 Thank you, German. It's a delight to be here today and to have the opportunity to cross some of the many many different exciting areas of research in the university. I thought for today what I would do is tell you a little bit about an area of research that's occupied me mostly over the last 10 years having been originally trained as a pediatrician and come to research rather by accident. I'll tell you a little bit about the problem of low glucose is in newborn babies. How the story of our research has evolved over the last 10 years or so and then a little bit about how it really happened, which isn't quite the way one would normally present it. So why low glucose is in newborn babies? Because they're really really common and if the glucose levels are low enough for long enough in newborn babies, that can cause brain damage. Because glucose is the major fuel for the brain. Babies have big brains. So if you don't have enough glucose and you're a baby, you have a brain problem. At least that's the principle. What actually happens is that babies go through this very dynamic period around the time they're born because before birth they get a continuous infusion of glucose across the placenta from their mothers and babies blood glucose levels are about 80% of mothers no matter what. Mothers go up and down so do the babies. But with events happening around the time of birth glucose levels go up with stress and you can see this is around the time of birth. There's a very wide range and the levels are generally high and then somebody comes along and cuts the cord. So you've stopped the supply of glucose across from the mother. Glucose consumption hasn't changed because the brain still needs to work. So glucose levels fall and they fall very dramatically reaching a low point somewhere around one to two hours of age and then in most babies they get themselves sorted out. They stop pretending they're a fetus. They start to make their own glucose. Indeed, they start to feed and glucose levels come back up. But they take two, three, sometimes four days to get up to adult levels and in the meantime some of them get into trouble. And this dotted line is a level at which we often define low glucose levels and you can see a lot of babies go down below that level for quite a long period. So the problem with low glucose in babies is that you can't tell it's there without doing a blood test. Babies can't tell you they feel grotty. They can't tell you about their symptoms. So you have to measure blood glucose levels. So we end up measuring glucose levels in lots and lots of babies. In fact, turns out to be about a third of all births are at risk of low glucose levels. And although we do that, we don't actually know for any individual baby how low is too low for this baby. And so we end up treating babies at an arbitrary threshold. And we do think that repeated or prolonged episodes might be problematic. And then we don't actually know what to do when we treat them. So that we surveyed about 10 years ago all units in Australia and New Zealand who look after these babies and we said if you have a baby like this or like this, what would you do? And you can see that there's all sorts of things people might do which really means we don't know what to do. So everybody has their own view, including not treating them at all. So that led over a long period of time to the sugar baby study led by Deborah Harris, who was a PhD student at this university. And we said could you treat low glucose levels with dextrose gel? This is a high concentration sugar gel rubbed into the baby's cheek. Would that be more effective than a placebo gel with a feed in reversing low glucose? And the short answer is it was. And in this trial, which we published at the end of 2013, you can see the incidence of treatment failure. Glucose not coming up. Was substantially reduced in the babies who got dextrose gel. And more exciting, the incidence of being admitted to an neonatal intensive care unit needing to be separated from their mother to have intravenous, to have a drip to treat their glucoses was markedly decreased as was the rate of formula feeding. Now the thing I was most worried about is whether we would stuff up breastfeeding by giving these babies dextrose gel. It was entirely possible that the babies would think dextrose gel was a good idea and they didn't need to feed and we would stuff up their breastfeeding and that would be a bad thing. It didn't happen. In fact, we improve breastfeeding. I thought very exciting. I think what it did was simply reverse the low sugars. So it took the pressure off the mothers. The babies fed better because they felt better. We didn't admit them to intensive care. So we didn't separate the baby from the mother and breastfeeding went much better. Very exciting. We were tremendously excited by this finding. We recommended that this might be first-line treatment for this group of babies. As I said, we published at the end of 2013. It's now the middle of 2016. So 18 months later, there are publications, publications from different parts of the world saying, guess what? They found the same thing. Admission rates fell, breastfeeding rates improved. Those of you who know anything about the slowness of medical practice to change anything will know how miraculous this seems to me. It's tremendously exciting. So with good evidence, a simple, cheap, effective treatment can be introduced into practice quite quickly. And we found that perhaps the most exciting part of that study. The other study I'd like to tell you about is the child study. And like many good researchers, it started with a good hypothesis that turned out to be totally wrong. So the straightforward hypothesis was that the outcomes, the development of children at two, ought to be related to the severity, duration, and frequency of low glucose concentrations in the newborn period. It was a perfectly reasonable hypothesis, I thought, and we started this study because we had over 600 children, all born at Waikato as it happened, who we had collected data on carefully, prospectively. We'd measured them very accurately. We'd looked after them. And we'd also included continuous glucose monitoring, which is not normally used clinically. And that gives you a reading every five minutes on what's happening to the baby's glucose. But that data was not available to the clinicians because you download the data at the end of the monitoring period. So it didn't change what people did in practice. They just did the usual thing with monitoring blood glucose. So within that study, there are babies who never got hypoglycemic because they were at risk and they didn't get low levels. There were those who did get low levels and were treated. And there was a smaller number who got low levels that we didn't know about because they were only detected on the continuous monitors and therefore were not treated. And we hunted these kids up and down the country and we saw them all at two years. And we said what was the relationship between development at two and glucose levels in the newborn period? And the short answer was