Effect of Acupuncture and Clomiphene in Chinese Women With Polycystic Ovary Syndrome

This study represented the largest randomized clinical trial ever conducted on acupuncture for fertility in women with polycystic ovary syndrome (PCOS), enrolling 1,000 participants across 21 centers in China. PCOS is the most common cause of anovulatory infertility, affecting 5-10% of women of reproductive age, and clomiphene is considered first-line treatment for ovulation induction. The methodology used a 2x2 factorial design, comparing active versus control acupuncture and clomiphene versus placebo, allowing assessment of both individual effects and any interaction between treatments. Participants were Chinese women diagnosed with PCOS by the modified Rotterdam criteria, including oligomenorrhea or amenorrhea, clinical or biochemical hyperandrogenism, and polycystic ovaries.

The active acupuncture protocol used deep needle insertion with manual and low-frequency electrical stimulation at specific points in abdominal and leg muscles, plus points on the hands and head, twice weekly for 30 minutes. The control group received superficial insertion of only 4 needles in the shoulders and arms, without manual stimulation and with simulated electrical stimulation. Clomiphene was administered at 50-150 mg for 5 days per cycle for up to 4 cycles. The primary outcome was live births after 20 weeks of gestation, with 10 months of follow-up after the last birth.

The main findings showed no significant interaction between acupuncture and clomiphene (p = 0.39), indicating that the treatments act independently. Clomiphene demonstrated efficacy superior to placebo, with a live-birth rate of 28.7% versus 15.4% (absolute difference 13.3%), confirming its position as standard treatment. However, active acupuncture did not show benefit over control acupuncture, with live-birth rates of 21.8% versus 22.4%, respectively (difference of -0.6%). Secondary outcomes of ovulation, conception, and pregnancy followed the same pattern, with significant advantage for clomiphene but not for active acupuncture.

The ovulation rate per cycle was 66.0% in the clomiphene group versus 33.8% in the placebo group, while between acupuncture groups there was no significant difference. Regarding safety, serious adverse events were rare and similar across groups. Active acupuncture caused more bruising (7.4% vs 1.8%) and diarrhea (5.0% vs 1.6%) compared with control but, interestingly, was associated with less low back pain during pregnancy. Four neonatal deaths occurred in the control acupuncture plus placebo group, related to preterm twin deliveries.

The clinical implications of this study are significant for patient counseling and health policy. The work provides robust evidence that acupuncture, even when performed by experienced practitioners at specialized centers, offers no additional benefit to conventional treatment of PCOS-related infertility. This contrasts with smaller earlier studies that suggested possible acupuncture efficacy. Limitations include the use of a fixed acupuncture protocol that may not reflect traditional individualized practice and a deviation from the original statistical plan.

Additionally, the study was restricted to Chinese women, limiting generalizability to other populations. The control used (superficial acupuncture) may not have been completely inert, potentially masking effects of active acupuncture. Despite these limitations, the high methodological quality, large sample size, and focus on the clinically relevant outcome of live births make this study an important contribution to evidence-based medicine in assisted reproduction.