Good afternoon and welcome to this session, An Approach to a Patient with Adrenal Insufficiency. My name is Lucinda Gruber and I will be your moderator. You'll be hearing a lecture from Dr. Richard Aukus on this clinically challenging subject. Dr. Aukus is an accomplished professor in translational medicine at the University of Michigan. He and his group have been involved in numerous projects ranging from basic chemical principles and steroid biosynthetic enzymes and steroid mass spectrometry, all the way to clinical trials and translational investigations and disorders of the pituitary, adrenal, ovary, and testes. Dr. Aukus also has a clinical interest in this area with a focus on caring for adult patients with genetic disorders of steroid biosynthesis and activity. I'm delighted he's been able to join us today to share his approach to a patient with adrenal insufficiency. Please join me in a warm and virtual welcome to Dr. Richard Aukus. Hello, everyone. I'm Rich Aukus from the University of Michigan, where as I record this virtual presentation, it is snowing in late April. And I'm gonna talk today about the approach to the patient with adrenal insufficiency. These are my disclosures. So the first point I wanna emphasize is that adrenal insufficiency is a rare disease. Now, unfortunately, adrenal insufficiency has become one of those pop culture diagnoses from non-endocrinologists. And so you will have patients referred to you with a putative diagnosis. And I want you to realize a couple of things. First of all, prevalence is about one per 10,000 individuals so far less than 0.1%. The incidence is about five per million per year. Now, and that's primary adrenal insufficiency. That's true Addison's disease. Now, or related diseases of the adrenal cortex. Secondary adrenal insufficiency is quite a bit more common and far and away that is due to exogenous glucocorticoids. So if you take that away, then we only have a few other things such as pituitary and hypothalamic diseases. And one important pearl to remember is that when it comes to loss of pituitary adrenal function from mass effects like pituitary tumors, that the adrenal axis is usually the last axis to lose. So usually the order of dropout is first growth hormone, then gonadotropins, then thyroid hormone, and finally adrenal function. With radiation, the thyroid function is last, but still it's pretty uncommon to have isolated adrenal insufficiency from hypothalamic pituitary disease unless you're taking exogenous glucocorticoids. Now, the one exception to that is lymphocytic hypophysitis. And in modern times, we also have immune checkpoint inhibitor induced hypophysitis as well. And because the ACTH producing cells are in the middle of the pituitary right under the stalk, when the lymphocytes invade through the stalk, they tend to hit the corticotropes first. Whereas when you have tumors around the periphery, they tend to spare the corticotropes until last. So thinking about primary adrenal insufficiency as a rare disease, that has some implications in terms of diagnostic testing. So the prevalence of primary adrenal insufficiency is one in 10,000, and you do a cosentropin stimulation test. Now, cosentropin stimulation test is a very good test. It has roughly 99% sensitivity and 99% specificity. It's hard to do better than that. So let's say you do 10,000 tests randomly on people in the population. You will get 100 positive tests or 100 failures. Now of those, 99% will be false positives, and only one will be a true positive. So your positive predictive value of a failed ACTH stimulation test without an index of suspicion, a high pretest probability is only 1%. So unless your pretest probability is high, you shouldn't be doing cosentropin stimulation tests. So who is at risk and what increases your index of suspicion? Well, other autoimmune disorders such as hypothyroidism and type 1 diabetes, people who are at risk for granulomatous infections, HIV patients who are not getting heart therapy reliably, people with immunodeficiencies, people getting anticoagulation or the antiphospholipid syndrome, and people who have specific signs and symptoms. So what are some specific signs and symptoms? So weakness and fatigue are universal, but anorexia and weight loss are seen in roughly 100% of people. So if people aren't losing weight, then it doesn't make a lot of sense to test them for primary adrenal insufficiency. So you notice the problems with abdominal pain, vomiting, salt craving, postural dizziness, and then less commonly muscle and joint pain. So again, people have anorexia, inability to maintain their body weight. So again, weight loss, 100%, hyperpigmentation. Most of the patients who have primary adrenal insufficiency will experience hyperpigmentation as well as hypotension. And then vitiligo is a quite specific finding on people with multiple autoimmune diseases. And so if you see somebody with weight loss and vitiligo, for example, especially if they have a history of hypothyroidism, then yeah, probably have primary adrenal insufficiency. So this is an example of the hyperpigmentation. You see hyperpigmentation in scars. Often the areola get hyperpigmented. Hyperpigmentation in the gums, in the skin folds. So in the knuckles and in the knees is common places. And then this is a patient who was an inpatient when I was attending several years ago who presented with primary adrenal insufficiency. So you can see how dark she is. You can see the hyperpigmentation in the palmar creases on the gums. And so somebody like this, it's a pretty easy diagnosis. You get a basal cortisol and ACTH. The cortisol is low, the ACTH is high. You give her 10 milligrams of hydrocortisone and she has a miraculous response. I would call it a Lazarian response. They get out of bed and they're eating and energetic for the first time in months. This is primary adrenal insufficiency. Now, a couple of other examples. Again, hyperpigmentation in bruises. Sometimes hyperpigmentation in the nail bed and hyperpigmentation in the gums in addition to the tongue, as I showed you before. So laboratory abnormalities. Hyponatremia is quite common. I wanna emphasize that in adults, hyponatremia is due to cortisol deficiency because when you're cortisol deficient, you can't suppress vasopressin. So it is a functional type of inability to suppress ADH, but it's due to the cortisol deficiency. So this occurs in both, oops, sorry, both primary and secondary adrenal insufficiency. Whereas the hyperkalemia is due to aldosterone deficiency because to excrete excess potassium, you need