Outcomes of segmentectomy and wedge resection for pulmonary metastases from colorectal cancer†

Abstract

INTRODUCTION

Pulmonary metastasectomy is generally considered for selected patients with pulmonary-limited metastases from colorectal cancers [1]. However, although there is a prospective study of demography and clinical characteristics [2], there have not been any randomized studies validating the survival benefit of pulmonary metastasectomy [3–6]. The 5-year overall survival rates have been reported to range from 27% to 68% [7–9]. Prognostic factors include the site of the primary tumour, R0 resection, disease-free interval (DFI), serum carcinoembryonic antigen (CEA) level, metastasis number, metastasis size and lymph node metastasis [8, 9].

Wedge resection is considered to be an adequate procedure for pulmonary metastases, and it is the most common procedure used for excision because of the following reasons: (i) as pulmonary metastases from colorectal cancers can occur at multiple locations metachronously, preservation of pulmonary function is required for a chance to undergo repeated pulmonary resections [10, 11]; and (ii) the role of lymph node dissection in metastasectomy is unknown [12–14]. However, a major problem with wedge resection is the risk of local recurrence, especially recurrence at the resection margin [15, 16].

Segmentectomy is generally indicated for multiple, centrally located or metastatic tumours too large for successful wedge resection. Although lobectomy reduces the pulmonary function, segmentectomy can obtain an adequate surgical margin without deteriorating pulmonary function. To date, it is unclear which procedure, wedge resection or segmentectomy, provides better outcomes after pulmonary metastasectomy. Therefore, there is a need to compare the surgical outcomes of wedge resection with those of segmentectomy. The objective of this study was to investigate (i) the survival rates of patients undergoing each type of surgical procedure for pulmonary metastases from colorectal cancer and (ii) the recurrence in patients who underwent either segmentectomy or wedge resection.

PATIENTS AND METHODS

The ethics committees of all the institutions participating in this study approved the study design and waived the need for informed consent from the patients, because the data on the patients remained anonymous.

This was a retrospective study and subset analysis based on Japanese nationwide retrospective study of resected pulmonary metastases from colorectal cancer from 46 institutions. Between January 2004 and December 2008, patients with pulmonary metastasis who underwent a first surgical procedure for pulmonary metastases were registered. Inclusion criteria of this study were as follows: first pulmonary metastasectomy from colorectal cancer, metastasectomy performed for curative intent, pathologically complete resection of pulmonary metastasis (R0) resulting in disease-free status and pathological diagnosis of pulmonary colorectal cancer metastasis.

The following parameters were collected from the participating institutions:

1. Patient’s characteristics (gender, age, Eastern Cooperative Oncology Group performance status and serum CEA level) immediately before metastasectomy;

2. Information on primary tumour (tumour site, stage, date of primary surgery, type of surgery, perioperative chemotherapy, extrathoracic lesions and disease curability);

3. Information on pulmonary metastasis (date of detecting pulmonary metastasis, number, laterality, maximum diameter, date of pulmonary metastasectomy, type of surgical procedure, extent of lymph node dissection, lymph node metastasis, postoperative complications, completeness of pulmonary resection and other treatments);

4. Outcomes (date of recurrence or last confirmation of disease-free status, sites of recurrence, date of death or last confirmed survival).

The DFI was calculated from the date of initial treatment for the primary tumour to the date of first detection of pulmonary metastasis. Recurrence sites after pulmonary metastasectomy were divided into the following three categories: intrathoracic (lung, intrathoracic lymph node and resection margin), extrathoracic and both. Resection-margin recurrence was defined as a recurrence developing in staple lines. Recurrence and the data for mortality were provided by each institution.

