European Journal of Cardio-thoracic Surgery 17 (2000) 550±556
www.elsevier.com/locate/ejcts
Long-term results of sleeve lobectomy for lung cancer q
FrancËois Tronc, Jocelyn GreÂgoire, Jacques Rouleau, Jean Deslauriers*
Division of Thoracic Surgery, Centre de Pneumologie de l'HoÃpital Laval, 2725 Chemin Sainte-Foy, Sainte-Foy, QueÂbec, Canada G1V 4G5
Received 27 September 1999; received in revised form 25 January 2000; accepted 22 February 2000
Abstract
Objective: Sleeve lobectomy is a lung saving procedure indicated for central tumors for which the alternative is a pneumonectomy.
Current controversies relate to the safety of the procedure and adequacy as a cancer operation. The aim of the study is to analyze long-term
survival after sleeve lobectomy, particularly in relation with nodal status and histological type. The incidence and patterns of recurrences
were reviewed. Methods: From 1972 to 1998, 184 patients (male 152, female 32) underwent sleeve resection for lung cancer. The mean age
was 60 ^ 10 years (11±78 years), and the indications for operation were a central tumor (79%), peripheral tumor with nodal involvement
(13%) and compromised pulmonary function (8%). The histological type was predominantly squamous (n ˆ 125, 68%), followed by nonsquamous (n ˆ 50, 27%) and carcinoid tumors (n ˆ 9, 5%). Resection was complete in 161 patients (87%). Results: The operative mortality
was 1.6% (n ˆ 3). Follow-up was complete for the remaining 181 patients (mean, 5.7 years; range, 1 month±26 years). The survival at 5 and
10 years of all patients was 52 and 33%, respectively. Theses rates for patients with N0 status (n ˆ 97) were 63 and 48%, and 48 and 27% for
those with N1 status (n ˆ 68; N0 vs. N1, P , 0:05). An 8% survival rate was observed with N2 status (n ˆ 19) at 5 years, with no survivors
after 7 years of follow-up. The 5 and 10 year survival was 56 and 34% for squamous carcinoma vs. 33 and 22% for non-squamous carcinoma
(P , 0:05). These rates were 58 and 38% for complete resection vs. 11 and 6% for incomplete resection at 5 and 10 years, respectively
(P , 0:05). Local recurrences occurred in 22% of cases, and the prevalence was statistically different between patients with N0 disease
(14%) and N1 disease (23%; P ˆ 0:03), but not between N1 and N2 disease (42%; P ˆ 0:2). When local and distant recurrence were pooled
together, the differences were highly signi®cant between N0 (22%) and N1 (41%) disease (P ˆ 0:007), and between N0 and N2 (63%)
disease (P ˆ 0:0002), but not between N1 and N2 disease (P ˆ 0:09). Conclusion: Sleeve lobectomy is a safe and effective therapy for
patients with resectable lung cancer. The presence of N1 and N2 disease, or of non-squamous carcinoma signi®cantly worsen the prognosis.
q 2000 Elsevier Science B.V. All rights reserved.
Keywords: Sleeve lobectomy; Bronchoplastic resection; Lung neoplasms; Survival; Recurrence
1. Introduction
Bronchoplastic procedures were originally designed for
cancer patients unable to tolerate pneumonectomy, or for
the treatment of uncommon benign endobronchial lesions,
such as adenomas or strictures. Indeed, the ®rst case of
sleeve lobectomy reported by Price Thomas in 1947 [1]
was carried out for a carcinoid tumor located in the right
main bronchus. In 1952, Allison [2] performed the ®rst
bronchoplastic procedure for lung cancer, and since then,
many reports [3±7] have suggested that sleeve resection
should be done more electively because it accomplishes
tumor and nodal clearance similar to that of pneumonectomy. In addition, the survival after sleeve resection is at
least equal to that reported after pneumonectomy, with the
q
Presented at the 13th Annual Meeting of the European Association for
Cardio-thoracic Surgery, Glasgow, Scotland, UK, September 5±8, 1999.
* Corresponding author. Tel.: 11-418-656-4747; fax: 11-418-656-4762.
added advantage of an improved quality of life and lower
incidence of long-term detrimental respiratory sequelae.
In the current literature, the 5 and 10 year survival ®gures
after sleeve lobectomy done for lung cancer vary quite
widely dependent on cancer stage and nodal status [3,6,8].
