Effects of acupuncture on rates of pregnancy and live birth among women undergoing in vitro fertilisation: systematic review and meta-analysis

As Professor Ernst, Dr. Renckens, and Dr. Hayhoe have suggested, it is possible that the effects of acupuncture on increasing pregnancy rates may be partially mediated through an expectation effect or by anxiety reduction. Our review offers indeed little insight into the working mechanism of acupuncture. However, the precise mechanism seems relatively unimportant from a clinical perspective. Furthermore, when a sham has effects that are part and parcel of the working mechanism of acupuncture (e.g. increased expectations, reduced anxiety), but without being a feasible alternative in clinical practice, you can learn little from sham- controlled trials. For clinically relevant conclusions, we need to compare realistic alternatives, like adjuvant acupuncture versus no adjuvant to IVF.

A second reason (raised by Ernst and Renckens) to prefer trials comparing acupuncture to sham or placebo treatment is the blinding of patients and outcomes assessments. A priori, we would not have expected these trials to be largely affected by a lack of blinding because the outcomes are entirely objective (that is, pregnancy and births), and therefore unlikely to be influenced by patients’ or investigators’ judgments. In contrast, for trials with subjective, patient-assessed outcomes such as pain, outcomes can be largely affected by judgments and expectations, and controlling for placebo effects by using sham controls is important. When pain is the outcome in an acupuncture trial, for example, a trial participant’s awareness of whether or not she was receiving acupuncture, a treatment that she may expect or prefer, can influence her later judgments of levels of pain. It seems much less likely, however, that a patient’s knowledge of whether or not she was receiving acupuncture would affect her ability to become pregnant. Indeed, last month, several leading methodologists from the internationally respected Cochrane Collaboration [1] published an empirical, meta- epidemiological investigation [2] that examined whether lack of blinding was associated with biased estimates of intervention effects in trials, and whether the association between lack of blinding and biased estimates varied depending on whether the outcome investigated was subjective or objective. These investigators found that there was indeed evidence of bias, suggested by exaggerated effect estimates, when there was a lack of blinding in trials assessing subjective outcomes, but that there was little evidence of bias associated with lack of blinding in trials assessing all-cause mortality or other objectively assessed outcomes. For the entirely objective outcomes of pregnancy and birth in this review, we believe that there is a low risk of bias due to lack of blinding. The forest plot in our review makes clear that we found very little differences between trials comparing acupuncture to no adjuvant treatment or to sham treatment. So it seems obvious that the discussion above has little impact on the interpretation of our review.

Drs. Ashworth, Heptonstall, and Hayhoe have suggested some potential mechanisms through which acupuncture may improve IVF success rates. In addition, many animal studies, human physiological studies, and retrospective clinical studies have suggested biological mechanisms through which acupuncture may increase pregnancy rates [3], and this groundwork has provided a scientific rationale for the initiation of the recent RCTs. Continuing to conduct basic research into mechanisms alongside with RCTs will help provide a more cohesive and complete biomedical explanation of acupuncture and may also improve treatment. For example, as Dr. Hayhoe has noted, better understanding of the mechanism may help guide the timing of the acupuncture treatments and the selection of the acupuncture points used in future trials.

Drs. Crestani and Ren raise the issue of optimal acupuncture point selection, and note that according to the ancient acupuncture texts, some of the points used in these trials are “forbidden points” for pregnant women. Although these points may be traditionally seen as forbidden for pregnant women, in these RCTs, the women were not yet pregnant at the time of receiving acupuncture but rather were trying to become pregnant. For all but one of the trials in the review, the acupuncture points used were the same, and were based on the points used in the trials by Paulus and colleagues. However, the one included trial that used a somewhat different selection of acupuncture points (Dieterle 2006) showed the largest benefits of acupuncture. Therefore, for future trials it would seem reasonable to use either those same points used in the Paulus trials (and all subsequent trials except for Dieterle 2006) or else to use those points used in the Dieterle 2006 trial.

Dr. Hayhoe suggested that more acupuncture treatments administered over a longer term (as opposed to the two to four treatments used in the trials in this review) may result in better outcomes. The number of acupuncture treatments to include in future trials is indeed an important issue for future trial investigators to consider. That is, although only two to four treatments showed a benefit in this review, tradition would suggest that a greater number of acupuncture treatments administered over a longer duration would be preferable. To be sure, acupuncture has traditionally been used over the long-term for gradual improvement rather than over a very short period of time to bring about an immediate effect, as was the case in these RCTs.

Professor Ernst noted several other concerns. First, he suggested that there was no assessment of study quality. But we did conduct a formal assessment of study quality (or more appropriately, risk of bias), using the internal and external validity criteria recommended by the Cochrane Menstrual Disorders and Subfertility Group. This Cochrane Group scale incorporates the criteria of the Jadad scale, and also includes many additional criteria. We described our risk of bias assessment in the methods section and presented the assessments in extra Table C (available at http://www.bmj.com/cgi/content/full/bmj.39471.430451.BE/DC2).

