| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2007;83:1971-1977
© 2007 The Society of Thoracic Surgeons
a Division of Cardio-Thoracic Surgery, Department of Surgery, University of Alabama at Birmingham, Birmingham, Alabama
b Department of Epidemiology, School of Public Health, University of Alabama at Birmingham, Birmingham, Alabama
Accepted for publication January 22, 2007.
* Address correspondence to Dr Cerfolio, Division of Cardiothoracic Surgery, University of Alabama at Birmingham, 1900 University Blvd, THT 712, Birmingham, AL 35294 (Email: robert.cerfolio{at}ccc.uab.edu).
Presented at the Fifty-third Annual Meeting of the Southern Thoracic Surgical Association, Tucson, AZ, Nov 8–11, 2006.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
Methods: We retrospectively reviewed a prospective electronic database of patients who underwent pulmonary artery resection. The technique used was an R0 resection with end-to-end anastomosis only if needed, distal control of the pulmonary artery by clamping the vein (not the artery), and no postoperative anticoagulation.
Results: Between October 1998 and June 6, 2006, 42 (3.2%) of 1328 patients who underwent lobectomy performed by one surgeon required resection of the pulmonary artery (38 partial, 4 circumferential) to achieve a margin-negative resection and avoid pneumonectomy. Of these, 41 had non-small cell lung cancer, and 23 (55%) had neoadjuvant chemoradiotherapy (median dose of 60 Gy). Right upper lobectomy was performed in 2 patients and a left upper lobectomy in 40. A negative bronchial and vascular margin was achieved in all. Morbidity occurred in 11 patients (atrial fibrillation in 6) and left recurrent laryngeal neurapraxia in 2. Aspiration resulted in one operative death. Follow-up (median, 48 months) showed no local recurrence on the pulmonary artery and normal blood flow through it. Five-year survival was 60%.
Conclusions: Pulmonary artery resection and reconstruction to avoid pneumonectomy can be performed safely, even in a highly irradiated field. Clamping of the remaining pulmonary vein for distal control is safe and affords more room. Circumferential resection with end-to-end anastomosis of the pulmonary artery is rarely required. Partial resection is safe, does not impede blood flow, and does not compromise local recurrence rates. Postoperative anticoagulation is not needed.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
Several studies have shown the oncologic soundness of these types of procedure [3, 4]. Rendina and colleagues in 1999 [5] and 2000 [6] reported several specific intraoperative techniques during sleeve resection of the PA. Others, such as Jimenez and colleagues in 2006 [7], showed a lower morbidity and mortality rates in patients who underwent sleeve resection of the bronchus with those who underwent pneumonectomy. Rendina and colleagues [6, 8, 9] reported 40 patients who underwent bronchovascular sleeves, and 15 required complete resection with end-to-end anastomosis. They also recommended an intraoperative dose heparin of 3000 to 5000 U/day, followed by 10,000 to 15,000 units/day of subcutaneous heparin for 10 days [6].
We reviewed our experience with partial as well as complete PA resection for pulmonary lobectomy using similar techniques to Rendina and colleagues. We evaluated our results because we were more liberal in the use of partial resection, used a lower dose of intraoperative heparin, avoided the use of postoperative heparin, and performed the procedures in patients who had undergone high-dose irradiation.
|
Material and Methods |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
Staging
Patients were staged as previously reported [10]. Entry criteria for this study mandated a dedicated fluorodeoxyglucose-positron emission tomography (FDG-PET) scan and a computed tomography (CT) scan with intravenous contrast and 5-mm collimated slices no more than 1 month before resection or nodal biopsy. All suspicious sites were investigated. Suspicious sites on FDG-PET were defined as previously described, and appropriate tests were performed as previously described [10]. Endoscopic transesophageal ultrasound was used to biopsy suspicious posterior aortopulmonary window nodes (n = 5), subcarinal (n = 7), periesophageal (n = 8) and inferior pulmonary ligament nodes (n = 9), as previously described [11].
