Recent reports on maternal mortality from the UK covering the years 2013–2015¹ and Australia for the years 2006–2010 and 2012–2014,^{2 3} highlight slight differences in the cause of maternal mortality in the two countries. Although the populations and demographics are significantly different, reviewing the reports together gives additional insights into the current causes of maternal deaths. It is important to stress that maternal deaths are rare in both the UK and Australia. Accordingly, the data should be interpreted with caution due to the relative small numbers, which can fluctuate, and the different methods of data collection.

Maternal deaths in the UK are reported to Mothers and Babies: Reducing Risk through Audits and Confidential Enquiries across the UK (MBRRACE-UK), who run the national program conducting surveillance into the causes of maternal and perinatal deaths. It is based at the National Perinatal Epidemiology Unit at the University of Oxford. The information provided to MBRRACE-UK is from the staff caring for the women concerned, or through other sources including coroners, Scottish procurators, pathologists and media reports. Identification of deaths is cross-checked with national records, such as death certificates. The importance of these additional sources is highlighted by the fact that the use of death certificates alone only identified 110 cases out of the final total of 202 maternal deaths.

In Australia, the data sources and the quality of maternal death reporting vary by state and territory. The initial process for data collection is through State and Territory Maternal Mortality Committees (STMMC). The data is obtained from a number of sources, including clinicians, midwives, coronial reports, and the Registry of Births, Deaths and Marriages. The data from the STMMCs is passed on to the Australian Institute of Health and Welfare, for compilation into a National Maternal Mortality Data Collection.

In Western Australia, only limited data on maternal deaths are available, due to the health and privacy legislation in that state. In addition, the Australian report for 2006–2010 noted that ‘for some states and territories, the maternal mortality committees and subcommittees were not active for periods during 2006–2013’. As an example, there was no committee active in the Northern Territory between 2006 and 2014. The formula for the calculation of the frequency of maternal deaths and the terminology differ slightly in the two countries. In the UK, the maternal mortality rate is the number of maternal deaths divided by the number of ‘maternities’ (women who were at least 24 weeks pregnant), multiplied by 100,000. In Australia, the maternal mortality ratio is the number of maternal deaths divided by women who gave birth (who were at least 20 weeks pregnant, or delivered a fetus weighing at least 400g), multiplied by 100,000.

In the years 2013–2015, the maternal mortality rate in the UK was calculated as 8.76 per 100,000 maternities. In the two Australian reports, the overall maternal mortality ratio was 6.83 per 100,000 births. However, it is generally accepted that the UK processes of identifying all maternal deaths within the country is more comprehensive than the systems in place in other countries.

In June 2015, the population of the UK was estimated as 65,110,000,⁴ and from 2013–2015, there were 2,305,920 maternities. In Australia, as of June 2015, the population was estimated to be 23,781,200.⁵ In Australia, apart from the reports mentioned above, an additional report⁶ was released covering the years 2008–2012. As these reports overlap, it was not possible to ascertain the exact details of the causes of death for the year 2011. Accordingly, a decision was made to use the two Australian reports covering the years 2006–2010 and 2012–2014. The total number of women giving birth in Australia in these two time periods was 2,368,540, which allows a reasonable comparison with the UK experience.

In the UK, 240 pregnant women died between 2013 and 2015. The deaths of 38 of these women were considered to be coincidental to the pregnancy, such as from motor vehicle accidents, leaving 202 maternal deaths. In the two Australian reports, there was a total of 198 pregnant women who died. The cause of death was considered to be coincidental in 36 cases, leaving 162 maternal deaths.

Maternal deaths are divided into a number of categories (Table I), as outlined by the World Health Organization classification (WHO, 1992).

Table 1. Definitions of maternal deaths (WHO, 1992).

Late pregnancy deaths are included in the UK report, but in Australia late pregnancy deaths are not identified.

The demographics of pregnant women in each country is outlined in the reports. In both countries, approximately three-quarters of the women who died were born in the country of study. The UK’s immigrant population comes largely from South Asia, Africa and Eastern Europe, while Australia’s immigrant population comes mainly from New Zealand and Asian countries. In the UK, women from Jamaica, Pakistan and Bangladesh were over-represented among the women who died during pregnancy. However, in Australia, the maternal mortality ratio of women who were born overseas is actually lower than that of women born in Australia. On the other hand, the maternal mortality rate among Aboriginal and Torres Strait Islanders is approximately three times that of non-indigenous Australian women, largely due to cardiac conditions and sepsis.

In the UK, the maternal mortality rate from 2013–2015 was found to be higher among older women, those living in the most deprived areas, and women from ethnic minority groups, especially women of African descent. Fifty-three per cent of the women who died were either obese or overweight. In Australia, the incidence of maternal death was higher for women over 35 and under 20 years of age, and for Aboriginal women. In the Australian report of 2012–2014, of the 33 women who died and whose BMI was calculated, 70 per cent were either overweight or obese.

