CEMACH

Chapter 6: Early pregnancy deaths

James Neilson

Summary of key findings for 2003-2005

The deaths of fourteen women who died of causes directly attributed to complications arising in early pregnancy, before 24 completed weeks of gestation, who were reported to the Enquiry during this triennium are counted and discussed in this Chapter. These are shown in Table 6.1. These include ten deaths from ruptured ectopic pregnancies, one following a miscarriage, two deaths after termination of pregnancy, one illegal, and a death from ovarian hyperstimulation syndrome (OHSS).

An additional 18 women died from thromboembolism occurring before 24 weeks’ completed gestation whose deaths are counted and discussed in Chapter 2 - Thrombosis and thromboembolism. Apart from the death following miscarriage attributable to anaphylaxis to analgesia which is counted in this Chapter, there were another five Direct deaths from sepsis associated with miscarriage which are counted and discussed in Chapter 7 - Genital tract sepsis. The case of a woman who died from anaesthesia following surgery for an ectopic pregnancy is counted and discussed in Chapter 8.

Table 6.1  Numbers of Direct deaths in early pregnancy counted in this Chapter by cause; United Kingdom: 1988-2005.

Triennium Ectopic
pregnancy
Miscarriage Termination
of pregnancy
Other All deaths
counted in
Chapter 6
Counted as
deaths from
sepsis in
chapter 7
1985-87 11 4 1 0 16 0
1988-90 15 6 3 0 24 0
1991-93 9 3 5 0 17 0
1994-96 12 2 1 0 15 0
1997-99 13 2 2 0 17 5
2000-02 11 1 3 0 15 2
2003-05 10* 1 2 1 14 5

Note: Up to 1994-96, early pregnancy deaths were defined as occurring before 20 weeks of pregnancy.

Since 1997-99, 24 completed weeks of gestation has been used as the upper limit. Thus, direct comparisons with data from previous triennia may be misleading.

* A further woman who died from anaesthesia for an ectopic pregnancy is counted among the anaesthetic deaths.

 

The women who died

The ages of the women who died ranged between 18-40 years with a median age of 31. In contrast to the last Report, all but two mothers were White. Most women were in long term stable relationships and a few were health care workers.

Sub-standard care

As in previous years, the major challenge is to reduce the number of deaths from ectopic pregnancy, especially those associated with sub-standard care. Overall, 11 of the 14 deaths counted in this Chapter were assessed as having sub-standard care. Seven of the ten deaths from ectopic pregnancy were associated with sub-standard care, as were all the deaths from other causes counted in this Chapter.

Ectopic pregnancy

The maternal death rates from ectopic pregnancy for this and previous triennia are shown in Table 6.2.

Table 6.2     Numbers of deaths from ectopic pregnancies and rates per 100,000 estimated ectopic pregnancies; England and Wales 1988-1990 and United Kingdom: 1991-2005.

Triennium Total estimated
pregnancies
Total
estimated
ectopic
pregnancies 
Ectopic pregnancies per
1,000 pregnancies
Deaths from
ectopic
pregnancies
Death rate per 100,000
estimated ectopic
pregnancies
  Number Number Rate 95 per cent CI Number Rate 95 per cent CI
England and Wales
1988-90 2,880,814 24,775 8.6 8.5 8.7 15 0.52 0.31 0.86
United Kingdom
1991-93 3,141,667 30,160 9.6 9.5 9.7 9 0.29 0.15 0.55
1994-96 2,917,391 33,550 11.5 11.4 11.6 12 0.41 0.24 0.72
1997-99 2,878,018 31,946 11.1 11.0 11.2 13 0.45 0.26 0.77
2000-02 2,736,364 30,100 11.0 10.9 11.1 11 0.40 0.22 0.72
2003-05 2,891,892 32,100 11.1 10.9 11.1 10 0.35 0.19 0.64

