A new study of acupuncture in IVF: pointing in the right direction
Article Outline
Abstract
New emerging evidence from clinical trials demonstrates that acupuncture performed at the time of embryo transfer does not improve the pregnancy or live birth outcome after treatment. This evidence raises questions regarding the futility of conducting further research in this area and the quality of evidence needed before any specific intervention is incorporated into routine clinical practice, particularly when a scientific rationale is lacking.
Keywords: acupuncture, complementary medicine, embryo transfer, IVF, non-penetrating acupuncture, outcome
Complementary medicine is popular amongst infertile patients. In particular, the use of acupuncture as an adjunct to IVF treatment has gained momentum after the publication in 2008 of a systematic review and meta-analysis (Manheimer et al., 2008) and a Cochrane review (Cheong et al., 2008), both of which have shown a beneficial effect of acupuncture at the time of embryo transfer on the clinical pregnancy rate after IVF.
Since then, a number of more recent publications have challenged the results of these two reviews (Domar et al., 2009, El-Toukhy et al., 2008, Pinborg et al., 2008, Renckens, 2009, So et al., 2009, Sunkara et al., 2009). In this issue of Reproductive BioMedicine Online, Anderson and colleagues (2010) report the results of a large high-quality multicentre randomized trial comparing acupuncture with a validated placebo (non-penetrating acupuncture) performed at the time of embryo transfer. With the exception of the study of So et al. (2009), which included 370 women, the study by Anderson et al. included more than twice as many women as any of the earlier studies on acupuncture. The authors concluded that acupuncture has no beneficial effect on the pregnancy or live birth rate after IVF.
The publication of this study raises two important clinical questions: one is specifically related to the role of acupuncture in IVF and the other addresses a more fundamental issue related to the practice of evidence-based medicine as a whole.
Firstly, does acupuncture enhance the success rate of IVF treatment when performed around the time of embryo transfer? According to the most recent meta-analysis (Sunkara et al., 2009) and after the inclusion of the study of Anderson et al. (2010), acupuncture does not improve IVF outcome (risk ratio for clinical pregnancy
=1.04, 95% CI 0.96–1.12). Using the results of the study by Anderson et al. to calculate sample size, any future high-quality and sufficiently powered clinical trial examining the value of acupuncture in IVF will need to recruit 2300 women in each arm to have an 80% power to detect a 4% difference in the clinical pregnancy rate at a double-sided alpha of 0.05. Indeed, larger numbers will have to be recruited to show the same effect on the live birth rate. The prospect of such study being conducted is unrealistic, indicating that the conclusion about lack of efficacy of acupuncture in IVF will prevail for the foreseeable future.
The second question raised is what level of evidence is required by clinicians to incorporate a particular intervention into clinical practice. It is accepted that systematic reviews and meta-analyses represent the highest level of evidence available (level 1a) and their results are meant to illuminate about best practice (Centre for Evidence Based Medicine, 2009). However, there is no doubt that the conclusion of a systematic review is dependent upon the quality of its component studies. Therefore, whenever a systematic review and meta-analysis of clinical trials claims ‘statistical’ evidence of efficacy for a specific intervention, its conclusions must be challenged if the included studies were small, of questionable quality, demonstrably heterogeneous or had small effect sizes (Ioannidis, 2005). This becomes particularly important when a sound and plausible biological rationale for the action of the intervention is lacking (Renckens, 2009) rendering the pre-study probability for true findings very low. Acupuncture in IVF is a stark example of such situation in many respects.
It is therefore incumbent upon all clinicians with critical appraisal skills, medical authorities developing clinical guidelines and the scientific community in general to scrutinize claims of therapeutic efficacy and delay the introduction of new interventions in clinical practice until there is not only sufficient evidence from adequately powered robust clinical trials or systematic reviews with little heterogeneity, but also a well-documented, clear and accepted mechanism of action.
References
- Anderson, D., Løssl, K., Nyboe Andersen, A., et al., 2010. Acupuncture on the day of embryo transfer: a randomized controlled trial of 635 patients. Reproductive BioMedicine Online, doi:10.1016/j.rbmo.2010.06.043.
- Centre for Evidence Based Medicine, University of Oxford. March 2009. <http://www.cebm.net/?O=1025> (accessed 15/5/2010).
- Cheong, Y.C., Ng, E.H.Y., Ledger, W.L., 2008. Acupuncture and Assisted Conception. Cochrane Database of Systematic Reviews Issue 4. Article No.: CD006920.
- The impact of acupuncture on in vitro fertilization outcome. Fertility and Sterility. 2009;91:723–726
- A systematic review and meta-analysis of acupuncture in in vitro fertilization. British Journal of Obstetrics and Gynaecology. 2008;10:1203–1213
- . Why most published research findings are false. PLoS Medicine. 2005;2:e124
- Effects of acupuncture on rates of pregnancy and live birth among women undergoing in vitro fertilization: systematic review and meta-analysis. British Medical Journal. 2008;336:545–549
- . Acupuncture with in vitro fertilization may increase birth rates, but guidelines should await the true results of ongoing trials. British Medical Journal. 2008;336:517
- . RCT of real versus placebo acupuncture in IVF. Human Reproduction. 2009;24:1238
- A randomized double blind comparison of real and placebo acupuncture in IVF treatment. Human Reproduction. 2009;24:341–348
- . Acupuncture and in vitro fertilization: updated meta-analysis. Human Reproduction. 2009;24:2047–2048
PII: S1472-6483(10)00433-5
doi:10.1016/j.rbmo.2010.06.024
© 2010 Published by Elsevier Inc.
