The quality of evidence is downgraded by risk of bias (several shortcomings in most of the studies), by heterogeneity, by imprecise results (limited study size for each comparison), and by potential publication bias.
A Cochrane review 1 included 42 studies with a total of 4640 subjects. 6 trials (n=477) reported on acupuncture (compared with sham or placebo control with inconsistent and inconclusive findings. The only study in the whole review at low risk of bias in all domains found no evidence of a difference between the groups at 3, 6 or 12 months. Acupuncture yielded lower pain scores than NSAIDs in visual analogue scale (VAS) in 7 trials and higher pain relief in 4 trials; OR 4.99 (95% CI 2.82 to 8.82; 10 trials, n=611; low evidence). There was an improvement in pain relief from acupressure compared with sham or placebo control in 6 trials (low evidence).
A meta-analysis 2 evaluating efficacy and safety of acupuncture included 49 trials. Manual acupuncture and electro-acupuncture were more effective at reducing menstrual pain compared to no treatment. Heterogeneity was high (I�= 98%).
A network meta-analysis 3 included 6 RCTs involving non-pharmacological therapies. In the Visual Analog Scale (VAS) score, acupuncture (SMD: -0.948,95% 95% CI -1.853 to -0.044) was more effective than placebo, and yoga (SMD: 2.634, 95% CI -4.28 to -0.988) was more effective than the blank control. Compared to the blank control, acupuncture (SMD: -4.81; 95% CI -6.63 to -3.00), auricular point therapy (SMD: -4.36; 95% CI -7.18 to -1.60), yoga (SMD: -2.12; 95% CI -3.13 to -1.09), and placebo (SMD: 3.10; 95% CI 1.03 to 5.27) proved to be a superior reduction in VAS.