| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J Thorac Cardiovasc Surg 2007;134:1266-1272
© 2007 The American Association for Thoracic Surgery
General Thoracic Surgery |
a Thoracic Surgery Department, European Institute of Oncology, Milan, Italy
b Division of Epidemiology and Biostatistics, European Institute of Oncology, Milan, Italy
c University of Milan School of Medicine, Milan, Italy.
Presented at the Thirty-second Annual Meeting of the Western Thoracic Surgical Association, Sun Valley, Idaho, June 21-24, 2006.
* Corresponding author: Lorenzo Spaggiari, MD, PhD, Thoracic Surgery Department, European Institute of Oncology, Via Ripamonti 435, 20100 Milan, Italy, phone +39.02.57489665, fax +39.02.57489698. (Email: lorenzo.spaggiari{at}ieo.it).
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Methods: We have defined "extended" pneumonectomy (EP) as the removal of the entire lung, associated with one or more of the following structures: superior vena cava, tracheal carina, left atrium, aorta, chest wall, or diaphragm. Our clinical database was retrospectively reviewed to identify patients who underwent EP to assess their postoperative morbidity, mortality, and long-term survival.
Results: Between 1998 and 2005, 47 EPs were performed. The "extended" procedure included left atrium resection in 15 patients, combined SVC and carinal resection in 9 patients, aortic resection in 8 patients (in 3 patients with prosthetic replacement), chest wall or diaphragmatic resection in 6 patients, SVC resection in 4 patients, and carinal resection in 4 patients. A partial esophageal muscular resection was performed in 1 patient. Overall 60-day mortality was 8.5%. Major postoperative complications occurred in 8 patients (17%). The 2- and 5-year survival rates for the overall population were 42% and 22.8%, respectively. Interestingly, long-term survivors were recorded only in the group of patients who received induction treatment.
Conclusions: Extended pneumonectomy is a feasible procedure with an acceptable risk factor. To improve the selection of patients, all candidates should undergo preoperative mediastinoscopy and induction chemotherapy. In patients with positive response to chemotherapy or stable disease, extended pneumonectomy may afford a radical resection in more than 80% of cases and may result in a permanent cure in some instances.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Complete tumor removal is the objective of any surgical resection when treating lung cancer.1 A tumor may infiltrate contiguous structures such as the superior vena cava (SVC), tracheal carina, left atrium, diaphragm, or chest wall. It has been demonstrated that in all of these situations, a radical resection can be achieved by combining an "extended" procedure such as SVC prosthetic replacement, tracheal "sleeve," or left atrium resection with lung surgery, resulting in acceptable morbidity and mortality rates as well as satisfactory long-term results.2-5
Of all anatomic resections, pneumonectomy is associated with the highest postoperative mortality rate,6,7 particularly after induction treatment,8 and it is often considered a cause of illness in itself, because of its adverse impact on the quality of life of long-term patients. When both an extended procedure and pneumonectomy are required to radically resect a tumor, surgical risk increases and the certainty of oncologic benefits diminishes.
To clarify the role of the extended pneumonectomy in the management of lung cancer patients, we reviewed our patient data (a) to assess the additional risk of extended pneumonectomy as compared with that of standard pneumonectomy in terms of postoperative morbidity and mortality, and (b) to evaluate long-term survival after such an invasive approach.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The clinical database of the Thoracic Surgery Department of the European Institute of Oncology was reviewed to identify patients who underwent extended pneumonectomy between January 1998 and March 2005. Our Ethical Committee was informed of the study and did not require approval. All patients gave their informed consent for the study.
Preoperative Management
Preoperative work-up consisted of brain, chest, and upper abdomen enhanced CT scan, and bronchoscopic scan. From 2000 on, positron emission tomography was routinely used. Functional evaluation was performed by spirometry, blood gas analysis, and lung perfusion scan. A predicted postoperative FEV1 value of less than 30% was considered a contraindication to pneumonectomy.