Because of these con¯icting results, we reviewed our
experience with these procedures done over a 25 year interval. The purposes of this review were to look at operative
mortality and morbidity as they relate to modern thoracic
surgery, to determine the adequacy of cancer clearance in
relation with the incidence of local recurrences and ultimately to analyze long-term survival ®gures.
2. Patients and methods
From January 1972 to July 1998, 184 patients underwent
sleeve resection as a primary treatment for bronchogenic
tumors. There were 152 men and 32 women, and the
1010-7940/00/$ - see front matter q 2000 Elsevier Science B.V. All rights reserved.
PII: S10 10-7940(00)0040 5-X
F. Tronc et al. / European Journal of Cardio-thoracic Surgery 17 (2000) 550±556
mean age at the time of operation was 60 ^ 10 years (range,
11±78 years). Bronchoplastic procedures were usually done
by choice in patients who would have been able to tolerate
pneumonectomy, but in whom the lesion could be completely resected by a lesser operative procedure (n ˆ 145). In
this group, the indications for operation included patients
with endobronchial tumors originating from a lobar
bronchus and extending into the main bronchus (n ˆ 71),
extraluminal extension to the outer wall of the main
bronchus undetected by bronchoscopic examination
(n ˆ 25), unsuspected involvement of the bronchial margins
diagnosed by frozen section (n ˆ 49), and peripheral tumors
with nodal involvement (n ˆ 24). In all of these individuals,
the tumor could not have been removed by conventional
lobectomy because the line of resection would have either
transected the tumor or provided an inadequate margin. In
15 additional patients with severe respiratory impairment,
the bronchoplasty was done as a compromise to pneumonectomy.
Mediastinoscopy, with random node sampling at three
different levels, was done in nearly every case (160/184).
In 24 patients, mediastinoscopy was not done because the
diagnosis was that of a carcinoid tumor, or simply because
the diagnosis of lung cancer had not been con®rmed preoperatively. Induction chemotherapy was used in three
patients with N2 disease documented at mediastinoscopy.
2.1. Operative procedure
Although the initial ®nding that identi®ed a possible
candidate for sleeve resection was the bronchoscopic
appearance of the tumor extending from a lobar ori®ce to
the adjacent main bronchus, the ®nal decision was always
made at the time of thoracotomy. The procedures were
considered complete when all gross carcinoma were
removed, and when both resection margins and the highest
node were free of the tumor.
The majority of sleeve resections were done for tumors
involving the origin of either upper lobe (n ˆ 152, 83%;
Table 1). In 17 patients, the middle lobe had to be taken
in continuity with the right upper lobe because of tumor
Table 1
Types of sleeve resections
Type of sleeve resection
Right lung
Upper lobe
Upper and middle lobe
Middle and lower lobe
Middle lobe
Main bronchus
Left lung
Upper lobe
Lower lobe
Main bronchus
551
growth across the ®ssure, or because of nodal disease
around the bronchus intermedius. Nine patients had a
lower lobe sleeve resection, and in four patients, a small
tumor located in the main bronchus could be resected without sacri®cing any of the distal lung. Four patients required
a concomitant sleeve resection of the pulmonary artery
(double sleeve resection). Frozen section analyses of
lymph nodes were routinely used during operation, and if
a patient was found to have surgical N1 or N2 disease,
radical lymphadenectomy was carried out.
Resection was complete in 161 patients, and incomplete
in 23 patients usually because of diseased bronchial margins
(n ˆ 12).
2.2. Follow-up
Patients were observed from the date of operation to the
time of death, or end of observation (July 1998), whichever
came ®rst. No patients were lost to follow-up. The mean
follow-up time was 2071 days and the maximum follow-up
time was 9559 days.
A local recurrence was de®ned as tumor growth within
the ipsilateral hemithorax or mediastinum, or both. For
patients who died, the cause of death and the site of ®rst
recurrence were determined from clinical records, death
certi®cates, or from information provided by family physicians or by relatives of the patient. Each death was classi®ed
as being due to: (1), lung cancer, including second primary
lesions and operative deaths; (2), causes unrelated to lung
cancer; or (3), unknown causes.
The sites of ®rst recurrences were classi®ed as being: (1),
local; (2), distant; (3), local and distant; (4), unknown sites;
or (5), second primary de®ned according to the criteria of
Martini and Melamed [9] as a tumor of different histological
type, or a tumor of similar histological type if it occurred
more than 2 years after the ®rst operation or if its origin
could be traced to a carcinoma in situ.
