| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Division of Cardiothoracic Surgery, University of Alabama at Birmingham, Birmingham, Alabama
Accepted for publication January 26, 2011.
* Address correspondence to Dr Cerfolio, Division of Cardiothoracic Surgery, University of Alabama at Birmingham, 703 19th St S, ZRB 739, Birmingham, AL 35294 (Email: robert.cerfolio{at}ccc.uab.edu).
Presented at the Fifty-seventh Annual Meeting of the Southern Thoracic Surgical Association, Orlando, FL, Nov 3–6, 2010.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
Methods: We retrospectively reviewed our experience in starting a robotic program in general thoracic surgery on a consecutive series of patients.
Results: Between February 2009 and September 2010, 150 patients underwent robotic operations. Types of procedures were lobectomy in 62, thymectomy in 30, and benign esophageal procedures in 6. No thymectomy or esophageal procedures required conversion. One conversion was needed for suspected bleeding for a mediastinal mass. Twelve patients were converted for lobectomy (none for bleeding, 1 in the last 24). Median operative time for robotic thymectomy was 119 minutes, and median length of stay was 1 day. The median time for robotic lobectomy was 185 minutes, and median length of stay was 2 days. There were no operative deaths. Morbidity occurred in 23 patients (15%). All patients with cancer had R0 resections and resection of all visible mediastinal and hilar lymph nodes.
Conclusions: Robotic surgery is safe and oncologically sound. It requires training of the entire operating room team. The learning curve is steep, involving port placement, availability of the proper instrumentation, use of the correct robotic arms, and proper patient positioning. The robot provides an ideal surgical approach for thymectomy and other mediastinal tumors. Its advantage over thoracoscopy for pulmonary resection is unproven; however, we believe complete thoracic lymph node dissection and teaching is easier. Importantly, defined credentialing for surgeons and cost analysis studies are needed.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
| Dr Cerfolio discloses that he has a financial relationship with Intuitive.
|
During the past decade, the use of robotic surgical systems has rapidly increased. Approximately 1500 robotic surgical systems have been sold in the United States [1], and this number is likely to continue to grow. The use of robotic surgical systems in general thoracic surgery is at its infancy. In 1998 less than 5% of prostatectomies were performed robotically, and in 2010 approximately 85% were performed robotically.
Will the trend that occurred in urology also occur in general thoracic surgery, and if so, for which procedures—thymectomy, esophagogastrectomy, or lobectomy—or all three? The reality is that despite the lack of any randomized data, minimally invasive techniques, such as video-assisted thoracoscopic surgery (VATS) and robotics, are becoming the preferred approach in many surgical disciplines [1]. Many in thoracic surgery are adopting robotics for thymectomy and for the removal of other mediastinal masses. In this report we review our experience in starting a robotic program and relate the lessons we have learned.
|
Material and Methods |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
The only patients who were not offered robotic resection were those who had a tumor in their segmental bronchus or more proximal or those who had chest wall involvement that required rib resection. (These two conditions no longer represent contraindications to robotic resection, but they did at the beginning of this study). Neither the size of the lesion, its location, the presence of N1 disease, nor the use of preoperative radiotherapy or chemotherapy was a contraindication to offer a robotic pulmonary operation. The indication for robotic thymectomy or for robotic resection of a mediastinal mass was the same as that used for open or video-assisted thoracoscopic surgical (VATS) resection.
The University of Alabama at Birmingham's Institutional Review Board approved this protocol (X100310009) as well as the prospective database used to collect information for this study (X030403013). Patient consent was waived for inclusion in this individual study; however, it was obtained to enter patient data in our prospective database.
The two attending surgeons and one thoracic fellow followed a standard pathway to obtain robotic training. This entailed online training (1 day), followed by on-site training in which the console and robot were used for manipulating an object (1 day), off-site cadaver training (1 day), an observational trip, and finally, several proctored operations. Our institution required five proctored operations for credentialing.
For this study, a newly modified procedure for pulmonary resection was used. We have termed it a completely portal robotic lobectomy using 4 robotic arms (CPRL-4). It was developed and modified several times during this study using the da Vinci robot (Intuitive, Sunnyvale, CA). The final CPRL-4 used for the last 25 patients is shown in Figure 1. This represented a slightly modified, completely portal robotic approach using 3 arms developed and championed by Ninan and Dylewski [4]. We set a time limit of 4.5 hours, after which if the operation was not almost finished, we decided to open to complete the lobectomy. The reason for opening was recorded.
|
Total operative time was recorded, defined as the time that the first incision was made until the time the last incision was closed. This included the robot docking, driving the robotic arm while seated at the console, robot de-docking, removal of the tumor, and skin closure.
