| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Department of Thoracic Surgery, Roswell Park Cancer Institute and University at Buffalo, Buffalo, New York
b Department of Biostatistics, Roswell Park Cancer Institute and University at Buffalo, Buffalo, New York
Accepted for publication May 20, 2009.
* Address correspondence to Dr Demmy, Department of Thoracic Surgery, Roswell Park Cancer Institute, Elm and Carlton Sts, Buffalo, NY 14263 (Email: todd.demmy{at}roswellpark.org).
Presented at the Forty-fifth Annual Meeting of The Society of Thoracic Surgeons, San Francisco, CA, Jan 26–28, 2009.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
Methods: From January 1, 2002, to September 30, 2008 at a comprehensive cancer center, 70 patients underwent pneumonectomy. Three patients were excluded for emergent operations. Thoracoscopic pneumonectomy was completed successfully in 24 patients and attempted in 8 others (25% conversion rate). Analysis was done on an intention-to-treat basis.
Results: By 2008, 75% of pneumonectomy cases were planned as TP while there were no conversions to thoracotomy. There was no difference in median blood loss between patients undergoing TP versus thoracotomy (325 vs 300 mL, p = 0.52), but operations were longer (286 vs 228 minutes, p < 0.01). Median intensive care unit stay was 2 days in both groups and median hospital stay was 5.0 days in the TP group versus 6.0 days in the thoracotomy group (p = 0.28). Major complications were similar between groups. The TP reoperations were for bleeding (2), bronchopleural fistula (2), empyema (1), and chylothorax (1). The only TP death occurred in an 83-year-old patient from respiratory failure. Neither the use of adjuvant therapy nor the time between surgery and commencement of adjuvant therapy was different between groups. Conversions alone compared with patients undergoing thoracotomy were associated with a moderate increase in blood loss and intensive care unit stay, but not in any major complications.
Conclusions: Thoracoscopic pneumonectomy can be done safely. The availability of this option is important especially in an era of multimodality therapy as more debilitated patients present for surgical therapy.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
Recent studies have reported on the feasibility and safety of thoracoscopic lobectomy compared with conventional thoracotomy for early stage lung cancer [1]. Benefits have included shorter hospital stay [2], decreased pain [3–6], decreased blood loss [7], earlier return to normal activities, tolerance of adjuvant chemotherapy [8, 9], and improved quality of life [10, 11]. There are still questions about whether the use of thoracoscopic surgery can be expanded to the treatment of more advanced disease, such as lung cancer requiring a pneumonectomy. The technique of thoracoscopic pneumonectomy (TP) has been described [12].
Our main purpose in this study was to assess the safety of TP in our experience. The second objective was to compare our early outcomes for patients undergoing thoracoscopic versus conventional thoracotomy for lung cancers requiring pneumonectomy.
|
Material and Methods |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
|
Analysis was done using an intention-to-treat basis. When important, patients undergoing conversions were compared with patients undergoing thoracotomy separately. Descriptive statistics such as frequencies and relative frequencies were computed for categoric variables. Numeric variables were summarized using simple descriptive statistics such as the median, standard deviation, range, etc. They are represented in the tables by their median values. The Fisher exact test was used to study the association between categoric variables. The Wilcoxon rank sum test was used to compare the groups in regard to numeric variables. A 0.05 nominal significance level was used in all testing. All statistical analyses were done using SAS (version 9.1; SAS Inc, Cary, NC).
|
Results |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
Of the 67 patients included in the final analysis, 32 were initially attempted thoracoscopically. Eight patients (25%) were converted from TP to thoracotomy. The reasons for conversions were tumor extension in 5 patients (2 with positive margins, 1 with lymphadenopathy, 1 with peritumoral inflammation, 1 with unclear extent), adhesions in 2 patients, and bleeding in 1 patient. Since the study design was intention-to-treat, the 8 conversions were placed into the TP group.
Demographic information on these groups is provided in Table 1. There were more female patients in the TP group compared with the open group but there were no other significant differences between the groups.
