Risks and rewards of the surgical treatment of lung cancer in octogenarians

Abstract

 

OBJECTIVES

Surgeons will face an increasing number of octogenarians with lung cancer potentially curable by surgery. The goal of this study was to evaluate short- and long-term outcomes after lung resection.

METHODS

We performed a single-centre study of consecutive patients ≥80 years old, surgically treated for suspected lung cancer between 2009 and 2016. Age, sex, performance status, lung function, surgical approach, type of lung resection, complications, in-hospital and 30- and 90-day deaths and long-term survival were analysed.

RESULTS

Two hundred and fifty-seven patients were enrolled. The median age was 82 years (range 80–97). One hundred and thirty-four patients were treated by thoracotomy and 123 by video-assisted thoracic surgery [10 (8.1%) converted]. Two hundred and thirty-two underwent lobar resection and 25 underwent sublobar resection. There were no intraoperative deaths and 9 admissions to the intensive therapy unit; 112 (43.6%) patients suffered complications: More complications occurred after lobar versus after sublobar resections [45.7% vs 24% (P = 0.037)] and in those with chronic obstructive pulmonary disease (COPD) [57.4% vs 40% (P = 0.02)]. The 30-day mortality was 3.9% (n = 10) and the 90-day mortality was 6.22% (n = 16). One hundred and sixty-seven patients died during the study period; patients with non-small-cell lung cancer (n = 233) had a median survival of 46.5 months with 67.2% alive at 2 years and 40.8% at 5 years. Pathological stage and R status were independent prognostic factors for survival.

CONCLUSIONS

Surgery for malignancies in octogenarians is feasible and safe with good long-term outcomes. The risk of postoperative complications, especially in those with COPD, is high but can be minimized with sublobar resection. Postoperative mortality is acceptable, and long-term survival is primarily governed by lung cancer stage. Age is no reason to deny patients surgery for early-stage disease.

Lung cancer is a common disease [1] in the Western world; 470 000 cases were diagnosed in Europe in 2018 [2].

INTRODUCTION

Lung cancer is a common disease [1] in the Western world; 470 000 cases were diagnosed in Europe in 2018 [2]. It is also a fatal disease: Worldwide, it accounts for 11.6% of new cancer cases but 18.4% of all cancer deaths [2]. In 2018, the United Kingdom (UK) saw 52 320 new cases and 37 688 deaths [2].

The incidence of lung cancer increases with age; in Europe, ∼6% of lung cancers were diagnosed in patients below 50 years of age, 20% in those 50–59 years of age, 29% in those 60–69 years of age and 44% in patients over 70 years of age [3]. Generally, in Western countries, a gradual increase in life expectancy has been reported. In the UK, there are 1.6 million people aged 85+; this group is predicted to double by 2041 and triple by 2066 [4]. The greater proportion (up to 38%) of squamous cell carcinomas in older patients with lung cancer means that they are more likely to have a localized, resectable tumour than their younger counterparts [5]. Older patients also receive increasing levels of investigations for comorbidities; hence more asymptomatic early-stage tumours will be detected.

These data suggest that an increasing number of octogenarians with limited stage disease will be discussed with thoracic surgeons. However, due to the comorbidities and frailty of octogenarians, surgeons have traditionally been sceptical about offering resection to octogenarians with localized lung cancer due to the risks involved and the belief that these patients would die of other diseases before they would die of lung cancer. With improved surgical management, such as minimally invasive surgery, limited resection and enhanced recovery after surgery programmes, lung resection can currently be offered to a greater number of older patients [6].

The aim of this study was to evaluate the safety and long-term outcomes of lung resection for suspected lung cancer in octogenarians.

