| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Washington Institute of Thoracic and Cardiovascular Surgery, George Washington University Medical Center, Washington, DC
Accepted for publication February 27, 2008.
* Address correspondence to Dr Gharagozloo, Washington Institute of Thoracic and Cardiovascular Surgery, 2175 K St NW, Washington, DC 20037 (Email: gharagozloo{at}aol.com).
Presented at the Fifty-fourth Annual Meeting of the Southern Thoracic Surgical Association, Bonita Springs, FL, Nov 7–10, 2007.
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and MethodsResultsCommentDiscussionReferences |
|---|
Methods: During a 41-month-period, 61 patients (27 men, 34 women; mean age, 68.2 years) underwent a robot-assisted video-assisted thoracic surgery lobectomy and complete mediastinal nodal dissection for early-stage lung cancer (stages I, II).
Results: Distribution of lobectomies was right upper lobe 14, right middle lobe 6, right lower lobe 9, left upper lobe 21, and left lower lobe 11. Operative times ranged from 3 to 6 hours (median, 4). There were 34 adenocarcinoma, 14 squamous cell carcinoma, 6 adenosquamous, 1 large cell, 2 bronchoalveolar, 2 poorly differentiated cancers, and 2 carcinoid tumors. Pathologic upstaging was noted in 10 patients (8 to IIb, 2 to IIIa). There were no emergent conversions to a thoracotomy. Complications included atrial fibrillation (4), atelectasis (4), prolonged air leak (2), pleural effusion (2), hydropneumothorax (1), and incisional bleeding (1). Mortality was 4.9%. Median hospitalization was 4 days. Follow-up was complete in 54 patients (88%). At a mean follow-up of 28 months, all patients were alive, and 4 had distant metastases. There was no local recurrence.
Conclusions: Robot-assisted vascular and nodal dissection during video-assisted thoracic surgery lobectomy for early-stage lung cancer is feasible. Greater experience and long-term follow-up is required to better evaluate patient selection, oncologic efficacy, and comparability with a conventional open approach.
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionReferences |
|---|
Video-assisted thoracic surgery techniques using conventional endoscopic instruments and two-dimensional visualization limit the surgeon in performing fine dissection and complex three-dimensional maneuvers in the chest. The use of robotic technology may obviate these difficulties. This report describes the application of the da Vinci Surgical Robotic System (Intuitive Surgical, Sunnyvale, California) to the dissection of the mediastinal and hilar nodes, and the dissection of the pulmonary artery and vein during a VATS procedure to achieve individual ligation anatomic lobectomy and complete mediastinal nodal dissection through endoscopic ports.
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionReferences |
|---|
Operative Technique
Equipment and personnel are positioned in a manner similar to that of other VATS procedures. One-lung ventilation is used, and the patient is placed in a lateral decubitus position. Three 2-cm incisions are made using a nontrocar technique. The videoendoscope is placed in the eighth intercostal space in the midaxillary line. Additional incisions are placed typically in the fifth intercostal space posteriorly and the sixth intercostal space anteriorly. The operation is performed in five stages as follows:
Stage I, routine VATS
A zero-degree Endoeye videoscope (Olympus America, Inc, Center Valley, PA) is inserted, and the chest is explored. If histologic confirmation is required, a biopsy is performed using a Microvasive Trucut needle (Boston Scientific, Watertown, Massachusetts.) A 1-cm chest tube incision is made in the one interspace above and three fingerbreadths anterior to the camera incision. An atraumatic paddle retractor (US Surgical, Norwalk, Connecticut) is introduced through this incision for retraction of the lung in an anterior and medial direction during robotic mediastinal dissection. At the end of the procedure, this incision is used for the insertion of a 28F chest tube. The paddle retractor is used to sweep the lung medially and enhances the exposure of the mediastinum, and is held in place by a self-retaining system (Mediflex; Velmed, Wexford, Pennsylvania).
Stage II, robotic mediastinal dissection
The robot is brought into the operating field over the patient's head (Figs 1 and 2).