Operative details are shown in Table 2. Median blood loss was similar between groups while operative time was increased in patients undergoing TP. When analyzing the patients undergoing conversions separately, both blood loss (600 vs 300 cc, p = 0.01) and operative time (340 vs 228 minutes, p = 0.01) were increased compared with the open group.
|
The LOS and ICU days were similar in the TP group compared with the open group (Table 3). When analyzed separately, the 8 conversion patients had a significantly longer ICU stay compared with the open group (3.5 vs 2.0 days, p = 0.01). And by their exclusion, median ICU stay for the nonconverted TP patients was one day.
|
Eleven patients required reoperations, 6 in the TP group and 5 in the open group. Of interest, of the 6 patients in the TP group, none were initially converted to thoracotomy. Two patients in each group were reoperated for bleeding. Both patients undergoing TP and developing a bronchopleural fistula (BPF) had delayed presentations (>30 days). One patient had adjuvant radiation treatment and was diagnosed with a BPF 9 months after her operation. This was treated successfully with bronchoscopy and fibrin glue. The other patient had adjuvant chemotherapy and developed her BPF 11 months afterward. This required repair through a thoracotomy. The sole patient in the open group developing a BPF presented early. One patient in each group developed an empyema requiring a reoperation. The remaining reoperations were in one TP patient who developed a chylothorax and one open patient who had a suspected BPF.
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
Technical feasibility has been a major concern in using thoracoscopy for locally advanced lung cancer. Even though our conversion rate of 25% is higher than the 5% to 15% rate in reports of thoracoscopic lobectomy for treatment of early-stage lung cancer [3, 13–16, 18], it is perhaps not surprising given the more advanced disease in these patients and the relatively limited experience with TP. In fact, the number could have been higher if a conversion was defined by the preoperative clinical decision. However, we believe the decision to proceed with a TP requires thoracoscopic assessment at the time of operation. As has been shown in many minimally invasive procedures, there is definitely a learning curve with thoracoscopic lobectomy and pneumonectomy [24, 25]. This is shown in our experience as well (Fig 1). The proportion of attempted thoracoscopic pneumonectomies increased and rate of conversions in our series generally decreased over the 6 years studied. When looking at our data thus far over the last 12 months, there have been no patients requiring a conversion. If this continues, our short-term results for TP may improve. For our group, the progression to thoracoscopic pneumonectomy has been a natural next step once the thoracic surgeon reaches the point on the learning curve where the majority of patients requiring lobectomy can expect a thoracoscopic approach without conversion. At this point, the same skills developed for difficult dissections and management of modest bleeding without conversion can be used for tumors that require pneumonectomy. In fact, as in open pneumonectomy, dissection of central airways and blood vessels can be easier than peripheral ones provided retractors are used to provide adequate exposure.
|
Thoracoscopic pneumonectomy also seems to be safe in this population. There were no statistically significant differences in any complication or overall complication rate between groups. As we found in our study, arrhythmias, in particular atrial fibrillation, have been reported in other series as similar between thoracoscopic lobectomy and lobectomy through conventional thoracotomy [2, 26]. We were not able to show a statistically significant decreased length of stay, ICU stay, or need for blood transfusions; findings other studies with thoracoscopic lobectomy have shown [2]. This is likely due to the limited statistical power of our study. As also reported by others, we found a low short-term mortality rate [27].
Other short-term benefits of thoracoscopic surgery may include more patients being able to receive adjuvant treatment and being able to receive this treatment faster [8]. Our study did not suggest this benefit, but with a larger sample size and more experience this may become evident.
