Moxibustion treatment for knee osteoarthritis: a multi-centre, non-blinded, randomised controlled trial on the effectiveness and safety of the moxibustion treatment versus usual care in knee osteoarthritis patients.

INTRODUCTION: This study tested the effectiveness of moxibustion on pain and function in chronic knee osteoarthritis (KOA) and evaluated safety. METHODS: A multi-centre, non-blinded, parallel-group, randomised controlled trial compared moxibustion with usual care (UC) in KOA. 212 South Korean patients aged 40-70 were recruited from 2011-12, stratified by mild (Kellgren/Lawrence scale grades 0/1) and moderate-severe KOA (grades 2/3/4), and randomly allocated to moxibustion or UC for four weeks. Moxibustion involved burning mugwort devices over acupuncture and Ashi points in affected knee(s). UC was allowed. Korean Western Ontario and McMaster Universities Questionnaire (K-WOMAC), Short Form 36 Health Survey (SF-36v2), Beck Depression Inventory (BDI), physical performance test, pain numeric rating scale (NRS) and adverse events were evaluated at 5 and 13 weeks. K-WOMAC global score at 5 weeks was the primary outcome. RESULTS: 102 patients (73 mild, 29 moderate-severe) were allocated to moxibustion, 110 (77 mild, 33 moderate-severe) to UC. K-WOMAC global score (moxibustion 25.42+/-SD 19.26, UC 33.60+/-17.91, p<0.01, effect size = 0.0477), NRS (moxibustion 44.77+/-22.73, UC 56.23+/-17.71, p<0.01, effect size = 0.0073) and timed-stand test (moxibustion 24.79+/-9.76, UC 25.24+/-8.84, p = 0.0486, effect size  = 0.0021) were improved by moxibustion at 5 weeks. The primary outcome improved for mild but not moderate-severe KOA. At 13 weeks, moxibustion significantly improved the K-WOMAC global score and NRS. Moxibustion improved SF-36 physical component summary (p = 0.0299), bodily pain (p = 0.0003), physical functioning (p = 0.0025) and social functioning (p = 0.0418) at 5 weeks, with no difference in mental component summary at 5 and 13 weeks. BDI showed no difference (p = 0.34) at 5 weeks. After 1158 moxibustion treatments, 121 adverse events included first (n = 6) and second degree (n = 113) burns, pruritus and fatigue (n = 2). CONCLUSIONS: Moxibustion may improve pain, function and quality of life in KOA patients, but adverse events are common. Limitations included no sham control or blinding. TRIAL REGISTRATION: Clinical Research Information Service (CRIS) KCT0000130.