In total, 1237 patients who underwent metastasectomy for pulmonary metastases from colorectal cancer at the participating 46 institutions were registered. Those with inadequate information about perioperative chemotherapy (n = 119), non-R0 resection (n = 116), inadequate information about postoperative follow-up (n = 47), surgery during an ineligible period (n = 34), a history of malignancies other than colorectal cancer (n = 12), non-first metastasectomy (n = 5) or other reasons (n = 6) were excluded. After exclusion, 898 patients were eligible. For this subset analysis, patients who received preoperative chemotherapy (n = 113), patients with insufficient data (n = 19) and patients with lobectomy (n = 213) were excluded. A total of 553 patients (44.7% of the database) who underwent segmentectomy (n = 98) and wedge resection (n = 455) were selected for this exploratory analysis. There was no predetermined protocol for pulmonary metastasectomy; the surgical indications, procedures, approaches and diagnosis of recurrence were determined at the discretion of each institution.

Statistical analysis

The chi-square test was used to evaluate the association between categorical variables and surgical procedures. The unpaired t-test was used to compare continuous variables between surgical procedures. Overall and recurrence-free survival were estimated using the Kaplan–Meier method. Overall survival was measured from the date of pulmonary metastasectomy to the date of death from any cause, or censored at the date of the patient’s last hospital visit. Recurrence-free survival was measured from the date of pulmonary metastasectomy to the date of detection of recurrence. The date of the patient’s last hospital visit was a censoring time if relapse is not observed. Univariable and multivariable Cox proportional-hazards regression analysis was used to identify prognostic factors for overall and recurrence-free survival. In univariable and multivariable analysis, patients were classified according to the following categories based on median values: age ≥ 66 and <66 years, tumour size < 15 and ≥15 mm, and DFI < 18 and ≥18 months. To control potential differences in the characteristics of patients treated with these two surgical procedures, we tried the propensity score matching, but a few factors (tumour location and pulmonary functions) essential for selection of the surgical procedure were lacking, and therefore we could not use the matching method in this study. Data were analysed using JMP software, version 11.0 (SAS Institute Inc., Cary, NC, USA). A P-value of <0.05 was considered statistically significant.

RESULTS

Characteristics of patients

The median age at the time of pulmonary resection was 66 years (range, 29–89). Of the 553 patients, 314 (56.8%) were male. The primary site was the rectum in 297 (53.7%), colon in 251 (45.4%) and both sites in 5 (0.9%) patients. The median DFI was 18 months (range, 0–190 months). The performance status of 509 (92.0%) patients was 0. Unilateral surgery was performed for 486 (87.9%) and bilateral for 67 (12.1%) patients. Based on the CEA reference values of each patient’s institution, a high preoperative serum CEA level was found in 178 (32.2%) patients. The median number of resected metastatic tumours was 1 (range, 1–8). The median size of resected metastases was 15 mm (range, 5–51 mm). There were extrathoracic metastatic lesions in 139 (25.1%) patients. Adjuvant chemotherapy was administered to 265 (47.9%) patients. Table 1 summarizes the characteristics of patients according to surgical procedure. Although the tumour size of segmentectomy cases was significantly larger than that of wedge resection cases, other characteristics of patients undergoing either procedure were not significantly different.

Postoperative complications

Postoperative complications developed in 36 (6.5%) of 553 patients, and the frequency of complications was significantly higher for those who underwent segmentectomy (Table 2, P = 0.001). Prolonged air leak developed at a significantly higher rate in those undergoing segmentectomy (P = 0.048). There were no deaths within 30 days of surgery, and 2 (0.4%) deaths occurred within 90 days.

Survival and recurrence

The Kaplan–Meier estimate of the median length of follow-up was 5.4 years (95% confidence interval, 5.2–5.6 years). One-hundred and sixty-three patients (30.2%) died. The 5-year overall survival was 70.5%. Figure 1 shows the Kaplan–Meier curve for overall survival according to the surgical procedure; the 5-year overall survival of patients undergoing segmentectomy was 80.1%, and that of patients undergoing wedge resection was 68.5%. Multivariable analysis revealed age, DFI, CEA level and extrathoracic lesions as the prognostic factors for overall survival (Table 3).