Finally, the question remains as to the long-term outcomes and oncologic equivalency of TP to conventional thoracotomy. There is a poorly understood, but well-documented interaction between limited pulmonary reserve after open lung resection and chronic chest pain lasting for years. Thus, avoiding chest wall perturbation with pneumonectomy seems prudent to us even if inpatient outcome data seem similar. Some have hypothesized that the reduced systemic inflammatory response with minimally invasive surgery may be associated with improved long-term outcome [28, 29]. The long-term outcome is critical in patients undergoing TP but was not the objective in this study given the limited sample size, experience, and follow-up. However, our preliminary analysis with a 14-month median follow-up shows that the TP group had a significantly better disease-free survival (26 vs 8 months, p = 0.04) and overall survival (not yet reached versus 13 months, p < 0.01) compared with the open group. We acknowledge that this might be secondary to factors other than the type of procedure the patients had; ie, differences in stage. Our population size was too small, however, to make any stage-specific comparisons. Thus we cannot say if the long-term benefits were due to TP alone. This remains an important question to answer.
There are some shortcomings in our study. First, it is a retrospective study. We tried to eliminate bias by excluding patients who could not be resected safely by both techniques (invasion into surrounding structures). Despite this, we understand there probably remains some bias in determining which patients received TP versus conventional thoracotomy, as evidenced by the larger and higher stage tumors in the open group. However, we acknowledge all patients needing pneumonectomies are not candidates for TP. There has to be a careful selection of patients, especially in one's early experience. The indications we adopted for this procedure as we began our series may be considered a guide.
In conclusion, TP for lung cancer is a feasible and safe operation. There is definitely a learning curve for these procedures. Improved surgical tools will broaden its use for treatment of lung cancer [30, 31]. More studies are needed to evaluate the role of TP for lung cancer and long-term follow-up of these patients needs to be performed to confirm the oncologic equivalence to conventional thoracotomy.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
The first is, and I'm sure you and Todd have heard this question before, that many would say that you really can't exclude the possibility that you could do a sleeve resection until a manual exploration was done. Now, I believe a lot of things can be done thoracoscopically, so I'm not criticizing the thoracoscopic approach. But how would you answer the question that you really could exclude the feasibility that an open resection would actually be able to achieve a sleeve resection?
The second question is, on a left pneumonectomy, I think the most important technical aspect is to get a proximal bronchial transection, which, an open procedure, could be performed behind the aortic arch. And once again, we can do a lot of things thoracoscopically, but I think that's actually relatively difficult to achieve; to get the same type of bronchial stump ligation thoracoscopically as can be done open.
And then, lastly, you weren't able to demonstrate quality-of-life advantages based on length of stay, and didn't measure pain or analgesic use, but I would suppose that the pain might be less than thoracotomy. As well, you didn't demonstrate significant oncologic advantages, such as the ability to administrate chemotherapy. Nor do you demonstrate reduced complications.
So how would you and your colleagues characterize why this operation would be a better operation, and how it should be used?
DR NWOGU: Thank you. I think those are great questions, Dr D'Amico. In terms of the assessment about whether a sleeve resection can be done, I think there are some cases in which the bronchoscopic appearance of the lesion makes the decision about a pneumonectomy clear. I think in cases where it's uncertain or there is a significant possibility of doing a sleeve resection, we would convert to an open procedure to make that determination.
And as our experience has grown, there are some cases where we do bronchoplasty through VATS [video-assisted thoracoscopic surgery]. And in the ideal setting, we would probably attempt to do a sleeve resection through VATS. In the cases that clearly require an open procedure to conserve pulmonary tissue, we would do that. We would open.
In terms of the length of the stump on the left side, that is definitely true. It's difficult. We do put a lot of traction on the lung and we use roticulated staplers to try and get as far down as we can. In those cases, we're also using some additional tissue to cover the stump.
But I think it's difficult to be as close to the carina with a VATS pneumonectomy as one would accomplish with an open procedure, so that is definitely a limitation. In terms of not demonstrating a clear-cut benefit in these patients, I think it mimics the doubts expressed when VATS lobectomies were introduced. It took a while before the benefits could be clearly demonstrated.
Similarly, here we have a limited number of patients. I don't think we can definitely say there is no benefit. I think with a larger sample size, we might be able to demonstrate the benefits.
