Outcomes of Lobectomy in ‘Active’ Octogenarians with Clinical Stage I Non-Small-Cell Lung Cancer

Abstract

Introduction: In octogenarians with early stage of lung cancer, sublobar resection has been shown to be no inferior survival to lobectomy. However, some octogenarians remain physically and mentally active.

Methods: We retrospectively studied 65 octogenarians who underwent curative resection for clinical stage I NSCLC (excluding adenocarcinoma in situ). All patients were assessed by cardiologists and underwent stair-climbing tests (five flights, 18 m) and cognitive function tests. Lobectomy was performed in patients who could climb 5 flights of stairs without stopping or oxygen desaturation of >4%.

Results: The actuarial survival rate was 68.6% at 5 years, and the median survival time was 109.2 months. Forty-three patients met the criteria for lobectomy. As compared with sublobar resection, lobectomy was associated with significantly higher rates of overall survival (78.4% vs. 48.5%; p = 0.02) and disease-specific survival (88.4% vs. 61.7%; p = 0.02) at 5 years. On multivariate analysis, male sex (hazard ratio, 3.827; 1.382–10.596) and sublobar resection (2.261; 1.054–5.360) were independent risk factors for survival. Mental disorders occurred in 6 patients (9.2%), and their score on preoperative cognitive function tests was significantly lower than that of patients without mental disorders (22.7 vs. 26.0, p <0.01).

Conclusion: Outcomes of lobectomy are good in physically and mentally competent octogenarians.

Introduction

With improvements in public health education and increasing medical resources and facilities, the elderly population has been growing in many countries. Surgeons face increasing numbers of elderly patients with lung cancer. Complete surgical resection, especially lobectomy, remains the curative treatment of choice for early-stage non-small cell lung cancer (NSCLC), excluding pathologically noninvasive adenocarcinoma. Recent improvements in patient selection, operative techniques, and postoperative management have contributed to lower mortality from lung cancer surgery. Octogenarians tend to have poor residual cardiopulmonary function and multiple comorbidities, potentially increasing the risk of operative morbidity and mortality and especially postoperative cardiopulmonary complications adversely affect the postoperative quality of life. Thus, some surgeons recommend sublobar resections for elderly patients owing solely to their chronological age.^1,2) On the other hand, the number of octogenarians with good physical and mental abilities who can tolerate lobectomy has also been increasing. It remains unclear whether sublobar resection should be performed in such octogenarians. Another very important factor in the treatment strategy for octogenarians with lung cancer is the occurrence of mental disturbance during postoperative follow-up. Few studies have examined the relation between the results of preoperative cognition tests and long-term mental disorders after surgery. We have prospectively defined strict criteria for deciding whether lobectomy is indicated in octogenarians. The present study was designed to elucidate not only postoperative mortality and morbidity but also long-term outcomes, including mental disturbance, in octogenarians with clinical stage I NSCLC who underwent lobectomy.

Patients and Methods

Our institutional internal review board approved this retrospective study. From January 2000 through December 2008, a total of 1453 patients underwent complete pulmonary resection for primary NSCLC at Kanagawa Cancer Center. Our standard procedure for stage I NSCLC was lobectomy, excluding cases with pure ground glass opacity (pure GGO) on high-resolution computed tomography, indicating adenocarcinoma in situ. Chronological age was not considered a contraindication for surgery. The same surgical team performed preoperative assessment, surgery, and postoperative management. All patients underwent a preoperative functional cardiologic evaluation, including ultrasonic cardiography and exercise electrocardiography, and pulmonary function tests. If the patients had the past history of ischemic heart disease, coronary computed tomography or coronary angiography was added. If the predicted postoperative forced expiratory volume in 1 s (FEV1) exceeded 500 ml/m² and the patient had no serious cardiopulmonary disease and successfully completed cognitive function and stair-climbing tests, lobectomy with mediastinal lymph node dissection was indicated. Stair-climbing tests (5 flights, 18 m) were performed in patients without leg disabilities to check their cardiopulmonary status. Patients were asked to climb 5 flights of stairs at a pace of their own choice; when they stopped, the reason was recorded. Pulse rate and capillary oxygen saturation were monitored by a portable pulse oximeter during the test. Patients who could climb 5 flights without stopping and with oxygen desaturation of no greater than 4% were considered candidates for lobectomy. Sublobar resection was done in patients who could not pass the stair-climbing test. Cognitive function tests (i.e., Mini-Mental State Examination) were also performed within 4 weeks before surgery. Scores of 25–30 were considered to indicate normal cognition, scores of 18–24 mild to moderate impairment, and scores of 17 or less severe impairment. Patients with severe delirium or an Eastern Cooperative Oncology Group (ECOG) performance status (PS) of 2 to 4 were excluded from surgical candidates. Only PS 2 related directly to leg disability was not considered a contraindication to surgery. All patients were followed up in our hospital’s outpatient clinic at least twice per year, and their physical and mental status was recorded.