Figure 1:

Overall survival according to the surgical procedure.

Open in new tabDownload slide

Table 3:

Open in new tab

Overall survival by prognostic factor^a

.	.	Univariable analysis .			Multivariable analysis .
.		HR .	95% CI .	P-value .	HR .	95% CI .	P-value .
Age (years)	<66/≥66	0.59	0.43–0.80	<0.001	0.62	0.45–0,86	0.004
Gender	M/F	1.28	0.94–1.75	0.123	1.25	0.91–1.73	0.170
DFI (months)	≥18/<18	0.61	0.45–0.83	0.002	0.59	0.43–0.81	0.001
PS	0/1, 2	0.63	0.39–1.08	0.092	0.88	0.53–1.53	0.627
Surgical side	Unilateral/Bilateral	0.61	0.42–0.93	0.021	0.78	0.46–1.32	0.355
CEA level	Normal/High	0.39	0.29–0.53	<0.001	0.43	0.31–0.59	<0.001
Tumour numbers	Solitary/Multiple	0.61	0.45–0.85	0.004	0.76	0.51–1.19	0.229
Extrathoracic lesions	−/+	0.68	0.49–0.96	0.028	0.64	0.46–0.91	0.014
Max tumour size (mm)	<15/≥15	0.79	0.58–1.07	0.132	0.78	0.57–1.08	0.136
Surgery	Segmentecomy/Wedge	0.56	0.33–0.90	0.014	0.65	0.38–1.05	0.080
Adjuvant chemoTx	+/−	0.92	0.68–1.24	0.057	0.85	0.62–1.17	0.316

.	.	Univariable analysis .			Multivariable analysis .
.		HR .	95% CI .	P-value .	HR .	95% CI .	P-value .
Age (years)	<66/≥66	0.59	0.43–0.80	<0.001	0.62	0.45–0,86	0.004
Gender	M/F	1.28	0.94–1.75	0.123	1.25	0.91–1.73	0.170
DFI (months)	≥18/<18	0.61	0.45–0.83	0.002	0.59	0.43–0.81	0.001
PS	0/1, 2	0.63	0.39–1.08	0.092	0.88	0.53–1.53	0.627
Surgical side	Unilateral/Bilateral	0.61	0.42–0.93	0.021	0.78	0.46–1.32	0.355
CEA level	Normal/High	0.39	0.29–0.53	<0.001	0.43	0.31–0.59	<0.001
Tumour numbers	Solitary/Multiple	0.61	0.45–0.85	0.004	0.76	0.51–1.19	0.229
Extrathoracic lesions	−/+	0.68	0.49–0.96	0.028	0.64	0.46–0.91	0.014
Max tumour size (mm)	<15/≥15	0.79	0.58–1.07	0.132	0.78	0.57–1.08	0.136
Surgery	Segmentecomy/Wedge	0.56	0.33–0.90	0.014	0.65	0.38–1.05	0.080
Adjuvant chemoTx	+/−	0.92	0.68–1.24	0.057	0.85	0.62–1.17	0.316

CEA: carcinoembryonic antigen; CI: confidence interval; DFI: disease-free interval; HR: hazard ratio; PS: performance status.

^aCox proportional hazards model.

Table 3:

