 I'm really looking forward to the next talk over coffee and I was talking to Chris Wood and I felt I'd had a really busy year in London and I'd done a hundred major renal receptions. And I asked Chris Wood over coffee how many he did last year and he said 350 and I felt rather intimidated. Chris, tell us about some of the 350 you've done. Well, thanks to the organizers for the kind invitation. This is actually my third time in Dublin and I've fallen more in love with the city with each visit. And I'm happy to talk to you today about the surgical strategy for the management of patients with infirivina cava invasion. When I first made this talk, I had something like a hundred slides. I didn't realize there was so much to talk about, but because of all the coffee I've been drinking from jet lag and the two points I tossed back in the afternoon, I think my bladder is going to make this a quicker talk than it originally was. So this is what we're talking about. This is a patient actually that I recently saw at MD Anderson, a locally advanced tumor involving a right kidney with the IVC thrombus extending up into the right atrium. Now you can see the locally advanced tumor in the right kidney again and on cross-section. These cases are probably one of the most daunting cases that urologic oncologists face. Probably this and post chemotherapy retroperitoneal lymph node dissection for testis cancer. Probably the two toughest cases that we face in our practice. Venous tumor thrombi are found most often with clear cell renal cell carcinoma, but they can occur actually with all types of kidney cancer. And it's not just limited to kidney cancer. A variety of other different non-renal urologic tumors will form tumor thrombi. I quote patients that the incidence of venous involvement is somewhere around 15%. It's anywhere from 4 to 36% in the literature, but IVC extension and subsequent atrial extension is obviously much more rare. The vast majority of these patients will present with blood in their urine or flank pain, but you can see that almost 20% of patients that will actually be in incidental finding with no symptoms. With regards to evaluation, probably the gold standard for evaluating a tumor thrombus remains the contrast enhanced MRI. Although with the developments that have been made with CT scanning, I often use CT scanning for my patients to try and identify the thrombus, the extent of involvement and so forth. But the MRI undoubtedly does give more information than a CT scan can, and you can see here the differentiation between the actual tumor thrombus which is vascularized and the associated IVC bland thrombus and the venous collaterals associated with IVC occlusion. It's very important to get updated imaging before taking these patients to the operating room. The tumor thrombus can progress rapidly, so frequently I'll get scans within a week of surgery. The further staging of the patient is for any other advanced renal mass. For those patients that have near occlusion of their IVC or have presence of bland thrombus, we often recommend anticoagulation. I can't tell you the number of times I've been called in the community by well-meaning urologists who want to transfer a patient with an IVC thrombus and they tell me, don't worry, we put a filter in the vena cava so the tumor thrombus won't move. Please don't do that. There have been some papers in the literature that have looked at the thickness of the IVC or the thickness of the renal vein to try and predict wall invasion. This is an important preoperative evaluation because it may give you some indication as to whether or not you're going to do IVC replacement or whether you'll need a vascular surgeon on standby. Although this still remains an imperfect science. There are a variety of staging systems that have been reported for characterizing tumor thrombi with regards to height of ascent. I think the vast majority of us use the Mayo system which is highlighted there. This is a summary of our experience where we've noted patients who present with tumor thrombi. The vast majority of them are level one and level two with a significant minority of patients having tumor thrombi extending up to the hepatic veins and then on into the atrium. Unfortunately, a significant percentage of our patients present with metastatic disease, either distant metastases, nodal metastases, or both. We had a significant or about 20% of patients were managed with preoperative embolization. This was largely before I came to MD Anderson and I'll talk a little bit more about that and the troubles it may cause later on in my talk. With regards to surgical technique, preoperative embolization for intracardiac or keboatrial thrombi used to be the standard at our institution. It was thought that perhaps it may decrease intraoperative blood loss. It may cause thrombus regression, particularly if it's vascularized, but we rarely use it anymore because of the complications that are associated with it. The tumor can be approached through either a midline or chevron incision and the real key to this surgery is early arterial control. We use ligature for collateralizing vessels and virtually all our patients are monitored with transesophageal echocardiography both to confirm the location of the thrombus and monitor for the devastating complication of thrombus embolization. And obviously sternotomy can be used for selected level 3 and 4 thrombi, although I find more and more we're actually able