Rheumatology (Oxford). 2003 Jan; 42(1): 141-8.

A randomized double-blind pilot study comparing Doloteffin and Vioxx in the treatment of low back pain.

Chrubasik S, Model A, Black A, Pollak S.

Department of Forensic Medicine, University of Freiburg, 79104 Freiburg, Germany. [email protected]

OBJECTIVE: This randomized, double-dummy, double-blind pilot study of acutely exacerbated low back pain was aimed to inform a definitive comparison between Doloteffin, a proprietary extract of Harpagophytum, and rofecoxib, a selective inhibitor of cyclo-oxygenase-2 (COX-2). METHODS: Forty-four patients (phyto-anti-inflammatory drug-PAID-group) received a daily dose of Doloteffin containing, inter alia, 60 mg of harpagoside for 6 weeks and 44 (non-steroidal anti-inflammatory drug-NSAID-group) received 12.5 mg/day of rofecoxib. All were allowed rescue medication of up to 400 mg/day of tramadol. Several outcome measures were examined at various intervals to obtain estimates of effect size and variability that might be used to decide the most suitable principal outcome measure and corresponding numbers required for a definitive study. RESULTS: Forty-three PAID and 36 NSAID patients completed the study. Ten PAID and 5 NSAID patients reported no pain without rescue medication for at least 5 days of the 6th week of treatment. Eighteen PAID and 12 NSAID patients had more than a 50% reduction in the week's average of their pain scores between the 1st and 6th weeks. The mean percentage decrease from baseline in the pain component of the Arhus Index was 23 (S.D. 52) in PAID and 26 (S.D. 43) in NSAID. The corresponding measures for the overall Arhus Index were 11 (31) and 16 (24) and, for the Health Assessment Questionnaire, 7 (8) and 6 (7). Tramadol was used by 21 PAID patients and 13 NSAID patients. Fourteen patients in each group experienced 39 adverse effects, of which 28 (13 in PAID) were judged to some degree attributable to the study medications. CONCLUSION: Though no significant intergroup differences were demonstrable, large numbers will be needed to show equivalence.