The camera arm is introduced through the camera incision. In the right pleural space, the right robotic arm with the robotic hook cautery is positioned through the anterior incision and the EndoWrist is positioned over the mediastinum. The left robotic arm with a robotic long forceps is positioned through the posterior incision over the paddle retractor, and its EndoWrist is also positioned over the mediastinum. Conversely, in the left side of the chest, the right robotic arm is introduced through the posterior incision and the left robotic arm is positioned through the right incision. At all times, the robotic arms are oriented so that the hook is manipulated by the surgeon's right hand and the forceps by the surgeon's left hand. A metal suction is introduced by the first assistant through the anterior incision below the robotic arm and is used to evacuate smoke, provide occasional retraction, and remove blood from the field. All visible nodal tissue from the subcarinal, paratracheal, and paraesophageal spaces is removed (Fig 3).
|
|
|
Stage IV, anterior hilar dissection
The robot is used to dissect the anterior aspect of the hilum, including the superior pulmonary vein and the proximal branches of the pulmonary artery.
Stage V, division of the vessels and bronchus by VATS
Once dissection is completed, the robot is withdrawn and lobectomy is completed using VATS techniques. At this stage a 30-degree EndoEye (Olympus) endoscopic camera is used. During this phase of the procedure, the camera is introduced through any port that will enable the best circumferential visualization of the anatomic structures. The pulmonary artery branches, the pulmonary vein, and the bronchus are divided using a stapler.
The operative sequence for VATS lobectomy after robotic dissection is as follows:
Right upper lobectomy: division of the pulmonary vein to the upper lobe; division of the transverse fissure above the right middle lobe pulmonary artery; division of the anterior ascending segmental pulmonary artery; division of the posterior-superior branch of the pulmonary artery; division of the upper lobe bronchus; and division of the posterior fissure above the superior segmental pulmonary artery.
Right middle lobectomy: division of middle lobe pulmonary artery; division of middle lobe pulmonary vein; division of the right middle lobe bronchus; and division of the transverse fissure.
Right lower lobectomy: division of the inferior pulmonary vein; division of the anterior fissure; division of descending pulmonary artery; division of the right lower lobe bronchus; and division of the posterior fissure.
Left upper lobectomy: division of the superior pulmonary vein; division of lingular pulmonary artery; division of the anterior fissure; division of the posterior fissure; division the superior segmental pulmonary artery; division of anterior ascending pulmonary artery; and division of the bronchus.
Left lower lobectomy: division of the inferior pulmonary vein; division of the posterior fissure; division of the descending pulmonary artery to the lower lobe; and division of the left lower lobe bronchus.
After the completion of the lobectomy, the intercostal portion of the anterior incision is enlarged to accommodate the resected lobe, and the specimen is retrieved using a lubricated double-walled bag. A fibrin sealant (Evicel; Johnson and Johnson, Sommerville, NJ) is applied to the bronchial and pulmonary staple lines. For postoperative analgesia, on-Q (i-Flo, Lake Forest, CA) pain control catheters are placed in a posterior subpleural tunnel encompassing intercostal nerves 2 to 8. All patients are extubated in the operating theater.
Follow-Up
Length of postoperative stay, all major and minor complications, and mortality were recorded for each patient. Follow-up was obtained from records of post discharge visits, interviews, tumor registry data, and regular radiographic and clinical follow-up. Computed tomographic scans were obtained for all patients at 3, 6, 12, 18, and 24 months postoperatively. They were included in the survival but not in the recurrence analysis.
Diagnosis of recurrent disease was made by radiographic and pathologic confirmation. Recurrence was defined as local when disease recurred at the pulmonary hilum or in the subcarinal space. Recurrence was defined as distant when disease developed in a separate lobe, in the contralateral lung, in N3 nodes, or in an extrathoracic site. Distinguishing second primaries from distant recurrence was difficult. These cases were recorded as distant recurrence unless specific criteria were met [9].
|
Results |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionReferences |
|---|
Distribution of lobectomies is listed in Table 1. There were almost equal number of lesions in the right and left sides (right: 29 of 61, 48%; left: 32 of 61, 52%). There was a preponderance of upper lobe lesions (41 of 61, 67%). Pathologic cell type of the resected tumors is shown in Table 2. Operative times ranged from 3 to 6 hours, with a median of 4 hours. There were no conversions to thoracotomy. Hospitalization ranged from 3 to 42 days, with a median postoperative length of stay of 4 days.
|
|
Complications
There were a total of 3 deaths (4.9%; 2 within 30 days, 3%, and 1 in hospital, 2%). The first death was of a 68-year-old man with forced expiratory volume of air in 1 second (FEV1) of less than 800 cc who underwent a left upper lobectomy for stage IA adenocarcinoma. Postoperative unsuspected liver failure developed, and he died on the 12th postoperative day. The second death was of a 75-year-old woman with preserved pulmonary function who underwent a left lower lobectomy for stage IIA adenocarcinoma. The patient died of unexplained cardiopulmonary arrest while undergoing a radiographic procedure on the fourth postoperative day. The third death was of a patient with FEV1 less than 800 cc who underwent a right upper lobectomy for bronchoalveolar carcinoma. The patient had postoperative pneumonia and respiratory failure and died on the 42nd postoperative day.
