Is it likely that, for instance, semaglutide, GLP-1 agents are causing this because they are, as it were, a "poisonous agent"? And the answer is, absolutely not! The bottom line here is that in those people who’ve already got longstanding diabetes, who’ve got a relatively high haemoglobin A1c, who’ve got established retinopathy, that one should be quite careful about sudden improvement in the glycaemia because those are people who are potentially at risk. Hello, I’m David Matthews. I’m emeritus professor of diabetes medicine from Oxford. I’m talking today about retinopathy and the way in which this can be deteriorating with the involvement of fast improvement of glycaemic control. The background to this is essentially shown in the DCCT, where over the first couple of years, they noticed that a subgroup of people were deteriorating in terms of their retinopathy, but that subsequently you were doing much better with good glycaemic control. When you looked at the data from the DCCT, it turned out that those people who hadn’t had previous retinopathy were not really at risk at all. But those people who got existing retinopathy, and indeed those people had diabetes for many more years, typically eight years diabetes, those are the people that were at risk of getting a deterioration of retinopathy, if you improve the glycaemia very quickly, and the DCCT did really improve the glycaemia in people with type 1 diabetes very quickly indeed. The UKPDS, by contrast, was just looking at people who were entirely newly diagnosed. And there was really no signal at all from the UKPDS, except a tiny suggestion that improvement of hypertension in some people might have given a signal for deterioration. Then, there were the agents that were improving glycaemia very quickly in terms of oral agents. And they included semaglutide and liraglutide as an injectable agent. And they had the same effect of improving glycaemia very quickly. And there was a signal within the semaglutide group that there was a deterioration in some people in retinopathy. And that led to the idea that it would be important to look at this as to whether this was related to agents or whether, in fact, it was just related to the sudden improvement of glycaemia. Is it likely that, for instance, semaglutide, and GLP-1 agents are causing this because they are, as it were, a "poisonous agent"? And the answer is, absolutely not, simply for two reasons: One is that this occurs very early on and doesn’t continue, despite the fact that you’re continuing to take these agents. And so, if you thought that the agent was poisonous, as it were to the eye, then you’d expect the deterioration to continue. With the issue with semaglutide and the possibility that there was an agent responsibility for the change, the drug company Novo Nordisk set up the FOKUS trial, which is a five-year trial, looking at glycaemia and retinopathy outcomes. And so, that should be a most interesting result because that will tell us specifically whether there is, first of all, an effect of sudden change in glycaemia versus the outcome in terms of retinopathy. And second, because it’s a five-year trial, will tell us also whether semaglutide is something that has got a background problem with retinopathy or whether, in fact, the only problem is relating to the fast and early glycaemic improvement. The question interestingly arises as to why you don’t see this with SGLT2 inhibitors. And the answer is, you may well do, but the cardiovascular outcome trials didn’t properly look at retinopathy. They were focused on cardiovascular outcome trials. And so, if you don’t have very detailed ophthalmoscopy at baseline and follow those carefully, then you miss that signal. But I suspect that anything that improves glycaemic control suddenly, if you looked at it carefully would demonstrate the same effect. The consensus now is that it really is about the sudden change in glycaemia and those people who’ve got existing retinopathy. And the reasons for thinking that is that first of all, it was shown in the DCCT as with a sudden change in glycaemia, a sudden improvement in glycaemia. The effect is also seen in people with pregnancy and with existing retinopathy, so sudden improvement in those people shows the same effect. And in fact, in bariatric surgery, we also have an effect where you get a sudden change in glycaemia and a deterioration in retinopathy in some. In terms of monitoring, I think that it’s important in those people who have got long-established diabetes that they need careful ophthalmic screening, that if one sees any significant retinopathy including microaneurysms, cotton ball spots, haemorrhages and so on, that one should be very cautious in terms of therapy and one should also be cautious in terms of how long one is going to wait before reviewing these. So, ophthalmological review should typically be within three months of instituting new vigorous therapy for improving glycaemic control. We should stress that it’s only in a subgroup of people that you’re getting this change and that overall, your chances of getting retinopathy are much improved by having good glycaemic control. So, the bottom line here is that in those people who’ve already got long standing diabetes, who’ve got a relatively high haemoglobin A1c, who’ve got established retinopathy, that one should be quite careful about sudden improvement in the glycaemia because those are people who are potentially at risk. However, better glycaemic control gives you better outcome both in cardiovascular disease and in microvascular disease, kidneys and what we’ve been talking about today, retinopathy.