Acupuncture with in vitro fertilisation

Abstract

May increase birth rates, but guidelines should await the results of ongoing trials

According to the European Society of Human Reproduction and Embryology, more than 300 000 treatment cycles of in vitro fertilisation and intracytoplasmatic sperm injection are performed each year in Europe.¹ Because in vitro fertilisation can affect physical and mental health, professionals delivering this treatment are trying to optimise the technology to increase birth rates.

About 90% of all assisted reproductive treatment cycles result in the transfer of at least one embryo, but only about 25% of all cycles end in implantation of the embryo and live birth. The main factor limiting the success of treatment is failure of implantation and not the lack of human embryos for transfer. Initiatives to improve rates of implantation have had varying success. In the accompanying systematic review, Manheimer and colleagues report improved pregnancy rates with in vitro fertilisation when acupuncture accompanies embryo transfer.²

As far as we are aware, this is the first systematic review and meta-analysis of the success rates of in vitro fertilisation with adjuvant acupuncture. It included seven methodologically sound randomised controlled trials with little clinical heterogeneity that studied 1366 women having in vitro fertilisation between 2002 and 2006. Adjuvant acupuncture at the time of embryo transfer significantly increased the rates of clinical pregnancy (odds ratio 1.65, 95% confidence interval 1.40 to 2.49) and live birth (1.91, 1.39 to 2.64). The number needed to treat with adjuvant acupuncture to achieve one more pregnancy was 10 and to obtain one more live birth this number was nine. A subgroup analysis restricted to the three studies with highest clinical pregnancy rates in the control groups, however, found no significant benefit of acupuncture (1.24, 0.86 to 1.77).

The pooled odds ratios on pregnancy rates after in vitro fertilisation with adjuvant acupuncture are higher than reported odds ratios for drugs or other procedures given to enhance the success of this treatment. For example, gonadotrophin preparations with luteinising hormone activity versus preparations without such activity gave a relative risk of clinical pregnancy of 1.17 (1.03 to 1.34)³; the use of assisted hatching (which prepares the embryos before transfer) gave an odds ratio of clinical pregnancy of 1.63 (1.27 to 2.09)⁴; and transfer of day 5 embryos (blastocyst stage) versus day 3 embryos (cleavage stage) gave an odds ratio for clinical pregnancy of 1.27 (1.03 to 1.55).⁵

Manheimer and colleagues have extensively dealt with and discussed the limitations of their meta-analysis and the possibility of biased results. We consider the results to be just as solid as any other meta-analysis of adjuvant treatment or medical procedures in assisted reproduction. However, although meta-analyses provide the highest level of scientific evidence, caution should be used when using data from small meta-analyses with a risk of publication bias to make daily clinical decisions.

The current meta-analysis was based on seven randomised controlled trials, and only three relatively small trials included a sham control group. Results of meta-analyses should always be interpreted in terms of how biologically plausible they are. In this instance, the biological mechanism is difficult to explain. Adjuvant acupuncture was given immediately before or immediately after embryo transfer. The effects of acupuncture are therefore most likely to involve uterine contractility rather than uterine receptivity. Acupuncture is unlikely to have exerted a central effect by mediating the release of neurotransmitters because the hypothalamic secretion of gonadotrophin releasing hormone would be “switched off” by the gonadotrophin releasing hormone analogues used during in vitro fertilisation and the high levels of oestradiol. Acupuncture may act by reducing the contractility of the uterus and thereby avoiding expulsion of embryos after transfer or through unknown effects on the blood flow to the endometrium.⁶ Animal and human studies investigating the effect of acupuncture through a direct or indirect mechanism on the uterus or the endometrium are needed.

National guidelines and recommendations should be based on systematic reviews and meta-analyses. Should adjuvant acupuncture now be included in national guidelines such as the National Institute for Health and Clinical Excellence guidelines from the British Fertility Society?⁷ We think that it is too early for such a recommendation. Publication bias may have influenced the results of the meta-analysis. A Danish randomised controlled trial on adjuvant acupuncture that includes more than 600 women having in vitro fertilisation (twice as many as in the largest randomised controlled trial included in the meta-analysis) is currently under way. Before adding adjuvant acupuncture for in vitro fertilisation to any national guideline we must wait for the results of this and other studies to clarify the value of this treatment.