Excellent. My name is Chris Chermansky, and I'd like to welcome everyone to today's webinar. It gives me great pleasure to introduce Dr. John DeLancey. We're going to be discussing pelvic floor MRI, and I'm today's moderator. Before we begin, I'd like to share that we will take the questions at the end of the webinar, but you can certainly submit them at any time by typing them into the question box on the left-hand side of the event window. Today's webinar, again, is on pelvic floor MRI, as presented by Dr. John DeLancey. Dr. DeLancey is the Norman F. Miller Professor of Obstetrics and Gynecology at the University of Michigan Health System, and he's one of the world's foremost surgeons and researchers studying pelvic floor disorders. He's been very prolific in this area. He is also past president of AUGS. With that, I would like to turn over the floor to Dr. John DeLancey. Great. Thanks, Chris. Thanks, everybody, for signing in. It's really a pleasure to be able to talk with you all about some of the things that we're learning from MRI. I wanted to thank AUGS for sponsoring these webinars, because I think it's a great way for people to keep up on some of the recent developments that are coming out. I have no disclosures that are relevant to this presentation, and I suppose you're wondering why I have a photograph of a bunch of pilots in a parachute. And the reason that I show this is because I kind of see our research group, like the pilots that you can see over on your left, they're trying to understand how the parachute works. You can see them pulling on some cords. You can see some of them kind of watching from the sidelines. They're trying to assess what's going on with the parachute because they know that that knowledge might be critical to them. And I think a lot of what our group has been doing and what I personally have been doing over the last 30 years is to try and understand how the pelvic floor works, because I don't think that my surgical results are as good as they possibly could be. And a lot of that has to do with understanding what it is that I'm trying to accomplish with an operation. Now the problem that we have with the pelvic floor is this. This is all we're really allowed to see. So that when you think about the ideas that you develop when you have a limited view of what's going on, you can see that it might be difficult to figure out how things work. And for example, when we look at this picture of a cystic seal, it would be perfectly obvious to think that there's a problem with the vaginal wall, because that's the only thing that we can see. And what I'd like to do today is to share with you the fact that now we can actually see the full picture of what's going on, that the advent of magnetic resonance imaging has allowed us to actually not only observe what's going on, but also to see in detail the inner workings that we haven't been able to see for so many years. And not only can we see things, but we can also put an axis on and make measurements. So if you look at where the dot is at rest, and then when you look at where the cervix goes, you can easily make a measurement about how much each and every part of the pelvic floor moves. And that's, I think, part of what's been revolutionary about this is the ability to look scientifically at what's going on. Now, I'd like to talk a little bit today about how MRI is changing what we believe about the pelvic floor and how it works. Talk a little bit about the anatomical foundations for MRI. How do we know what we're looking at? I'll talk a little bit about what we use static MRIs for and something about the different sequence types that could be helpful with those scans. We'll also talk about how to get good images, the role of dynamic imaging, which you've just seen. And then I think the technique that has really, at least for me, revolutionized what I thought about things and has allowed us to be able to actually do studies to prove what theories are right and what theories are wrong. The technique of stress 3D MRI, how is it done? And then what have we learned? Now, a lot of what I'm going to talk about is a shift. And this is a shift from saying what fell, that is, we look at this and we say that it's a cystocele, to saying, why did it fall? To try and understand what it is. And I'd like to give you an illustration about what I mean, because currently what we do in clinical practice is to say what fell without necessarily being able to know why it fell. This is that same kind of situation. Here is the Tacoma Narrows Bridge and you can see here that it failed. And so the engineers, of course, were interested in trying to understand what happened. And they did a very detailed analysis. And they could see that there was a fraying of the primary cables. They could see that there was buckling of the roadbed and they could see that there were individual failures of the secondary cable. And so you could ask which of these is really at fault. But when you actually look at a video, you can determine what was the cause and what was in effect. Because actually what was wrong was this harmonic oscillation of the roadbed that happened with high winds. And that all of the other things were the effect of this oscillation. You may notice how thin the roadbed is, and it's not nearly as thick as we currently see. Because once they realized this problem, all of the roadbeds were stiffened and this has never happened again. So this is the value of understanding the difference between cause and effect. You're also probably familiar with this with echocardiography. If you get the echocardiogram report, you'll see all of these different measurements along with normal ranges that allow you to tell exactly what's going on. So if you look at the echocardiogram, they'll talk about the movement of the ventricular walls and their thickness, their diameters, valve morphology and function, ejection fraction. And based on those individual problems, you will then choose a course of treatment. So this is what I'm now talking about with pelvic floor MRI is that we're going to evaluate each of the individual parts of the vaginal support mechanism. And we're going to start to have measurements about what things are and we're going to start to say, is it normal or isn't it? We'll look at