| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
General Thoracic Surgery Division, Massachusetts General Hospital, and Department of Surgery, Harvard Medical School, Boston, Massachusetts
Accepted for publication July 28, 2009.
* Address correspondence to Dr Wright, Department of Thoracic Surgery, Massachusetts General Hospital, 32 Fruit St, Blake 1570, Boston, MA 02114 (Email: cdwright{at}partners.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: This study is a retrospective review of 196 patients who underwent sleeve lobectomy. One hundred twenty-five patients had non–small cell lung cancer. There were 117 men (59.7%) and 79 women (40.3%) with a mean age of 54 years. Sixteen patients (13%) received neoadjuvant therapy. Fifty-six patients with N1 disease underwent sleeve lobectomy.
Results: There were 4 (2.0%) postoperative deaths. The postoperative morbidity rate was 36.7%. Four patients (2.0%) experienced bronchopleural fistulas. Multivariate analysis demonstrated that age older than 70 years (p = 0.02) and the diagnosis of non–small cell lung cancer (p = 0.0002) were risk factors for postoperative complications. Multivariate analysis also demonstrated that neoadjuvant therapy predicted anastomotic complications (p = 0.01). For non–small cell lung cancer patients, the 5-year survival rate was 44%. The 5-year survival rates for patients with pathologic N0 disease and N1 disease were 52.6% versus 39.3%, respectively (p = 0.205).
Conclusions: Sleeve lobectomy can be performed with minimal bronchial anastomotic complications and low postoperative mortality. In our study, neoadjuvant therapy for non–small cell lung cancer adversely influenced the rate of anastomotic complications. Performing sleeve lobectomy for patients with N1 disease was not associated with decreased overall survival rates.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
Despite the demonstrated efficacy of sleeve lobectomy, many concerns linger because of the potential for serious bronchial anastomotic complications such as bronchopleural fistula, which could result in fatal hemoptysis or sepsis. Many factors have been shown to influence the incidence of bronchial anastomotic complications, including nodal metastases, smoking, cancer histology, and positive bronchial margins [6, 7]. Neoadjuvant therapy, however, has not been associated with increased rates of bronchial anastomotic complications in recent series [8–10]. The efficacy of sleeve lobectomy for N1 disease remains controversial [11]. In a report by Okada and colleagues [3], nodal status was the most significant factor affecting survival. Gaissert and colleagues [4] also concluded that N1 disease was associated with lower overall survival among patients undergoing sleeve lobectomy.
The purpose of this study is to report our experience with sleeve lobectomy for NSCLC and low-grade neoplasms. We determined the operative morbidity, mortality, and long-term survival of sleeve lobectomy. In addition, we examined the effect of neoadjuvant therapy on the rate of bronchial anastomotic complications and the impact of N1 disease on the overall survival rate in patients with NSCLC.
|
Material and Methods |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
Preoperative Evaluation
The preoperative evaluation of patients undergoing sleeve lobectomy consisted of physical examination, chest radiography, computed tomography of the chest and upper abdomen, pulmonary function tests and electrocardiography. Bone scans, computed tomography or magnetic resonance imaging of the brain, and positron emission tomography were performed in selected cases. Quantitative ventilation and perfusion scans were performed in 108 patients (55%). Patients with a percent predicted forced expiratory volume in the first second (FEV1) less than 55% or a postoperative predicted FEV1 of less than 800 mL/s were considered to have compromised pulmonary function. Thirty patients in our series had compromised pulmonary function; therefore, they were not considered as candidates for pneumonectomy. Patients with a history of coronary artery disease underwent echocardiography, thallium stress testing, and selective coronary angiography. At our institution, mediastinoscopy is performed routinely in NSCLC cases to biopsy mediastinal lymph node stations 4R, 4L, and 7.
