Does nonsurgical treatment improve longitudinal outcomes of lateral epicondylitis over no treatment? A meta-analysis.

BACKGROUND: Lateral epicondylitis is a painful tendinopathy for which several nonsurgical treatment strategies are used. Superiority of these nonsurgical treatments over nontreatment has not been definitively established.

QUESTIONS/PURPOSES: We asked whether nonsurgical treatment of lateral epicondylitis compared with observation only or placebo provides (1) better overall improvement, (2) less need for escape interventions, (3) better outcome scores, and (4) improved grip strength at intermediate- to long-term followup.

METHODS: The English-language literature was searched using PubMed and the Cochrane Central Register of Controlled Trials. Randomized-controlled trials (RCTs) comparing any form of nonsurgical treatment with either observation only or placebo at followup of at least 6 months were included. Nonsurgical treatments included injections (corticosteroid, platelet-rich plasma, autologous blood, sodium hyaluronate, or glycosaminoglycan polysulfate), physiotherapy, shock wave therapy, laser, ultrasound, corticosteroid iontophoresis, topical glyceryl trinitrate, or oral naproxen. Methodologic quality was assessed with the Consolidated Standards of Reporting Trials (CONSORT) checklist, and 22 RCTs containing 2280 patients were included. Pooled analyses were performed to evaluate overall improvement; requirement for escape interventions (treatment of any kind, outside consultation, and surgery); outcome scores (Patient-Rated Tennis Elbow Evaluation [PRTEE]; DASH; Pain-Free Function Index [PFFI]; EuroQoL [EQ]-5D; and overall function); and maximum and pain-free grip strength. Sensitivity analyses were performed using only trials of excellent or good quality. Heterogeneity analyses were performed, and funnel plots were constructed to assess for publication bias.

RESULTS: Nonsurgical treatment was not favored over nontreatment based on overall improvement (risk ratio [RR] = 1.05 [0.96-1.15]; p = 0.32), need for escape treatment (RR = 1.50 [0.84-2.70]; p = 0.17), PRTEE scores (mean difference [MD] = 1.47, [0.68-2.26]; p < 0.001), DASH scores (MD = -2.69, [-15.80 to 10.42]; p = 0.69), PFFI scores (standardized mean difference [SMD] = 0.25, [-0.32 to 0.81]; p = 0.39), overall function using change-from-baseline data (SMD = 0.11, [-0.14 to 0.36]; p = 0.37) and final data (SMD = -0.16, [-0.79 to 0.47]; p = 0.61), EQ-5D scores (SMD = 0.08, [-0.52 to 0.67]; p = 0.80), maximum grip strength using change-from-baseline data (SMD = 0.12, [-0.11 to 0.35]; p = 0.31) and final data (SMD = 4.37, [-0.65 to 9.38]; p = 0.09), and pain-free grip strength using change-from-baseline data (SMD = -0.20, [-0.84 to 0.43]; p = 0.53) and final data (SMD = -0.03, [-0.61 to 0.54]; p = 0.91).

CONCLUSIONS: Pooled data from RCTs indicate a lack of intermediate- to long-term clinical benefit after nonsurgical treatment of lateral epicondylitis compared with observation only or placebo.

LEVEL OF EVIDENCE: Level II, therapeutic study.