And, welcome everyone to this, discussion on new monitoring options in canine endocrinology. I, I really see this as, as the road ahead. This is the West Highland Way in Scotland, a long road that I, I was, happy enough to, walk this summer, and This is really looking at the road ahead.
Some of the things that we're going to be talking here today are not mainstream. They may become mainstream, and I hope at the end of this talk that, some of you will, will perhaps, think about these in, in your work. it's just, it's really great for me to be talking in this, webinar today because, the, for some reason, yes.
Why, why, why webinars? Well, at the moment in Scotland, we're, we're in the grip of, a good Scottish winter. And while this may be great for the dogs, I can't get out my driveway.
It is a sheet ice down my driveway, and, if we had to do this as a, as a live CPD event, as it were, in, in a, in a, in a typical way, you wouldn't have a speaker here today because I couldn't get out. So, a few disclosures before I start. I think it's important that we, we do this.
I am a consultant, for, for DeRA. I have a clinical trials sponsored by them. I've spoken for them.
I, I also have some, equipment and tests that have been supplied to me by, by Siemens, and that will come up during the course of this, conversation that we're going to have. So, I thought I had to declare that as well. It's also was, was worthwhile pointing out that there are some unlabeled and unapproved uses, in different parts of the world.
this, this may be important, it may not be important to where you are. So just with that and things, let's move on. I'd like to start with, an introduction, looking at the importance of interpreting test results in their context, improving the accuracy of clinical histories.
I'd then like to go and talk about three new monitoring options pre-veteral cortisol, haemoglobin A1C, and continuous glucose monitoring. And for each of these three, I would like to talk about why we change, why we should change, why we should consider changing, how you should perform them, and how to interpret them, and I'll come back at the end and with a, with a wee summary. So what I'd like to do, looking at those three now, is, is to ask you all a question just to know where, where we're coming from here.
And that question is which of the following do you currently use? I sometime in the last year, say, you've used, pre-veteral cortisol, haemoglobin. A1C or continuous glucose monitoring.
You can vote for 12 or 3 of these depending on what you're doing and hopefully James will then fire, fire back and tell me, what, what the results are. So James, if you could launch the poll, please. Yeah, the poll is launched.
People are voting, there's no need to be shy anyone because, we can, we can't see who's answering these. We won't shame you. Yes.
And, of course, if, if everyone answers all three, they're all using all three, then we can end the talk really quickly. We can just, I was gonna say go home, but we're at home. Yeah, we are at home.
I can't go anywhere but home at the moment. OK, I think Most people have voted now. I'm gonna stop it there, because we've got quite a clear, clear response which is that, most people are not using any of those, so I think your talk is going to be very useful, but, quite a few people are using pre-veterol cortisol, and so about 40% of people said they're using pre-veterol cortisol, 40%, did you say?
40 40% of, yeah. 40, wow, OK. Half said that they're not doing any of these tests at the moment.
So I'm sure your talk will be very useful and not very many people are using the haemoglobin, A1C or the continuous glucose monitoring. So again, that'll be very helpful. OK then, thank you very much.
That's, that's useful information. So I, I, I am, 40% is, is, really high. I was, I was not expecting that high.
I, I will confess, I'm pleased to hear it's that high, but, so, that's good. OK then, let's just talk about a little introduction to endocrine monitoring. The truth is that much of what Peter talked about in his talk, about all of what Peter talked about in his talk, was about diagnosis.
But the truth is that once I've diagnosed an endocrine disease, I very rarely managed to cure it. I can control it. I can manage it, but I can't cure it.
And as a result, I spend far more time in consultation monitoring endocrine disease that I'm treating than I do in the actual diagnosis. So from an economic point of view, From a veterinary involvement point of view, monitoring outweighs diagnosis in some respects. But if you look in the scientific literature, it's all about diagnosis.
It's about the difficulties, the sensitivities and species, all that stuff that, that Peter talked about, which was, was just a great talk. I, I loved it. but it's mainly about the diagnosis, about that sort of thing.
And Even when they talk about monitoring, they talk more about tests than they do about clinical presentations during those monitoring. And, a lot of the, literature, is concerned with using the tests that are useful in diagnosis without often necessarily, using those, being shown to be useful in monitoring. So for example, we, we do know that for example, the low dose dexamethasone suppression test in Cushing's disease is a diagnostic test, but as a monitoring test, it's, it's pretty useless and nobody, not many people would use a low dose dexamethasone suppression test as monitoring, and I wouldn't encourage that anyway.
Whereas fructosamine, at least in canine diabetes, and emphasising canine diabetes, fructosamine is not really a diagnostic test, but it is a good or might be a good monitoring test or it might be used as a monitoring test. It might change that during the course of this talk, but, we're quite happy with that. But there are some tests and I would flag up the ACTH stimulation test as one, which are Demonstrated to be effective in diagnosis, but, the demonstration for them being useful in monitoring, is, is not as good.
And of course, when we look at the costs of treatment, the costs of the monitoring of the treatment can be a significant part of the costs of the overall treatment. And, if we're going to encourage owners to, treat their pets, and, and after all, we are all in the business ultimately of trying to make these animals better, then the costs of monitoring have to be considered and making that monitoring effective, but still cost efficient, is going to be important. So that comes down to the next question.
What, what do vets want out of a consultation? I mean, mostly what we're doing, I think, is checking for signs of an underdose or an overdose. We, we're, we're trying to better control the animal and likelihood is that we're going to either change the treatment, give more treatment, perhaps give less treatment, but that's what when we walk into a consultation, that's what we want out of it.
What do owners want out of the consultation? I think a lot of owners actually just want reassurance that they're doing the right thing, that the dog is well. They want advice on other aspects of their dog's life and not on the necessarily on tweaking the treatment that much.
I'm constantly amazed about how badly controlled some diabetics are from my perspective, but the owners are very happy with them. They, they, they're just better than they were when they were diabetic. They, they, they, they're still not good from my point of view, but the owners, because they feel the dogs are better than they were when they were overtly diabetic, they're actually happier than perhaps they should be, and that comes down to this idea of reassurance.