Patients having one or more mediastinal lymph nodes with a diameter larger than 1 cm in their short axis on CT scan underwent mediastinoscopy. When N2 disease was detected, they underwent induction chemotherapy in three cycles of cisplatin and gemcitabine (cisplatinum 80 mg/m2 days 1,21 and gemcitabine 1250 mg/m2 days 1,8,21) and, in the case of tumor reduction or stable disease, they underwent surgery. In patients without mediastinal involvement, induction chemotherapy was discussed case by case. The presence of pleural effusion or the clinical suspicion of pleural disease was investigated by video-assisted thoracoscopy in all cases.
Restaging after chemotherapy was performed by brain, chest, and upper abdomen enhanced CT scan and bronchoscopy. The final decision on indication for surgery was taken after multidisciplinary discussion. The delay between the end of chemotherapy and surgery was 4 to 5 weeks.
Postoperative Complications
Postoperative death was defined as any death occurring during hospital stay, or within 30 days after surgery. Sixty-day death was defined as any death occurring within two months from the day of pneumonectomy.
Postoperative complications were classified as: (1) respiratory (acute respiratory failure, ARDS and ALI, as defined by the American European Consensus Conference on ARDS in 1994,9 pneumonia, sputum retention, pulmonary embolism, pulmonary oedema, chronic respiratory failure), (2) cardiac (cardiac arrythmia, angina, myocardial infarction, cardiogenic shock), (3) surgical (SVC thrombosis, hemothorax, bronchial fistula, empyema, chylothorax, cardiac dislocation), and (4) others. Respiratory, surgical, and cardiac events with the exception of cardiac arrythmia were considered to be major complications. All the others were defined as minor complications.
Intraoperative Management
Intraoperative management was focused on reducing the risk of damage to the controlateral lung; fluid administration was in the order of 5–7ml/kg/hour cristalloids infusion, not exceeding a total amount of 1500 ml in all cases.
In patients requiring SVC resection, intraoperative fluids and vasoactive agents administration was managed differently until SVC replacement was completed. The objective was to obtain a mean arterial pressure before clamping of 80 mmHg, to compensate for the expected drop in arterial pressure at SVC clamping, due to the reduction in cardiac output.10
Ventilation was managed using a protective-ventilation strategy (a tidal volume 
6ml/kg, driving pressure < 20 cm H20 above the PEEP value, permissive hypercapnia, and the preferential use of pressure-limited ventilatory modes).11
The bronchial stump was covered in all cases. Because extended pneumonectomy is often performed intrapericardially, the availabilty of autologous pericardium is often limited and for this reason it was rarely used. The preference was to cover the stump using a pedicled mediastinal fat pad. When it was not available, a pedicled parietal pleura flap was used.
Statistical Analysis
The impact of the following 10 variables on postoperative morbidity was verified (age, sex, induction treatment, preoperative FEV1%, side of pneumonectomy, carinal reconstruction, SVC resection, atrial resection, chest wall resection, and diaphragmatic resection).
Patients were divided into two groups based on the presence or lack of major postoperative complications and were compared for all relevant variables using frequency tables for categorical variables and summary statistics for continuous variables. Chi-square or Fisher’s exact test were applied when appropriate.
Postoperative complications were considered as outcome variables in a logistic regression model, using the defined covariates. The Odds-ratio and a corresponding 95% of CIs were reported for covariates considered clinically relevant or statistically significant at the 0.05 significance level (Wald chi-square test) and then included in the final multivariate model.
Overall survival and disease-free intervals were estimated from the date of surgery using the Kaplan–Meier survival analysis method. Survival comparisons by stage were analyzed by log–rank test; the difference was considered statistically significant when p-value was less than 0.05.
Follow-up information was obtained via telephone contact on October 2005.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
After surgery, postoperative complications occurred in 27 patients (57.4%). Major complications occurred in 8 patients (17%), most of them surgical: 3 early bronchopleural fistulas, 2 hemothoraces, and 1 cardiac herniation, due to pericardial prosthesis rupture. Five major respiratory complications were recorded (10.6%): respiratory failure in 3 cases, ARDS in 1 case, and pulmonary embolism in 1 case. Four patients required temporary tracheostomy for prolonged mechanical ventilation. Univariate analysis did not identify any risk factor for the occurrence of major postoperative complications.