Patients who had biopsy-proven N2 disease underwent neoadjuvant radiochemotherapy as previously described [12]. In general, patients with N2 disease underwent neoadjuvant radiation and chemotherapy using doses of 60 Gy or higher of radiation, and patients with biopsy-proven N1 disease (with Wang needle biopsy during bronchoscopy) received neoadjuvant chemotherapy alone. All patients were restaged after neoadjuvant therapy, and resection was generally reserved for those who were down-staged as previously described [13].
At the time of right thoracotomy, all lymph nodes contained in the #2R, #4R, #7, #8, #9 stations were completely removed. At the time of left thoracotomy all lymph nodes contained in the #4L, #5, #6, #7, #8 and #9 lymph nodes stations were completely removed. Pathologic review was performed with standard techniques, and immunohistochemical staining was used when appropriate. The pathologic stage was assessed using the international staging system [14].
Definitions
For a procedure to be labeled as "a pulmonary artery resection" the following was required: (1) proximal and distal control (by clamping the remaining pulmonary vein of the lobe[s] that is not resected) of the PA was obtained, (2) some part of the right or left PA was removed using a knife or scissors, and (3) the artery was reconstructed to reestablish blood flow to the remaining pulmonary parenchyma. Procedures that used proximal and distal control to "just safely" tie the PA branches were not included in this study. Operative morbidity was defined as any mortality within 30 days of surgery or during the same hospital admission.
Surgical Technique for Pulmonary Artery Resection
When safe conventional dissection of the pulmonary arterial branches can not be performed and yet lobectomy could yield complete resection of the cancer, the following technique is used. Proximal control of the main PA is obtained extrapericardially. On the right side, proximal control is obtained posterior to the superior vena cava; and on the left side, it is obtained just distal to the ligamentous arteriosis. Care is taken to avoid the left recurrent laryngeal nerve.
The pulmonary vein of the lobe to be removed is encircled and then divided. We prefer using a vascular stapler. Then 1500 U (we have not based the dose of heparin on the patient’s weight) of intravenous heparin is given. Approximately 1 minute later, a Satinsky clamp is placed on the proximal PA and positioned so that its handle is facing in the opposite direction of the Satinsky to be placed on the vein (as shown in Figure 1). This allows more room for the operating surgeon.
|
A knife is used to cut out the part of the PA that has cancer invading its surface, and then the bronchus is cut as well. The large lobe with the cancer is removed from the operative field. This opens up the entire surgical field and affords more room for the surgeon to work. At this time, separate frozen section margins of the bronchus and the PA should be sent. If sleeves of the bronchus and of the artery are both required, we prefer to perform the bronchus first, before the PA is sewn back together. Rendina and colleagues prefer the artery first, which decreases the clamp time. We prefer the bronchus first, even though it prolongs the clamp time, so a freshly sewn PA does not have to be retracted to perform a left upper lobe bronchial sleeve resection.
The PA is now examined from inside to determine if a partial or circumferential resection is needed. If removal of part of the PA narrows the lumen by more than 20% to 30% of its original size, than we prefer a circumferential resection. Another option is a patch angioplasty using a piece of bovine pericardium. If a circumferential resection is required the PA is sewn back together using 5–0 Prolene (Ethicon, Somerville, NJ), as shown in Figure 2.
|
|
Follow-Up
Patients were followed up after resection, which consisted of chest and abdominal CT every 6 months for the first 2 years and yearly afterwards. FDG-PET was used in some patients based on insurance coverage and availability. Data were obtained from multiple sources, including clinic letters, follow-up scans, hospital computer information systems, tumor registry, Social Security Death Index, telephone calls, and letters from oncologists and other physicians. The time to disease recurrence (defined as biopsy-proven recurrent NSCLC), progression, or death from lung cancer was recorded. All information was entered into our prospective database.
|
Results |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
The indication for surgery was NSCLC in 41 (98%) of the 42 patients and for tuberculosis with a destroyed left upper lobe in 1 patient. Table 1 summarizes patient characteristics, including history of neoadjuvant therapy and the type of procedure performed. PA resections were done in 42 patients, of whom 4 required a circumferential resection and 7 required bovine pericardium patch angioplasty. Six patients had a concomitant sleeve of the both the artery and the bronchus. An intercostal muscle was applied to the bronchial closure in 32 of the 42 patients, and in some, it acted as a buffer in between the vascular repair and the bronchial closure. Table 2 summarizes the histopathology of these 42 patients. Of interest was that a significantly greater number of patients had squamous cell carcinoma (64%) compared with other types of NSCLC.
