Diabetic Ketoacidosis

With over 400,000 type 1 diabetics living with the disease in the UK, the likelihood that paramedics and emergency responders will encounter a patient in crisis is quite high. Diabetic ketoacidosis is undoubtedly a problem that you will have encountered in your own practice, and effective pre-hospital recognition and management is key to starting a potentially life-saving pathway of care. This month we're taking a look at DKA. We look at the pathology of the disease, what common triggers are that can send a patient into DKA, and how we can recognise and treat a patient who's suffering with it. So, if you don't know why we need to tailor our fluid rehydration carefully in this patient group, if you're not sure how you take a history to look for the common precipitating causes, or you just want to know how these patients are managed in ED or ITU, then keep listening. This episode is going to be sweet. I'm your Lens General Broadcast, and it vehicles available to Broadcast or come clear for an outstanding, cash free run emergency. So, hello and welcome to General Broadcast, my name's Josh. I'm Simon, and I'm George. And George, you're a new voice to the podcast, so I don't you tell people that are listening a little bit about who you are. Yeah, thanks. I'm a new voice around here. It's great to be part of the team. I've just joined. I'm George. I'm a doctor working as a clinical fellow in intensive care, and my previous life I was an emergency department charge nurse. Great. This month we are talking about DKA, which is a topic that many people will be aware of. It's definitely something that's covered in most of our training, in medical school and paramedic training. But George, it's probably something that you see a fair amount of working on ITU. Certainly, DKA is very prevalent. Not every patient with DKA comes to intensive care, but many get reviewed by intensive care and are admitted to hospital every day up and down the country. The ones that come to intensive care are usually the ones managed by me. It's because you keep them alive long enough to get up to theirs, isn't it? Absolutely. That's exactly what I meant. Before we get started, just a quick thanks to our sponsors, Gigi Medex, who continue to support this podcast, keeping it free and open access to our listeners. Thank you to those of you listening that have supported us on coffee.com. We really appreciate any coffees that you do by us. As we say, they go straight into helping support our hosting fees and our website hosting fees. So, if you have had benefits from this podcast, if you like what we're doing, and if you met us in your life and you buy us a coffee, why not buy us one online? Go to coffee.com/generalwalkost and you can follow the details there on how to do it. Okay, well, we said we're talking about DKA this episode. Why don't we start off by just a couple of minutes revision on what diabetes is? Now, you can find a far more comprehensive review of this in our HiPoglycemia podcast, which will be linked in the show notes for this one. But diabetes in 60 seconds or thereabouts? Well, we know in normal physiology, insulin is produced and secreted by the pancreas. It has exocrine functions that perform things that help with digestion in the stomach and endocrine functions that involve the secretion of insulin in cells called the eyelet of Langerhans. Insulin is necessary to drive glucose into cells where it can be utilized for cellular function. This is all cells in the body apart from those in the brain, red blood cells, liver and kidneys and also the eyes because they use a different transporter molecule. But otherwise, all other cells require insulin in order to uptake glucose. Diabetes is an issue with this insulin production or utilization and type 1 diabetes, which is what we're going to be focusing on in this podcast, is a problem with the production of insulin, often caused by an autoimmune disorder that causes the destruction of these insulin producing beta cells in the eyelet of Langerhans. Now, without the production of insulin cells, cannot effectively take up glucose, resulting in a high level of glucose in the bloodstream. This causes a variety of micro and macroscopic issues and, as is the focus of this podcast, an inability for cells to get access to glucose molecules for normal function. And this means that they need to resort to other methods of energy production. So Simon, with that in mind, do you want to talk to us more about the pathology of DKA? Basically, there is a significant insulin deficiency and an increased counter-regulatory hormone release, things such as cortisol. And this results in increased gluconeogenesis, which is essentially the creation of glucose. But the key part of that is that it comes from non-carbohydrate sources and increased glycogenolysis, which is the breakdown of glycogen into glucose. And that's why we give glucose gone, one of the reasons we give glucose gone for hyperglycemia, is because it kind of stimulates that process. But this effectively increases glucose, but obviously, due to that insulin insufficiency, we can't uptake that into cells. And therefore, the blood remains hyperglycemic, but the cells are effectively staffed of glucose. At the same time, because of there is insufficient glucose utilization, our backup systems kick in and free fatty acids are released into the circulation from adipose tissue. And this is broken down in a process called lipolysis, so effectively lipofat, lysis, breakdown. These undergo hepatic fatty acid oxygenation, and that creates two predominant substances known as beta hydroxybutyrate and acetate. And these are collectively known as ketone bodies. And both of these ketone bodies are acids, and because they're acids, this raises the level of hydrogen ions in the blood, making the pH become more acidotic, and this is the resulting metabolic acidosis that we see in DGA. A C2 acetate apparently has that characteristic acetone smell on the breath. There's some people can smell, and I've been told it smells like nail polish. I personally can't smell it or haven't been able to smell it. What do you guys think? So George, can you smell it? Yeah, Simon, I'm glad you brought that up. I think exactly the same thing. It's the classic sort of Dr. House diagnosis. Oh, I can smell paedrops on people on this patient's breath. This must be DGA. In my time, I don't think I've ever reliably smelt the paedrops smell. So, coming back, the hyperglycemia itself causes increased osmolality. This is effectively a measure of the concentration of dissolved particles in the blood. And due to this, higher osmolality, this creates a concentration gradient which moves water from cells in the circulation and then as it passes through the nephron in the kidney, the glucose enters the urine, pulling water and electrolytes with it, resulting in an osmotic diuresis and polyurea. This eventually causes dehydration, which triggers excess thirst and polydipsia. Also in the urine, ketones are excreted. So we can test for this with a urine dip and this will show us ketoneuria. We can also, however, detect