And I guess this tension between vets and owners in this and things about what we want out of the consultation really can be crystallised into into this question is what is a well-controlled diabetic? What do we mean by a well-controlled or a well controlled Cushing's dog or a well-controlled anything really? Is it a fructosamine concentration, haemoglobin concentration, an animal that is not having hypos, an animal that's not PUPD, a dog with a blood glucose curve.
And I think we need to answer that question for ourselves and for our owners, and the closer that we agree with the owners on what, and that includes both moving the owners to our position as well as us moving to the owner's position, as it were, the closer that we agree what is a well controlled diabetic, the better the chances are that the owner will be satisfied with monitoring. If we have diametrically opposed views that actually all the owner wants to do is to come back into your practise for needles and syringes. But we're trying to establish a blood glucose curve that looks whatever you like it, like, then we're going to have a tension there.
And I think that tension needs to be resolved as soon as possible so that we may not have to have a brilliant blood glucose curve, but also the owners appreciate that just coming back into insulin practise for insulin and needles is probably quite a good idea. It's important that we stop constantly trying to improve these animals as well. I think I see a lot of attempts by people always they will try to change something.
They will always try to improve the diabetic. Diabetics are never going to be perfectly controlled. It's doubtful such a state exists.
And if so, if you can't achieve perfection, when is good enough? And that's a discussion that one needs to have with owners. But as I say, there's very, very few papers looking at, at this sort of question.
One of the best papers that's out there, and I would encourage, everyone, if you, if you read no scientific paper this year, read this one, please. And, and, and it's one of the, the, a, a real sort of, watershed moment is actually rather than taking some, some biochemistry test or some clinical, parameter to measure, these, researchers, looked at, the history and physical exam findings, and it, and it's so often, and important. And what they did was they took 25 dogs that they felt were well controlled and 28 dogs that were poorly controlled.
And they did a whole load of measurements pre-insulin glucose, glucose measurements through the day, which then took a mean off. They didn't look at hypos or hypos, but just the mean, the fructosamine, and this concept of glycosylated haemoglobin and we'll come back to that. They were able to demonstrate that the blood glucose concentration was different between the two groups.
It's significantly different. But there was also overlaps. There were some dogs here that had a, a poor, If we can just take a wee pointer option here.
so there were some dogs here that, that were apparently. Poorly controlled, but there they are sitting down with a mean blood glucose concentration and equally well there were other dogs here that were . Well controlled But apparently had a high glucose concentration.
And the same thing went for other conditions such as the fructosamine and the glycosylated haemoglobin and so forth. And if you looked at these, the glycosylated haemoglobin here, again, significant overlaps, even more so when you start looking at this, the, the fructosamine concentration, look at the overlaps there. Quite, quite significant, overlaps there.
And, so their, their, their conclusion, and quite rightly was that, when they looked at it, glucose measurements and glycated proteins were correct, ie correctly. Predicted 60% of well-controlled dogs, but only 39% of poorly controlled dogs. And what they concluded was that the reliance on history and physical exam findings and changes in body weight are effective for initially accessing control of glycemia.
And that I think is an important paper to emphasise that these tests must always be interpreted in the light of the history and the clinical exam findings, and that of course assumes that the tests that you are measuring are accurate. Many people will use glucometers to measure the blood glucose, and yet if you compare a glucometer to a reference blood glucose, the results can be quite different. Sometimes, sometimes they're quite, quite reliable.
But, if you actually compare using one of these bland Altman plots, which is, so what you expect is these, these, dots here would all fall on this line here. That would be nice. That would be a perfect, perfectly, agree, agreeable test.
Or if there was a bias that they would all form on another horizontal line. And yet, you can see here that, some, some, sometimes the glucometer is reading significantly less than the reference method. Sometimes it's reading significantly more than the method.
And given how much these can affect your decision making, the reliability of these tests needs to be considered. I put two glucometers here, 4.7, 5.8.
That's the same blood sample. Yes. So it's difficult to just look at these and say, oh, yes, that we know what we're, we're doing.
In your mind, therefore, this clinical signs must always outweigh the results. And I'm going to quote someone who has, is probably listening at the moment, and that's Peter Graham, who once stood up and said, in the battle of vets' hearts, the numbers so often win and the animal so often loses, and I absolutely support that. It, it is really important.
That we consider those signs first and then look at the results. Can we just use clinical signs? Well, I think if you're talking to specialist endocrinologists, probably the answer is quite a lot of them do place great emphasis on the clinical signs and, and are prepared to ignore laboratory tests.
But I also appreciate that there are colleagues of mine in practise who, who do things like anaesthetize animals and do surgery on them and do go out and see cows and horses and sheep and so forth. Therefore, their familiarity with the clinical signs, their subtleties, and their ability to spend a long time with their clients may be somewhat limited. And so I think it depends on your experience and your case load and your type of practise, about how much you can trust the clinical signs.
Furthermore, even in, even among specialists, using clinical signs does, I'm afraid, rely on owners and, and that, of course, is very variable between, between owners. Clinical signs, for the most part, do not give us early warnings. So subtle changes in cortisol concentrations or glucose concentrations that start to go below the cutoff, below or whatever, don't give us an early warning that we're about to tip over the edge.
And so by the time we see a dog with hypoglycemia or ketoacidosis or hypoadrenal corticism, we, we've really gone too far and we don't want that. So, I think that's, that's, an important factor to consider as well. And lastly, whether you're using, Trista or, thyroxine or, insulin, there are no indications for endocrine, increments of changes in dose based on clinical signs alone.
And so I think, having some sort of numeric. Value in there can help with that, increment. But, the, the, the, I'll come back to the, the reliance on the owner.
And if we look at the reliance on, on the owner, then we have to think, well, how do we get that history? Now, when we, when we, when we see an animal, the, the clinical history recorded in our records is filtered by, goes through the prism of the owner's perceptions. If the owner, If for example, a diabetic herself, then that, that, that may, may influence how she sees the animal.