aldosterone. So to excrete excess free water, you need cortisol. To excrete excess potassium, you need aldosterone. And then these other things less commonly, eosinophilia is a tip off to primary adrenal insufficiency because of the autoimmune nature. And so the sort of common presentation for a new onset diagnosis in somebody who has an adrenal crisis as their initial event is that they have weeks to months of fatigue that evolves into nausea. Then they have an illness and they start vomiting during this illness, usually a gastrointestinal illness like a stomach virus. And then they have vascular collapse, which is the adrenal crisis. So it is the combination of the cortisol deficiency plus a stressor event that leads to volume depletion. Then there is inability to maintain the intravascular volume because you need cortisol to respond to catecholamines. And without that cortisol presence, you don't get the pressor effect of catecholamines. Now, so the precipitants of an adrenal crisis, mainly infection, even in 2021, infection and GI infection tend to be the leading causes. Other inter-current illnesses such as a myocardial infarction, a pulmonary embolus, or other severe illness, surgery, but it usually has to be a major body cavity that is operated on. So abdominal surgery, thoracotomy, something like that, not like a dermatologic surgery or a surgery on an ankle or carpal tunnel release or something like that. Severe trauma, a car accident, somebody falling off a bridge, environmental volume depletion. So somebody who's out exercising or playing golf in very hot climate, like in the Southwest of the United States where it's 100 degrees and the humidity is low and people don't notice that they're losing a lot of fluids. And then drugs, for example, rifampin, which induces cytochrome P453A4, that is one of the things that metabolizes cortisol, so it increases the clearance of cortisol, or etomidate, which is an 11-hydroxylase inhibitor in addition to being an anesthetic, and it will impair cortisol synthesis. So just in comparing primary and secondary adrenal insufficiency and primary adrenal insufficiency, everything the adrenal cortex makes, including cortisol, aldosterone, and DHEA sulfate, is reduced, ACTH is high, and renin is elevated because of volume depletion. Whereas in secondary adrenal insufficiency, ACTH drive is low, renin-angiotensin-aldosterone axis is intact, and so cortisol is low and DHEA sulfate is low because it is an ACTH-driven steroid. Now, we still don't know what DHEA sulfate does in human physiology, but we make a lot of it. It's useful as a biomarker to tell you how much ACTH you're making on a long-term basis, so it's useful both evaluating autonomous cortisol production from an adrenal adenoma as well as the potential for having secondary adrenal insufficiency. So nowadays, we actually can make most of our diagnoses based on simply basal hormone testing if we get everything together at the right time. So what I do is I try to get by 8 a.m. in the morning a cortisol, a DHEA sulfate, and an ACTH. So cortisol less than 3 is low. Cortisol greater than 14 is normal. And values in between, we need to adjudicate in some way. So how do we adjudicate? Well, one thing is to use DHEA sulfate. If it's over about 60 micrograms per deciliter, assuming the patient is not taking DHEA supplements, then that rules out adrenal insufficiency. I'll show you the data on that. So that's one way to adjudicate these intermediate values. Now, the ACTH is then going to be useful if we have diagnosed adrenal insufficiency to tell whether it's primary or secondary. If it's low, then the adrenal insufficiency is primary. And low, I mean less than 10 picograms per mil or so. And if it's high, meaning over 100 or so, then it's a primary adrenal insufficiency. Renin and aldosterone are useful in confirming primary adrenal insufficiency, that the renin should be high and the aldosterone low. In secondary adrenal insufficiency, both are usually normal or they can be high. They even can be low if someone is on a high salt diet. Now, so occasionally we can't tell, and then we do a cosentropin stimulation test where we administer 250 micrograms of cosentropin. And a cortisol value greater than 14 micrograms per deciliter is normal. Now, you might say, where did I get 14 micrograms per deciliter? I thought it was 20 or I thought it was 18. Well, actually new cortisol assays that are being used commercially for our monoclonal antibody assays or mass spec assays. And in fact, when they rolled these out, they didn't tell anybody, but they did a comparison between the old assay, which was a polyclonal antibody in the new assay and the new assay read about 20, 25, 30% lower. So they looked at a bunch of people who had cosentropin stimulation tests, and they had a bunch of people with abnormal responses and a bunch of people with normal responses. And with using 500 in the old assay to get optimal performance characteristics with the new test, they had to use a value of about 374 nanomoles per liter. That's a little too hard to remember. So I just remember 400 nanomoles per liter, which is about 14 micrograms per deciliter. And this has been confirmed. There's actually now in this month's journal of the Endocrine Society, there is a paper from the Wisconsin group, from the Milwaukee group, Jay Finlay and his colleagues, which confirmed that 14 or 15 micrograms per deciliter in multiple current assays, including mass spec assays is what we should be using as cutoff for normal. Now, what about free cortisol? Is there any advantage to free cortisol? Well, usually no, but occasionally yes. And this was a very nice study done from Mayo Clinic where they compared total cortisol versus free cortisol. Now notice this curvilinear relationship. This curve here, when you get to about 12 or so micrograms per deciliter of total cortisol is when you saturate cortex of steroid binding globulin. And then after that, there's a linear correlation, okay? Whereas before that, you get a curvilinear relationship. So 12 or so micrograms per deciliter is where you saturate CBG. And they had a couple of patients who had discordant results between the free cortisol and total cortisol, mainly it was in women taking birth control pills who have elevated CBG. So that's a time when you might wanna get a free cortisol in addition to a cortisol. Now, remember that saliva is an ultrafiltrate of plasma. And I think we should all be comfortable now using late night salivary cortisol to diagnose cortisol excess in Cushing syndrome. But some recent