Professor Ernst also wrote that there were considerable variations in both the acupuncture treatments and IVF procedures. In terms of the acupuncture treatments, we documented homogeneity across the trials in terms of the acupuncture points used (i.e. similar in all but one trial) the number of treatments (i.e. two to four), and the timing of treatments (i.e. all immediately before or after embryo transfer). In terms of IVF procedures, there were indeed geographical differences in the nature of the IVF procedures typically used, and this likely explained the substantial heterogeneity of baseline rates. However, we pooled the studies’ results because of the relative consistency of the effect of acupuncture across trials as well as because of the relative homogeneity of the acupuncture interventions.

Professor Ernst commented that most of the studies were published in the same journal, Fertility and Sterility. We do not consider this a weakness of our review. Fertility and Sterility is one of the top two highest impact journals in the field of reproductive medicine [4]. Finally, Professor Ernst stated that private fertility clinics would have a financial interest in the results of these trials. Indeed, four of the trials were at least partially funded by private IVF clinics. However, whether this source of funding would lead to the potential for bias is debatable. That is to say, positive results from these studies might benefit IVF clinics in general, if, for example, such positive results would make it easier for IVF clinics to charge extra by adding acupuncture as an adjuvant procedure to IVF. But such positive results would be unlikely to result in direct financial gain to the individual clinics supporting and conducting the studies because acupuncture, unlike pharmaceuticals and medical devices, is not a proprietary therapy and therefore it cannot be patented and sold by an individual, clinic, or company.

Reference List

1. Clarke M, Langhorne P. Revisiting the Cochrane Collaboration. Meeting the challenge of Archie Cochrane--and facing up to some new ones. BMJ 2001;323:821.

2. Wood L, Egger M, Gluud LL, Schulz KF, Juni P, Altman DG, et al. Empirical evidence of bias in treatment effect estimates in controlled trials with different interventions and outcomes: meta-epidemiological study. BMJ 2008;336:601-5.

3. Chang R, Chung PH, Rosenwaks Z. Role of acupuncture in the treatment of female infertility. Fertil Steril 2002;78:1149-53.

4. Buster JE. Fertility and Sterility: an evaluation. Fertil Steril 2006;86:790-4.

Competing interests: None declared

The retrospective analysis of Manheimer et al states the effects of acupuncture on rates of pregnancy and live birth among women undergoing in vitro fertilization (IVF), which is a contribution to medicine [1]. Interestingly, there are several rapid responses referring to this article. As a Chinese health-care worker, I realize that some opposite responses hold opinions that are self-contradictory and not convincing. For example, Cees N.M. Renckens [2] seems to believe that acupuncture is a useless art only based on one of the WHO reports (he stated that the report contents are not all peer- reviewed). I am not intending to disregard the WHO report, however, this only means that more clinical trials and evidence are needed for understanding the mechanism of acupuncture and there is no doubt of its effect in therapy and alleviation of clinical symptoms for various diseases.

For gyneacological problems, Francesco Crestani [3] has supported and proved the effect of acupuncture on female reproduction by providing more evidence-based literature, although the author mentioned the question regarding the potential contraindications of this therapy. Although "forbidden points" needs further clarification in the future, none can underestimate the medical contribution of acupuncture. The concern of Edzard Ernst [4] about acupuncture for IVF actually well agrees with my previous opinions that the effects of acupuncture on different diseases may be rather discrepant on various diseases. It is necessary to make a long-term follow-up study for decreasing the risk after acupuncture [5], and nothing is fully understood in medical research so far. In addition, the publishing of the rapid responses by BMJ demonstrates academic freedom, maintaining the journal's prestige.

In summary, many ancient and recent, but clear evidence supports the effectiveness of acupuncture in medicine, including gynaecology. The opposite opinions from some of the rapid responses are not convincing and they are merely trying to play to the gallery.

References

[1] Manheimer E, Zhang G, Udoff L, Haramati A, Langenberg P, Berman BM, Bouter LM. Effects of acupuncture on rates of pregnancy and live birth among women undergoing in vitro fertilisation: systematic review and meta- analysis. BMJ. 2008 Feb 7 [Epub ahead of print].

[2] Cees N.M. Renckens. Acupuncture helpful in IVF: an absurd claim! BMJ rapid response 18 February 2008.

[3] Francesco Crestani. What about "forbidden points"? BMJ rapid response 14 February 2008.

[4] Edzard Ernst. Acupuncture for IVF? BMJ rapid response 12 February 2008.

[5] Xiaofeng Ren. An old, but effective skill in use.BMJ rapid response 11 February 2008.

Competing interests: None declared

In 2003 the WHO published its report ‘Acupuncture: Review and Analysis of Reports on Controlled Clinical Trials’, which contained a list of diseases, symptoms or conditions for which acupuncture had been ‘proved’— through controlled trials — to be an effective treatment (1). These indications included adverse reactions to radiotherapy and/or chemotherapy, allergic rhinitis, biliary colic, acute bacillary dysentery, rheumatoid arthritis, tennis elbow and many others.