the cardinal and utero sacral ligament length, paravaginal descent. We'll look at the width, length and stiffness of vaginal wall factors as well as the levator ani. And with each of those things, eventually you would have a normal range, you'd see an abnormality, you treat the abnormality and then you evaluate the results. So the first thing is how do we know what we're looking at? Many of you have looked at an MRI scan and certainly you'll feel comfortable in seeing where the uterus is or you'll feel comfortable in where the bladder is, but you may not feel as secure in knowing where the levator ani muscles are or what the perineal membrane is. So in the beginning, we had the same problems, but by doing some actual anatomical cross-sections and then comparing them directly with MRI scans, we were able to decode what it was that we were seeing in the MRI images so that we could look at the exact details. This is just kind of gross anatomy, but it was actually also possible to look at some of the fine detail and here what you're seeing on the right-hand side is a histologic section of the middle of the urethra. And down at the bottom, down here is the vaginal wall. This is the urethral sphincter here. This is the urethral lumen here. And you can see this MRI image, which was taken of this urethral specimen shows exactly the same layers. And so by doing anatomical correlations with actual anatomical specimens, over the years, it's been possible to build up a fairly precise knowledge about actually what it is that we're seeing in these magnetic resonance imaging scans. Now the type of scan sequence that's being used matters. And I just am comparing two of those here so that you can see some of the differences. On the left-hand of the side of your screen is what's called a T1-weighted image. And on the right-hand side of the screen is a T2-weighted image. And what I do is to call your attention to this central area here and compare it with this central area here. Now certainly you can see the thigh muscles and the fat very well in both of these. But as you start to look at the urethra and the bladder base and the other structures around this area of the perineal membrane, there's much more detail in the T2-weighted scans than there is in the T1-weighted scans. Now there are hundreds of different scans. And so if you're starting to do MRIs with your radiologist, you'd want to look at some of the articles in the literature and find out which specific scan types they use. And then the radiologist can help you with obtaining those in the studies that you're going to be doing. So the type of scan actually matters in terms of what you're trying to do. So with a static scan like this, the question is what can you do with it? What kinds of questions can you answer? Well, the first questions that we ran across were during a study that we were doing that was looking at women who developed new stress incontinence after their first delivery. And when we wrote the grant, I believe that we could on static scans see paravaginal defects, something that I realized pretty quickly that we actually could not do. What we didn't know is that we were going to find levator defects because those really haven't been described before in MRI. But if you look at the normal scan that's over here on your left, you can see the urethra, the vagina and the rectum and then you see a levator muscle on this side and levator muscle on this side. Whereas if you look at this individual, the urethra is in the front, the vagina is in the middle, the rectum is in the back and there's no levator on either this side or on this side. So it became pretty obvious to us that there was something that was wrong about individual patients and once we had figured out what the anatomy was, we could show that that was a defect in the levator ani. Over here, you can see an asymmetry where there's a levator muscle that's present over here. You look at the urethra here in the middle, it kind of looks like the head on a little person. There's the obturator internus out here and there's muscle between there and here. When you see the head, you see the obturator internus over here. There's no muscle there and the vagina has spilled out to contact the obturator internus. So static scans are able to tell you when there is a specific defect in muscle or something that you can actually see without necessarily having motion or being able to look at things under load. Well, why haven't you and I picked up on that before and I think part of it is that we never really were trained to look and I'd like to give you an example. All of us, if I told you that this was the vaginal lumen up here and that this is the anterior wall and posterior wall would have an easy time saying that that was a rectocele. We similarly have an easy time saying that this was normal support and I can't tell you how many textbooks I've seen that said that the rectocele was because the perineal body was deficient and too thin, hence the rulers. So here we can see that the normal perineal body is two and a half centimeters. Perineal body with the rectocele is three centimeters. So no difference there. But what's radically different is the fact that in the normal woman, the front of the perineal body is two centimeters from the urethra, whereas in the woman with the rectocele, it's five and a half centimeters from the urethra. So this is something else that fundamentally got us thinking about whether or not just calling attention to the pink bulge that we could see was really the full picture about what was going on. But the question then was what are the structural differences between the woman on the left with normal support that has failed in the woman on the right-hand side that's allowed the vagina to open and allowed the posterior wall to become visible? Well, if you look anatomically, you can see the vulva here in this dissection. This is the anus. The anal sphincter has been taken off and then this is the pubic portion of the levator ani. It's a muscle that has constant activity and so it normally, as you can see from its origin and its course, would hold the vagina and the pelvic floor closed by its constant activity. So it would seem logical that if the muscle broke, that things would collapse downwards. Now here's a