|
|
|
|
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
Lobectomy offers less operative risk and improved postoperative quality of life. Oncologic surgical rules mandate that all cancer must be removed and all margins cleared. Once in the chest and faced with a large central tumor or nodes stuck to the PA, the surgeon must resist the temptation to opt for the less-time-consuming and easier-to-perform pneumonectomy. A sleeve lobectomy, even in a highly irradiated field, is safe. Often times the only way to know if a sleeve is possible is to try. This can, however, take a significant amount of time and effort, especially if one does not have proximal and distal control of the artery. We searched for a way to technically speed the operation without compromising patient safety or violating oncologic principles. Initially, we were hesitant to clamp a highly radiated main PA; however, as shown in this report, it appears to be a very safe maneuver.
In our study we found it much more likely to require this inflow occlusive technique for a left upper lobectomy than for a right (40 on the left, only 2 on the right). In the 1999 article by Rendina and colleagues [5], PA resection was performed on the right in only 12 patients and on the left in 40. So it seems to be more commonly required on the left side. Another interesting similarity between their report and our reports is that 33 (63%) of the 52 patients in their series had squamous cells tumors and only 15 (29%) had adenocarcinoma. Similarly, 27 (64%) patients in our report had squamous cell lesion and 8 (19%) were adenocarcinoma. Only 3 of their 52 patients underwent preoperative radiation, however. This was one reason for our initial trepidation of a clamp injury in the irradiated PA. In our series, 17 patients had preoperative radiation and the median dose was 60 Gy; in fact, 14 of the 17 patients had doses of 60 Gy or higher. Thus, this technique does seem safe in a highly irradiated chest.
Standing on the shoulders of Rendina and colleagues’ expertise and their outstanding report, as well as the experience of many other outstanding surgeons, we modified these techniques just a little. First, we noticed that most NSCLC did not seem to involve the entire PA, but rather only the anterior aspect of part of the PA. Thus, we did not see the need for complete pulmonary resection with end-to-end anastomosis in most patients. Our goal was to achieve a negative margin, and we did not specify the extent of that margin, only that it was negative. Circumferential resection was only required in 4 patients (9.5%) in our series to achieve a negative margin compared with 19 (29%) of 65 patients in the Rendina and colleagues report [6].
We first applied this PA inflow and outflow occlusion technique to left upper lobectomies that did not necessarily require PA resection but presented a difficult dissection of the PA branches. We found it unnecessary to use much if any intraoperative heparin and did not use any heparin or even aspirin postoperatively.
We then applied the technique to patients who required PA resection. We preferred clamping of the inferior pulmonary vein as a way to obtain distal control after the superior pulmonary vein was ligated. This technique, which was described in length and detail in our "Methods" section, affords more room for one to work on and inside the PA. It is easier to assess the precise extent of the disease and it is much easier to assess the tension of the PA when a complete 360-degree PA resection is needed with end-to-end anastomosis. It avoids placing a clamp on the more distal PA, which is smaller and more fragile than its more proximal component. This also avoids ripping this thin-walled irradiated vessel distal to the suture line.
In our earlier experience, we noticed oozing from the lymph node basin, especially from the #5, #6, and #7 stations after the intravenous heparin was given. Because we do a complete thoracic lymphadenectomy before resection (often to ensure there is no residual or recalcitrant N2 disease) this was problematic. We postulated that perhaps a smaller does than 5000 U might be as efficacious for preventing clots in remaining PAs but would have less bleeding. We have reduced our dose to 1500 U now for most patients, and it appears to work well. The complete elimination of heparin may very well be safe. No further heparin is needed. We have avoided heparin reversal with protamine as well and have not used any anticoagulation postoperatively. Patients wear self-compressive leg device to help prevent deep venous thrombosis and pulmonary embolism.