Operative mortality was de®ned as a death occurring
within 30 days of the operation, or a death directly related
to the procedure even if it occurred more than 30 days after
the operation.
2.3. Statistical analysis
Number of patients
113
17
3
1
1
39
6
4
The results are presented as means ^ SD for continuous
variables, and as percentages for categorical data. The Chisquare test was used to compare these proportions. A Pvalue of #0.05 was considered signi®cant. The life-table
method for longitudinal data was obtained using the
Kaplan±Meier method and all causes of death were
included. Differences between survival curves were
obtained by computing the con®dence limits on the lifetable values. The data were analyzed using the SAS statistical package (SAS Institute, Inc, Corey, NC).
552
F. Tronc et al. / European Journal of Cardio-thoracic Surgery 17 (2000) 550±556
Table 2
Pathology and stage of 184 patients included in the comparison of survival by nodal status
Characteristic
Nodal status
Histological type
Squamous carcinoma
Non-squamous carcinoma
Carcinoid
T status
T1
T2
T3±T4
N0 (n ˆ 97)
N1 (n ˆ 68)
N2 (n ˆ 19)
Total (n ˆ 184)
69
19
9
49
19
0
7
12
0
125
50
9
6
77
14
6
53
9
3
12
4
15
142
27
3. Results
3.1. Histopathology and stage of disease
There were 125 patients (68%) with squamous cell carcinoma (Table 2), 50 patients with non-squamous carcinomas
(adenocarcinoma, 19; large cell carcinoma, 20; mixed cell
type, 11), and nine patients with typical carcinoid tumors.
All tumors were pathologically staged according to the
most recent tumor node metastasis (TNM) classi®cation
[10]. Ninety-seven patients (53%) had N0 disease, 68
(37%) had N1 disease and 19 had N2 disease (10%). The
stage distribution shows that the majority of patients are
either in stage 1 (45%, 82/184) or in stage 2 (39%, 72/
184) categories.
3.2. Operative morbidity and mortality
Three deaths occurred after operation, resulting in a
hospital mortality of 1.6%. Two of the three deaths were
the result of infectious complications in the reimplanted
lung, while the third operative death was due to a pulmonary
embolus.
Non-fatal complications were seen in 26 additional
patients, and these are listed in Table 3. Late anastomotic
strictures developed in four patients, all after right upper
lobe sleeve resection. These complications were treated by
the endoscopic removal of granulomas in two patients and
dilatation of the stricture in one patient. Completion pneu-
monectomy was required in the remaining patient. Including
the two patients with early bronchopleural ®stula, both
managed conservatively, six patients (3.2%) had complications related to the anastomosis.
3.3. Survival
The observed survival rates for the entire group and each
subset of the N descriptor are shown in Figs. 1 and 2. The
actuarial survival rate for all 184 patients was 52 ^ 4% after
5 years, and 34 ^ 4% after 10 years. The median survival
time was approximately 5 years.
For patients with N0 status, the 5 and 10 year survival was
63 ^ 5 and 43 ^ 6%, while for patients with N1 status, these
survivals were 48 ^ 6 and 27 ^ 6%, respectively. Among
the 19 patients with N2 status, none survived more than 7
years after operation, and the 5 year survival was only 8%.
The differences in survival for N1 or N2 patients compared
to N0 patients were signi®cant. These differences become
signi®cantly different after 6 (N1) and 2 years (N2), respectively, using the Bonferroni correction.
For squamous carcinomas, the 5 and 10 year survival was
56 ^ 5 and 34 ^ 5%, and 33 ^ 8 and 22 ^ 7% for nonsquamous carcinomas (Fig. 3). The curves differ signi®cantly (P # 0:05) after 2 years of follow-up. All patients
with carcinoid tumors survived 10 or more years after
operation. No signi®cant difference in survival was
observed between patients with right- or left-sided sleeve
Table 3
Major operative complications after sleeve resection
Complication
Early (n ˆ 26)
Pneumonia
Empyema
Respiratory failure
Bronchopleural ®stula
Others
Late (n ˆ 4)
Granulation at suture line
Bronchial stricture
No of patients
9
4
4
2
7
2
2
Fig. 1. Life-table analysis showing the percentage of all patients remaining
alive after sleeve resection of bronchogenic carcinoma.