Surgical Technique
Our operative technique for sleeve lobectomy has been described in previous reports [11, 12]. Rigid and flexible bronchoscopy was performed to evaluate the airway anatomy and tumor infiltration. Most patients underwent double-lumen endotracheal intubation and standard posterolateral thoracotomy. The resectability of the tumors was determined before any structures were divided, and standard lobectomy was performed until the bronchus was encountered. The bronchus was completely divided with a sharp knife, and the proximal and distal margins were sent for frozen-section evaluation. The bronchial blood supply was preserved as much as possible to ensure adequate bronchial healing at the anastomosis. Incising the inferior pulmonary ligament was performed routinely to minimize tension on the anastomosis. If tension remains an issue, a U-shaped incision in the pericardium below the inferior pulmonary vein (hilar release) is usually performed. The anastomosis was performed in an open manner to facilitate precise placement of the sutures. The anastomotic sutures of interrupted 4-0 Vicryl (Ethicon, Somerville, NJ) were placed sequentially with the knots outside of the bronchial lumen. The bronchial anastomosis is typically wrapped with either pedicled pleura or pericardial fat. In cases involving preoperative chemotherapy or radiation, a pedicled intercostal muscle flap or omentum is used to cover the bronchial anastomosis. Table 1
lists the bronchial anastomosis coverage techniques that were used in this study. Routine bronchoscopy was performed to assess the healing of the bronchial anastomosis before discharge.
|
2 test. Multivariate analysis of categorical variables was performed using forward multiple logistic regressions. Survival rates were calculated by life-table analysis. Kaplan-Meier curves were plotted and compared using the log-rank test. Independent prognostic variables for overall survival were assessed by using the Cox proportional hazards stepwise model. Statistical analysis was performed using MedCalc for Windows, version 9.5.0.0. (MedCalc Software, Mariakerke, Belgium). Differences were considered significant when the probability was less than 0.05. |
Results |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
One hundred twenty-five patients were diagnosed with NSCLC, and there were 6 patients with other malignancies. Table 2 summarizes the histologic diagnoses of NSCLC and secondary lung cancer. The pathologic stage distribution of the patients diagnosed with NSCLC is listed in Table 3. Based on the TNM staging system for NSCLC [13], 48 patients (38.4%) were stage I (1 IA and 47 IB), 48 patients (38.4%) were stage II (3 IIA and 45 IIB), 24 patients (19%) were stage III (18 IIIA and 6 IIIB), and 1 patient (0.8%) presented with stage IV NSCLC (resected synchronous brain metastasis). Four patients who received neoadjuvant therapy before sleeve lobectomy had a complete pathologic response on final histologic examination. In terms of N status, 54 patients (44.7%) were N0, 56 patients (46.3%) were N1, and 11 patients (9%) were N2. In terms of T status, 4 patients (3%) were T1, 97 patients (80%) were T2, 13 patients (11%) were T3, and 7 patients (6%) were T4. A total of 65 patients were diagnosed with low-grade neoplasms. Postoperative histologic examination demonstrated that 52 patients were diagnosed with carcinoid (80%), 7 patients (10%) with mucoepidermoid carcinomas, 1 patient (1.5%) with cystadenoma, 1 patient (1.5%) with leiomyoma, 2 patients (3%) with histiocytomas, and 2 patients (3%) with hamartomas.
|
|
Type of Lung Resection
Table 4
shows the frequency of the different types of lung resections that were performed in the study. Right-sided sleeve lobectomies (65%) were performed more frequently than left-sided resections (35%). Upper lobectomies (64%) were performed more frequently than any other lobectomy. Table 5
shows the associated procedures that were performed with sleeve lobectomy. There was 1 patient who required a pulmonary artery sleeve resection in addition to a sleeve lobectomy. An additional 5 patients underwent tangential pulmonary artery resections.
|
|
|
|
|
|
|
|
|
Recurrence in Non–Small Cell Lung Carcinoma
The median and mean recurrence-free interval for patients with NSCLC (n = 123) was 33 months and 60 months, respectively (range, 1 to 324 months). Forty-three patients (34.4%) had a distant recurrence, and 22 patients (17.9%) had locoregional recurrences. Six of the locoregional recurrences involved the bronchial anastomosis, and 11 involved the lung or mediastinum. None of the 6 patients with bronchial anastomotic recurrences underwent a completion pneumonectomy owing to poor pulmonary function, and they were treated with palliative radiation therapy. One patient required a completion pneumonectomy for a recurrence in the ipsilateral lung, which could have represented a second primary lung cancer.