MATERIALS AND METHODS

An analysis of our thoracic surgery unit registry was conducted to identify all patients aged ≥80 years who underwent lung resection for suspected or known lung cancer between February 2009 and December 2016. Data were collected retrospectively from hospital records and prospectively from a compiled computerized database. Clinical records were analysed for sex, age, performance status (PS) (Eastern Cooperative Oncology Group scale), comorbidities, side, approach [video-assisted thoracic surgery (VATS) vs thoracotomy] and extent of resection, length of hospital stay (LOS), complications, readmissions and long-term survival. Survival was measured from the date of surgery. Complications were divided into cardiac and respiratory complications. Grade 1 or 2 complications were managed with medical treatment on the ward, and grades 3 and 4 required readmission to intensive care or reoperation. Chest infection and pneumonia were defined as chest X-ray findings of atelectasis or consolidation, raised white cell count, temperature >38°C and increased oxygen requirement. Prolonged air leak (PAL) was defined as air leak for >5 days. VATS was the standard approach for small peripheral tumours; tumours above 5 cm or bulky nodal disease were approached by thoracotomy. All of the patients were followed up until death or until May 2020 using the UK NHS tracking system. This research project has been approved as a clinical audit (project number 7753) at our institution.

All potential prognostic indices were measured at the time of surgery and evaluated as categorical variables. Continuous data were reported with medians and ranges, whereas categorical data were reported with counts and percentages. The Chi-square test was used to compare categorical variables. Survival and prognostic factors were analysed by the Kaplan–Meier method; survival comparison between groups of patients was performed by means of the log-rank test. Factors that significantly associated with survival in univariable analysis (at P < 0.10) were tested for their independent role in multivariable analysis using the Cox proportional hazards model. The stepwise backward procedure based on the likelihood ratio was used to assess the significance of covariates included in the model. Hazard ratios and 95% confidence intervals (CIs) were used to quantify the observed differences. Probability values <0.05 were considered statistically significant. All analyses were conducted using the SPSS (IBM-SPSS Inc., Armonk, NY, USA) software package.

RESULTS

Study population

Two hundred and fifty-seven patients were enrolled in the analysis [median age 82 years (range 80–97)]. Forty-nine (19.1%) patients were older than 85. One hundred and six (41.2%) patients were women. The median PS was 1 (range 0–3), 31 patients were PS 2 and 1 was PS 3. Other characteristics are summarized in Tables 1 and 2. One hundred and fifty (58.4%) lesions were located on the right side and 107 (41.6%), on the left side. One hundred and thirty-four (52.1%) patients underwent thoracotomy, and 123 (47.9%) had VATS with 10 patients (8.1%) converted due to intraoperative bleeding (n = 5), oncological reasons (n = 3) and extensive adhesions (n = 2). Two hundred and thirty-two (90.7%) patients underwent a lobar resection or greater (bilobectomy n = 8 and pneumonectomy n = 2) and 25 (9.3%), a sublobar resection (12 segmentectomy and 13 wedge resection). The surgical approaches and the types of resections according to PS are shown in theSupplementary Material, Table 6; wedge resection was performed mainly in patients with PS 2 and 3. All patients underwent surgery for presumed primary lung cancer, although 10 had lung metastases and 3 had benign disease based on the final pathological reports.

Table 1:

Patients characteristics (n = 257)

Patient characteristics	Number of patients	Percentage
Age group
80–84	208	80.9
≥85	49	19.1
Sex
Female	106	41.2
Male	151	58.8
Performance status
0	59	23
1	166	66.6
2	31	12
3	1	0.4
Lung function tests
FEV1	189	110 (40–174)
FVC	164	105 (55–171)
TLCO	180	69 (31–125)
Comorbiditiesa
Hypertension	91	35.4
AF	28	10.9
Other cardiovascular disease	69	26.8
2 or more CVD	30	11.7
COPD	54	21
CKD	10	3.9
DM	34	13.2

^aSome patients have >1 comorbidity.

AF: atrial fibrillation; CKD: chronic kidney disease; COPD: chronic obstructive pulmonary disease; CVD: cardiovascular disease; DM: diabetes mellitus; FEV1: forced expiratory volume in the first second; FVC: forced vital capacity; TLCO: carbon monoxide transfer capacity.

