| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2003;76:343-349
© 2003 The Society of Thoracic Surgeons
a Department of Surgery, University of Virginia School of Medicine, Charlottesville, Virginia, USA
* Address reprint requests to Dr Jones, Assistant Professor of Surgery, Department of Surgery, PO Box 800679, University of Virginia, Charlottesville, VA 22908-0679, USA.
e-mail: djones{at}virginia.edu
Presented at the Forty-ninth Annual Meeting of the Southern Thoracic Surgical Association, Miami Beach, FL, Nov 7–9, 2002.
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and methodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
METHODS: We retrospectively reviewed the records of 61 patients (5% of all anatomic resections) undergoing 62 segmentectomies from 1991 to 2001. Wedge resections were excluded. The operative indications were suspected or known lung cancer (43), benign disease (12), and metastatic cancer to the lung (7). Median follow-up for patients with malignancy was 28 months (range 1 to 89 months). Actuarial survival was calculated using the Kaplan–Meier method.
RESULTS: Segmentectomy was performed in 43 lung cancer patients with pathologic stages of Ia-22, Ib-14, IIa-2, IIb-1, IIIa/IIIb-2, and IV-2. All resection margins were negative for malignancy. Segmentectomy for benign diseases included granulomatous disease (5), pulmonary abscess (2), plasmacytoma (1), and others (4). Complications occurred in 39% (24/62) of patients, including atelectasis requiring bronchoscopy (10/62, 16%), pneumonia (9/62, 14%), air leak more than 7 days (5/62, 8%), and supraventricular dysrhythmias (6/62, 10%). In-hospital mortality was 3% (2 patients). Median length of hospital stay was 6 days (range 4 to 66 days). In the lung cancer patients the locoregional recurrence rate was 0% and the distant recurrence rate was 20%. The 4-year actuarial survival for patients with lung cancer was 72%.
CONCLUSIONS: Pulmonary segmentectomy has acceptable morbidity and mortality in selected patients with both benign and malignant lung disease and remains a useful procedure in a thoracic surgeon’s armamentarium. Distant, not locoregional recurrence, was responsible for the cancer deaths.
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and methodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
We have traditionally performed a lobectomy or pneumonectomy for primary lung cancer in "good-risk" patients and reserved the use of pulmonary segmentectomy for patients with NSCLC who have limited pulmonary reserve, or in patients with central metastatic lesions and benign processes that are not amenable to wedge resection. Therefore, the purpose of this study was to evaluate the results of our experience with pulmonary segmentectomy in this select group of patients.
|
Patients and methods |
|---|
|
TopAbstractIntroductionPatients and methodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
Preoperative evaluation
All patients had a directed history and physical examination, a chest roentgenogram, computed tomography (CT) of the chest, and pulmonary function testing. The percent predicted postoperative forced expiratory volume in 1 second (FEV1-ppo) was calculated by using quantitative perfusion scans or by the method of Juhl and Frost [5] that uses the number(s) of resected segments to calculate the FEV1-ppo. Extrathoracic staging was performed if clinically indicated. Mediastinoscopy was performed in patients with radiographic evidence of pathologically enlarged lymph nodes (
1 cm short-axis diameter) or for tumors rated T2 or worse as measured by CT scan.
Surgical technique
After informed consent, an epidural catheter was inserted, followed by general anesthesia with selective lung ventilation. All patients underwent bronchoscopy to rule out other endobronchial lesions. A variety of incisions were used, but we now favor a vertical axillary incision through the fourth intercostal space. Exploration of the hemithorax was performed and intrapleural adhesions were lysed carefully to avoid visceral pleural injury that may result in postoperative air leaks. After confirmation of the segmental location of the lesion, an attempt was made to confirm by frozen-section analysis a histologic diagnosis of malignancy, if this was not known preoperatively. Tissue samples were obtained through either an excisional or core-needle biopsy. The segmental artery was then identified, ligated, and divided. The arterial supply to the affected segment can be variable and anomalies do occur; therefore, it is imperative to be certain of the patient’s anatomy before ligation of the segmental artery. The segmental bronchus is generally 1 to 2 cm peripheral to the origin of the segmental artery from the pulmonary artery and dissection is performed at this level. We have found that traction on the affected segment with a Duval lung clamp facilitates identification and subsequent dissection of the segmental bronchus, thus avoiding injury to adjacent bronchi and unnecessary parenchymal dissection. For patients with known or suspected lung cancer, all interlobar (station 11), lobar (station 12), and segmental (station 13) lymph nodes were swept distally and taken en bloc with the specimen.
