Acupuncture: Another Clinical Trial Fails

Review of: Stener-Victorin E, Waldenström U, Nilsson L, Wikland M, Janson PO. A prospective randomized study of electro-acupuncture versus alfentanil as anaesthesia during oocyte aspiration in in-vitro fertilization. Hum Reprod 1999; 14: 2480–2484.


Here is another of many articles purporting to demonstrate the clinical value of acupuncture. It is also another of many such articles that fail to support the authors’

This Swedish study consisted of randomizing 150 women undergoing ultrasonically guided ovum retrieval as part of an in vitro fertilization (IVF) protocol to receive either electro-acupuncture (EA) or alfentanil, a fast-acting narcotic agent, as anesthesia for the procedure. The authors used a visual analog scale to assess “stress,” pain, and nausea. Their reported findings were

that although preoperative “stress” was higher among the group subjected to acupuncture, “no differences in pain directly related to oocyte aspiration, abdominal pain, or degree of nausea were found between the two groups.” Those who received EA instead of alfentanil also “had a significantly higher implantation rate (P<0.05), pregnancy rate (P<0.05), and take-home baby rate (P<0.05).” Stener-Victorin and her coauthors therefore “suggest[ed] that EA may be a good alternative to conventional anaesthesia during oocyte aspiration.”

Sadly, this study suffers from major defects that render it useless for reaching the conclusions desired by its authors. To begin with, all of the study subjects had undergone ovum retrieval in previous IVF cycles. All of them, therefore, were already familiar with the procedure and its rigors and, apparently, were not deterred by them. But the two groups differed somewhat with respect to this inasmuch as nearly two-thirds of the EA group had already been through 2 or more cycles, as compared to only slightly more than half of those in the alfentanil group.

These facts take on added significance inasmuch as no placebo interventions of any kind were used, despite the availability of placebo acupuncture needles.1 Yet patients undergoing an unpleasant procedure for the second, third, fourth, or even ninth time, as 2 subjects in the report did, might very well be inclined to report less pain with the use of a different anesthetic method. Given the similarities of proposed mechanisms of action and effectiveness of acupuncture2,3 and other placebo interventions for pain,4 this is a critical shortcoming.

Another very troubling aspect of this report is that 17% of the women in the EA group actually required and received alfentanil. Twenty-eight percent of those in the alfentanil group also required additional narcotic after the initial dosage. Besides undermining the authors’ conclusions of the efficacy of acupuncture, this strongly suggests that they were comparing it to inadequate initial doses of alfentanil. Since all patients received paracervical infiltration of local anesthetic, the results could also be interpreted to show that for some women this alone is sufficient analgesia.

The women who received EA experienced a higher implantation, pregnancy, and take-home baby rate than those who did not. But they were also, on average, a year younger—age being a well-known variable that affects the success of assisted reproduction. The number of ova retrieved and embryos transferred were similar between the two groups, but in the absence of additional prognostic information it is impossible to attach any significance to these findings. The statistical significance, in any case, was weak.

Curiously, the most statistically significant finding of Stener-Victorin and her coauthors was that “the EA group experienced discomfort [a euphemism for pain!] for a significantly longer period during oocyte aspiration (P<0.01).” But this does not appear to have affected the authors’ conclusions that “EA has been shown to be as good an anaesthetic method as alfentanil” for ovum retrieval. It is reasonable to wonder why.

Acupuncture may or may not be little more than a dramatic placebo. Some argue that placebos have their uses,5 though the ethical issues that arise from that argument are seldom addressed. Another issue concerns whether less invasive methods such as acupressure, transcutaneous electrical nerve stimulation, and other interventions that are alleged to increase endogenous opioids are just as effective. Given the vast amount of time, trouble, and effort spent on the study of acupuncture, it is disappointing, to say the least, that these questions remain unresolved. This report does little to address these problems, is seriously deficient methodologically, and, like other clinical trials of this technique, once again fails to demonstrate the dramatic results that characterize the typical anecdotal reports of acupuncture’s usefulness.


  1. Streitberger K, Kleinhenz J. Introducing a placebo needle into acupuncture research. Lancet. 1998; 352: 364–365.
  2. Petti F, Bangrazi A, Liguori A, Reale G, Ippoliti F. Effects of acupuncture on immune response related to opioidlike peptides. J Tradit Chin Med. 1998; 18(1): 55–63.
  3. Ulett GA, Han S, Han JS. Electroacupuncture: mechanisms and clinical application. Biol Psychiatry. 1998; 44(2): 129–138.
  4. ter Reit G, de Craen AJ, de Boer A, Kessels AG. Is placebo analgesia mediated by endogenous opiods? A systematic review. Pain. 1998; 76(3): 273–275.
  5. Brown WA. Harnessing the placebo effect. Hosp Pract. 1998; 33(7): 107–116.