to manage these through a transdiaphragmatic approach without opening the chest. So it's critical to assemble an experienced team that may include surgeons that are familiar with hepatobiliary techniques, obviously vascular surgery and when it goes to the heart cardiac surgery. It's important to operate on the vessels first to ligate the renal artery, to isolate the venous structures and remove the thrombus, and then the cancer operation begins with removal of the kidney and the lymph node dissection. I borrowed this from Brad Lipovich. This basically describes the management of the inferior vena cava during thrombus resection. In group A, if there's no evidence of any clot within the vena cava, you do a simple caveotomy, extract the thrombus and close. If you have distal tumor thrombus in the iliacs, frequently we'll put in a green field filter or a dewey's clip to prevent subsequent post-op embolization. For those patients that have partial or total IVC occlusion, frequently we will just staple across the IVC or actually physically interrupt the IVC, again to prevent post-operative embolization. Again, a word about pre-operative embolization, as I said, it used to be the standard of care in our institution before I came there, but in this study you can see that in patients who underwent pre-operative embolization, it was not associated with decreased blood loss. In fact, blood loss was increased. There was also increased complications and mortality. In the multivariate analysis, there was a five-fold increased risk of pre-operative death in those patients that were embolized pre-operatively. With regard to technique, level one and level two thrombites were relatively straightforward. Frequently, we'll mobilize the caudate low off the vena cava, tying off those short hepatic veins, and then it simply involves including the left renal vein, the vena cava below the thrombus and above the thrombus, and also securing the lumbar vessels and then extracting the thrombus. It becomes a bit more complicated with level three thrombite. There is a procedure to it. First again, renal artery ligation, making sure that you get all the renal arteries going to the kidney with the tumor, mobilizing the liver to allow retro-epatic and super-epatic IVC access, and then you include the left renal vein, the distal IVC, make sure to get all the lumbars in that area. You do a Pringle maneuver to cut off hepatic inflow, and then finally ligate, or not ligate, but include the super-epatic vena cava to remove the thrombus. Frequently, after extraction of the thrombus, you can get control of the vena cava below the liver and allow return of liver blood flow by releasing the Pringle. With regards to level four techniques, veno-veno bypass has been described. I've never used it. We use cardiopulmonary bypass, rarely with hypothermic circulatory arrest, although the reported advantages of hypothermic circulatory arrest are resection in the bloodless field. The problem is, after the arrest is over, it's anything but bloodless, but it does allow you a significant period of ischemia time to extract the thrombus. There was a study done by Brian Souk that demonstrated that the hypothermic arrest was associated with a longer overall survival and a significant reduction in perioperative mortality, but that has really not been my experience. Just a word about the devastating potential complication of tumor thrombus embolism. This was a review in one of the thoracic surgery journals where they looked at nine patients who presented with pulmonary emboli in addition to IBC thrombus. They did an aggressive surgery, which included pulmonary embolectomy. The question of whether or not that embolus represents bland thrombus versus tumor, and this particular very limited series of patients, nine and all, the vast majority ultimately went on to recur and die of renal cell carcinoma, suggesting that the thrombi were in fact associated with tumor thrombi. In contrast, we did a retrospective study looking at our patients who presented with pulmonary emboli, and we noted absolutely no difference with regards to recurrence-free and cancer-specific survival that we did not present with a pulmonary emboli. So I would not suggest to you that all patients who present with a pulmonary emboli should be assumed to have metastatic disease because they don't. What about the robot? Is there any role for robot surgery in the management of IBC thrombi? This was one of the first series that was published with regards to management of IBC thrombi using the robot. It was by Ronnie Abaza who used to be at Ohio State and now is in private practice. He reported on five patients and according to the comments that were made before, we haven't heard about the disasters that he may have had. But in those five patients, he had very respectable blood loss, very respectable operative time, and the mean length of stay was 1.2 days. Presumably these patients are highly selected. He went on to report partial nephrectomy with associated venous tumor thrombus. Again, reporting on four cases of note, two of those four patients developed metastatic disease in less than one year. So it is technically feasible. It doesn't mean just because you can do something, you should do something. I think the jury is still out. This was a review that was published by Indy Gill and he noted that there were a total of 78 minimally invasive IBC thrombus cases reported in the literature. You could see that the