picture of a fairly typical systocele and those are actually my fingers and I had actually done her exam, but it was only sometimes afterwards when we were looking at a videotape that I noticed that the systocele had captivated my attention and I'd completely missed the fact that her perineal body was at a 45-degree angle. It kind of looks like on one side there's something attached that's broken on the other side. Well, the reason we had a video is that she was in one of the MRI projects and so we pulled her MRI. And if you find the little head, the little bullseye up at the top, and then if you look behind that at the vagina and the rectum, and then if you look on the right-hand side, you'll see a levator ani muscle. And if you look on the left-hand side, as you look at the picture, you'll see that there's not a levator ani muscle. So here's the urethra, here's the vagina, here's the rectum, here's the levator on our right side. You can see it attaching to the pubic bone there. And when you look at the other side for where there should be a muscle attaching, there isn't. So it isn't that we couldn't see these things, it's just that we hadn't necessarily understood what it was that we were seeing. And so one of the things that I've learned a lot from MRI is once I've noticed something on an MRI scan, I can then go back and see if there's any correlate to that on a physical examination that I may be able to pick up. And these defects are actually quite palpable once you learn what the normal muscle feels like and what the abnormality feels like. But we still didn't know whether that was important or not. And so again, we were fortunate to have funding from the NIH. We recruited 134 women who had normal support. We had 151 with prolapse. And you can see that there is an excess of 40% of these kinds of muscle injuries in the individuals who have prolapse or an odds ratio of 7.3. This is a dramatic difference. And in all the research that we've done since then, we've never found anything that is as strongly associated with prolapse as these injuries to the levator ani muscle. Now the fact that 55% of people have these injuries means that 45% of women with prolapse don't have these injuries. And so there's still a lot of other factors that are involved. I'm not trying to imply that the muscle injury is all of it. But now we have a test that allows us to be able to see who has this injury and who doesn't have this injury. And that then starts to subdivide people with prolapse in a new way that we haven't been able to look at before. So another way that MRI has been helpful is actually in studying how things are injured. And I'd like to share some work that Janice Miller in our group did in looking at the mechanism of injury at the time of vaginal delivery. Because there was a lot of speculation about what it was about vaginal birth that caused these injuries. Some people said that it was because of muscle compression or stretching or neuropathy. And by using some different scan times, and you'll see that we scan early and then we scan later on and using a specific type of scan that generally is referred to as a fluid sensitive sequence, we could actually see whether there was edema. And I'll kind of play that out for you now. So this was a publication from 2010 that now has been replicated in a larger population to be able to prove that the early results were correct. But I'm just going to show you these early findings because they're quite simple. So the basic way that MRI helped with this is that we took 19 primiparous women who were at high risk for levator injuries so that we knew that we would have some women who did have injuries. And part of the strategy was to scan them soon after the delivery of one month and then repeat the scan six months later and I'll tell you why that was important in just a minute. We did anatomical sequences like you've been seeing and then we did these fluid sensitive sequences to look for edema. So here's the basic way that this study was going to be helpful. So if the hypothesis that this was a muscle tear, if the muscle tore, we would expect that it would be abnormal right after the delivery. So we would see it early and then we would also see that that tear was still present later on. If it was a denervation injury, which was one of the dominant theories at the time, then you should normally see a muscle that had its normal morphology early. But then with nerve related atrophy, it would shrivel up later on. So you'd see a big muscle that then turned into a small muscle. There were people who had hypothesized that this was compression of the muscle by the head against the pelvic floor. And if that were true, we would see muscle edema from the compression. And because both the levator and the obturator internus are inside the pelvis and right next to one another, the edema would involve both muscles. So here's an example scan of an individual, the same person both early and late. And you can see over here again, we can see the little target of the bullseye of the urethra. We can see the levator ani over on this side. And we can see that the levator ani is missing on this other side. Then when we look at six months, again, we can see the normal muscle over on this side. And we can see that the muscle is still missing on the other side. So that is definitely consistent with the muscle tear hypothesis. But you can see that the muscle here hasn't gotten any thinner. There's not any loss of muscle bulk. And if anything, you can see that this muscle is a little bit paler than the internal obturator muscle over here. And here you can see that they have equal density. And so that's a subtle sign of edema. But I'll show you with the fluid sensitive sequences in a second how that becomes dramatic. So again, we can see a normal muscle on one side, we can see that both early and late. There's no muscle early, there's no muscle late indicating that this is consistent with the tear hypothesis. So on your left hand side of these pair, you can see a normal anatomical sequence like we've been looking at one of the proton density scans. And on the right hand side, you can see a fluid sensitive scan that's for edema. And notice how the levator muscles here now are bright white, in contrast to the other muscles over