Response to Letter by Marossezky et al
The letter by Marossezky et al takes issue with the design of our recent randomized trial of splinting early after stroke,1 suggesting that our findings failed to show that “routine splinting soon after stroke should be discontinued.”
They believe that we did not define “contracture.” We take this opportunity to clarify that we use the term contracture to refer to a reduction in the slack length or compliance of the relaxed muscle.2 Clinically, contracture manifests as a loss of range of motion or an increased resistance to passive movement.3 Much of our article refers to muscle extensibility to make clear the intent of our study.
In addition, Marossezky et al suggest that the participants of our study “did not appear to have, nor be at risk of, developing excessive wrist or finger contracture” and were “unlikely” in the acute stage of stroke rehabilitation to develop contracture. If our only objective was simply to reverse contracture, their argument concerning the minimal contracture at baseline could be well taken. But in fact our intention was broader in scope because we wished to examine the effect of splinting on preventing as well as reversing contracture. In acute rehabilitation settings, people with no active movement after stroke are typically splinted because they have no active wrist or finger extension; our study sought to test the effectiveness of this common clinical practice. Further, in line with previous work by Turton et al4 and Ada et al,3 our data showed that there was a progressive reduction in mean wrist range of motion during the 6-week phase of the study. By the end of the study, the mean wrist range of motion was about 22° less than at baseline.
Their criticism that our study is “underpowered” should also be discounted because it overlooks the narrow confidence intervals around the estimate of the mean treatment effect for the primary outcome of muscle extensibility.5 We did not design the study to detect subgroup effects, partly because of the well-known problems associated with tests of subgroup effects in clinical trials.6
Finally, the criticism that the study design did not provide the expected levels of acute neurological rehabilitation suggests that they are viewing current rehabilitation practices through rose-colored glasses. Splints were worn in the study for an average of 11 hours, far greater than the stretch commonly provided in rehabilitation. The study deliberately controlled the amount of wrist extension practice to 10 minutes per day; in Australia at least patients receive between 4 and 11 minutes of upperlimb therapy per session during the period of acute stroke rehabilitation.7 Despite the assertions of Marrossezky et al, there was no limit imposed on “functional use” of the upper limb in our studies. These observations suggest to us that it is not “current acute neurological rehabilitation” practice for patients to receive greater amounts of therapy than that provided to participants in our trial, and in fact many receive less.8
We stand by our assertion that the routine practice of hand-splinting to prevent muscle contracture during acute rehabilitation after stroke should be discontinued.
Lannin NA, Cusick A, McCluskey A, Herbert RD. Effects of splinting on wrist contracture after stroke: a randomized controlled trial. Stroke. 2007; 38: 111–116.
Gordon J. Disorders of motor control. In: Ada L and Canning C, eds. Key Issues in Neurological Physiotherapy Oxford: Butterworth-Heinemann Medical; 1990: 25–50.
Turton AJ, Britton E. A pilot randomised controlled trial of daily muscle stretch regime to prevent contractures in the arm after stroke. Clinical Rehabilitation. 2005; 19: 600–612.
Journal review: a randomised controlled trial of splinting the hand in brain damaged patients. Advances in Clinical Neuroscience and Rehabilitation. 2003; 3: 32.