Context • Primary dysmenorrhea (PD) is one of the most common complaints among young women. Acupuncture has been widely applied as a therapeutic modality in China and abroad for PD; however, the evidence for its benefits is still not convincing. Objective • The study intended to conduct a systematic review of randomized, controlled trials (RCTs) to evaluate the evidence regarding the use of acupuncture in treating PD. Design • The research team retrieved reports for RCTs published in 7 databases from their inception to March 2016, with no language restrictions: PubMed, Medline, Embase, the Cochrane Central Register of Controlled Trials, the Chinese National Knowledge Infrastructure database, the Chinese Biomedical database, and the Wanfang database. Setting • The study was conducted at the Beijing University of Traditional Chinese Medicine (Beijing, China). Participants • Participants in the reviewed studies were women aged 14 to 49 y who had received a diagnosis of PD in the absence of any visible pelvic pathology. Interventions • The types of acupuncture included traditional acupuncture, electroacupuncture, ear acupuncture, scalp acupuncture, superficial acupuncture, electrosuperficial acupuncture, wrist-ankle acupuncture, and abdominal acupuncture. Outcome Measures • The primary outcome was pain relief measured using a visual analogue scale (VAS), a verbal rating scale (VRS), or a numerical rating scale (NRS). The secondary outcomes included (1) overall improvement as measured by the short-form McGill pain questionnaire or symptom scale based on the Clinical Study Guideline for New Developed Chinese Medicine, (2) menstrual distress as measured by the Menstrual Distress Questionnaire, (3) quality of life as measured by a validated scale (eg, the short-form 36), and (4) adverse effects. Results • Twenty-three trials enrolling a total of 2770 patients were included in the review. Overall, most trials were of poor quality. Among the trials, only 6 were evaluated as having a low risk of bias, 3 of which indicated that acupuncture was statistically more effective than sham acupuncture-mean difference (MD), -3.51; 95% confidence interval (CI), -5.27 to -1.75; P < .0001; I², 0%-or no treatment-MD, -21.95; 95% CI, -25.45 to -18.45; P < .00001; I², 0%-on the VAS (0 to 100 mm). Acupuncture also showed superiority to the control arms on the VRS, the NRS, and the McGill pain questionnaire, but those findings had been influenced by methodological flaws. Conclusions • The available evidence suggests that acupuncture may be effective for PD and justifies future high-quality studies.