Thirty-one randomised controlled trials (RCTs) were included in the review (n=3,013 participants). Of these, two RCTs did not provide sufficient information to be included in meta-analyses. Methodological quality of trials varied; 20 RCTs scored 1 on the Jadad scale, the lowest score possible. Six RCTs scored 2 and five RCTs scored 3 or higher.
Acupuncture compared to lifestyle controls: Acupuncture significantly reduced average body weight compared to lifestyle controls (weighted mean difference 1.72 kg, 95% confidence interval (CI): 0.50 to 2.93; p=0.006, five RCTs) and combined with diet, significantly improved weight loss compared to diet alone (relative risk 2.57 kg, 95% CI: 1.98 to 3.34; p<0.00001, five RCTs). Subgroup analyses were inconsistent across the different outcome measures used. Statistical heterogeneity and publication bias were not observed for most outcomes, however, when acupuncture was combined with diet, funnel plots suggested a lack of smaller sized studies.
Acupuncture compared to placebo or sham treatments: Compared to placebo or sham treatments, acupuncture significantly reduced average body weight (weighted mean difference 1.56 kg, 95% CI: 0.74 to 2.38; p=0.0002, three RCTs). No heterogeneity was observed. Weight loss was also significantly improved in patients receiving acupuncture (mean difference 3.66, 95% CI: 3.05 to 4.27; one RCT) and also patients receiving both acupuncture and auricular acupressure (relative risk 17.0, 95% CI: 2.49 to 115.86; one RCT) compared to wait-list control
Acupuncture compare to medication: Acupuncture showed more improved outcomes for body weight (WMD 1.90 kg, 95% CI: 1.67 to 2.13; p=0.0002, two RCTs) and for weight loss (relative risk 1.13 kg, 95% CI: 1.04 to 1.22; p<0.00001, four RCTs) than conventional medication. No statistical heterogeneity was observed.
Acupuncture compared to other treatments: Compared to other treatments, acupuncture was significantly beneficial compared to herbal tea, herbal supplements and transcutaneous electrical nerve stimulation treatment. No significant differences were reported between acupuncture and ultrasound stimulation therapy.
Adverse events: Among trials reporting adverse events, four RCTs reported minimal adverse events and two RCTs reported no adverse events. Redness, pain or discomfort, bleeding in ears, dry mouth, headaches, sleepiness, hypertension and dizziness were reported.