Late maternal deaths: a neglected responsibility.

2072 www.thelancet.com Vol 387 May 21, 2016 Maternal mortality, no matter when and where it occurs, results in sequelae that extend beyond the loss of the life of a single woman. The death of a mother adversely aff ects the ability of her family to survive and thrive, especially under conditions of socioeconomic deprivation. Documentation of data on maternal mortality has helped identify areas of socioeconomic inequity and serves as a barometer of a society’s health system. Avoidable deaths from pregnancy complications occur on a global scale, with the greatest burden of mortality among women in low-to-middle-income countries. Most countries record maternal death only up to 42 days post partum because of the assumption that avoidable death in pregnant women occurs during pregnancy or shortly thereafter. Although limited, the available data suggest otherwise. Globally, there are more postpartum and late maternal deaths from direct and indirect obstetric causes than maternal deaths during pregnancy. Post-partum and late maternal deaths have not declined in the past decade, whereas deaths during pregnancy and the puerperium have. Estimates of post-partum and late maternal deaths are likely to be underestimated because late mortality has been variably specifi ed and either counted or discounted in reporting systems used in the recent past. This problem was highlighted in reports by WHO and the US Centers for Disease Control and Prevention on maternal mortality surveillance. Interestingly, the introduction of a check box indicating pregnancy in the past year before death on national death certifi cates in some US states led to an increase in reported late maternal deaths in those states. Currently, physicians can be unclear about what counts as a late maternal death. The WHO Working Group on Maternal Mortality has suggested International Classifi cation of Diseases (ICD) coding principles that defi ne maternal death up to a year after delivery from causes directly related to pregnancy or indirectly precipitated by the eff ects of pregnancy on underlying diseases; coincidental deaths are not included. The ICD10 code makes it obligatory to document the occurrence of pregnancy within a year of the death of any woman. These principles and the system of reporting have been tested against existing databases and reviewed by professional bodies, including the International Federation of Gynecology and Obstetrics, the American College of Obstetricians and Gynecologists, and the UK’s Royal College of Obstetricians and Gynaecologists. However, the use of ICD10 coding of late maternal death is generally not applied globally, and so far there is no large data series outlining the specifi c causes leading to late maternal death on a global scale. What is known is that late maternal deaths fall into four main categories: cardiovascular causes, thromboembolism, cancer, and suicide. Pregnancy can trigger cardiovascular disease (eg, hypertensive disorders leading to heart failure), aggravate underlying disease (eg, rheumatic heart disease, congenital heart disease, or pulmonary arterial hypertension), or cause specifi c diseases, such as peripartum cardiomyopathy (PPCM). The latter disease typically presents only 1–3 months post-partum, with mortality rates of about 10–25% within 6 months after diagnosis. PPCM is the largest contributor to cardiovascular maternal death in South Africa, but because it often occurs outside the 42-day post-partum period women who die from PPCM are not usually reported as late maternal deaths in South Africa and elsewhere. Thus, epidemiological estimates of the burden of disease causing maternal mortality are skewed by the exclusion of deaths caused by PPCM. This situation is of concern because no matter how late these deaths occur, they are related to pregnancy. Maternal deaths related to mental disorders have recently been assessed as part of the Confi dential Enquiry into Maternal Deaths in the UK and Ireland. Almost a quarter of maternal deaths that occurred between 6 weeks and 1 year after pregnancy in 2011–13 in the UK and Ireland were due to psychiatric Late maternal deaths: a neglected responsibility 8 Cauchemez S, Besnard M, Bompard P, et al. Association between Zika virus and microcephaly in French Polynesia, 2013–15: a retrospective study. Lancet 2016; published online March 15. http://dx.doi.org/10.1016/ S0140-6736(16)00651-6. 9 Schuler-Faccini L, Ribeiro EM, Feitosa IM, et al. Possible association between Zika vírus infection and microcephaly—Brazil, 2015. MMWR Morb Mortal Wkly Rep 2016; 65: 59–62. 10 Victora CG, Schuler-Faccini L, Matijasevich A, Ribeiro E, Pessoa A, Barros FC. Microcephaly in Brazil: how to interpret reported numbers? Lancet 2016; 387: 621–24. 11 Brasil P, Pereira JP Jr, Gabaglia CR, et al. Zika virus infection in pregnant women in Rio de Janeiro—preliminary report. N Engl J Med2016; published online March 4. DOI:10.1056/NEJMoa1602412.