Thirty-five studies were included in the review. Studies evaluated 22 different types of herbal medicinal product, 17 of which were discussed within the paper. Twenty-six of the 35 studies used a parallel design and eight used a crossover design. The review stated that results evaluating trial quality were presented in tabulated form, but this table was missing. A written summary stated that six of the 35 studies met all criteria; other studies were variable.
Statistically significant results based on meta-analysis of more than one trial:
SK1306X: two trials (n=388 participants) reported a statistically significant reduction in pain using the visual analogue scale (MD -17.36, 95% CI -22.57 to -12.15) and improved physical function using the Lequesne index (MD -2.73, 95% CI -3.71 to -1.74).
Topical capsaicin: a statistically significant reduction in pain scores using the visual analogue scale was reported four weeks after intervention (MD -8.26, 95% CI -14.88 to -1.65; two studies, n=179 participants).
Avocado-soybean unsaponifiables: within two studies (n=326 participants) pain scores using visual analogue scale showed a statistically significant difference after three months favouring treatment (MD -10.79, 95% CI -14.91 to -6.66),
Statistically significant results based on a single trial:
Ginger extract: improvements were reported in terms of WOMAC stiffness, WOMAC total scores, pain after walking 50 feet and patient global assessment.
Boswellia-Curcuma mixture: improvements were reported in minutes of pain-free walking time after two and three months.
Boswellia serrata: improvements were reported in pain, loss of movement, and swelling.
Reumalex: improvements were reported in AIMS 2 pain scores.
Avocado-soybean unsaponifiables: improvements were reported in pain scores after six months.
Statistically significant findings favouring the following herbal medicinal products were also reported for cat's claw extract, comfrey extract and stinging nettle leaf. Full results were available in the paper.
No serious side effects were reported with any herbal intervention.