 Thank you very much for giving me this opportunity to come and talk here. I am very impressed by the previous talk, so it's very big shoes to fail. I'm actually a clinician, I see patients here on campus and essentially for the past many years I'm from India, I used to be a clinician there, came here, trained again and then over the past many years I've been trying to, kind of I do patient-based research in epidemiological research at a much smaller scale and essentially I'm going to talk to you about this condition which is called pre-eclampsia which is high blood pressures in pregnancy. I was in Boston for many years working in a lab and studying pre-eclampsia and then for and I'll show you data about some long story short I went to Haiti and was very very surprised by the prevalence of disease in Haiti and how it's actually being treated. So hence the title, try to compare the outcomes of patients who have the same diagnosis but in two very different settings. So I am a maternal-fetal medicine specialist and I was working at Beth Israel Deaconess before I came to UNICEF Chicago. Let's talk about, so maternal mortality just for all of us, all the non-clinicians in the audience is defined as death of a mother of a woman which happens while she's pregnant or within 42 days of termination of pregnancy and I kind of like the slide because it's not just the mom or the baby that dies but obviously there's so many other things and people around her such as this little baby that she obviously is about two to three years old. So some of the key facts about from WHO there are about a thousand women who die every day from preventable causes that are related to pregnancy and childbirth so that's a very large number. In 2013 about 300,000 women died during childbirth or during pregnancy and we know that 99% of these deaths are actually occurring in developing countries and 50% of them in sub-Saharan Africa. So very very large burden of maternal death and this again I kind of stole it from the WHO slide from 2014. So why are we talking about this is because hypertensive disorders of pregnancy is the number two cause for maternal death. First is severe bleeding but in fact a lot of patients who have high blood pressure then ultimately die from bleeding so a lot of that bleeding complications that mothers die from are related to high blood pressures and pregnancy. So we're going to talk about this condition called pre-eclampsia which is one of the most common medical complications of pregnancy. So in a very true form it is characterized by hypertension and proteinuria so sometime in the third trimester the women develop high blood pressure and protein. In US the prevalence is about two to three percent and so globally if you look at maybe the average prevalence is about five percent. It is responsible for about 70,000 maternal deaths per year worldwide and majority of the deaths occur from seizures and bleeding. So women with preeclampsia can actually have seizures which is called eclampsia and they can also have abruption and bleeding. In US it is the leading cause of prematurity and I won't go into much of obviously the treatment and stuff but it is it can often present atypically and a lot of research I I'll some of the research I'll present is just trying to figure out how you can find these patients and there is no treatment and in general it is also associated with long-term complications such as cardiac and renal complications. So when I was in the lab with a little molecular biology slide here so we believe that it's a placental dysfunction. So the primary problem is placental dysfunction with this causes release of certain proteins which we call biomarkers. So these are the two common biomarkers that we've been working on for example soluble fit and soluble endogline and we believe that these biomarkers then cause maternal endothelial dysfunction. So there is widespread endothelial dysfunction for the mother and then she suffers all these different kind of clinical signs that you see when a woman has preeclampsia. So high blood pressure's protein it can cause liver dysfunction and cerebral edema and seizures. So early on when these proteins were discovered in the lab from Dr. Karamanji's lab in Boston I was a clinician I got to the lab and we were trying to figure out how you can actually use these biomarkers to predict for example onset of preeclampsia. One of the things we did was we said okay well in the third trimester the woman has signed in symptoms of preeclampsia how perhaps we can use these biomarkers. So we did a large cohort study about thousand patients and we looked at patients who were coming to triage so triage is in kind of a semi-emergency room who are pregnant in their third trimester and they have some sign and symptoms of preeclampsia and what we were able to show that these biomarkers so the ratio of soluble fit PLGF was very high in women who developed preeclampsia within two weeks of arrival to triage and not just the diagnosis of preeclampsia but these patients who had adverse outcomes had a very high ratio compared to patients who did not have adverse outcomes so it can actually be used as a discriminatory test to figure out whether you're at risk for adverse outcomes and then what we did was we kind of compared it so the problem with preeclampsia and why the burden the disease is so large is because it's very deceiving and the things that the clinicians have such as measurement of blood pressure's measurement of protein doesn't really correlate with adverse outcomes it's very hard to kind of figure out who is that patient who is at risk for adverse outcomes and then it's until it's literally too late so we compared our ratio of soluble fit PLGF to other known things that we use such as systolic blood pressure measurement of creatinine uric acid and by itself it was actually had a much higher AUC compared to other things so telling you that perhaps measurement of soluble fit PLGF is better than the clinical tests that are