Before division of the bronchus, we occasionally clamped the segmental bronchus and then ventilated the ipsilateral lung to ensure that the correct bronchus had been isolated. This maneuver, while intuitively obvious, may not be effective in patients with underlying obstructive pulmonary disease secondary to ventilation into the occluded segment through the pores of Kohn. The segmental bronchus was either transected and oversewn with absorbable interrupted suture (n = 22), or stapled and divided (n = 39).
Identification with separate ligation and division of the segmental pulmonary vein were rarely performed. We routinely use the stapler to divide the intersegmental parenchymal plane that effectively incorporates, and divides, the respective segmental pulmonary vein. In patients with severe emphysema we use bovine pericardial strips to buttress the staple line, thus minimizing postoperative air leaks. Identification of the most central aspect of the planned parenchymal division is facilitated by placement of, and tension on, a traction suture placed in the distal segmental bronchus of the segment being resected. Every effort was made to achieve a 2-cm margin around the tumor. On occasion this goal required an extension of the resection margin into the adjacent segment(s) as has been described by Okada and colleagues [6].
Once the specimen and all corresponding nodal tissue were removed, mediastinal lymph node sampling or dissection was performed in most cases and in all cases since 1995. Lymph node dissection was not performed for benign or metastatic lesions. Routine frozen-section analysis of lymph nodes or surgical margins was not performed, and only done if the surgeon suspected that the final resected margins might be close.
Follow-up
Follow-up, complete in 98% of the study patients, was obtained from review of the medical records, tumor registry, and telephone interviews. Median follow-up for patients with primary lung cancer was 28 months (range 1 to 89 months). Locoregional recurrence was defined as the development of an additional carcinoma in the ipsilateral hemithorax (lung and mediastinum). Tumor recurrence was documented either histologically or radiographically. Actuarial survival was calculated by the Kaplan–Meier method.
|
Results |
|---|
|
TopAbstractIntroductionPatients and methodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
|
|
|
|
|
|
|
Comment |
|---|
|
TopAbstractIntroductionPatients and methodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
Since the seminal report of Jensik and colleagues [8] nearly 25 years ago, no other reports have critically evaluated the postoperative complications related to performing a segmentectomy. As we have shown, the most common complication is the need for postoperative bronchoscopy for retained secretions and atelectasis. These patients frequently have a poor cough despite epidural anesthesia, early mobilization, and chest percussion. We have recently begun to use a 4.0-mm minitracheostomy (SIMS Portex Inc, Keene, NH) on these patients and anecdotally we believe the need for bronchoscopy has decreased. The incidence of pneumonia and supraventricular dysrhythmias is similar to that seen after lobectomy or pneumonectomy [9]. Finally, although segmentectomy has been thought to increase the incidence of postoperative air leaks, our 8% significant air leak rate is similar to the 7% rate reported by others [8, 9]. Our low incidence of significant air leaks for this procedure is likely related to division of the lung parenchyma with the stapler, as well as our low threshold to use bovine pericardial strips to buttress the staple lines [10].
Although the role of segmentectomy for selected benign lung diseases is not debated, controversy continues to exist regarding the role of segmentectomy in the treatment of primary lung cancer. The group at Rush-Presbyterian examined the utility of segmentectomy compared with lobectomy in "good-risk" patients, and concluded that patients undergoing segmentectomy had an increased risk of locoregional recurrence (23% versus 5%) and decreased 5-year survival [11]. Interestingly, in this retrospective study, when only the patients with pathologic stage Ia disease were compared, no survival difference was noted between segmentectomy and lobectomy. In 1995 the Lung Cancer Study Group reported the results of their prospective study (again in good-risk patients) that examined the role of "limited" resections compared with lobectomy for pT1N0 mol/L0 lung cancers [3]. This study suggested that a limited resection, which included wedge resections and segmentectomies, resulted in a threefold increase in locoregional recurrence. In addition, the increased death rate for patients with limited resections compared with lobectomy became apparent only 4 years after the operation. Together, these two studies provided the evidence base for which lobectomy, not a "limited" resection, is the preferred oncologic procedure for pathologic stage Ia/Ib NSCLC in "good-risk" patients.