Open in new tab

Overall survival by prognostic factor^a

.	.	Univariable analysis .			Multivariable analysis .
.		HR .	95% CI .	P-value .	HR .	95% CI .	P-value .
Age (years)	<66/≥66	0.59	0.43–0.80	<0.001	0.62	0.45–0,86	0.004
Gender	M/F	1.28	0.94–1.75	0.123	1.25	0.91–1.73	0.170
DFI (months)	≥18/<18	0.61	0.45–0.83	0.002	0.59	0.43–0.81	0.001
PS	0/1, 2	0.63	0.39–1.08	0.092	0.88	0.53–1.53	0.627
Surgical side	Unilateral/Bilateral	0.61	0.42–0.93	0.021	0.78	0.46–1.32	0.355
CEA level	Normal/High	0.39	0.29–0.53	<0.001	0.43	0.31–0.59	<0.001
Tumour numbers	Solitary/Multiple	0.61	0.45–0.85	0.004	0.76	0.51–1.19	0.229
Extrathoracic lesions	−/+	0.68	0.49–0.96	0.028	0.64	0.46–0.91	0.014
Max tumour size (mm)	<15/≥15	0.79	0.58–1.07	0.132	0.78	0.57–1.08	0.136
Surgery	Segmentecomy/Wedge	0.56	0.33–0.90	0.014	0.65	0.38–1.05	0.080
Adjuvant chemoTx	+/−	0.92	0.68–1.24	0.057	0.85	0.62–1.17	0.316

.	.	Univariable analysis .			Multivariable analysis .
.		HR .	95% CI .	P-value .	HR .	95% CI .	P-value .
Age (years)	<66/≥66	0.59	0.43–0.80	<0.001	0.62	0.45–0,86	0.004
Gender	M/F	1.28	0.94–1.75	0.123	1.25	0.91–1.73	0.170
DFI (months)	≥18/<18	0.61	0.45–0.83	0.002	0.59	0.43–0.81	0.001
PS	0/1, 2	0.63	0.39–1.08	0.092	0.88	0.53–1.53	0.627
Surgical side	Unilateral/Bilateral	0.61	0.42–0.93	0.021	0.78	0.46–1.32	0.355
CEA level	Normal/High	0.39	0.29–0.53	<0.001	0.43	0.31–0.59	<0.001
Tumour numbers	Solitary/Multiple	0.61	0.45–0.85	0.004	0.76	0.51–1.19	0.229
Extrathoracic lesions	−/+	0.68	0.49–0.96	0.028	0.64	0.46–0.91	0.014
Max tumour size (mm)	<15/≥15	0.79	0.58–1.07	0.132	0.78	0.57–1.08	0.136
Surgery	Segmentecomy/Wedge	0.56	0.33–0.90	0.014	0.65	0.38–1.05	0.080
Adjuvant chemoTx	+/−	0.92	0.68–1.24	0.057	0.85	0.62–1.17	0.316

CEA: carcinoembryonic antigen; CI: confidence interval; DFI: disease-free interval; HR: hazard ratio; PS: performance status.

^aCox proportional hazards model.

The mode of recurrence is shown in Table 4. The overall recurrences developed in 325 patients (63.7%). The recurrences were more frequently seen in patients undergoing wedge resection than segmentectomy (61.8% vs 44.9%, P = 0.003). Among patients developing intrathoracic recurrences, resection-margin recurrences were more frequent in wedge resection than in segmentectomy (7.3% vs 2.0%, P = 0.035).

The 5-year recurrence-free survival was 38.2%. The 5-year recurrence-free survival of patients undergoing segmentectomy was 48.8%, and that of patients undergoing wedge resection was 36.0% (Fig. 2). Multivariable analysis showed that DFI, CEA level and surgical procedure (wedge or segmental resection) were significant factors for recurrence-free survival (Table 5).

Figure 2:

Recurrence-free rate according the surgical procedure.

Open in new tabDownload slide

Table 5:

Open in new tab

Recurrence-free survival by prognostic factor^a

.	.	Univariable analysis .			Multivariable analysis .
.	.	HR .	95% CI .	P-value .	HR .	95% CI .	P-value .
Age (years)	<66/≥66	0.86	0.69–1.07	0.181	0.91	0.72–1.14	0.401
Gender	M/F	1.14	0.91–1.42	0.259	1.22	0.97–1.54	0.086
DFI (months)	≥18/<18	0.55	0.44–0.68	<0.001	0.51	0.41–0.64	<0.001
PS	0/1, 2	0.89	0.61–1.35	0.558	1.09	0.73–1.70	0.683
Surgical side	Unilateral/Bilateral	0.60	0.45–0.81	0.001	0.86	0.58–1.28	0.458
CEA level	Normal/High	0.59	0.48–0.75	<0.001	0.60	0.48–0.76	<0.001
Tumour numbers	Solitary/Multiple	0.63	0.50–0.80	<0.001	0.70	0.52–0.96	0.030
Extrathoracic lesions	−/+	0.80	0.63–1.03	0.080	0.75	0.59–0.97	0.028
Max tumour size (mm)	<15/≥15	0.90	0.89–1.39	0.339	0.91	0.72–1.14	0.403
Surgery	Segmentecomy/Wedge	0.63	0.45–0.85	0.004	0.63	0.44–0.87	0.005
Adjuvant chemoTx	+/−	1.06	0.85–1.31	0.626	1.02	0.81–1.27	0.890

.	.	Univariable analysis .			Multivariable analysis .
.	.	HR .	95% CI .	P-value .	HR .	95% CI .	P-value .
Age (years)	<66/≥66	0.86	0.69–1.07	0.181	0.91	0.72–1.14	0.401
Gender	M/F	1.14	0.91–1.42	0.259	1.22	0.97–1.54	0.086
DFI (months)	≥18/<18	0.55	0.44–0.68	<0.001	0.51	0.41–0.64	<0.001
PS	0/1, 2	0.89	0.61–1.35	0.558	1.09	0.73–1.70	0.683
Surgical side	Unilateral/Bilateral	0.60	0.45–0.81	0.001	0.86	0.58–1.28	0.458
CEA level	Normal/High	0.59	0.48–0.75	<0.001	0.60	0.48–0.76	<0.001
Tumour numbers	Solitary/Multiple	0.63	0.50–0.80	<0.001	0.70	0.52–0.96	0.030
Extrathoracic lesions	−/+	0.80	0.63–1.03	0.080	0.75	0.59–0.97	0.028
Max tumour size (mm)	<15/≥15	0.90	0.89–1.39	0.339	0.91	0.72–1.14	0.403
Surgery	Segmentecomy/Wedge	0.63	0.45–0.85	0.004	0.63	0.44–0.87	0.005
Adjuvant chemoTx	+/−	1.06	0.85–1.31	0.626	1.02	0.81–1.27	0.890

CEA: carcinoembryonic antigen; CI: confidence interval; DFI: disease-free interval; HR: hazard ratio; PS: performance status.

^aCox proportional hazards model.

Table 5:

Open in new tab

Recurrence-free survival by prognostic factor^a

.	.	Univariable analysis .			Multivariable analysis .
.	.	HR .	95% CI .	P-value .	HR .	95% CI .	P-value .
Age (years)	<66/≥66	0.86	0.69–1.07	0.181	0.91	0.72–1.14	0.401
Gender	M/F	1.14	0.91–1.42	0.259	1.22	0.97–1.54	0.086
DFI (months)	≥18/<18	0.55	0.44–0.68	<0.001	0.51	0.41–0.64	<0.001
PS	0/1, 2	0.89	0.61–1.35	0.558	1.09	0.73–1.70	0.683
Surgical side	Unilateral/Bilateral	0.60	0.45–0.81	0.001	0.86	0.58–1.28	0.458
CEA level	Normal/High	0.59	0.48–0.75	<0.001	0.60	0.48–0.76	<0.001
Tumour numbers	Solitary/Multiple	0.63	0.50–0.80	<0.001	0.70	0.52–0.96	0.030
Extrathoracic lesions	−/+	0.80	0.63–1.03	0.080	0.75	0.59–0.97	0.028
Max tumour size (mm)	<15/≥15	0.90	0.89–1.39	0.339	0.91	0.72–1.14	0.403
Surgery	Segmentecomy/Wedge	0.63	0.45–0.85	0.004	0.63	0.44–0.87	0.005
Adjuvant chemoTx	+/−	1.06	0.85–1.31	0.626	1.02	0.81–1.27	0.890