Table 2:

Oncological characteristics (n = 254)

Histological diagnosis	N	%
Adenocarcinoma	122	47
Squamous cell carcinoma	99	38.5
Typical carcinoid	7	2.7
Large cell cancer	7	2.3
Othera	9	4.3
Metastasis	10	3.9
Benign pathologyb	3	1.2
R status
0	238	93.7
1	16	6.3
Pathological stagec
IA1	17	7.3
IA2	32	13.7
IA3	9	3.9
IB	67	28.7
IIA	21	9.0
IIB	51	21.9
IIIA	27	11.6
IIIIB	9	3.9

^a n = 1 adenoid cystic carcinoma, n = 1 pleomorphic cancer, n = 2 small cell cancer, n = 5 adenosquamous carcinoma.

^b n = 1 granulomatous inflammation, n = 1 Ig4-related lung disease, n = 1 acute inflammation and scarring.

^cPrimary lung cancers.

Early postoperative course

Complications

There were no intraoperative deaths. One hundred and twelve (43.6%) patients suffered postoperative complications, 36 (12%) of whom had multiple complications. Nine patients were admitted to the intensive therapy unit for respiratory failure (n = 3), hospital-acquired pneumonia (n = 2), cardiovascular instability (n = 2), aspiration pneumonia (n = 1) and sepsis (n = 1). Four patients required a reoperation for bleeding. The complication rate was similar for patients <85 years [87/208 (41.8%)] and patients ≥85 [25/49 (51%)] (P = 0.24).

Factors associated with complications

Whereas sex and PS were not correlated with the complication rate (P = 0.24 and P = 0.78), chronic obstructive pulmonary disease (COPD) was significantly correlated with postoperative complications compared to patients without respiratory comorbidities: 31/54 (57.4%) versus 81/206 (39.9%); P = 0.02. Other comorbidities (cardiac, renal, diabetes), laterality of surgery and pathological stage did not correlate with the complication rate (Table 3). Postoperative complications were reported in 59 (44%) patients after open surgery (53 lobar resection, 6 sublobar resections) and in 53 (43.1%) after VATS (all lobar resection) (P = 0.9). Chest infection and PAL rates were 14.6% (n = 18) and 12.2% (n = 15) in the VATS group and 10.4% (n = 14) and 13.4% (n = 18) in the thoracotomy group (P = 0.2 and P = 0.4), respectively. The atrial fibrillation (AF) rate was marginally increased after a thoracotomy compared to VATS: 15.7% (n = 21) versus 11.4% (n = 14) (P = 0.2). The complication rate was significantly higher after resection of a lobe or more compared to sublobar resection: 45.7% (n = 106) versus 24% (n = 6) (P = 0.028). The most frequent complication in lobar resections was chest infection/atelectasis (n = 34, 25.7%), followed by PAL (n = 33, 25%) and AF (n = 25, 18.9%). In the sublobar group, the most frequent complication was AF (n = 4, 16%). In the subgroup of lobectomy only (n = 232), comparing VATS (n = 112) to open surgery (n = 120), 53 (47.3%) patients who had VATS lobectomies had complications compared to 53 (44.2%) patients who had open lobectomies (P = 0.36). The most frequent complications were (VATS compared to open) as follows: PAL in 17 (15.2%) versus 16 (13.3%) patients, pneumonia in 15 (13.4%) versus 19 (15.8%) patients and AF in 12 (10.7%) versus 13 (10.8%) patients, respectively (P = 0.6).

Table 3:

Complications in the study population (n = 257)

Variables	Number of patients (n = 257)	Number of complications (n)	Complications (%)	P-value
Sex				0.24
Male	151	69	45.7
Female	106	43	40.6
Performance status				0.78
PS 0	59	24	40.7
PS 1	166	74	44.6
PS 2–3	32	14	43.75
Comorbidities
None	62	22	35.5	0.18
Cardiac	153	68	44.4	0.16
Renal	10	6	60	0.2
Diabetes mellitus	34	15	44.1	0.5
COPD	54	31	57.4	0.02
Laterality				0.23
Left side	107	42	39.3
Right side	150	70	46.7
Surgical access				0.9
VATS	123	53	43.1
Open	134	59	44
Type of resection				0.037
Lobar	232	106	45.7
Sublobar	25	6	24
Histological type				0.18
Adenocarcinoma	122	55	45.1
Squamous cell carcinoma	99	48	48.5
Adenosquamous carcinoma	5	2	40
Lung metastases	10	3	30
Benign	3
Other	18	4	22.2
Staging				0.29
IA1	17	8	36.4
IA2	32 (12.8)	13	39.4
IA3	9 (3.5)	3	30
IB	67 (26.8)	29	42
IIA	21 (8.6)	7	31.2
IIB	51 (20.2)	26	50
IIIA	27 (10.9)	18	64.3
IIIB	9 (3.5)	5	55.6

COPD: chronic obstructive pulmonary disease; PS: performance status; VATS: video-assisted thoracic surgery.