here. That indicates that there's a lot of excess fluid in the muscle compared to the other muscles, for example, the obturator and ternus muscle here. So what you can see is that something affected the levator, but it didn't affect the obturator. And if the baby's head was pressing, it would press on both of those equally. And so the fact that it only involves the levator means that the levator was probably stretched, but compression was not the mechanism of injury. The other thing that those of you that have been looking closely at the pubic bone have noticed is that the two sides of the pubic bone are quite different from one another. This actually happens to be a woman who had significant pain after a delivery and the radiologists were able to diagnose an internal fracture inside the pubic bone that explained the pain that she had. We hadn't planned to be able to make those diagnoses, but we've had three or four patients who had difficult deliveries and in whom these internal pubic bone fractures explained why they had so much pain afterwards. So again, you can see the obturator here a little bit lighter than the, excuse me, the levator here, a little bit lighter than the obturator. And here in the edema scan, you can see how it's quite different than everything beside it. So what were we able to learn? Well, six of the 19 individuals had a levator tear. Three of them were high grade, meaning more than half of the muscle was lost and none of them resolved. So that supported the hypothesis that levator tears are a mechanism that's responsible for this injury. When we looked at the compression hypothesis that there would be high signal intensity, all of the patients had edema in the levator. All of it went away by the late scan and it was never present in the obturator. So this then rejects the hypothesis that muscle compression is responsible for this levator injury. Delayed atrophy, seeing a normal muscle that shriveled was never seen in any of the 19. Now, this doesn't mean that there aren't nerve changes in the pelvic floor with delivery because we know from EMG studies that there is alterations in neuromuscular function, but it does say that it's not responsible for this visible defect that we can see in the levators. So let's talk a little bit about some of the adjuncts about how to get better scans. There are a couple of tricks that are helpful in specific situations. And in the top two images here, you can see on the left an individual who did not have contrast placed in the vagina. And in the upper right, you can see an individual who did have contrast placed in the vagina. And then in the bottom two images, you can see the same thing. So this is just ultrasound gel, the regular ultrasound gel that you use for transvaginal ultrasonography. And it actually shows up very well in the vaginal lumen. So you can see that although here you could kind of guess where the lumen was, over here you can see quite exactly where the lumen is. And this was very important to us when we were starting to look at some of the stress 3D MRI. Over here, you can see a lot of, you know, lines here and you're really not sure which one is the lumen. Whereas over here with the gel in the vagina, you can tell very exactly where the lumen of the vagina is. It's important when you're doing this not to put too much gel in the vagina because you can get a big bolus of gel up at the top, which distorts things some. But if you put about 10 to 15 cc's in and then distribute it with your fingers, you can get a nice image like this. You can also if you want to get really high resolution of specific structures like the urethra, you can use an endovaginal coil. And this is the endovaginal coil, which is a soft compressible rubber device that fits into the vagina. And if you compare what you can see in the urethra here with the coil and the same individual here without the coil, and then at higher magnification here, you can see a lot of detail here within the urethra. You can see the striated sphincter right here, starts here, ends here. You can see the circular striated muscle and you can see the longitudinal striated muscle, a lot of stuff. And you really can't see that in that kind of detail without the coil. And so if you have a specific question that you need high resolution, you can do things like putting a coil in the vagina to be able to see that. So let's talk a little bit about dynamics now. And dynamic MRI I think is the one that opened a lot of our eyes to the variability between different individuals. And I'm going to show you a few examples and I'm just going to talk through them while the video plays. There are three different subjects here. One is a woman with normal support and then two women who have prolapse and you'll see three valsalvas for each of them. And I'll just kind of run through those again, point out different things as we go along. So this is a normal individual, pubic bones over here, bladders here, uterus is here. Everybody can see that it's, you know, it's kind of funny. Why is this not all the same in the urine? And that has to do with the fact that urine actually can be moving and that affects the way that the MRI scan sees things. But let's just take a look now at what happens when we move. So notice when she pushes down that you can see everything in the abdominal pelvic cavity kind of pushing downwards. You can see the bowel up above the bladder and uterus. Here in an individual who has prolapse, you can see that everything including the uterus and the bladder move downwards. And here we can again see that the bladder and the uterus move downwards and you can see a little enterocele sneak down. So we're going to start over with the whole sequence again. Now look at the levator back here this time. Watch this, the levator plate as we go along. See how in a normal woman that holds itself in place. The vagina and the rectum kind of get pushed against it. Now look at this next one and see how it just moves out of the way and everything kind of plops down into that space that's been created. See how that has moved out of the way. Now look at this one. Notice that her levators don't move at all. They're holding the fort but everything is coming out in front of the perineal body. So let's take a look at it again. Here's a