.	.	Univariable analysis .			Multivariable analysis .
.	.	HR .	95% CI .	P-value .	HR .	95% CI .	P-value .
Age (years)	<66/≥66	0.86	0.69–1.07	0.181	0.91	0.72–1.14	0.401
Gender	M/F	1.14	0.91–1.42	0.259	1.22	0.97–1.54	0.086
DFI (months)	≥18/<18	0.55	0.44–0.68	<0.001	0.51	0.41–0.64	<0.001
PS	0/1, 2	0.89	0.61–1.35	0.558	1.09	0.73–1.70	0.683
Surgical side	Unilateral/Bilateral	0.60	0.45–0.81	0.001	0.86	0.58–1.28	0.458
CEA level	Normal/High	0.59	0.48–0.75	<0.001	0.60	0.48–0.76	<0.001
Tumour numbers	Solitary/Multiple	0.63	0.50–0.80	<0.001	0.70	0.52–0.96	0.030
Extrathoracic lesions	−/+	0.80	0.63–1.03	0.080	0.75	0.59–0.97	0.028
Max tumour size (mm)	<15/≥15	0.90	0.89–1.39	0.339	0.91	0.72–1.14	0.403
Surgery	Segmentecomy/Wedge	0.63	0.45–0.85	0.004	0.63	0.44–0.87	0.005
Adjuvant chemoTx	+/−	1.06	0.85–1.31	0.626	1.02	0.81–1.27	0.890

CEA: carcinoembryonic antigen; CI: confidence interval; DFI: disease-free interval; HR: hazard ratio; PS: performance status.

^aCox proportional hazards model.

DISCUSSION

As developments in diagnostic modality, surgical procedures and systemic therapy over the past two decades have greatly improved the outcomes of metastatic colorectal cancer patients, we believe that it is important to reevaluate the efficacy of surgery for colorectal pulmonary metastasis. Our group performed a retrospective multi-institutional study to assess the outcomes of patients undergoing pulmonary metastasectomy for colorectal pulmonary metastasis during this new era. Our database has been established to survey the outcomes of pulmonary metastases from colorectal cancer after the introduction of chemotherapy with oxaliplatin to Japan. The study subjects were the patients who underwent R0 pulmonary metastasectomy from colorectal cancers. Because our database is not a national registry but an academic database [17], it could not avoid a patient selection bias. The 5-year survival rates of patients undergoing surgery for colorectal pulmonary metastasis were previously reported to be about 50% [7–9]. Our study found a higher 5-year survival rate after pulmonary metastasectomy. The better outcome of our study might be linked with a highly selected patient population based on better imaging and effective chemotherapy.

Wedge resection has been the most common procedure for pulmonary colorectal metastases, whereas segmentectomy has been less frequently performed. Several studies have reported proportions of segmentectomy performed for colorectal pulmonary metastasis ranging from 2% to 20% [18–22]. In this study, 98 of 553 patients (17.7%) underwent segmentectomy. To the best of our knowledge, ours is the largest study of segmentectomy for colorectal pulmonary metastasis. Although the tumours were larger in the patients who underwent segmentectomy, multivariable analysis revealed that recurrence-free survival was significantly longer in patients who underwent segmentectomy. Moreover, lymph node evaluation was performed in 72 of 98 (73.5%) segmentectomies and 25 of 455 (5.5%) wedge resections. Possible explanation for long recurrence-free survival in segmentectomy was sufficient surgical margins and lymph node evaluation. However, there might be other unknown confounding factors affecting the outcomes after surgery.

