Response to Letter by Shah
Dr Shah makes 5 criticisms of our randomized trial of splinting the wrist after stroke.1 We respond to each in turn.
Randomization was based on the diagnosis and time from onset, not the degree of spasticity: The randomization schedule in our trial was unrestricted; it was not, as Dr Shah suggests, “based on the diagnosis [or] time from onset.” Dr Shah insists that randomization “should be based on the degree of spasticity.” Presumably, he means that we should have stratified allocation by spasticity, but he does not provide a coherent justification for that assertion. There is no methodological imperative to stratify in a randomized trial.2
Reference to “wrist contracture” in the title is misleading: The trial was designed to determine if splinting the wrist after stroke prevents contracture (in people who do not yet have contracture) or reverses contracture (in people who already have contracture). We did not distinguish between these aims of splinting because splinting is widely used for both aims in current clinical practice. Thus, we feel our title (“Effects of splinting on wrist contracture after stroke… ”) is justified.
It seemed “most unusual” that most screened patients could not actively extend the wrist: The 95 patients were those who potentially met the inclusion criteria. Inclusion criteria included no active wrist extension. Patients with active wrist extension were therefore not usually screened.
The splint should not be called a volar splint: The terminology we have used is consistent with the terminology used by others.3,4
The neutral and 45° positions are inappropriate positions for splinting these patients: The neutral position is the position in which wrist splints are most commonly applied after stroke. We agree that there is little in the way of a strong rationale for splinting in this position but we wanted to test the effectiveness of splinting as it is commonly applied. We also tested the effect of splinting the wrist and fingers in an extended position based on animal studies of responses of muscles to sustained stretch. We hypothesized (incorrectly, as it turned out) that the application of sustained stretch to the extrinsic wrist and finger flexor muscles in this position would provide the stimulus required to prevent or reverse contracture.
Dr Shah insists that we should have splinted the wrist “just beyond the point of the stretch reflex,” and asserts that wrists should not be splinted in extension. We do not find his justification (that the extended wrist position is “beyond the point at which the stretch melts away from the homonymous inhibitory Golgi tendon organ”) convincing. Stretch reflexes are highly variable and the angle at which a stretch reflex is elicited depends on many factors including the person’s emotional state, medication levels, and the muscle’s stretch history. We would be very surprised if it was possible to identify “the point of the stretch reflex” with any accuracy. Moreover, we question the relevance of the stretch reflex, as animal studies show that longitudinal growth in muscles appears to depend on muscle length but is independent of the degree of muscle contraction.5