Hello, everybody. My name is Niki Karavitaki. I'm an endocrinologist from the University of Birmingham and the Queen Elizabeth Hospital, Birmingham, UK. I have the pleasure to co-chair, along with my colleague, Dr. Ismat Shafik, endocrinologist from the University of Rochester in New York, the first of 2021 webinar series organized by the Adrenal and Pituitary Special Interest Group Committee. So a very warm welcome from both of us and from the whole team. This webinar is devoted to Cushing's, which is always a hot topic. And today we will cover the often painful journey of diagnosis. We have two eminent speakers who we will introduce to you in a second, Professor Johnuel Price and Professor Roberto Salvatori. Just to remind you that during the presentation, we would like you to put questions and answers through the Q&A facility. We will work with these questions through after the presentation, after each presentation. And for those questions who have not been answered, then Drs. Johnuel Price and Roberto Salvatori will be answering them in the discussion forum of the Community Connect for two days after the end of this event. So now time to move on. And it is indeed a great honour for me to introduce Professor Johnuel Price, a well-known expert in the field. John is a professor of endocrinology at the University of Sheffield and the Sheffield Teaching Hospitals NHS Foundation Trust in the UK, where he's the lead for specialised endocrine services and for the Sheffield INED European Centre of Excellence for Neuroendocrine tumours. His research interests include glucocorticoids, pituitary and neuroendocrine tumours, and he's widely published in these areas. He was a senior editor of clinical endocrinology and also a co-author on clinical guidelines for the diagnosis of Cushing's, the treatment of Cushing's for the endocrine society, and for adrenal incidentaloma for the European Society for Endocrinology. He serves on the Endocrine Society's Board of Directors and he was the chair of the Annual Meeting Steering Committee for ENDO 2018 in Chicago. So John, thank you very much for accepting our invitation and we are all very eager to learn from you today. Thank you. So Nikki, thank you for that very kind introduction and my thanks go to you and ISMAT and the Endocrine Society for setting this up and asking me to speak. I expect that there are people from all different time zones and different areas. It's the most beautiful day in the UK at the moment, so I hope it's a beautiful day wherever you're watching this from. So I'm going to be talking about the diagnosis of Cushing's syndrome and this will lead into Roberto's talk. If I have the control of the slides. Okay, so here are my disclosures. So the learning objectives that I'm going to speak to are what are the clinical features that you can use in your day-to-day practice to help diagnosis? Which are the patients that you should be considering for testing for a diagnosis of Cushing's? And then to go in some detail as to what are the pitfalls of testing and of the tests that you use and which tests you use and then what are you going to do with the information you get from those tests? So my first question to you, and this is just something for you to be thinking about whilst you're looking at the screen, is a consideration. If I were to show you this slide, do you think this person's got Cushing's? So it's a question and I'm going to come back to this question and answer the question right at the end. I just want you to hold that image in your mind as we go through the talk. And the real key principles when one approaches the diagnosis is it has to start with clinical suspicion and sometimes I feel this step is missed. But the first question is does this, whoops let's just go back. I'm not sure I can reverse the slide slightly. There we go, great. So is there a clinical suspicion? And if there is a clinical suspicion then you're going to consider this patient may then be tested for endogenous hypercholesterolemia or Cushing's syndrome which of course is the prolonged, protracted, inappropriate high levels of circulating cortisol in an individual. And then only then if you get this established would you then move to the cause. And the critical thing is you do not move from step two to step three unless you've absolutely established the fact that the patient has Cushing's syndrome. And you don't cross that. It's an absolute issue. You do not cross from one to the other. The reasons why, well if you do that you're going to get into an absolute mess because in patients who do not have Cushing's there's around about a 10% rate of pituitary incidental illness. And all of the different tests, differential diagnostic tests, CIHS, whatever you choose actually are validated for patients who have proven Cushing's syndrome. So I implore you not to go from point two to point three unless you're absolutely sure of point two first. So the issue then is, struggling a bit, there we go, is in terms of just thinking about Cushing's I just show this table to remind us that we divide Cushing's into ACTH dependent in the the top row of this table and ACTH independent in the bottom row. We're going to be focusing on ACTH dependent for this afternoon. As you know the vast majority of these patients have Cushing's disease from the pituitary and there's a significant female preponderance in patients with Cushing's disease. Interestingly that's reversed in the pre-pubertal era with a preponderance for boys compared to girls in the pre-pubertal presentation of Cushing's. We're not going to be discussing paediatric Cushing's today. So I just illustrate that because of course these different ratios need to be thought about a bit when one is considering the likelihood or otherwise of different diagnoses. So I'm going to present a case and then after this case I'm going to go through it with asking you some questions as we go through so you're about to get to vote. So this is a 54 year old man. He presents with new onset type 2 diabetes and is managing pretty well with that by diet. However he has had hypertension for many years and he drinks two or three large glasses of wine most nights and at some weekends and he's previously been documented to have some chronic remand band. These are the medications that the patient is taking and you'll see here that he's on a proton pump inhibitor, a thiazide diuretic, a beta blocker, aspirin and a statin. And then when he was referred I saw this man and examined him and when I did so, I'm trying to get the slide to advance, really he appeared to be pretty well but there was truncal obesity and his arms were really quite thin as were his legs and he had quite a lot of bruises. So with that background there is a question here is do you think with that background this is someone who you should consider for testing for Cushing's and this is going to be a binary question so this is a yes no answer which you can now vote on. We'll hold it just for a few seconds whilst you get a chance to vote. So we've probably had nearly sufficient time I guess. So we have the results of the vote. So here we go so that's interesting. So the majority of people feel that they should be but there are some people, 14% of people also who felt that actually there wasn't a need and so the question then is well what would trigger the idea of actually looking into this person and so when I saw him I felt that the signs of protein wasting in his arms, the truncal obesity and particularly the bruises persuaded me there may be something going on. Now this isn't the case of course for every single patient with type 2 diabetes or indeed every single patient with hypertension. I haven't shown you a picture so those 14% who didn't feel that he should be investigated that's okay. It's a question of what you personally think the index of suspicion is for the diagnosis of Cushing's. I'm afraid these slides aren't advancing on this remote so advance. So then the question is well if you have or you're going to do these testing what are you going to do with the results because of course it's important to pre-empt what you're going to do with the results depending on what they're likely to be and I'll come back to that and illustrate what I mean by that shortly. So it's important to consider what influences a test performance and that is the prevalence of a condition in the test population. So for Cushing's the prevalence is vanishingly low in most patients or most places you're going to consider it and by definition the tests are then up against it in terms of their ability to be able to discriminate between Cushing's and non-Cushing's and that is important because how likely is a Cushing's present well that's depending on your clinical judgment. How good are the tests or what are the characteristics of the test you're using? Are you using the right test in the right person? And then of course there are patient characteristics which may influence the way that the test actually performs. So there are of course important things to remember that you can get hypercortisolism without Cushing's syndrome so I list those here. So they definitely have hypercortisolism. The amount of cortisol in their bloodstream is high and yet they do not have Cushing's and they have these different conditions that account for that. But they may of course have quite a few symptoms or clinical features that may look similar to Cushing's and in contrast you have other patients which I list here which have a variety of other causes for hypercortisolemia but actually it's very unlikely that they'll have the feature of Cushing's. So it's really this top panel that causes diagnostic difficulty and here you'll see particularly depression, some alcohol dependence but really poorly controlled diabetes with morbid obesity. The type of patients who you worry might have Cushing's but of course chances are on a probabilistic basis still are relatively small. So who are you going to test? Well we discussed this and thought about it very carefully in the Endocrine Society guidelines which are now 13 years old but actually I think these recommendations still hold true. So the recommendations are to test in patients who have unusual features for age. So for