Previous studies have reported that the rates of a resection-margin recurrence after pulmonary metastasectomy were 9% [15] and 28% [16]. Because segmentectomy can provide adequate resection margins, we predicted a lower rate of resection-margin recurrence after segmentectomy than after wedge resection, and found respective recurrence rates of 2% and 7.3%. Aerogenous spread of floating clusters of cancer cells in pathological findings was reported to be associated with local recurrence at the resection site [16]. The importance of accurate intraoperative cytology for preventing local recurrence has been reported [15, 23]. Higashiyama et al. [24] recommended clear surgical margins of 10 or 20 mm if possible. Pathological findings indicate that both small and large metastases require clear 3-mm and 8- to 10-mm margins, respectively, to prevent local recurrence [25]. Unfortunately, our database did not record details on the surgical approach, pathological findings and the extent of the surgical margins. Whether or not local recurrence affects survival has not been sufficiently investigated. As the type of surgical procedure appears to affect rates of resection-margin recurrence, every possible effort for preventing resection-margin recurrence should be made. Especially in cases with a centrally located tumour, or when an adequate surgical margin cannot be obtained by wedge resection, segmentectomy should be carried out without hesitation.

Although the rate of postoperative complications among the patients who underwent segmentectomy was higher than among the patients who underwent wedge resection, there were no fatal complications, and the rates of postsegmentectomy pneumonia and prolonged air leak were low, approximately 5%. These results indicate that segmentectomy for pulmonary metastasis is acceptable enough in the safety issues.

This study has limitations. Firstly, this study was a subset analysis, and we did not collect the detail surgical and pathological findings. This is a great limitation of the study. Secondly, segmentectomy was performed at the discretion of each institution considering several factors such as location, size and number of metastases, and we could not capture the criteria used to decide upon a wedge resection or segmentectomy; thus, the characteristics of the two groups of patients studied were different. We could not apply propensity-score matching, because these important factors for matching were lacking in our database. Moreover, there is a remarkable numerical difference between segmentectomy cases (n = 98) and wedge resection cases (n = 455). It makes the statistical analysis rather unreliable. Thirdly, the methods of follow-up were based on the discretion of each institution, and there were no specified diagnostic criteria for recurrence. Fourthly, although we assume that an anatomic segmentectomy was performed, we could not evaluate the technical aspects in each case. From these points of view, our results should be cautiously interpreted.

In conclusion, although wedge resection is a widely accepted surgical procedure for pulmonary metastases from colorectal cancer, recurrence especially at the resection-margin was more frequent than segmentectomy. However, it remains an open question whether this difference of recurrence rates affects survival outcomes. Because a randomized study of segmentectomy versus wedge resection would be hard to perform, a well-planed and high-quality prospective observational study should be considered, including many factors essential for selection of surgical procedures.

ACKNOWLEDGEMENTS

The authors thank all the participating investigators listed below for data collection. The authors also thank Ms Mizuki Aida for data preparation. Hirotoshi Horio, Department of Thoracic Surgery, Tokyo Metropolitan Cancer and Infectious Diseases Center Komagome Hospital; Mitsuhiro Takenoyama, Department of Thoracic Oncology, National Kyushu Cancer Center; Motohiro Yamashita, Division of General Thoracic Surgery, Shikoku Cancer Center; Takehisa Hashimoto, Division of Thoracic and Cardiovascular Surgery, Niigata University Graduate School of Medical and Dental Sciences; Atsushi Fujita, Department of Thoracic Surgery, Gunma Prefectural Cancer Center; Meinoshin Okumura, Department of General Thoracic Surgery, Osaka University Graduate School of Medicine; Kazuhito Funai, First Department of Surgery, Hamamatsu University School of Medicine; Satoshi Shiono, Department of Thoracic Surgery, Yamagata Prefectural Central Hospital; Hisatoshi Asano, Department of Surgery, University Hospital, Jikei University School of Medicine; Makoto Suzuki, Department of Thoracic Surgery, Kumamoto University Hospital; Yuji Shiraishi, Chest Surgery Division, Respiratory Disease Center, Fukujuji Hospital, Japan Anti-Tuberculosis Association; Mitsuo Nakayama, Department of General Thoracic Surgery, Saitama Medical Center, Saitama Medical University; Shunsuke Yamada, Department of General Thoracic Surgery, Tokai University Hachioji Hospital; Eishin Hoshi, Department of Thoracic Surgery, Saitama Cardiovascular and Respiratory Center; Nobuhiro Yamazaki, Department of Surgery, Saitama Medical University International Medical Center; Toshihiro Matsuo, Department of Surgery, Kurume University School of Medicine; Hideki Miyazawa, Chest Surgery, Toyama Prefectural Central Hospital; Yukio Sato, Department of Surgery (Thoracic Surgery), University of Tsukuba; Motoshi Takao, Department of Thoracic and Cardiovascular Surgery, Mie University School of Medicine; Haruhiko Nakamura, Department of Chest Surgery, St Marianna University School of Medicine; Haruhiko Nakayama, Department of Thoracic Surgery, Kanagawa Cancer Center; Kimihiro Shimizu, Department of Thoracic and Visceral Organ Surgery, Gunma University Graduate School of Medicine; Takehiro Watanabe, Division of Chest Surgery, Nishi-Niigata Chuo National Hospital; Hiroyuki Suzuki, Department of Regenerative Surgery, Fukushima Medical University; Masafumi Kataoka, Department of Surgery, Okayama Saiseikai General Hospital; Yoshio Tsunezuka, Department of General Thoracic Surgery, Ishikawa Prefectural Central Hospital; Shinji Akamine, Department of Chest Surgery, Oita Prefectural Hospital; Mitsutaka Kadokura, Division of Chest Surgery, Department of Surgery, Showa University School of Medicine; and Masao Nakata, Department of General Thoracic Surgery, Kawasaki Medical School.