them in the blood as ketonemia. And in fact, many modern glucometres will measure both glucose and ketones. Certainly, the ambulance service where I work locally, their monitors can measure both. And this is really useful for the pre-hospital increased diagnosis of DKA as opposed to just hyperglycemia. So whilst we're a little bit of a way from routinely measuring blood gases in pH pre-hospital at the moment, unless you've got critical care with near-patient testing, you know, it does add another one of the diagnostic features, which we'll talk about in a second. We should be aware, however, that despite the fact I said that there is osmotic diuresis and this causes polyurea and increased urine output, eventually they'll become a point where the patient becomes so profoundly dehydrated that shock develops in oliguria, will and stew. So coming back to that diagnosis of DKA then, so we need three of the following criteria, hyperglycemia with a glucose over 11 millimoles for liter. So we'll come back to that one in a minute, because actually you don't always need hyperglycemia to have a DKA. But in general, hyperglycemia with the glucose over 11 millimoles a liter, you need a ketonemia of 3 plus millimoles a liter or a urine ketones of 2 plus or more on a dipstick and you need a metabolic acidosis with a pH of under 7.3 and/or a bicarbonate of under 15. In response to the development of DKA and a metabolic acidosis, the change in pH will trigger the respiratory center to increase respiratory rate and volume and attempt to blow off carbon dioxide. Carbon dioxide, when mixed in a solution of water, is a carbonic acid and this reduces hydrogen ions by blowing off additional carbon dioxide and an attempt to restore pH back to normal levels. This process, therefore, is known as a metabolic acidosis with respiratory compensation and this gives us that classic deep-sying, kush-smile pattern of breathing that we see in these DKA patients, which we need to make sure isn't being confused with a respiratory presentation, especially the times I've seen it in like children. So, and a first presentation of DKA, and a lot of people work them up for kind of respiratory conditions 'cause they come in with like a packet near and short of breath, but actually the underlying cause is undiagnosed type one diabetes, and they are in a DKA, and that's their first presentation. Other symptoms the patient may experience is non-specific abdominal pain, as a result of the ketone bodies and acidosis, which irritates visceral organs, and stimulates nociaceptive pathways. And this causes that diffuse abdominal discomfort, as well as that osmotic dioresis, which causes electrolyte loss, resulting in hyperclemia, and hypophosphatemia. And depletion in these substances impairs gut metility, smooth muscle contraction, which leads to cramping abdominal pain, and even inducing a paralytic Ilius and gastric stasis. Furthermore, significant volume depletion, and hypobelemia, may cause mesenteric hypopuffusion, the measentary being the gut's blood supply, and this splunch-nic low blood flow, can cause ischemia and therefore pain. The same process is caused gastric irritation, delayed bowel emptying, and elevated counterregulatory hormones, which contribute to nausea and vomiting. Finally, the neurologic deterioration that we see in patients with DKA, is caused by that extreme metabolic acidosis, and its complications, such as the cerebral edema, which we'll talk about later when we're talking about fluids. It's worth mentioning about the counterregulatory hormones, which include cortisol. Cortisol is a steroid, and steroid increase hypoglycemia, which in themselves raise blood glucose. And therefore, it's one of the reasons we need to kind of give extra monitor repercussions to patients if we are supplying or prescribing them, with steroids for other conditions, when they are also concurrently diabetic patients. There is a caveat to what I just said. So we talked earlier about the diagnosis of DKA and having those three criteria. Those were hypoglycemia with a glucose over 11, ketonemia of three millimoles a liter, and a metabolic acidosis. So there are times when a patient can actually present in DKA, but with a normal or only mildly raised glucose level when testing. And this is known as euglycemic DKA. The euglycemic DKA is relatively rare, but effectively, it is the same process, but patients who, for example, are taking SGLT2 inhibitors, those are drugs like dapaglyphlosen, which some diabetics are on. It effectively reduces the amount of glucose that we find in their blood. It also increases the urinary excretion of glucose, and therefore glucose levels are much lower than what we'd expect even at times normal levels. But the same response mechanism of ketone body production and metabolic acidosis occurs because the cells themselves are still not getting any glucose uptake. So just keep that in mind. I've only ever seen it once myself in clinical practice. I don't know if you guys have come across it at all. Yeah, it's like you say it's really rare. There are more patients that are taking SGLT2 inhibitors, but it's rare that we see it. Yeah, that's really interesting, George, because actually, that means then that we may actually see an increase in how many of these euglycemic DKA patients we come across as more patients are put on these types of drugs. Sure, and I also wonder maybe how many patients, certainly in the emergency department who might come in in an early DKA, which we never really pick up because their glucose was never high. So no one ever really thought to check their ketones. Brilliant. Thanks for scaring me that how many of these I might have missed over the last six months. Right then. So finally, although we're not going to focus on this topic, in fact, it's probably worthwhile of a separate podcast itself. We should be aware that there is another condition that causes a hyperglycemic emergency. And this is that of HHS or hyperglycemic hyporosmotic state, previously known as HONK. And basically in HHS, there is still insulin deficiency, but it's too a lesser degree. And therefore, because some insulin is effectively able to break down glucose, lipolysis is prevented, meaning that those ketone bodies, which are essential to the formation of DKA, are not created. It's a slower process than DKA, but over time, it often presents with a much more profound hyperglycemia because there's a higher level of glucose over time, a larger osmotic effect on fluid levels. And this results in significant polyureria and fluid loss. And these patients are often significantly dehydrated by the time they seek medical help and need often absolutely liters of fluid. But the key is that this significant dehydration, unless there's obviously evidence of shock, needs relatively slow replacement over time. Textbooks say that DKA is more common in type 1 diabetics, HHS is more common in type 2. So, but I think that, you know, actually, it's just being aware that both conditions can exist and we'll probably look at doing another podcast on HHS itself. We'll probably release that as a short 10 