Acupuncture and PCOS: signals, but not for live birth

Polycystic ovary syndrome is a heterogeneous disorder spanning reproductive, endocrine and metabolic problems, in which autonomic, neuroendocrine, inflammatory and metabolic abnormalities may all converge on ovarian and endometrial function. That complexity is exactly why non-drug adjuncts such as acupuncture keep being proposed and studied. The difficulty for anyone reading the literature is that mechanistic plausibility and clinically meaningful benefit are constantly blurred together, and much of the supporting material comes from animals or from unrelated conditions. This review set out to sort the evidence by its source and directness, and to be explicit about which outcomes actually matter to patients.

The authors conducted a narrative review rather than a systematic one, searching PubMed, Embase and Web of Science from inception to 31 March 2026 and supplementing this with backward citation tracking. They deliberately prioritised human PCOS trials and mechanistic studies first, then PCOS-specific preclinical evidence, and included findings from other conditions only when explicitly labelled as indirect evidence. Emphasis was placed on clinically meaningful reproductive outcomes such as ovulation, pregnancy and live birth. Importantly, no formal risk-of-bias assessment and no meta-analysis were undertaken, so the synthesis is interpretive rather than quantitative.

Clinical evidence most consistently suggested possible effects on menstrual or ovulatory function, sympathetic activity and insulin resistance, though results varied across acupuncture protocols and PCOS phenotypes. Evidence for pregnancy, ongoing pregnancy, embryo quality and live birth was limited and inconsistent, and the review states acupuncture has not been established as a replacement for guideline-based fertility treatment. Human data directly linking acupuncture to immune-cell reprogramming, endometrial immune tolerance or correction of window-of-implantation timing were described as scarce. PCOS-specific animal studies offered plausible signals involving hypothalamic-pituitary-ovarian regulation, ovarian sympathetic signalling, inflammatory pathways, oxidative stress and insulin signalling, which the authors caution cannot be assumed to translate.

Rather than settling the question, this review reframes it by separating relatively supported clinical domains from mechanisms that remain preclinical or indirect. The neuroendocrine-immune account of how acupuncture might act in PCOS is presented as a hypothesis-guiding map, not an established explanation. That complicates enthusiastic readings of the mechanistic literature while leaving room for the more modest clinical signals around cycles and insulin resistance. The open questions are named directly: future trials should use standardised acupuncture protocols with phenotype stratification, prespecified ovulation, ongoing-pregnancy and live-birth outcomes, and longitudinal autonomic, metabolic, inflammatory, ovarian and endometrial measurements.

If you have PCOS and have seen acupuncture promoted for fertility, this is a reason to be sceptical of confident claims about pregnancy or live birth, because the review found that evidence limited and inconsistent. The somewhat better-supported areas are cycle regularity, sympathetic activity and insulin resistance, and even there the results shift with protocol and PCOS subtype. The elaborate immune and implantation mechanisms often cited come largely from animal work or indirect sources. This is background for a conversation with a clinician about guideline-based care, not a basis for choosing or dropping any treatment.