studies have looked at salivary cortisone. Now, why would you wanna look at salivary cortisone instead of salivary cortisol? Well, remember that in serum, in plasma, the F to E ratio, the cortisol to cortisone ratio is about four to one. But in saliva, the salivary glands are a tissue that transports ions. And in those tissues, the enzyme 11-beta-hydroxysteroid dehydrogenase type two is present just as it is in the kidney. And that enzyme converts cortisol to cortisone so that it doesn't bind to the mineralocorticoid receptor and allows aldosterone to regulate ion transport. So in saliva, there is much more cortisone than cortisol. There's about six times as much cortisone than cortisol. So actually, the dominant glucocorticoid that one can measure in saliva is actually cortisone. So studies have been done now looking at normal volunteers and at patients receiving either IV or oral hydrocortisone. And the conclusion is that salivary cortisone correlates best with serum cortisol. In fact, even better than salivary cortisol. So let me show you some of the data on that. So this is a correlation matrix. Of course, the best correlate of serum cortisol is serum cortisol. So that correlation is one and so on for serum cortisone and free cortisol index, free cortisol, salivary cortisol, salivary cortisone. If you look about what correlates best in terms of salivary steroids, salivary cortisol is not quite as good as salivary cortisone. Now, people taking oral hydrocortisone, because now you're sticking some cortisol in your mouth, there's a very poor correlation with salivary cortisol because of that little bit that adheres to the lips and the gums, where salivary cortisone correlates much better. And then finally, people receiving IV hydrocortisone, still salivary cortisol correlates better than salivary cortisol. So now it turns out that most labs that are measuring salivary cortisol by mass spectrometry, they actually also measure salivary cortisone. They just don't report it, but you can ask them to report it. So I think we're gonna see a lot more use of this in the future. In fact, there has been the development of a completely non-invasive cosyntropin stimulation test that uses intranasal cosyntropin and salivary cortisone that was published earlier this, well, last year in the JCM. And I think that in the future, this is gonna be a way that we can do completely non-invasive testing at home or in the office without having to stick in any IVs. All right, so what other diagnostic tests do we need to consider? There's the insulin tolerance test or the insulin hypoglycemia test. And we don't do this a lot anymore. It's still the gold standard for secondary adrenal insufficiency. And it's still used in some specific instances. We give 0.1 to 0.2 units per kilogram of regular insulin as an IV bolus. And we measure glucose and cortisol every 15 minutes for 90 minutes. The current gold standard is a cortisol greater than 18, but as I mentioned, with the new assays, it's probably less, it's probably closer to 15, and to define it as normal. But in order to fail the test, you need to document a glucose of less than 40 milligrams per deciliter. Now, of course, if the cortisol goes way up to the 20s, and you don't find, you didn't catch the glucose at its nadir, it's still a pass. But in order to call it a fail, you have to document that the glucose didn't fall adequately. So similarly, the miteripone test, miteripone is kind of hard to get these days. You give 30 milligrams per kilogram up to three grams at 11 p.m., and you get blood tests at 8 a.m. And by being 11-hydroxylase inhibitor, it should block that last step and cause accumulation of 11-deoxycortisol. And so if the 11-deoxycortisol rises to greater than seven micrograms per deciliter, that's a normal response. Now, in order to fail the response, you have to have an 11-deoxycortisol less than seven, as well as a cortisol less than five to document adequate inhibition of 11-hydroxylase. There's also the one-microgram cosytropin stimulation test. I don't recommend this because the medication has to be diluted and the timing of the cortisol has to be very accurate. And I think DHEA sulfate is a better secondary test than a one-microgram test. So as I said, DHEAS reflects integrated ACTH secretion, and it has very little diurnal rhythm. So even though ACTH stimulates both DHEAS and cortisol production, only cortisol provides negative feedback. So for example, if someone's taking prednisone chronically, lowering their ACTH, lowering their serum cortisol will also lower their DHEAS. And similarly, if someone has a pituitary tumor, it's causing multiple pituitary hormone defects. If they're cortisol deficient, they'll also have a low DHEAS. So this is beautiful data from Baha Arafat's group in Cleveland, where he looked at DHEAS and all these patients had an insulin tolerance test. God bless them doing all these insulin tolerance tests. So he knew who had adrenal insufficiency and who did not. And so the dots are people who had a normal insulin tolerance test. And the triangles are people who failed an insulin tolerance test. And so you see about 60 or so, everybody of 60 had a normal insulin tolerance test. And people below 60, some of them had a normal test and some of them had an abnormal test. So a DHEAS less than 60, particularly in older individuals who normally have a low DHEAS does not prove that you have adrenal insufficiency. A DHEAS over 60 can rule it out. So etiologies, primary adrenal insufficiency. You don't just get primary adrenal insufficiency. You just don't wake up one day and have adrenal insufficiency like you could have COVID or something. Autoimmune is the most common in developed countries, either isolated or very commonly as part of the polyglandular syndrome. And in these cases, they should have positive 21-hydroxylase antibodies. Why 21-hydroxylase antibodies? Well, the adrenal is the only place in the body that makes 21-hydroxylase. So if you have adrenal insufficiency, you should have positive 21-hydroxylase antibodies. Infection, things like tuberculosis. We used to see CMV, adrenalitis in HIV patients before heart therapy became widely available. We can still see it in some places. Fungal, such as histoplasmosis still occurs. And then infiltrative diseases such as hemochromatosis and amyloidosis. Acutely, if we see acute adrenal insufficiency in someone that's not of autoimmune onset, we think about destructive causes such as surgery, hemorrhage, including the Waterhouse-Friedrichson syndrome in association with meningococcemia. But