There are differences in the classification of suicide. In 2012, WHO recommended classifying all maternal suicides as direct maternal deaths.⁷ This classification was adopted in the UK report. However, the Australian National Maternal Mortality Advisory Committee, with advice from the Royal Australian and New Zealand College of Psychiatrists, came to the conclusion that puerperal psychosis is rare. The committee concluded that suicides where there was evidence of a pre-existing mental health disorder should be regarded as indirect deaths, whereas suicides in the setting of no previously diagnosed mental health illness should be regarded as direct deaths. Accordingly, suicides in Australia associated with a previous psychiatric illness are classified as indirect deaths, while suicides due to the development of a puerperal psychosis are classified as direct deaths.

In addition, in Australia, maternal deaths associated with external events, such as homicide, are classified as either an indirect or incidental death, depending on the circumstances. A homicide occurring in a situation of domestic violence is classified as an indirect death, while homicide occurring outside a domestic setting is classified as a coincidental death. In the UK report, homicide is classified only as a coincidental death. In the Australian reports, women who died from haemorrhage are categorised as dying from either obstetric or non-obstetric haemorrhage. A non-obstetric haemorrhage is bleeding from a site other than the uterus. A cerebral haemorrhage associated with severe pre-eclampsia is categorised as a direct death, while a cerebral haemorrhage without associated hypertension is considered an indirect death. In the UK reports, the term non-obstetric haemorrhage is not used. Patients who died from non-obstetric haemorrhages in the UK, such as from a ruptured splenic artery aneurysm, are included in the category of ‘Other indirect causes’.

Among indirect deaths, the UK report has a category entitled ‘Neurological conditions’, which is not used in the Australian reports. These neurological cases comprise mainly patients who died following strokes or epilepsy. Deaths due to epilepsy in Australia are included under ‘Other causes’. The UK report also includes chapters that review some of the conditions that can cause maternal death. Some of these chapters also include data from the Republic of Ireland. The causes of direct and indirect maternal deaths in the UK and Australia are listed in Table 2.

Table 2. Causes of direct and indirect maternal deaths in the UK and Australia.
* In the Australian reports, 12 cases of suicide were classified as indirect deaths. All suicides in the UK report were considered direct deaths.
# The haemorrhage resulted from blunt trauma to the uterus.

Results

The traditional ‘big five’ causes of direct maternal deaths (thrombo-embolism, obstetric haemorrhage, hypertension, sepsis and amniotic fluid emboli) continue to make up the majority of cases of maternal deaths in both countries. In the Australian report, 57 out of 71 direct deaths (80 per cent) were from these causes, while in the UK, they accounted for 68 out of 88 deaths (77 per cent).

Thrombo-embolism is the leading cause of direct deaths in both countries. In the UK, there were 26 deaths between 2013 and 2015. In the two Australian time periods, there were 15 deaths. The recognised risk factors, such as obesity, older maternal age, smoking and operative delivery, were prevalent in the women who died from this condition in both countries.

In the UK, between 2013–2015, 21 women died from obstetric haemorrhage. In the Australian reports, 13 such deaths were recorded. The UK report included a chapter⁸ that reviewed the details of deaths from obstetric haemorrhage in both the UK and the Republic of Ireland. The breakdown of 22 cases was: three deaths from placental abruption; nine deaths from placenta praevia/accreta; nine deaths from postpartum atony (of which five were post-caesarean section); and one death from genital tract trauma.

In the two Australian reports, there was a total of ten deaths from hypertensive disorders, such as eclampsia. There were only three deaths from hypertensive causes in the UK. Hypertensive deaths are now much reduced in the UK. From 1985–1987, there were 27 such cases.⁹

In both the UK and Australia, there were 14 deaths following sepsis, of which four were indirect deaths in both countries. Three of the Australian deaths were due to H1N1 influenza (swine flu), and there was one confirmed case among the British deaths. Most cases of direct obstetric sepsis occurred postpartum. Group A beta haemolytic streptococcus was the most common pathogen associated with maternal mortality, with five cases in Australia and two in the UK.

There were 11 deaths from amniotic fluid emboli in the Australian reports, and eight in the UK. Among the UK cases were two women undergoing induction because of intra-uterine deaths, who received excessive doses of misoprostol and collapsed following hyperstimulation. Two multiparous women, who were induced to establish labour, also developed hyperstimulation after the use of prostaglandins and died from amniotic fluid emboli.

The number of suicides in the UK report was 12; in the Australian reports, it was 14. However, in the UK, there were 46 suicides between six weeks and 12 months postpartum, almost four times as many as during the pregnancy. Many of these patients had pre-existing psychiatric histories. It appeared that many women with postpartum psychiatric conditions that placed them at risk of suicide were not recognised, and were not ‘owned’ by any one group of the woman’s healthcare team.

