Prolotherapy is defined as iatrogenic stimulation of wound healing and tissue repair through injection of an irritant solution into damaged ligaments, tendons, and joints to encourage healing and repair. Although evidence to support efficacy of prolotherapy exists, most studies suffer from small sample size, inadequate controls, and inconsistent study design. This pilot study was designed to critically compare prolotherapy vs. corticosteroid therapy in the treatment of chronic lateral epicondylitis in a non-inferiority trial using a standardized treatment protocol. Objective and subjective outcome measures were used to describe strength, function, and pain. We hypothesized that prolotherapy is as effective as corticosteroid therapy.
Materials and methods: Patients with chronic lateral epicondylitis were recruited using strict inclusion and exclusion criteria via advertisement at local tennis clubs and direct physician referral. IRB approval was obtained by Spaulding Rehabilitation Hospital. Subjects underwent 3 office visits and 1 phone follow-up. The study was double blinded. All patients received one of two study medications: (A) P2G/Sodium Morrhuate or (B) DepoMedrol 40mg/ml, each combined with local anesthetic solution (Procaine 0.9% NACL preservative-free 1% solution), administered in a standardized injection protocol. Primary outcome measures were VAS and DASH; secondary outcome measure was grip strength dynamometry.
Results: A total of 24 patients were recruited; 18 patients completed the study. T-test analyses were conducted. For change in VAS, an observed difference between groups A and B was 0.4, (p-value 0.7; CI -2.70-1.96). For change in DASH, the observed difference was 5.6 (p-value 0.5; CI of -21.9-10.6).
Conclusions: Although our pilot data did not achieve statistical significance, the observed differences between groups A and B were clinically significant and consistent with the non-inferiority hypothesis. This suggests that prolotherapy may be as effective as corticosteroid therapy. Further studies with larger sample sizes are warranted to investigate whether a true difference exists.