example the young man in his early 20s who has some osteoporotic crush fractures of his vertebra would be really worrying. Patients with multiple and progressional features say multiple features that fit with Cushing's particularly the ones which are more diagnostically discriminating and I'll come back to those. In children there are very few things which mean that you put on weight and decrease your high percentile. In fact it's virtually pathognomonic of excess cortisol. And patients with adrenal incidentaloma compatible with adenoma which is worth testing but is an entirely different talk to the one I'm going to give this afternoon. And the recommendations are not to do widespread testing in many other patient groups unless there are clinical features which would suggest it. Before you test it's really important you have a thorough history and examination and really important that one looks for and asks about excessive exogenous glucocorticoids. So for example I have had people present who have been taking excess hydrocortisone. One measures that in the bloodstream and it is cortisol and they have a cushionoid appearance. That is a real diagnostic challenge whereas of course if someone presents extremely cushionoid and their cortisol and their cortisol is suppressed that automatically tells you that they've got exogenous unmeasured glucocorticoids. And very importantly it's important to check for a history of cyclical disease because that will influence your interpretation of the test and indeed may influence your interpretation of the outcome of treatment that patients get ultimately with pituitary surgery etc. So what are the discriminatory signs? So this is taken directly from the guidelines that you may or may not be familiar with but really if we look of the discriminatory signs and these are the ones to focus on. Facial plethora, proximal myopathy, continuous stray and bruising and in children this weight gain and reduced height percentile. And actually it wasn't really emphasized but I would say that thin skin really is a very good discriminating signs and they're all down to protein wasting. So if one considers the signs here these are all down to protein wasting as a consequence of a high level of cortisol. And we've got all these other symptoms and complications which are less discriminatory of course if they all tend to be or if they are all present that is further evidence but it's really focusing on these discriminatory signs. And in symptoms as I mentioned the osteoporotic scrush fracture particularly in the young man that would be a really worrying feature that may prompt investigation all of itself but multiple features in the same person raises the probability that this person has excess cortisol in their bloodstream. And I just show here skin fold thickness. Patients with Cushing's skin fold thickness measure in millimeters. It's often easiest to measure just looking at the back of the skin just by pinching the back of the skin and trying to measure it. Some people actually get out calipers to measure it. I just say it's worth just measuring it. I can usually just do it by by pinching the back of the hand or on the back of the index finger. And you'll see that in many other conditions which may to certain extent resemble Cushing's really the skin fold thickness usually is thicker. So skin fold thickness is a useful sign in my opinion. I illustrate here some really nice data that came from Francesco Aguri-Giraldi and others working in Milan now many years ago but still to highlight some of the particular things which appear to be a bit more common in men. So men appear to get more in the way of the of striae, more in the way of muscle atrophy and more in the way of osteoporosis. Just go back one. So what about type 2 diabetes? Well of course the patient I've described to you has got type 2 diabetes. There have been numerous studies looking for Cushing's in diabetic populations. I think I can summarize by saying that actually the likelihood in completely unselected patients with type 2 diabetes, the likelihood of finding Cushing's just by screening is very very low. Hence the reason for the recommendation not to do indiscriminate widespread screening. And really the same holds true for patients with obesity. Many many studies have been done. I think the biggest type of studies have been done really have failed to identify Cushing's unless there are other features. So again I think this emphasizes the recommendation to be screening for Cushing's in patients in whom you consider it is possible or indeed likely that they may have Cushing's. So this algorithm again appears and I suspect is well known to people. We're just going to walk through it. So if you have someone, whoops go back, I'll just go back one. In someone who has got suspected Cushing's without exposure to exosgenic glucose, the recommendation is to perform one of these tests. I'm going to go through these tests in more detail in a moment. And if that test is normal you're pretty likely to have excluded Cushing's. But if it's abnormal you're going to choose another one of the tests and if they are concurrently, if both are abnormal, you effectively have established a diagnosis for Cushing's syndrome. If they're both normal then really you've excluded Cushing's. But if one's normal and one's abnormal or one's very close to being normal you've got some discrepancy and you're probably going to go around this circle again and repeat the tests at an interval but not cross to the differential diagnosis until you're sure the patient has Cushing's. And I'm going to come back to this recommendation here because I'm not convinced that this recommendation should remain as a recommendation anymore in light of data that have been gathered since the guidelines were actually published. So let's just have a look at how these tests, these tests, a 24-hour urinary free cortisol which is giving you an integrated measure of the amount of cortisol in the bloodstream and the free cortisol saturates the cortisol binding spills out in urine or the one milligram overnight dexamethasone suppression test or the 48-hour two milligram per day dexamethasone suppression test or the late night salivary cortisol. How do these, how do these tests do? And so I think this has been very nicely addressed by this, this very nice piece of work that came from the Mass General Group looking at how the different tests. So on the left hand, far left hand column we've got urinary free cortisol, late night salivary cortisol, dexamethasone suppression test, the 48-hour dex test, midnight salivary cortisol, and dexCRH. And I've highlighted here the sensitivity and the specificity of these tests. And the first thing to say is they're all pretty sensitive tests. There isn't a vast amount to choose them. You can see that actually the one milligram overnight test is really very, very sensitive, slightly less specific than some of the other tests according to this analysis. So really that is a very good test. The other tests are also good. Arguably the urinary free cortisol is somewhat less sensitive than the other tests. So if you've got mild disease, it's certainly true that you find that on occasion, or more than occasion, the urinary free cortisol value is actually within the normal range of the particular assay being used. But I also emphasize just here, the fact that this dexCRH test, if you look at it, has a similar sensitivity, but actually is less specific than the standard testing. And I have to say in my practice, I do not use the dexCRH test. And I know that it's used by many people. But I think these data really support the notion that you're probably not getting any or really much added value from the dexCRH test in these patients. So overall, these tests have high sensitivity, a lower specificity. And of course, it's that specificity which is difficult to deal with, particularly when you're considering whether someone who has perhaps less severe clinical features may or may not have Cushing's. So talking to that is this particular slide. And I'm going to walk this through, because I think it illustrates a really important point. The important point comes back to this clinical assessment that's going to influence your decision as to how likely it is the patient in front of you has Cushing's. I'm going to show you three scenarios. But you have to remember that in each scenario, the test is the same. The actual test is performing the same. But you'll see that the outcome of that test is entirely dependent on what you think is going on before the test is done. So if we start here, this is a Fagan nomograph. And on the left hand part of this, you've got what's called the pre-test probability. This is the clinical judgment that you feel the patient does or does not have in the way of Cushing's. And in the middle here is what's called the likely odds ratio of the performance of the different tests, overnight, DEX test, urinary free cortisol saliva, pretty much together for either a test that's positive for a diagnosis of Cushing's syndrome or a test that is negative for Cushing's syndrome. So let me illustrate this. So imagine you've got someone, you're pretty sure they've got Cushing's. You've used your clinical acumen, you've thought about the different various clinical features they have, and you're really pretty sure they got Cushing's. If you do one of these tests, and I'll do it, I'll put the boundaries of the probability for the urinary free cortisol value, you're pretty much home and dry. You are almost certain that this patient has got Cushing's. But the issue is if they test negative, then you don't completely remove the chance they got Cushing's. In fact, it's still quite a high probability. If you think they've got really strong chance of Cushing's, even though this test is negative, they may still have Cushing's. Let's just take another scenario. So here we go, where you've got someone you're not sure, 50-50. Remember the test, nothing's changed, same patient, you decide differently. If they test positive, well, they probably have got Cushing's. But if they test negative, although it's a good chance they've not got Cushing's, you haven't completely removed the possibility. And then I