In 2013–2015, eight women in the UK with epilepsy died during pregnancy or in the immediate postpartum period. A further five women with epilepsy died between six weeks and one year after delivery. The two Australian reports recorded four epilepsy deaths.

The UK report included a chapter¹⁰ on sudden unexplained death in epilepsy (SUDEP). SUDEP is defined as the sudden and unexpected death during pregnancy of a woman with epilepsy, without a toxicological or anatomical cause of death detected during pregnancy. The cause is not understood. Among the UK cases were several women who discontinued their anti-epileptic medication without specialist advice. Another risk factor for a pregnant woman with epilepsy is drowning. In these reports, two women in Australia drowned in the bath and one such case was reported in the UK.

One of the least appreciated causes of maternal death is from a ruptured splenic artery aneurysm, which is not on the radar of many obstetricians. In Australia, from 2006–2010, five women died from this cause. The UK report¹¹ noted that in the UK and Republic of Ireland, nine women died from this condition. Most cases are asymptomatic prior to rupture. In many cases, the accompanying acute abdominal pain was misdiagnosed. Where patients had already collapsed, emergency staff were often focused on other diagnoses, such as a pulmonary embolism or an amniotic fluid embolism. A literature review of reported cases of splenic artery aneurysm rupture in pregnancy, published in 2009,¹² found that early involvement of a general or vascular surgeon reduced mortality. An earlier report from 2003 argued that ‘it is therefore important to increase awareness of this condition, so that obstetricians and other front-line staff can entertain the diagnosis of a ruptured splenic artery aneurysm in any pregnant woman who presents with severe upper abdominal pain’.¹³ That sentiment remains valid today.

In developed countries, cardiovascular disease is currently the most common cause, not just of indirect deaths, but of maternal deaths overall. In the UK, between 2013–2015, 51 women died. In the Australian reports, there were 25 deaths. The main contributing factors appear to be increasing maternal age, obesity and the increasing number of women with congenital heart defects who, following surgical treatment, survive into adult life, and who subsequently become pregnant. Both reports recommended that pregnant women with pre-existing cardiac conditions should be reviewed by physicians with particular experience in treating cardiac conditions in pregnancy.

Most early deaths were from ruptured ectopic pregnancies. There were four deaths in the UK report. In Australia, there were five deaths due to ectopic pregnancies and one death following complications associated with termination of pregnancy. The UK report also listed a small number of rare causes of maternal death from medical causes that most obstetricians would probably not encounter during their career. These include deaths from cystic fibrosis, sickle cell anaemia, systemic lupus erythematosus, Addison’s disease, pancreatitis and thombo-cytopenic purpura. Four women in the UK from ethnic minorities died from haemophagocytic lympho-histiocytosis. Managing these unfamiliar conditions would be problematic for most obstetricians. In the face of atypical symptoms and disorders, the appropriate management would be to obtain an opinion from a physician experienced in medical conditions in pregnancy.

The two Australian reports recorded 16 deaths among the Indigenous population. This was almost ten per cent of the overall total of 162 deaths. The Aboriginal community constitutes approximately three per cent of the Australian population.

Ninety-three of the women in the UK report were delivered by caesarean section. Of these operations, 35 (38 per cent) were performed as perimortem procedures. Of the babies born following perimortem caesarean sections, 14 survived. In the Australian report for 2012–2014, 23 women were delivered by caesarean section. Of these, six (26 per cent) were perimortem operations and four babies survived.

Conclusions

Beyond the statistics, it is stating the obvious to say that a maternal death is a catastrophe, leaving behind a devastated family and, often, other children who no longer have a mother.

In the UK, the Confidential Enquiries into Maternal Deaths in England and Wales began in 1952. Since that time, maternal death rates in the UK have fallen from approximately 90 per 100,000 women giving birth¹⁴ to around nine per 100,000 currently.

Maternal mortality data was first recorded in Australia for the years 1964–1966.¹⁵ In that report, the maternal mortality rate for direct deaths was reported as 30.3 per 100,000 confinements. In the 2012–2014 report, using the same definitions, the incidence was 6.8 per 100,000 women giving birth.

The reduced numbers of maternal deaths in the UK and Australia over the last 50 years is, of course, very welcome. However, the decrease is predominantly due to a reduction in direct deaths. There has not been a similar reduction in indirect deaths. To reduce the number of indirect deaths, multidisciplinary care coordinated across primary and secondary healthcare teams remains a priority. A similar situation applies to deaths in late pregnancy. In the UK, from 2013–2015, 326 women died between six weeks and 12 months postpartum, over 60 per cent more than the 202 women who died during the pregnancy. As mentioned above, almost four times as many women in the UK committed suicide in late pregnancy (n=46), as during the pregnancy (n=12). Over the last half-century, obstetricians have understandably focused on addressing the ‘big five’ direct obstetric deaths. The ongoing challenge is to direct the same energy to addressing indirect and late pregnancy deaths.