normal. Here's one of the prolapses, and both of these patients had about the same POPQ, this one and the other woman with prolapse. So even though the two of them, this patient and this patient, have basically the same POPQs, you can see that mechanistically one has an intact levator and one has a dead levator. And which one do you think is going to have a better long-term surgical outcome? I think it doesn't take many of us very long to say that we'd rather have the last patient in our surgical trials rather than our second one. So that's an illustration how what you can see may be different than what you've thought about before. Now let's take a look at what we can do once we start to say let's make measurements. So one of the things that you can do then is to look at resting anatomy and compare it to straining anatomy. And you can look at normal women because you're not doing an x-ray, so you can ethically get normal volunteers to volunteer for these studies and you can look at individuals with prolapse. I show this, this is just a screenshot of the straining images of 136 mid-sagittal MRIs at Maximal Valsalva out of the 600 examples that we have. And the reason that I show this is that when you're examining patients, you see a patient, you make some observations and then she's gone. But one of the brilliant things about MRI is that we can do exactly what I showed you before and you can sit there and study. Patient's not uncomfortable, you don't have another patient that's waiting, you can actually study in these 600 examples of normal individuals in every conceivable kind of prolapse what's going on to be able to see what the variability is and you can start to make measurements. So let me give you just one example. This is a paper that Anne Simarco, our now third-year fellow, published last year. And we were interested in this idea about the yielding of the levators and the downward descent of the levators. So what we did is to make a bunch of different measurements here. The levator hiatus, many of you know, has been measured on ultrasound and a lot of different things. The urogenital hiatus from the pubic bone to the front of the perineal body is another measurement. But we're actually also going to measure the area below the skip line, which is a reference, and above the levators to look at that downward descent. Now the levator hiatus tells us about the puborectal muscle and the urogenital hiatus tells us about the pubofisceral or the pubococcygeal muscle. The other thing that we were curious about is what changed more. What we have been doing is evaluating the red out here, what we see come below the hymenal ring. And in somebody with normal support, there isn't much below the hymenal ring. And over here in a woman with prolapse, you can see that there's a lot of stuff that's come below the hymenal ring. But you can also see that this area that represents the area above the levators and below the reference line also gets a lot bigger. Now this we tell with the POPQ quite easily and this currently is not something that clinically people measure or pay attention to. So just to give you some idea about the results that you can get from that, this is looking at that blue levator area here. You can see that in the controls, it's about 21 and in the women with prolapse, whether they had anterior prolapse or posterior prolapse, it's 35. So that's an increase of about 15 square centimeters over and above the normal amount that's there. And when you look at the protrusion area, there's only about an increase of 10 square centimeters. So there's actually more that goes on above the levator than there is that goes on below the levator. And these are just the amounts that things have increased that I said over here. So the levator area difference from the normals were 13 and 14. And the difference in protrusion area were only 9.3 and 9.5. So again, this has shown us that something that's going on that we previously haven't been able to measure. Now whether or not any of our surgical interventions alter this is an open question and one and I think we're very interested in trying to be able to look at in the near future. Because with both the NIH's clinical trials network using abdominal sacral culpapexy considered to be the gold standard operation as a 25% operative failure rate at seven years and our similar results with vaginal repairs have a similar failure rate. Understanding failures is really what the future game is going to be all about. So this is just a reminder that the dynamic scans are only as good as the instructions that you give the patients and we usually will rehearse the patients ahead of time to make sure that they know what the instructions are going to be. You want to give at least four good pushes to make sure that you get the prolapse to come all the way out. And then we always have somebody if it's a research MRI who's got the pop cue there to make sure that the prolapse develops itself completely. Okay. So let's talk a little bit about some of the tools that you can go as we start to move into looking in three dimensions. And 3D slicer which is a free software allows you to be able to look at these images in very unique ways. The first thing you can see is this is just a screenshot of one of the studies. You can simultaneously see the axial, the sagittal and the coronal images and then you can actually put those all into the same space so that you can see what it is that you can see in the sagittal scan. What does that look like for example in the coronal scan? And this is very useful as you're trying to sort out some of the off-scan planes. Through the Society of Gynecologic Surgeons we have a working group that's starting to get people's skills up on slicer because I believe that this is going to be a very powerful investigative tool for the people who learn to be able to use the software and make measurements. One of the nice features about slicer is that you can actually, as you see here in the 3D box, you can see that the axial plane over here has now been tipped so that it's actually going along the axis of the anal sphincter. Because if you cut straight across on the sagittal scan to look at with the axial scan, you'll really see that you're not in the right orientation to understand what's going