Ninth Annual Maternal and Child Health Epidemiology Conference / December 10-12, 2003
VIOLANDA GRIGORESCU: Good afternoon. My presentation, as Susan mentioned is going to be an investigation of a rare disease. The investigation was conducted while I was assigned to work with Georgia Division of Public Health MCH epidemical (inaudible) and I would like to acknowledge my colleagues from Georgia, *Hina *Yoshi, MCH epidemiologist, Emily *Conn, chief MCH epidemiology section, *Irma (inaudible) with the hospital “A.” We’re going to call that hospital, hospital “A” for confidentiality reasons from Georgia and Melissa *Tobindangelo. I will also like to acknowledge my colleagues from CDC Division of Reproductive Health, especially Dr. William Saffenfield and Dr. *Calligan. Maternal mortality is a reportable condition in Georgia. One of some maternal deaths in 2002 was reported by hospital “A.” It was the first death after 40 years and it was of great concern. The hospital epidemiologist performed a search looking to hospital discharge data and findings suggested a potential cluster of peripartum cardiomyopathy cases. As a consequence, Georgia Division Polyhealth MCH epidemiology section was informed.
The question was what is peripartum cardiomyopathy? When we find out that we might need to do an investigation. What we’re going to investigate if we go to talk to hospital. Heart failure in the (inaudible) has been recognized since the 18 century, but cardiomyopathy was not identified as it cost anti-(inaudible) publish in 1937. Peripartum cardiomyopathy is now considered to be a cardiomyopathy of unknown cause that occurs in the peripartum period in women without pre-existing heart diseases. It’s a relatively rare, but can be very devastating. In April of 1997 the National Heart Lung and Blood Institute and the Office of Rare Disease of the National Institute of Health convened their workshop on peripartum cardiomyopathy to foster a multidisciplinary review. Experts in cardiovascular medicine, obstetrics, hematology and pathology method discussed the available information and make recommendations.
The recommendations were published in JAMA in 2000 and this was kind of baseline or the main information we had and still is the only information available who covers all information or all data necessary to understand peripartum cardiomyopathy. The (inaudible) participants agreed on a standardized definition of peripartum cardiomyopathy, which is very important. There are four criteria listed in these lights. I would like to draw your attention to the interval, which is very important. Last month of pregnancy or within five months after delivery. Why this is very important. It is important because it helps us exclude preexisting causes of cardiomyopathy that be exacerbated by pregnancy rather than arising as a result of pregnancy. For example, heart failure, occurring early on in pregnancy may be caused by previously and suspected dilated cardiomyopathy and masked by the (inaudible) or hormonal stress of pregnancy. Peripartum cardiomyopathy is considered only when it occurs in (inaudible) with no prior health conditions. The diagnosis of peripartum cardiomyopathy rests on the echocardiographic identification of new left ventricular systolic dysfunction. Again, this is something really very important and is a last criterion of peripartum cardiomyopathy diagnostic.
The incidence is not known because population based estimates are not available and the diagnosis of the grave disease is not always straight forward. Despite the standardized definition and criteria I just shared with you in my previous slide, peripartum cardiomyopathy is still a challenge because many women in the last months of a normal pregnancy may experience symptoms identical to heart failure. Therefore, go unrecognized in incidents and underestimated. The information we have the incidence rates are reported in individual studies and are based on the experience at the particular institution. It may reflect (inaudible) bias, as well as individual practice patterns. The only accepted incidence is the last one I’m aware of is the estimated incidence of 1 per 3,000 to 1 per 4,000 live births, which can be translated to between 1,000 and 1,003 women affected each year in the United States. The etiology is unknown.
A number of possible causes have been proposed by the (inaudible) participants, including those listed here like myocarditis, familial cardiomyopathy, abnormal immune response to pregnancy and so on. There is not one close very well known and identified. There are some risk factors identifying the literature that include advanced maternal age African-American race, multi-fetal pregnancy, multi-parity, preeclampsia and gestational hypertension. It is unclear though whether race represents and independent risk factor or it is in an interactional phase with hypertension, and I haven’t found anywhere dealing specific information about this. There is no way to sugar coat the seriousness of peripartum cardiomyopathy. Its mortality rate is high. It varies geographically. In the United States, the reported mortality rate varies between 6 and 50 percent with nearly half of these deaths occurring in the first three months after delivery. Prognosis also appears to be related to the severity of the left ventricle dysfunction and this is just something else very important. The critical window for opportunity is the first six months after diagnosis. If a woman’s heart recovers normal heart function during this period, her chances of survival improve significantly. However, half of these women who have peripartum cardiomyopathy can remain with persistent left ventricle of this function and these women have a high mortality rate in the following five years of 85 percent.
The question was again what is the risk for subsequent pregnancies and is not anything fit in the literature. Research studies and case studies indicate that it depends on the severity of cardiomyopathy and whether it reverses itself. If a woman develops a severe case and it does not improve after six months, she probably be advised not to have more children, but as I said it depends of follow up and how she recovers. Now we feel a little more comfortable to move forward in to do the investigation. We kind of have a sense of peripartum cardiomyopathy is and what we need to look for. We investigated the peripartum cardiomyopathy suspected cluster further. In the peripartum cardiomyopathy cases we are supposed to investigate what identified by hospital “A.” They thought (inaudible) hospital discharge data and these are that diagnostic hospital epidemiologists used to identify how many cases that hospital had. As methodology, we did a comprehensive chart obstruction of data containing the charts provided by hospital “A” epidemiologists. No interviews were conducted to either women who experienced a peripartum cardiomyopathy or physicians. Therefore, all diagnostic cases were accepted as recorded in charts by the physicians who provided care, and as I said again, we are looking only to charts provided by hospital epidemiologists.