Two postoperative deaths were recorded (overall mortality rate 4.2%) because of ARDS after re-thoracotomy for fistula in one case and simultaneous pulmonary embolism and tracheobronchial fistula in the other. The average ICU was 3 days (range 0–42), with the average hospital stay 10 days (range 5–60). Two subsequent deaths were recorded after discharge, within the 60-day margin: in 1 patient death was due to cardiac failure after left atrium resection, and in 1 patient sudden death occurred at home, with the cause remaining unclear.
A radical resection was obtained in 87% of cases. Final pathologic details are given in Table 1.
Complete follow-up was achieved in all but one patient. 47% of patients were still alive (22/47) at the mean follow-up time of 19 months. Eighteen (38%) were disease free, whereas four had experienced relapse. Out of the 25 recorded deaths, 14 were due to recurrent disease and 11 due to causes unrelated to cancer. In 1 case, the information was not available. Apart from the 4 patients who died within 60 days of surgery, 4 patients died because of respiratory causes, 1 of cardiac causes, 1 of sudden death.
The 2-year and 5-year survival rates for the overall population were 42% and 22.8%, respectively. The overall median survival time was 22 months (95% CI 18–36).
Regardless of the negligible statistical significance, two variables seem to affect long-term survival: lymph node status and type of extended resection. Patients who had mediastinal nodal metastases (N2) had a lower 5-year survival rate (41.2%) compared with those without mediastinal involvement (8.1%, p = 0.13, figure 1). Moreover, patients who underwent carinal resection had a better prognosis as compared with that of those who underwent left atrial resection (34% and 17%, respectively, p > 0.05). Pathologic T status did not affect long-term survival (22% and 26%, respectively, at 5 years for pT < 4 and pT4 tumors, p > 0.5).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
First, when dealing with patients with suspected mediastinal infiltration, staging of the tumor is extremely difficult because of radiologic inaccuracy.12,13 This inaccuracy can potentially translate into over-treatment in the cases of SVC, atrial, or carinal resection in "false T4" patients (patients with clinical T4 disease that is not confirmed by pathologic testing), which in our series was in the order of 15%. Moreover, the majority of these patients received chemotherapy before surgery, making the distinction between the response to chemotherapy and an erroneous tumor staging even more problematic.
Second, surgeons who consider clinical T4 tumors as inoperable by definition probably deny permanent cure to a certain proportion of patients. This attitude is not completely justified, given the fact that a T4 tumor can sometimes be radically removed by a tangential SVC14 or a partial left atrium resection,15 two procedures that do not increase morbidity and that should be in the repertoire of every thoracic surgeon.
The disadvantage of surgical exploration in clinical T4 patients is that the rate of exploratory thoracotomy is higher. During the study period, we performed 5 exploratory thoracotomies in cT4 patients, mainly due to infiltration of the aortic arch (4 cases), which means unresectable disease in almost 10% of cases. A more extensive use of VATS as a staging tool could probably reduce the rate of exploration in case of suspected aortic infiltration. Morbidity of exploratory thoracotomy when the tumor infiltrates the mediastinum is significant (60% in our experience), because operability can be diagnosed only after extended dissection and intrapericardial vessels isolation. In tumors infiltrating both SVC and the carina, the final judgment on the airways is possible only after SVC resection.
This type of surgery requires highly specialized centers, skilled surgeons, and skilled anesthesists, given the complexity of intraoperative and postoperative management. For example, pneumonectomy requires intraoperative fluid restriction, whereas, at the same time, SVC crossclamping requires generous fluid administration in order to overcome arterial tension drop and the risk of cerebral edema at SVC crossclamping.10 Anesthesists should be familiar with this problem to reduce the risk of neurologic consequences or early postoperative ARDS.
Once adequate standard of care is estabilished, maximum attention should be placed on properly selecting patients most likely to benefit from extended pneumonectomy. As previously reported, the most important predictor is mediastinal nodal status,16,17 the impact of which was suggested but not statistically confirmed in our study because of the small dimension of the population. Patients with persistent mediastinal nodal involvement after chemotherapy have a poor long-term prognosis and should not be considered as candidates for extended pneumonectomy. Given the fact that these tumors are close to the mediastinum, the PET scan is rarely useful for preoperative detection of N2 disease. Consequentially, mediastinoscopy should play a central role in preoperative staging. The best strategy for restaging after chemotherapy remains an open question.