Six complications were observed in 14 patients (22%) and are noted in Table 3. The most common complication was atrial fibrillation in 6% of patients. All complications were minor.
|
A total of 4 distant metastases (7%; 2 brain, 1 liver, 2 bone) have been identified in the follow-up period. There has been no local recurrence.
|
Comment |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionReferences |
|---|
Among these techniques, individual ligation VATS lobectomy with a utility thoracotomy has been shown to be safe and feasible for patients with early-stage lung cancer [11, 12]. When compared with lobectomy by thoracotomy, this technique for VATS lobectomy is associated with shorter hospitalization, decreased postoperative pain, preservation of pulmonary function, fewer complications, and shorter duration of hospitalization and chest tube use [13, 14, 15]. However, a number of difficulties remain.
Owing to the technical difficulties of dissecting the hilum by endoscopic VATS techniques, some investigators have opted for a less than complete hilar dissection or have needed a utility thoracotomy. Even with a utility thoracotomy, pulmonary vascular dissection has been challenging. This shortcoming of the VATS technique stems from the instruments being introduced through ports or small incisions, which amount to holes in the chest wall. The instruments pivot at the entry hole and can be moved in four directions. The limited mobility of conventional endoscopic instruments, whether in the abdomen or the chest, has been referred to by some investigators as "chopstick surgery." The chopstick nature and the limited maneuverability of the effector instruments stem from the rigid shaft axis fixed to the thorax by the entry hole. This technical shortcoming limits the surgeon in performing fine dissection and complex three-dimensional maneuvers. Pivoting instruments on the chest wall results in a large radius of curvature for the tip of the instrument and makes fine dissection in deep spaces such as the mediastinum very difficult and even dangerous. Indeed, it is this fact that has necessitated the need for a utility thoracotomy. Using the utility thoracotomy, the surgeon is able to utilize conventional instruments and his or her own wrist to provide additional degrees of freedom. During the early phase, VATS lobectomy was offered to patients with stage I lung cancer with no evidence of nodal disease on preoperative imaging studies. This strategy was followed because the technical limitations of VATS may make a complete mediastinal and hilar nodal dissection difficult. Another shortcoming of the VATS technique is the lack of the three-dimensional visualization. The surgeon has to use two-dimensional information from the video monitor to create a three-dimensional mental image. That requires significant experience and can prove to be a source of fatigue for the surgeon. Most importantly, using such indirect means of depth judgment is rarely equivalent to binocular vision. In the thoracic hilum, binocular vision is paramount for lymphadenectomy and individual vascular and bronchial dissection.
The use of the robotic technology may obviate these difficulties. The daVinci surgical robot represents an ideal tool for dissection of the subcarinal space, the pulmonary artery in the fissure, and the pulmonary veins. The ultimate goal of completely endoscopic lobectomy can be realized by incorporating the daVinci robot into the VATS platform and combining it with the VATS techniques.
Although robotics provides many potential benefits, a balanced assessment of the benefits of this technology needs to consider a number of issues.Three-dimensional endoscopy is available without the robot. This technology along with high resolution videoendoscopes may provide similar benefits for binocular visualization as are obtained with the robot. Tremor control may not be very useful in thoracic surgical procedures. The lack of haptic feedback can result in excessive pressure and stress on the fragile intrathoracic structures. Robotics removes the surgeon from the field and requires a highly trained assistant at the operative field. That necessitates a highly organized and rehearsed approach to the conduct of the operation, especially in the event of complications such as catastrophic bleeding.
The distribution of the robotic lobectomies was similar to lobectomies for the same stage performed previously by conventional techniques. After the learning curve, the operative times for robotic lobectomy were comparable with those for open lobectomy.
One of the most important aspects of the robot-assisted VATS lobectomy has been the depth and accuracy of hilar and mediastinal nodal dissection. Ten patients with clinically understaged disease were upstaged after robotic mediastinal exoneration. We do not use mediastinoscopy routinely. We reserve mediastinoscopy for patients with N2 disease as diagnosed by staging computed tomography and positron emission tomography scans. It is possible that mediastinoscopy combined with VATS may have upstaged the patients who were upstaged as the result of robotic mediastinal dissection. The depth and extent of mediastinal and hilar nodal dissection may have been responsible for the lack of local recurrence in this series.