If the owner has, a busy family life and, you know, the dog is just there in, in the kitchen, and, unless the dog is urinating on the kitchen floor, the owner hardly will hardly notice what how the dog is doing. That is going to influence those perceptions. And so I think in order to get a good clinical records, we need to think how we can get the owners to try to be as less an influence on that clinical record, less of a, of a prison, and try to get the owners to help us more, and getting the owners to help us more to get That accurate history really relies on a more consistent picture over a period of time that we can build up.
If we ask an owner how a dog is doing when they come to see us, then the owner usually will tell us about the last week. If we push it and ask them how are they doing the week before, they may remember. If we ask ourselves, what were we doing 29 days ago?
Can anyone remember what you were doing 29 days ago? And yet, what was happening 2 and 3 months ago can be as important as what's happening right now. We're not, when we stabilise a dog, with its endocrine problem, be asking for Every day to be perfect, we're asking for an average stability without any crises.
And we need to know if this animal has had minor crises in the past, which maybe the owner forgets to tell us about. And therefore, with any endocrine disease, I think it's important that owners keep home records. Diabetics, Cushing's, hypothyroidism, Addison's disease, have the owner keep her home records.
Have the owner keep their home records about all the things that are important, the thirst, the appetite, the drinking, the urinating, the activity levels, if the dog's happy or sad, if the dog's exercising well or not, the more we can get owners to do with this, the more trusting we are of the owner's history. And so we are really keen on owners keeping home records, and it's something that I'm pushing very heavily in my own clinic, and I would encourage you to do the same. Commercial companies will support you with this, the companies that make insulin and the companies that, that, that make other endocrine drugs, they will support you in this with, with documentation to give your owners, but it's the best.
I am getting now something like, an email a week, from one owner, very intense person, but they're sending me Excel spreadsheets of how these owners, dogs are, are doing. I try to get it to be less frequent than that, but I think it's really useful over time to have these home records so that you can see how they're doing. If owners are, you find it difficult to get owners to, to keep home records or they're not keeping home records for whatever reason, then the other thing that, that can help quite a, quite a lot is to use, owner questionnaires.
So these are, these are designed for, dogs coming into the clinic. They're given the questionnaire before they meet you. So this is not something that I would ask owners to particularly fill out in the consulting room, but rather to fill out beforehand.
So we arm our receptionists with these owner questionnaires, and that way we make sure that all the necessary questions are asked. So about appetite and thirst and so forth, and it's surprising how much these questionnaires are consistent between endocrine cases. We make sure that every owner is asked the same way.
So they have the questionnaire, they look through it, they've got time to fill it in, and they have time to think about it. And this actually, although it may seem like a lot of effort, saves time in consultations. So if you're one of those practises that has a relatively short consultation, maybe only a, a, a, a 7 minute consultation, then being able to get the most out of that consultation.
A little bit of preparation beforehand, and this is something you could put on your, your website for your owners to download. You don't have to give it to each of them. They could come with a questionnaire already filled in, but it helps to get the information into the clinical record to make sure that we can trust those signs more and perhaps spend less time worrying about very precise figures.
So that's just a little introduction into, into how we should be thinking about endocrine monitoring and how to improve our clinical history, histories. So make sure that we're more, more accurate because in these dogs with, Cushing's and diabetes and so forth, it is the history rather than the physical examination that is really going to weigh heaviest on our minds. So let's talk about pre-veteral cortisol monitoring.
Why, why you might want to change and, and if you do want to change how you might perform and interpret it. For the 40% of you are already using it, this is, this is something that may be a little bit old happen to you now, but, for the others, this may be new. So, before 2013, veteral triloine was monitored using a combination of clinical signs and ACTH stimulation tests.
Polyuria polydipsia should expect to resolve within two weeks of getting the right dose, aloplecia within a month. That's my expectation. And if the animals show signs of polyphagia or lethargy, alopecia, panting, what, whatever, then that is an unstable, Cushing's dog.
It's not getting the full benefit of, of the treatment. I do not advocate, routine biochemistry unless the dog is ill. And if you do do routine biochemistry, do not expect the alkaline phosphatase to become normal.
They only sometimes become normal. Many dogs, despite being well controlled on their veterol, do not become, do not develop a normal alkaline phosphatase. I don't know why, but, but they don't.
So, ACTH stimulation tests, we, we, we would use, before 2013, at 2 weeks, 1 month, and every 3 to 4 months. And it was generally accepted the target range was about, a post ACTH cortisol concentration. Of about 40 to 120 animals per litre, and we would adjust the dose to try to achieve that.
And I often came across dogs where that was done despite the clinical signs not agreeing with the ACTH stimulation test. In autumn 2013 in Europe, the ACTH, which until then had been very cheap and we'd used widely, and was really almost the standard of practise, and no one argued really. That ACTH preparation suddenly became hard to get hold of.
It went off the market and, and what little supplies we could get. were extremely expensive. And this had obvious implications for the diagnosis of Addison's disease because there really isn't a good alternative yet to an ACTH stimulation test.
It had some implications for the diagnosis of Cushing's disease and, for those people who used ACTH stimulation tests, as part of their diagnostic investigations, that was obviously an issue, but, there were alternatives and Peter's covered, some of the issues around some of the alternatives. But when I looked at it, at least at Glasgow vet school, 9 bottles out of 10 of synaxin were actually being used for the monitoring of Cushing's disease. And of course, therefore, if we could avoid the use of ACTH stimulation tests in Cushing's disease monitoring, that's so much the better.
Of course, for some people in some parts of the world, they've been spending a lot of Money on ACTH for a long time and therefore, for many people, the monitoring of veal using a a a bottle of ACH was always an expensive and difficult option. And as a result, monitoring was somewhat difficult. But of course, even in 2013, we already knew.
That ACTH stimulation tests were really rather variable in these dogs being treated with veterol. It would happen, that if you, did a test at 4 hours and then repeated the test at 24 hours, you would get very different results, in both controlled, in the thin line here, and in uncontrolled dogs, with, with Cushing's disease. And furthermore, shortly after this, this problem happened in 2013, Rebecca Hesse's group published a paper which showed that dogs that some dogs at least, which had a ACTH stimulation test that was very low.
