Efficacy of repetitive transcranial magnetic stimulation for improving lower limb function in individuals with neurological disorders: A systematic review and meta-analysis of randomized sham-controlled trials.

Two-thirds of the trials did not describe how the randomisation sequence was generated and about three-quarters did not say how allocation was concealed, so the review judged a moderate to high risk of selection bias to be present. Only 11 of the 27 included trials provided data that could be pooled, and in every category except balance data could not be obtained for more than half of the identified comparisons. The authors say this means the pooled effects may be over- or under-estimated. The trials varied widely in who was treated, how long the intervention lasted, how many sessions were given and the stimulation settings used, and the gait results were moderately to substantially heterogeneous. The pooled effects are pre-to-post changes, not changes to last follow-up, because follow-up time points differed between trials. The authors note that an analysis to follow-up might have given different results. Samples were small to medium, and some results rest on one or two small trials, including the large muscle strength effect (2 trials, 50 people) and the freezing of gait effect. In several included trials rTMS was delivered alongside physical therapy, treadmill training or motor imagery practice, in both the real and the sham arms, so the pooled effects describe rTMS added to rehabilitation rather than rTMS on its own. No data could be pooled for people with multiple sclerosis or ALS, and the spasticity analysis rests on a single study, so the review says nothing about rTMS in those groups.

The paper declares that the authors have no conflicts of interest. No funding source or sponsor is stated in the text provided.

Do not read this as proof that rTMS works for walking. Data could be pooled from only 11 of the 27 trials, more than half of the comparisons the review identified had no usable results, and most trials did not report how patients were allocated to real or sham stimulation, so the pooled effects may be over- or under-estimated. The review also found no effect on mobility or balance, and the strong muscle strength result rests on two small spinal cord injury trials.