think it's most useful then to think about the 10%. That's probably the most common patient that we're going to face was the patient who has got poorly controlled diabetes, they have hypertension, other things. You're really keen not to miss that patient with Cushing's amongst all these things, the 10%. What happens if they test positive? If they test positive, you're in a real potential mess because you haven't confirmed Cushing's, and you haven't really, you know, they may not have Cushing's. But in this patient, if they test negative, that's extreme use of information, because you really have got a very strong chance of excluding. So the test in each of these scenarios is the same, it's just the difference in the way that you've assessed the patient, which then influences how you're then going to interpret that test. And I think it's very important to remember this, because it influences what you're going to do with the information after the test has been done. So for example, in this patient, if they test negative, you're not going to say, well, you don't have Cushing's, definitely not. You're then a bit uneasy with that test result. Do you repeat it? Do you think of another test? And in this patient here, if they test negative, you're probably going to say to them, I'm pretty sure you've not got Cushing's. But before you do the test in this patient, you may need to go through where the doubt may look, in the sense that if they test positive, it doesn't mean they necessarily got Cushing's, because you need to manage that patient's expectations. And I suspect many of us familiar with patients have been to see many, many people, they've had lots of tests done, ultimately don't have Cushing's, but they've been through many people having borderline positive tests. And that's where some of the issues lie. So managing that patient expectation, and indeed managing your own interpretation of the test is crucial importance. So going back to my question, I'm now going to give you some data. So here are some data for you to interpret on this particular patient. And you'll see here that the patient's creatinine is elevated, giving him an impaired EGFR of 40, whereas the normal is 80 or 90. I've given the units for cortisol in both micrograms per deciliter and nanomoles per liter. So the nanomoles per liter in brackets, and I've given the normal ranges here. And you'll see that on the dexamethasone suppression test, the baseline cortisol was 18.7 micrograms per deciliter or 15, 516 nanomoles per liter. And after 0.5 milligrams every six hours for two days, you could have just done an overnight dexamethasone suppression test. In this circumstance, it's a 48-hour test. There was a failure of suppression. It so happened because it was a 48-hour test, there was a cortisol measured at the baseline. Of course, we wouldn't normally measure ACTH until one has actually confirmed the diagnosis of Cushing syndrome. And you wouldn't normally be measuring ACTH on an overnight dexamethasone suppression test. However, we have the information here that the ACTH is not suppressed. So my question is, with these data, which are in front of you here, what do you think is going on? And what is the potential likely diagnosis? And the options are one, depression, two, alcohol-induced hypercortisolemia, three, renal impairment, four, Cushing's disease, and five, ectopic ACTH syndrome. So maybe a moment more on the result. There we go. Okay. So I think that's an interesting thing. So one person thought depression, I mean, of course, yes, depression can cause things like this. There wasn't a history of depression, but I, and so I think depression is unlikely. 15% thought this could be down to the alcohol. So alcohol stimulates the HPA axis, and therefore you can get ACTH-driven hypercortisolemia. 20% felt that this could be down to renal impairment. So certainly, renal impairment can affect dexamethasone suppression testing. Around half the audience felt that this could be Cushing's disease. And a few people thought, well, it could be ectopic. And of course, the issue is that, yes, you can get patients with ectopic disease that looks very similar to Cushing's disease, although, of course, typically, ectopic ACTH is more florid, although it is a possibility. So if I could close those. I felt that with those results, but I'll move on a slide. Claudia, it's not advancing again, let me just try again. There we go. I felt the most likely with those things is either alcohol, that could be doing it, or the renal function or Cushing's disease. I don't think it's depression. And actually, the features were just not really consistent, except that ectopic ACH is a possibility. But if I was trying to assess from a probabilistic viewpoint, those were my leading things that I was thinking was going on. So the question now is, what do you do? Well, you've got a problem on your hands because urinary free cortisol can cause false positives. And he hasn't yet, we haven't done any in this patient. I list here the fact that if you've got very high urinary volume, you can get a false positive. There's a problem with some immunoassays cross-reacting with synthetic glucocorticoids. If you're using an HPLC assay, you can cross-react with this, but increasingly people are using time and max spectrometry, so the precision and the accuracy and the specificity is very, very high. And of course, if there are drugs and compounds that inhibit 11-lead HST type 2, these aren't used very much, but that can actually cause a problem with excess cortisol in the urine. The causes of false negatives, as well, you need to do more than one collection to avoid missing mild disease. So if you do two or three collections and they're all normal, then that would be very reassuring that we recognize that you may miss mild disease with urinary free cortisol. And of course, if you've got an impaired renal function, you may well then reduce the urinary free cortisol. And we recommended the guidelines that if it's less than 60 mils per minute, that may then actually significantly reduce your urinary free cortisol and thereby reduce your sensitivity. And we also recognize in work that Stephen Peterson led that there's a big variability, around about 50% variability in the urinary free cortisol values in patients with Cushing's disease. So those are some of the pitfalls of urinary free cortisol. And it's important to check creatinine and volume to assess the collection adequacy. And the advice is to use the upper limit of your diagnostic range. And of course, that depends on how that's been established, but you need to know what the actual range is for your local assay. How about dexamethasone suppression tests? Well, they come in two flavors, the one milligram overnight test with a milligram at 11 o'clock at night and serum cortisol measured at nine. Or the classic test as in this patient had, who had 0.5 milligrams every six hours for 48 hours. In either case, we're looking for a serum cortisol for a normal response of less than 50 nanomoles per liter or less than 1.8 micrograms per deciliter. And we recognize, and I'm just showing you data showing that it's a very sensitive test. So probably only about one or 2% of proven Cushing's disease will suppress the values less than that. But of course, in some assays, that's quite near the threshold. I think this statement is probably less true increasingly because assays are improving all the time. If you've got drugs that induce or inhibit P450, CYP3A4, then you will get either false positives or false negatives because you're gonna either inhibit the metabolism or increase the metabolism of dexamethasone. And estrogens taken orally particularly will increase cortisol binding globulin and therefore give you a false positive. And these need to be stopped around six weeks before doing dexamethasone suppression testing. How about salivary cortisol? Well, I showed you that salivary cortisol is another good test, but just to here is a graphical illustration that it's not perfect. In patients with different conditions, there is overlap with patients with Cushing's syndrome. So it will be very, very sensitive, but there will be some overlap. It's not gonna be at all 100% specific. And another issue with salivary cortisol is this very nice paper that came out now, the best part of 15 or more years ago. So these are patients who do not have Cushing's. And these are different thresholds for the late night salivary cortisol shown in these different colors. And this is the whole group. But if you then break it down in these groups of patients, 187 patients, if they have diabetes or they become older or they're diabetes and they're older or they're diabetes, they're older and they have hypertension, the numbers who've got an abnormal serum cortisol generally increase by whichever threshold you use. And so if you like, this is a health warning, that using salivary cortisol is a very good test, but it becomes less accurate in the more elderly population and as other comorbidities increase. So for the young 20-year-old, it's gonna be a very good test. For the 80-year-old, it's just not gonna be as good a test. And so overall, it's important also to remember that renal failure is associated with a blunting of the normal circadian rhythm, and therefore you have higher level of salivary cortisol than you would have controls. So with those things in mind, people always ask about the DEXA-CRH and decimopressin test. They ask about the word pseudocushions. You'll notice I'm not using that word because we don't use it in the guidelines. We would use the word of non-tumorous hypercortisolemia. They have different cutoffs. The responses vary. There's a difference between the responses of human-sequenced CRH versus ovine-sequenced CRH. Ovine-sequenced CRH is no longer available in the United States. There's some issues with other medication. And really, the cutoff, which originally was described as 38 nanomoles per liter, that's below the dexamethasone suppression test already. So the only people this could be of value on is the few patients with Cushing's disease who have already suppressed on DEX. If you haven't suppressed on dexamethasone, there's no actual value to this test. And really, in answer