Ten women died from ruptured ectopic pregnancies during the period of this Report. Another woman, whose death is counted in the anaesthetic Chapter (Chapter 8), died of the consequences of the anaesthetic she received for treatment of an ectopic pregnancy. Among the ten deaths counted in this Chapter was one corneal (interstitial) pregnancy. Another woman had a heterotopic pregnancy (combined intra- and extra-uterine pregnancies). She presented with diarrhoea and vomiting, as did three other women with (non-heterotopic) ectopic pregnancies in whom misdiagnosis occurred. Recent Reports have repeatedly emphasised the importance of diarrhoea and vomiting as a possible, atypical clinical presentation of ectopic pregnancy: 

A woman who had diarrhoea and vomiting also had vaginal bleeding, fainting and severe abdominal pain. She was known to be pregnant and was diagnosed as having gastroenteritis by a very junior gynaecologist in an Emergency  Department (ED), and discharged. She returned to the ED the following day with increased pain, having collapsed at home. She was found to be hypotensive and tachycardic by the nursing staff, given a very large amount of intravenous fluid for what was thought to be dehydration from ‘gastroenteritis’ and again discharged. Her haemoglobin was not checked. She was tachycardic throughout. She returned a few hours later in extremis. Autopsy revealed a large haemoperitoneum from a ruptured tubal pregnancy.

It is important to re-emphasise that early pregnancy plus fainting points to ectopic pregnancy until proven otherwise. The importance of observation of vital signs also needs to be remembered. There is a recurring theme in this Report of junior medical staff disregarding important, basic clinical signs – tachycardia, hypotension, rapid respiration – this is one such case.

There were also potentially avoidable deaths in women who were under the care of specialist gynaecological services:

One woman had a modestly raised β hCG level, no evidence of an intrauterine pregnancy and a pelvic mass on ultrasound compatible with ectopic pregnancy. While in hospital and awaiting a repeat β hCG assay she collapsed with a ruptured tubal pregnancy from which she could not be resuscitated.
Another woman was having medical treatment with methotrexate for a known ectopic pregnancy. Her β hCG levels rose rather than fell after one week of treatment, but it was not clear who saw these results. She became unwell, with diarrhoea and vomiting, and subsequently collapsed, but phone calls to the early pregnancy unit in the local hospital did not elicit the appropriate responses. Her GP found her shocked at home and she had a fatal cardio-respiratory arrest in the ambulance en route to hospital.

Many ectopic pregnancies in modern practice follow a benign clinical course which allows a more conservative approach to management. However, ectopic pregnancy remains a dangerous condition and these two cases are tragic reminders that deaths still occur. Medical treatment, in particular, must be based on strict adherence to protocols and immediate access to hospital services1.

Another woman underwent salpingectomy at laparotomy. Although she had a large haemoperitoneum, her vital signs were stable during the procedure. She had been extubated and was breathing, when she had a cardiac arrest. The cause of death is not certain but the possibility of a cardiac arrythmia resulting from (relatively) cold intravenous fluids needs consideration.
Miscarriage

One woman had an anaphylactic reaction to an opioid analgesic administered by a paramedic. She was known to have a probable anembryonic pregnancy and was, appropriately, awaiting a further ultrasound, one week after the first, for definitive diagnosis. She was given the opioid for severe pain associated with spontaneous miscarriage.

One of the women whose death is counted in Chapter 7, Sepsis, was initially and correctly thought to have had a septic miscarriage when seen in the Emergency Department, but the diagnosis was revised to probable ectopic pregnancy, despite a temperature of over 40°C.

Termination of pregnancy

The 50-year review of Confidential Enquiries in the last Report noted that ‘the most striking change during the past 50 years has been the disappearance of unsafe, illegal abortion as a cause of early pregnancy Direct deaths in this country. The first full working year of the Abortion Act was 1969, but it was not until 1982-4 that no deaths from illegal abortion were recorded’. There have been no further such deaths until now:

A woman who had not been long in the UK, but who spoke English well, requested termination of an early pregnancy because of her social circumstances. She was referred from a community clinic to a hospital clinic for a surgical termination. In the meantime, she attended an Emergency Department with rigors and severe low abdominal pain, where she was found to be markedly pyrexial and tachycardic. She declined admission and was discharged by a junior gynaecologist on a broad spectrum antibiotic for what was thought to be a urinary tract infection. She returned to the hospital the next day, still very pyrexial, and intravenous fluids and antibiotics were given after significant delay. Shortly after a pelvic ultrasound examination, which showed free fluid in her pelvis, she suffered a cardiac arrest from which she could not be resuscitated. A laparotomy was done in case she had an ectopic pregnancy. Although there was blood in the pelvis there was no ectopic pregnancy. Microbiological cultures grew an unusual organism, usually found in water, from several tissues, and careful inspection at autopsy showed evidence of unusual trauma in her genital tract.