We retrospectively studied postoperative morbidity, mortality, outcomes, and mental status by reviewing the clinical characteristics of the patients (age, sex, smoking history, smoking index, forced vital capacity [FVC], forced expiratory volume in 1 s [FEV1], results of stair-climbing tests, results of Mini-Mental State Examinations, partial pressure of oxygen in arterial blood [PaO₂], co-existing cardiopulmonary disease, and clinical TNM stage), surgical records (extent of resection, operation time, blood transfusion volume, blood loss), and pathological records (histologic diagnosis, TNM stage). Statistical analysis was performed with the c² test, Fisher’s exact test, and Student’s t-test. Logistic-regression analysis was used for multivariate analysis, performed with Stat View for Windows (version 5.0; SAS Institute Inc., Cary, North Carolina, USA). P values of less than 0.05 were considered to indicate statistical significance. Actuarial survival curves were constructed by the Kaplan-Meier method, and all the figures were terminated at the point of 100 months. Univariate analyses were performed using the log-rank test and multivariate analyses were done with a Cox proportional-hazards model. The 7th edition of the UICC TNM staging system was used to stage disease.

Results

Patient characteristics

Among 1453 patients with primary NSCLC who underwent pathologically complete resection, we prospectively studied 65 (4.5%) patients with clinical stage I disease who were 80 years or older. Patients with a diagnosis of adenocarcinoma in situ were excluded. Table 1 shows the patients’ characteristics. Three patients with osteoarthritis of lower extremities had a PS of 2. Among 14 patients with chronic obstructive pulmonary disease, 8 patients received oral bronchodilators or inhaled medication. Eleven patients (16.7%) had previously been treated for other malignant diseases (gastric cancer in 3 patients, colorectal and breast cancer in 2 patients each, and laryngeal, uterine, thyroid, and renal cancer in one patient each). None of these patients had evidence of cancer recurrence at the time of lung resection. The stair-climbing test could not be done by 3 patients who had severe leg disability. Forty-five of the 65 patients could climb 5 flights of stairs without stopping or oxygen desaturation of greater than 4%. Twelve patients had oxygen desaturation exceeding 4%, 3 patients stopped climbing because of leg fatigue, and 2 patients stopped climbing because of breathing difficulty. On the Mini-Mental State Examination, 32 patients showed normal cognition, and 16 patients had mild to moderate cognitive impairment. Patients with severe cognitive impairment were excluded from surgical candidates.

Table 1.

