OBJECTIVE: To evaluate the efficacy and safety of short-wave therapy with sham or no intervention for the management of patients with knee osteoarthritis.
METHODS: We searched the following databases from their inception up to 26 October 2016: MEDLINE, CENTRAL, EMBASE, Physiotherapy Evidence Database, CINAHL and OpenGrey. Studies included randomized controlled trials compared with a sham or no intervention in patients with knee osteoarthritis. The results were calculated via standardized mean difference (SMD) and risk ratio for continuous variables outcomes as well as dichotomous variables, respectively. Heterogeneity was explored by the I2 test and inverse-variance random effects analysis was applied to all studies.
RESULTS: Eight trials (542 patients) met the inclusion criteria. The effect of short-wave therapy on pain was found positive (SMD, -0.53; 95% CI, -0.84 to -0.21). The pain subgroup showed that patients received pulse modality achieved clinical improvement (SMD, -0.83; 95% CI, -1.14 to -0.52) and the pain scale in female patients decreased (SMD, -0.53; 95% CI, -0.98 to -0.08). In terms of extensor strength, short-wave therapy was superior to the control group ( p < 0.05, I2 = 0%). There was no significant difference in the physical function (SMD, -0.16; 95% CI, -0.36 to 0.05). For adverse effects, there was no significant difference between the treatment and control group.
CONCLUSION: Short-wave therapy is beneficial for relieving pain caused by knee osteoarthritis (the pulse modality seems superior to the continuous modality), and knee extensor muscle combining with isokinetic strength. Function is not improved.
Certainly knee arthritis is a common problem, and current treatments are limited and not necessarily effective, so finding other treatments is a laudable goal. However, I'm not sure that this article adds much. I can't tell how much of the information in this review is new compared with the prior review of 2012 that the authors referenced. Despite the fact that this was a systematic review, the number of studies and patients was small, so there is a good chance of bias. Also, some of the studies did not have a control group, and some did not disclose their funding source, so these are other sources of bias. I'll be interested in seeing new information, but this will not change my current practice. PS--it was difficult to read in places because the translation to English was not very good.
Standardized mean differences are difficult to interpret whereas mean differences have more direct meaning, e. g. for VAS. It would have been useful to have direct metrics in addition to the standardized ones.