Media outlets are reporting that Chinese herbs may relieve symptoms of endometriosis, using a Cochrane review of 2 research articles. Headlines read "Chinese herbs show early promise for endometriosis" and  "Chinese Herbs May Relieve Endometriosis Symptoms, Review Finds".

Endometriosis is a medical condition in which some of the endometrial cells (typically found in the uterus under the fluctuating influence of female hormones) are found outside of the uterine cavity. Symptoms include many nonspecific complaints such as pelvic pain, infertility, nausea, unusual menstruation, chronic fatigue, mood swings, back pain, ovarian cysts, constipation, urinary tract infections, diarrhea, anemia, etc. Appropriate diagnosis is by laparoscopic biopsy - a doctor will use a laparoscopic instrument to remove suspected extrauterine endometrial cells and examine them.  Treatments vary and can include hormonal treatments or surgery to remove the cells. In China, treatment of this disorder with Chinese herbal medicine (CHM) is routine.

Both of the media reports linked to above say that the Cochrane reviewers found some evidence that CHM has comparable benefits to conventional drug therapy after laparoscopic surgery for people with endometriosis but that the review has limitations. Andrew Flower, the primary author of the study is quoted as saying "I think the positive message is that Chinese herbal medicine may offer equivalent benefits to conventional medicine but with fewer side effects."

I found the Cochrane review [1] and noted that reviewers collected 110 studies for review and graded them based on methodological criteria. They dropped all but two of the 110 studies due to excluding trials with poor methodology, unconfirmed randomization procedures or ones diagnosing endometriosis without an appropriate laparoscopic biopsy.

Did the two retained articles feature research with superior methodology? First, let's look at what those two articles were, and what the author concluded from this review:

The first article [2] had two treatment arms: women treated with CHM orally (2x/day) and via enema (1x/day) after laparoscopic surgery versus women treated with gestrinone (2x/wk) after laparoscopic surgery for 3 months. The results showed no difference between rates of symptom relief or pregnancy in either group.

The second article [3] had three treatment arms: women treated with CHM orally(2x/day), women who treated with CHM orally and via enema(1x/day), and women treated with danazol(1x/day) for 3 months. These women did not undergo laparoscopic surgery, but instead were only biopsied for diagnostic purposes.  Women obtained greater symptomatic relief with oral and oral plus enema CHM versus danazol, oral plus enema CHM shower a greater reduction in dysmenorrhoea pain scores than danazol and shrinkage of adenexal masses. There were no differences for other factors (lumbrosacral pain, rectal discomfort, vaginal nodules).

The author concluded that post-surgical administration of CHM may have comparable benefits to gestrinone but with fewer side effects, that oral CHM may be better for treatment than danazol and may be more effective at relieving dysmenorrheal and shrinking adnexal masses when used with a CHM enema.

So, what are our weaknesses?

1.    No placebo control: There was no arm of the first study which looked at women receiving laparoscopic surgery alone without CHM or danazol, and no arm of the second study which looked at women receiving no treatment or a placebo pill treatment.

2.    Poor blinding: I should not have to point out that if you enroll in a study that has a pill treatment arm and an enema treatment arm, it is impossible for the participants to be blinded to which treatment group they are in. And enema, as you probably know, is a procedure in which liquids are forced into the rectum through the anus. It might be possible to blind participants to whether or not they are getting CHM versus the other medications, but I bet most people can tell the difference between a Chinese medicine pill and the other pills in the study. The researchers were also not blinded as to which treatment group women were in, though the paper indicates the assessors were blinded to which treatment group the women were in.

3.    Inadequate comparison treatments: Danazol is no longer commonly used as a treatment for endometriosis, and gestrinone is not available in the USA. These studies would have been much more robust had they compared it to typical drug treatments for endometriosis. In the world of conventional treatments for endometriosis, these two drugs can hardly be called conventional.

4.    Poor outcome measures: In both of the studies, a clinical outcome of "no effect" was recorded if there were no change in symptoms or if the symptoms became worse. Recording worsening symptoms as "no effect" biases the data toward a positive outcome.

I think that the most appropriate take home message or finding of the study is this: the massive stockpile of clinical trials that explore CHM for treating endometriosis have serious methodological shortcomings.

The author's main conclusion (and the conclusion parroted by the press), that CHM may work to alleviate symptoms of endometriosis, seems spurious in light of this.

Additionally, it appears that researchers used a specific mixture of herbs (Nei Yi) in the two studies, which raises the question: why the author did not title his paper "Nei Yi for endometriosis"?  Perhaps he wanted his readers to focus on the fact that this was a Chinese herbal medicine versus a "conventional" medicine. While I applaud the authors for making the weaknesses of the studies they looked at apparent, I do not applaud the way they framed their conclusions.

Lastly, it is worth noting that although the authors of the review state that there is no conflict of interest in the publication of this review, the primary author is an acupuncture and Chinese medicine practitioner at a center for Chinese medicine in the UK.


[1] Flower A, Liu JP, Chen S, Lewith G, Little P. Chinese herbal medicine for endometriosis. Cochrane Database of Systematic Reviews 2009, Issue 3. Art. No.: CD006568. DOI: 10.1002/14651858.CD006568.pub2.

[2] Wu SZ,Chen XL,Chen WZ, Li SY.Clinical analysis of the treatment of endometriosis using Nei Yi pills and Nei Yi enema. Journal of Liaoning University of TCM 2006;8(7):5-6.

[3] Wu SZ, Chen XL, Chen WZ. Clinical observation of Nei Yi pills combined with Nei Yi enema in the treatment of endometriosis. Chinese Archives of TCM 2006;24(3):431-3.

Christina Stephens, OTD/s blogs at