Eleven RCTs were included in the review (n=475). Sample sizes ranged from 26 to 80 participants. Trial quality varied, but was on average moderate (with a mean Jadad score of 3 out of 5, and a range of 1 to 5). OPVS scores ranged from 6 to 15 (out of 16). The degree of confidence that acupuncture was applied appropriately ranged from 60 per cent to 90 per cent.
Overall, bee venom acupuncture was associated with reduced pain compared to control treatments in 10 trials. A statistically significant difference in pain was seen for the four trials comparing bee venom acupuncture plus classic acupuncture with saline injection plus classic acupuncture, WMD 14.0mm (95% CI: 9.5, 18.6, p<0.001), the two trials comparing bee venom acupuncture plus classic acupuncture with classic acupuncture alone, WMD 23.4mm (95% CI: 16.2, 30.7, p<0.001) and the three trials comparing bee venom acupuncture with classic acupuncture, SMD 0.55 (95% CI: 0.16. 0.95, p<0.01).
No heterogeneity was found for any of the analyses. Only four trials reported adverse events. Itching and skin hypersensitivity were the most common.