F. Tronc et al. / European Journal of Cardio-thoracic Surgery 17 (2000) 550±556
Fig. 2. Life-table analysis by nodal status showing the percentage of all
patients remaining alive after sleeve resection of bronchogenic carcinoma.
Signi®cant differences in survival were noted between N0, N1 and N2
patients.
resections (Fig. 4). Only 11% of patients with incomplete
resection survived 5 years.
3.4. Sites of recurrence and cause of death
Currently, 73 of the 184 patients are alive. Of the 111
patients who have died, recurrent lung cancer was the cause
of death in 61, and second primary cancers were the cause of
death in 18 patients.
Local recurrence occurred in 22% of patients, and the
distribution was signi®cantly different between patients
with N0 (14%) and patients with N1 disease (23%)
(P ˆ 0:03; Table 4). In comparison, eight of 19 patients
(42%) with N2 disease had a local recurrence (N0 vs. N2,
P ˆ 0:005). When local and distant recurrence are pooled
together, the recurrence rates were 22, 41 and 63% for N0,
N1 and N2 groups, respectively, (N0 vs. N1, P ˆ 0:007; N0
vs. N2, P ˆ 0:0002; N1 vs. N2, P ˆ 0:09). The ®rst site of
recurrence for the entire group was mostly locoregional
(locoregional, 41/61, 67%; distant, 20/61, 33%). The most
common site of distant metastases was the brain (n ˆ 9).
After complete resection, 26 patients (16%) had a local
recurrence, while, in comparison, 15 patients (65%) had a
local recurrence (P , 0:05) after an incomplete resection.
Fig. 3. Life-table analysis by histological type showing the percentage of
all patients remaining alive after sleeve resection of bronchogenic carcinoma. Signi®cant differences were noted between the two groups.
553
Fig. 4. Life-table analysis by side of operation showing the percentage of
all patients remaining alive after sleeve resection of bronchogenic carcinoma.
4. Discussion
Sleeve lobectomy can be used as an alternative to pneumonectomy in carefully selected patients with bronchogenic
carcinoma. It preserves the functional lung, and it has been
previously shown that the reimplanted lobe(s) contribute
signi®cantly to postoperative lung function [5,7]. In addition, it may allow a second procedure to be done in patients
who develop a second primary, the incidence of which has
increased in recent years [11].
Sleeve resection is a technically more demanding procedure than standard lobectomy, and the bronchial anastomosis can give rise to speci®c complications not seen after
conventional lobectomies. This is one of the reasons why
bronchoplastic operations must have a low operative
morbidity and mortality, as well as provide adequate survival in order to be recognized as valid cancer operations.
In a collective review of 1915 patients submitted to
bronchoplastic procedures over a period of 12 years
(1980±1992), Tedder and colleagues [12] reported a 30
day mortality of 5.5%, with a range of 0±11% [7,13,14].
The causes of death were mainly respiratory or cardiac
related events. In the current series, the operative mortality
of 1.6% is similar to that of more recent reports [5,8,13], and
compares favorably with the 6% 30 day operative mortality
reported by the Lung Cancer Study Group after pneumonectomy [15].
Persistent atelectasis is one of the most common morbidities reported after bronchoplasty [12,16], as it relates to the
interruption of the ciliary epithelium and lymphatics, partial
denervation of the reimplanted lobe(s), or anastomotic
edema. In this series, atelectasis occurred in 10% of
patients, but in the majority of cases, the problem was
minor and could be solved by the bronchoscopic removal
of the retained mucus.
In the review of Tedder and colleagues [12], anastomotic
complications included bronchopleural ®stula in 3% of
patients and late stricture in 4.8% of cases. In the present
series, these problems occurred in 1.1 and 2.2% of patients,
respectively. This favorable outcome is likely to be the
554
F. Tronc et al. / European Journal of Cardio-thoracic Surgery 17 (2000) 550±556
Table 4
Site of ®rst recurrence in 184 patients included in the comparison of survival by nodal status a
Site of ®rst recurrence
Nodal status
Locoregional recurrences
Distant recurrences
a
N0 (n ˆ 97)
N1 (n ˆ 68)
N2 (n ˆ 19)
Total (n ˆ 184)
14 (14)
7 (7)
19 (23)
9 (13)
8 (42)
4 (21)
41 (22)
20 (11)
Figures in parentheses represent percentage values.
result of careful construction of the anastomosis, and the use
of resorbable suture material.