on with the anal sphincter. But if you tip the plane to this angle and then view it here, you can see it along the muscle fiber direction. So a very nice feature of the new way that slicer is able to tip planes. So just a brief appetizer about the kinds of things that once you have the scan that you can do. As I said, 3D slicer is a free download. There's no cost to it. And once you have the scans, you can look at as many scans with it as you want to. Now, I'd like to talk a little bit about model building and stress 3D MRI, which has really changed a lot of the ways that we've thought about the pelvic floor. Now, the concept of stress 3D MRI is pretty simple. So first of all, let's talk about model building. So model building is really fun. What you do is you take an image and you simply outline it in slicer. And then each thing that you outline in slicer, if you outline it on each of the different slices, you can push the I'd like to see a model button. And in less than a second, it will produce an exact three dimensional model of the tracings that you've made. So with a modest amount of work, you can then have a 3D object that tells you exactly what the geometry of that individual is. You can spin it around, you can make things transparent, you can cut through it, a lot of different things you can do. But what we would really like to know, and this is the stress 3D MRI, is that we don't really want to look at it at rest. What we'd really like to be able to do is to look at it at Maximal Valsalva. And this took a little bit of ingenuity and working with the MRI team to be able to get sequences that could be taken in the length of time that a woman could hold a breath hold and hold her prolapse all the way up. But we've been able to do that. And we've done that in probably about 100 women now. And then what you have is a three dimensional model of the prolapse at Maximal Valsalva. This is tremendously powerful, as you'll see. If you can recreate what's happening in women with prolapse, you can then start to make measurements. So now we have the ability to compare normal to an individual who has prolapse and start to make measurements to test some of the hypotheses that we've all been wondering about for decades about what's really wrong inside. So to do that, you set up a reference system based on the bony pelvis. Here we put in a line that's basically where the arcus tendineus fascia pelvis would be. You can then say, okay, let's measure the length of the cardinal and the utero sacral length. And so we measure from their origins out on the pelvic side wall at the top of the greater sciatic foramen where the dots are. And we can measure down to where the cervix is. We can also measure the distance that the vagina is below where it normally should be. And I've just shown one location here, but we actually do that in five locations on each side. You can also measure the length of the vagina and you can measure the width of the vagina. And here again, we've done that at five different locations. So this is a normal scan. So we can get all of these data in 30 normal women. And then we can get those measurements in 30 women that have pelvic organ prolapse so that we can start to see what's wrong. So how would this help? Well, if you all took a bunch of different textbooks, you could see that there are many different thoughts that people have had about what this patient's problem is. Some people would just say that it's a stage three cysticeal. Other people would tell you that it's a midline defect in the fascia or a parafascial defect or an apical problem or levator injury. And I don't know about you, but I had back pain and I was really interested in whether I had a ruptured disc or not. And an MRI was able to show that there was a protrusion at the L3 or L4, L5 inner space. And they were then able to know what to do in that specific situation. So in order to be able to see what was going on, we took 30 women with anterior predominant prolapse and 30 women with normal support and made these measurements. So let's take a look first at what we found about the length of the vagina and the width of the vagina. So here we've got the length and we've got the width at five different locations. So the anterior vaginal wall is about 25% longer in the women who have a cysticeal than the normal women. So that's a definite difference. So what do you think you see in the women who have prolapse, excuse me, in the width? You can see a little bit of difference up near the anterior fornix, but which one is wider? It's actually the normals that are wider. And as you get down to the distal vagina, that goes away. Now, that's not to say that there aren't some women who have a wider vagina, but it's certainly not as impressive as what you see with the length of the vagina where there's a clear difference in the intended direction. So what about paravaginal defects? Well, this is someplace where we really now know that the paravaginal defect is proven. There are huge differences in both the apical location here, the apex being, you can see here about three to four centimeters lower in this group. And over here, you can see that the paravaginal location is quite a bit lower in all these individuals with P values of less than 0.001 in all of these different places. So is it a width problem? It's not a width problem. Is it a paravaginal location problem? It is a paravaginal location problem. And the paravaginal location problem looks very much like an apical problem. But think about this. Here is one person's paravaginal separation that's normal. And here is another person who has a cysticial. So that although the means are quite different, there are individual stories within these data. Now what that means is that we're now starting to be able to define a normal range and then say which individuals are outside of that normal range. So think back to the echocardiography example where you might have one person who has a problem with vaginal length but has a normal apical location and another person who doesn't have a problem with vaginal length who does have an apical support problem. Now you've got a number to tell you what to do. It's like having a number for the ejection fraction or for the size of the mitral valve or for the thickness of the ventricular