Patients’ characteristics

Demographic variable
Age, years
Range	80–90
Mean	82.2 ± 2.2
Sex Male/Female	35/30
Performance status 0/1/2	53/9/3
Smoking history	39
Smoking index
Range	40–2175
Mean	574
Clinical stage IA/IB	35/30
Pathological stage
IA/IB/IIA/IIB/IIIA	33/20/3/3/6
Histology AD/SQ/Others	53/6/6
Procedure (Lobectomy/Seg/WWR)	43/3/19
Pulmonary function
FVC (L)
Range	1.69–4.35
Mean	2.62 ± 0.53
% FVC
Range	64–137
Mean	99.1 ± 18.2
FEV1
Range	1.20–3.53
Mean	1.91 ± 0.49
FEV1%
Range	53.5–92.0
Mean	73.3 ± 7.1
Mini-mental state examination
Normal/mild to moderate/Severe	32/16/0
Comorbidity total	45 (69%)
Hypertension	36 (55%)
Arrhythmia	6 (9.2%)
Ischemic heart disease	2 (3.1%)
COPD (FEV1/FVC <70%)	14 (21%)
Prior tuberculosis	5 (7.7%)
Post lobectomy	2 (3.1%)
Diabetes mellitus	7 (10.8%)
Renal failure	1 (1.5%)
Arteriosclerosis obliterans	1 (1.5%)
Hypertrophic cardiomyopathy	1 (1.5%)
Ischemic cerebral disease	1 (1.5%)

AD: adenocarcinoma; SQ: squamous cell carcinoma; Seg: segmentectomy; WWR: wide wedge resection; FVC: forced vital capacity; FEV1: forced expiratory volume in one second; COPD: chronic obstructive pulmonary disease

Surgical procedures, morbidity, and mortality

Lobectomy with mediastinal lymph node dissection was performed in 43 patients (66.2%), all of whom passed the stair-climbing test. Three patients underwent segmentectomy, and 19 patients underwent wedge resection because of poor performance on the stair-climbing test, insufficient cardiopulmonary reserve, and complex comorbidities. Mean operation time was 139 ± 58 min (range: 52 to 280 min). Mean blood loss was 44 ± 63 ml (range: 0 to 260 ml). Eight patients (12.3%) had cardiopulmonary complications, and 12 (18.5%) had multiple complications. Details of postoperative complications are as followed; 3 arrhythmia, 2 pneumonia, 2 prolonged air leak, each one of sputum retention, postoperative bleeding, empyema, transient cerebral ischemic attack, anuresis and delirium. There was no in-hospital death or evidence of any adverse effects on cardiopulmonary complications (Table 2).

Table 2.

Risk factors for postoperative cardiopulmonary complications

Variable	Complications (n = 8)	No complication (n = 57)	p value
Age (years)	82.6 ± 2.4	82.2 ± 2.2	0.60
Sex (M/F)	7/1	28/29	0.06
Smoke (Y/N)	7/1	32/25	0.13
%FVC	107 (± 15.9)	100 (± 17.9)	0.17
FEV1%	72.9 (± 6.6)	73.4 (± 7.3)	0.85
PaO2 (mmHg)			0.57
Mean	87.3	82.5
Range	72–100	64–93
Co-existing cardiopulmonary disease (+/–)	6/2	38/19	>0.99
Clinical stage (IA/IB)	3/5	33/24	0.45
Cognition test (normal/mild to moderate)	2/4	30/12	0.09
Procedure (lob./sublobar.)	7/1	36/21	0.25
Operation time (min)
Mean	154.8	137.8	<0.45
Range	93–280	52–267
Blood transfusion (+/–)	0/8	1/56	>0.99
Blood loss (g)
Mean	78.8	39.6	0.10
Range	10–255	5–260

Data are presented as mean (± standard deviation). FEV: forced expiratory volume in 1s; FVC: forced vital capacity; PaO₂: arterial oxygen tension; lob.: lobectomy; sublobar.: sublobar resection

Long-term outcomes

Median follow-up was 62.8 months, and there were 21 deaths during follow-up. The causes of death were recurrent lung cancer in 14 patients and other diseases in 7 patients (respiratory diseases in 3 patients, other cancers in 2 patients, heart failure in one patient, and cerebrovascular disease in one patient). The overall survival rate at 5-year was 68.6%, and median survival time was 109.2 months. When survival was analyzed according to the extent of resection, the rates of overall survival and disease-free survival were significantly higher in the lobectomy group than in the sublobar resection group (Figs. 1 and 2).

Fig. 1.

Overall survival curves according to the extent of surgery in clinical stage I non-small cell lung cancer (NSCLC). The 5-year survival rate and median survival time were respectively 78.4% and 109.2 months in the lobectomy group and 48.5% and 56.6 months in the sublobar resection group.