I Addison's. But were clinically well, that if you checked their ACTH stimulation test a few hours later, it was back to normal. And indeed, they, they, and some might say this was quite brave of them.
They left those 13 well dogs with apparently Addisonian responses, and they found that after a period of greater than 88 days, 12 out of the 13 were still on the same dose without any problem. So the idea that the ACTH. Stimulation test, somehow acted as, as the, the, the most important determinant of stopping the veteral at 3 to 6 hours, also was, was being challenged here.
And, and I have plenty of other cases where, which would fit in with those 13 dogs, where the ACTH stimulation test, at, 3 to 6 hours was low, but in fact these dogs were doing fine on the dose, and we, we, we, we were aware in Glasgow of quite a number of dogs that have been taken off their veterol or the dose reduced considerably, and the clinical signs had come back when such results had been received. And so the dog was often getting veterol, being charged for it, but was getting not the benefit that we might expect. So with this, with this in mind, we decided to have a look at the ACTH stimulation test and see if there was some alternative.
And of course, if you're going to have an alternative, you need to have some sort of gold standard. And we took the gold standard of, of an owner's questionnaire. Now this was a questionnaire that we developed over, over a number of years with our patients before this.
And, using, a panel of, veterinary people, we decided to ask all the questions that you hopefully, you'd expect to ask on these, things, and we were able to derive a score. And, the, the score from that panel was then, a number of these, examples were put to the, the panel, and they were asked whether they would increase the dose or decrease the dose or keep it the same. And from that, we look back at the scores that we had achieved, and they were then dividing the cases into three groups.
So these are groups that on average, vets would or would not change the dose of, of Vetterol. So we had this questionnaire set up, it took us a long time to develop it, and we then looked at a group of dogs that had had ACTH stimulation tests and we compared their scores. With their post ACTH cortisol concentrations, that which we'd been using up until then.
To monitor dogs with veterol. And for all of you who are doing ACTH stimulation tests in dogs with, with Cushing's disease being treated with triosta, I would invite you to look at this graph very carefully and consider how you interpret your ACTH stimulation tests in those, in these animals. Indeed, if you take the sort of Published, opinions on how you should interpret those, tests, then, you know, the, and you place them in these green zones, these are the ones that are correctly identifying the right ACTH cortisol concentrations for the.
Right sort of clinical scores, you'll find about 28% are correct, meaning that the rest are incorrect. And some of these are, are very incorrect. Of course, you, you have animals here that are, for example, this one here has a clinical score that would suggest his dog's really well controlled, but has a high post ACTH cortisol concentration and, and there is a danger, of course, that a clinician might be overly influenced by the ACTH cortisol and may even consider increasing the dose.
more worryingly here, perhaps on, on this end here is a dog that is showing signs of, Cushing's, but the post ACTH cortisol concentration is very low. And so with this dog, the danger is that people reduce the dose of veterol if they look at the post ACTH concentration. And actually, the dog's already got clinical signs and, and they, they may even get worse.
The degree of, the degree of wrongness, if you like, the agreement within the group is, is really, quite, quite large, and we therefore have quite a problem with ACTH stimulation tests if we're going to use these, in our interpretation of how well stable these animals are. So what we did was instead of looking at the ACTH stimulation test, we took a sample of blood before the dog had its veterol. So this was either at 24 hours after the last dose or if the dog was on twice daily, 12 hours after the last dose.
We then gave the dog it's peter. And then 3 hours later, we did an ACTH stimulation test. And the first thing we did was then to look at exactly those, those clinical scores again and to look at the degree of agreement with the pre-veteral cortisol, but using cutoffs that were applied to the post ACTH stimulation.
And we found not 28%, but 47%. OK, still not good. But if you look at the spread, the spread is less.
And so the, the, the number of really wrong results is, is far, less. On the basis of this work, we went forward and last year, the year before last, now actually, Laura McFarlane, my resident, and Tim Parker and I published a paper in the veterinary record, looking at 110 of these tests now performed on 67 individual dogs. So some of these dogs got multiple tests.
Some of these dogs were unwell for some other reason, so they had pancreatitis or diabetes or whatever, and had at the time they had problems going on. As far as we know, none of those 17 dogs at the time were diagnosed with Addison's disease, i.e., their electrolytes were normal and they were There was clearly other reasons for that, that, that their illness.
These dogs have been diagnosed using, standard tests, and because, we had recruited from other practises, the Small Animal Medicine Society members in the UK were very generous in, in helping us with this, this work. These were, we, these were, 73 of these tests were, were done outside of Glasgow in first opinion practise. And this is important because, of course, one of the, things is that maybe dogs referred to a university context are not typical of the population, whereas, we were using dogs, with a lot of dogs from first opinion practise.
It's not common to treat dogs twice daily yet in the UK with veterols, so only 16 tests, in this population were done on twice daily dogs. Others in, in Zurich now have taken this forward and have done far more tests on twice daily dogs, and, we're therefore very confident that twice daily is, is, the same, gives you the same results as once daily. So it doesn't matter whether it's once or twice daily.
And again, if you, if you look at the adrenal dependent and pituitary dependent dogs there, only 4 were confirmed as adrenal. That has since been extended. So this works in adrenal and pituitary, dependent dogs.
This is, what's called an ROC curve, which compares the sensitivity and specificity that Peter was talking about earlier on at different cutoffs. If you take a diagnostic test, and you alter the threshold at which you make the diagnosis, then you will alter the sensitivity and you alter the specificity. And, if you take a test that gives you random results, let's say your, your test for whether a dog has Cushing's or does not, is to toss a coin.
If you toss a coin, you get the coin toss, then you will end up with a line, this line here, the, the horizontal, sorry, the, the diagonal line across here is what you'd expect with a coin toss. If you look at using an ACTH stimulation test to diagnose Addison's disease, which is probably a, one of the most specific and sensitive tests for, for, any endocrine condition, then you'd expect to find a, a curve that was somewhere up here. These three curves there are, are therefore not, not brilliant, any of them, but it's worthwhile noticing that the post ACTH stimulation test, here, throughout most of its range is performing, worse than the 3 hour, post-trialists.