DR D'AMICO: So with a larger sample size looking ahead, what do you think the benefits would be? Do you think they would be the same as thoracoscopic lobectomy?
DR NWOGU: I think we could definitely show a statistically significant increase in the number of patients that get adjuvant chemotherapy and how quickly they started the chemotherapy.
I think if we analyze our data and show the disposition of patients when they go home, a larger number of patients are independent. So I think there are some benefits that we can show as time goes on. And right now our data is not mature. We have only about 14 months of follow-up, but the survival curves are very promising. But it's too early and the selection is somewhat biased, so it's hard to prove that at this point.
DR D'AMICO: Good work.
DR SCOTT W. SWANSON (Boston, MA): Chumy, excellent presentation and interesting ideas. Just to get a little bit more of what Tommy D'Amico was saying, during this time period, how many sleeve lobectomies did you do in your institution, any idea?
DR NWOGU: I don't have those numbers.
DR SWANSON: Do you have a guess, less than 5, more than 10?
DR NWOGU: Even before we started doing the VATS pneumonectomies, I think there was only a small percentage of patients that were ideal for sleeve lobectomies. Overall, our proportion of patients that get sleeve resections has not decreased.
DR SWANSON: And the 3.8-cm size for a pneumonectomy seems a little on the low side compared to my practice. Do you have any sense of what type of tumors these were and what the range in size was?
And along the same lines, did the utility incision change at all to do a pneumonectomy in terms of the length versus a standard utility incision for a lobe?
DR NWOGU: Some patients had relatively smaller lesions, but they were so proximally located and the involvement of the PA especially, the main PA was such that a pneumonectomy was required for relatively smaller tumors.
In terms of the size of the incision, earlier in our series, maybe the size was a little longer. We didn't do any rib spreading. But in general, in our current practice, it's about 5 cm for the pneumonectomies, 4 cm for the lobectomies.
DR SWANSON: And one last question, the bleeding seems interesting. The number that you reported was lower, 200 I think. Was that, I assume, the surgeon's estimate?
But it seemed like, if I read the table right, about a third or 33% of the patients were transfused, so I was just trying to make sense of those two numbers and if you could shed any light on that.
DR NWOGU: Yes. The numbers were obtained from our OR records. At the end of the case, there is typically a discussion between the surgeon and the anesthesiologist about the volume in the drainage canisters and the volume of irrigation used. And whatever they agree is the blood loss, that's what the nurse documents. So when our fellow pulled the data, it was from our OR records.
But I think the proportion of patients that received blood transfusions was somewhat high, given the average blood loss, so I appreciate what you're saying.
DR DANIEL L. MILLER (Atlanta, GA): I enjoyed your presentation, but I think the main issue is that a lot of us here are concerned in regards to the safety of doing this procedure.
The issue is, as you know, for lung surgeons, we live and die by the stapler, and there is not a perfect vascular stapler that fires successfully 100 percent of the time. So my main concern would be going across a main pulmonary artery and stapler problem would occur, which could be fatal.
What techniques do you at that time of ligation and division of the pulmonary artery to prevent a fatal complication? Do you have a Satinsky clamp ready? Have you opened the pericardium? How do you prepare for a bleeding problem and if it occurs what would do you do?
DR NWOGU: Well, the key thing that we determine before doing this procedure is adequate proximal control of the PA [pulmonary artery]. We have to be very comfortable passing the stapler around the PA.
And I know that personally, in some cases I've applied also a vascular clamp proximally. Sometimes I just have it there without closing it. When I started doing these procedures initially, I would apply the clamp, fire the stapler, and then be very satisfied with the staple line, then release the clamp. But as time went on, I was more comfortable with doing the procedure, and I don't do that routinely.
But I think it's important to have very good proximal control. And there is enough room to put a Satinsky clamp there before firing the stapler.
