Low Dose of Caffeine Blocks Acupuncture’s Pain Relief

Recent research in mice suggests that ingesting even a low dose of caffeine may interfere with the analgesic effects of acupuncture in acute and chronic pain models.

The researchers used male and female mice that were 8–12 weeks old. In some experiments, animals were given drinking water supplemented with caffeine while other mice were given water without caffeine. The scientists induced acute inflammatory pain via injection directly into the animals’ left hind-limb ankle joints; they induced chronic inflammatory pain via injection into the left hind-limb knee joints. In other experiments, mice were injected with caffeine or saline at a specific acupuncture point. Mice also received sham acupuncture treatment, in which a needle was inserted into the same acupuncture point but not rotated (as it was during active treatment).

For the acute pain model, mice that did not ingest caffeine showed a 35 percent reduction in sensitivity following acupuncture treatment; mice that drank caffeine-laced water showed no such reduction with acupuncture. In the chronic pain model, a low dose of caffeine—0.1 mg/mL—blocked the analgesic effects of acupuncture, and at higher doses (0.2 and 0.3 mg/mL), acupuncture appeared to worsen the pain. A follow-up series of experiments suggested that the pain-relieving effects of acupuncture were restored after eliminating caffeine exposure.

The scientists speculate that even a small amount of caffeine in the body, even hours after consuming one cup of coffee, could potentially reduce the efficacy of acupuncture. They suggest that, for some people, eliminating regular consumption of caffeine may enhance the efficacy of acupuncture for pain relief. However, clinical studies are needed to confirm this, and to see how long people would need to stop ingesting caffeine to improve acupuncture’s effects.



Fujita T, Feng C, Takano T. Presence of caffeine reversibly interferes with efficacy of acupuncture-induced analgesia. Scientific Reports. 2017;7:3397.