minute topic podcast just to supplement this to go over those differences. Cool, so not me doing it then, since I've waffled so long on DKA, I don't know if I can fit HHS into 10 minutes. Right then, so we can't say anything in 10 minutes. No, I can't, though. I can. I can fit something three times into 10 minutes. (laughing) Welcome to the podcast. No, leave that in the studio, I George. You can leave that in your, as my introduction, right? Yeah, this is the last time you'll hear George on the podcast, everyone. Cool, so we've talked about the pathophysiology of DKA, but let's now have a look into the common triggers and what to look out for. With some of break almost over and a new academic year starting, that means a whole new set of oskies to prep for. So why not give yourself a head start with a clinical skills bundle from Geeky Medics? They have over 1,300 practice scenarios, interactive checklists for practicing with peers, 3000 flashcards and their virtual AI patient. And I can't rate highly enough the AI patients. These are a fantastic learning tool, whether you're just starting out with history taking all like me, just wanted to brush up on it before you're prescribing oskies. So check out the link in the show notes below. And this month, why not try to bag yourself a free book? Geeky Medics are giving away copies of their clinical examination and oski prep books to our listeners. Just share a post on Twitter, Facebook, or Instagram advertising this episode or share your favorite general broadcast episode with your followers. Just remember to tag us in the post to be in for your chance to win. Okay, back to the podcast. - So let's think about some of the common triggers 'cause we've got a really good understanding of the pathophysiology and some of the things that we're gonna elucidate in the history there, Simon. But what would be really useful is to know what are the common situations that people resent with DKA. And the number one situation that the main thing that we're looking for here is some form of stresser on the body. Now in about 30% of cases, certainly I think all of the ones that come to mind when I see DKA, this is pretty much always as a result of an infection. Is that fairly, I mean, there are other causes, but infection is often the most obvious and the one I see. Does that ring true with your sort of experience? - I know you're gonna talk about it in a minute, but actually my experience is probably an even split between infection and compliance. So I think that's my experience. - Yeah, so we're looking for stresses on the body. As I say, infection is probably one of the most common. This is as a result of an immune response. So as we know, when we mountain immune response, we get a release of cortisol when adrenaline. And this will affect our ability to convert to glycogen from the liver and will increase. We know cortisol increases our insulin resistance. This is all key to the fight or flight response, as it ensures that we have enough glucose available for the brain and skeletal muscle. And again, that rings true when we think about what we know about the sympathetic response to stresses. However, over days of illness, this will accumulate resulting in a hyperglycemic state. Now the most common sources of infection are often respiratory tract infections, UTI or skin infections. And this hyperglycemic state is magnified or compounded when we introduce symptoms of DNV. Simon, you talked about how vomiting particularly is a symptom to look out for in DKA. Well, if we've got DNV as a result of whatever our illness is, then this is going to compound the effect by further dehydrating our patient. So that's infection sources as a potential stresser. Another common stresser is surgery. We know surgery is a massive stress on the body. And if someone is known to be diabetic, in hospital, they'll take certain precautions to mitigate this hyperglycemic state. George, I don't know if maybe that's something you could expand on a little bit. What precautions? are taken when we do surgery on a known diabetic? Yeah, absolutely. A lot of it comes down to what you do preoperatively, so certain medications might be held, leading up to surgery, some will be continued. But certainly postoperatively, we try and get that patient back onto their normal regime as soon as possible. And that's all to sort of prepare or mitigate the sympathetic response that will calm as a result of surgery and tissue trauma and things like that, is it? Yes, in part, in part, and the other important factor is to make sure that that person is to label to uptake insulin as they would have been able to before. Historically, patients were held off their antibiotics for days around their operation and they had worse outcomes. So other stresses when we think about it, we need to think about physical injury. Even there's been cases where severe emotional trauma has resulted in DKA. And we know that, when I read that, or certainly used to read that, you think, "Oh, that's a little bit woo-woo, isn't it?" That emotional trauma can create this physical state. But we know that there's good evidence for this if we think about, you know, Takasubo's Cardio-ML, the severe emotional states can bring around physical changes and that's exactly the same for a diabetics. And another one that we really need to have in the presence of our mind is an MI. And that's really quite clear in the literature is that we can miss MI's in diabetics. They don't present as typically because of that neuropathic state that we know diabetics can get into when they're exposed to consistently high blood sugars. MI's don't present as classic and it may be that the first sign these patients are presenting to us as a result of having an MI or having had an MI might be when they're starting to tip over into DKA. And then another common cause, Simon, as you alluded to, is treatment non-compliance. So this might be accidental or deliberate. So we should consider self-neglect in this instance. So this might be somebody who's just missed time there in Chilean, hasn't factored in the correct amount of insulin for things that they'd eaten, maybe that they've had a particularly, it might be that they've had a particularly highly excessive sugar-filled diet for a short period of time which goes away from their normal diabetic diet. Or as we say, it could be deliberate or accidental self-neglect of their insulin regimes. Yeah, and I think like in ED, my experience of seeing a lot of these cases are around teenagers. So like when people become adolescents and then they start to feel like they can't do necessarily the things that their friends and other people their age are doing and they feel inhibited by their disease that they've had since maybe when they were young, that maybe their parents used to manage it and they've been given a bit more freedom. Or they're just neglecting to manage it themselves because they are either not managing it because they kind of are irresponsible in the management of it or they become like really resentful against their diabetes and therefore choose to not manage it correctly. So I