available so from this study and I don't want to bore you too much with science scientific kind of journey here but just kind of laying the background that we were able to show that in our cohort of patients if they presented early the ratio was associated with diagnosis of preeclampsia but more importantly identified all patients who were at risk for adverse outcomes so while I was doing all this research this was 2008 2010 I came across this article from Dr. Gordon Smith who's actually in London and he wrote this article saying that while all the research is being done in developed countries women are dying in developing countries and then he suggested he also does research on preeclampsia and he said perhaps what we should do is we should identify women who are at high risk for preeclampsia and somehow transfer them to a higher care facility for early delivery and proper treatment so there are actually a lot of studies looking at if a woman has a seizure at home she's a much higher chance to die compared to if she has a seizure in the hospital so just because you are in a in a higher level facility health care it by just by itself it can improve the outcome so we I was then I proposed a question saying what is the role of these biomarkers and we wrote a while we were in Boston we wrote lots of papers about kind of different clinical scenarios we asked this question what is the role of biomarkers and resource poor countries for prediction of adverse outcomes in women with preeclampsia and you measure it late in pregnancy so somehow the patient can actually be transferred and baby can be delivered and use it as a rule-in test so when I say rule-in test it'll be people that if the test is positive you can say maybe you're at high risk and you need to get treatment so again long story short the reason I went to Haiti was at the time when I was doing research in preeclampsia was also interested in cardiovascular disease in women and seems like Haiti has the worst largest incidence of peripartum cardiomyopathy so that happens when your heart fails and when you're pregnant and I wrote this paper and one physician from New York said you should go to Haiti and study preeclampsia he was a medicine person and he said there is so much preeclampsia in this hospital you should definitely go to Haiti and see what's going on there so I went there it's a beautiful country that's my fellow Melissa who actually we just literally bought a ticket and flew down there that's the front of the hospital I was standing there so Haas was the hospital we went it is in central Haiti it's about 90 miles up north from Port-au-Prince so you take a nice a drive from there and it serves a very large population of 300,000 people in a large 610 square mile central Haiti area so their obstetrical unit has they do about 900 to a thousand deliveries per year they have a seven bed labor and delivery unit and when I say live in delivery unit that is the labor and delivery unit so they have a bed and then there's a patient next in the next bed they have this fetal monitors but they don't have paper tracing so you kind of essentially just listen to the fetal heart rate there's no actual recording going on they do have an operating room where they perform C-sections and that's Dr. Narcisse he's one of the obstetricians there so kind of the one of the ways I kind of looked at this from a clinician's scientist's point of view is that I wanted to look at the prevalence of the disease so first of all define what the disease is so at that time when I looked at the world map to figure out what is the prevalence of preeclampsia in different countries Haiti did not even show up because nobody has actually described the prevalence of preeclampsia I did find a paper that was published in 2003 from investigators at Yale so I called them and asked them are you still studying preeclampsia and they said oh no that was you know something we did like few years ago so I said fine we'll do a first kind of a demographic study looking at what is the prevalence of preeclampsia and what is the prevalence of adverse outcomes in this particular hospital but more importantly I was also very interested in figuring out if the angiogenic factors that I've been studying here in US do they actually correlate with adverse outcomes especially severe adverse outcomes occurring in Haiti and then perhaps which we have not worked on but are currently working on perhaps developing models of care at a level of a physician at a level of a hospital where perhaps you can stratify patients at risk based on the biomarkers so the first study like I said was a retrospective study we looked at the medical records all the caveats of looking at medical records and in this hospital we essentially wanted to evaluate the women who had a diagnosis of preeclampsia eclampsia either before or after delivery so we looked at a two-year record from in 2011 and 2012 so essentially what we found was there were in that two-year period there were about 1700 deliveries now astronomically high rate of preeclampsia eclampsia and obviously the prevalence is difficult to define because I don't really know exactly the total number of patients who were pregnant at that time period but at least in the hospital out of the 1700 deliveries that they did 270 patients had preeclampsia eclampsia what was even more surprising that about 30% of these patients had eclampsia which is seizures during pregnancy very very high rate and then what was also very interesting to me as a clinician I've never seen this in US was a very high rate of postpartum eclampsia so that's where you deliver and then you seize after delivery which is kind of a very enigmatic disease just from like scientific point of view so obviously I had all this data in Haiti and then I had all this data in Boston and I was like wow this is like really very different these patients are not similar to the patients I was seeing in Boston so I kind of just simply compared the clinical characteristics