So the pre pre-ACTH cortisol concentration. And throughout much of the range is also performing worse than the pre-veterol or pretrial is staying cortisol concentration. So the point of this curve is that it really doesn't matter where you draw the threshold that the pretrial is staying cortisol, the pre-veteral cortisol, nearly always performs better than the post ACTH stimulation test cortisol.
Using this, we can also, go forward and, identify then the, the, ability of the, Sorry, the, test, to, to come up with a range. And using this, we, we, we were able to work out the, the thresholds, at which, we, we would, confidently diagnose that animals above that threshold. Were to have a dose increase, were needing their clinical signs would indicate a dose increase, and the threshold for pre-veal cortisol was 138.
Below that threshold, that there, there were dogs that had clinical signs, and, some dogs that were well controlled. But above that threshold, we really only found dogs that were poorly controlled. So that's a, that's a safe threshold, if you like.
We didn't want to encourage people to change doses too quickly, so we set the threshold at a fairly high level. If you wanted to be more strict, you could bring that threshold down, but the risk of that is that you may start increasing the dose, and we didn't want to do that because we wanted clinical signs to determine more than anything else what was going on. And for those of you who are using ACTH stimulation tests, then if the pre-ACTH is above 62 or the post ACTH is above 130, then those were, were good indications that you should increase the dose or more accurately, they were good.
Of the fact that these dogs would have clinical signs of Cushing's disease. But as you'll see from those sensitivity and specificity figures, they're not that good at it. they're, they're less good than the pre-veteral cortisol at almost all points.
Of course, the more important aspect of monitoring is in detecting that overdosed animals. Underdosed animals, you could detect by clinical signs alone if you wished, but detecting the overtreated animals was more important. Now what, what this shows is that the pre-ACTH cortisol is particularly sensitive, too sensitive.
And so many of these dogs were diagnosed, if you just looked at the pre-ACTH as being overtreated. And I think that's most people's common impressions of the dogs that are getting this stimulation test done is that many pre-veteral cortisol many, sorry, many pre-ACTH cortisols are very low. So that's not, not really very useful.
We found 8 dogs which said the post ACTH cortisol was too low. These were all well dogs, and 6 dogs where the pre-veteral cortisol was too low. And of course, some dogs had both.
Most of those dogs had a dose reduction because vets, and this is a, this really shows that most vets reacted to that ACTH stimulation test and they lowered the dose. Even though the dogs are clinically well, they lowered the dose. Just two stayed with having a low pre-veteral cortisol, but these two had a post ACTH cortisol that was above that 1.4 mcg or 40 nan per mL, and the vets kept on going.
They ignored the pre-veteral cortisol. One dog, actually a week later, had the veter withdrawn and showed a marked clinical improvement. Even though the owner had said it was well, it actually got better in that the owner then said, actually, I didn't realise that my dog was sick.
It was lethargic, it wasn't doing well. And actually, when you took away the veterol, I realised how bad it had been. The second dog developed signs of Addison's disease 3 months later on the same dose.
So a pre-veteral cortisol of less than 40 was certainly able to detect hypocortisolism in those, those two dogs, where the post ACTH cortisol had missed it. We extended this, this work, and took a, a, a new study, forward with, the, groups working in Zurich, Bologna, and Utract. And these, were, this study was rather more focused and so we only had 35 dogs, but we've done 123 monitoring tests on those dogs.
This, this data is going to be published. it's only an abstract form at the moment, but it will be published, or submitted to publication, sorry, in due course. So, this is currently not in the literature, except as an abstract.
Of those dogs, we were able to identify 21 dogs that had 39 pairs of test results where they had the same daily dose of Vetterol, i.e., we had, a dog treated for 33 weeks at least.
Had a test and then had another test 3 or 4 weeks later on the same dose. So you should expect those two results to be broadly similar. OK, they should be have a low repeatability, a coefficient of variation should be very low.
and actually what we found was that the pre-ACTH and the post ACTH were very variable, that one week's results, were not the same as, dogs, done, 22 or 3 weeks later, suggesting that it wasn't particularly repeatable. And if we, look at the Results and we take a coefficient of variation of 25% or 0.25%, then you'll find that the pre-veteral samples were more often repeatable than the pre-ACTH or the post ACTH, suggesting that the pre-vegetable, as well as being a predictor of Addison's was also a more repeatable animal.
So at least at Glasgow vet school now, the ACTH stimulation test is no longer performed as a monitoring technique for, Cushing's dogs. We only measure a pre-veteral cortisol, and we take our target range of 40 to 138. above that, above that target range, we will, look very carefully at whether we would increase the dose.
Below that, we, we look very carefully at the clinical signs. And I must emphasise again and again and again, it is the clinical signs. The control of Cushing's disease is not a preveteral cortisol in target range.
The control of Cushing's disease is where you take an Akita like this and you turn that dog into an Akita like this. And if you achieve that, then that's your, your, your aim. It is not a pre-veteral cortisol, but that preveteral cortisol can help you achieve that aim.
So the way we do this is to book an appointment just before the dog's next veterol is due. If it's given at an inconvenient time like 6 o'clock in the morning, I don't work at 6 o'clock in the morning. Then I ask the owner to give a convenient time to give the veterol at a convenient time, at least one day, if not to the day before.
So many of these dogs will come in on Monday, having had the weekend to give it at a reasonable time from my point of view. And if the owner's 2 or 3 hours either way, I don't get too hot and bothered. And that's because, the, the, variation in cortisol is, is still quite, dramatic, but it, there's no significant difference between samples taken at 22 hours.
Or 24 hours based on this work from Reto *** and his group when he was working at the Royal Veterinary College, and they showed that there was, there was little variation. There's a lot of effective veterol in the 1st 2 to 6 hours, but beyond that, it it it's not so clear. When the, animal comes in, we ask them to fill in the, quality of life questionnaire that I showed you.