see that quite a lot in ED and that's often a predisposing factor for some of the younger DKAs I see. While we're talking about that group of patients, probably worth mentioning about alcohol consumption. So typically in diabetics, excessive alcohol consumption will often cause a rebound hypo, so the liver will generally stop converting glycogen stores whilst processing the alcohol, which when consumed on an empty stomach can cause a hypo, I've always called them rebound hypo's, but I suppose they're not really. So that's generally how we may find them presenting. It might be that they've taken additional insulin anticipating that they're going to have a heavy night and that they're going to take a lot of alcohol. So generally they'll present as a hypo, but as alcohol contains a lot of sugar in excess, it can depress the central nervous system and that in turn may impact on their treatment protocols, such as taking insulin. So you can result in a hyper glycemic state as a result of excessive alcohol consumption. So that's probably just something to bear in mind. While we're talking about insulin and treatment non-compliance, insulin pumps are an excellent thing that I suppose they're not really new anymore, but definitely becoming more and more common and has definitely helped people with treatment compliance. But again, these devices can malfunction, so it could be that the insulin pump has disconnected from the site. The cannula may have kinked any kind of device issue. I'm led to believe that they do have inbuilt safety measures to sort of warn or combat against this, but you know, we know things can go wrong, so it could be even though they've got an insulin pump and they might just fire and forget broadly with that because it does make your treatment compliance so much easier, it could be that the pump has malfunctioned and that's something to bear in mind. And then finally, about 25% of DKA presentations can be as a result of a new or missed diagnosis of diabetes. So Simon, sort of as you referred to earlier, it could be that this patient doesn't have a diagnosis of other type one or type two diabetes and we've missed some of the early warning signs of that and their disease has now progressed so much to the point that they're now presenting in a crisis state, which is, you know, this is hopefully common sense to most people, but this is why it's vitally important, particularly in children presenting with vomiting or abdominal pain, we're making sure that we get those blood sugar readings just to assess for this and indeed assess that they are not tipping over into a high-po. For the remaining sort of quarter of cases, there's no known precipitating event. This might be that we just haven't discovered it yet or there's not enough information in the history to to elucidate exactly what it is. So there are some of the common triggers, just to quickly cover those infection is probably one of the most common. Remember, we're looking for stresses on the body. So infections, possibly one of the most common, this can include surgery, severe physical injury, emotional trauma and do not forget about subtle MIs presenting in this population. Could be as a result of treatment non-compliance, either accidental or deliberate. It could be their first presentation of a diabetic problem so they don't have a diagnosis of diabetes but they are now presenting in crisis and then for some cases we're just not going to know what the precipitating event was. So there are some of the common causes to bear in mind but let's now think about physical assessment and what we're actually looking for in this patient group and what our treatment optimization is going to be. So George, do you want to talk to us about physical assessment? Yeah, absolutely and thanks for that Josh. Thanks for covering some of those common causes of DKA because as you rightly say, patients with DKA usually have another illness running in parallel. DKA rarely happens on its own. So once you're examining patients with DKA or with suspicion of DKA, you may also find symptoms that helped elicit what's caused this DKA. You couple this in with the fact that patients with DKA usually have a reduced GCS, might not be able to give you a full history. And so the physical examination here is really important. And we'll look at this from the perspective of working pre-hospitalia. I think on well patients like this it always makes sense to think about your examination in an A2E assessment. So we'll start with airway. Reduced GCS is possible with DKA. If the patients are alert and they're able to speak to you, fine, moving on. But if not, you need to consider if airway maneuvers here are warranted. Formatting is also a common feature with DKA. So sometimes they're best cared for in the lateral position, especially whilst traveling into hospital. Breathing, Simon, you spoke really nicely in your talk about pathophysiology on the breathing that we see in patients with DKA. Remember that DKA has a profound acid-based disturbance and a significant metabolic acidosis. And so these patients work really hard from a respiratory point of view to compensate for this. They do this by trying to remove as much carbon dioxide as possible. And that's why we see this labored cuspal breathing. It's deep, it's labored and they move large volumes of gas. They're often be tecnic and the breathing is deep and noticeable. Oxygenation, therefore, is not usually a problem in DKA because patients are shifting such large volumes of gas. So if you see hypoxia, for example, it's important to think about the other causes of that. For example, could this be pneumonia? And again, Simon, you mentioned about the ketones on the breath. I'd like to touch on that again here. It's really not a very reliable sign, so don't ever count on not smelling ketones. For ruling out DKA. I don't think I have ever, not least, I've seen loads of DKA. It's like the classic thing that everybody talks about when they talk about DKA. And that really is not going to be what gives you your tell-tale signs and makes you think about DKA. I think the only time I've ever smelled sweet sort of sweet breath has been because the patient's been maybe thinking they had a high poll or something and had a load of gummy sweets. I've never never seen it. The description of cosmobreathing is exactly the give away. It's classic in every case of DKA. And yeah, like you say, George is that it's literally the body doing a brilliant job of trying to blow off that gas blow off that acid to compensate for the metabolic acidosis and it's exactly what you do if you've finished a run or something and you're trying to get your breathing under control, it's what you naturally do isn't it? That big deep sying gasp, so that's really the thing to focus on and have clear in your mind. Absolutely, I totally agree. Yeah, I know as a student I was like, "I've never smelled this paired up thing, maybe I'm a really crappy paramedic, I just don't think it's as common as we're led to believe." Yeah, Josh, I completely agree with that actually. The smell of ketones on the breath would not be diagnostic, but