Robotic dissection of the pulmonary artery and its branches is precise, and as the result of the magnified three-dimensional visualization, it is inherently more accurate than dissection with conventional instruments through a "utility" thoracotomy. When using the robot, we have had to reappraise the conventional approach to the pulmonary artery. Conventional teaching is to approach the artery through the oblique fissure and to localize it at the junction of the three lobes. However, since during robotic surgery the artery is viewed using a cephalad viewing camera placed over the diaphragm, we have discovered an approach to the localization of the artery that is more applicable to robotics. With the camera viewing the fissure in a cephalad direction, the most accessible and superficially positioned part of the artery is the anteromedial branch of the inferior pulmonary artery. After dissection of the pleura overlying this vessel, the vessel is followed to localize the remainder of the pulmonary artery in the oblique fissure. Furthermore, we have determined that regardless of which lobe is being removed, the extent of the pulmonary artery dissection should encompass all the main branches. This technique facilitates greater mobility of the hilar structures, prevents excessive tension on the branch points, and is paramount in preventing inadvertent avulsion of the branches.
We have routinely used three incisions for VATS lobectomy. Robot-assisted hilar, vascular, and mediastinal dissection necessitates an additional 2-cm port. As the robotic camera and arms occupy the three ports, the additional port has been used for the placement of an endoretractor. At the conclusion of the procedure this port is used for the tube thoracostomy.
There has been concern about the excessive force that can be exerted onto the port sites, especially by a poorly positioned robot. Although this concern needs to be studied in the future, we have not found the additional port or the use of the robot to be the cause of greater pain and morbidity when compared with our conventional VATS approach. Our post-VATS analgesia strategy is based on prolonged (5-day) infusion of a local anesthetic (0.5% bupivacaine, 4 mL/h) through catheters (on-Q Painbuster; i-flow) that are placed in a subpleural tunnel encompassing intercostal spaces 2 through 8. This technique has been shown to be associated with shorter hospitalization and greater patient satisfaction [16].
We routinely seal the pulmonary parenchymal staple lines with human fibrin sealant (Evicel; Johnson and Johnson) This strategy has been shown to be associated with greater aerostasis [17].
Two of the 3 deaths in this series were in the early part of the experience, at a time when the procedure was offered to patients with borderline lung function. This approach represented poor patient selection and flawed reasoning as to the benefits of robotic lobectomy in this high-risk population. Although greater experience is necessary, we have observed that in patients with poor FEV1 and diffusion capacity of the lung for carbon monoxide, the lung does not collapse, and the robotic and endoscopic VATS maneuvers are hindered by the confined pleural space. As the result, for these patients, the operative times were longer and may have contributed to the postoperative complications and the poor outcome. As Demmy and Curtis [18] have observed, a utility thoracotomy may be instrumental in applying VATS lobectomy in high-risk patients with poor pulmonary reserve.
Based on this limited experience, we can conclude that robot-assisted VATS lobectomy is feasible for patients with stage I and stage II lung cancer. Furthermore, robot assistance may facilitated nodal and vascular dissection during VATS lobectomy. Study of this technique in terms of safety, potential advantages, and oncologic efficacy should be the goal of further investigation.
|
Discussion |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionReferences |
|---|
While the robot is technically very easy to use and actually a lot of fun, many of the benefits that he describes are for the surgeon and not necessarily for the patient. In our experience, the technological advances afforded by the robot have not translated into decreased chest wall trauma, length of stay or postoperative pain when compared with the VATS technique. The only variables to which the robot has added are increased operative time, increased number of operating room personnel, and cost. So to state that the robot is the be-all or end-all right now, I think, is a bit premature and would hinder further advancement in technology. So I think we need to find out what it is good for, what it is not good for, and create newer instruments to help take over its shortcomings. That being said, I congratulate you on a tremendous effort and a unique contribution to thoracic surgery.
I have two technical questions. First of all, in your manuscript you state that you don't use a utility port, all the incisions are 2 cm in size, and you state that the VATS technique requires a 4-cm incision for dissection, which I actually disagree with. The 4-cm incision is used to extract the lobe as a whole. The only lobe where we do not use a utility incision is the middle lobe because it is so small. So my question is, how do you get, let's say, a big left upper lobe out of a 2-cm incision? Do you morcellate it, or what do you do with it?