to the question, I see there's already a question in the Q&A, how would I use this test in pseudocushions? Firstly, I won't call it pseudocushions. Secondly, I don't use this test. I don't see that it's got added value. And the recent analysis published last year would, I think, support that. And I accept many centers and many experts will use this test. I don't have a working experience of this test, and so I won't be able to speak to it. I haven't got working experience in using the dexamethasone test, but it is widely used in the context of trying to ascertain whether there is tumorous versus non-tumorous hyperplasmia. But I don't have any practical working knowledge of it myself. So let's go back to this patient. So how on earth are we gonna approach this patient? We've already decided to go down the route of investigation. We're worried it could be down to alcohol. We're worried it could be down to the renal function. We're worried he could have Cushing's disease. We don't think he's got depression. Possibly he could have ectopic, but less likely. So what do we do in this circumstance? And you might say, well, do a survivory cortisol one. And that's true. We could do a survivory cortisol, but we recognize in the data I've just shown you that with comorbidities and renal impairment, the survivory cortisol may give you a false positive. If you look at these, you say, well, his EGFR is 40, and that's below the recommended level. I mustn't use urinary free cortisol. Well, I thought about this, and actually that's exactly what I did use. And my reasoning was is that if the urinary free cortisol is still elevated despite an impaired EGFR, that would be very useful information. So I did four urinary free cortisols, and on every occasion, urinary free cortisol was elevated. So this is where I'm using a test, but I'm thinking through how I'm using it and in which place I'm using it and what I'm gonna do with the information. If the urinary free cortisol had been normal, that wouldn't have been helpful to me, but as it turns out, it was. And the other thing is I persuaded him to abstain from alcohol entirely for three weeks and repeated the dexamethasone suppression test. And you'll see here that the baseline value was a little bit lower, and actually there was better suppression, but you'll still see there was a complete lack of suppression. And so what I would say from this is that this is consistent with endogenous hypercortisolemia, ACTH driven, because we know that his ACTH was measured, but at least some of it was being driven partly by alcohol, or he's got a bit of cyclical disease, but we've got concordant abnormalities in a patient in whom I made a clinical judgment that there were abnormalities. And in fact, this patient did turn out ultimately to have some abnormalities. So my question then is on the basis of this, what's the diagnosis? And it's exactly the same question again. And we were gonna vote on this, but I think in view of the time, we should just probably move on. And I'm gonna suggest to you that actually, we just move on to the next slide. It's not working. That actually the diagnosis is not depression. It's not alcohol induced hypercortisolism. It's not down to the renal failure. It is most likely Cushing's disease. I suppose there's still this chance that it could be ectopic ACTH syndrome, but really through using that test, I was happy that it was one of these. Actually, it turned out to be histologically proven Cushing's disease ultimately. So let's just go back to this one slide that I showed right at the beginning. And the question here you may be thinking is, does this person have Cushing's when you're looking at it? And I show this just to illustrate how important it is to keep clinical context and consider hypercortisolemia. Here is the explanation for this appearance. And you'll see here, if I click, hopefully, I'm afraid the slides aren't advancing, is this was a poster I saw on the London Underground, the Tube, shortly before I was giving a talk at the Pituitary Society. And I thought that made a great illustration because this was in fact a campaign by the company Mothercare for body proud mums. And I was looking at this thinking, actually, that makes quite a good illustration of the striae that one gets in pregnancy, which is a hypercortisolemia state. But of course, that's physiological hypercortisolemia with this proud mum with a beautiful baby who's just been born who definitely does not have Cushing's. And so I'd like to conclude that the tests are imperfect. You need to think probabilistically. The clinical assessment is really critical. Focus on those discriminatory signs. History and examination need to be very robust. You may need to employ the diagnostic dimension of time. So if the patient is clinically mild and you're unsure, far better to repeat the test an interval or ask someone else, do not go down the temptation to start ordering MRI scans, the temptation to start doing any other differential diagnostic tests. And once Cushing's confirmed, well, if it's confirmed to be ACTH, so on the algorithm, you would then check ACTH and if it's detectable, you've got ACTH and Cushing's, then you would move to differential diagnosis, which is going to be the subject of Roberto's talk. And with that, I thank you for your attention. Thank you, John. That was brilliant. We already have a lot of questions. We have around 10 minutes to go through them. And as we mentioned before, those that will not be covered, hopefully in the next couple of days, this could be done online. So let me start with the first question that came up. Could you comment on how you would approach in terms of workup and diagnosis, cyclical Cushing's? So the short answer is that's a long question and it's not an easy question. And in fact, I was asked to give a talk on that for the British Endocrine Meeting, gave a whole talk on it. So in terms of cycle, the first thing it is to establish in the history, is there really any evidence of cyclicity? So are you seeing real things within the history and or some of the diagnostic tests that other people may have done that would fit with it? So for example, the patient who presents with hypertension that improves, the patient who has sort of transient impairments of the blood glucose or not, or people who present with really profound changes in their mood or their cognition or other things. You're looking for something really to focus on that could be cyclical. I think this word cyclicity is difficult because I think true cyclicity, where you go from very high levels to very low levels or indeed adrenal insufficiency is pretty uncommon. Variable levels of course is very common indeed, but that's not true cyclicity. In women is the changes in the menstrual history and that type of thing. So that's on the history. Of course, the examination, if you happen to be seeing patients, they may not necessarily have such severe signs and that's a challenge. How do I approach it from if then I feel there's enough evidence to investigate, I think salivary cortisol is a very useful thing because of course the patients can have these at home. They can measure them over quite a number of days. They can send them in the post and you're able to pick up a picture of whether or not there is a real evidence that there's hypercortisolemia. Again, there's a warning here. So I have been referred patients who have had 200 late night salivary cortisols, one of which was positive. That patient does not have cyclical depression. So I think there needs to be some common sense applied as well. I think salivary cortisol is useful. Dexamethasone suppression test is a bit of an issue and probably not as useful because you may just be expressing out when the patient's cycling out. So there isn't a straight and easy answer. I think it's a question of really listening to the patient. And then I think of the various tests that are used, salivary cortisol has got a lot of merit. Great, thank you. Another question, how useful is basal cortisol for the diagnosis of Cushing's? Well, if it's very high, it's very useful. So if you've got someone who's not otherwise acutely unwell and they've got a serum cortisol of a greater than a thousand and they've got some features of Cushing's, I don't think you need any tests at all. If someone's acutely unwell, of course there's a problem there because of course the axis will be elevated. If someone's got florid Cushionoid features and the cortisol is undetectable, that's very useful because they're taking dexamethasone or prednisolone or something else. But between that, a normal value of cortisol isn't so useful. In my experience, it is unusual for a patient with active Cushing's to have a low normal morning cortisol. It just is unusual. It's normally in the upper end of the normal range or above the normal range. But it is not impossible for a patient with Cushing's to have a sort of a serum cortisol in the lower end of the normal range. It's just not very common or not very likely. So it doesn't, in some circumstances, I think you can absolutely use it to make a diagnosis in some circumstances to say that they've got exogenous Cushing's, but otherwise a one-off basal cortisol is not gonna be useful in the morning. Of course, if you're assessing it at late night when the circadian rhythm should be quiescent and you've got very high levels, well, that's much more likely to be Cushing's. Great, thank you. Another question, is there any recommended period for a patient to stay away from alcohol before we repeat the test? Yeah, that's a good question. Because you could argue that if you come off alcohol and you've got withdrawal symptoms, you might really rev up your HP axis and you may still lack of suppression. So there are data. There are data from the early seventies that one week off alcohol was sufficient to convert a lack of suppression to a suppression on dexamethadone suppression testing. In this particular patient, I was trying to remove as much doubt as possible. So I wanted a longer period to try and do it, but a week or two is probably sufficient. Great, we have a plethora of questions. I'm afraid you will have a lot of work to do in the platform, but I will focus on as many