actually that deep labored customer breathing would certainly be the sign and in fact if I was seeing a patient in hospital who had a profound acidosis like you see in DKA but wasn't in fact breathing in that way, I would be extremely worried that they're now getting very tired. So moving on to assessing circulation. As we've previously mentioned, as Simon's picked up on, the high blood glucose and metabolites from the other sources of respiration caused this significant osmotic dioresis. This means that these patients passed lots and lots of urine, like you'd see in an undiagnosed or poorly treated diabetic. And later on they start vomiting as well. So because of this, patients with DKA are often profoundly dehydrated. In fact sometimes up to 8 to 10 litres fluid deplete. By the time we see them, they're often no longer polyureic, in fact they're oliguric. Because of this profound dehydration, they'll often be tackycardic and hypotensive. Their skin and mucus membranes will be dry, their tongue will be cracked. Once patients are in a established DKA, they're really unlikely to be polyureic. And then additionally due to this dehydration, older patients, those with cardiac history and even as a consequence of the severe electrolyte disturbance, these patients can end up in arrhythmias such as AF. They can be bradycardic. They can even have serious dysrhythmias, even like Torsarts. These are the circulatory effects of DKA, but it's really important to also remember that there may be circulatory effects as a consequence of the illness that's triggered DKA, for example if they've had an MI. Quite a lot of the time we've seen these patients in very established DKA, if they're quite lethargic and reduced-disch GCS, pre-hospitalia, they've been incontinent right now and that's our big giveaway there. Obviously we don't have the ability to measure urine output, but often I've seen these patients have been incontinent and it's not huge amounts of urine that they're passing, but it's like you say George, they're no longer polyuretic because they pass that point. Certainly, and I can remember the last lady that I saw in A&E with DKA who we actually did bring up to the unit even after four litres of IV fluids, she still hadn't made any urine. So assessing disability, many patients with DKA can be alert enough to give you a history, even if that just means that they're responsive to voice. But a reduced level of consciousness is often seen. I think it's really useful to get an accurate GCS here because the changing GCS can help us track a response not only to the treatment, but also as an effective treatment. We can see severe electrolyte disturbance and rapid fluid resuscitation can cause cerebral edema. I won't touch on that too much now because Josh and I know you're going to pick that up shortly. I personally think getting an accurate GCS score really early on is really helpful in patients with DKA because a changing GCS can help us track their response not only to the treatment that we're giving, but as a consequence of the treatment that we're giving. As we resuscitate these patients, we can see severe electrolyte disturbances and we can also see consequence of rapid fluid resuscitation. We can cause cerebral edema from that. And actually in children and adolescents, cerebral edema is one of the biggest forms of morbidity from DKA. If the level of consciousness is severely impaired in a patient with DKA, it's definitely important to involve critical care at that point. When we're assessing patients' disability, obviously, but that's when we're going to check glucose and repeatedly now, or more often we see that ambulance trusts are also able to check ketones as well. I know Simon, you said the ambulance service in your local area can check ketones on the road. So say at this point, as Simon mentioned earlier, we do see patients with you glycemic DKA. So if you've checked your glucose and it's normal, if the history and the story fits and they're diabetic, I certainly wouldn't be afraid to check a ketones anyway, especially if they're taking an SGL to inhibitor, such as dapagloflosen. So once you've assessed those features, it's really important to assess everything else. Again, a really thorough clinical examination is so important here. For example, are they febrile? Are they jaundiced? And do they complain of pain? Coming back to the abdominal pain that Simon mentioned earlier. Just like nausea and vomiting, abdominal pain is a really common feature in DKA. But it's really important not just to include abdominal pain as a symptom of DKA and to put it down to them having DKA. As a reminder, we may still be trying to establish what the cause of the DKA was. And illnesses such as appendicitis, colisistitis or bowel obstruction can all lead to DKA. So that's probably the crux of assessing a patient with DKA. And now why don't we throw that into the pre-hospital management? Pre-hospitality, in general, the goal of care when managing a DKA patient. Obviously you've talked about any AB or C requirements at George, but generally our goal of care is all around rehydration, electoral-like correction and preparing them for insulin therapy for glycemic control to stop that ketone production. That will obviously take place in the ED department. But yeah, our focus pre-hospitality, good ABC care and then appropriate rehydration with fluids. And I've said appropriate rehydration there because depending on the circulatory status for patients, will depend on how rapidly we're going to rehydrate these patients. So fluids are really important for rehydrating the cells, as we've talked about these patients are really, really dehydrated. Potentially to the point that they are on the border of circulatory collapse with reduced blood pressure and impaired tissue perfusion. Insulin therapy will also cause further dehydration. So that's why you see in every guideline, do not start insulin therapy until you've given at least I think it's like a liter of fluids until you've started this rehydration therapy because yeah, the treatment in one of the better words for DKA will further potentially rehydration. So pre-hospitality is really helpful if we can get this started in good time and start this, start this while the patient's in the R care, even if their blood pressure isn't through the floor. Fluids additionally help to dilute ketones in the blood, reducing the relative acidemia and improving perfusion through the kidneys, improving that process of ketone clearance, like we've already discussed is important. We look at what guidance is out there to direct after the rehydration. Well, if the patient's clinically shocked, we're going to give them a large fluid bolus, so for adults, this is going to be 500 ml over 15 minutes and then repeated until that blood pressure is somewhat stabilized and we're then going to start rehydrating them to rehydration amounts. So that would be about a liter over 60 minutes. I think most ambulance trusts carry bags of 500, so as long as you're dribbling a bag of 500 through over about half an hour time period, then you're hitting the correct