there wasn't one. Told you it was a perfectly good hypothesis that ruined by a bunch of facts. That happens to be a lot. So there is no relationship between these outcomes. These are the mean and 95% confidence intervals and these measures of performance at two. But undaunted and because we'd extracted enough money from the NIH, we saw them all again at four and a half. Which is a miracle in itself because the HRC would never have funded us to see them at four and a half. And at four and a half, there is a relationship and there's a dose response. So here, for example, are the different outcomes, visual motor difficulty, executive function. This is related to severity, none mild severe, none mild severe. And the same is true of frequency, none, one or two episodes, three or more episodes. So it turns out to my great surprise that five years later, the hypothesis is in fact supported by the data. It's just it got stuffed up along the way, which is a good reason to hang in there and to keep looking if you think it ought to be there. And perhaps most alarming from a clinical perspective was this finding at two years. The red ones are the children who had some problem at two and the blue ones are the children who didn't. This is the continuous monitoring data, so it measures every five minutes over the first 48 hours. And what you see in all children is the ones who ended up having a problem at two actually had higher, not lower, glucosees in the newborn period. And that was even more marked in the babies who had low glucosees in the newborn period. So if you started low and then went high, you did much worse in our study. That's quite important. It's led to me to all sorts of hot water because the American Pediatric and Endocrine Society, who are terribly powerful in this area, last year published a recommendation that all newborn babies should keep their glucosees above four, above four after day three. And we're going, um, maybe not. It's an ongoing debate, but it does show the importance of the discussion. So what are we doing now? We're busy writing a new ground application, of course. We're always writing new ground applications, but we now really do think we should see these children again. And that the oldest one turns nine at the end of this year. But of course, nobody would give us any money until we'd done the study and shown there was something there to look at. So we're now aiming to look at them at nine. And we're writing the applications and piloting the protocol because we can't wait until we've got the money, or they'll be ten by then. We're doing a trial of dextrose gel to prevent low sugar levels. If it works to treat, could it work to prevent those changes and prevent that damage? We're doing a randomized trial of implementation strategies. How do we get people to actually use dextrose gel now we've shown it works? And that's been a very interesting project. About half of the units in New Zealand are now using it, and we're busy trying to get that up. I've been co-opted on to international steering groups where we might think about doing a different trial as to what level we should treat, and we're going opportunities in the commercial world for other treatments in this area, because the Americans have woken up to the fact that something is happening down in New Zealand that might be useful to them. So all sorts of surprising things come out of what started off as a very simple question. How did it really happen? It wasn't like that at all, of course. It was an important problem. One third of all babies getting a blood test when we don't actually know how to treat them or whether the treatment makes a difference. And I'd been interested in it since I was a student. But I wouldn't have actually got into this research at this time if I hadn't had a very pushy master's student, who having done her masters kept coming back saying, I really want to do my PhD, give me something that's interesting and important clinically. And I kept saying, I think you should work with somebody else, and I think you should come up with a different approach. And she kept coming back until I eventually said, well, here's a really big one if you really want to get into this. And it grew from there. Determination in the sense that you can't wait till you've got the money to start the study. None of this wouldn't have happened if we hadn't said, alright, this is what we want to do. We better get on with it, figure out a way to do it, and figure out how to get the money and the people and the support afterwards. And I have time and time again met people who say, if I get the money, I'll do the study. I think the only way to get it done is to say, we're going to do the study, where is the money? When I wrote the application for the funds to do the two-year follow-up, there were two areas where I had not a clue. I knew that I needed somebody who knew how to do the detailed analysis of the continuous glucose monitoring. Those every five minutes. You end up with many thousands of points. And I had not a clue how to do that. And all of the literature at the time said that vision was the thing that was most affected by low glucose levels, and I had no idea how to measure vision in two-year-olds. Not many people do, I have to say, but I didn't have a clue where to start, so I knew. But I already started seeing the children in piloting the study and writing the grant applications. And there were two things that made a huge difference. One was a stray advertising journal that crossed my desk. You'll get them. Here's our new journal. Please read it and subscribe. And I picked it up on its way to the rubbish bin. I looked at the front and it was a bioengineering journal and it was a whole issue on continuous glucose monitoring. So I looked through the table of contents and one of the authors lived in Christchurch. So I phoned him up. So that's how we got into continuous glucose monitoring. Not carefully planned. The other was what I call a reluctant cocktail party. I went to an event, a drinks event at the university, somewhat reluctantly because I was DVCR and I had to go and I wasn't really that interested, but I got talking to this really interesting young man who turned out to be a vision scientist who got very excited about the idea of measuring vision in two-year-olds and developed all the methodology that we used and is still collaborating with us ten years later. So don't avoid the cocktail parties and do talk to people. It helped that I had contacts around the world because I'd been going international meetings for years. I would never have got the NIH grant without knowing people to know who to ask. And those international contacts are terribly important and they're really important over a long period of time. You never know quite how they're going to be important. This turned out to be useful years after I met some of these people. And, of course, there's a fantastic team. We've now had more than 50 people working on these projects over the last ten years and I just get to come and talk about it, but they really do all the work. Thank you very much.