Superficial readers of the research paper of Mannheimer et al. (2) might be tempted to add IVF to the in my opinion disgraceful (and not at all the result of peer review) list of the WHO. They should however better not do so because this meta-analysis of RCT’s on IVF and acupuncture sets another example of Skrabanek’s dictum ‘Extraordinary claims require extraordinary evidence, and randomised clinical trials, applied to absurd claims, are more likely to mislead than illuminate’ (3). Unless there is a priori doubt about the outcome, any RCT is superfluous and even ethically dubious, which is therefore the case in trials of acupuncture. Chinese acupuncture is an atavism based on a ‘theory’ that is completely irreconcilable with our present biomedical science and the suggestions done by the authors as to how acupuncture might enhance fertility are far- fetched, to put it mildly. Without any apologies the authors do mention acupuncture-points that are known to influence fertility and points that do not have such reputation. But the existence of acupuncture-points has never been demonstrated, leave alone the possibility that these two kinds of non-existing points can in any rational way be compared as to their effect on success-rates in IVF!

The fact that in their review the two trials of better quality (W5 and W6) show less (and no longer significant) effect than the others confirms a recognisable characteristic of trials on acupuncture: the better the methodology is, the less results are found, and a perfect trial will most likely not show any effect above the placebo-effect(4).

The ‘preliminary’ positive conclusions of Mannheimer can easily be explained by publication-bias, a well-known phenomenon in the world of IVF, a treatment-modality that is nearly everywhere offered in a commercial and competitive setting. We strongly regret the fact that a prestigious journal as the BMJ accepts articles like this, which will drive many desperate infertile women into the hands and needles of practitioners of an ancient but nevertheless useless art.

1. Anonymus. ‘Acupuncture: Review and Analysis of Reports on Controlled Clinical Trials’ 2003. Edited by WHO, Geneva.

2. E. Manheimer, G. Zhang, L. Udoff, A. Haramati, P. Langenberg, B. M Berman, and L. M. Bouter. ‘Effects of acupuncture on rates of pregnancy and live birth among women undergoing in vitro fertilisation: systematic review and meta-analysis’ BMJ 2008; 0: bmj.39471.430451.BEv1

3. Skranabek P. ‘Demarcation of the absurd’. Health Watch Newsletter (5) 1990, 7.

4. G. ter Riet, J. Kleijnen, P. Knipschild. ‘De effectiviteit van acupunctuur. Nawoord en aanbevelingen’. Huisarts en Wetenschap 1989;32(8):308-312.

Competing interests: None declared

The meta-analysis by Manheimer et al of 7 RCTs suggests that acupuncture improves the success of in vitro fertilisation (IVF).[1] Remarkably, the authors provide no formal assessment of the methodological quality of the 7 studies (the BMJ subscribes to the QUORUM guidelines; why does it make an exception for acupuncture?). They merely state that “the included trials generally had sound methods” and that “any minor design concerns would not be expected to result in a substantial risk of bias”. Even though not all of the studies have been published, the 5 that are available, in my view, raise grave concerns.

Four of the 7 RCTs made no attempt at all to blind patients or control for placebo effects. Manheimer et al believe that “the necessity to blind participants … is arguable when the outcomes are entirely objective”. Considering that one of the mechanisms explaining the effects of acupuncture in IVF is stress reduction (Dieterle), this statement may seem to be misleading. The only two trials which adequately control for placebo-effects (Paulus 2003, Smith 2006) fail to demonstrate significant effects.

Other important concerns abound. There are considerable variations in both acupuncture treatments and the IVF procedures. Yet the authors claim that there was clinical homogeneity amongst the primary studies. Most trials fail to explain how many acupuncturists performed the experimental and control treatments and whether they could have influenced the outcome. Several studies report that acupuncture always elicted ‘dequi’, yet this sensation is not normally experienced by all patients. Most of the primary studies were published in the same journal and many were conducted in private fertility clinics. The authors state that “a primary reason for orientation bias is widely acknowledged to be competing financial interest and as acupuncture is a non-proprietary treatment these might be reduced”. Yet private clinics could have a very real financial interest in the results of these studies.

In conclusion, my reading of the primary studies (which were not provided in the online version but had to be retrieved through literature searches) suggests that the best available evidence fails to suggest that acupuncture has a strong influence beyond placebo on the success of IVF.

Prof Edzard Ernst - Edzard.Ernst@pms.ac.uk

References

[1] Manheimer E, Zhang G, Udoff L, Haramati A, Langenberg P, Berman BM et al. Effects of acupuncture on rates of pregnancy and live birth among women undergoing in vitro fertilisation: systematic review and meta- analysis. BMJ 2008;Feb 7 [Epub ahead of print]- doi:10.1136/bmj.39471.430451.

Competing interests: None declared