vast majority were level one and to date only nine robotic cases have been reported. So is it possible yes? Should we do it? I'm not sure. This is our experience with management of patients with venous tumor extension. We reported on 605 patients who presented with venous tumor extension and had a follow-up of two years. Only 45% of our patients had no evidence of metastatic disease and you can see the different level of thrombi. These are very challenging operations. You can see that our median blood loss was almost the leader. Hospitals stayed with six days. Complications do occur with this surgery. We had 25% within the first 30 days. 10% within the first year. Almost 60% of our patients had the transfusion and the 30-day mortality rate was 2.6%. If you don't have metastatic disease the outcomes from this surgery can be excellent. If you do have metastatic disease the outcomes are significantly less so. We did an analysis trying to predict which patients would have complications and in our multivariate analysis we noted that those patients who were older than the age of 60 and those patients who underwent preoperative embolization were far more likely to have complications. There were three complications. Our embolization was also associated with an increased incidence of minor complications. 12 patients died. One was intraoperatively. 10 were within the first 30 days postoperatively and we did have one late death from a massive PE at three months out from surgery. We looked at a variety of different predictors of overall survival. Those predictors are listed there but in our multivariate analysis these were the factors that predicted outcome of IVC thrombi. Clear cell subtype actually provided a protective phenomenon whereas those patients who had non-clear cell histology and venous tumor involvement had a dismal prognosis. Patients who had high grade sarcomatoid de-differentiation perinephric fat invasion nodal metastases or distant metastases all were associated obviously with an adverse outcome. From graphically here we noted that those patients who had involvement of tumor thrombi to the atrium had a significantly worse prognosis than those patients where the tumor was located to the IVC below the diaphragm. Our results are comparable to numerous other ones that have been published in the literature historically where patients who don't have metastatic disease can have a significantly better outcome than those patients who present with metastatic disease at the time of presentation. This is another study that was published out of Europe where they noted that any involvement of the IVC not just limited to atrial involvement but any involvement was associated with a worse prognosis. In their multivariate analysis they noted that IVC invasion, tumor size, fat invasion, and metastatic disease all were associated with an inferior outcome. These are data that again Brad Leibowitz let me borrow where they noted that the height of the thrombus and also perinephric fat invasion was associated with outcomes. So here you can see this is a level 0 thrombus limited to the renal vein. Patients who had fat invasion had a worse prognosis than patients who had only venous involvement. And then you can go on from there where if you had fat invasion with a level 0 thrombus your outcome was the same as if you had IVC involvement. And if you had IVC involvement with fat invasion your outcome was the same as those patients who had a level 4 thrombus and then T4 disease in this series did the worst. And the factors that predicted their outcome in their particular series included TN and M stage, grade, sarcomatoid features, the presence of histologic necrosis, performance status, fat invasion, and histologic subtype. But by and large, as I stated over and over again, patients who don't have metastatic disease can be cured by this operation. And I quote patients that without metastatic disease there's a 50 to 65% 5-year disease free survival but those patients with metastatic disease obviously do worse. So we looked at some of the factors that predicted outcome in our series of patients who underwent IVC thrombus resection. We looked at 270 patients who had known some metastatic disease at the time of surgery and we noted a very surprising finding. 19% of these patients actually had cancer invading the wall of the vein at the margin of resection. Positive margins were more likely in patients who had a higher level tumor thrombus and also in those patients who had a higher tumor grade. And you can see here that the risk of the local recurrence was significantly higher in those patients who had a positive vein margin of metastatic progression. Initially this led me to believe that we should be sending frozen sections at the time of surgery to rule out the presence of venous invasion at the resection margin. But subsequently you can see from these data that that probably is not going to impact outcome. The vast majority of these patients present with metastatic disease or local and metastatic disease and it's actually rather relatively uncommon to just have a local recurrence. So from my perspective, invasion of the vein wall is more a depiction of a tumor's biology rather than poor surgical technique. I did mention that histology played a role in predicting outcome for these patients. This was a retrospective study that was recently published in European Urology and the bottom line is papillary histology with venous involvement