positive antiphospholipid antibodies and the antiphospholipid syndrome predisposes people as do anticoagulants for bilateral adrenal hemorrhage. Cancer rarely causes adrenal insufficiency because the tumors tend to grow around the adrenal cortex rather than in, although lymphoma, which can be diffuse in the adrenal cortex can cause adrenal insufficiency and some other tumors on rare occasion, particularly if they compromise the blood supply. There's developmental and genetic causes and enzyme deficiencies. These should be picked up in childhood. So these developmental causes, DAX1 mutations, often with the contiguous gene syndrome cause what's called adrenal hypoplasia congenita. And this also causes hypogonadotropic hypogonadism. Adrenal leukodystrophy, which is an X-linked disease due to mutations in the ABCD1 gene. This is a paroxysomal long-chain fatty acid transporter. And these people will have elevated very long-chain fatty acids in the circulation. So again, you think of young males with adrenal insufficiency, particularly if they also have neurologic problems. Transcription factors, such as mutations in the SF1 protein. And they often have, particularly with SF1, they often have disordered gonadal development. In fact, they more commonly have gonadal dysgenesis than adrenal insufficiency, but they can have both. There are the familial glucocorticoid deficiencies. So isolated familial glucocorticoid causing a primary adrenal insufficiency. Mutations in MC2R, which is the ACTH receptor, or its associated protein MRAP, and a couple of other genes. In fact, mutations in the STAR gene can cause a type of non-classic lipoid congenital adrenal hyperplasia that is very similar to the familial glucocorticoid deficiency. In fact, it's one in the same. And then the Aladdin syndrome, the AAA syndrome, which is adrenal insufficiency, alacrima, and achalasia, which is something for the fellows to think about for their boards. I'm gonna show you some CT scans of normal and diseased adrenals. So this is a normal right adrenal. It's usually alongside the liver. It looks like a wishbone. It's behind the inferior vena cava. This is the aorta. This is the diaphragm, the limbs of the adrenal, and normally about the thickness of the diaphragm. This is the left adrenal. It's normally sitting in the fat. It's anterior and medial of the kidney and posterior to the pancreas. It looks like a Y. Now, these are not normal adrenals. These are large destructive lesions, and this turns out to be a man who had histoplasmosis and stopped taking his fluconazole because he was feeling good, and then he wasn't feeling so good anymore because he has an undetectable cortisol level. This is an acute onset of enlarged adrenal glands that have blood in them based on this MRI scan, and they don't enhance with contrast, and that's what they look like on pathology. This is bilateral adrenal hemorrhage, so acute onset kind of foggy, strandy areas in the adjacent tissues and low cortisol and aldosterone. So this is that lady that I showed you that we gave 10 milligrams of hydrocortisone to, and she jumped out of bed and started dancing. She has very, very tiny adrenals, just little wisps that you can see. You can't even see both limbs on the left adrenal, and you can barely see one limb on the right. So the dominion of adrenals that you would see in autoimmune adrenal insufficiency or chronic glucocorticoid therapy. So when you see someone with primary adrenal insufficiency, if it's infants and select children and adults, then you wanna first think about genetic causes. So you measure 17-hydroxyprogesterone because the most common cause would be 21-hydroxylase deficiency or some other form of cortisol deficiency, AH. If that's negative, then you think about these other genetic syndromes such as adrenal hyperplasia genita. And if that's negative, then it's some undescribed or form of primary adrenal insufficiency, or it's just something you can't test for. It may be one of the familiar glucocorticoid deficiency syndromes, your genetic testing didn't quite get around to getting those rarer genes. Now, let's say you have an adult, we usually get 21-hydroxylase antibodies, and most of the time they're positive and that diagnoses autoimmune adrenal insufficiency. And we of course have to think about polyglandular syndromes in young adults who also have mucocutaneous candidiasis and or primary hypoparathyroidism. When you think about polyglandular syndrome type one, in older adults who have vitiligo, egg-blind diabetes and or hypothyroidism or maybe Sjogren's syndrome and maybe pernicious anemia, we think about polyglandular syndrome type two. If those are negative, if you have a young male, you should measure very long chain fatty acids to diagnose adrenal insufficiency. And this is a hugely important diagnosis. About 9% of young men with primary adrenal insufficiency will have ALD. And this can be life-saving because they won't necessarily die of the adrenal insufficiency, but could die of the adrenal leukodystrophy if they have the cerebral form and a bone marrow transplant can be life-saving in these individuals while you give them cortisol replacement. If those are negative, then we do a CT scan of the adrenals. We can pick up things like infiltrative diseases, hemorrhage infections, and malignant tumors. Secondary adrenal insufficiency, drugs, drugs, drugs, drugs, drugs, glucocorticoids, 99%. Also things like migestral acetate when it's used in high doses for its appetite effects or medroxyprogesterone acetate when it's used at quite high doses to treat things like endometriosis and so on. Narcotics is becoming recognized now as a cause of adrenal axis suppression. Usually these people make some cortisol and they usually can respond to stress, but it's quite variable and it's something that you should think about. Mostly we see it as sort of artifacts and kind of low baseline cortisol values in people who are on chronic narcotics. And if you think, well, will I make these people better by giving them glucocorticoids? Well, chronic pain people, a little bit of hydrocortisone is not gonna make a huge difference in their life, but it's worth thinking about. Obstruction from surgery, tumors, pituitary apoplexy and radiation therapy, lymphocytic and granulomatous hypophysitis. TB and other infections can also involve the pituitary and the pituitary stalk. Chemochromatosis and sarcoidosis can be infiltrative causes of secondary adrenal insufficiency and then genetic or developmental causes. Isolated or combined pituitary hormone defects, things