In contrast to the above studies, the Japanese have prospectively evaluated the role of segmentectomy for tumors of a maximum size of 2 cm with negative nodes (pT1N0M0) and found no difference in survival between segmentectomy and lobectomy [5, 12–14]. In addition, Miller and colleagues [15] recently reported results in 100 patients with tumors of 1 cm or less and found no difference in overall 5-year survival (57% versus 71%), lung cancer-specific 5-year survival (75 versus 92%), or locoregional recurrence rates (8% versus 13%) between segmentectomies and lobectomies, respectively. Differences in locoregional recurrence rates and survival were significant only when wedge resections were compared with lobectomy in this patient population. Thus, contemporary, compelling data suggest that for smaller T1N0 mol/L0 lesions staged intraoperatively, segmentectomy may be an acceptable oncologic procedure in patients with adequate pulmonary reserve.
We have routinely performed a segmentectomy in patients with NSCLC who are considered by the surgeon to have limited pulmonary reserve, a marginal performance status, or both. Although we prefer to perform the procedure on smaller T1 tumors, as shown in Table 5, more than 30% of tumors in our series were T2 lesions. Our definition of limited pulmonary reserve is similar to that used by Cerfolio and associates [7], less restrictive than described by Miller and Hatcher [16], and more restrictive than that described by Landreneau and associates [17]. As shown in Table 2, the median FEV1-ppo was 42%, which approaches the 40% value considered to be an acceptable breakpoint for a significant increase in morbidity and mortality after lobectomy or pneumonectomy [18].
A careful review of the literature highlights the difficulty in separating the survival and recurrence rates in patients with NSCLC between segmentectomy and wedge resections, irrespective of whether the patient is considered "high-risk." This dilemma is largely because the two procedures have been combined, and together have been classified as "limited" or "nonanatomical" resections [3, 15, 19, 20]. We believe that segmentectomy is an anatomic resection, and to that end should not be categorized with nonanatomic thoracoscopic or open wedge resections. Miller is the first author to attempt to separate the two procedures, and found that even in patients with the earliest stage NSCLC (tumors of 1 cm or smaller), a wedge resection resulted in a 5-year survival rate of 27% and a 31% incidence of locoregional recurrence compared with a 5-year survival rate of 57% and 8% locoregional recurrence for a segmentectomy [15]. Another study from the Mayo group found that a wedge resection in "high-risk" patients with NSCLC was an independent predictor of worse survival [7]. It is beyond the scope of this retrospective study to definitively state that a wedge resection is inferior to an anatomic segmental resection, but based on these studies as well as our own experience, we eschew performing nonanatomical wedge resections. Instead we prefer, whenever possible, performing an anatomical segmentectomy with lymph node dissection in selected patients with NSCLC.
This study demonstrates that in high-risk patients with NSCLC a pulmonary segmentectomy can achieve an acceptable survival rate without the significant locoregional recurrence rates previously reported by others [8, 11]. The limitations of this study include the fact that it was a retrospective analysis and the possibility exists that a locoregional recurrence was missed or will occur with longer follow-up. Our median follow-up was only 28 months, but most, although not all, recurrences occur within the first 2 years after segmentectomy [11, 20]. Another potential criticism of the study is that a small number of patients may have had adequate pulmonary function to undergo a lobectomy instead of a segmentectomy. The choice to perform a segmentectomy in these patients was based on the fact that they had had previous anatomic pulmonary resections, or may have had an adequate %FEV1-ppo, but had a poor DLCO. The ultimate outcome in those few patients does not appear to be compromised as evidenced by a 72% 4-year survival for all patients with lung cancer, and the fact that segmentectomy has a good 5-year survival in highly selected (pT1N0M0), "good-risk" patients [12, 13, 15].