rehydration target. So just to say that again in adults, we're going to give them a 500 ml bolus over 15 minutes if they are hypertensive and clinically shocked and then we're going to start them on a rehydration regime which is about targeting 500 ml over about 30 minutes and I can't imagine in the UK most people are taking longer than an hour to convey them to their nearest Nazi department. So for kids, we're going to be a bit more cautious with that and that's for the reasons that you've alluded to George which is cerebral edema. So it is a rare one but it's one of the most common causes of death and significant injury in children presenting with DKA. So about in some studies, the size nearly 1% of children presenting with DKA will go on to develop clinically significant cerebral injury and exactly why this occurs isn't so clear but it's believed to be associated with over-resuscitation and the speed of fluid re-resuscitation. Now I take that with a little bit of a pinch of salt because there's also some studies out there that have looked at rapid versus slower rehydration regimes in children that didn't have a clinically significant difference in cerebral injury but all of the guidance out there is designed to give. All of the guidance I've seen out there treats children under a slightly slower fluid re-resuscitation regime because of this risk of cerebral edema and cerebral injury. But it's important to caveat that with the fact that if children are volume-deplete and shocked then it's still really important that we provide appropriate rehydration and intravenous fluid. So if we think about who's at risk for cerebral injury, well it's people with new new onset diabetes who've had a longer duration of symptoms as George said children and adolescents but particularly children under five are at a higher risk of developing cerebral edema. If we provided the insulin therapy prior to appropriate rehydration that's an increased risk factor and as I've already said a rapid administration of fluids can increase that risk. I think that's a really good point because actually if you look at the best pair the pediatric guidelines compared to adults in ED we will generally fluid resuscitate adults is the first step but we also pretty much give insulin therapy at the same time whereas in children we actually start with the fluid therapy and then start the insulin therapy slightly later. It is interesting how much more sort of cautious everything is when it comes to children in fact don't do your Canada George but working in Canada now getting my head round a new set of guidelines pre-hospitalia they don't give fluids to children in DKA unless they are shocked so they don't provide pre-hospital fluid rehydration here unless they are hope-otentive and I suspect the reasoning behind that is to just facilitate a slower more controlled rehydration in ED but that's not necessarily the case with UK guidance so J.R. Calc would suggest children and young people presenting with DKA may present with significant dehydration. If they are shocked we should give an initial bolus of 10 mils per kilogram over that 15 minute timeline again and if we need to give further fluid rehydration past that they suggest discussing with a senior clinician so as you suggested George if you've got someone who's really really sick with DKA particularly sick kids it would be good to get critical care out to see them and support with support with that care. If the patient isn't shocked then that 10 mil per kilogram bolus should be given over 30 minutes and again that is probably a reasonable conveyance time for most people working in the UK. I'll come on to talk about some of the warning signs for cerebral injury in a second or cerebral adema in a second but I just think we can't emphasise this point enough we should be rehydrating these patients at an appropriate level for their circulatory status. I think the big risk is if we particularly in children is if we've got a very high potential patient and the temptation is to just bang fluids in like we've all been in that situation where you've got a poorly person in front of you you may be a bit max it's just you in the back and you are used to putting a large amount of fluids into adults. The temptation might be to just bang a load of fluids in get the blood pressure up and then lose track of the amount of fluid and the speed at which you're rehydrating these patients. Like I say that there's varying an evidence on whether infusion speed really does affect cerebral adema rates but the the current working thought and in every guideline around children is that the way that we might be able to mitigate this is with appropriately tailored fluid therapy. So you need to decide do they fall into the slow rehydration format or the resuscitate there the shock to patients and then start them on this slower rehydration therapy and you need to take account of the amount of fluid that you've put into them to get the blood pressure up when you're calculating your rehydration fluids. Just before we go on to talk about ED and ITU management then let's have a quick think about the warning signs for cerebral adema and cerebral injury. This comes back to your point George of why you absolutely should have a cast iron GCS calculated and one that you trust because that is going to be our way to really note whether or not we need to slow down our fluid resuscitation or take a different tack. So altered level of consciousness can be common in DKA but if their level of consciousness is getting worse or particularly has altered after an initial improvement to our therapy that's something that might be a warning sign for cerebral adema. Severe headache or worsening headache after the commencement of our fluid resuscitation is something to take note of in children particularly irritability or significant vomiting might be signs of that. And again this is all hard to determine isn't it because there's so much crossover with a lot of the symptoms that we've already discussed but our main ones are going to be altered level of consciousness, severe worsening headache and then our cushing's triad signs. So if our patient is starting to get hyper-tensive, bradycardic and have respiratory depression then again that might be signs that we're developing cerebral adema and we need to reduce our fluid resuscitation or even stop it if the patient's no longer shocked. So let's go on to talk about how ED management and ITU management might differ. So Simon we've brought this patient in, they've started pre-hospital fluid. It's not a huge much else that we're able to do past good ABC care. What are they going to be having done to them in the ED department and then George what kind of patients go up to ITU and what treatment they receive there? So I think this is a fantastic example of good continuation of care because we're going to carry on all the good things that have been started pre-hospital. We're going to obviously reassess that patient with an A to A approach, manage any kind of persistent shock that's still present on arrival. We're going to carry on with those fluids that