like septo-optic dysplasia from all those mutations. Isolated ACTH deficiency due to mutations in CBX-19, which is also known as TPIT or POMC. And then autonomic failure, interestingly enough, sort of like narcotics can cause kind of a wimpy hypothalamic pituitary adrenal axis, but usually they respond to stressors. So if you think about, what about people who just had pituitary surgery? Did the surgeon damage the pituitary gland? So a lot of work has been done on this. The ARIFA group has done a lot of work. We've done some work on this. So we try to minimize their perioperative glucocorticoid replacement, and then we can test them perioperatively. And a cortisol less than about 100, which is three or four, especially if there's multiple pituitary defects, especially if they have cortisol deficiencies before surgery, they're not recovering and they need full replacement. Remember, people can actually recover adrenal axis function if they have partial adrenal insufficiency, partial central adrenal insufficiency from a mass effect prior to surgery. You can get recovery of function. If the cortisol is over about 285, which is like nine micrograms per deciliter, they're okay. And you can confirm that later on, but they probably don't need any treatment. If they're in between, then we can either test them initially or get a DHEA sulfate. And a DHEA sulfate above 1,500 nanomoles per liter, which is again about 60 micrograms per deciliter, they should be okay. At some of these people, you just can't tell, you might have to do dynamic testing like an insulin tolerance test. Another way to look at this is just measure morning cortisol and measure a DHEAS in the people who are in between. And if they're normal, you don't need to test them. If they're low, you should probably do an insulin tolerance test before you step into lifelong glucocorticoid because it is a big deal. They need the chronic replacement therapy. They can't get life insurance. They can't serve in the military or be on the fire department. Maybe they don't wanna do those things, but you've really committed them to a lifetime of a major health burden. And it's best to let them recover function if they really don't need it. So today, this was a recent review from German registry from Stephanie Hanner's group just about five years ago. People still mainly die of gastrointestinal or other infectious illnesses. Now, then you have a smattering of other things like wasp stings, hot weather, sport, pregnancy, alcohol intoxication, and then things like emotional... I mean, that was the only identifiable thing. They probably had an occult infection, but don't know. Surgery were not that common, but it still happens. And so I think there is a range of things, but still far and away, astroenteritis and other infectious illnesses in people with known adrenal insufficiency are the main causes of adrenal crisis. So we wanna treat that by giving people their emergency parenteral hydrocortisone, and it's supposed to be given IM. But what if you can't give it IM? Can you give it sub-2? Well, the answer is yes, because Stephanie showed that, that you can give sub-2 hydrocortisone and you get roughly the same pharmacogeny. It's a little bit slower onset. You get peak cortisol levels that are well above 30 micrograms per deciliter. And so if they can't give it IM, it's okay to give it sub-2. Now, what about new treatments? So you may have heard about modified release, hydrocortisone preparations, and continuous subcutaneous infusions of hydrocortisone. Well, so these are measurements of salivary cortisol to reflect the delivery of the subcutaneous infusion. And you can get a smoother curve, sure, than you can with oral hydrocortisone. But the oral hydrocortisone is pretty good, and people ask about this. And there's some studies that show that people feel better with it. And I'm not knocking it completely, but, you know, our type 1 diabetics, they love their pumps. But if I could tell them that you could just take two or three pills a day instead of the pump, every one of them would take the pump off. So I'm not sure what's so great about the subcutaneous hydrocortisone. The study from Australia showed no improvement in quality of life. Other studies have shown improvement. It's useful in patients with 21-hydroxylase deficiency that have rapid cortisol metabolism. That's probably the best documented use. It's out there for what it's worth, and there are some people that benefit from this. So what about critical illness? I could spend a whole nother lecture on critical illness, but what I want to emphasize is you should consider special circumstances. So medications predisposing to adrenal insufficiency, recent glucocorticoids, high-dose migestral acetate, ketoconazole, etomidate, high-dose narcotics, when the critical illness is due to meningococcemia or other forms of sepsis, when they have hypoproteinemia, such as liver disease, malnutrition, nephrosis. Now, in these cases, the total cortisol is low, but actually the free cortisol is normal. So this is an artifact. These people do not actually have adrenal insufficiency. They have a low total cortisol, but they have a normal free cortisol, particularly when the albumin gets to be less than 2.5. A total cortisol of nine or 10 is totally adequate. So how do we recommend testing? Look, if someone's stressed and critically ill, a random cortisol is usually sufficient. And again, if the albumin is low, cortisol is less than even 10, it may be normal. So you rarely need cosyntrope stimulation testing, and the delta in the stim test in someone who's critically ill just tells you how sick they are. The sicker they are, the less they stimulate. So I gave two multiple choice questions. And since I have a little bit of time, I'm gonna run through these for the benefit of the audience. 