Pulmonary Function Tests
The mean preoperative FEV1 was 2.28 L ± 0.81 L (range, 0.9 to 5.95 L) and the mean percent predicted FEV1 was 71.9% ± 19.5% (range, 27% to 124%). Thirty-five patients underwent postoperative pulmonary function testing (range, 3 months to 2 years) for FEV1, which allowed for direct comparison with the preoperative FEV1 values. Figure 5
demonstrates the box-whisker plot comparing the mean preoperative FEV1 with the mean postoperative FEV1 in the same cohort of 35 patients who underwent sleeve lobectomy. The mean preoperative FEV1 was 2.54 L, and the mean postoperative FEV1 was 2.22 L. The preoperative and postoperative FEV1 values were very similar, and there was no significant difference (p = 0.081).
|
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
In our study, we reported an operative mortality of 2.0%, which is well within the range of other recent reports. Deslauriers and colleagues [2] reported an operative mortality rate of 1.3% for sleeve lobectomy compared with 5.3% for pneumonectomy in a series of 1,230 patients. Yildizeli and colleagues [15] reported an operative mortality of 4.1% in a series of 218 patients who underwent sleeve lobectomy for NSCLC. Tedder and colleagues [16] reported an operative mortality rate of 5.5% in a review of 1,915 patients who underwent sleeve lobectomy for malignancy. None of the 30-day mortalities in our study were the direct result of bronchial anastomotic complications.
The major postoperative complications that have been reported for sleeve lobectomy include pneumonia, bronchial complications, and atelectasis requiring bronchoscopy [6, 15]. We reported an overall complication rate of 34.6%, and postoperative pneumonia was the most frequent complication at 9.0%. The clinical definition of pneumonia varies depending on the report [7]; however, we defined pneumonia as a new significant infiltrate on chest radiography and a positive sputum culture. The major independent risk factors that predicted postoperative complications were an age older than 70 years and a diagnosis of NSCLC based on multivariate analysis. Compromised pulmonary function was not a significant risk factor for complications in our study as determined by multivariate analysis. Hollaus and colleagues [7] and Yildizeli and colleagues [15] demonstrated that compromised pulmonary function does predict an increased overall complication rate, which contradicts the findings in our study.
Bronchial anastomotic complications are probably the most serious complications associated with sleeve lobectomy. Anastomotic dehiscence and bronchovascular fistula often require completion pneumonectomy and can result in postoperative death [15]. We reported a bronchopleural fistula rate of 2.0%, and there were no bronchovascular fistulas in our series. Yildizeli and colleagues [15] reported a bronchial complication rate of 6.4%, which included bronchopleural fistula, bronchovascular fistula, local necrosis, and bronchial rupture. Ludwig and colleagues [17] reported a bronchial leakage rate of 6.9%, and Yatsuyanagi and colleagues [18] reported an anastomotic dehiscence rate of 8.5% and a stenosis rate of 8.5% in a sleeve lobectomy series for NSCLC. Our low rate of bronchial anastomotic complications is a direct result of strict adherence to the basic principles of bronchial anastomotic technique. We routinely avoid tension on the anastomosis by performing a hilar release maneuver whenever necessary and incising the inferior pulmonary ligament to provide added length. In addition, precise dissection, preservation of bronchial blood supply, and the placement of vascularized tissue between the bronchial anastomosis and the pulmonary artery are important principles in bronchial anastomotic technique.
Many clinical risk factors, such as positive resection margins, metastases to N2 lymph nodes, and compromised pulmonary function have been implicated in the formation of bronchial anastomotic complications [7, 18]. In our report, metastases to N2 nodes, compromised pulmonary function, age, side of operation, and smoking are not independent risk factors for the formation of bronchial anastomotic complications. Based on univariate and multivariate analysis, neoadjuvant therapy was a significant risk factor for the formation of bronchial anastomotic complications. Yamamoto and colleagues [19] demonstrated that preoperative chemoradiation decreased bronchial mucosal blood flow and compromised bronchial healing more than preoperative chemotherapy given alone. In contrast to our report, there are recent studies demonstrating that preoperative chemoradiation does not increase the rate of bronchial anastomotic complications. Burfiend and colleagues [8] reported a series of 19 patients who received neoadjuvant therapy before sleeve lobectomy in which the overall bronchopleural fistula rate was 1.3% and the neoadjuvant therapy group had a rate of 5.2%. Ohta and colleagues [10] reported a series of 20 patients receiving neoadjuvant therapy before sleeve lobectomy and did not find a significant increase in bronchial anastomotic complications. Given the small numbers of patients receiving neoadjuvant therapy in our retrospective report and other reports, the true implication of neoadjuvant therapy on bronchial complications may not be known, and only prospective, randomized studies can resolve the controversy. Despite the contrast in results of the clinical impact of preoperative chemoradiation on bronchial complications, radiation therapy has been shown to damage the bronchial microcirculation and impair bronchial healing after surgery [19]. In addition, Martin and colleagues [20] demonstrated that neoadjuvant therapy significantly increased mortality for patients undergoing right pneumonectomy. The authors of this report believe that neoadjuvant therapy, particularly with radiotherapy, significantly increases the risk of bronchial complications; therefore, the bronchial anastomosis should be covered with well-vascularized tissue, such as an intercostal muscle flap or omentum in cases involving neoadjuvant therapy. The authors believe that sleeve lobectomy can still be performed in patients who receive neoadjuvant therapy.