We take a history. We may quickly examine the dog if, if the dogs, if the dog's owner reports to us that the dog is unwell. But what we're mainly looking for is a history and clinical signs that are, about the Cushing's disease.
And, looking at a dog like this, obviously, it tells us that when we're not getting great control, we ask about the polyurea polydipsia, we ask about the polyphagia, we ask about the lethargy, we ask about any skin changes, but we're not actually handling the dog that much because I want to keep the stress levels for these dogs as little as possible. And, and of course, any other signs that the dog had been, recorded with having Cushing's, before, before the diagnosis. Really important point.
If you've not done this test, then when you change, it is important to make sure the owner has not given the veter all that morning and that nothing stressful has happened that morning, e.g., the dog hasn't vomited that morning or being stressed in some way.
We had one owner who turned up, and the owner was obviously very stressed and upset. The dog looked fine, but the owner was very stressed and upset, and we asked what happened, and she'd been involved in a car crash on the way to the vet school. As such, measuring, pre-veteral cortisol in, in a, in a dog that has sat in a car while it's had a crash is probably not a good idea.
It is important that these dogs are, are unstressed, and also that, if they're particularly aggressive and it takes 2 or 3 nurses to hold the dog down to get the sample, then it probably is not a good idea to do pre-veteral cortisol on those animals. But assuming that's the, the case, then we would take a, a sample immediately after the examination. I will often have my owners hold the dog.
I'm quite happy if they're, if they're able to hold the dog and, the dog's good, then I think it's better done in a consulting room rather than, getting, nurses to hold and so forth. But it's up to you, you, yourself. But, if you're the more you're going to do, of, of getting nurses involved and, getting, the distressing the more important is to do the sample first, when they come in and don't worry about the history and clinical exam, do those afterwards.
we send the sample to an external laboratory, that's participating in external quality assurance schemes, such as the ESVE scheme that, Peter Graham is so heavily involved with, and preferably using a a laboratory that's using a Siemens simulite or a method that has been validated against that machine. And that's because Different cortisol, the different samples sent to different laboratories can give you different results. And the work that's being done at Zurich and at Glasgow and at Utrecht and Bologna were all done using semen simulites, and we know that they're very consistent between centres.
How do we interpret this? Well, we start with the history and the clinical examination. If there are signs of poor health, then we stop the veterol and we do an ACTH stimulation test.
If there are no signs of Cushing's disease, so the dog is, is clinically asymptomatic, then if it's in the target range, no change in dose is needed. If it's low, then we should go back. We should reevaluate the dog and probably lower the dose.
If it's high, then again, reevaluate the case. Many dogs with no clinical signs, sorry, many dogs with prevagal cortisols above 138 do have clinical signs. So go back and ask the questions again.
Consider twice daily dosing in those animals or a small dose increase. If there are clinical signs of Cushing's disease, then we use the pre-veteral cortisol to determine the magnitude of our increase in frequency and then in dose. So we'll go from once a day to twice a day very quickly and then beyond that, we'll, we'll up the dose.
If there are clinical signs of Cushing's disease and the pre-veteral cortisol is, less than 40, then that's kind of, inconsistent. I haven't come across that in all the cases we've done, never really come across that, really. But, I, I think if you do, then you should probably contact, an, an experienced endocrinologist, or, the people who supply you with veterol, or repeat the test.
OK, so that's pre-vectoral cortisol. Let's just talk quickly about haemoglobin A1C. We're all very familiar with the fact that glucose binds to protein.
We use it every time we measure fructosamine. That also happens with haemoglobin, and the haemoglobin that's created is called HbA1C. In human beings, HbA1C has been shown to be linked to the risk of complications.
So the higher humans, human diabetics, HbA1c, the more chance they have of having, complications. And that, goes through to survival as well. And if you look at, the, death rate in, in humans, from, diabetes, or, or sorry, all causes death.
Then you see that the human diabetics who keep their HbA1C, within a, a narrow band, range tend to do better over a, a, a longer period of time than the, individuals who have a high HbA1c and that high, high HbA1C may be because of, excessive insulin doses or underdosing. HbA1c has been measured or glycosylated haemoglobin has been measured in dogs and cats for a number of years, but unfortunately these systems were all difficult to use, expensive, and indeed now are no longer available. Indeed, in human medicine, these have also been withdrawn largely because they were unreliable.
But when we did have those tests, we were able to demonstrate a clear difference between, diabetic patients and non-diabetic patients, and in some cases between treated patients and non-treated patients. But as I said, the problem is none of these machines are now available. Some time ago I had the owner of a diabetic dog who came in to see me who turned out to be a consultant in diabetic medicine at the University of Dundee.
And it's one of those consultations where I ended up getting more out of it than I think they did. And one of the questions she had was, why aren't you measuring HbA1C? And I said, Well, we haven't no methods.
She said, why don't you try the vantage, the DCA vantage? And sure enough, we were able to get a few samples and put them through, and it became fairly clear to us pretty quickly that there was a quite a The difference between the HbA1C that we were measuring in these dogs in diabetics, compared to normal, normal dogs. And indeed, we were able to put one dog with Cushing's through there.
And on the basis of that, I was able to get a grant, from, my, my own vet school, and we were able to extend this study to, to a larger population of hospital and diabetic animals. And these were diabetic dogs. This test does not work in diabetic cats.
And we were able to, show, a nice, difference there. And one of the groups, obviously, I was worried about, was, was, was groups with, with dogs with, which might have a degree of insulin resistance, such as Cushingoid dogs. And we found that, actually the, the Cushingoid dogs tended to, to, to be only just slightly higher than the hospital population, and the, the diabetic population were, were, were clearly different.
On the basis of this, with some help from Siemens, we were able to, extend this out to about 60 normal dogs, from which we developed a reference range of diabetes, sorry, reference range for normal dogs of 9 to 18 millimoles per mole. Now, this is different units to the percentage that human diabetics tend to use, but actually we're, we're ahead of the curve, and that human diabetics are now starting to talk about millimoles per mole. What I can't tell you from this is what the target range for diabetics is.