40-year-old Latin American man, he's had adrenal insufficiency for five years due to hemorrhage, improved after treatment, but not baseline. He's on hydrocortisone 15 and five, and fludrocortisone at one milligram a day. He wakes up refreshed, but he has fatigue during the day. So he doubled his hydrocortisone for three months, and it didn't help. On exam, he doesn't have any hyperpigmentation. He has some mild skin thinning. His seated blood pressure is 125 or 80 with a heart rate of 68. His standing blood pressure is 105 or 60, heart rate of 92. His sodium is 140. His potassium is 4.7. His glucose is normal. His ACH is high. His DHEA, of course, is undetectable. And he has primary adrenal insufficiency, so his aldo is undetectable, despite a renin that's elevated. And his testosterone is normal as well. And so you give him his hydrocortisone dose, and six hours after the dose gets up to 43, and then six hours later, it's nine micrograms per deciliter. So my question is, what's wrong? What would you do next? Would you divide the hydrocortisone three times a day? Would you switch him to prednisone? Would you increase his fludrocortisone, or would you add DHEA? And so the point here is that this man is volume depleted. He has orthostatic hypotension. He has a potassium that's on the high end of normal, and he has an elevated plasma renin activity. He's not getting enough fludrocortisone. So this is what you see when people are chronically volume depleted, that they're okay first day, but then they get tired as they go through the day. And when you increase the hydrocortisone, it doesn't help. You've already proven he's got good glucocorticoid exposure. So this is not gonna help. In fact, prednisone's a lousy choice for adrenal insufficiency. And so I would increase his fludrocortisone. And then finally, a 28-year-old white woman who has adrenal insufficiency of autoimmune etiology for nine years. She's 15 weeks pregnant. She feels fine. She's on hydrocortisone 10, five, and five and fludrocortisone 0.1 a day. She has some slight hyperpigmentation, but no bruising. Her pressure is normal sitting and standing. Her labs are, her electrolytes are normal. And of course, you know, her ACTH is still high and her aldosterone is still low. She's never gonna make KGAS. So, and her renin is normal. So what do you advise her to do? Do you want her to double her hydrocortisone throughout pregnancy? Do you want to increase her hydrocortisone in third trimester? Do you want to switch her to dexamethasone? Or do you want to give her some DHEA so she can make estriol? And so the answer is, you know, she's doing well, but you don't need to increase her hydrocortisone now. But commonly in the third trimester when CBG gets really high and they're getting very large, they need a little increase later on. So it's sort of the opposite of thyroid hormone where we think of increasing the thyroxine early in pregnancy. This, we usually increase the hydrocortisone a little. You don't have to, but it's often necessary. Or at least you should make sure that she's okay in the third trimester. But most of the time they need a little increase. You wouldn't put her on dexamethasone. That's contraindicated during pregnancy because it can cause midline defects in the child. And then DHEA that fuels estriol synthesis in the mother actually comes from the fetal. So she'll make estriol, even though she's not making DHEA because the baby is going to be making DHEA for her. So that's all I have. And now I'll close for questions. All right, very good. Well, we're ready for our question and answer. And we have a few that have rolled in. So one question, thinking of hypercortisolism, would salivary cortisone be used instead of salivary cortisol? With the normal cortisol cutoff for adrenal insufficiency due to better assays, should cutoffs for hypercortisolism be adjusted as well? Wow, that's, okay. So that's a great question. Actually, some of the great minds think like, it's something that I thought of. You know, Jay Finling's group also published a paper recently where they looked at salivary cortisol and cortisone by mass spec versus the immunoassay, the salimetrix cortisol for diagnosing Cushing's. And they found that salivary cortisone performed about as well, maybe slightly poorer than the salimetrix immunoassay, which again is a little nonspecific because it cross-reacts with the cortisone. So I actually think that that needs some further looking at, whether salivary cortisone is what we should be measuring to diagnose Cushing's. But that really seems like it should perform better. Now, the saliva assays nowadays mostly are performed by mass spec. And so they have their own kind of normal range. And I think it depends who you talk to, but generally, you know, if it's, most people when you measure salivary cortisol at midnight, it's zero and it's undetectable. And so we usually say like sort of two to two and a half times the upper normal is confidently abnormal. So whatever we get when they go around, get around to doing the normal ranges for salivary cortisone, I think we're going to have to have, you know, some buffer between the upper limit of normal and what we're going to call diagnostic, kind of like we do with urinary metanephrine. But that is a really great question. I think we're going to see a lot of change over the next couple of years about the saliva assays. Thank you. All right, next question. So for patients with joint injections, cosyntrope and stimulation tests results may be flawed. So how long do you typically wait and what would you advise that we do? Right, so the narcotic effect is a very interesting one. I didn't talk about that that much today. It's usually partial. And in the old studies that were done, there's not a lot of literature. People still like, for example, on insulin tolerances, they still rise on insulin tolerances. These don't rise as much. So I think this is, you should look at this as sort of a partial central adrenal insufficiency in most people rather than a complete. And so it's not really clear if these people are at risk for adrenal crises. That's my point because they still make some cortisol. So it's not as bad. Now, how long to wait? Well, with the narcotic effect actually is immediate. It's rapid on rapid off. So as soon as they stop, if they can stop, you should have a normal response. And the exception would be if people are on really high dose with combinations of narcotics and prednisone where their axis has been chronically suppressed for a long time. If it's been partially suppressed and intermittently partially suppressed, it should recover right away once you stop the narcotic. Now, the joint injections, when they give glucocorticoids in the joint, this is tricky because it depends on the glucocorticoid. So methylprednisolone semisuccinate, which is solumedrol, solumedrol is a soluble drug. And so it has a predictable clearance. The thing that kills us, and you remember this, is Kenalog. Kenalog is triamcinolone acetonide, and it is a suspension. So Kenalog is like a milky thing, and the orthopedics love Kenalog. And because it's a suspension, it is not soluble. It is highly variable how long it takes to be absorbed. So people who get a shot of Kenalog intramuscularly or like a facet joint injection, that can last a