We studied 14 of the first 16 patients who underwent partial pulmonary artery resection using intravenous CT scan within 6 weeks of surgery. None of these patients had PA clots or obstruction to their PA blood flow through the narrowed, partially resected PA. The remaining patients all had intravenous CT scans at 6 months, and there was no evidence of lack of pulmonary arterial flow. We thus hypothesized that this primary closure technique was safe. From an oncologic standpoint, our results suggest that local recurrence is not higher if the PA is not circumferentially transected as long as frozen section analysis assures negative margins of the PA.
The decision of when to perform a partial PA resection compared with when to perform a patch pulmonary arterioplasty or a complete resection is determined by the amount of PA wall involved with tumor. If more than one fourth of the anterior wall has to be resected to achieve an R0 resection of the PA, we favor patching or a complete resection and then performing an end-to-end anastomosis. The decision about when to perform a patch pulmonary arterioplasty instead of a complete resection with end-to-end anastomosis is arbitrary, however. The advantage of the patch is that it does not leave a posterior suture line on the artery, which can rub against the sutures or staples used to close the left upper lobe bronchus. Although we almost always place an intercostal muscle over the bronchial closure, the intercostal muscle must run under (posterior to) the artery, and if it is bulky, it can lift or elevate the artery to a higher than desired position.
In conclusion, this surgical technique of PA resection is safe and effective and can be performed in a patient that has undergone more than 60 Gy of preoperative radiation to the chest. Partial resection of the PA is safe and is all that is needed in most patients to achieve an R0 resection. Patch angioplasty using bovine pericardium is another option. Circumferential PA resection with end-to-end anastomosis can be reserved for more extensive lesions. Clamping the remaining pulmonary vein instead of the PA distally is safe in a highly irradiated chest and affords more room for the surgeon to work. This technique does not lead to compromised PA blood-flow and does not seem to compromise the local recurrence rate or long-term survival. Because we have gained comfort and confidence in this technique, we now apply it quicker then we did before, especially for left upper lobectomy. Further refinements and corroboration from other centers are needed.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
At the present time bronchial and vascular reconstructive procedures are performed with excellent results in many centers around the world, and Dr Cerfolio has modified and refined some of these techniques and has now reported on his results using these modified techniques in 42 patients over the last 8 years. I have several questions.
Recognizing that the decision to perform a pulmonary artery resection or reconstruction is almost always made at the time of surgery based on intraoperative findings, have you found preoperative imaging studies such as pulmonary arteriograms, MR angiograms, or CT angiograms useful in surgical planning of these procedures?
Two, an important aspect of your technique is the use of very low heparin doses prior to clamping the pulmonary artery and the elimination of any anticoagulation postoperatively. At many lung transplant centers, heparin is not routinely given when performing lung transplantation when cardiopulmonary bypass is not required. Do you think at some point you will completely eliminate the use of all systemic anticoagulation?
Number three, in your manuscript you mentioned that you perform a complete thoracic lymphadenectomy prior to resection. Do you obtain frozen section analysis of all of these N2 nodal stations prior to proceeding on with the decision to perform a PA reconstruction, and do you consider N2 nodal involvement a relative contraindication to proceeding with a PA reconstruction? What is your practice when you encounter residual N2 disease in patients who have been treated with induction therapy or you find unexpected N2 disease at the time of surgery?
Next, Dr Rendina’s group uses low-dose steroids postoperatively as well as inhaled steroids preoperatively as a way of reducing swelling around the bronchial anastomosis and reducing airway secretions. In cases where you are doing both bronchoplastic and angioplastic procedures, have you incorporated that into your practice?
And finally, did any of the patients in your series require use of a prosthetic conduit for reconstruction after PA resection? What technique would you recommend if an end-to-end anastomotic technique could not be done because of excessive tension? Specifically, would you modify your strategy as far as anticoagulation was concerned if a prosthetic conduit was required?
I would like to thank the Association for the opportunity to discuss this paper and wish to congratulate you on your outstanding results in your efforts to further refine and advance these important lung-sparing techniques.
DR CERFOLIO: Dr Shen, thank you very much. You have given me five questions. I will answer them in order.
The first question is about the use of a preoperative imaging modality. The answer is no. We use CT scan with 5-mm cuts and integrated PET-CT in everybody, but I have not used MRI or CT angiogram. I don’t know if it would change my management. I will use MRI for a T status T4 into vertebral a body or a Pancoast into the brachial plexus, then we do use MRI.