wall. We can now start to individually say what's wrong. We can also look at surgical failure and this is work that Liu-Yung Chen in our group has been doing. And it's looking at doing MRI scans before and after an operation. So we took preoperative and postoperative comparisons. The majority had some form of apical suspension and none of the patients had a paravaginal defect repair. Now these were all repairs that were done by one of our female pelvic medicine subspecialty certified surgeons. All experienced people who have gain and there were a variety of techniques. Most of them being vaginal but some of them being abdominal but we just wanted to see what was going on. Now if we grade ourselves on the apex, we can see that these individuals had apexes that were below the normal range. The normal range being indicated by this dark blue which is the interquartile range or the light blue which is the range of the normal values. The dotted lines mean that they did not have an apical repair. The solid lines mean that they did. 90% of the patients were back inside the normal range postoperatively. If you look at the paravaginal locations, this is up near the anterior fornix. This is down in level three down by the urethrovesical junction. You can see that we get great success in restoring the paravaginal location at the top of the vagina but not so much at the distal vagina. I can tell you the reason for this is this is the area where the levator muscles have to hold the pelvic floor closed. If you don't have levators holding the pelvic floor closed, it's not as good a story. The other thing is that despite not doing a paravaginal repair by re-elevating the apex, the paravaginal descent was corrected. That just shows the intimate relationship between apical descent and paravaginal descent. What's the clinical relevance of this? As I said, if we just took apical location, for example, we know that apical location is responsible for 78% of what's going on with systoceles but we also know that not all of the people with systoceles have an apex that is outside of the normal range. This then would give you a test to be able to say who needs an apical repair and for whom would apical repair add morbidity but maybe not add anything in terms of success. Vaginal length is another way that this has helped us and it's kind of a conceptual help in that we had always been sewing from side to side during our anterior repair but once we started to understand that the vagina was too long in the people with a long vagina, we started sewing from the top to the bottom. What that does then is that shortens the length from the top to the bottom. And if you look at preoperative and postoperative, let me just move through this in the interest of time, length of the anterior vaginal wall, you can see that we were able to normalize vaginal wall length, something that for the first 20 years of my practice I never thought about but with the biomechanics we really started to understand played an important role as well. So, what have we learned? Well, I've come to realize that MRI is a very powerful research tool because it has moved us from the era of opinion when all of us, me included, loved to guess about what we thought was wrong and now we can actually do a study to find out what the truth is of these different hypotheses. We're shifting from what fell down to why did it fall down, what was broke that resulted in that and using now moving towards an echocardiography model where we actually say what's wrong in each person. Is this somebody with congestive heart failure because of mitral stenosis or whether this is somebody with a very thin and a contractile ventricle. I don't think that MRI is ready for full-time, prime-time and clinical practice but I can say without any question that despite being a very experienced surgeon I've changed what I do over the last 10 years by the lessons that I've learned from the research that we're doing. We've published with native tissue vaginal repairs long-term success rates in large studies that are equal to that of mesh-augmented repair and I think some of that success comes from knowing exactly what it is that we're trying to do. Ultrasound clearly can do some of this and I think that over the course of time ultrasound will get to be more sophisticated and be able to make more of the measurements that we're currently doing. Well I hope like me you're curious. I hope that like these pilots that are trying to understand what's going on that you find that these kinds of approaches to things are useful and that in the future as you see more information coming out from MRI scans that you'll be able to understand how that plays into this bigger picture of what's going on with the pelvic floor and that all of us can look forward to a day when we have a specific treatment plan for each person that's based on the specific measurements that they have so that we're not just guessing anymore about what's going on but we actually know. Well again I wanted to thank Augs for the chance to be able to present this and then Chris I know is going to be looking at the questions that you type in and then he'll be able to relay those to me so that I can answer them with the audio. Thank you all very much. And Chris are you there? Yeah I am here. Thank you so much Dr. DeLancey for your presentation. We do have a few minutes for questions and there have been two submitted so far. The first question comes from Cedric Oliveira and he asked about why compression would cause edema. I believe this is in reference to the study you were discussing earlier that you did with Janice Miller in 2010. So if you could provide an answer to that. Yep and good question Cedric. It's actually the thing that causes the edema in the levator is actually stretching and the levator gets stretched but the obturator internus doesn't get stretched so that's why you see it in just the levator and not in the internal obturator. I'm sorry that I hadn't made that clear. Good question. Excellent and I have another question from Caroline Elmer-Lyon and she was asking a question specifically about the ultrasound gel that you used as a contrast for MRI imaging and she wanted to know what the best way was to insert the gel so that it reaches all the fornices. Yeah we