Fig. 2.

Disease-free survival curve according to extent of resection in clinical stage I non-small cell lung cancer (NSCLC). In the lobectomy group, the disease-free survival rate at 5 years was 88.4%, and median survival time was not reached. In the sublobar resection group, the disease-free survival rate at 5 years was 61.7%, and median survival time was 109.7 months.

The rate of mortality from recurrent lung cancer was slightly but not significantly lower in the lobectomy group (n = 6, 14.3%) than in the sublobar resection group (n = 8, 36.3%). Mortality from other diseases also did not differ significantly between the groups (4 deaths in the lobectomy group vs. 8 deaths in the sublobar resection group). Postoperative cardiopulmonary complications did not adversely affect survival. The 5-year survival rate according to clinical stage was 68.2% in stage IA and 68.0% in stage IB. Median survival time in stage IB was 95.7 months (not reached in stage IA). The 5-year survival rate according to pathological stage was 71.2% in stage IA, 71.1% in stage IB, 66.7% in stage IIA, 33.3% in stage IIB, and 66.7% in stage IIIA. On univariate analysis, sex (p <0.01) and extent of surgery (p = 0.03) were significantly related to overall survival. Multivariate analysis showed that male sex (hazard ratio 3.827, 1.382–10.596) and sublobar resection (hazard ratio 2.261, 1.054–5.360) were independently related to overall survival.

During follow-up, 6 patients (9.2%) had severe mental disturbance that requiring familial and social support. The median time from operation to the onset of mental disturbance was 3.7 years (range: 1.2–8.3 years). Five patients had mild-to-moderate cognitive impairment before operation, and the other patient had normal cognitive function. The incidence of severe cognitive impairment in the remote period after operation was significantly higher in the subgroup of patients with mild-to-moderate cognitive impairment before surgery than in the subgroup with normal cognitive function (p <0.01). The median score on the Mini-Mental State Examination was significantly lower than in the 6 patients with severe mental disturbance after surgery than in patients without severe mental impairment (22.7 ± 2.07 vs. 26.0 ± 2.35, p <0.01). The extent of surgery was unrelated to the incidence of severe mental disturbance (p >0.99).

Discussion

The elderly population is increasing in many advanced countries. The growth rate of the elderly population is estimated to be highest in Japan, and clinicians have to meet the challenge of treating more elderly patients with lung cancer. The White Paper on the National Lifestyle 2009, issued by the Japanese Ministry of Health, Labour and Welfare, reported that the average life expectancy at 80 years of age was 8.66 years in men and 11.68 years in women.

Limited surgery for early lung cancer has been reported to be equivalent to lobectomy.^3,4) Phase III clinical trials comparing lobectomy with limited resection are ongoing.^5,6) At present, lobectomy remains the standard procedure for early lung cancer other than adenocarcinoma in situ. In high-risk patients unable to tolerate lobectomy, limited surgery is performed.⁷⁾ The rate of limited surgery is particularly high among elderly patients. In our series, limited surgery was performed in 33.8% of patients 80 years or older, as compared with 16.7% of those younger than 80 years. The main reasons for performing limited surgery are the avoidance of postoperative complications in elderly patients with reduced reserve capacity and the maintenance of patients’ quality of life after surgery.²⁾ The outcomes of limited resection have been reported to be similar to those of lobectomy, the standard procedure in elderly patients with lung cancer.^2,3) In addition, a study based on the Surveillance Epidemiology and End Results (SEER) database failed to demonstrate the superiority of lobectomy.¹⁾ The potential contribution of lobectomy to survival is most likely offset by factors such as the shorter life expectancy of elderly patients and the increased risk of serious complications.

