DAVIS, Calif., July 29 -- For patients with knee osteoarthritis, an extract of the Indian frankincense plant gave significant pain relief and reduced levels of a marker of joint pathology, researchers here said.

Patients in a randomized, double-blind, 90-day trial showed significantly greater reductions in pain scores with the agent than with placebo, reported Siba R. Raychaudhuri, M.D., of the University of California Davis, and colleagues online in Arthritis Research and Therapy.

Synovial fluid levels of matrix metalloproteinase-3 also declined significantly in patients receiving the agent, whereas MMP-3 levels rose in the placebo group.

This was the agent's first randomized, placebo-controlled trial in osteoarthritis of the knee, the researchers said.

The agent, tradenamed 5-Loxin, was derived from Boswellia serrata, a plant that has long been prominent in the Ayurvedic system of traditional Indian medicine. It is already widely marketed in the U.S. as a nutritional supplement, with purported benefits including enhanced joint mobility and function.

The extract contains 30% 3-O-acetyl-11-keto-beta-boswellic acid, identified in laboratory and animal studies as the anti-inflammatory component, according to Dr. Raychaudhuri and colleagues.

In particular, the compound appears to inhibit 5-lipoxygenase, a key intermediate in the inflammatory cascade.

The study assigned 70 patients to placebo or to 100 mg or 250 mg of the extract, given orally in divided doses.

Patients at baseline had mean pain scores, on a visual analogue scale, of 56 to 57 points.

At the end of treatment, mean scores in the placebo group were 41.8 (SD 16.0), compared with 21.4 (SD 7.1) in the low-dose treatment group and 14.2 (SD 6.8) for those in the high-dose group. The declines in both treatment groups were significantly greater than in the placebo group (P<0.0001).

Similar results were also seen in the pain subscale of the Western Ontario and McMaster Universities Osteoarthritis Index (WOMUOI).

Improvements in pain scores with the agent were noted in one week in some patients, the researchers aid.

Functional assessments, including the WOMUOI stiffness and functional subscales and Lequesne's Functional Index, also showed significant advantages for the plant extract versus placebo.

For example, mean stiffness scores in the placebo group declined from 33.2 (SD 2.7) at baseline to 24.5 (SD 2.4) after treatment, whereas in the low-dose treatment group, mean scores fell from 31.8 (SD 3.6) at baseline to 14.1 (SD 3.7) following treatment (P<0.0001 versus placebo).

As with the pain scores, greater improvements were seen in the high-dose group: stiffness scores decreased from 27.8 (SD 3.4) at baseline to 9.24 (SD 2.1) after treatment (P<0.0001).

Mean MMP-3 levels in synovial fluid increased slightly in the placebo group, from 902.1 ng/ml (SD 275.1) at baseline to 928.5 (SD 216.0) following treatment.

In the low-dose and high-dose treatment groups, MMP-3 levels fell substantially: from 893.6 (SD 270.1) and 926.9 ng/ml (SD 270.5) at baseline, respectively, to 637.2 (SD 224.5) and 497.5 (SD 167.5) ng/ml after treatment (both P<0.0001 versus placebo).

MMP-3 is a cartilage-degrading enzyme. Its presence in synovial fluid is a marker of joint destruction.

"5-Loxin has potential efficacy in terms of reducing pain and improving the physical ability of osteoarthritis patients," Dr. Raychaudhuri and colleagues wrote.

In combination with animal experiments that found no mutagenic effects or other major toxicity concerns, the adverse event profile in the clinical trial indicates that the compound "is potentially safe in the treatment of osteoarthritis in humans," the researchers said.

Dr. Raychaudhuri and colleagues said the most common adverse effects were diarrhea, nausea, abdominal pain, mild fever, and general weakness, without significant difference between placebo and the active treatment groups.

The study was funded by Laila, developer of 5-Loxin. Several co-authors were Laila employees and others received research funding and/or had other relationships with the company. No other potential conflicts of interest were reported