DR GHARAGOZLOO: Thank you, Dr Flores. I agree that this is just the beginning. I believe that the robot is an excellent tool to be added to the VATS armamentarium. It is very important to view the robot as a tool for dissection that can enhance the VATS platform and obviate the shortcomings of the conventional videoendoscopic instruments, as opposed to a "new" platform.
The answer to the question about how you remove the lobe is actually multifaceted. First, as I showed, we use a double bag system. We have experimented with different materials, and presently use two bags which are commercially available. A LapSack, which has high tensile strength, is placed inside an Endocatch. The openings of the two bags are attached with four sutures. The Endocathch has a spring-loaded opening mechanism that allows both bags to be opened simultaneously when inside the chest. The lobe is placed inside the inner bag (Lapsack), and the entire system is pulled out of the anterior incision. A lubricant is applied to the outside of the inner bag in the interface between the bags and the inner bag is pulled out of the outer bag. The outer bag is used to protect the wound and to decrease friction.
Second, with the larger lobes we try to take out the tumor as a wedge excision first. This allows for the hardest part of the lobe to be removed separately. The remaining lobe is spongy and will not form a "knucke" under the intercostal space.
Third, if the lobe is too big, we will cut it within the inner bag, which is very strong, and remove the two pieces in tandem. Whether the incision is 2 cm at the outset of the procedure is a moot point. The issue is the skin stretches very well. Furthermore, removal of the specimen through the anterior hole takes advantage of the pliability of the costal arch.
DR FLORES: The other question I had is your conversion rate to thoracotomy was zero, and I think that is because you excluded certain patients. Did you have any conversion from robot to just a straightforward VATS lobectomy?
DR GHARAGOZLOO: No. In all patients who undergo robotics we explore the pleural space by VATS first and make the decision whether to proceed. Once we proceed, we believe in responding to complications using videoendoscopic techniques as apposed to resorting to a thoracotomy.
The key is to make sure that one can actually accomplish the operation safely at the outset of the operation and not to try to go on as long as one can until a problem occurs. The second key for the low conversion rate is to learn how to deal with problems, like bleeding, like bronchial problems, et cetera, using video techniques. It is very important for us to learn these new tricks so we don't open the patient. This philosophy is not because I think a thoracotomy is bad, but I think that if you are doing a thoracotomy in an emergent situation, you are not serving the patient's safety. You need to learn to fix the problem with the new techniques.
DR FRANCIS ROBICSEK (Charlotte, NC): It is somehow "déjà vu all over again." Robotics was introduced in cardiac surgery 8 years ago, and now after 8 years we are doing 7.2 procedures per year in the United States for every $1.5 million robot. However, there is one paper written for every 20 cases. My question to you is, what is the advantage of your 3- to 6-hour operation performed for double the price over a 1 to 1-and-a-half-hour simple VATS lobectomy?
Also, don't you think that we should concentrate on developing robotic hand-operated instruments to further improve VATS lobectomy, instead of trying to operate from the next room using a robot?
DR GHARAGOZLOO: Thank you for the question. Obviously, you are not the first one to ask that question. Surgery is a living and ever-changing field. As we go forth with robotics, it is important to note the remarks of two great surgeons Dr Billroth who said: "A surgeon who operates on the human heart should be banned from surgery," and Dr Charles Mayo who said: "The only thing that is constant in surgery is change." I would like to begin my answer with the latter.
Anyone who has a lot of experience with VATS lobectomy will honestly tell you it is not as good as an open operation. I think that this fact is the elephant in the room and it needs to be acknowledged. A VATS lobectomy is not as good as a lobectomy by thoracotomy from an oncologic standpoint. That is why there has been reluctance among thoracic surgeons to embrace VATS lobectomy.
DR FLORES: I disagree.
DR GHARAGOZLOO: Except for Dr Flores.
DR ROBERT J. CERFOLIO (Birmingham, AL): And I suspect many others disagree as well ... Sorry to interrupt, go ahead.
DR GHARAGOZLOO: At least in my opinion, the issue is, how can we improve on the VATS platform and get closer to an open procedure? This phase is not the end-all of our efforts, but I think it is certainly a good beginning.
DR CERFOLIO: One option, guys, is to do a thoracotomy, spare the intercostal muscle, the rib, and the nerve, and everything is great.
|
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  F. Gharagozloo, M. Margolis, B. Tempesta, E. Strother, and F. Najam Robot-assisted lobectomy for early-stage lung cancer: report of 100 consecutive cases. Ann. Thorac. Surg., August 1, 2009; 88(2): 380 - 384. [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 |