Your second question was how we have lowered the dose of heparin and do we think we could eliminate it. There is one patient we didn’t give any heparin to and that patient did fine. And I think you are right. I think because the standard of care is to give some heparin, I have lowered the dose, but I think one could easily get away with using no heparin intraoperatively, and I am not talking about just irrigating distally with heparin but completely eliminating heparin altogether. Although I don’t have data to support that, that is my opinion—but beware of the surgeon with opinions without supporting data.
Number three, what about lymph nodes? As you know, we have a real interest in patients with N2 disease and we favor preoperative chemoradiotherapy. In general, if patients have residual or recalcitrant N2 disease, we do not proceed with resection. However, most times you can remove the subcarinal, the No. 7, the 8, and the 9, and the 2s and 4s— but often those 5s and 6s are just plastered sometimes to the PA, and you don’t know until you have clamped it proximally and cut part of it out. At that point I would proceed with resection. I think we had one person in the series that had that. I have been impressed with the high sterility or CR rate with the high dose radiation. So in a young patient with low comorbidities that I think I can resect, I now finally feel pretty comfortable that I can do this technique safely. I may proceed with lobectomy even with recalcitrant N2, but in general the answer would be we would stop with recalcitrant N2.
Your fourth question was about the use of steroids, whether we use inhaled steroids preoperatively. The answer is no. We presented a paper at the STS about giving intraoperative high-dose steroids prior to clamping the PA for pneumonectomy, to help prevent postpneumonectomy pulmonary edema, and we still do this on every pneumonectomy, but we have not used inhaled steroid. And since some of these patients are getting sleeve or bronchoplastic procedures, I would argue not to use the steroids. We do use an intercostal muscle flap to buttress all those patients and to put it in between the PA and the bronchus.
Your fifth question was have I used any prosthetics in this series to bridge a gap. The answer is no, we may use a patch but we have not used a cylindrical tube (ie, a homograft). The offshoot of that question was what if I couldn’t get the ends together. Well, you are in trouble then, but I think a pulmonary artery homograft would be an option. You could, of course, do a tunnel, a circumferential tunnel over a chest tube with bovine pericardium, but I would probably use a homograft. I will tell you I know a lung transplant surgeon who had a real problem a few months ago. They couldn’t get the bronchus back together, and they had to reoperate on this patient two or three times, and a pulmonary homograft was placed to bridge this gap on the bronchial anastomosis after a transplant. So far it has worked and there is no malacia— so far. Thank you.
DR RODNEY LANDRENEAU (Pittsburgh, PA): I was wondering if you use that same isolation technique on difficult pneumonectomies, because I have found that very valuable in our setting of postinduction therapy when you have really got a tough dissection around the PA to take that extra time to get the veins isolated, controlled, and have proximal control of the PA, and then it is a bloodless field if you should get into the PA and have to do some pulmonary resection associated with that.
DR CERFOLIO: The answer is yes. In fact, I have used this now in a lot of difficult left upper lobes. You want to stay efficient, you have other patients waiting to go into other operating rooms, and it is a difficult left upper lobe, now, they wouldn’t get a PA resection, but we have done that. You just take a knife and cut all the branches. You have to control the vein first. You have to divide the superior pulmonary vein first, put a clamp on that inferior pulmonary vein and one on the main PA, and you can go very quickly. And we have done that on patients where sometimes you don’t know if you can get a negative margin. You think you are going to do a sleeve but you do not have to cut any PA out. So I think it is a very good and versatile technique just for a tough lobe. Those patients are not in this series. Thank you.
|
Acknowledgments |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  R. Nakanishi, T. Yamashita, and S. Oka Initial experience of video-assisted thoracic surgery lobectomy with partial removal of the pulmonary artery Interactive CardioVascular and Thoracic Surgery, December 1, 2008; 7(6): 996 - 1000. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Rea, G. Marulli, M. Schiavon, A. Zuin, A.-M. Hamad, G. Rizzardi, E. Perissinotto, and F. Sartori A quarter of a century experience with sleeve lobectomy for non-small cell lung cancer Eur. J. Cardiothorac. Surg., September 1, 2008; 34(3): 488 - 492. [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 |