have been started. Now depending on how much fluid's been given pre-hospital I'm probably going to try and get in a non-shock patient, a good leader into the patient. And then it's important that we start looking at blood gases and we start the patient on some insulin therapy. So we normally give a certain dose of Ac trapid which is the most common insulin that we've ministered for this alongside that fluid resuscitation. So once the patient's got some insulin going in, they've got some fluids going in, we then need to relook at our blood gas and have a look at our potassium levels. So as we manage DKA and we give insulin this is going to shift potassium. So we may end up with a hyperkalemic patient so we're going to need to do some potassium replacement potentially. So depending on that potassium level and how it comes back over the next couple of hours is where we're going to start to add, you know, carry on with just normal fluids or where we're going to give fluids with some potassium added in. And like I said, as long as the patient's not shocked, obviously if they're shocked we're going to carry on regressively resuscitating them. But if they're not shocked then this is just going to be a slow process. And then we kind of start to overlap with critical care and things. So we need to look at those patients that are going to likely need our critical care colleagues and intensive or at least high dependency management. So some of that might be done in the ED but obviously, you know, a lot of it depends on what patients need to go there. So George, if we carry on that transition into ITU management and where we overlap in kind of our ED care into critical care, DKA is a very common presentation that we'll get a referral for to the emergency department. And actually as you were saying before Simon, this is a really nice way of seeing the continuity of care. If really good pre hospital management has happened, especially with fluid resuscitation and early treatment in the emergency department has occurred, after three or four hours, this person may very well have turned around and never needed to come to critical care. However, with that said, I do think critical care and at least a high dependency area is really good for most of these patients because we can put in lines like arterial lines or central lines and we can check their blood regularly without them needing to have multiple stabs with needles. We can monitor their heart rate and cardiac rhythm. They get better nursing ratios, etc. The patients we most commonly will get referrals for or the ones we certainly should be more aware of and should see are young adults, especially those between 18 and 25. They count still within the adolescent period and are more at risk of cerebral edema. The elderly, the pregnant, those with significant comorbidities, if the starting GCS is already less than 12, we should definitely be seeing those. Those with key tones that are higher than six, a bicarb that's less than five, a pH that's less than seven, these are all just markers of how severe the illness is, not really any value in remembering what these numbers are. Yeah, go ahead, Josh. So it was just a question around the GCS. So probably most people were we familiar with the phrase that is probably still true but going out of fashion, which is GCS less than eight, intubate. And we've done fear podcasts in the past. What is the threshold for intubating these patients or will they ever be intubated? Is that something we should be considering pre-hospitalia or is it really based on their response to treatment? Thanks for bringing that up. That's a really good question. And obviously the old, the old adage exists for a reason, but actually patients with DKA, we very rarely end up intubating and ventilating them. They're already doing such a good job themselves of controlling their pH to the best of their ability from a respiratory point of view. of you that if we put them off to sleep and switch them to a ventilator, we are really going to struggle to achieve the same CO2 level that they're already achieving without seriously overventilating them. So we risk calming them actually by putting them off to sleep, even from that period of induction. There are times where we do have to inch-bait and ventilate patients with DKA and that's usually either because they have got so tired that they're no longer controlling their acidosis even partially with their respiratory compensation or if they need an operation. So sometimes patients with DKA need an operation and that's the reason for their illness, for example, about perforation. But we will try for as long as possible to correct as much of that acidosis as possible before we put them off to sleep. There's something that we consider and I'm seeing ED for RSI. It's people with that level of metabolic acidosis. As you rightly said, George, they are compensating with that attack at near those Kushmau breathing we were talking about earlier and they've got that respiratory compensation and even knocking that out for a few moments in order to RSI them, that drop can cause significant cardiovascular collapse and you might end up with a really, really downward trending RSI into significant problems really quickly. And the only thing that I've kind of been looking at is if a patient does need that for an absolute emergency as you've kind of alluded to the reasons we might need to do that. There is some good evidence out there about ventilator assisted preoxygenation to try and really hyperventilate that patient in order to kind of maximize that before we knock off their respiratory drive with a kind of muscle relaxant before we induct. So I think it's just worth consideration. So if we can avoid it, avoid it. And if we can't then try and find strategies to really make sure that we optimize physiology because this is going to be a really physiologically difficult RSI. Yeah, exactly. And we'll avoid it as best as possible. Great. So let's say this patient has actually come up to intensive care. What are we going to do on intensive care? Well, we're just going to carry on doing exactly what's been done in the emergency department already. We're going to continue this fix rate insulin infusion that's intravenously. We'll leave that running and we will keep correcting the fluid and potassium levels until the decay has resolved. We used to use variable rate insulin infusions where we would change the rate of the insulin going in based on what the ketones and glucose were doing. But now actually we don't do that. The evidence space has changed and we'll initially start people on a 0.1 units per kilo per hour per hour insulin infusion. We will continue that to the decay resolves. In critical care, sometimes we change that about a little bit. So if it's not resolving as quickly as we'd like, or if the glucose is falling too rapidly, we might change it a little bit, which I guess kind of argues against me saying it's a fixed rate, but that's quite rare. And certainly if a patient goes to come in under general medicine to acute medical unit, they'll generally stay on that fixed rate