The long-term survival of NSCLC patients undergoing sleeve lobectomy has proven to be similar to or better than long-term survival of pneumonectomy in most recent reports [1–4]. In our study, we reported a 5-year survival rate of 44% for patients undergoing sleeve lobectomy for NSCLC. The long-term survival achieved in our report is slightly lower than the survival rates reported by Yildizeli and colleagues [15] and Fadel and colleagues [6]. In addition, Kaplan-Meier survival curves for N stage did not demonstrate significantly worse survival for patients with N1 disease compared with N0 disease, and Cox proportional-hazards regression analysis demonstrated that stage III NSCLC was predictive of a worse survival. N1 and N2 nodal involvement was not a significant predictor of worse survival in our study. This finding contradicts recent reports that demonstrated nodal metastases predicted worse survival in patients undergoing sleeve lobectomy for NSCLC [3, 4, 21, 22]. Van Schil and colleagues [21] reported that patients with N1 and N2 disease had significantly lower long-term survival after undergoing sleeve lobectomy for NSCLC. Similarly, Okada and colleagues [3] demonstrated that patients with metastases to N2 nodes had significantly worse survival compared with patients with N0 or N1 nodal status. In our study, the insignificant impact of nodal metastases on overall survival may be attributed to the inherent selection bias of a retrospective review of a large surgical series. In addition, the study may not have captured enough patients to demonstrate a statistically significant difference in overall survival between N0 and N1 patients.
The rate of locoregional recurrence is another controversial topic involving sleeve lobectomy. The rate of local recurrence for sleeve lobectomy ranges from 5% to 51% [15]. In our study, we reported a locoregional recurrence rate of 17.9%, which is well within the range of other recent reports. Deslauriers and colleagues [2] reported a locoregional recurrence rate of 22%, and Ferguson and colleagues [1] reported a locoregional recurrence rate of 20% in a meta-analysis of sleeve lobectomy. Kim and colleagues [14] demonstrated that patients with N1-positive lymph nodes, who underwent sleeve lobectomy, had a higher rate of locoregional recurrence.
In conclusion, we demonstrated that sleeve lobectomy can be performed with minimal morbidity and mortality. The rate of postoperative complications was significantly increased in patients with NSCLC and in patients 70 years of age or older. The rate of bronchopleural fistula was increased in patients who underwent neoadjuvant therapy based on univariate and multivariate analysis. In our study, overall survival after sleeve lobectomy in NSCLC was comparable to most recent reports, and N1 disease did not predict a worse overall survival rate.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
DR MERRITT: The majority of the stage III patients were N2 by mediastinoscopy, and there was a small cohort of stage III patients that were T3 N1. So those were the two groups that comprised the stage III group.
DR SWANSON: And did all the N2 patients have induction therapy?
DR MERRITT: Ten of the 11 N2 patients received induction therapy, which consisted of cisplatin-based chemotherapy and concurrent radiation therapy.
DR M MICHAEL S. MULLIGAN (Seattle, WA): Can you postulate why N1 disease had no significant effect statistically on survival? That's kind of surprising.
DR MERRITT: Yes, our observation that N1 disease did not result in worse overall survival was counterintuitive. In previous studies involving sleeve lobectomy, it was clearly demonstrated that patients with N1 disease had increased postoperative local recurrence rates and worse overall survival.
In our study, we concluded that N1 disease did not impact local recurrence or overall survival, and I think it's just the fact that we didn't have enough patients to show a statistical difference in survival. There was a trend towards better overall survival in patients with N0 disease, but it just wasn't statistically significant.
DR SETH D. FORCE (Atlanta, GA): Did your patients that received neoadjuvant therapy, did they get neoadjuvant chemo and radiation therapy? And if so, kind of similar to what we do with pneumonectomies now, are you going to switch to just chemo neoadjuvant given the higher complication rates?