I, I, I think it might be like one of those who use fructose means slightly higher than the, the reference, range, but we don't know. We're also looking at the effective management. Alterations on HbA1C.
We show in a number of diabetic dogs now, we've shown the ability of better control to bring the HbA1C down. But, how much it comes down and to what our target range is, we're not sure yet. And I must emphasise that if you have a high level of HbA1c in a dog, what you need to do is to consider the whole routine.
It is not a question of increasing the insulin dose necessarily. It may be a different insulin, it may be different feeding. It may be something.
It's not a, a, a, a thing that tells you what to do. It tells you how well you're doing. It, it's a score.
You need to go back to the animal and then to work out what you need to change. And the thing about HbA1c is it gives you a measure of glycaemic control over 2 to 3 months rather than 2 to 3 weeks, which you'd expect with fructosamine. Just quickly then, I'll finish on, on continuous glucose monitoring, to say that as many of you, I'm sure, know that if you do a glucose curve on, on a dog and then repeat it the next day, you get a different result.
This is Linda Freeman's study, from 2003 looking at, three visits by a group of dogs, and it was a group, but this is just one example from that paper. Showing how different these blood glucose curves are. These are all the same dog being treated with the same insulin and the same food, the same exercise, no change, week apart, and then doing a blood glucose curve on day 1 and on day 26 curves, 6 totally different results, 6 totally different ways of interpreting those results.
Sometimes you'd increase the dose, sometimes you'd lower the dose, sometimes you'd keep it the same. So our blood glucose curves, the gold standards, well, I think that's very doubtful. They're tedious, stressful, and expensive, and it's particularly the effect of stress that makes it hard to interpret these curves, because if you're doing them in a hospital, you're going to be causing stress.
So we only should use blood glucose curve and when necessary, they're not a measure of stability. They're an investigation technique when you know they are unstable. Some of you may be using these blood glucose curves at home, but we know from studies that there's only a few owners that can do these reliably.
Moreover, it doesn't decrease veterinary visits. It doesn't improve your control, and there's no evidence that it actually results in improvement in life. So that brings us on to continuous glucose monitoring, and the principle of continuous glucose monitoring is that a glucose sensor inserted into the skin, coated with glucose oxidase, creates an electrical current within the interstitial fluid, which can be picked up and calibrated against a known blood glucose concentration and therefore gives you an estimation of blood glucose concentration.
But you are measuring the glucose in the interstitial fluid. Continuous glucose monitoring systems do not measure blood glucose. They measure tissue glucose.
I prefer to refer to my owners as skin glucose because they can understand that skin glucose is not the same as blood glucose. There is a dynamic relationship between the two, but there's a lag time, and the lag time is about 5 to 12 minutes. It, it varies from, from patient to patient and indeed from day to day, depending on the hydration status of the dog.
So it's, it's not a perfect estimate of, of blood glucose, but it is, a, a, a, a good estimate. Continuous glucose monitoring systems have been used for some time now in small animals. What I'm telling you is nothing new, but I can tell you that the price tag for all of these run into these several thousands of pounds, the Guardian, the MiniMed, the iPro and I've had some experience with the Dexcom G4, but these were all very expensive.
What's new is that the, there is now a, a system available in the UK and Europe, and I believe it's, it's coming in, it's in the states as well, which, is, is cost efficient. The, the disc that you plant onto the, the skin of the dog, the little white disc here that you, you can see here on this little border terrier. It is extremely sticky.
So the dog finds it, almost impossible to get off if you cover it with, with a, a string vest or a, or a t-shirt. And the monitor that you can buy. It is also about 50 pounds.
So this, this total kit costs about 100 pounds. And for that, you will get 14 days of glucose measurements. If you have an Android, unfortunately this does not work on our iPhones, but if you have an Android, then you can download an app and your Android phone can read this disc, so you don't even need this little monitor to do.
So that makes continuous glucose monitoring now a particularly useful, useful thing to consider. And there's no need to calibrate it. It's a factory calibration.
And that's a good thing from many respects, but I would just caution you that if you think the dog is going too low, then you really should check it with a blood glucose before acting. We found a number of these will have low, low readings, but in fact, when you check them on a blood glucose, they're fine. The, the metre seems to underread by about 2 millimoles per litre, and of course that's quite a significant difference at the low level.
The sensor stays in place compared to, previous, versions of this machine. It doesn't work. The only thing, I, I, I would say is that the company are very unkeen on vets using this.
So, we tried to get the owners to buy these machines for themselves, and, to buy, to buy the discs, and to register their, their, their dogs, as if they were, were, diabetics themselves. I, I, the company are unwilling to support it. They don't know about canine diabetes, and because they are, unsure of the advice to give owners on this, but they, of course, all diabetic monitoring equipment is not going to be licenced for dogs.
We never will have a diabetic, thing that's, licenced for dogs, and we're very used to using that situation, using human equipment, for the, for a, a dog, and so I'm quite comfortable with that. But you have to be care on the interpretation. We don't know that these are exactly like blood glucose curves, so you have to look very carefully at your interpretation and perhaps redefine your criteria for what exactly, you're looking for.
So are continuous glucose curves, continuous glucose monitoring systems, useful and unstable? Are they, are they blood glucose curves made easy? Well, I think we still have to say there's a bit of a question mark about that, about looking at these curves and trying to define criteria for things, but there's certainly an exciting option in the future.
So quickly then, as I appreciate that I'm coming to the end of my presentation, some general points then a good clinical history is paramount. Don't rely on the numbers. Preveteral cortisol may replace ACH stimulation tests, but be careful, don't use them if the animal is unwell or very stressed.
They're, they're not appropriate. Haemoglobin A1C, may well replace fructosamines. they've no help in telling you what to do, just when you should do something.