of patients who came to Haiti compared to patients who were in Boston and as you can see women in Haiti were younger so they were 27 on an average compared to 32 they were in terms of parity the little bit more Paris there was no real difference in gestational age of admission or delivery and again you have to understand this is very biased I did not look at how many women died at home or how many women went to some other hospital there are no other hospitals by the way around this hospital but perhaps to the dispensaries or something else what we were amazed by that the level of hypertension was immense so in Haiti the average blood pressure was 160 or 110 compared to Boston for example 147 over 92 and then the other thing was obviously very clear that the number of women who received prenatal care was very low so in Haiti only 23% of the women received prenatal care an average number of prenatal visits for the only two compared to Boston where 99% of the women received prenatal care with an average number of prenatal visits of 10 and that's pretty average in the United States a lot of it's like a pyramid that you as your pregnancy advances the number of times that you're going to see your physician actually keeps going up and up but in terms of adverse outcomes so we obviously were interested in figuring out what is going on in in in Haas so one interesting fact we found was this diagnosis of mild preeclampsia so mild preeclampsia is an asymptomatic disease so you don't have any symptoms as a patient but when you go see the doctor they will find high blood pressures and protein and they'll say oh you have preeclampsia so we found that in in Boston about 43% of the women were diagnosed with preeclampsia well in Haiti mild preeclampsia did not exist nobody asymptomatically is presenting to the doctors for labeling them with preeclampsia what was there was very high prevalence of severe preeclampsia very high prevalence of anti-partimaclampsia and postpartimaclampsia and then also obviously unfortunately there was very high rates of adverse outcomes so when you look at the fetal outcomes there were 51 babies who died among 270 patients but only three babies died out of 117 in Boston so very very dramatic obviously very very high rates of IUFT 20% so one in five babies are dying and then the rates of and unfortunately also for example when you look at the maternal death the rates of abruption were high now I was a little bit surprised by it wasn't very high like corresponding to this but because the abruption is a very clinical diagnosis but also pathological diagnosis so a lot of times the placenta obviously they don't have a pathology department there so the placenta is not routinely examined by by pathology so I think it's a pretty fairly under diagnosed but unfortunately there were five maternal deaths out of those 270 women but obviously no maternal deaths in Boston the other interesting thing we kind of did was we looked at the biomarkers so I'm calling them biomarkers because blood pressure is a biomarker is predicting something we found that there was actually no association of blood pressure systolic or diastolic blood pressure with any of the adverse outcomes which again people have written about it hundreds of times in multiple different scenarios and it is a very very poor biomarker for predicting adverse outcomes in fact with blood pressure in 13 the positive predictive value is only about 20% so clearly it was very clear that these women were actually very very sick and they're dying and there we looked at all different outcomes it was pretty much crossing one so hypertension and progenuria was a poor predictor of association with adverse outcomes so we went back to answer the second question whether angiogenic factors are actually associated with adverse outcomes in Haiti so this was a prospective study that we did there we went back to Haas we enrolled patients after their consent we drew their blood we kept the blood in minus 20 they did have a lab and we transported the samples back to Boston and measured the angiogenic factors so that was the lab there and that's my fellow Carl so what we found was again kind of expected I knew it it will be but what we found was then patients who have early onset preeclampsia so women who had less than 34 weeks they have a very very high levels of angiogenic factors so very very abnormal angiogenic profile and similar results with late onset preeclampsia and also interestingly the early onset preeclampsia again it's known is a very kind of a bad disease so their levels were higher than late onset preeclampsia but what was kind of interesting to show and kind of move the field forward was that it was not just the diagnosis of preeclampsia but the adverse outcomes so what we did was we characterized adverse outcomes based on kind of some kind of scientific knowledge so we had severe hypertension is category one but we kind of looped in punched in all these adverse outcomes together for category two and I believe that eclampsia is a separate disease I really truly believe there's a biomarker out there for for eclampsia so we I characterize it separately and then we felt maternal fetal neonatal death is a bigger outcome than for example IUGR so we kind of group them together what was interesting then all patients who had preterm adverse outcomes all of them had very very abnormal angiogenic profile compared to patients who did not have adverse outcomes and this is also true for the late onset adverse outcomes so I think we came back and we kind of concluded that there is a presence of angiogenic imbalance in the women who have a diagnosis of preeclampsia but more importantly there was a profound angiogenic imbalance among all women who had preeclampsia related adverse outcomes regardless of gestational age so when we wrote this paper there was lots of debate in and in literature people saying that well you know we're sure it can predict help but it doesn't predict x y and z so I think and