is a death sentence. In our series all of the patients who had papillary histology were dead at five years. There are other studies that look at tumor thrombus consistency. This was one retrospective study that demonstrated a friable tumor thrombus when compared to those that had a solid tumor thrombus had a significantly worse outcome. In a more recent series looking at that they actually looked at some of the markers that might predict for a solid versus a friable tumor thrombus. They noted that there was increased expression of the cell adhesion molecule you can hear in and also increased collagen in the solid tumor thrombi. And in fact they noted that in their multivariate analysis the only factors that predicted for outcome were the priability of the thrombus as well as the presence of tumor necrosis. Okay, I've told you what can happen if you do operate. What happens if you don't operate? Well these patients do dismal. This was a retrospective study that looked at patients who were resected versus those that were treated with quote-unquote conservative management and you can see that the survival for the no surgery group was horrible. And this is a retrospective study out of UCLA that basically demonstrates the same thing be it just renal vein involvement or IBC involvement. So in the time that I don't have left I'm going to talk to you just briefly about the role of targeted therapy in the management of tumor thrombi. We've all seen these retrospective studies, anecdotal studies that have demonstrated these remarkable regression of tumor thrombi associated with targeted therapy. So perhaps we should be giving this to all patients who present with locally advanced disease with venous tumor thrombi to get thrombus to regress. Well we recently published a multi-center experience with UT Southwestern in Wisconsin looking at patients who are managed with the venous tumor thrombi with targeted therapy, the vast majority receiving synitinib. And you can see from these data that the vast majority of these patients had little to no response in their venous tumor thrombi after being treated with targeted therapy. We recently updated this looking at an experience of 48 patients who were treated with a tumor thrombus in place and again the same story the vast majority had little to no response in their primary tumor suggesting that targeted therapy is not a way to get the tumor thrombus to regress and make the surgery easier. So if there's no evidence of metastatic disease the preferred option for these patients is surgical resection receiving, reserving systemic therapy only for those patients who are not really surgical candidates. And just a word about Bud Kiari syndrome Bud Kiari syndrome relates to hepatic venous outflow occlusion which is associated with abdominal pain, ascites, hepatomegaly, lower extremity edema and abnormal LFTs. In our hands the in-hospital mortality rate for taking these patients to the surgery is in excess of 80%. So we no longer operate on patients who present with Bud Kiari syndrome but instead perform angioembolization and if they have regression of their tumor thrombus and reperfusion of their liver and reversal of their symptoms then we take them to the operating room. So in summary radical nephrectomy with IV sheet thrombectomy is a technically demanding surgery that can be associated with significant morbidity and mortality. In the absence of metastases nodal disease sarcomatoid de-differentiation and invasion into the perinephric fat or vein wall appear to predict prognosis. Minimally invasive approaches are described but patient selection clearly is critical. And pre-surgical treatment with either embolization or targeted agents does not appear to be helpful and in fact may worsen prognosis. Thank you very much for your attention. I just copy. Fantastic. Is anyone in the audience want to ask Professor Wooder a question? I just want to ask a little more about embolization. You don't think there are confounding factors the embolization were performed earlier in the series before you started with the surgery and it might be a selection of patients and when did you do the embolization the day before the surgery or earlier and otherwise what should be the reason for the bad performance of the embolization? So you know when I was a fellow the person that did these before me did embolize and I was honestly never impressed with I don't think it decreases blood loss I think it does make the surgery more difficult there's some question as to whether or not the edema that it causes may sort of help you with the surgical planes but quite honestly I think that the problems it causes are not worth any potential benefits I've never saw any dramatic regression of our tumor thrombus never saw our approach change so when I joined the faculty after the first sort of few cases that I did the old way because that's the way I was taught decided this is probably not the way to do things I think that with embolization you basically can operate at two different times you have to operate within 48 hours because after 48 to 72 hours the tissue is basically like wet Kleenex trying to sell or you operate three months later we approach with patients who have Bud Kiari syndrome we embolize them, we wait three to four months re-image them and if they have reperfusion of liver and resolution of their symptoms we take them to surgery but if they don't then we treat them with systemic therapy but I just have