over with clinical stage I non-small cell lung cancer: prognostic factors for overall survival and risk factors for postoperative complications. J Thorac Oncol  2009; 4: 1247– 53. Google Scholar CrossRef Search ADS PubMed  4 Sawabata N, Miyaoka E, Asamura H, Nakanishi Y, Eguchi K, Mori M et al.   Japanese lung cancer registry study of 11,663 surgical cases in 2004: demographic and prognosis changes over decade. J Thorac Oncol  2011; 6: 1229– 35. Google Scholar CrossRef Search ADS PubMed  5 Saji H, Ueno T, Nakamura H. Committee for Scientific Affairs TJAfCSTJ: prospective observational cohort study of postoperative risk and prognosis scoring for elderly patients with medically operable lung cancer (JACS1303). Gen Thorac Cardiovasc Surg  2016; 64: 634– 5. Google Scholar CrossRef Search ADS PubMed  6 Girones R, Torregrosa D, Gomez-Codina J, Maestu I, Tenias JM, Rosell R. Prognostic impact of comorbidity in elderly lung cancer patients: use and comparison of two scores. Lung Cancer  2011; 72: 108– 13. Google Scholar CrossRef Search ADS PubMed  7 Miyazaki T, Yamasaki N, Tsuchiya T, Matsumoto K, Kunizaki M, Taniguchi D et al.   Inflammation-based scoring is a useful prognostic predictor of pulmonary resection for elderly patients with clinical stage I non-small-cell lung cancer. Eur J Cardiothorac Surg  2015; 47: e140– 5. Google Scholar CrossRef Search ADS PubMed  8 Haruki T, Yurugi Y, Wakahara M, Matsuoka Y, Miwa K, Araki K et al.   Simplified comorbidity score for elderly patients undergoing thoracoscopic surgery for lung cancer. Surg Today  2017; 47: 718– 25. Google Scholar CrossRef Search ADS PubMed  9 Travis WD, Brambilla E, Burke AP, Marx A, Nicholson AG. WHO Classification of Tumours of the Lung, Pleura, Thymus and Heart , 4th edn. Lyon, France: International Agency for Research on Cancer, 2015. 10 TNM Classification of Malignant Tumours , 7th edn. Oxford, UK: Wiley-Blackwell, 2009. 11 Charlson ME, Sax FL, MacKenzie CR, Fields SD, Braham RL, Douglas RGJr. Resuscitation: how do we decide? A prospective study of physicians' preferences and the clinical course of hospitalized patients. JAMA  1986; 255: 1316– 22. Google Scholar CrossRef Search ADS PubMed  12 Colinet B, Jacot W, Bertrand D, Lacombe S, Bozonnat MC, Daures JP et al.  ; oncoLR health network. A new simplified comorbidity score as a prognostic factor in non-small-cell lung cancer patients: description and comparison with the Charlson’s index. Br J Cancer  2005; 93: 1098– 105. Google Scholar CrossRef Search ADS PubMed  13 Kawashima M, Murakawa T, Shinozaki T, Ichinose J, Hino H, Konoeda C et al.   Significance of the Glasgow Prognostic Score as a prognostic indicator for lung cancer surgery. Interact CardioVasc Thorac Surg  2015; 21: 637– 43. Google Scholar CrossRef Search ADS PubMed  14 Fukuse T, Satoda N, Hijiya K, Fujinaga T. Importance of a comprehensive geriatric assessment in prediction of complications following thoracic surgery in elderly patients. Chest  2005; 127: 886– 91. Google Scholar CrossRef Search ADS PubMed  15 Hurvich CM, Tsai CL. Regression and time series model selection in small samples. Biometrika  1989; 76: 297– 307. Google Scholar CrossRef Search ADS   16 The 2015 annual report from the Ministry of Health, Labour and Welfare in Japan, http://www.mhlw.go.jp/toukei/saikin/hw/life/life15/. 17 Brunelli A, Kim AW, Berger KI, Addrizzo-Harris DJ. 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. Google Scholar CrossRef Search ADS PubMed  18 Guerra M, Neves P, Miranda J. Surgical treatment of non-small-cell lung cancer in octogenarians. Interact CardioVasc Thorac Surg  2013; 16: 673– 80. Google Scholar CrossRef Search ADS PubMed  19 Rueth NM, Parsons HM, Habermann EB, Groth SS, Virnig BA, Tuttle TM et al.   The long-term impact of surgical complications after resection of stage I nonsmall cell lung cancer: a population-based survival analysis. Ann Surg  2011; 254: 368– 74. Google Scholar CrossRef Search ADS PubMed  20 Lugg ST, Agostini PJ, Tikka T, Kerr A, Adams K, Bishay E et al.   Long-term impact of developing a postoperative pulmonary complication after lung surgery. Thorax  2016; 71: 171– 6. Google Scholar CrossRef Search ADS PubMed  21 Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis  1987; 40: 373– 83. Google Scholar CrossRef Search ADS PubMed  22 Birim O, Kappetein AP, Bogers AJ. Charlson comorbidity index as a predictor of long-term outcome after surgery for nonsmall cell lung cancer. Eur J Cardiothorac Surg  2005; 28: 759– 62. Google Scholar CrossRef Search ADS PubMed  23 Jacot W, Colinet B, Bertrand D, Lacombe S, Bozonnat MC, Daures JP et al.   Quality of life and comorbidity score as prognostic determinants in non-small-cell lung cancer patients. Ann Oncol  2008; 19: 1458– 64. Google Scholar CrossRef Search ADS PubMed  24 Battafarano RJ, Piccirillo JF, Meyers BF, Hsu HS, Guthrie TJ, Cooper JD et al.   Impact of comorbidity on survival after surgical resection in patients with stage I non-small cell lung cancer. J Thorac Cardiovasc Surg  2002; 123: 280– 7. Google Scholar CrossRef Search ADS PubMed  © The Author 2017. Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. All rights reserved.

Journal

European Journal of Cardio-Thoracic SurgeryOxford University Press

Published: Apr 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 12 million articles from more than
10,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Unlimited reading

Read as many articles as you need. Full articles with original layout, charts and figures. Read online, from anywhere.

Stay up to date

Keep up with your field with Personalized Recommendations and Follow Journals to get automatic updates.

Organize your research

It’s easy to organize your research with our built-in tools.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

Monthly Plan

  • Read unlimited articles
  • Personalized recommendations
  • No expiration
  • Print 20 pages per month
  • 20% off on PDF purchases
  • Organize your research
  • Get updates on your journals and topic searches

$49/month

Start Free Trial

14-day Free Trial

Best Deal — 39% off

Annual Plan

  • All the features of the Professional Plan, but for 39% off!
  • Billed annually
  • No expiration
  • For the normal price of 10 articles elsewhere, you get one full year of unlimited access to articles.

$588

$360/year

billed annually
Start Free Trial

14-day Free Trial