after that obviously a lot of patients people have written a whole bunch of papers about angiogenic factors so I think I'm just going to conclude with two other slides so what did we learn from Haiti so I think and I didn't show you some last year we got funded from the global health department here and we went back and we looked at postpartum hypertension so again persistence of blood pressures after delivery it's very very prevalent again very high blood pressures so I think we learned that the prevalence of hypertension during pregnancy is very high and I again didn't show you the data but the treatment is not universal so even in the hospital where there's availability of anti-hypertensives patients are not being treated with with blood pressure medication and given for example magnesium for prevention of seizures so what would be the reasonable steps as a scientist I asked perhaps we should explore the existing and I know they're like you know societal burdens and national burdens and poverty burdens I'm talking about what you can do in terms of being a physician and in the situation so perhaps we can look at existing hypertension protocols in the hospital if they have have them looking at the existing knowledge among care providers and dangers of hypertension do they understand it and I really truly want to understand what are the barriers for treatment for hypertension and again other thing is patient education are there any instructions given to patients about dangers of high blood pressures and then obviously this was my basic question and I'll show you one more slide as to how I think it needs to work is in weaker health systems where you don't have so much resources that you can potentially see every patient every week are their role of biomarkers that you can predict and actually act on patients who are at high risk so there was a very recent study that was published in Hypertension Journal in April of 2017 with this very large group they went to Mozambique in a clinic and they took patients who were in their third trimester coming to this clinic and measured their plgf and at the plgfs low they're at high risk for adverse outcomes they send them to the hospital and say go get evaluated for preeclampsia and they were able to show that patients who got delivered had a much much lower rates of adverse outcomes including abruption they didn't really show a big difference in maternal death but they were able to show some of these very bad adverse outcomes are happening much lower rates in women who had the accessibility to for example this biomarker so I wrote an editorial attached to that paper and I think this is what is going on so for the prenatal care the way we perceive it for the past many many years is we have this conventional approach so what's going on is you repeat measures of blood pressure you repeat measures of protein and you evaluate the patient's pace on non-specific sign-in symptoms so any of the sign-in symptoms that we educate our patients they're very non-specific and happen when you have migraine or something else going on it's very resource intense so you have to do it multiple times it actually has very low sensitivity in specificity it's not at all cost effective because you have to have clinics you have to have all these people it requires constant access to high quality care so perhaps come to my clinic and I can pick up preeclampsia faster you go somewhere else and perhaps they'll miss the disease and then unfortunately all patients are treated the same way so everybody goes through the cycle over and over again it is a model that we practice here in United States I see my patients a lot of them on a weekly basis and I think it's a it's a reasonable model if you don't think about cost but also because majority of the patients are receiving prenatal care though our caveat there I want to point out I don't have a slide but United States is the only nation in the developed countries where maternal mortality is actually going up and a large majority of it is actually delayed diagnosis and preeclampsia and hypertensive disorders what I think is we need to have something close to personalized medicine where we can have and I don't say that the biomarkers that we propose maybe there are other biomarkers but have some sort of a biomarker based risk assessment whereby maybe will require less resources they are very sensitive so the if you kind of look at adverse outcomes and the sensitivities are up to 80 percent it'll be cost effective if you know the company is actually selling cheaper it does have a very high positive and negative predictive value and I think it'll require less expertise so if you have a test and if you say if your number is 85 you're at high risk and if your number is two you're at low risk so anybody can kind of interpret it and then essentially you can divide patients into rather than treating them all the same way you can stratify them as high risk and low risk majority of the patients are actually low risk they can deliver in their own place wherever the plan was to deliver but among the patients who are high risk I think if transferring them to a higher level facility to refer them to physicians with knowledge and perhaps access to high level maternal and neonatal care will actually reduce the maternal and neonatal death rate so where am I going next so first of all before I say that I want to acknowledge obviously a lot of people but essentially all the medical staff at Haas they're fantastic people my department of OBGYN here gives me all the time and resources to do this kind of work Dr. Karamanthi has been my mentor for the past 15 years and then my hospital in Boston at BIDMC and my funding so where am I going next I actually was just visiting HUM I we have an MOU I'm told is almost signed so that's Meredith is fantastic she runs the maternity ward there so we're trying to figure out if we can come up you know it's a bigger hospital more patients to figure this out so thank you very much