PURPOSE: To quantify the effect of saline solution injections on patient-reported outcome measures (PROMs) and to determine whether this effect is clinically relevant by comparing it with minimal clinically important difference (MCID) criteria.
METHODS: A systematic search identified randomized controlled trials of lateral epicondylitis interventions comparing saline solution injections with nonsurgical injection therapies. Among included studies, saline solution was compared with platelet-rich plasma, autologous conditioned plasma, corticosteroid, and botulinum toxin injections. By use of data from included studies, a random-effects model was used to calculate overall mean differences (MDs) in pre- and post-injection PROMs in a pair-wise fashion. Calculated MDs were then compared with MCID criteria.
RESULTS: Of 458 identified studies, 10 met the inclusion criteria and encompassed 283 patients. At 1, 3, 6, and 12 months, statistically significant improvements in MDs in visual analog scale (VAS) scores were noted as follows: MD of 16.11 (95% confidence interval [CI], 8.29-23.93) at 1 month; MD of 22.50 (95% CI, 11.45-33.55) at 3 months; MD of 40.40 (95% CI, 27.48-53.32) at 6 months; and MD of 47.04 (95% CI, 39.43-54.66) at 12 months. At 6 months, Disabilities of the Arm, Shoulder and Hand scores showed a statistically significant improvement (MD, 23.92; 95% CI, 9.47-38.37).
CONCLUSIONS: Improvements in Disabilities of the Arm, Shoulder and Hand scores at 6 months (23.92) surpassed MCID criteria for conservatively managed upper-extremity musculoskeletal pathology (10.83)-suggesting that saline solution injections have a clinically relevant effect. VAS MCID criteria are poorly established, but VAS scores at 6 and 12 months surpassed MCID criteria for conservative treatments for common orthopaedic conditions. In all but 1 study, no statistically significant difference in PROMs was found between saline solution and non-saline solution injections.
LEVEL OF EVIDENCE: Level II, meta-analysis of Level I and II randomized controlled trials.
As a rheumatologist, I do not consider these findings to be very relevant as the effect observed may be due to a placebo effect. It is unlikely this study will modify clinical practice.
The improvements may be due to the natural history of the condition. Placebo effect was also not adequately emphasized and the conclusion may be misleading considering that saline injection was not compared to doing-nothing.