just take a catheter tip 60 cc syringe and put depending upon how big you think the vagina is between 10 and 20 ml just hook onto the end of that an 18 or a 20 French red rubber Robinson catheter you cut about two-thirds of it off so you've got the red rubber Robinson on the catheter tip syringe. Try and be careful not to have too much air in it and then you just put the red rubber Robinson into the vagina and inject and start up at the fornix and then pull the catheter out as you're injecting and then once I've done that I usually just put my finger in and distribute the gel so that it distributes itself in the vagina a little bit better. So really easy very well tolerated and easy to get the gel. Great. I have another two questions here. Next one comes from Cherelle Carter-Brooks here at Pitt. She wants to know how you've changed your surgical practice based on this information thus far. I know you've alluded towards the end of your talk there how it's been helpful with your native tissue repairs but if you could go into a bit more detail on that. Well I think probably the most specific thing is how much attention we pay to returning the vagina to a normal length and we actually oftentimes will make measurements pre-operatively and post-operatively and it's surprising that once you get the hang of it just by using either those side-to-side stitches or the top-to-bottom ones that you can achieve a normal six and a half centimeter anterior wall length. I also have reduced the number of people that I do apical suspensions on just with the idea that there are many people who have normal apical support and Pam Fairchild published our series of about 200 individuals who we did a vaginal hysterectomy anterior posterior culporaphy and did not use either a high utero sacral or a sacrospinous ligament suspension and compared those to people who did have a high utero sacral and or a sacrospinous to make sure that we were not getting worse results by selecting these patients appropriately and there were about I can't remember exactly but about 200 in each group and the results were exactly the same that if you choose the patients appropriately you can avoid the ureteral injuries of the high utero sacrals and the potential bleeding and neurologic risk of the sacrospinous ligament suspension. Now that's obviously not everybody but the people you know my rule of thumb cut off is if the cervix is not more than four centimeters outside the traction in the operating room and if the ligaments look pretty good then shortening the ligaments and reattaching them works with a little bit less morbidity. I guess that's the other thing. There are a lot of frustrations that I have and then if I'm operating on somebody with dead muscles I don't have a way to fix that and so that hasn't really changed my practice but it is something where I have a little bit better sense about who's not going to have as good a result. Good question. Yeah. Here's another great question from a Tariq Khalif asking if you would agree that for the time being that MRI would be used better for prolapse failures or would it perhaps be more helpful to use it to evaluate prior to doing the initial prolapse surgery for the index surgery. Yeah. Yeah. It's a good question. You know part of it is what are you going to ask what questions you're going to ask at the MRI scan. We have thought in the beginning that levator injury was going to be a strong predictor of operative failure and it turns out that there's no question that levator defects are associated with prolapse but we're not seeing that levator defects are highly associated with operative failure and I think it intuitively makes sense but on the other hand you know I have to be transparent and say I don't get MRIs on my clinical patients at this point but I think probably as we start to go further it will be things like seeing what is specifically about that person that wasn't corrected that needs to be corrected. And then one last question John from Andre Player asking about if the increased detection of paravaginal levator defects led to any specific surgical correction of the levator ani such as more paravaginal suturing or levator placations in your surgeries. Yeah. I have had many different phases in my career and there was a phase where I probably did 400 transabdominal paravaginal defect repairs so I know that operation well. I also was honest with myself in that when I had done an operation for prolapse and not done a paravaginal defect repair that the results were pretty good so I actually wasn't surprised that not doing paravaginal stitching didn't result in a high failure rate. So I think that's one of the things is that the paravaginal defect in my mind is a result of apical descent that as the apex comes down the paravaginal gap widens and that if you get the apex back up the paravaginal gap is returned to more normal. I do think that there will probably be a subset of people who have levator injury where we continue to have problems in level 3 and level 3 I think is the next horizon in terms of our understanding because it's in my practice despite, you know, I've done 2,000 posterior culporaphys. I know the anatomy really well. I personally consider that I'm probably at least an average if not a better than average surgeon and the place that we don't get back to normal very well is getting the hiatus returned to normal despite putting stitches in to try and narrow the hiatus. We certainly improve it but I don't think that we get it back to normal and I think that's the area that at least our group is starting to look at to say can we learn a little bit more about that. It's a good question and one that I think we all need to think carefully about. Excellent. Well, that's all we have time for. So, on behalf of the Ogg Scientific and Education Committees, I would like to thank Dr. DeLancey and everyone for joining us today. Our next webinar will be Sexuality in the Aging Female presented by Dr. Carol Kuhl on May 23rd next month. So, with that, thank you so much everybody. Have a great evening. Thank you everybody and thank you Chris. Thanks John.

pelvic floor MRI

pelvic floor disorders

MRI scan sequences

contrast gel

dynamic MRI

levator ani muscle tears

paravaginal defects

3D model building

stress 3D MRI