year. You know, 40 milligrams of Kenalog is a year's supply of glucocorticoid, literally. And so people can be suppressed chronically. You can measure it in the urine. You can send to Mayo their synthetic glucocorticoid, and you can measure it months after they get the injection. So that's the one to worry about really, really prolonged suppression with. All right, thank you. I think we have a couple of questions that came up about adrenal insufficiency in the inpatient setting. So when you're concerned about that in someone who's hospitalized, albumin levels may be low for various different reasons, how you would assess that patient? Yeah, so Amir Hamrahi in a classic study from the New England Journal, he showed people with an albumin less than 2.5, their stem tests are like 10 is normal because their free cortisol is actually proportionally greater. So, you know, I think this is part of being a clinical endocrinologist is that you have to make clinical decisions there. You know, DHEA sulfate does not help you in that situation because there is actually a dissociation. You actually make more DHEA in critical illness than DHEA sulfate. And ACTH rises initially, and then it falls and cortisol stays up. So Greed Vanderburg did some great studies that cortisol clearance is very much impaired in critical illness. So that little bit of cortisol stays around a long time. So I think, you know, you look for the classic feature, you know, the hyponatremia, the hypoglycemia, things like that. And then, you know, I think it's okay to give people a test dose, but, you know, a test dose, again, they should have that Lazarian response. And if they don't, then it's not the cortisol. All right, you know, if they're hypotensive and you think it's because of adrenal insufficiency, one dose, they should respond. Their blood pressure should be normal. If they don't, then I would say, stop it. So I think it's okay to do that when you have these questions. But I think, you know, again, people place too much emphasis on the laboratory tests these days, and they don't place as much emphasis on the clinical evaluation. You really need to trust your clinical evaluation. Thank you. All right, it looks like we have some questions about antibody testing and primary adrenal insufficiency. So you had mentioned earlier that with primary, often 21-hydroxylase antibodies will be positive, but a couple of people are interested to hear your thoughts on times when there's kind of a classic presentation for primary adrenal insufficiency with negative antibodies or positive antibodies, but normal testing. So, right, they're not always positive, that's true. And, you know, this is where, once you've diagnosed adrenal insufficiency, you're trying to figure out the etiology. If the adrenals are atrophic and they have negative antibodies, it's probably autoimmune. If they have other, if they have vitiligo, if they have hypothyroidism, it's probably autoimmune. All right, so, but, you know, it's a matter of whether you should keep searching. So in cases like that, you might want to do a CT scan where they have negative antibodies to make sure they don't have big adrenals, in which case you're worried about fungus balls and disseminated infections like that. But if they're small, then I don't think I would mess with it anymore. A very interesting question about what do you do with people with positive antibodies? Because we've had situations where we're testing first-degree relatives of type 1 diabetes in a trial net study where they're looking at other antibodies, you know, that, and positive, if you have a first-degree relative with type 1 diabetes and positive 21-hydroxylase antibodies, you're actually very high risk of developing type 1 diabetes but you're not a high risk of developing adrenal insufficiency. So we don't quite know what to do with those people right now. The one exception though, is if you have autoimmune polyglandular syndrome type 1, which is the one that causes mucocutaneous candidiasis and hypoparathyroidism, if you have positive 21-hydroxylase deficient, positive 21-hydroxylase antibodies, you will get adrenal insufficiency, like 100% guaranteed that you will get that. That is the one situation where it's highly predictive. So that is where I would pay much attention. Other people, I think I would just get a morning cortisol and DHEAS once a year, because we really don't know what the predictive value of that is in those cases. All right. And moving on, we have a good question about ACTH, just kind of issues with the actual assay. So do you have any recommendations for how ACTH should be drawn and stored on its way to the lab? Yeah, there've been some issues with this and there are some, you know, like there are sometimes there are these, you know, interfering antibodies and, you know, Jay Finlay and a couple other people published paper on this recently and they saw some artifacts with that. So remember, yeah, I mean, I think that if you get a result that doesn't make sense, that try sending it to a different lab or ask your lab to do the polyethylene glycol precipitation procedure to get rid of those human anti-mouse antibodies that could be interfering with that. That's my best suggestion. In terms of handling the specimen, it's best to draw it, spin it and freeze it. Kind of like a plasmarine, draw it, spin it, freeze it as quickly as you can. I don't think it matters what temperature so much you use centrifuge at, as long as it's not warm. But, you know, rather than, I don't think it's so critical to ice it as soon as you draw it, but it's important to freeze it as soon as you separate. All right. We have a question about non-oral contraceptives. So things like IUDs, subcutaneous implants and how those impact cortisol binding globulin and total cortisol compared to oral estrogens. Well, most of those actually don't contain any estrogen. Most of the IUDs and stuff are progestin only and the implants are progestin only and they don't have much impact. I think topical estrogens hasn't been studied that well because, I mean, you know, the patches and the gel, you know, I think that they still do affect the CBG. They may just not affect it quite as much. All right, thank you. Would you be able to comment on some of the recent studies in the UK and the US that have suggested prednisone dosing at four or five milligrams a day in the morning may be a better physiologic mimic than our TID hydrocortisone dosing? You know, that was prednisolone, okay? They don't use prednisone in the UK and I'm actually okay with using