as I can. So another question, is there a role for DDAVP either in the differential diagnosis of ACTH-dependent Cushing's, and probably we'll leave this for later, or in ruling out non-Cushing's hypercortisolism? Yeah, so there are lots of data using the peripheral desmopressin test to try and distinguish between non-tumorous hypercausalemia and Cushing's disease. And as I said in my talk, it's not a test that I use. It's a test that's used very well by colleagues of mine, such as Anton Tabaran and others in France. And yes, there are data, which you can look up at the various cut points about using it. It's just not a test that I'm familiar with or use. Should I be using it? Maybe, but I just haven't been to date. Of course, I've published papers using desmopressin, but in the context of differential diagnosis. The time when I don't think a peripheral desmopressin test is useful is in the context of ACTH-dependent Cushing's where you're trying to differentiate between ectopic and pituitary, because we recognize there are some ectopic tumors that express V2 receptors that can respond and therefore look like they're responding. I think that's different when you're using patronal sinus sampling, but I'll leave that to Roberto, because I think that certainly desmopressin has a role in that. And then another question on the clinical picture. Dark-skinned individuals and purple striae. Can you give us some light on that? Yeah, it's tough. It's much harder. I can't be clearer than that. You just have to really be very careful. And of course, if there are other features such as the truncal obesity, thinning of the skins and muscle wasting, it is harder, of course, if some of the features that you may want to use are less apparent in a given individual. I just, I haven't got any other clever things to say apart from that. Thank you. And then a further question, other than medications, are there any factors that can increase the dexamethasone metabolism, genetic factors, for example? So there will be some rapid metabolizers. There are data where the post-dexamethasone, serum dexamethasone value is measured. And there are various papers quoting different levels to be sufficient using different normative data to be sufficient to suppress people normally on a dexamethasone suppression test. So if you're taking a serum sample for cortisol, if it happens that there is a dexamethasone assay available to you, then of course, you can measure dexamethasone at the same time to see what the level of dexamethasone is. That gives you the reassurance they've taken the dexamethasone first and foremost. And if the dexamethasone value is very low, then perhaps you are dealing with someone who's a rapid metabolizer, and you may then have to choose a different test to use other than the dexamethasone suppression test. Great. Another question, how do you come up with a pre-test probability? So do you weight each comorbidity? Do you give it a particular percentage and then that's how the pre-test probability comes? I don't know if we need any mathematical modeling here, but it will be interesting to know your thoughts. So it's not a precise estimate. I'm afraid what you call clinical judgment and experience is about as precise as it can be. There are papers and there are several papers actually where people have designed clinical risk scores for the diagnosis of Cushing's based on various different parameters. And that makes it more or less likely. And of course you could apply those scores and that will, I think, enhance people's ability to try and then screen the right people. I imagine if that was done well enough, particularly if you had a very clever piece of AI, they could really maximize those things. But it's just down to the overall picture, listening to the length of the history, the features in the history that make it more or less like the Cushing's and the various diagnostic signs. I have to say that I see many people referred to me with, could this be Cushing's? Or they've seen other people with Cushing's. And there are quite a lot of people I see who I do no tests in at all, because I think that the chances of Cushing's is sufficiently low. And in those people, continuing to test them just runs the risk of false positives. And my approach in that circumstance, particularly if there aren't comorbidities which are desperately poorly controlled, was to review them again at an interval and see again. So can I ask precisely how I have it in my head? No, but it's putting together the features in the history together with those different amounts of, the more you've got of those different discriminatory signs, the more confident I'll be. And of course, I get it wrong. Everyone gets it wrong at times. So, you know. Sure. A further question on hair cortisone. What is the value of hair cortisone for cyclical Cushing's? Yeah, so it's a good question. I took a slide out. And so I had a slide showing hair cortisone. So hair cortisol is a really interesting concept. I find it truly amazing that you can abuse your hair with all sorts of hair products. And yet you can still measure cortisol. So the hair grows at approximately a centimetre a month in some people, not myself, obviously. And you integrate some cortisol into that hair follicles. You can measure the hair. So if you like, you can then take the hair from the nape of the neck. So you have 10 centimetres and someone's got long hair, divide it into one centimetre things. And you've got a picture over the last 10 months of what the cortisol values are. And of course, if you measure those and they're all up and down, that might be consistent with cyclical Cushing's. I think it's a fabulous research tool. I'm not yet convinced we're fully able to use that in clinical practice, but I think it's a fabulous tool. I think it may yet be very useful. It's not a tool that I've been using. I think it's a very elegant approach. Great. Thank you, John. We could go on and on for at least an hour. I'm sure we've got a lot of questions which haven't been answered, but we have to move on. So again, many, many thanks for the brilliant and so informative presentation. Just to remind our audience to send questions to the Community Connect after this. And now the floor goes to my colleague, Ismat Shafik to continue with Professor Salvatori. Thank you again, John. Many thanks. Thank you, Niki. And that was excellent, Dr. Newell-Praus. Thank you. So now that you have diagnosed the patient by a chemical hypoglycemia, the question comes where the source is. And that's the reason Dr. Salvatori is going to spend a good 40 minutes with us to tell us how to find it. Dr. Salvatori has graduated summa cum laude from Catholic University School of Medicine in Rome. Then he came to United States to do his internal medicine residency at Albert Einstein College of Medicine in New York, followed by his fellowship at John Hopkins University School of Medicine. He decided to stay at John Hopkins, rising to the present rank of Professor of Medicine and Neurosurgery. He has founded and co-directed the John Hopkins Pituitary Center, and his major focus is clinical and research activities on the diagnosis and treatment of pituitary and adrenal disease. He has authored more than 200 original papers and review articles, and has several chapters in endocrinology and neurosurgery textbook. Thank you, Dr. Salvatori. Thank you, Ismat. Thank you, Niki. And thanks to the Endocrine Society for organizing this interesting webinar. This is Harvey Cushing in the Hunterian Building at John Hopkins Hospital, probably around 1912. And we are proud to be standing on his shoulders, all of us, all of us in the endocrine. Do I have a control on this slide? Great. These are my disclosures. And I want to start with a case. This is a woman I saw a few years ago. She was a young woman, 34, who was noted to have Cushing gout features by her primary care doctor. When she complained of about 30 pounds weight gain, she also developed a metatarsal fracture without any recollection or trauma, and a DEXA scan has shown osteopenia. On a real system, she had weight gain, tiredness, irregular periods, and she also reports some easy bruising. And on her exam, her blood pressure was mildly elevated. She had abdominal obesity, a mild retrocervical fat accumulation, and some supracavicular fullness. She had no obvious striae, no prosthetic muscle weakness, and no obvious irritation. And I just want to remind you that Dr. Cushing described this syndrome in his paper published in the Bulletin of John Hopkins Hospital in 1932. He already had a very good understanding of what patients with Cushing syndrome have, and you can read it by yourself, but he describes the rapidly acquired peculiar dysplasia adiposity, a tendency to become round-shouldered, meaning osteoporosis, sexual dystrophy, hypertrichosis, striae, hypertension, erythremia, backaches. And remarkably, before he could measure any hormone, he wrote an excess of anterior pituitary hormones, secondary effect of the function of the cortex with established certainty, whereas nothing comparable to these occurs in the reverse direction. Remember, just remember that this was when there was no hormonal assay whatsoever, so this is quite remarkable. So we do have a laboratory evaluation, and this lady had obvious Cushing syndrome. She had a frankly elevated 24-hour urinary free cortisol, about three times normal. She had abnormal bedtime salivary cortisol. She had a random cortisol, which was in the upper range of normal with the CTH in the upper range of normal as well, and she failed the dexamethasone suppression test with appropriate dex label. So I think nobody would argue that this lady has a CTH-dependent Cushing. Her pituitary MRI was negative, and abdomen and chest CT were also negative. So my question to you is, what would you do next? Would you do a high-dose dexamethasone suppression test? Would you do an octorotized scan? A desmopressing test? Would you just refer it to neurosurgery or order an in-vitro platelet-rich cell sampling? Great. So, uh, a lot of people do a high dose, the suppression test, um, probably just almost as many, a little more, we'll do an IPSS. Um, very few would do a ultrasound. Nobody would do a decimal