Adverse outcomes from IVFBMJ 2011; 342 doi: http://dx.doi.org/10.1136/bmj.d436 (Published 28 January 2011) Cite this as: BMJ 2011;342:d436
- Susan Bewley, consultant obstetrician1,
- Lin Foo, academic foundation trainee2,
- Peter Braude, professor of obstetrics and gynaecology3
- 1Kings Health Partners, Women’s Services, Guy’s and St Thomas’ NHS Foundation Trust, London SE1 7EH, UK
- 2Royal London Hospital, Barts and London NHS Trust, London, UK
- 3Kings College London, Division of Women’s Health, St Thomas’ Hospital, London UK
The fundamental measure of women’s health and maternity services—maternal mortality—has been creeping up for two decades in the United Kingdom,1 and more recently in the United States, Denmark, Austria, Canada, and Norway.2 The first published case report of a maternal death related to in vitro fertilisation (IVF) predicted that maternal morbidity and mortality would rise with increasing use of assisted reproductive technologies as a result of pregnancies at an older age, multiple pregnancies, and pre-eclampsia.3
The most comprehensive work on this topic to date, which used multiple sources of information, including 24 years of the Dutch maternal mortality database,4 examined the late effects of ovarian hyperstimulation syndrome and other deaths related to IVF to derive estimates of mortality. With 23 pregnancy related deaths and conservative assumptions, the study showed convincingly that overall mortality in IVF pregnancies was higher than the maternal mortality rate in the general population in the Netherlands. Compared with the national maternal mortality rate (12.8/100 000), it was estimated that 6/100 000 were directly related to IVF and 42.5/100 000 were related to IVF pregnancy. No other sizeable studies have directly tackled this question.
The NHS Choices website emphasises that maternal death is rare,5 and it found the lack of deaths from ovarian hyperstimulation syndrome after 1997 in the Dutch study reassuring, notwithstanding the continued occurrence of such deaths in the UK. The last Confidential Enquiry into Maternal Death recorded four deaths directly related to IVF via ovarian hyperstimulation syndrome and three deaths related to multiple pregnancy after IVF.1 Thus, more deaths were related to ovarian hyperstimulation syndrome than to abortion (two) despite many fewer procedures (for example, 48 829 IVF cycles v 198 500 abortions were performed in the UK in 2007).6 IVF associated maternal deaths may be underestimates, because record linkage is not allowed by the Human Fertilisation and Embryology Act, and some women do not disclose assisted reproduction or egg donation to maternity services. The assessors noted “important lessons for the provision of infertility treatment . . . which is not currently subjected to such critical review,” and that “many of the women who died from pre-existing diseases or conditions . . . did not receive any pre-pregnancy counselling. In particular this was the case for several women with major risk factors who received treatment for infertility.”1
Women undergoing assisted reproduction should be healthier than average, as a result of self selection and medical screening. A “healthy patient effect” has been observed for all cause death,7 suggesting that the unhealthiest women in the population are deterred from undergoing assisted reproduction. The global industry has operated on an assumption that this ensures safer pregnancies, but the high maternal mortality rate suggests otherwise. Age is associated with poorer pregnancy outcomes across the board,8 women who undergo assisted reproduction are more likely to enter pregnancy with a chronic condition,9 ovum donation increases the risk of pre-eclampsia, and all complications barring postmaturity are increased in multiple pregnancy.10
Paradoxically, in the developing world women die from complications of excess fertility and unsafe abortion, yet in post-industrial societies, infertility treatment now poses a higher risk for maternal death. Even though IVF pregnancy in the UK is still very safe, deaths may reflect a far greater burden of severe adverse morbidity. Long term analysis shows increasing numbers of women are having infertility treatment even later in life (mean age (peak age range) in 1992 v 2004: 33 (31-34) v 34.8 years (36-40)).11 Women with severe ovarian hyperstimulation syndrome, complicated pregnancy, and extremely premature babies usually end up in NHS hospitals because of the expense of looking after them privately.
Aside from public health initiatives for avoiding infertility, what would temper the dangers to patients? Better information is unlikely to deter many women, but they are entitled to know the risks before starting fertility treatment, particularly when using centres in less regulated jurisdictions. Practitioners of assisted reproduction need to identify high risk women more efficiently: those with pre-existing medical conditions have ample opportunity for preconceptual counselling, yet many are alerted to potential complications only when pregnant. It is important to communicate with women’s previous doctors and heed the advice of clinical teams that provide pregnancy care. Deaths are too rare to identify specific substandard procedures or centres, but market pressures can conflict with best and safest practice. Obstetricians and neonatologists would welcome more single embryo transfer as the norm to prevent death and disability. Ovarian hyperstimulation syndrome may arise from a reluctance to abandon cycles because of financial payment structures, lack of appropriate cycle monitoring, and competition within the Human Fertilisation and Embryology Authority’s “league table.”
The authors of the Dutch study advise that all iatrogenic complications and lethal outcomes after IVF should be reported so lessons can be learnt, and this would be bolstered by formal register linkages. One small step would be for ovarian hyperstimulation syndrome to become notifiable. More stringent attention to stimulation regimens, preconceptual care, and pregnancy management is needed so that maternal death and severe morbidity do not worsen further.
Cite this as: BMJ 2011;342:d436
Competing interests: All authors have completed the Unified Competing Interest form at www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author) and declare: no support from any organisation for the submitted work; no financial relationships with any organisations that might have an interest in the submitted work in the previous three years; SB is a member of the NICE Infertility Guideline Development Group. PB is a member of the HFEA Clinical and Scientific Advisory Committee and chaired the Expert Committee on Multiple Births after IVF.
Provenance and peer review: Not commissioned; externally peer reviewed.