Morbidity was defined using The Society of Thoracic Surgery database's definitions (version 2.8) [6]. Operative mortality was defined as death within 30 days after the operation from any cause or before discharge. Outcomes reported included intraoperative time, estimated blood loss, if patient was converted from robotic to an open approach and the reason why, postoperative morbidity, operative mortality, pathologic analysis, and hospital length of stay. Data were collected and analyzed using Excel software (Microsoft Corp, Seattle, WA). Descriptive statistics were used to report the medians and standard deviations of the continuous study variables and number and percent of categoric variables.
|
Results |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
|
|
|
There were no operative deaths for patients who underwent an operation that was completed robotically. However, one operative death in this series occurred in a patient with poor pulmonary function who we converted to open thoracotomy because we could not divide the fissure with a stapler secondary to the thickness and fibrosis of the lung. It was divided with a knife. The patient was admitted to a regular floor bed and did well initially for 2 days; however he sustained a myocardial infarction and died on postoperative day 5.
The details pertaining to the 13 patients who underwent conversion from the robotic procedure to an open procedure are reported in Table 3. There was one conversion from a robotic mediastinal resection to a right thoracotomy for concern of major bleeding in a patient with an 11-cm thymoma. The patient received no intraoperative blood transfusions, and when we opened because we believed we had avulsed a venous branch off of the superior vena cava, no tear or bleeding was evident.
|
All patients in this series with cancer had complete thoracic lymphadenectomy and all had an R0 resection. A median of 14 lymph nodes were removed during a CPRL-4 for non-small cell lung cancer. All visible mediastinal N2 lymph nodes were completely resected (not sampled), and all hilar lymph nodes were resected along with appropriate lymph nodes in the part of the lung removed.
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
Survival data are lacking. Initial reports in 2009 from Gharagozloo and colleagues [7] on a series of 100 patients with a median follow-up of 32 months were favorable. They reported that that only 1 patient (1%) died of recurrent cancer, 6 had distant metastases, and a second lung primary cancer developed in 2. However, the learning curve—the time it takes for the surgeon to be comfortable with the entire operation from beginning to end—is steep. The increased operative times at first are frustrating, and they should not be taken lightly.
Many general thoracic surgeons are eager to learn robotic surgical systems and many centers are starting to use them. This zeal must be tempered with proper training, careful patient selection, and intelligent program building. The credentialing of robotic proficiency needs to be systemically promulgated. Currently, the decision regarding credentialing is made on an institutional level only. Although the literature is now peppered with articles that show the technical and oncologic safety of robotic pulmonary resection and robotic thymectomy, [5, 8–13], proper training of the entire team is critical to maintain patient safety.
We believe that the proper sequence of training helps leads to a program's success. There are many lessons we learned that cannot ascribe a p value or can be proven using the time-honored scientific evidence-based method. The following points thus represent our opinion after having performed more than 150 robotic operations and after having proctored many surgeons at different centers. Some are enumerated in Table 4. Some are important and painful lessons we learned in our initial experience that perhaps we can help others avoid in theirs (Table 5).
|
|
After these two steps have been completed, off-site training using a cadaveric model is usually next. It is critical. Although in the future the console will probably be used as a simulator, the software has not yet been fully developed to replace cadaveric training. A software package does exist that can be snapped onto the back of a robotic console and provides a series of exercises that can measure and document robotic skills and speed. This software summarizes the efficacy of the surgeon and it also provides metrics that document robotic proficiency. However, the current software is not yet capable of mimicking an entire robotic lobectomy or robotic thymectomy. Until this type of software is available and the console simulator is fully developed, cadaveric training will remain a critical part of the training.
Once the surgeon and his or her team return from the cadaveric training, the next step is to ensure that all of the equipment needed for the planned procedures is ready and actually in the operating room and working properly. We have seen too many centers that are "ready to perform their first robotic operation" but do not have all of the necessary trocars, attachments, and equipment. These gadgetry type problems are minimized by having the entire team at the bedside to simulate an operation. They should be familiar with all the gadgets and how to quickly and safely place the robotic instruments.