pricing test. That's interesting. And, uh, only very few people will refer the patient to neurosurgery on that. Um, probably it's a very early, uh, and I agree, you know, what they say, if you go to the barber, he's going to cut your hair and I'm not sure we are there yet. Alrighty. So, um, the focus of my talk would be, uh, exactly what to do in a situations like this, when there is no obvious CTH source. And, uh, can I get Claudia able to, um, give me the next slide. Somehow I lost control. Okay. So there are several, uh, approaches to this. Just remember that a significant percentage of patients have normal MRI, the literature say about 35%. It may be a little lower now with modern, uh, 3T, uh, MRIs. Also remember what, uh, Dr. Newell-Price told you that up to 10% of people are walking around with a small pituitary adenomas that are just innocent bystander doing nothing. And of course, uh, a similar percentage of patients with a topical CTH producing tumor will have abnormal pituitary MRIs. So the MRI by itself is not a good test to diagnose Cushing disease. And these are sort of reiterates the big no, um, uh, that, uh, Dr. Newell-Price mentioned to you, not, not to go to imaging testing unless you have a proven, um, uh, diagnosis. So you can do a variety of tests to differentiate, uh, to determine the origin of a CTH. Uh, you can do it. Uh, there is a literature on CRH tests, there's more pressing tests, high dose, XML suppression tests, or others. The caveat is that no test is 100% sensitive or specific, or you can go directly to IPSS. And what is IPSS? Um, uh, the radiologist insert a catheter in a patient's groins, threads the catheter up to the, um, uh, venous system, uh, up to the, um, sinus, which drains, uh, pituitary blood. And, uh, a CTH is measured, uh, both at baseline and after stimulation, which can be done with a CRH or Desmo pressing. And we're going to be talking about the specificity of the sensitivity of these tests. And this test is, um, not available, of course, to everybody, um, is mostly done in the larger centers. Uh, I actually would argue that should be done in larger centers where, um, radiologists, interventional radiologists can gain, uh, significant expertise in doing them. And, uh, as a occasional, but potentially, uh, dangerous side effects, uh, pontine infarct has been described, DVTs I've seen in a couple of patients, pulmonary embolism and, uh, nerve palsy, which are usually transient. So the question is, do I really need IPSS in a patient like this? And do I need to refer the patient to a major center where the IPSS is available, or I can get away without it? There is a paper that was published, uh, very recently, um, uh, by, uh, French, two French groups led by Dr. Tabarin, um, that sort of questions whether IPSS are necessary. These are very interesting paper that I encourage everybody to read and digest very carefully. I'm going to try to outline just the main messages, but really there is so much meat and so much data in this paper that I really encourage, uh, uh, you to take, um, an hour or two to read it. They looked at 167 patients with Cushing disease and 27 patients with atopic STD syndrome, which we were investigating to French center in the past 15 and during a period of 15 years, all these patients underwent the peripheral CRH and desmopressing test. And they did IPSS in 92 of these 194 patients. And they also, this patient underwent, uh, CT scanning and this CT scan identified, uh, the source of ectopic CTH in 67% of these patients with absolutely zero false positives. And as you can see here, the idea of the CRH and the desmopressing tests are very similar. Um, essentially, uh, there is no perfect separation between, uh, uh, Cushing disease and ectopic CTH syndrome, but in, in general patients with ectopic CT syndrome tend to respond less to CRH and to desmopressing than patients with, um, Cushing disease. But as you can see here, there is a overlap between the two tests. But when they actually, um, looked at the best cutoff for these tests, they, um, determined the best cutoff in their hands was an increase in cortisol or 17% from baseline, uh, which, you know, it's really not very marked, uh, with a CRH and a 10% increase with desmopressing and a 37% increase in a CTH with, um, uh, CRH and a 33% with desmopressing. And what they found is that when both cortisol and the CTH were above this threshold, no matter which test you used, there was a sensitivity and specificity of these tests that was satisfactory, but certainly not perfect between 81 and 85%. So this is something that gives you some information, but of course doesn't tell you the whole story. However, when they put everything together, they found that, um, the using this test, using MRI, using a CT scan of the chest, they were able to identify the source of a CTH in more than 50% of these patients without the need of, um, IPSS. And interestingly enough, this was also in patients with negative pituitary MRI. So the, um, non-invasive strategy that they call, uh, the positive predictive value for Cushing's disease in a hundred percent in patient with positive response to both peripheral tests and negative, and negative pituitary MRI, and the negative predictive value, a hundred percent in, in patients with a positive CT scan. So using this strategy, they concluded that IPSS could have been avoided in about 50% of patients who actually underwent the test. So their conclusion was that in collaboration with expert radiology, and this, of course, an important aspect of this work that they have expert radiologists, the non-invasive investigational strategy that they proposed was able to substantially reduce by 50% the need for IPSS and still maintaining a very high positive predictive value for Cushing's disease. Now, if you decide to do IPSS, uh, as you know, this test was developed at NIH by a group led by the late, uh, Dr. Doppler, who was a legend in neuroradiology, uh, and a wonderful man on top of it. And, um, the test, uh, in the, in the initial, uh, evaluation was really a very, very, uh, accurate, but you know, in real life may not be as accurate as you, uh, can imagine. These are retrospective analysis of IPSS performed over a larger, longer period of time at the Massachusetts General Hospital. And as you can see here, uh, and I'll just outline the bottom line that out of, uh, about a hundred IPSS, which were successful, meaning that the radiologist was confident that they had incandelated the, uh, inferior pedresal sinus, there were, uh, two full positives. So patients with ectopic CTH who had a positive test, and there were actually nine false negative. And these were patients where the IPSS seemed to show no gradient between the central and the periphery. Uh, extensive search was done for ectopic source of CTH. They were not found and family patient underwent pediatric surgery and then CTH signaling adenoma was discovered. So, um, when they looked at their ROC curves and they calculated the specificity and sensitivity, you can see that, um, the, um, uh, accuracy of the test is not perfect and is better after CRH than pre-CRH. And this is important because as you know, in the U.S. and I guess in other countries, uh, CRH is no longer available currently. So, um, you may need to do this test without CRH. You can either just do baseline or use desmopressin. We'll talk about that. In their hands, the, uh, sensitivity specificity was best using a central to peripheral ratio of 3.3. Um, and the negative predictive value was best when there was a ratio below 1.8, and this is all these patients had CRH stimulation. Interesting enough, there was really not very, um, uh, much different in terms of MRI findings in patients who were ectopic versus, um, um, uh, pediatric disease. And as you can see here in patients with a positive MRI, um, only 48% of cushy disease is a positive MRI and 22% of patients with ectopic, uh, had a positive MRI. So, um, again, you know, showing that MRI positivity may be present in ectopic CTH. They also noted similar to the French group, and this was done many years prior, that in a CTH increase of above three, above 35%, this is during the IPSS, but measuring peripheral CTH. So this is sort of a expensive way to do, uh, uh, a CRH stimulation test. Uh, they found that there was an increase in the 35% in 62 patients with pediatric disease and in no patients with ectopic disease. And this was also in eight patients in which the IPSS was negative, meaning there was no gradient between the center of the periphery. But then when they looked at the increase of CTH in the periphery, they found that it was an increase above 35%. And this patient, um, ended up having cushion disease. So I think that, uh, uh, um, interpreting the CRH stimulation test, if CRH is available to you, a cutoff of a 35% increase is what probably you should memorize. Um, in terms of performance of the IPSS, there are several papers that are published. Uh, this is a large paper. I think the one with the largest number of patients from a single institution that looked at 331 IPSS done in 327 patients, some patients had it twice, and some patients, uh, had no stimulation. Most of them had stimulation. Some of them with CRH, some of them with a desmopressing. They didn't specify what percentage was CRH, what percentage was desmopressing. And again, the unstimulated sensitivity was 89%. The stimulated sensitivity was 96%. I think the most important piece of information in this paper is this bar graph on the right. It tells you that the shows the increase in CTH after a stimulation. And as you can see the peak stimulation happens at three minutes. So I think it's important that you discuss with your neuroradiologist that they have to be very timely and they have to get a three minute, um, um, uh, blood draw to maximize sensitivity. There is a, some literature that you may have noted about, um, making sure that you are in the right place. And one way to make sure that you are in the right place during the procedure is to measure a prolactin level. And in this paper from the UK published a few years ago, what they did, they, uh, measure both a CTH and prolactin. And then they measure, uh, the sensitivity of the test, uh, on whether they just