Physiotherapy is an established component of stroke rehabilitation but uncertainties remain about the most appropriate intensity of therapy input. We conducted a systematic review of the randomised trials of physiotherapy after stroke where qualitatively similar therapy regimens were provided at different levels of intensity. A heterogeneous group of seven randomised trials (597 patients) was identified. Dichotomous outcomes (death or the combined poor outcome of death or deterioration) were analysed by use of the odds ratio and 95% confidence interval. Patients subjected to more intensive physiotherapy input showed a non-significant reduction in case fatality (odds ratio 0.60; 95% CI 0.33 to 1.09) and a significant reduction (OR 0.54; 95% CI 0.34 to 0.85; p < 0.01) in the combined poor outcome of death or deterioration by the end of follow-up. Two statistical techniques were used to identify patterns within the continuous data. Firstly, impairment and disability scores were converted to a standardised measure of 0 to 100 and the weighted mean difference (WMD) between the scores in the intensive and conventional physiotherapy groups were then calculated. Modest improvements were observed in both the impairment (WMD+5; 95% CI-1 to 11) and disability scores (WMD+5; 95% CI 0 to 10) recorded at the initial review (median 3 months post-stroke), but not at the final review (median 1 year post-stroke). Secondly, Fisher's inverse Chi-squared test was used to combine the p values from individual trials; this confirmed the above findings (p < 0.05 at initial review; p > 0.05 at final review). More intensive physiotherapy input was associated with a reduction in the combined poor outcome of death or deterioration and may enhance the rate of recovery. These observations warrant further investigation.

Full text (sometimes free) may be available at these link(s):      help

• PubMed
• PDF locator
• publisher

A brief summary and a critical assessment of this review may be available at DARE