DR MERRITT: In our series, 14 of the 16 induction therapy patients received combined chemotherapy and radiation, which consisted of cisplatin-based chemotherapy with two to three cycles. The average dose of concurrent radiation therapy was 45 Gy.
Since we did see an adverse effect on the anastomotic complication rate with induction therapy, we suggest that the surgeon adheres to the basic principles of performing a bronchial anastomosis. The bronchus should be covered with either an intercostal muscle flap or omentum, and there should be meticulous anastomotic technique to minimize the risk of postoperative anastomotic complications.
But we don't advocate against N2 patients who receive neoadjuvant chemoradiation therapy from having a sleeve lobectomy. We think that chemotherapy and radiation is probably more effective than just chemotherapy alone in the neoadjuvant setting for stage III NSCLC.
DR KALLIOPI ATHANASSIADI (Athens, Greece): Congratulations for your series. My question is, during a sleeve lobectomy, for instance, sometimes you need to do also sleeve for the artery.
So in these cases, in our series, we had a problem with thrombosis of the artery after neoadjuvant chemotherapy. We had no problem with bronchial anastomosis but only thrombosis and sometimes after 14 days later.
And one comment. In your abstract, you state that in 64% of your cases it's squamous cell carcinoma. Are you not afraid of local recurrence by doing sleeve? Some years ago we learned that when we have squamous carcinomas, try to avoid the sleeve resection. Thank you very much.
DR MERRITT: Yes, thank you.
Before we perform our bronchial anastomosis, we routinely check the bronchial margin, both proximal and distal by frozen section, to confirm that there is no tumor at the site where we're going to perform the anastomosis. And if the frozen section is negative, then we will proceed with the anastomosis.
I think using this technique to confirm that there is no cancer at the bronchial margin has limited our postoperative recurrence rate at the bronchial anastomosis.
DR GIUSEPPE CARDILLO (Rome, Italy): Congratulations for your paper. My question is about the cutoff that you have chosen for the age, 70. Actually, we operate [on] very old patients, so I think that a 70-year-old cutoff seems to be too young for our patients.
So why don't you investigate a different cutoff, I mean, 75 or 80 years old, because now we are operating on patients that are very, very old.
DR MERRITT: We clearly found in our series that patients that were older than 70 years old did have a higher risk of having postoperative complications. A lot of these patients have comorbidities such as coronary artery disease, impaired pulmonary function, and diabetes. I think that these and other comorbidities contribute to the increased risk of postoperative complications in patients over 70 years of age.
DR JOSHUA R. SONETT (New York, NY): Excellent presentation and results. I have one question.
So it says 19 patients underwent sleeve lobectomy for suspicious N1 nodes. So are you recommending a sleeve lobectomy for anyone with N1 disease?
DR MERRITT: In our series, there were 19 cases where the surgeons encountered large suspicious nodes at the lobar division point, which were not pathologically confirmed to be N1 disease, but they were highly suspicious. Instead of performing a lymphadenectomy and potentially having a positive margin, those patients underwent sleeve lobectomy to minimize the risk of having a postoperative local recurrence.
DR SONETT: So this is more like a bulky N1 disease?
DR MERRITT: Correct.
|
References |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  E. Storelli, M. Tutic, P. Kestenholz, D. Schneiter, I. Opitz, S. Hillinger, and W. Weder Sleeve resections with unprotected bronchial anastomoses are safe even after neoadjuvant therapy Eur J Cardiothorac Surg, January 20, 2012; (2012) ezr291v1. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. E. Merritt, S. I. Reznik, M. C. DaSilva, D. J. Sugarbaker, R. I. Whyte, D. M. Donahue, C. D. Hoang, W. R. Smythe, and J. B. Shrager Benign Emptying of the Postpneumonectomy Space Ann. Thorac. Surg., September 1, 2011; 92(3): 1076 - 1082. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Martinod, D. M. Radu, K. Chouahnia, A. Seguin, A. Fialaire-Legendre, P.-Y. Brillet, M.-D. Destable, G. Sebbane, S. Beloucif, D. Valeyre, et al. Human Transplantation of a Biologic Airway Substitute in Conservative Lung Cancer Surgery Ann. Thorac. Surg., March 1, 2011; 91(3): 837 - 842. [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 |