looked at the absolute value of a CTH, or if they corrected the CTH value based on the prolactin levels. And as you can see, they saw that, um, 66% out of 83 had a positive IPSS when they use uncorrected CTH level, but there were six additional patients for 10% more patients about, uh, whose test became positive only when there was a correction according to prolactin level. So they recommended that a prolactin adjusted, uh, IPSS to peripheral ratio in a CTH below 0.8 suggested ectopic CTH syndrome, a level above 1.3, uh, is diagnostic of, um, uh, pituitary cushion and level below between 0.8 and 1.3 need further investigation. Another, um, important aspect of the IPSS is whether it can help your neurosurgeon. And of course, if there are neurosurgeon, uh, here in the, in the audience help you determining which side the pituitary adenoma is. And, uh, um, there are several papers that looked at this. This is one of the first papers was published in a JCNM, uh, almost 20 years ago where they retrospectively looked, um, 90, 90 IPSS studies done between 1988 and 2011. And all these had CRH and they consider positive, uh, pre-CRH, uh, gradient of 2.1 and a post-CRH gradient of 3.1 as most, uh, uh, literature recommends. And then they consider, uh, lateralization if there was an inter-sinus gradient between one side and the other before CRH of above 1.4. And, uh, this is a picture from one of their patient that shows that, uh, not everybody's anatomy is the same. And, uh, this is the inferior pituitary sinus, uh, on, uh, on the left. And this is on the right, an example of plexiform, uh, right IPSS. And, uh, they, uh, in their series, they found that approximately 30% of subject have an asymmetric venous damage, which of course is going to affect your ability to determine which side is the highest CTH level. So when they looked at, um, the overall lateralization, uh, the ability of the IPSS to predict which side the adenoma was, they found that in all camera, the, um, uh, accuracy was only 57%. And just if you flip your coin, it would be 50%. So it's not much better than flipping a coin. And it was better away, but in patients who had the asymmetric venogram, um, and catheters where the lateralization success was approximately, uh, um, 71% interesting enough in a subset of patients after CRH was injected, the lateralization inverted. And, uh, uh, it seemed from this series that the post, uh, stimulant stimulation, um, lateralization was less accurate than baseline, possibly because, uh, the stimulus, um, acts on, uh, suppressed, uh, cortical cells that otherwise would not be active. And this is another paper from, uh, San Francisco, looking at very similar, um, uh, data in a large number of IPSS. And essentially, if you just look at the last right column, you can see that, uh, pre-CRH and post-CRH, the accuracy is not outstanding, um, in between 66 and 80%. And of course, it's a little higher when there is symmetrical drainage in a yellow versus, uh, all cameras, which include of course, people without symmetrical drainage. And, uh, again, this is from the, uh, the same paper. I'll skip it. So, um, actually, no, I'm sorry. I, uh, this is an important actually slide because, um, what they did in some of these cases, they, and they just went ahead and do the hemiapophysectomy. They removed just half of the pituitary gland being guided by, uh, the result of the inferior petrocell sinus sampling. And as you can see here, no matter whether there was a symmetrical damage or all cameras, the cure rate was about 50%. So, um, uh, you know, the, the lateralization from the iPSS was not very helpful in determining, um, which half of the pituitary to remove. A more recent paper, uh, that was published just a couple of years ago, uh, showed a surgical strategy in, uh, 22 patients with Cushing in whom, uh, at surgical exploration, the surgeon could not find the adenoma. And what they did in these patients, they did, uh, they started doing a cut on one side, a vertical cut on one side of the pituitary, uh, trying to explore the side and the decision of which side was based on either the MRI funding or the iPSS data. If they didn't find anything, they went on the other side, they did another cut exploring the right side of the pituitary. If they didn't find an adenoma, they do the horizontal cut, um, looking for the adenoma. And lo and behold, in 22 of these patients, after these three cuts, the three slices, they didn't find an adenoma. And what they did, they just removed the two lateral part and the inferior part. And they essentially removed two thirds of the gland. And interestingly enough, they had an overall remission rate of 80%. And it was a hundred percent when, uh, the adenoma was found in pathology, but a subset of patients had their remission, even if the adenoma was not found in pathology, probably the adenoma ended up in a structural cavity. And with this strategy, only two patients, um, which will be only 9% of patient had new hormone deficient. One patient developed new hypogonadism and one patient developed a new central hypothyroidism. So this is a strategy that your neurosurgeon may want to, may want to apply when the adenoma is not visible at a time of surgery. And even more recent paper tried to be a little more, um, uh, conservative. And what they did in this paper actually did the one third hypothyroidism based on IPSS or MRI data. And, uh, with one third hypothyroidism, the remission rate was much lower. It was only 31% and still, uh, about 15% in new hormone deficits. So, um, I think these are papers that you should discuss with your, with your neurosurgeon, uh, because it's always good to have a good strategy before going to surgery, a strategy that neurosurgeon discuss with endocrinologist, you know, what am I going to do if I don't find the adenoma, how aggressive you want me to be? I'll never forget one time where, uh, I was paged on, uh, the, um, the Evo Thanksgiving holiday by the neurosurgeon who told me that he explored this patient, uh, of mine that I refer with a diagnosis of pituitary cushion, and he didn't find anything. And he did a total hypophysectomy. And I almost had a heart attack and I spent all weekend, uh, praying that somebody will be found. And in fact, a three millimeter CTH adenoma was found, but of course the patient ended up with a panhypopituitarism. So, uh, in terms of lateralization, there is a paper from, uh, the Cleveland clinic led by Dr. Ryan, who now joined us at Hopkins, um, trying to determine whether a prolactin, uh, adjustment helps lateralization. And in fact, what they found in these relatively small study of 28 patients when they went IPSS with CRH stimulation, looking at the, uh, left to right gradient, they found that if there is an adjusted gradient of about 1.4, the predictor, the ability of predicting it was pretty good. And if there was a ratio of a CTH adjusted between the two sides, about 1.4, just by prolactin, nobody had an adenoma on the opposite side. So, uh, again, um, something to think about when you, uh, look at the result of the IPSS study. And since Dr. Ryan joined our group, we have been actually measuring a prolactin level before IPSS, uh, during IPSS to adjust our CTH levels. Now, as I mentioned, there is no CRH available in the United States. We don't know if, and when this will be available. So, uh, how about this more pricing? How does this more pricing work? Well, normally, um, uh, normal cortical cells respond to vasopressing, but only, uh, the expression of the V2 receptor would give response to this pressing a normal cortical cells in theory are not supposed to respond, but these are supposed to be expressed by subset of patients, a subset of a cortical tropadenoma. So that's the idea of using this more pricing tests. And, uh, I'll skip this one. Um, there are a couple of papers that looked at, um, the, uh, uh, performance of the decimal pricing test and they are a little old paper when there was another period, uh, where, uh, CRH was not available. Um, in this particular paper from France, they looked at 43 patient with a CTH dependent cushing. Um, and, uh, they use a gradient central to peripheral 2.1 and using post decimal pricing and using this gradient, uh, the test in their hands was a 97% sensitive, 100% specific. And in a similar paper from Brazil published more or less in the same period, looking at 56 patient in this particular paper, they use the post, um, there's more pricing gradient of about 3.1. And that's what we used. Uh, they had a sensitivity for cushion syndrome of 92% and a specificity of a hundred percent. So, um, I think there's more pricing is the way to go these days where a CRH is not available. I also was under the impression that there's more pricing was so much cheaper than a CRH. Uh, in fact, my pharmacist told me recently that the vial of CRH a hundred microgram is, uh, uh, between seven and $800 and a 50, uh, microgram vials. So there's more pricing is a $400. And because of course you do only one test and the rest gets thrown away, probably the difference in expense is not as dramatic as I initially suspected. Um, can you do, uh, internal jugular sampling? If you are in a center where the radiologist is not so skilled and, uh, um, uh, doesn't want to go all the way to the inferior pitotal sinus, can you get blood from the, um, jugular vein? And we looked at that. We, uh, our radiologist for a while was drawing blood both from the cavernous sinus and from the internal jugular and the sensitivity of the test is much lower. Um, so, um, went down to 68% from 93%. This is done in the same patients. So certainly if it's positive, it helps you, but if it's negative, it doesn't help you. So it's not something that would recommend routinely. So now I'm going to ask you the same question. We have this patient who has obvious ACTH dependent coaching. Um, she has a negative MRI, negative chest CT. And, uh, what would you do next? Great, so many less of you elected high dose suppression test and that's probably right because I think this test doesn't really help in differentiating match between the eutopic and ectopic CDH, little more people start thinking about it as more pressing test and we are actually starting to do this more pressing