Funding

This work was supported by the non-profit organization Tsukuba Cancer Clinical Trial Group.

Conflicts of interest: none declared.

REFERENCES

APPENDIX. CONFERENCE DISCUSSION

Dr L. Molins(Barcelona, Spain): We in Spain have published experience with a prospective trial and a review of more than 500 lung metastases collected over 2 years. We also found that lobectomy patients had better survival than patients with lesser resections, and this was the theory of the cluster of cells that are still in the same lobe.

So to my question. You have 213 lobectomies that were analysed in your series and only 93 segmentectomies. My understanding is that you have not analysed the survival of those lobectomies versus lesser resections; have you done it? If not, do you think that this could be a better study, comparing the anatomical versus lesser resections?

Dr Shiono: I read your paper and I am very interested that lobectomy has better survival than limited resection. Unfortunately, this is a subset analysis for colorectal cancer pulmonary metastases, so we could not compare lobectomy versus segmentectomy survival. But the patients who have pulmonary metastases have multiple nodules and some patients are in a very bad condition and have chemotherapy. So, in my opinion, I prefer limited resection, but lobectomy might have better survival than limited resection.

Dr D. Gossot(Paris, France): I have two questions. In daily practice, has your decision-making altered after this study? Will you do segmentectomy whenever you can do it? Secondly, what about a solitary metastasis that is really easy to remove by a large wedge resection? What do you prefer?

Dr Shiono: Sorry, second question once more?

Dr Gossot: In your daily practice, let’s assume you have a small solitary metastasis from colorectal cancer which is easy to remove with a large wedge resection: what will you do in practice? Do you prefer to perform a segmentectomy or a large wedge resection?

Dr Shiono: The first question is the indication for segmentectomy?

Dr Gossot: The first question is, now that you have done this study, will you do a segmentectomy any time you can do it, rather than a wedge resection?

Dr Shiono: Wedge resection is a fairly easy procedure, particularly for peripheral nodules. When doing a wedge resection, the surgical stump should be carefully checked. I think local recurrence should be avoided.

Dr F. Le Pimpec-Barthes(Paris, France): The question is what is the indication for segmentectomy? If you have the possibility to do a wedge resection, how do you decide between segmentectomy or wedge resection?

Dr Shiono: Okay, I got it. It depends on the location. In the case of central nodules very close to the hilum, we would choose segmentectomy, but for peripheral nodules, we should choose wedge resection.

Author notes