not been impressed with the potential benefits of embolization and I think that the side effects and the potential for increased morbidity and mortality are significant Bill Hayden, you have a question and now I had the answer to Sven this made a systemic review of the literature finding that there is more bleeding and complications after embolization so that's in the guidelines, the EU guidelines also Yes, I know that but all patients that are embolized are selected patients they are not randomized serious comparing embolization One observation I would have about embolization is that it means different things to different radiologists and I think the cases I've done that have been embolized it's extraordinary, in some cases it's made things a lot easier or it felt like it's made things a lot easier and there are many cases where it seems to have made no difference at all or made it worse I think I'm not sure there's a standardized technique of embolization that has been described which can produce a consistent result within the kidney what do you think about First of all are we all talking about renal embolization or is it because I think we're also talking about the liver embolization for the bad chiari renal artery embolization So I think the best way to do it is go and embolize with a couple of coils in the vein renal artery to reduce perfusion and this is the way that what we're aiming is to reduce vascularity but not cause an immediate ischemia so from my experience this has been from the feedback that I have this has been the best way of treating high vascular lesions I think the problem is that at least what our problem was when we embolized the artery our thought was okay the artery is controlled we can go up to the venous structures when in fact many times it wasn't controlled and then you're opening up the vein there's still artery going into the kidney blood's going everywhere it's just miserable I agree with that can I ask you a question the cases I really struggle with where it's cytoreductive big venous thrombus is it the right thing to operate there are some metastases it's going to be complicated how do you make your mind up in those situations whether to do complex venous surgery I mean 55% of your patients have metastases so how do you make your mind up about whether you're going to offer that surgery to people who already have metastases because many of them may never get the TKI's if they get a complication that's true but I think complications can happen with any surgery the group from UCLA published on this and reported that patients even with IBC thrombi and extensive nodal disease can still safely undergo cytoreductive surgery and I think that the complications of a thrombus that progresses and causes hepatic outflow obstruction lower extremity edema and so forth those are not insignificant and we've already shown as I showed in my last couple of slides that targeted therapy is not going to impact that thrombus it's going to continue to grow continue to cause problems so I think if you can get the pay you know I'm not advocating this to be done out in the community but if you can get the patient to an experience center who has a lot of experience with IBC tumor thrombi management those patients can get through surgery relatively well I've asked Christopher what his indications or strategy is with vascular replacement of the caver with resection the Marsden we tend to be very aggressive in terms of resection even the absence of obstruction not just with these but also RPLND and many of the retroperitoneal sarcomas just resect at the patients generally experience leg edema for about 3 months but by 12 months they largely got pretty much a normal lower limb function right and that's basically our experience we do try to rebuild the vena cava if we can typically using bovine pericardium again prior to my coming there some of the vascular surgeons were using teflon and cortex that doesn't work it clots off so basically if we can reconstruct with either a tube graft or a partial graft with bovine pericardium we'll do that and put the patient on aspirin after surgery but we've also been in scenarios where there's still flow through the vena cava but we had to resect a portion of it and just left it resected and I agree with you those patients are miserable for about 3 months but they develop collateralization and subsequently a lot of those symptoms will go away Joe Jack do you want to bring things to a close? Just a question do you think that there are some predictive factors in metastatic disease that could help for selecting both patients who are going to benefit from surgery? I mean I published a few articles on it but I don't think it relates to IBC thrombus I think that the factors that you have to look at are presence of liver metastases presence of CNS metastases presence of nodal metastases those are all predictors of bad outcome extensive bone metastases poor performance status absence of clear cell histology presence of sarcomatoid those are the features that we use to select patients for pseudo-reductive nephrectomy but I think just the presence of a tumor thrombus is not in and of itself a contraindication Okay thank you So we are out of time now and I will close the session I thank all the speakers of the session and we have I think a good cover of all the the minimative treatment of small venal tumor and also the practical approach for the coliadvance disease Thank you very much