Another important step is an observational experience. The entire team should observe an experienced team perform a few robotic operations so questions can be answered and the actual teamwork needed for a robotic operation is observed, first hand. Finally, it is our opinion that the thoracic surgeon will develop safe robotic skills quicker if the robot is available to him or her at least once a week and several operations and proctors are lined up very soon after training.
One advantage the robot offers the general thoracic surgeon and the most common application of it to date is for the resection of mediastinal lesions, especially thymectomy. We believe it is in the anterior and posterior mediastinum—in tight confined spaces—that the robot offers its greatest advantage over VATS. We prefer the right chest instead of the left for thymectomy. We believe it is safer, especially when learning, because there is more room to allow for the safe entry of the robotic instruments. Also, the identification of the innominate vein is easier for the right side than the left side. However, one disadvantage of the right-sided robotic thymectomy is that the entire left phrenic nerve cannot always be visualized. Likewise, neither is the all the right phrenic nerve visualized from the left chest. These limitations also apply when performing a VATS thymectomy. This problem can be remedied by placing a VATS camera in the opposite chest and importing the image to the robotic console as a tile if needed.
For lung resection, in our opinion, the advantage of a CPRL-4 over a VATS lobectomy is that it is easier to remove all of the hilar (N1) and mediastinal (N2) lymph nodes. However, many adroit VATS surgeons are able to accomplish an excellent N2 and N1 nodal resection with VATS. We added the fourth robotic arm to the well-described technique of Ninan and Dylewski [4], who use a completely portal technique featuring 3 robotic arms. The fourth arm allows the surgeon to retract for him or herself. In addition, we moved the extraction of the tumor to an area above the diaphragm instead of through part of it.
During the period of this study, we modified the CPRL-4 many times, and this led to longer operative times initially. Over the past month we have performed eight CPRL-4 operations with a median time of 118 minutes, and no further modifications have been made.
The main weaknesses of robotic pulmonary resection compared with VATS lobectomy resection is the need for a bedside assistant who fires the stapler on pulmonary vessels, the inability to feel any of the lung when performing a completely portal approach, and the added time and cost of the operation. The cost of a robotic operation has to be mentioned. It includes the capital cost of buying the robot and 2 consoles (about $2.2 million), the contract to maintain the robot (often about $100,000/year), and the instruments (8-mm instruments are about $2000 each).
Park and Flores [14] reported in 2008 that the robotic operation cost more than VATS—but less than thoracotomy—for pulmonary lobectomy. Increased use of the robot purportedly reduces this cost. In the Park and Flores study, the robot added about $3981 to each operation, $3880 of which was on the day of operation. However, they found that a robotic lobectomy was $3988 less per patient than lobectomy through a thoracotomy, which still remains the most common manner in which lobectomy is performed in the United States.
In summary, this study has shown that the current robotic system available to us today is safe and effective for pulmonary lobectomy and for the removal of many types of mediastinal tumors. Several important technologic advancements are missing, such as a robotic lung clamp, a robotic sucker, and a robotic stapler. These issues and the added costs remain an obstacle to quicker adoption. The increasing use of robotics in general thoracic surgery is undeniable, however. Thus, the proper training of residents and fellows, and the sensible and safe credentialing of practicing thoracic surgeons, is immediately needed to protect patients and surgeons alike. Furthermore, its future use as a simulator may represent a true paradigm shift in surgical training. Academic institutions need to develop programs and unite to ensure the adequate training of future robotic surgeons and to agree on a robotic nomenclature for operations.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
There have been several attempts in the literature to codify how you set up programmatic changes in order to build and adopt a robotic program. Ali and colleagues in 2006 described a program with didactic and hands-on training. They spent 8 hours on simple tasks with increasing difficulty and 2 hours with actual didactic lecturing, and then they went on to each "first assist" 30 robotic laparoscopic gastric bypass procedures. Recently in another paper by Landry and colleagues, they discussed their implementation of a program in robotic thyroid surgery at MD Anderson Cancer Center. They offered several models, and I would like you to comment on these later. These include preceptorship, proctorship, or observation, which you are expert at, and simulation training. They stressed that the importance of team training and cadaver training was key, and they even adopted a new lingo and a new shorthand in order to understand when they needed to do things quickly in the operating room without wasting words. Finally, they pointed out that most programs that failed had insufficient preparation as a team. I have several questions.
Number one, can you share with us some comparative data on your open thoracotomies during the same period or from a prior cohort? You have alluded to us that you currently have a better lymph node dissection than you had previously. I would like you to explain that to us with some data.