Another debateble argument is the use of preoperative chemotherapy in patients requiring extended surgery. Even in the absence of concrete evidence from literature, we advocate the use of chemotherapy in T4 lung tumors for four main reasons. The first is the clinical observation that a significant proportion of patients who undergo extended surgery develop distant metastases that had not been evident at preoperative staging, after surgery. The second reason is the theoretical advantage of induction treatment (decreasing tumor size, increasing the likelihood of negative margin, sterilizing micro-metastatic disease, defining tumor response to chemotherapy), which may facilitate surgery and exclude patients with rapidly evolving disease. The third consideration is that a negative impact of induction chemotherapy on postoperative mortality has not been clearly demonstrated. Finally, the benefits of chemotherapy in terms of survival have been demonstrated in early stage,18 stage IIIa,19,20 and in stage IV.21 Why should T4 tumors represent an exception? The supposed survival advantage of induction chemotherapy was not evident in our series, and it was instead due to the bias of selection and to the limited dimension of the population. Only a prospective randomized trial could confirm the actual advantages provided by preoperative induction treatment. Unfortunately, its feasibility is limited by several factors, the first being the difficulty of precise clinical staging in T4 tumors.
In our experience, 60-day postoperative mortality after extended pneumonectomy is in the order of 10%, doubled as compared with that of standard pneumonectomy.22 This increased mortality rate is due to a higher rate of surgical and respiratory complications. In terms of surgical complications, the most dramatically negative event remains bronchopleural fistula, particularly in the case of carinal resection, because no effective salvage repair procedure exists. The increase of respiratory complications is probably linked to preoperative chemotherapy by the mean of Dlco impairement.23 It is advisable that all candidates for extended pneumonectomy are submitted to Dlco assessment before and after chemotherapy, because patients with a Dlco loss > 20% are probably at higher risk.
From an ethical point of view, is it acceptable to propose a procedure with a 10% postoperative mortality rate? It depends on the alternative, which is chemoradiotherapy with curative intent. Five-year survival after chemoradiotherapy in stage III NSCLC is 3–10%.24 Our reported surgical survival, comprising R+ patients, was 22%. There is no mean to define whether surgical treatment translates into better cure, but patients are given a chance of permanent cure, which is exceptional by the mean of the other treatment. This is what patients and surgeons want.
This study is limited by 3 factors: its retrospective design, the small number of cases, and the combination of T3 (chest wall and diaphragm) and T4 tumors. The first factor is common to all the other published studies, and it is probably counterbalanced by the fact that our series was collected over a relatively short period of time. Concerning the limited number of patients, as far as we know there are few studies published that focus on the combination of "extended" procedures and pneumonectomy,25 regardless of the type of extended procedure. As such, we believe that the information available from this series could be useful for further studies. Finally, the decision to consider chest wall or diaphragmatic resections as extended procedures was arbitrary but justified by their morbidity when associated with pneumonectomy. Two of the 4 patients who died within 60 days after surgery underwent such a type of procedure.
In conclusion, "extended" pneumonectomy is a feasible procedure with an acceptable risk. To facilitate an accurate selection of patients, all candidates should undergo preoperative mediastinoscopy (excluding from surgery patients with mediastinal nodal metastases) and induction chemotherapy. In patients with response to chemotherapy, or with stable disease, extended pneumonectomy may afford a radical resection in more than 80% of cases and may result in permanent cure.
| Earn CME credits at http://cme.ctsnetjournals.org
|
|
Acknowledgments |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Related Article
This article has been cited by other articles:
|
|
 |
|
  A. Toker, S. Tanju, Y. Ozluk, and S. Kaya Partial resection of the aorta and patch plasty with a simple clamping technique in a lung cancer patient Interact CardioVasc Thorac Surg, May 1, 2011; 12(5): 827 - 828. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Y. Perentes, C. C. Erling, H.-B. Ris, J.-M. Corpataux, and L. Magnusson A simple bypass technique for superior vena cava reconstruction Interact CardioVasc Thorac Surg, January 1, 2011; 12(1): 15 - 19. [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 |