Length of stay

The median LOS in the study population was 7 days (range 2–54). In patients <85 (n = 208), the median LOS was 7 days (range 2–54 days) and, in patients ≥85 (n = 49), the LOS was 8 days (range 3–34) (P = 0.6). The median LOS after thoracotomy was significantly greater than that after VATS [8 days (range 3–54) and 6 days (range 2–34), respectively (P = 0.004)]. There was also a statistical difference in the LOS between lobar versus sublobar resection [7 days (range 2–54) vs 5 days (range 3–15) (P = 0.04)].

Mortality

The postoperative mortality in the hospital was 3.11% (n = 8); only 1 patient died after 30 days in the hospital [<85 (n = 6, 2.9%) and ≥85 (n = 2, 4.1%), P = 0.47]. The causes of death were hospital-acquired pneumonia (n = 4), aspiration pneumonia (n = 1), acute renal failure (n = 1), right heart failure (n = 1) and multiorgan failure following intraoperative liver injury and intraoperative bleeding (n = 1). Seven patients were males and 7 deaths followed right-sided procedures. Three patients had no major comorbidities and 5 had more than 1 comorbidity. Two patients had a VATS lobectomy and 6 had an open lobectomy.

The 30-day mortality was 3.89% (n = 10) including 7 who died in-hospital: 8 were males. Three patients had no comorbidities, and 7 had >1 major comorbidity. Three of the 10 patients had VATS lobectomies and 7 had open lobectomies (1 sleeve). Three patients died after discharge at home. Seven patients who died during the first 30 postoperative days were <85 (3.36%), and 3 patients were ≥85 (6.97%) (P = 0.29). The 90-day mortality was 6.22% (n = 16). Twelve of the 16 patients were <85 (5.76%), and 4 were ≥ 85 (9.03%) (P = 0.36). Seven patients had no major comorbidities, 2 patients had only 1 major comorbidity and 7 had >1 major comorbidity. Three patients had a VATS lobectomy and 13, an open lobectomy.

Readmission

Eight (3.11%) patients were readmitted to our surgical centre within 30 days of surgery. The median age was 82 (range 81–84). In this group, the patients had a history of COPD (n = 4), hypertension (n = 1), ischaemic heart disease (n = 1) and diabetes mellitus (n = 1). All of the patients had a lobectomy for primary lung cancer (VATS n = 5 and thoracotomy n = 3). Four patients were readmitted with chest infections treated with intravenous antibiotics, 2 following discharge with a drain in situ for PAL, 1 with postoperative empyema and 1 with sputum retention causing lobar atelectasis.

Long-term survival

The median overall survival (OS) of the study population (n = 257) was 47.8 months (95% CI 38.8–57.2): 167 (69.5%) patients died and 81 (31.5%) were alive at the last follow-up. Survival was similar in the age groups <85 and ≥85: 51.4 months (43.4–59.4) versus 40.4 months (95% CI 24.5–56.3) (P = 0.2, Fig. 1). Excluding carcinoid and small-cell lung cancer, we calculated the median OS in patients with non-small-cell lung cancer (NSCLC). The median OS of patients with NSCLC (n = 233) was 45.7 months (95% CI 38–54.4). The survival rate at 2 and 5 years was 67.2% and 40.8%. Survival was similar in the age groups <85 and ≥85: 47.8 months (95% CI 40–55.5) versus 40.4 months (95% CI 11.5–69.4) (P = 0.21). The survival at 2 and 5 years in the age groups <85 and ≥85 was 68.6% versus 56.8% and 41.8% versus 36.3%, respectively.

Figure 1:

OS related to age according to Kaplan–Meier (n = 257). OS: overall survival.