Number two, how do you explain the fact that robotic surgery actually decreased length of stay?
And then, finally, although you have shown us your learning curve, can you articulate a thoracic surgery-specific training paradigm that would introduce robotics into our specialty?
And on a lighter note to quote Dr DeBakey in paraphrase, "did you really need a robot, Cerf, to kill that one old lady?"
I would like to thank the Association for the privilege to discuss this paper.
DR CERFOLIO: Dr Krasna, thank you very much for your kind comments. We have done thousands of VATS and still use VATS for benign disease or for diagnostic wedges all the time—but the truth is the lymph node dissection in my hands for VATS lobectomy for cancer, and in the vast majority of surgeon's hands who I talk to, and that is a lot of thoracic surgeons, is not the same as open. Thus, I can go to all the VATS courses in the world, and I have taught many of them, and it would not change those facts. Now we could spend an hour talking about training, because the credentialing of a robotic surgeon is a critical issue and one that has a problem right now. I will make you aware that there is a software package coming out called Mimic that will really allow us to train residents, fellows, and attendings to apply metrics to their skills set. This will be the best simulator for surgeons in the world very soon, and will set the standard as to how to teach surgeons. I wish we had something like this for open or VATS surgery, but we do not. It goes through a series of skills and measures and reports the time a surgeon takes, the efficiency and accuracy of hands and foot movements, etc. It then generates a computer printout for the resident and states, "Oh, you didn't do so well, go back; I want to get you at this speed, this accuracy" etc. It is going to protect us, the patient, and our specialty and document when someone is able to be a robotic surgeon who can start operating on people. It will be out in January.
But, your point of coming and watching is critical because there are lots of things to be learned by the whole team. Proctorships are also important. And although the credentialing is the hospital's job, it is really our job as well to credential one another to protect the patient. So I think all that is coming.
You talked about team training, and this is the ultimate team support, which is what I love about it because you know I love sports. There is a thing called SI Connect that the Intuitive da Vinci people have come up with that will allow surgeons in a room to be able to have the entire operation transmitted over the Internet to other teams of surgeons and their assistants and nurses—it does not just show what the surgeon sees in the console but there are, in addition, cameras in the room and microphones in the room for the nurses to talk to our nurses and communicate. So that will allow teams to be trained very, very quickly with use of the Internet.
Your next question was about comparative data. I didn't put this up, but we sent a paper to the Annals that does just that. I will say, just read it, but it does exactly what you want. It is a comparative data of 4:1 propensity-matched thoracotomy patients to CPRL-4 patients and the robotic patients did better, with fewer complications and less pain and shorter length of stay.
You asked about lymph node data. We looked at the number of lymph nodes. I think that is a little bit of misnomer, if you want my honest opinion. VATS and open surgeons often quote this, but you and I know that only the surgeon really and honestly knows if he removes all of the lymph nodes or not—we have had over 50 visitors come to our place, mostly VATS surgeons, and they are all amazed at how we get all the nodes out and do such an aggressive lymph nodes dissection. The number is misleading, one could measure weight or volume but we have not done that; thus, it is hard to prove that our lymph node dissection is better—but come watch it anytime, or I will show you our videos. Come to Birmingham, we will show you.
Length of stay is less because you could really send them home the day after surgery. Their pain is so much less and their chest wall mechanics are better because we make no access incision. We sent the first 10 patients home on postoperative day 1. Two came back, 1 because of atrial fibrillation, 1 with some subcutaneous air. So now I keep most, except the right middle lobectomy patients, 2 days post-op. But they have a couple of little band aids, so they can go home much quicker.
And finally, you talked about a specific algorithm for robotic thoracic team training. We propose that a surgeon receives on-line training, then training on the robot with a model, and then training on a cadaver performing several thoracic operations, such as a thymectomy and a few lobectomies. Then the surgeon and a few key team members observe an experienced team performing several robotic operations—that is a critical part of this. When they return home, they perform a few simple operations like a mediastinal cyst, robotic wedge, or lymph node biopsy—all of which are proctored—and then a lobectomy that is proctored. I propose that, and I direct you back to the idea of SI Connect and what you can do just by getting the team together as opposed to just the surgeon.
DR W. RANDOLPH CHITWOOD, JR (Greenville, NC): I enjoyed your presentation. The thing that I liked the best is to have a religious naysayer converted into an apostle. I don't think you can have a better apostle for robotic thoracic surgery than Dr Cerfolio. He will go spread the word throughout the world.
