Electroacupuncture for the treatment of perimenopausal syndrome: a systematic review and meta-analysis of randomized controlled trials

To assess the efficacy, comparative effectiveness and safety of electroacupuncture (EA) in the treatment of perimenopausal syndrome (PMS).Nine databases were searched until June 2019. Only relevant randomized controlled trials (RCTs) of EA for PMS were included.Twelve trials involving 746 women were included. EA and hormone therapy (HT) did not significantly differ in terms of effective rate (risk ratio (RR) = 0.98, 95% confidence interval (CI) = 0.93 to 1.04), Kupperman index (KI) (mean difference (MD) = -0.25, 95% CI = -0.76 to 0.26) and serum levels of follicle-stimulating hormone (FSH) (MD = -3.80, 95% CI = -11.59 to 3.98) or luteinizing hormone (LH) (MD = -2.51, 95% CI = -10.72 to 5.70). Serum estradiol (E2) levels were significantly lower in EA versus HT groups (MD = -60.58, 95% CI = -71.93 to -49.23). Compared with sham EA, EA had a significantly greater effect on reductions in KI (MD = -4.71, 95% CI = -6.57 to -2.86) and hot flushes score/24 h (MD = -2.43, 95% CI = -2.93 to -1.93). There were no significant differences between EA and manual acupuncture (MA) in terms of effective rate (RR = 1.14, 95% CI = 0.98 to 1.33) or serum FSH (MD = -2.87, 95% CI = -29.65 to 23.91), LH (MD = 2.73, 95% CI = -9.65 to 15.11) or E2 (MD = 26.80, 95% CI = -12.06 to 65.65). However, it seemed that EA had a better effect than MA on KI (MD = -2.44, 95% CI = -4.80 to -0.08). Subgroup analyses indicated that EA may have more of a benefit in the pre-menopausal state (hot flushes score/24 h: MD = -1.66, 95% CI = -3.49 to 0.17) compared to post-menopause (p > 0.05).The effect of EA appeared broadly similar to HT and MA in the treatment of PMS, although EA-associated reductions in KI were superior to MA and sham EA, suggesting effects beyond placebo. The evidence base is limited by a small number of eligible studies, risk of bias and clinical/statistical heterogeneity, limiting our ability to draw firm conclusions. As such, additional larger scale, high-quality RCTs are needed.