Factors associated with long-term survival in patients with non-small-cell lung cancer

In the patients with NSCLC, pathological stage correlated significantly with median OS (P = 0.004) (Fig. 2) as did resection status (R0 vs R1/R2) (P = 0001). Surgical approach and sex showed a trend towards an association with survival: VATS 57.1 months (95% CI 41.2–73) versus open 39.3 months (95% CI 28.1–50.5) P = 0.072 (Fig. 2); female versus male, 58.5 months (95% CI 49.3–67.7) versus 41.7 months (95% CI 33.9–49.5) P = 0.068. PS did not correlate with survival (P = 0.26). Neither did cardiac comorbidities (P = 0.85), COPD (P = 0.53), chronic kidney disease (P = 0.73), diabetes (P = 0.68), side of disease (P = 0.99), lobar versus sublobar resection (P = 0.60) and histological diagnosis (P = 0.43) (Table 4). On multivariable analysis, the pathological stage and resection status were independent prognostic factors for survival (Table 5). Approach and sex were not significantly correlated.

Figure 2:

OS related to pathological stage in non-small-cell lung cancer according to Kaplan–Meier (n = 233). OS: overall survival.

Table 4:

Univariable analysis in patients with NSCLC (n = 233)

Variables	Number of patients	Overall survival (months)	95% CI	P-value
Age				0.21
<85	189	47.4	39.2–55.5
≥85	44	40.4	11.4–69.4
Sex				0.068
Male	94	58.5	49.3–67.7
Female	139	41.7	33.9–49.5
PS				0.27
0	50	54.3	39.8–68.8
1	152	47.9	36.3–59.5
2	31	38.6	28–49.2
Cardiac comorbidities				0.85
No	167	46.5	36.9–56.1
Yes	66	45.5	28.8–62.2
COPD				0.53
No	182	51.7	42.2–61.2
Yes	51	42.8	34.7–50.8
CKD				0.73
No	200	50.2	42.7–57.6
Yes	10	43.9	24–63.7
Diabetes				0.68
No	181	51.7	43.2–60.2
Yes	32	42.4	26.2–58.6
Side of disease				0.99
Right	134	42.8	32–53.6
Left	99	50.2	39.3–61.1
Type of approach				0.072
VATS	115	57.1	41.2–73
Thoracotomy	118	39.3	28–50.5
Type of resection				0.60
Lobar	213	45.7	37.5–53.9
Sublobar	20	47.8	27.8–67.7
Histological type				0.92
Adenocarcinoma	122	44.1	34.8–53.4
Squamous cell carcinoma	99	52.5	39.4–65.6
Adenosquamous	5	23.1	18.4–27.8
Large cell carcinoma	7	26.7	14.4–39
Stage				0.004
IA1	22	94.4	40.6–148.2
IA2	32	62.8	42.3–83.3
IA3	9	38	26.4–49.7
IB	67	53.7	44.8–62.6
IIA	21	62.8	56.9–68.7
IIB	51	33.9	17.9–49.9
IIIA	27	16.1	5.6–26.6
IIIB	9	10.7	8.6–12.7
R status				0.001
R0	218	51.7	43.1–60.3
R1/R2	15	13.6	0.3–26.8

CKD: chronic kidney disease; COPD: chronic obstructive pulmonary disease; NSCLC: non-small-cell lung cancer; PS: performance status; VATS: video-assisted thoracic surgery. Marginally significant or significant values are highlighted in boldface.

Table 5:

Multivariable analysis in patients with primary lung cancer (n = 233)

Variable	HR	CI 95%	P-value
Gender	1.305	0.942–1.807	0.11
Surgical approach	1.098	0.787–1.531	0.58
Staging	1.153	1.055–1.260	0.002
R status	2.558	1.451–4.510	0.001

CI: confidence interval; HR: hazard ratio. Marginally significant or significant values are highlighted in boldface.

