 and thank you for being here for this talk the picture that you've been seeing all days the picture of peak actually so this this was one of them in the in the MSF setting so MSF France has been working in Coutiala in Mali since 2009 and we have a pediatric project that has that focuses on curative care for underfives at hospital level at health health center level and in the community through community health workers a big part of the focus is on malaria and another big part of the focus is on nutrition we also do preventive care with a project in in Contagela where we do vaccination nutritional supplements bed nets and general follow-up and since 2012 we've been implementing at district level so outside of the in the five health areas but also outside of them seasonal malaria chemo prevention which is being transferred to the Ministry of Health this year we also do operational research it's it's our biggest pediatric project with transversal surveys the Ronnie that I presented before research attached to see to SMC antibiotic resistance and peak which we did in 2014 so what is peak it's portable eye examination kit play on words form from the ophthalmologist side it was developed by LSH TM and Stuart Jordan and people at the University of Strathclyde to help in and diagnostics for eye disease and low resource settings basically what it is it's a sleeve and I'm sorry that I don't have a sample here but it's a sleeve that adapts to a smartphone and that helps channel the light from the torch of this of the smartphone and amplify the image to give a an accurate image of the retina of the back of the eye there's also a software part with an app that serves as a that serves as a database and also helps capture image there have been trials in Kenya and in Botswana and use for adult eye care mainly for diabetes care so why why would we need to use peak in in MSF settings well any tool that can help more accurate diagnosis will be an interesting tool for for us to use especially if it's simple and if it can be a point of care tool funders examination for all of us clinicians is something quite difficult we learn how to do it in a residency or in medical school and then we'd never use it because it's hard to do when it's yeah it's hard to do and we don't really know what we're looking at a lot of the times the and the retina is an accessible window to what's going on in the brain so a lot of the changes that we see in the retina is actually what's happening in the brain and allows us not to have to do a brain biopsy which is not very plausible in our settings or in any settings for that matter so this is how peak comes in we had a few of the sleeves here so okay so now to the malaria side so what is malaria retinopathy so in the late 90s a group of clinicians and ophthalmologists saw that in there were changes in the retina that happened in cerebral malaria so the definition of cerebral malaria is patient with a positive malaria test and coma that cannot be associated to anything else as simple as it seems as a definition in the field we find a huge range of conditions that get classified as cerebral malaria without us being able to really discern what was cerebral malaria and what wasn't so what they saw was that there were three main changes in the back of the eye that happened only and there were pathognomonic so that happened only in people who had in kids who had cerebral malaria and these were hemorrhages white patches and abnormal vessels so these are the hemorrhages quite easy to see here and in these nice pictures so these pictures are taken with special cameras and using quite expensive materials that of course is not available to us in the field these are the white patches and also the hemorrhages and in this one you see abnormal vessels if you're not trained in seeing these it's not very easy to see and especially not with the tools that we usually have and if you have to look at it you have to look at the patients back of the eye while the patient might be my what while the patient's eye might be moving you can only get a peek at that point in time and then try to assess what that what what you think what you think you actually saw without being able to compare with any of your colleagues so what they saw when they did the studies was that there were there were some kids with cerebral malaria who didn't have these changes but that some that that the other changes were quite were present quite frequently and this obviously doesn't add up to 100 but it's because some kids presented with more than one change so what's interesting with this is that they there is an association between the presence of these changes and the and the risk of mortality so in terms of prognosis it is quite a good tool to use so yeah so 36% of the kids with papillodema and malaria retinopathy died 15% with only malaria retinopathy ended up dying and 44% of the kids with papillodema ended up dying I will explain a little bit of papillodema if I time at the end so yeah so another study this is more for the background also showed the 25% of the kids who had been diagnosed with cerebral malaria and who died did not have cerebral malaria on autopsy which means that maybe they could have been treated for another condition and they weren't because they it was just assumed that they had cerebral cerebral malaria because they had a positive malaria test this was one of the this is one of the algorithms algorithms that was proposed after that study all of the cases when they're positive for malaria they're treated for malaria but other causes will be looked at if there's no malaria retinopathy so now coming to our study what were we trying to do we were trying to see whether using peak as a tool as a simple ready-to-use tool by non-ophthalmologist clinicians was as sensitive and as specific as using the binocular indirect ophthalmoscopy which today is the gold standard and which is a tool that's used by ophthalmologists and relatively it's relatively expensive but not only that it's quite difficult to to use and difficult to have in the field we did this in in Mali where we have in in the hospital where we have approximately two thousand cases of cerebral malaria diagnosed per year mostly during the malaria season we had 61 inclusions we did this in 2014 had 61 inclusions and we were trying to validate peak as a as a tool we were also trying to assess what training needs the