In their review, Tedder and colleagues [12] report overall
5 year survival ®gures of 40% when sleeve lobectomy is
done for bronchogenic carcinoma. In the absence of nodal
involvement (N0 status), these rates can reach up to 60%.
Our overall survival ®gures of 52 and 33% at 5 and 10 years
are similar to ®gures recently reported by other institutions.
In 1999, Suen and associates [17] reported an overall 37% 5
year survival in 58 patients, while this was 42% in the
Gaissert and colleagues series [5] of 72 patients, and 46%
in the series of Van Schil et al. [6] (Table 5). Older series
include those of Watanabe and associates [4] (5 year survival, 45%; n ˆ 79), Maggi and associates [16] (5 year survival, 40.8%; n ˆ 69), and Naruke and associates [13] (41%;
n ˆ 91). In general, survival ®gures after sleeve resection
are comparable with those seen after conventional lobectomy, but much better than those reported after pneumonectomy [18].
The use of sleeve resection when the carcinoma has
spread to bronchopulmonary or hilar nodes (N1) is still an
area of controversy, even when both tumor and nodes can be
completely resected. Studies of pulmonary lymphatic drainage show that upper lobe tumors seldom metastasize to
nodes located below the oblique ®ssures, but rather have
relatively constant drainage to ipsilateral para-tracheal
nodes [19]. Thus, if adequate tumor clearance can be
achieved at the margins of bronchial excision, sleeve lobectomy of upper lobes should be an adequate cancer operation.
In 1983, Firmin and associates [8] reported a 5 and 10 year
survival of 17 and 10% when the tumor had metastasized to
hilar nodes, but more favorable outcomes were reported by
Faber et al. [3] (23% for N1 disease at 5 years), and Gaissert
and associates [5] (38% at 5 years). The results in the
present series are concordant with those series, and that of
Van Schil [6], who reported a 5 and 10 year survival of 31
and 10%, respectively, for patients with N1 disease. For
patients with N2 disease, the survival is signi®cantly
worse, and ranges from 0 (20) to 33% (13) at 5 years, and
from 0 (21) to 13% (6) at 10 years. This is consistent with
nearly all survival analyses, which show that N2 disease is
highly predictive of distant recurrences after resection.
In contrast to our previous report [21], we now ®nd a
signi®cant difference in the long-term survival between
patients with N0 and N1 disease. In their series of 145
patients, Van Schil and colleagues [6] showed by multivariate analysis that nodal status was the most signi®cant factor
related to long-term survival, both N1 and N2 disease
having a de®nite negative impact. Our results are similar
to these, although patients with N1 status can still expect a
survival of equal or better to that reported after pneumonectomy for similar stage tumors [22,23]. For patients with
N2 disease, the difference in survival at 5 years between our
series (8%) and that of Van Schil [6] (31%) might be
explained by the unsettled controversy in determining
whether central nodes, especially in the right lung, are to
be included in station 10 or 4 of the nodal maps, and whether
these nodes are hilar or mediastinal. If one groups the
patients with pN1 and pN2 status together, the survival at
5 and 10 years after operation is quite similar between the
current series and that of other series [5,6,13].
The issue of local tumor recurrence after sleeve resection
has been extensively investigated. In their review, Tedder
[12] and colleagues reported local recurrence rates of
12.5%, ranging between 5 and 51% [14,16], while Faber
and colleagues [3] observed local recurrence rates of 9%
Table 5
Survival rates after bronchoplastic procedures by lymph node status
First author (reference)
Firmin [8]
Voyt-Moykopf [20]
Naruke [13]
Van Schil [6]
Gaissert [5]
Present series
Year
1983
1986
1989
1996
1996
1999
Number of patients
90
248
111
145
72
184
5 year survival (%)
10 year survival (%)
N0
N1
N2
N0
N1
N2
71
37
50
62
57
63
17
30
46
31
38
48
±
0
33
31
43
8
49
±
±
51
±
43
10
±
±
10
±
27
±
±
±
±
±
0
F. Tronc et al. / European Journal of Cardio-thoracic Surgery 17 (2000) 550±556
in 101 patients, and Gaissert and associates [5] had a local
recurrence rate of 14% in 69 patients. In many of these
papers, local recurrences are not clearly de®ned, some
authors considering only tumor at the suture line as a local
recurrence, while others include nodal disease or disease at
pulmonary sites which are remote from the initial resection.