In conclusion, using the technique described, pulmonary segmentectomy can be performed for benign and metastatic lung lesions with low morbidity and mortality. While technically more challenging than other anatomic pulmonary resections, the procedure is valuable and should not be forgotten by current-generation thoracic surgeons. In addition, in patients with limited pulmonary reserve and primary lung cancer, segmentectomy can be performed with an anticipated 4- to 5-year survival of 50% to 80% and a low chance of developing a locoregional recurrence (Table 6). As with all early-stage NSCLC, distant, not locoregional, recurrence remains the primary reason for cancer-related deaths after segmentectomy. Finally, no current or pending randomized trial has compared segmentectomy with wedge resection in high-risk patients with NSCLC. However, based on our experience and the reports of others, we advocate performing a segmentectomy with lymph node dissection whenever feasible in patients with NSCLC and limited pulmonary reserve.
|
|
Discussion |
|---|
|
TopAbstractIntroductionPatients and methodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
First of all, what is a segment and when is segmentectomy appropriate? I think we all agree what is a formal posterior segmentectomy of an upper lobe or a superior segmentectomy of the lower lobe: we find the artery, we find the bronchus, we even identify the vein, then we do a formal, "straight-up'' segment. Then we have segments, such as in the basilar part of the lower lobes where we improvise; it is not much different than a giant wedge. Could you talk to us about, what really defines a segment?
And then second, when do you do a segmentectomy? We have seen many essential reports over the last few years—''how low can you go" sort of papers— with increasingly lower postoperative predicted FEV1%. I have become impressed that almost anybody can undergo a lobectomy with epidurals and all our other anesthetic and postoperative care. So please talk about criteria for segmentectomy.
And that leads me to my last few questions about the use of frozen sections. My preference has always been to look at the N1 lymph nodes in the part of the segments that I am not removing and send those N1 nodes in for frozen sectioning. I also ask the pathologist to check the staple line on frozen sections, because if those are positive, usually I will do a larger resection. The question is, can the patient tolerate a larger resection and if so why not just do that in the first place, and is lobectomy a better operation? Why do you not evaluate the bronchial margin, the N1 nodes that stay in, or the staple lines by frozen section?
We also check N2 lymph nodes on frozen sections, and I noticed that you did not mention that evaluation. We especially check them on patients with poor pulmonary function tests, because if their N2 nodes are involved, we usually will not resect them. I think we are offering patients only local control with significant morbidity and the concomitant risk for a disease that takes their life with systemic disease.
Finally a few other questions about the vertical axillary thoracotomy. Can your resident or fellow really see when you use that approach?
Thank you.
DR JONES: Thank you, Dr Cerfolio, for your comments. Let me try to address them. A segment is fairly well defined as a region of the lung that possesses its own bronchus and blood supply. Most of our segmentectomies were performed in the upper lobes and occasionally in the lower lobe. We have not done individual segments of what is considered to be the basilar segment. A basilar segmentectomy is, in essence, multiple segmentectomies. We have not tried to do, for instance, an anteromedial basilar segmentectomy.
When do I recommend doing a segmentectomy? There is, as you know, no pulmonary function test criterion that completely defines which patient should get a wedge, a segmentectomy, or a lobectomy. I agree with you that the most commonly performed procedures are lobectomies. This is evidenced by the fact this paper represents only 5% of all the anatomic resections performed at the University of Virginia. I consider a segmentectomy when the patient has significant comorbidities, when their predicted postoperative FEV1 or predicted postoperative DLCO is approaching 35% to 40%, when their performance status is 1, 2 or greater, or they arrive in a wheelchair. Under those circumstances I consider some type of limited, anatomic resection, such as a segmentectomy.
With respect to the frozen-section analysis, we believe that segmentectomy is the maximal resection that these patients will tolerate. We always sample the N2 nodes and would agree with you to not perform a segmental resection if we found N2 nodes in multiple nodal stations. If, however, we found an N2 node in only one station after a negative pre-resection mediastinoscopy, we would go ahead and resect that patient’s tumor and perform a lymphadenectomy. If we have any concerns about our parenchymal staple line or our bronchial margin, we obtain a frozen section, but with most of these lesions that was not a concern and we did not routinely obtain a frozen section.
And finally, with respect to the vertical axillary incision—it is a great incision. Two surgeons sometimes have difficulty seeing simultaneously, but in general we manage.