test nowadays, and somebody or a little more people actually refer to neurosurgery and about two thirds of you opted for IPSS and in fact IPSS is what I had done in this patient and just to give you a real life picture, the IPSS in our hands was flat, there was no gradient and I was surprised because the pre-test probability as Dr. Newell-Price told us in a young woman with a CDH dependent cushion to have a pediatric cushion is very high and so I refer her to another center where she had an IPSS and there was an obvious gradient and so I went to speak with my neuroradiologist and the neuroradiologist sheeply admitted that he was actually flying to Japan for a meeting that day and one of his junior faculty just hired had done it and then he reviewed with me the images and he realized that they were not where they're supposed to be. So I think this really shows how important it is the skill of the neuroradiologist and how important it is that you talk with your neuroradiologist and ask, you know, well, do you think this test was successful, do you think you're successful in cannulating inferior pitot-cell sinus, was there any problem, was the catheter dislodged? So I'm going to close this and go to the next slide. Claudia, I'm having a hard time in going to the next slide, I don't know why. Seems to happen before, after the question. Okay, so in conclusion, there is no perfect way to work up a CDH dependent cushion, IPSS may not be needed even when the P2D MRI is negative, peripheral post-desmopressant value may help if there is an increase of about 35% or also CRH value, post-CRH, IPSS is accurate but it's not 100% sensitive and it probably is more specific than sensitive. Some data on low sensitivity of the test may be missing the past because people, patients were classified as probable ectopic, may have had actually cushion disease. The source of, the use of desmopressant is recommended in absence of CRH during IPSS. Lateralization data is only approximately 70% reliable, less when the venous drainage is not symmetric, it may be improved by prolactin adjustment, hemi-epiphysectomy based on lateralization should not be done, but two-third epiphysectomy should be considered, can be considered. And the inferior jugular vein sampling is less sensitive than IPSS. And I like to conclude with a couple of quotes, one from Harvey Cushing and a letter to his colleague Dr. Christian on November 1911, he wrote, of all the subjects, that of endocrinology particularly lends itself to temptation of impressionistic speculations. And actually I should tell you that Dr. Cushing despised endocrinologist, he called us endocrinologist and he had some good reasons because, you know, it was all about humors and there was no measurement and there was a lot of hand-waving, but no real science at that time. And a quote from Dr. James Finding from Wisconsin, which I offer a quote to my fellows, if you never missed the diagnosis of Cushing syndrome and you've never been fooled attempting to establish its cause, you should refer your patients with suspected hypercortis limb to somebody who has. And I thank you for your attention. Thank you, Dr. Salvatore. That was great. I hope we answered our audience questions about the challenges with IPSS and the unavailability of CRH. There are a number of questions and they are on the interpretation of IPSS. The first question is, what is the diagnostic value of IPSS in cyclical Cushing's and would you do IPSS in a patient where you have clinically high suspicious, but their diagnostic tests are negative? Yeah. So the first question is very interesting and there are case reports of patients with secret Cushing with negative IPSS because they were tested during an off cycle. So some centers measure 24 hour urine free cortisol the day before the IPSS. We can do this because the turnaround time for UFC is by no means enough. So what we do, we measure a serum cortisol the day of the test and we proceed with IPSS if the cortisol is above 10 microgram per deciliter. If it is below, then we postpone the test. And the reason that we do this is because our laboratory runs cortisol level around the clock and we usually are able to get it within a couple of hours from the drawing. So that's for the cyclical Cushing. And the other question, I'm sorry, it was the second was, if you have high clinical suspicious that patient has Cushing's, but the biochemical testings are negative, do you do IPSS in them? No. I mean, IPSS is not a test to diagnose Cushing. And I've seen patients refer to us who already had IPSS, who ended up to have a non-tumorous hypercortisolism. And think about it. If you have a non-tumor hypercortisolism, your SEDH will come from the pituitary. So you are not going to be able to differentiate between Cushing and non-tumorous hypercortisolism. IPSS only differentiates between Cushing disease and ectopic CTH syndrome. That's only use of IPSS. I agree. And the second question is also an IPSS interpretation where they say that the baseline central to peripheral ratio is less than two, but the ratio is more than three after CRH is given. What do you think is the probability of having such IPSS result to have Cushing's? I think it's very high. I think that you don't have to meet both criteria. You can only meet one of them. So that's why we do the stimulation test, because we want to look both at the pre and a post stimulation. And if you have no gradient or a very small gradient pre, but you have above three after stimulation, I think that's the diagnostic of Cushing disease. And the third one is that, is there any newer data with endoscopic transspinoidal resection on improvement with blind resection? Because the blind resection data is coming from older studies. And do we have newer data with endoscopic TSS that those surgeries are more successful? Yes. I'm not aware of any. I mean, that paper that I showed you with a two third is, it was published in 2018, but it looks experienced in the previous 20 years. I think that every surgeon has got their own preference, both in terms of technique and in terms of approach. Some neurosurgeons rely heavily on their frozen section, others rely more heavily on their visual appearance of seeing an adenoma. Of course, not being a neurosurgeon, I can't comment on which one is better. I think this is an answer that should be given by a pediatric surgeon rather than an endocrinologist. And one of the questions is that, what do you think about methionine PET CT? That's a great question. And there is some literature actually from the UK about this, Dr. Mark Grinnell has published on this and he has some fascinating initial data on this. You need an accelerator close by to generate the tracer. And so I suspect that it's not going to be a widely used technique, but he has some fascinating pictures of patients where the MRI did not show the adenoma and the nuclear medicine combined with MRI found the target. And I am not competent enough in this to comment, but it's fascinating research that is going on there. And one of the questions is that, do you use ketoconazole or any other adenolytic agents in patients who have Cushing's, but their MRI is negative, to see if you can visualize the tumor on repeat MRI? Yes, this is a strategy that I've used in several patients. You can use an adenolytic drug like ketoconazole, you can use Mifepristine. So buying yourself time and possibly sort of creating your own little Nelson syndrome, control Nelson syndrome, allowing the adenoma to grow because it will grow or to grow because you have dramatically reduced the patient's cortisol level and rescan the patient maybe a year or two later. I've done that. And I remember one case where it was obviously successful. Of course, you know, you can also rely on different MRI techniques. I'm another radiologist, of course, but, you know, people are starting to publish on the T7 Tesla. My impression with this very high definition MRI is that they sometimes end up murking in the water more than clearing because they show probably too much. But certainly that approach of reducing cortisol burden and waiting for the adenoma to reveal itself is a very reasonable approach. Okay. One last one. Do you think that high dose dexamethasone suppression test has any value? Well, there was some old data from the NIH that put a lot of weight on the degree of suppression. And I think in that literature, if somebody suppressed more than 70%, there was no ectopic. Of course, there is no perfect test. There are some tests, some pituitary glands that think they're ectopic, some pituitary tumors think they're ectopic and some carcinoid that really think they're pituitary glands. So the high dose dexamethasone suppression test is not perfect. I don't use it often. I find myself sometimes in a young woman in whom the predisposibility is very high with a three millimeter adenoma, I want to convince myself further, I will do a high dose test. If it is positive, I'll pat myself on the back and I'll say, yes, it's the pituitary cushioning. If it is negative, I'll say, well, this is still probably pituitary cushioning just with a negative high dose. So I don't put too much weight on the high dose and I would love to hear what John has to say about that. I don't use the high dose dexamethasone. Got it. So this was great. I think we learned a lot from today's conference. I want to thank both speakers, Dr. Newell Price and Dr. Salvatore for the amazing talk. I also would like to thank Dr. Karavattaki for being my partner today and hosting this interesting seminar. Finally, I would like to thank our Endocrine Society staff, Claudia, who spent a lot of time with us organizing it, as well as Courtney and Rodnika, to arrange this webinar. And of course, we cannot forget about our community members because they are the ones who initiated this topic and wanted to learn more about it. Have a great day and a good weekend. Thank you.

Cushing's syndrome

diagnosis challenges

clinical suspicion

history and examination

discriminatory signs

urinary free cortisol

dexamethasone suppression test

late-night salivary cortisol

case studies

desmopressin test

ACTH-dependent Cushing's syndrome

IPSS

collaboration between endocrinologists and neurosurgeons