For the purpose of clarity, we included all patients with
recurrent disease within the ipsilateral hemithorax as having
a local recurrence. By using this de®nition, our results correlate with the experience of other groups [6,24]. Despite
meticulous operative evaluation through the liberal use of
frozen sections, there were nine instances of truly anastomotic recurrences. The majority of these occurred in
patients with incomplete resections based on positive resection margins, and it is of interest to note that some of these
individuals had negative frozen sections, but turned out to
have positive margins at the ®nal microscopic analysis
obtained 3 days after operation. This would support the
reluctance that some surgeons have to completely rely on
frozen sections for ®nal decision making.
Overall, the incidence of cancer recurrence is similar or
even less after sleeve resection to what is reported after
conventional resection for stage 1 and 2 carcinomas of the
lung [22,25]. However, the patterns of failure are slightly
different. After complete resection of early stage lung
cancer, ®rst recurrences tend to be distant for patients
with N0 status, and local for patients with N1 status [25].
Immerman and colleagues [25] reported that after the
complete resection of non-small cell stage 1 and 2 lung
cancers, the site of ®rst relapse was local in 18% of patients
and distant in 26%. In our series, the site of ®rst relapse is
more often locoregional, even for patients with N0 status.
This pattern of recurrence in patients with N0 or N1 disease
is different from that reported by Van Schil and colleagues
[6] where distant metastases were the main cause of death.
5. Conclusion
Sleeve lobectomy is a valuable alternative to pneumonectomy, with excellent short- and long-term survival
results. The operative mortality is low and the radical nature
of lung cancer treatment is not jeopardized. Bronchoplasty
with node dissection achieves the same surgical margins as
pneumonectomy and the long-term results are strongly
in¯uenced by nodal status.
References
[1] Price-Thomas C. Conservative resection of the bronchial tree. J R
Coll Surg Edinburgh 1955;1:169±186.
[2] Allison PR. Course of thoracic surgery in Groningen. Ann R Coll
Surg 1954;25:20±22 Quoted by Jones PW.
[3] Faber LP, Jensik RJ, Kittle CF. Results of sleeve lobectomy for
bronchogenic carcinoma in 101 patients. Ann Thorac Surg
1984;37:279±285.
[4] Watanabe Y, Shimizu J, Oda M, Hayashi Y, Watanabe S, Yazaki U,
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
555
Iwa T. Results in 104 patients undergoing bronchoplastic procedures
for bronchial lesions. Ann Thorac Surg 1990;50:607±614.
Gaissert HA, Mathisen DJ, Moncure AC, Hilgenberg AD, Grillo HC,
Wain JC. Survival and function after sleeve lobectomy for lung
cancer. J Thorac Cardiovasc Surg 1996;111:948±953.
Van Schil PE, Brutel de la Riviere A, Knaepen PJ, Van Swieten HA,
Reher SW, Goossens DJ, Vanderschueren RG, Van den Bosch JM.
Long-term survival after bronchial sleeve resection: univariate and
multivariate analyses. Ann Thorac Surg 1996;61:1087±1091.
Deslauriers J, Gaulin P, Beaulieu M, Piraux M, Bernier R, Cormier Y.
Long-term clinical and functional results of sleeve lobectomy
for primary lung cancer. J Thorac Cardiovasc Surg 1986;92:871±
879.
Firmin RK, Azariades M, Lennox SC, Lincoln JCR, Paneth M. Sleeve
lobectomy (lobectomy and bronchoplasty) for bronchial carcinoma.
Ann Thorac Surg 1983;35:442±449.
Martini M, Melamed MR. Multiple primary lung cancers. J Thorac
Cardiovasc Surg 1975;70:606±612.
Mountain CF. Revisions in the international system for staging lung
cancer. Chest 1997;111:1710±1717.
Jensik RJ, Faber LP, Kittle CF, Meng RL. Survival following resection for second primary bronchogenic carcinoma. J Thorac Cardiovasc Surg 1981;82:658±668.
Tedder M, Anstadt MP, Tedder SD, Lowe JE. Current morbidity,
mortality and survival after bronchoplastic procedures for malignancy. Ann Thorac Surg 1992;54:387±391.
Naruke T. Bronchoplastic and bronchovascular procedures of the
tracheobronchial tree in the management of primary lung cancer.
Chest 1989;96:53S±56S.
Weisel RD, Cooper JD, Delarue NC, Theman TE, Todd TRJ, Pearson
FG. Sleeve lobectomy for carcinoma of the lung. J Thorac Cardiovasc
Surg 1979;78:839±849.