Nonetheless, some elderly patients can adequately tolerate lobectomy because of good cardiopulmonary capacity and extremely high mental and physical performance. The selection of patients likely to have improved outcomes after lobectomy is thus very important. We used stair-climbing tests in addition to standard electrocardiography and pulmonary function testing to evaluate cardiopulmonary function in elderly patients. The stair-climbing test can be simply and easily performed to comprehensively evaluate oxygen-transport status, without the need for special devices or equipment. It has been reported to be superior to ergometry for the estimation of maximum oxygen consumption (VO₂ max).^8,9) Thoracic surgery should not be performed in patients with a VO₂ max of less than 15 mL/min/kg. Patients who can climb 5 flights of stairs have been reported to have a VO₂ max of greater than 20 mL/kg/min, making then suitable candidates for surgery, and a decrease in percutaneous oxygen saturation by more than 4% during stair-climbing testing has been linked to an increased risk of postoperative complications.^10,11) On the basis of these findings, since 1999 we have prospectively performed lobectomy in elderly patients 80 years or older who can climb 5 flights of stairs without stopping or oxygen desaturation of >4%.

Decreased cognitive function may lead to an increased incidence of postoperative complications in elderly patients, and the Mini-Mental State Examination has been reported to be useful for predicting the risk of postoperative complications in elderly patients with lung cancer.¹²⁾ In our study, decreased cognitive function before surgery was not a significant risk factor for postoperative complications, but was associated with a further decline in cognitive function in the late phase, negatively affecting patients’ quality of life. The extent of resection was apparently unrelated to decreased cognitive function in the late phase, but longer survival accompanying improved survival rates might also lead to an increased incidence of cognitive dysfunction. This problem is specific to elderly patients, and effective solutions remain elusive. Preoperative Mini-Mental State Examinations may provide useful information for deciding the treatment strategy in elderly patients, but larger studies are needed to confirm our findings.

In our series, the incidence of postoperative complications was 18.5%, and 6.2% of patients had life-threatening complications, but there was no in-hospital death. Our results compare favorably with the findings of previous studies (Table 3).^3,13–17) In general, the onset of complications increases oxygen consumption, and patients with compromised cardiopulmonary function, such as elderly individuals, cannot meet the increased oxygen demand, resulting in multiple organ failure.¹⁸⁾ The height of the stairs able to be climbed during stair-climbing testing has been reported to be a predictor of the risk of complications after lung resection. The use of stair-climbing tests for the selection of patients might therefore contribute to a decreased incidence of postoperative complications.^19,20) As for the outcomes of resection in our study, the 5-year survival rate was 68.6%, with a mean survival time of 109.2 months. These results compare favorably with those of previous studies of surgical outcomes in stage I NSCLC (Table 3).^3,13–17) As compared with limited resection, lobectomy had better outcomes in terms of not only overall survival, but also disease-free survival (Figs. 1 and 2). Better outcomes after lobectomy are probably attributed to the fact that the results of preoperative stair-climbing testing as well as the findings of conventional cardiopulmonary function testing were considered in the selection of candidates for lobectomy. Our results indicate that lobectomy can lead to cure in patients with stage I NSCLC who have adequate cardiopulmonary function, even if they are elderly.

Table 3.

Summary of stage I non-small cell lung cancer resection in octogenarians

First author	Year	Number	Morbidity (%)	Mortality (%)*	5-year survival (%)
Aoki13)	2003	49	41	2	52.5
Brock14)	2004	68	44	8.8	34
Port3)	2004	61	38	1.6	46
Matsuoka15)	2005	40	20	0	56.9
Mun16)	2007	55	26	3.6	65.9
Okami17)	2009	367	8.4	1.4	55.7
Present	2013	65	18.5	0	68.6

*Death within 30 days after initial operation or in-hospital death

Our study had several important limitations. It was retrospective, and the results may have been affected by selection bias. Because it was an observational study, patients were prospectively enrolled, but were not randomly assigned to surgical procedures. In addition, information on outcomes is unavailable for patients who did not undergo resection. However, because our trial was conducted at a single center, it had several strong points. For example, surgical procedures and postoperative care were carefully managed in accordance with well-defined treatment guidelines and detailed information on clinicopathological characteristics and long-term follow-up were available.

Conclusion

Lobectomy was associated with good long-term outcomes in physically and mentally competent elderly patients with stage I NSCLC. Decreased cognitive function at the time of surgery might contribute to a poorer quality of life in the late postoperative phase.

Disclosure Statement

None declare.