Acupoint stimulation intervention for people with primary dysmenorrhea: Systematic review and meta-analysis of randomized trials.

BACKGROUND: Primary dysmenorrhea is a common gynecologic complain in women of reproductive age. Acupoint stimulation therapies might be an effective intervention for primary dysmenorrhea.
OBJECTIVE: The aim of this study was to determine the effectiveness of acupoint stimulation for primary dysmenorrhea. DATA SOURCES: All searches in the Cochrane Library, MEDLINE, PubMed, CINAHL Plus with Full Text, and CEPS databases (inception to March 2011). STUDY SELECTION: Randomized controlled trials (RCTs) included were comparing acupoint stimulation with non-acupoint-related stimulation or medication. DATA EXTRACTION: Data were abstracted independently by two authors onto standardized forms, and disagreements were resolved by discussion. DATA SYNTHESIS: Thirty RCTs met the selection criteria, and 25 reported sufficient data for pooling. The main outcomes assessed were cure rate, total effective rate, pain intensity, menstrual pain, plasma PGF(2α)/PGE(2) ratio, and adverse events. According to the type of outcome, the strength of a relationship between two dichotomous variables was described by odds ratios and 95% confidence intervals, and continuous variables were expressed as mean±standard deviation. Fixed-effects models were used to perform meta-analysis.
RESULTS: Twenty-five RCTs with a total of over 3000 participants were included for the meta-analysis. Acupoint stimulation when compared with non-acupoint-related stimulation or medication had significant effects. Moderator analysis further confirmed that invasive and noninvasive acupoint stimulation was effective separately, with the latter being more effective. The most common adverse events were hemorrhage and hematoma. LIMITATIONS: Papers written in language other than English or Chinese were not included. CONCLUSIONS AND IMPLICATION: This finding indicates that acupoint stimulation, especially non-invasive acupoint stimulation, could have good short term effects on pain of primary dysmenorrhea. Inference in some studies was somewhat restricted due to low methodological rigor. We suggest well-designed, methodologically rigorous, large trial, evaluating both short and long-term effects on pain and other outcomes in comparison with the available standard treatments.