It is not known why this woman opted for an unsafe abortion, having already engaged with NHS services. One can only speculate if there may have been cultural issues, or coercion. Globally, unsafe, illegal abortion is very common and is one of the leading causes of maternal death. With increasing migration, this sad case is an important reminder to clinicians that this can happen in the UK too. Another woman died following a legal termination of pregnancy:

A young woman who underwent a medical termination of pregnancy was subsequently readmitted to hospital with vaginal bleeding. Because the bleeding settled and because ultrasound examination showed no evidence of retained tissues, she was discharged without further treatment. She re-presented again with a pounding headache and breathlessness on exertion, and was found to be pale and tachycardic. She was given a large amount of intravenous crystalloid and colloid while a blood count was analysed and re-analysed. The haemoglobin concentration indicated severe anaemia. She had a cardio-respiratory arrest before a blood transfusion was started. Although she was resuscitated, she developed overwhelming pulmonary oedema. Profound cerebral damage became evident. Her death was attributed to severe anaemia from haemorrhage from retained products (found at autopsy), exacerbated by intravenous fluid infusion.

This case again illustrates a failure to recognise basic clinical signs and symptoms. The initial extremely low haemoglobin result should have prompted a re-inspection of the conjunctivae, urgent blood transfusion, and senior review rather than additional venepuncture and further delay whilst awaiting re-analysis in the haematology laboratory.

Ovarian hyperstimulation

There has been debate as to whether deaths from ovarian hyperstimulation should be included in this Report. As discussed in Chapter 1, in the view of the assessors these deaths should be reported to, and assessed by this Enquiry, as they occurred as a direct consequence of a woman trying to become pregnant. Further, they contain important lessons for the provision of infertility treatment, a growing area of medical intervention which is not currently subject to such critical review.

One death reported to the Enquiry was of a woman who had undergone ovarian hyperstimulation and intrauterine insemination. There are conflicting reports about whether a pregnancy test was positive, or not, at the time of her death. She was admitted with ovarian hyperstimulation syndrome and deteriorated over three days in the gynaecology ward before being transferred to Critical Care Unit, extremely ill. She did not receive thromboprophylaxis. Autopsy showed fluid in body cavities and patchy infarction throughout the body, including bowel. Women admitted to hospital with severe ovarian hyperstimulation syndrome should receive thromboprophylaxis.2

Box 6.1

Learning points: early pregnancy deaths

Clinicians in primary care and Emergency Departments, in particular, need to be aware of atypical clinical presentations of ectopic pregnancy and especially of the way in which it may mimic gastrointestinal disease. This needs to be taught to undergraduate medical and nursing students and highlighted in textbooks.

Fainting in early pregnancy suggests the possibility of ectopic pregnancy.

More conservative approaches to the treatment of ectopic pregnancy should not blind clinicians to the dangers of this condition. Laparoscopy or laparotomy should be undertaken without delay if there are clinical signs suggestive of tubal rupture. Medical treatment of ectopic pregnancy should be based on strict adherence to protocols, with women having immediate access to in-patient facilities if complications occur.

Illegal, unsafe abortion is common globally; it can occur in the UK.

Acknowledgements

This Chapter was seen and discussed with Dr Davor Jurkovic (London).

References

  1. RCOG Guideline No. 21. The Management of Tubal Pregnancy. London: Royal College of Obstetricians & Gynaecologists; May 2004. www.rcog.org.uk
  2. RCOG Green-top Guideline no. 5. The Management of Ovarian Hyperstimulation Syndrome. London: Royal College of Obstetricians & Gynaecologists: September 2006. www.rcog.org.uk