DISCUSSION

Octogenarians are a growing part of the Western population; therefore, despite their greater comorbidities and frailty, they will have to be considered for surgical resection. Currently, the rate of cancer treatment, including surgical resection, falls as the population ages [7], with a reported resection rate for octogenarians as low as 6% in one cancer registry [8]. Teams treating lung cancer require data on the risks and benefits of surgical resection in this challenging group of patients to inform decision making. With few exceptions [9], case series published on this subject tend to be small [10–12] and, although registry data are more plentiful, they tend to be less detailed and can suffer from inconsistent collection or incomplete entry [8, 13–15]. We present the short- and long-term outcomes of one of the largest, contemporary, single-institution series of octogenarians undergoing pulmonary resection.

Patient population

Our patients were carefully selected; hence only 32 had a poor Eastern Cooperative Oncology Group PS (2 or 3), but the rate of medical comorbidities was high, which reflects the population from which we take our referrals. All but 62 patients (24.1%) had at least 1 significant comorbidity, which shows a limitation of using PS alone as a metric in an older surgical population. Hino et al. [16] reported a similar rate of comorbidity in their series whereas Dominguez-Ventura et al. [9] did not report an overall morbidity rate but did report a cardiovascular comorbidity rate of 33.5% and poor lung function in 19.3% of their series of octogenarians. Our reported results for the subgroup of 49 patients aged ≥85 years represent a category of results for which there is a paucity of data reported in the literature.

Complications

The postoperative complication rate in our study was 43.6%, increasing with age from 42% in patients aged <85 years to 51% in patients aged ≥85 years; the difference was not significant, as was also shown by Ichinokawa et al. [17]. There is a large range of complication rates reported in the literature; ours was higher than the 8.4% reported by Okami et al. [15], the 16% reported by Cohen et al. [10], the 27.7% reported by Hino et al. [16] and the 30.4% reported by Fanucchi et al. [11]. Birim et al. [18] reported a morbidity of 57% (13% serious) in octogenarians whereas Dominguez-Ventura et al. [9] reported a rate of 48%, with the most common complications being arrhythmia (19.8%) and pneumonia (7.1%). In our study, PAL and pneumonia were the most common complications followed by AF. We have shown that a formal preoperative diagnosis of COPD was significantly associated with postoperative complications; in contrast, Dominguez-Ventura et al. [9] showed that the complication rate rose with a reduction in FEV1. This finding would be expected because COPD predisposes the patient to both pneumonia and PAL. The rate of postoperative AF after anatomical lung resection in our unit was previously reported as 11.4%, with age demonstrated to be an independent risk factor [19]. Thus, it is no surprise that the rate in this study of octogenarians was 13.6%, slightly more frequent after thoracotomy compared to VATS. Indeed, Park et al. [20] showed a correlation of the frequency of postoperative AF with increased age but not with surgical approach.

Several authors have shown a lower incidence of morbidity with a VATS approach, compared to an open procedure, both in the general thoracic surgical population [21, 22] and after pulmonary lobectomy in older patients [Cattaneo et al. [6] (28% vs 45%)]. In our study, the surgical approaches showed a very similar complication rate (43.1% VATS vs 44% open). Detillon et al. [12] showed that their 29.9% complication rate was not altered significantly by patient age but was reduced by the use of VATS and limited resection. We also found that the extent of resection correlated with postoperative complications in patients with lung cancer. Complications were significantly more common following lobar resections compared to sublobar (45.7% vs 24%), an effect that was previously reported [9, 23]. Even though Suzuki et al. [24] reported a similar rate of complications with lobectomy and segmentectomy, their study contained no wedge resections. At 7 days, our median postoperative LOS is similar to those of other published studies [25–27], but reduced compared to that of Cohen et al. [10], who reported an 11-day stay using a geriatric care protocol. We believe the extent of resection rather than the surgical approach is the main driver of complications and consequently of LOS.

Early mortality

Our overall 30- and 90-day mortality was 3.9% and 6.2%, respectively, which is similar to the published mortality of 3% and 5.7% following 303 579 lung cancer resections on patients of all ages in the United States between 2004 and 2013 [28]. Both Dominguez-Ventura et al. [9] and Detillon et al. [13] reported 30-day mortality in octogenarians of 6% in their series, although Dominguez-Ventura et al. [9] had a greater proportion of pneumonectomies than did our study, whereas Bokx et al. [8] reported a 30-day mortality of 4%.