the whole time. Whilst these patients are having their insulin infusions, the potassium level can drop quite rapidly. As Simon mentioned earlier, we get this profound hypocholemia. So we replace potassium as we go along. Patients with DKA can be up to 800 millimoles of potassium deficient, which is significant depletion. So sometimes we need to put a central line in and just replace that potassium more aggressively than you can do through a cannula. But generally it's pretty easily corrected once this therapy has started, and most DKA cases should have resolved within 24 hours. Now the underlying cause of the DKA may not have, for example, if they've got colysis titus pneumonia or had an MI. I don't know about you guys, but I think one of the things I actually find most satisfying about DKA is that they actually resolve quite promptly once the initial treatment has been started and is continued. And most DKAs will resolve within 24 hours. In fact, the guidelines do even say if the DKA has not resolved within 24 hours, you should definitely be seeking expert help, i.e. from an endocrinologist. So yeah, the DKA itself may well have resolved within 24 hours, but the underlying cause will still need addressing and often still need streetings. So these patients may stay in hospital for several days whilst they're treated for their pylonofritus, their MI, their bowel obstruction, etc. So DKA really is only half of the picture. And I really hope that we've managed to hammer home the point today that the DKA can really sometimes be a very nasty symptom of an underlying illness. So how do these patients do? Well, once DKA is a life threatening condition that can prove fatal, the outcomes in Western countries and in developed countries is actually very good with a less than 1% mortality. The outcomes are worse if you're elderly and also slightly worse if you are from the young and adolescent population. And that's usually because of morbidity, specifically cerebral oedema. And for those that die in hospital from a presentation that involved DKA, generally for those patients, the cause of death is not from the DKA, but from the cause of DKA. George, I did have one question for you. So it's mainly about the use of sodium bicarb. Obviously, we've got a metabolic acidosis at LAPH. You could logically think that giving some bicarb and it is going to help correct that pH faster. Now, personally, I'm a bit anti-bicarb. I know it's not in a lot of the recessed guidelines, but what about in ITU and critical care? Do people use it? Is it something that's in standard guidelines or standard practices? Does it fit into the treatment of this? Yeah, that's a really good question, actually. So in some causes of metabolic acidosis, we do use sodium bicarb. The bicarb works for those who aren't aware, sodium bicarb acts as a buffer. It basically accepts those hydrogen ions so that we start to neutralize that pH. Now, actually, we don't really use bicarb, sodium bicarb. As often as people might think, and b, in DKA, the acidosis in DKA is caused by the ketones, the ketoecasses, really, rather than by how we might classically see metabolic acidosis, such as in tissue ischemia or profound hypervelemia. That's why you sometimes see these patients with DKA with a far worse pH than you can fathom them being able to survive with. I'm sure Simon, you've seen patients in recess with worse pHs with DKA than those who have come in in cardiac arrest. Oh, yeah, yeah. Absolutely. Sometimes I look at them and go, "How is this person functioning?" And we're talking high six points, not even sevens. Yeah, exactly. That's because the cause of this acidosis is from something different, rather than from the bigger picture of this systemic hypo perfusion that's causing a profound acidosis. We don't really use bicarb in DKA because the cause of the acidosis is not profoundly harmful and actually they will recover as you treat the case. That said, there are some intensivists who like to use it, particularly if you've had to put them on to filtration, etc. It's like you say, Simon, it's slightly touchy subject. Some people love it. Some people don't, but it doesn't really feature in the guidelines for that reason. The Mar-Mite of the Resource World. Cool. Yeah, I appreciate that because I don't use it, but I was just wondering whether it's always good to just question some of those, and then you can reflect because if other people use it all the time, maybe I need to look at my own practice and change. Okay, so let's summarize. So to wrap up in DKA, a lack of insulin leads to high blood sugar, an acid buildup from fat breakdown resulting in a metabolic acidosis. This triggers dehydration and electrolyte loss through osmotic dioresis, with key signs, including excessive thirst, frequent urination, and ketones in the blood and urine. There's a number of causes of DKA that can precipitate the condition. All of these are generally stresses on the body. So this can be infection, it can be surgery, recent physical injury, or emotional trauma. Myocardial infarction is definitely something that we need to have in the back of our mind because as we've discussed, this can present differently in diabetics. It can be from either accidental or deliberate treatment non-compliance, such as expired or improperly stored insulin. It can be the first presentation to us of an undiagnosed diabetic. And for some cases, there's just no known precipitating event. We need to keep in mind when we're taking a history, as we've discussed that we shouldn't assume all symptoms that we see are as a result of DKA. Medical pathology can present that has patentiated the DKA, so the DKA might be secondary to a medical complaint. Abdominal pain, for example, could be from the arsidosis, or it could be appendicitis, bowel obstruction, or colysis thytis. Managed then in general the goal of care in DKA is rehydration, electrolyte correction and insulin therapy for glycemic control and preventing further ketone production. And our focus pre-hospitali is good ABC care and rehydration with fluids. Now this rehydration needs to be appropriate to the clinical situation. If the patients in circulatory collapse we need to rehydrate them slightly faster. However, if the patients maintaining their blood pressure that doesn't mean that we don't need to give them fluids we should still provide some fluid rehydration to dilute those ketones and prepare the patient for an insulin infusion in ED, which without rehydration at long side will result in the patient dehydrating further. And we need to be careful not to overresuscitate these patients as particularly younger patients and those with the risk factors we discussed earlier may be susceptible to the development of cerebral edema and devastating cerebral injury. That's all for this month. Thank you again to our supporters on coffee.com and of course our sponsors Geeky Medics who all helped to keep this podcast free and open access. As always you can find the article with references used in the podcast and links to further learning. But thanks for listening and join us again next month. [BLANK_AUDIO]