Comparative efficacy of acupuncture therapies in premature ovarian failure: A systematic review and network...

Premature ovarian failure (POF) is becoming more common and has a major negative impact on women's mental and physical well-being. The use of acupuncture therapies to treat POF has gained popularity. However, the optimal treatment remains uncertain. This study aimed to systematically review the literature and conduct a network meta-analysis (NMA) to compare the efficacy of different acupuncture and related therapies and identify methodologic weaknesses in previous studies.

We systematically searched six databases from their inception to April 2024 for randomized controlled trials (RCTs) of acupuncture therapies for POF. We assessed methodological quality and risk of bias using the RoB 2.0 tool. The NMA was conducted using R and STATA software based on frequency theory, focusing on overall effectiveness rates and hormone levels. Additionally, we critically reviewed methodological limitations and potential biases in the studies included.

Thirty-seven eligible studies involving 2419 patients were included in this NMA. The NMA results indicated that moxibustion_catgut implantation at acupoint (MB_CIA) (P-score = 0.88; pooled mean difference (95 % CI): 22.07 (1.61-302.48)) showed the highest improvement in overall effectiveness, while warming acupuncture (WA) (0.95; -27.56 (-45.35 to -9.78)) improved follicle-stimulating hormone (FSH) levels best. WA_MB (0.85; -9.35 (-15.08 to -3.62)) and WA (P-score = 0.77) were most effective for luteinizing hormone (LH) and estradiol (E2) levels, respectively. Confidence in evidence ranged from moderate to very low, with low overall confidence. Key methodologic weaknesses included insufficient allocation concealment, lack of blinding, and small sample sizes.

Acupuncture therapies may effectively improve POF. MB_CIA, WA, and WA_MB seem to be the most effective. However, severe methodological constraints, such as insufficient randomization and a lack of blinding, may reduce trust in these results. To offer robust evidence, high-quality RCTs must overcome these limitations.

[PROSPERO], identifier [CRD42024550180].