There is no question that robotics has a place in the evolution of closed chest surgery, and we have shown this in cardiac surgery. We have trained over 400 surgeons worldwide in this technique. The new visualization techniques and ergonomic instruments are truly facilitating. When you are operating with this device, you become ensconced completely in the operative environment—an entirely new ambiance—this new environment and new methods for operative facilitation. I believe that this will be a major part of every operating room for both thoracic and cardiac surgery. It certainly is in gynecology now, and over 85% or 90% of prostatectomies in the United States are now done robotically.
It is not a robot. It the surgeon using a new fashion of "Metzenbaum scissors." Surgeons must be able to do the traditional operation first. This is key. If you can't do a VATS lobectomy, you can't do a robotic lobectomy. If you can't do a mitral valve repair well and be intuitive in doing these repairs, then you should continue using traditional methods: sternotomy or minimally invasive, but not robotic.
The team training is key; there is no question about it. Every member has to be in synchrony; you have to be in the "rugby scrum" together. This is where many programs fail. About 150 cardiac programs in the United States have started and failed, with almost all of them failing because of inadequate team training.
I congratulate you on the paper. I look forward to reading it in the Annals of Thoracic Surgery.
DR CERFOLIO: Thank you. I think one of the important points is if you are going to get trained, if you are going to wait a month before you come back from training and do a case, that is not the way to do it. You have got to get trained, not just you but you and your team, and come back and have several cases immediately that week and the next week, so everybody can get comfortable and get used to it. And in general thoracic surgery, you should start off with a wedge, you should start off with a couple of very simple mediastinal cysts and do the easier operations first. Let your nurses and your assistants all get comfortable with this giant machine, your anesthesiologist get comfortable with being separated from the patient because of extension tubing, and then as that paradigm shift becomes accepted, then do the more difficult operations.
DR ROBERT S. POSTON (Baltimore, MD): With the growth of your volume fairly rapid over about a one-year period, you probably were involved with marketing.
DR CERFOLIO: No, not at all.
DR POSTON: Okay. Well, that takes care of my question.
DR CERFOLIO: We didn't do any marketing for this. Our hospital didn't do any. We were seeing the same number of patients. As I showed you, I have done about 1100 operations a year for the past 6 years, but if you look at my numbers this year, I will be under 1000 operations for only the second year in a while. My numbers are down because I have been away lecturing so much about this and it takes time to do robotics, it takes longer, and we have developed a robotic program and we have essentially developed this CPRL-4 (completely portal robotic lobectomy using 4 robotic arms) with new port placements, etc. Developing the operation we now use really took some time. So it is just that I am applying it to more and more patients.
There really are no longer exclusion criteria for a robotic lobectomy, which is very different than VATS. The size of tumor is not a reason not to do robotics. As I showed you, we can take out 9- and 10-cm tumors. If you look at all the VATS series, they are limited by size. The tumor can be in the airway and you can still do a bronchoplasty or a sleeve with a robot because it is easier to sew with the robotic arms. We have even done chest wall resections from inside the chest and avoided cutting the muscle above the ribs that are involved with cancer. These would be very difficult to do with VATS, although there are a few surgeons that can do it, the majority cannot. So this allows more patients to get minimally invasive surgery, it allows more surgeons to do minimally invasive surgery, and it probably extends the lifespan of a surgeon to do minimally invasive surgery because of the ergonomics of sitting, the robot improves your vision as you get older with the 10 magnification, and if you get a tremor as you get older, the robot has software that filters that out.
So I think the real big issue is the cost. The cost and the credentialing are the Achilles' heel of this, and the fact that the stapling of vessels for now is done by the bedside assistant until we get the robot stapler. I think those things will get better as we use it more and more, maybe as there is competition, and as we come together as a Society to promulgate how we should really credential this.
Finally, if a patient has multiple small nodules in a lobe that is not to be resected and lung palpation is needed because tattooing or marker is not feasible, or perhaps even for metastasectomy, perhaps the robot is not ideal for these patients because you cannot palpate the lung.
|
Acknowledgments |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  R. J. Cerfolio, A. S. Bryant, and D. J. Minnich Operative techniques in robotic thoracic surgery for inferior or posterior mediastinal pathology J. Thorac. Cardiovasc. Surg., May 1, 2012; 143(5): 1138 - 1143. [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 |