Pages et al. [14] reported a 30-day mortality of 3.85% for 597 octogenarians undergoing VATS lobectomy and 7.9% for 2963 octogenarians undergoing open lobectomy for lung cancer, a statistically significant difference. When we considered patients aged 85 years and older, our mortality rates increased to 6.9% and 9% at 30 and 90 days, a finding not reported in most major series.

Long-term survival

The reported OS following surgery for NSCLC in octogenarians is generally good; in case series, the reported 5-year survivals range from 27% [11] through 47% [8] to 58% [10]. The 5-year survival of our patients undergoing resection of NSCLC was 40.8%. We believe that this number is secure because all patients have been followed up to death or for a minimum of 40 months using their UK NHS number. We found pathological stage and resection status to be the 2 independent determinants of survival, further evidence that the cancer rather than comorbidities is the primary determinant of these patients’ survival. The 5-year survival for stage 1 disease following resection is reported as 55.7% [15] and 69% [16]; in our case, it is slightly lower at 46.8%. Interestingly, despite their higher postoperative mortality, patients aged 85 years or older did not have a significantly reduced 5-year survival compared to younger octogenarians (36.3% vs 41.8%), suggesting that, even in patients of this age, successful treatment of their lung cancer is imperative for improved survival [17].

Limitations

Our retrospective design limits the conclusions we can draw. The influence of case selection on these results as well as the lack of data on patients who were referred but did not have surgery must be acknowledged. Although there was a significantly lower in-hospital complication rate associated with sublobar resection in our patients, the group was too small to determine if this operation would provide the satisfactory 5-year survival achieved with lobectomy. Our quoted post-discharge readmission rate is low, and we believe that the majority of readmissions were dealt with in local hospitals and only transferred back to our care when surgical complications required expert intervention. The fact that this is a report from a single institution means care must be taken in generalizing the results. However, the assessment and the operative and perioperative care of the patients were homogeneous and the data collection was robust, so the results are highly accurate.

CONCLUSION

In this large, single-institution experience of pulmonary resection in octogenarians over an 8-year period, the early mortality is acceptable. Care must be taken with those aged 85 years or more, but surgery can be safely offered as a radical treatment option. The 5-year survival achieved in octogenarians undergoing lobectomy for NSCLC indicates the high rate of postoperative complications seen in this challenging, but growing, cohort of patients. Hence, we believe that surgery for early-stage lung cancer should be offered to octogenarians adequately counselled as to the risks of the procedure.

SUPPLEMENTARY MATERIAL

Supplementary material is available at ICVTS online.

Conflict of interest: none declared.

Author contributions

Igor Saftic: Conceptualization; Data curation; Formal analysis; Investigation; Methodology; Project administration; Resources; Validation; Visualization; Writing—original draft; Writing—review & editing. Andrea Bille: Conceptualization; Data curation; Formal analysis; Methodology; Project administration; Resources; Supervision; Validation; Writing—original draft; Writing—review & editing. Nicole Asemota: Data curation; Formal analysis; Investigation; Methodology; Project administration. Loreto Berjon de la Vega: Conceptualization; Data curation; Formal analysis; Investigation; Methodology; Project administration. Tom Routledge: Conceptualization; Data curation; Resources; Supervision; Validation. Juliet King: Conceptualization; Data curation; Investigation; Methodology; Resources; Validation. Karen Harrison Phipps: Conceptualization; Data curation; Methodology; Resources; Validation. John Pilling: Conceptualization; Investigation; Project administration; Resources; Supervision; Validation; Writing—original draft; Writing—review & editing.

Reviewer information

Interactive CardioVascular and Thoracic Surgery thanks Cheng-Long Huang, Haruhisa Matsuguma and the other, anonymous reviewer(s) for their contributions to the peer review process of this article.

Abbreviations

AF

Atrial fibrillation

COPD

Chronic obstructive pulmonary disease

ECOG scale

Eastern Cooperative Oncology Group scale

LOS

Length of hospital stay

NSCLC

Non-small-cell lung cancer

OS

Overall survival

PAL

Prolonged air leak

PS

Performance status

UK

United Kingdom

VATS

Video-assisted thoracic surgery