Acupuncture and moxibustion for lateral elbow pain: a systematic review of randomized controlled trials

This systematic review represents the first comprehensive analysis of the English and Chinese literature on acupuncture and moxibustion for lateral elbow pain (LEP), a condition known as lateral epicondylitis or tennis elbow. LEP affects 1-3% of the population and represents a significant health burden, impacting workplace productivity and patient quality of life.

The research analyzed 19 randomized controlled trials conducted between 1994 and 2011, involving 1,190 participants. Of the included studies, 14 were published in Chinese and conducted in China, four in English (three in Germany, one in Canada), and one in Italian. Sample sizes ranged from 16 to 120 participants, with ages between 17-76 years.

The studies were categorized into three main comparison groups. First, three studies compared acupuncture with sham acupuncture, all demonstrating superiority of real acupuncture. The Molsberger study found 55.8% pain reduction in the acupuncture group versus 15% in the sham group immediately after treatment. Irnich demonstrated significant improvements in elbow pain and mobility compared with sham, maintained for two weeks.

Fink reported superior improvements in maximum strength, pain intensity, and arm function at two weeks, although the differences were not maintained at two months.

Second, seven studies compared acupuncture with conventional therapies such as corticosteroid injections, pulsed ultrasound, and anti-inflammatory drugs. Three studies showed significantly higher cure rates for acupuncture compared with prednisolone injections and oral meloxicam. An Italian study found significant reduction in the Maigne functional score and pain on the visual analog scale at six-month follow-up.

Third, three studies evaluated moxibustion versus conventional therapies, finding no significant differences between groups. Additionally, six studies compared acupuncture combined with moxibustion (ACM) versus acupuncture alone, all demonstrating superiority of the combination. Interestingly, a three-arm study showed that electroacupuncture was more effective than ACM.

The most commonly used acupuncture points were local tender points (Ashi points) in 14 studies, followed by He Gu (LI-4), Shou San Li (LI-10), and Qu Chi (LI-11) in ten studies. Only one study used exclusively distal points (GB-34). The number of sessions ranged from 1 to 36, with frequency ranging from once daily to once every three days, and total treatment duration from 1 to 37 days.

Methodological quality was evaluated using the Cochrane risk of bias tool, revealing significant limitations. All studies except one presented at least one domain classified as high risk of bias. Main problems included inadequate random sequence generation, absence of allocation concealment, and difficulties in blinding due to the nature of the intervention. Telephone contact with Chinese authors revealed that two studies used alternating allocation, not being truly randomized.

Evaluation using the revised STRICTA criteria (2010) showed that no study reported acupuncture procedures with sufficient detail. Essential information such as number of needles, retention time, and needle type was not reported in approximately one-third of the studies. Adverse events were rarely documented, with only four studies reporting safety.

Clinical implications suggest that acupuncture may offer benefits for LEP, particularly when compared with sham interventions. The combination with moxibustion may potentiate the effects, although the mechanisms remain poorly understood. Possible mechanisms include activation of diffuse noxious inhibitory control, release of endogenous opioids, anti-inflammatory effects via cholinergic stimulation, and local modulation through neuropeptide release.

Limitations include low methodological quality of most studies, lack of standardization in outcome measures, absence of clear definition of primary outcomes, variation in control treatments, and study duration. Meta-analysis was not possible due to methodological heterogeneity and inadequate study quality.