non-ophthalmologist clinicians would need to be able to use peak and then eventually think about other potential uses of peak if this was a feasible tool so the inclusion criteria were kids from six months to five years presenting at the Kutela district hospital with either a coma or repeated convulsions that required administration of diazepam and were assessing the coma scale was not was not possible and with consent provided by a guardian they did not need to have a positive malaria test because we were trying to see if the non-ophthalmologist clinicians would identify malaria retinopathy in kids that did not have malaria and the exclusion criteria was kids that with hypoglycemia that and coma that resolved quickly after administration of glucose so all of the kids received if they were positive from from malaria they received injective laureates in it and usually they received septraxon which is part of the of the guideline with today we give kids in a coma with a positive malaria test treatment for malaria and treatment for meningitis the pupils were dilated and then there was an examination done by Susan the one in who's one of who's the ophthalmologist who first identified first described malaria retinopathy and the two non-ophthalmologist clinicians and data from peak was registered in the app and from bio it was registered on on paper and then all mixed up together so this is Susan using the indirect ophthalmoscope it's this device that she's wearing on her head and plus a lens that she needs to use needs to use in her hand see whatever she can see only she can see and it can't be compared with no other colleague can have a peek at what peak sorry sorry for that at what she's and what she's looking and this is her doing this is when we were doing the training and looking at peak and having the other the the other doctors look at what she's looking and it's also recorded so we can look at it afterwards so the results which you can't really see I can see it here this the this is the these are the sensitivity so this is clinician number one and this clinician number two they received a two and a half day training and this is the sensitivity that we found comparing their their outcomes with those of Susan's and that's the specificity so quite oh great quite high specificity and sensitivity above 80% which for people who have never looked at the back of the eye before is pretty good so continue with the results when when comparing and that was for all types of malaria retinopathy when we compared each one individually the the the highest agreement was on the hemorrhages which is probably the easiest to see papillodema which we were hoping would have a better result because then it would help papillodema is a sign of intracranial hypertension so we thought if we can actually help identify this it will mean that we can decide whether or not it helps you decide whether or not you can do a lumber puncture so we thought if we get good results on that we can also use it for clinical changes but we didn't get very good results and the images are still being reviewed today by at morefields eye hospital in London for for further assessment and so we will have we will refine the results a little bit more so this is well this is not a test but if anyone wants to shout out what they think they're seeing you can go for it that's the back of my eye for yeah this is a video that we put together to try and show the rest of the clinicians in the hospital what was what we were looking at because everyone was very curious of what and what so that's normal I hope that's a child's disc which is also normal one of the things that was very interesting is that we were also teaching or reviewing anatomy which sometimes we forget about so that's a hemorrhage which is quite easy to see now if you have the time and the the time to look at it and to discuss with other colleagues it might not be so easy to see if you have a couple of seconds you also see that these are some of the the hemorrhages and some whitening which is not so easy to see but when you can look at it after having looked at the child it's yep there you go makes it a little bit easier there nobody's guessing it was all it also showed that it's sometimes it's quite difficult some of these kids they're in a coma but they they have quite a lot of nistag most so it's quite difficult to see and at least this can capture it for longer period time and the vessels were the most difficult to see but right it's not even even with pick it's not so easy to see right there there you go okay so in conclusion it seems like peak can be a good tool to improve our our diagnosis we've been talking a lot about data but data is only going to be as good as whatever you introduce in it so if we're making if the clinicians are making a wrong diagnosis of cerebral malaria we're going to think we have huge amounts of very little cerebral malaria when when we actually don't know what what we're actually looking at it will not change the clinical management of cerebral malaria any child with a coma with a positive malaria test will still be treated with injectable artesanate and with and with septoriaxone but it will help refine the diagnosis and help us talk the diagnosis in the prognosis and help us communicate with the family and see what we're what we think we're seeing and what we think the patient's outcome is going to be and it can also help improve the epi follow-up of cerebral malaria like I was saying before if we're seeing in some of our projects with the introduction of some preventive measures that it seems like the proportion of cerebral malaria is increasing which is one of the reasons why we did why we did this study we're not really sure if that's true or if that's just something that's just what the what the clinicians are diagnosing so having tools like this that help refine our diagnostics will be will be of use in these in these kind of follow-ups and we also found that the two-day initial training was insufficient and a four-day training would be would be more appropriate and recommend it and what's most interesting I think about this is that a peak seems to be a feasible tool for us to use in the field so hopefully once once the tool is available we will be able to use it for diabetes follow-up and for for HIV and probably for more more conditions in the future and that's it I think so thanks a lot great thank you