Ginsberg RJ, Hill LD, Eagan RT, Thomas P, Mountain CF, Deslauriers J, Fry WA, Butz RO, Goldberg M, Waters PF. Modern 30 day
operative mortality for surgical resections in lung cancer. J Thorac
Cardiovasc Surg 1983;86:654±658.
Maggi G, Casadio C, Pischedda F, Cianci R, Ruf®ni E, Filosso P.
Bronchoplastic and angioplastic techniques in the treatment of
bronchogenic carcinoma. Ann Thorac Surg 1993;55:1501±1507.
Suen HC, Meyers BF, Guthrie T, Phol MS, Sundaresan S, Roper CL,
Cooper JD, Patterson GA. Favorable results after sleeve lobectomy or
bronchoplasty for bronchial malignancies. Ann Thorac Surg
1999;67:1557±1562.
Shields TW. Carcinoma of the lung. In: Shields TW, editor. General
thoracic surgery, Philadelphia, PA: Lea & Febiger, 1972. pp. 837.
Nohl HC. An investigation into the lymphatic and vascular spread of
carcinoma of the bronchus. Thorax 1956;11:172±185.
Vogt-Moykopf I, Fritz T, Meyer G, Bulzerbruck H, Daskos G.
Bronchoplastic and angioplastic operation in bronchial carcinoma:
long-term results of a retrospective analysis from 1973 to 1983. Int
Surg 1986;71:211±220.
Mehran RJ, Deslauriers J, Piraux M, Beaulieu M, Guimont C, Brisson
J. Survival related to nodal status after sleeve resection for lung
cancer. J Thorac Cardiovasc Surg 1994;107:576±583.
Iascone C, DeMeester TR, Albertacci M, Little AG, Golomb HM.
Local recurrence of resectable non-oat cell carcinoma of the lung: a
warning against conservative treatment for N0 and N1 disease.
Cancer 1986;57:471±476.
The Ludwig Lung Cancer Study Group. Patterns of failure in patients
with resected stage 1 and 2 non-small-cell carcinoma of the lung. Ann
Surg 1987;205:67-71.
Muller C, Schinkel C, Hoffmann H, Dienemann H, Schildberg FW.
Bronchoplastic resection for non-small-cell lung cancer. Thorac
Cardiovasc Surg 1996;44:248±251.
Immerman SC, Vanecko RM, Fry WA, Head LR, Shields TW. Site of
recurrence in patients with stages 1 and 2 carcinoma of the lung
resected for cure. Ann Thorac Surg 1981;32:23±27.
556
F. Tronc et al. / European Journal of Cardio-thoracic Surgery 17 (2000) 550±556
Appendix A. Conference discussion
Dr P. Van Schil (Antwerp, Belgium): I have three questions. In contrast
to earlier reports from our institution, you presently found a signi®cant
difference between N0 and N1 disease. Is this due to longer follow-up
time, and did you look at the causes of death in relation to nodal involvement?
Secondly, do you advocate adjuvant therapy in N1 or N2 disease after
sleeve resection?
Thirdly, the group from Paris, from Professors Levasseur and Icard,
recently proposed to initiate a prospective randomized trial between sleeve
lobectomy and pneumonectomy in N1 disease for patients who can tolerate
a pneumonectomy. What is your opinion about this kind of study?
Dr Tronc: First of all, yes, we found a difference between N0 and N1
with longer follow-up, exactly. For the second question, we do radiotherapy
for all patients with N2 status studied at mediastinoscopy or thoracotomy.
For N1, this is not always the case. For the third question, if I understand it,
pneumonectomy for patients who are not able to tolerate a sleeve lobectomy?
Dr Van Schil: Yes. Due to the controversial results of N1 disease, they
propose to initiate, a prospective study with patients randomized between
sleeve lobectomy and pneumonectomy when N1 disease is found during the
operation.
Dr Tronc: I think the results of sleeve lobectomy are good enough and
in this series perhaps not needed.
Dr K. Al Kattan (Riyadh, Saudi Arabia): I noted there is a big difference
in the 10 and 5 year survival, and we always deal with lung cancer so that
after 5 years the cause of death related to cancer becomes much less. Have
you analyzed the cause of death in the patients who had a 10 year survival?
Dr Tronc: Most of the patients died from cancer in this series, of course,
and my opinion is, what is cause of death after 5 or 10 years? We don't
exactly analyze the question.