And continuous glucose monitoring curves may replace conventional, glucose curves, but be careful about how you look at those, those results and do check them, if there is, is a particularly if there's a low value, then I think you need to make sure that, it is genuinely a hypoglycemia. So with that, I hope that I've given you some idea of the road ahead. It may be clear for some of you, for some of you it may more look like this, the West Highland way a few days later, and it's a question of looking at the literature and making some decisions and making any changes gradually.
So. Having done that, James, I'd just like to throw one last question to, to, to, to the audience. Which of the following do you think you will use in the next year?
Have I changed your minds? Has anyone thought that, gosh, I will have a go at this, or, or have we failed? OK, Ian, I've, I've launched that poll for everyone to look at.
Thank you very much for a fantastic presentation. It's really exciting to hear kind of the new developments and also that they don't necessarily come with kind of huge amounts of extra expense like a lot of new developments do. So, it's good to, to hear from you about all of those.
People are just voting now, so hopefully I'll be able to give you a result in a few seconds. I, I, I, I do. That, that, that point about cost is, is, is important.
we, we know from commercial sources that one of the things that stops people treating their dog is the cost of monitoring. They can understand paying for the treatment. It's the cost of monitoring that hurts them.
So all these three methods are cheaper than the current existing methods. Haemoglobin A1C is cheaper than fructosamine, in a commercial lab. Which, as I say, is, is so rare these days in medicine.
Can I just ask you a couple of questions whilst the votes are coming through that have that have been, so one person has asked, I, I think is a quite a, a good question for you which is, . You said that you'd be worried about stressed or aggressive dogs and using, the kind of cortisol level that you were checking. So what do you recommend to people who have dogs that get particularly stressed or aggressive in terms of their Cushing's monitoring?
Well, OK then, so for the aggressive dogs, I, I, I would not recommend anything that involves sticking a needle in, in the vein. I think, If, if you are putting yourself or your staff at risk by doing that, then that is unreasonable and unfair to expect human beings to be at risk of injury. And I think you should be clear with owners of aggressive dogs that they should, that the level of care that you can give is necessarily less.
and I have no compunction about that. the technical thing I would probably do for the really stressed animal at the moment, my best guess is to use haptoglobin. And the reason, and I'd still measure cortisol, but I would measure haptoglobin.
And that's because we've got enough data and it's not been published yet, but that, that haptoglobin. Is, is a, is a, is a possibly useful as a secondary thing, but more importantly, it can pick up those dogs that are very high cortisols, which are stress-induced because the haptoglobin would remain low in those animals, and therefore, that, that difference, might be picked up. So that's the way I would solve it in clinical pathology terms.
I mean, of course, even more important to rely on clinical sciences, even more important to rely on what the home records are, even more important to, to, to be that. There is no good answer. But I am quite interested in hapoglobins and I'm showing some fairly good levels of correlation with clinical scores with hamoglobins.
Excellent, thank you. Jill has asked if you would mind putting the flow chart for the interpretation of the pre-veterol cortisol up again. And presumably it's also in the notes that people.
I, I believe it was in the notes. I think if you get the handout notes, and the other thing I, I'm sure Dera would like me to point out is that, your local Dera rep should have copies of this flow chart. they can't, in some areas, in some jurisdictions, they can't.
Give you the flow chart unless you ask for it. But if you ask for it, you can get it. And there's, theirs is much nicer than mine.
It's all, it's all, lovely, lovely stuff. So it's either in the it's either in the notes or get your local, local decor rep. I can put it up here.
OK, thank you. 11 final question before I give you the poll results, which is Richard has asked, if you're going to switch a dog to twice daily better, . How what dose do you use?
How do you switch them over? Well, that depends on this, if we go back to the prevaginal cortisol concentrations, and we're assuming here that we've got a dog with clinical signs of Cushing's disease. OK.
So if I have a dog with clinical signs of Cushing's disease and a pre-veteral cortisol greater than 138, then to be quite honest, I would simply go from, say, 30 milligrammes once a day to 30 milligrammes twice a day. I'm quite brave. I think too many people are taking tiny little baby steps in moving up the dose.
And in fact, you know, before we had all of this sort of detailed monitoring, we went from 30 to 60 milligrammes without with confidence, and we, we didn't really cause much. We didn't see much need for 40 and 50 milligramme doses. So the most common doses are 30 or 60, so I go from 30 milligrammes once a day to 30 milligrammes twice a day.
If the dog's got clinical signs of Cushing's. And it's pre-veteral cortisols between 40 and 138. Now, then I will take a smaller step.
So I'll go from, say, 30 milligrammes once a day to 20 milligrammes twice a day. And then 30 milligrammes twice a day, that, that, that, thereafter. So I, I think it's, it's important that, we, we accept that this is, this is purely my best guess.
It is not research. What I'm telling you now has no basis in any research. We haven't looked at that, how we affect the preverteral cortisol with changes in dose.
that's one of the more interesting things that we need to, to be. Looking at over the next 10 years, but it's going to take a long time to get that kind of information together. It's, it's very hard work doing this kind of, stuff.
And, I, I think the advantages of the pre-veteral cortisol are now clear enough, but, we should be using it, but 30 milligrammes to 60 mg, sorry, 30 milligrammes once a day to 30 milligrammes twice a day if the preveteral cortisol's over 138. Perfect, thank you very much, Ian. I know you're dying to hear what people have said, so I'll give you the results of the poll.
You'll be pleased to know that you've convinced a lot of people, 95% of people are now planning to use the pre-veterol cortisol, and you've also got high take up rates of continuous glucose monitoring, 35% and haemoglobin A1C 25%, and very few people, have, not planning to take forward your, your new test. A successful Saturday morning then. Absolutely.
Thank you very much everyone for listening. I, I, I put up a slide of acknowledgements. None of this work would have been possible without, without, those, people listed in the slide.
and I, thank the webinar vet and, the virtual congress, for, for allowing me to speak. Thank you very much. It just leaves me, time to say thank you very much again to Ian and obviously to Peter, earlier on for two fantastic talks, on endocrinology.
I think we've, we've all learned a lot and it's been, fantastic. And thank you also to Holly and Megan who've been behind the scenes helping at the webinarves. Thank you.
Thank you.