DR MILLER: And the second question is, you had a very high rate of atrial fibrillation, probably one of the highest rates of any VATS lobectomy, well, VATS series. This is a pneumonectomy. Because most are usually in the single digits and less than 15. You're at 33%, which is pretty much what the open is especially from that series that was presented this morning. How do you attribute that?
DR NWOGU: Well, in looking at the literature while preparing the manuscript, I think it's been difficult for people to demonstrate in large VATS series of lobectomies that there is a difference in afib rates between VATS and open procedures. So, it is controversial whether VATS really reduces the incidence of atrial fibrillation.
But in our open series, we had a high incidence of atrial fibrillation, so we actually started a prophylactic amiodarone protocol in our patients.
So, we were not able to demonstrate that VATS decreased the incidence of atrial fibrillation in our series.
DR MILLER: And then last is what Tommy brought up in regards to the bronchial stump and especially on the left side as he brought up. Because even in an open case, you're trying to get as deep as you can. You are underneath the arch and so forth.
And also, too, is making sure that that end of the double-lumen tube is up far enough so you're not going across that. And if you have an inexperienced anesthesiologist, I'm sure there is a possibility you can go across the tip.
But what do you do in the VATS pneumonectomy on the left side to assure that, and when you do that, how do you fire your stapler across? Is that from anteriorly? Because I think coming from posterior, the aorta would be in the way. You would get a tangential cut and so forth.
DR NWOGU: Well, we do the bronchus last, so everything else has been divided. We tend to use a triangular laparoscopic liver retractor to encircle the hilum, and that allows us to really lift up the lung to really put some traction on it, and we use a roticulating stapler.
Depending on what the anatomy is, we introduce the stapler from one of the two lower ports. It's interchangeable. It depends on what the anatomy allows us to do.
We definitely ask the anesthesiologist to pull the double-lumen tube back into the trachea before we fire. We do that routinely after we have our stapler in place.
So those are some of the maneuvers we use. With a combination of those things, we are successful in getting really close to the carina, but it's not right flush with the carina.
DR MILLER: And were your two BPFs [bronchopleural fistula], were those both left-sided?
DR NWOGU: They occurred 9 and 11 months after their resections, respectively. One was left-sided.
DR MILLER: Which would make—if you had a long stump, then that one—
DR NWOGU: Okay. The second one was right-sided.
DR MILLER: Yeah, yeah, yeah.
DR D'AMICO: Just a quick comment to clarify the question about atrial fibrillation after lobectomy. We presented our work at the Western Thoracic Surgical Association in June, so it should be in the literature soon. We used a propensity-matched analysis, which controls for all preoperative variables and stage.
In VATS lobectomy patients, the incidence of atrial fibrillation is 9%. In open lobectomies, it's 22%, statistically different.
And at the General Thoracic Surgical Club, we'll present our data on the over 70 group, which is similar and also statistically significant.
DR NWOGU: I appreciate that. As expected, in our experience there is a higher incidence of atrial fibrillation in pneumonectomies versus lobectomies. That will also be reflected as we do VATS pneumonectomies. Those patients have a higher risk for atrial fibrillation.
|
References |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  T. L. Demmy, I. E. Platis, C. Nwogu, and S. Yendamuri Thoracoscopic Extrapleural Pneumonectomy for Mesothelioma Ann. Thorac. Surg., February 1, 2011; 91(2): 616 - 618. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. E. Nwogu, S. Yendamuri, and T. L. Demmy Does Thoracoscopic Pneumonectomy for Lung Cancer Affect Survival? Ann. Thorac. Surg., June 1, 2010; 89(6): S2102 - S2106. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. M. Rueth and R. S. Andrade Is VATS Lobectomy Better: Perioperatively, Biologically and Oncologically? Ann. Thorac. Surg., June 1, 2010; 89(6): S2107 - S2111. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. L. Demmy, C. E. Nwogu, and S. Yendamuri Thoracoscopic Chest Wall Resection: What Is Its Role? Ann. Thorac. Surg., June 1, 2010; 89(6): S2142 - S2145. [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 |