DR GEOFFREY M. GRAEBER (Morgantown, WV): I appreciated your fine talk, Dr Jones, and your interesting data. You bring up one important question that I would like you to address.
A number of these patients with bad COPD do have prolonged air leaks, granted that number is small. How do you handle those air leaks and how do you get those patients ambulatory and out of the hospital?
I enjoyed your presentation.
DR JONES: That is a good question, Dr Graeber. We manage the air leaks in patients with COPD by first dividing their lung parenchyma with bovine pericardial strips as described by Miller and Landreneau. This tactic helps us by decreasing the amount of air leaks before leaving the operating room. We have also used Focal Seal and other pulmonary sealants. We are liberal about creating pleural tents and, in some cases, a pneumoperitoneum for these basilar segmentectomies that were performed. All of these strategies have led to a low incidence of significant air leaks in what traditionally would be considered to be an operation associated with a higher incidence of air leaks.
|
Acknowledgments |
|---|
|
TopAbstractIntroductionPatients and methodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
|
References |
|---|
|
TopAbstractIntroductionPatients and methodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  M. Shapiro, T. S. Weiser, J. P. Wisnivesky, C. Chin, M. Arustamyan, and S. J. Swanson Thoracoscopic segmentectomy compares favorably with thoracoscopic lobectomy for patients with small stage I lung cancer. J. Thorac. Cardiovasc. Surg., June 1, 2009; 137(6): 1388 - 1393. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Watanabe, S. Ohori, S. Nakashima, T. Mawatari, N. Inoue, Y. Kurimoto, and T. Higami Feasibility of video-assisted thoracoscopic surgery segmentectomy for selected peripheral lung carcinomas Eur. J. Cardiothorac. Surg., May 1, 2009; 35(5): 775 - 780. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. A. D'Amico Thoracoscopic Segmentectomy: Technical Considerations and Outcomes Ann. Thorac. Surg., February 1, 2008; 85(2): S716 - S718. [Full Text] [PDF]
|
|
|
|
|
|
B. Z. Atkins, D. H. Harpole Jr, J. H. Mangum, E. M. Toloza, T. A. D'Amico, and W. R. Burfeind Jr Pulmonary Segmentectomy by Thoracotomy or Thoracoscopy: Reduced Hospital Length of Stay With a Minimally-Invasive Approach Ann. Thorac. Surg., October 1, 2007; 84(4): 1107 - 1113. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Okada, T. Mimura, J. Ikegaki, H. Katoh, H. Itoh, and N. Tsubota A novel video-assisted anatomic segmentectomy technique: Selective segmental inflation via bronchofiberoptic jet followed by cautery cutting J. Thorac. Cardiovasc. Surg., March 1, 2007; 133(3): 753 - 758. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Okada, T. Koike, M. Higashiyama, Y. Yamato, K. Kodama, and N. Tsubota Radical sublobar resection for small-sized non small cell lung cancer: A multicenter study J. Thorac. Cardiovasc. Surg., October 1, 2006; 132(4): 769 - 775. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 O. Schussler, M. Alifano, H. Dermine, S. Strano, A. Casetta, S. Sepulveda, A. Chafik, S. Coignard, A. Rabbat, and J.-F. Regnard Postoperative Pneumonia after Major Lung Resection Am. J. Respir. Crit. Care Med., May 15, 2006; 173(10): 1161 - 1169. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. E. Martin-Ucar, A. Nakas, J. E. Pilling, K. J. West, and D. A. Waller A case-matched study of anatomical segmentectomy versus lobectomy for stage I lung cancer in high-risk patients Eur. J. Cardiothorac. Surg., April 1, 2005; 27(4): 675 - 679. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Okada, W. Nishio, T. Sakamoto, K. Uchino, T. Yuki, A. Nakagawa, and N. Tsubota Effect of tumor size on prognosis in patients with non-small cell lung cancer: The role of segmentectomy as a type of lesser resection J. Thorac. Cardiovasc. Surg., January 1, 2005; 129(1): 87 - 93. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Iwasaki, T. Shirakusa, T. Shiraishi, and S. Yamamoto Results of video-assisted thoracic surgery for stage I/II non-small cell lung cancer Eur. J. Cardiothorac. Surg., July 1, 2004; 26(1): 158 - 164. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
