Constraint-induced movement therapy (CIMT) for hemiparetic children in comparison to bimanual treatment – what is (more) effective?

Objectives: The efficacy of CIMT in children with cerebral palsy has been proven in several studies (Taub et al 2004, Eliasson et al 2005, Charles et al 2006). Yet it is unclear what the main principles of efficacy are – restraint, structured therapy or high therapy intensity? To clarify the importance of restriction we have compared our child-friendly interdisciplinary (kid-)CIMT program with an equally intensive well-structured bimanual program.

Method: Prospective, randomized, controlled intervention study.

Participants: 42 children with unilateral cerebral palsy or other central hemiparesis, aged 3 to 12yrs.

In the (kid)-CIMT group (n=24) children were treated with restraint of the non-involved arm 4 hrs daily over 3 weeks in single-therapy sessions. In week 4 bimanual training of daily activities was performed with the same intensity.

In the bimanual group (n=18) children also received therapy for 4 hrs daily over 4 weeks with emphasis on everyday life and playing activities.

Assessment instruments: Melbourne Assessment of Unilateral Upper Limb Function (MA)

Assisting Hand Assessment (AHA)

Results: Both groups showed clear improvement in hand and arm motor function after treatment.

In the evaluation with the Melbourne Assessment (task-related isolated movements of the involved arm) the CIMT group performed significantly better than the bimanual group: +7.7 vs. +2.8 points (p<0.02). However in AHA outcome (use of the paretic hand in bimanual practice) there was no superiority for the CIMT group: +3.8 vs. +3.1 (n.s.)

Comparing the 50% patients with better and the 50% with poorer AHA scores before treatment the best outcome could be found for the most impaired children under CIMT therapy (+ 5.2 points).

Conclusions: Our study reveals similar improvements for CIMT and bimanual treatment. Only for isolated functions of the paretic arm the CIMT group shows significantly better results, yet in bimanual functions and activities of daily living there is no significant difference.

Thus the restraint in CIMT seems actually to be important for the development of new motor functions, but children do not benefit from this advantage in spontaneous bimanual activities. We conclude that hand-arm motor training in hemiparetic children should be much more tailored to the individual child and that CIMT is not the only effective approach.