prednisolone, which nowadays, you know, the pills have gone outrageously expensive and you have to use the liquid. I actually use a lot of methylprednisolone because that is cheap. That's the generic of Medrol and a lot of my CH patients are taking a morning dose of methylprednisolone and a very small dose at bedtime. I think that's okay. I really don't have any problem with that. Just remember that it's easier to overdose people with those more potent regimens. It's very hard to overdose people on hydrocortisone and make them cushy, but it's very easy to do that with something like prednisolone or methylprednisolone. So it's okay. Careful that you don't overdo it. What's your strategy for monitoring the potassium level in patients who are on potassium-sparing diuretics and also have adrenal insufficiency? Oh, that is a tricky situation. I don't like when that happens. So, you know, again, I think you sometimes, so there's occasionally a CH patient that comes to me taking spironolactone. I was like, whose idea was this? So, which is one of the reasons why we're trying to develop better drugs for that. But I think you sometimes do have to raise the flucocortisone dose to compensate for it. And so, again, standing blood pressure, number one, potassium, number two, plasmarin, and number three. Those are your three ways to monitor treatment. And I think that's generally a bad idea. I would try to find something else. You know, you can get like, if you need an androgen antagonist, you can use bicalutamide off-label. That's what I sometimes do in a situation like that. Do you have a recommendation about the ideal frequency of steroid replacement, whether that be twice a day or three times a day? Well, I mean, three times a day is ideal, but very few people can do that. I think twice a day is pretty good in most people and it suffices. And this relates back to the prednisolone question. We don't really have an ideal pharmacokinetics from our hydrocortisone preparation. So these new products might be a good thing. I actually think these delayed release, modified release hydrocortisone are gonna be great for adrenal insufficiency, for Addison's type patients, because I think they're gonna be much smoother pharmacokinetic you take it at bedtime, it doesn't come into your system until around 3 a.m. when cortisol is normally rising. I think there's a long way to go on that. Neither is great. Again, it's more, it's just part of the negotiating that you do with patients. If people are willing to take three times a day, they can. I'll tell you one thing that I've had some CH patients that use this company called PillPack. I think they're in the Northeast. They actually will make blister packs for you where they'll break the pills and everything and package them in blister packs. I had a student at U of M that she would just rip off a strip in the morning, take her first pill, go to class, and then her phone would ring when she had to take the second and third. She actually took four doses a day and she was extremely compliant with that. So there are kind of low tech solutions to this that we can use with our current regimens. So I think it depends on how motivated the patient is and how well they're willing to split things up like that. All right, thank you. So for patients with adrenal insufficiency who are increasing their physical activity, so things like running for a long distance or for a long time, going on a long hike, do they need to stress dose before activity? You know, a lot of my European colleagues say no. They have people that do orienteering, which is pretty intense when you do it professionally, and they don't actually stress dose them. I tell people if it's gonna be, like if you're gonna do, you know, I've done a couple of half Ironmans, you know, and if you're gonna do something where you know, like if you can take your dose before that, right? If it's usually these races are more, you take your dose 30 minutes before the race starts, then you're usually covered unless it's like a six or seven hour race. And then I say, if it's more than four hours after you've taken your dose, then take another five milligrams if you can. That's really the only thing I say. So I think their usual dose is adequate. It's just a matter is if they run out, you know, at the end of that duration, because it's not so much, they don't like need a cortisol of 30. They need a cortisol of 10, but they need it for six or so hours. And that's not the way the pharmacokinetics are. You're gonna get a peak and a trough. And so you just gotta boost that to get them up above 10 or so when they start to drop. Thank you. For patients who are basically flipping shifts, so they're working night shifts, they're working day shifts. What are your thoughts about their hydrocortisone replacement? Yeah, these are the ones that give me headaches. So I try to always give them their largest dose first thing in the morning, whenever they're on awakening, on arising. So if they take a big dose in the morning, a little dose in the afternoon, and then they're gonna have a short day and they're gonna go back to work, take another morning dose when you get up and go to work. And then you take a smaller dose in the morning, then you're gonna go to sleep in the afternoon. So I think the key is that they always take the large dose when they wake up, okay? And so you may skip an afternoon dose. You may give two afternoon doses, but then they get back to their big morning dose. Never take a big dose before you're trying to go to sleep. All right, perfect. So I think we have time for one last question. Should renin be routinely followed as you adjust fludrocortisone? I do. I mean, certainly standing blood pressure and potassium. With video visits, I don't have the standing blood pressure, so I've been ordering it routinely. I think once people are on autopilot, you may not need to do that routinely, but I find it useful. I don't know if it has to be done at every visit, though. I think certainly in the new patient who you're treating, people who are symptomatic, they don't feel good, and you're not sure what's going wrong, you should check that. But I don't know if it needs to be done every visit. All right, great. Well, thank you to Dr. Aukas for being here for this session and thank you to all of you for the great discussion. Hope you all have a good meeting. Okay, thanks for attending.

adrenal insufficiency

diagnostic testing

autoimmune

21-hydroxylase antibodies

adrenal crisis

management

steroid replacement therapy

electrolytes

blood pressure

diagnosis