Short-term effect of prism adaptation treatment on severity of unilateral spatial neglect following right hemispheric stroke: a systematic review and meta-analysis.

Only 7 trials with 227 participants were eligible, and some subgroup analyses rested on as few as 2 trials, so those results could change with more studies. None of the included trials was judged at low risk of bias. Three were rated high risk, mainly because the randomisation process was not described, so the trials may not have been truly comparable at the start. The authors rated the certainty of the evidence very low for the main analysis of neglect symptoms and low or very low for the subgroups. The positive pooled result is therefore weak evidence, not a settled finding. Some trial results were available only as graphs and were read off with a digitising tool, which may have changed the numbers. Only articles published in English were included, so negative trials published elsewhere may be missing. The chronic-phase finding rests on 2 trials and did not reach significance (P = 0.08), so it cannot support a claim that prism adaptation works better later after stroke. The review addresses short-term effects only. It says nothing about whether any benefit lasts.

The authors declare they have no conflicts of interest. The paper does not state who funded the work.

Do not read the significant pooled result as proof that prism adaptation works for neglect. The authors rated the certainty of that result very low, none of the 7 trials was judged at low risk of bias, and the gain was on paper-and-pencil neglect tests: the pooled result for daily function was null. The finding that prisms of 10° or more worked is a subgroup result from a handful of small trials, not a direct comparison of prism strengths, so it does not establish that a bigger prism is better.