Acknowledgements

I would like to gratefully acknowledge the assistance of Prof Marian Knight of the National Perinatal Epidemiology Unit/MBRRACE-UK.

In Australia, I would like to acknowledge Prof Georgina Chambers of the National Perinatal Epidemiology and Statistics Unit, as well as all editors, authors and contributors to the maternal mortality reports in the UK and Australia.

I would also like to thank Prof Roger Pepperell for reviewing an earlier draft of this article.

1. Knight M, Nair M, Tuffnell D, et al (Eds.) on behalf of MBRRACE-UK. Saving Lives, Improving Mothers’ Care – Lessons learned to inform maternity care from the UK and Ireland Confidential Enquiries into Maternal Deaths and Morbidity 2013–15. Oxford: National Perinatal Epidemiology Unit, University of Oxford 2017
2. Johnson S, Bonello MR, Li Z, Hilder L, Sullivan EA. 2014. Maternal deaths in Australia 2006– 2010, Maternal deaths series no. 4. Cat. no. PER 61. Canberra: AIHW.
3. Australian Institute of Health and Welfare 2017. Maternal deaths in Australia 2012–2014. Cat. no. PER 92. Canberra: AIHW.
4. Office of National Statistics. Population estimates for UK, England and Wales, Scotland and Northern Ireland: mid 2015. Available at: www.ons.hov.uk/ population and community.
5. Australian Demographic Statistics, June 2015. Available at: www.abs.gov.au/absstats/abs@nsf/mf/3101.0.
6. AIHW: Humphrey MD, Bonello MR, Chughtai A, et al. Maternal deaths in Australia 2008–2012. 2015. Maternal deaths series no. 5. Cat. no. PER 70. Canberra: AIHW.
7. The WHO application of ICD-10 to deaths during pregnancy, childbirth and puerperium: ICD MM. WHO, 2012.
8. Knight P & Paterson-Brown S on behalf of the Haemorrhage and AFE Chapter-writing Group. Messages for care of women with haemorrhage or amniotic fluid embolism. In Knight M, Nair M, Tuffnell D, et al (Eds.) on behalf of MBRRACE-UK. Saving Lives, Improving Mothers’ Care: Lessons learned to inform maternity care from the UK and Ireland Confidential Enquiries into Maternal Deaths and Morbidity 2013–15. Oxford: National Perinatal Epidemiology Unit, University of Oxford 2017:74-81
9. Nair M & Knight M. Maternal Mortality in the UK: Surveillance and Epidemiology. In Knight M, Nair M, Tuffnell D, et al (Eds.) on behalf of MBRRACE-UK. Saving Lives, Improving Mothers’ Care: Lessons learned to inform maternity care from the UK and Ireland Confidential Enquiries into Maternal Deaths and Morbidity 2013–15. Oxford: National Perinatal Epidemiology Unit, University of Oxford 2017:6-21.
10. Kelso A, Wills A, Knight M on behalf of the MBRRACE-UK Neurology Chapter-writing Group. Lessons on epilepsy and stroke. In Knight M, Nair M, Tuffnell D, et al (Eds.) on behalf of MBRRACE-UK. Saving Lives, Improving Mothers’ Care: Lessons learned to inform maternity care from the UK and Ireland Confidential Enquiries into Maternal Deaths and Morbidity 2013–15. Oxford: National Perinatal Epidemiology Unit, University of Oxford 2017:24-36.
11. Knight M & Nelson-Piercy C on behalf of the Medical and Surgical Chapter-writing Group. Lessons for the care of women with medical and general surgical disorders chapter writing group. In Knight M, Nair M, Tuffnell D, et al (Eds.) on behalf of MBRRACE-UK. Saving Lives, Improving Mothers’ Care: Lessons learned to inform maternity care from the UK and Ireland Confidential Enquiries into Maternal Deaths and Morbidity 2013–15. Oxford: National Perinatal Epidemiology Unit, University of Oxford 2017:50-58.
12. Ha JF, Phillips M, Faulkner K. Splenic artery aneurysm rupture in pregnancy. European Journal of Obstetrics Gynecology and Reproductive Biology 2009;146:133-7.
13. Selo-Ojeme DO, Welch CC: Review: spontaneous rupture of splenic artery aneurysm in pregnancy. Eur J Obstet Gynecol Reprod Biol. 2003;109(2):124-7.
14. Ministry of Health. (1957). Report on Confidential Enquiries into Maternal Deaths in England and Wales 1952–1954.
15. NHMRC (National Health and Medical Research Council) 1966. Maternal deaths in the Commonwealth of Australia, 1964–1966. National Health and Medical Research Council. Canberra: NHMRC.