The long term implications of fertility therapy for the health of women

Cardiovascular disease mortality is not increased in women undergoing IVF, but vigilant surveillance is nonetheless required

Since the first birth facilitated by in vitro fertilisation (IVF) in 1978, the use of IVF has spread across the world and more than eight million babies have been born.1 The use of medically assisted reproduction, including IVF, currently accounts for about 7% of all births in Australia, and the annual number of births involving assisted reproductive technology has increased by 55% over the past ten years.2 IVF is a recognised medical technology with a reasonable success rate in younger women, partially funded by Medicare, widely available across Australia, and well accepted by the Australian community.

Modern IVF technology was optimised in Melbourne, where for several years Australia was a world leader in successful IVF programs. As laboratory and clinical practices improved, the focus increasingly moved to enhancing peri‐ and neonatal outcomes. There were initially concerns that fertility interventions involving hormonal preparations, ovarian puncture, and the extra‐uterine growth of embryos might lead to unforeseen poor health outcomes for women and children. Melbourne researchers were among the first to investigate the long term health of women in IVF programs, publishing several reports about breast cancer and a variety of other health outcomes that were reassuring,3,4 and Alison Venn and her colleagues were among the first to explore the concept of the “healthy patient” effect in the context of fertility treatment.3

In this issue of the MJA, Yiallourou and colleagues report further comforting evidence from Melbourne, this time with respect to all‐cause and cardiovascular disease mortality among women who have been stimulated with hormones as part of the IVF procedure.5 In their cohort study of more than 44000 women who registered for fertility treatment during 1975–2018, the authors assessed outcomes for those who ultimately received fertility drugs or did not, analysing linked Monash IVF clinical registries and National Death Index mortality data. All‐cause and cardiovascular disease mortality were each lower for women registered for fertility therapy than for women of similar age in the general population. Further, all‐cause mortality was similar for registered women who underwent IVF and for those patients with infertility who registered at the clinic but did not receive specific therapy. Cardiovascular disease mortality was lower among women who underwent IVF than for those not exposed to the hormones used in an IVF cycle. All‐cause mortality also tended to be lower among women from areas of lower socio‐economic disadvantage.5

These findings will reassure women and their doctors making decisions about IVF treatment, and they accord with reports from elsewhere,6 although some concerns remain with respect to ovarian tumours.7 Retrospective studies have clear disadvantages in terms of limitations in the collection of pertinent data and personal histories to inform assessment of the underlying disorder and the metabolic status of women seeking fertility treatment. In the early days of IVF, much of the information we need to refine our understanding of long term outcomes was not available. In addition, the cost of IVF largely restricted its use to women with more secure financial backgrounds, which may explain the healthy patient effect.3 Attitudes regarding the advisability of IVF for women with high body mass index (BMI) or significant underlying disease may also have led to the selection of healthier patients. It is essential that fertility clinicians now collect more information on the genetic, metabolic, and medical histories of the women with infertility they see.

Many women with infertility do not need IVF and consequently may not be represented in the study by Yiallourou and colleagues. Of note is the polycystic ovary syndrome (PCOS), in which metabolic syndrome, high BMI, and hypertension are strongly associated with increased prevalence of insulin resistance.8,9 These women respond well to lifestyle interventions and ovulation‐inducing agents that lead to weight loss and ovulation,10 although many may need to undergo IVF for other reasons, including male factor infertility or age‐related reduced fertility. The prevalence of subclinical cardiovascular disease is higher among women with PCOS, including changes in flow‐mediated dilation, arterial stiffness, higher coronary artery calcium scores, carotid intima media thickness, and visceral and epicardial fat.11 Although it included women with infertility of ovarian aetiology, women with PCOS were probably underrepresented in the study by Yiallourou and colleagues.

Modern fertility treatment differs in certain respects from the management experienced by the women in this historical study. More recent practice prefers frozen embryos to fresh transfers, and there is evidence that using frozen embryos during hormone replacement cycles is associated with higher risks of hypertensive disorders during pregnancy, which are in turn risk factors for future cardiovascular disease.12 This problem is less marked with natural or induced ovulation cycles, which are now increasingly used. Further, the use of donor gametes has also increased, and transferring an embryo grown from a donor egg is associated with increased risk of high blood pressure during pregnancy.13

While IVF technology is becoming more efficient and more accessible, ongoing surveillance of the long term health of women receiving these treatments is still required.