We devised 19 charts from 1997 to 2002 and we identified 13 cases of peripartum cardiomyopathy, 12 of them were diagnosed during post partum, one ended in a maternal death in 2002. If we like to look by year, how these cases were distributed from 1997 to 2002, we identified that each year for first four years they were just one case of peripartum cardiomyopathy with four and five respective in 2001 and 2002. Looking to the number of live births that hospital “A” had, it was an average of 2159 live births and 107 fetal deaths. But if you at number of live births, this hospital had a slight increase each year, but if you go back and think to estimated incidents accepted of 1 case per 3,000 live births, obviously in 2001 and 2002 there were too many cases, was above the accepted incidence and we considered that was a cluster. We looked to all these risk factors identified in the literature for these 13 cases identified and this how these factors were listed. The age of those clients mirrored the age distribution of all women having live birth. They were whites than black with a white/black ratio of 2.25, but this reflects again, the way that hospital offered services more to white women than black women. Method of delivery for most of the cases was vaginal, pregnancy outcome most of them were at term birth, seven had primipara with primipara preterm births, four cases in 2002 had pregnancy-related hypertension.
This as I said, are all 13 cases we identified, highlighted in like orange color, there are nine cases identifying 2001 and 2002 who are actually the cluster. If we looked at those orange cases, orange risk factors, for cases in 2001 and 2002, there were nine peripartum cardiomyopathy cases, which made up 69.23 percent of all cases we identify in that time frame. There is distribution again, mirror the (inaudible) distribution of pregnancies in general, white/black ratio was too, but at hospital in live birth had a white/black ratio of six, that means again there were more cases actually (inaudible) to this ratio in black than in white, which mirrors actually the findings published in the literature. Only one was a multiple pregnancy. There were four multiple parity cases, four cases had pregnancy‑related hypertension and four cases had fever, but without infection confirmed by laboratory tests. None of these factors really pop up as the one significant, the one we can chose and say this is the cause of cluster in 2001 and 2002.
If we like to summarize, we would say that hospital had the higher incidence of cluster of peripartum cardiomyopathy cases in 2001 and 2002 compared to the national estimated incidents. We are not able to identify, of course, a number where it’s too small to make any inferences, but that was a beginning and a good collaboration between Georgia Division of Public Health MCH epidemiology section and the hospital “A.” Findings were presented to OB/GYNs at hospital “A” in Georgia and we had very interesting discussion. We suggested that we may need to go further if like to identify what happened in 2001 and 2002, but we find out and we’re really pleased to find out that hospital went ahead and they developed their tracking system looking in 2003 if they have any cases of peripartum cardiomyopathy and they didn’t identify any other cases and we discussed a lot about having or not having peripartum cardiomyopathy registry. Of course, first question was why do we need a registry?
This is why we need that registry, to prospectively capture all women with peripartum cardiomyopathy, to better understand the incident, the prevalence, the causes, what are risk factors, to better evaluate even the therapeutic interventions. We haven’t made any decision, but hospital was very open and they were very happy that for first time we had such a good collaboration talking about some public health problems and sharing with hospital epidemiologists and with OB GYNs. Besides developing a registry or along with a registry, by having a strong education provided to all professionals involving the care of women of childbearing age, it may be of great help. Is this really necessary to have more timely recognition and diagnosis of peripartum cardiomyopathy cases and thus be able to develop better prudential strategies. Education about family planning.
Very important piece and we talk about this in all publication I found about peripartum cardiomyopathy, in many cases they discuss about having post partum educational family planning, well it may go beyond and we might need to be more careful how we address and what issues and what problems and what to discuss with all these women during our family planning visits, and again, remember subsequent pregnancy among women diagnosed with peripartum cardiomyopathy (inaudible) controversial is not proved, but this is why need to pay more attention. What lessons to learn. AS we all know and as we heard today, maternal death is the major indicator used to measure maternal health. It’s true, it’s easy measure from epidemiological prospective, but for every woman who dies, many suffer serious life threatening complication of pregnancy. Maternal morbidity is a public health problem.
Nearly 1.7 million women annually are affected by different maternal morbidity problems, and just to remind you, women comprise over half of the United States population and although they live longer than men, the rate of morbidity and the need of health services is much greater. In addition, women have the major healthcare decision makers in their families and thus, need access to timely and reliable information. Although, some advances have been made in addressing the problems that women face in the healthcare systems, service delivery continues to be fragmented for most of these women. Have we made any progresses? Yes. I’ll go back and I’ll look to healthy people 20/10 goal. We have a goal to reduce not just maternal death to reduce maternal morbidity too, and if you look to 16/5 post partum complication, I feel that peripartum cardiomyopathy could be easily included in this goal to decrease and understanding better and to have better prevention strategies. What public health implications this type of investigation had?
We understood by looking to this fatal health problem, such as peripartum cardiomyopathy, the need for improved long term follow up for women. All these women, the majority of women, we look into charts, women with peripartum cardiomyopathy were followed for probably six weeks with the recommendation if something happened in the following six months, to see a cardiologist. It may not be enough. We need to have successful intervention to reduce maternal morbidity in each effect. How can we do it? Only by having a strong collaboration between clinitions, hospital and public health professionals, and I felt really, while I was doing this investigation that this type of case reviews can be used in MCH as a tool to improve this collaboration and to develop better strategies. I would like to challenge you to think a little bit more not just to medical aspect of peripartum cardiomyopathy, to think why it’s important, how many women having peripartum cardiomyopathy go unrecognized, undiagnosed, how many of them become pregnant again and increase our pregnancy death. As just Meena mentioned the highest percentage is during pregnancy. How many of them probably have some problems before unrecognized, undiagnosed, we didn’t know, we didn’t follow up. Thank you very much for your attention.