Background The presence of significant forearm bone torsion might affect planning and evaluating treatment regimes in cerebral palsy patients. We aimed to evaluate the influence of longstanding wrist flexion, ulnar deviation, and forearm pronation due to spasticity on the bone geometries of radius and ulna. Furthermore, we aimed to model the hypothetical influence of these deformities on potential maximal moment balance for forearm rotation. Methods Geometrical measures were determined in hemiplegic cerebral palsy patients (n = 5) and healthy controls (n = 5). Bilateral differences between the spastic arm and the unaffected side were compared to bilateral differences between the dominant and non-dominant side in the healthy controls. Hypothetical effects of bone torsion on potential maximal forearm rotation moment were calculated using an existing anatomical muscle model. Findings Patients showed significantly smaller (radius: 41.6%; ulna: 32.9%) and shorter (radius: 9.1%; ulna: 8.4%) forearm bones in the non-dominant arm than in the dominant arm compared to controls (radius: 2.4%; ulna 2.5% and radius: 1.5%; ulna: 1.0% respectively). Furthermore, patients showed a significantly higher torsion angle difference (radius: 24.1°; ulna: 26.2°) in both forearm bones between arms than controls (radius: 2.0°; ulna 1.0°). The model predicted an approximate decrease of 30% of potential maximal supination moment as a consequence of bone torsion. Interpretation Torsion in the bones of the spastic forearm is likely to influence potential maximal moment balance and thus forearm rotation function. In clinical practice, bone torsion should be considered when evaluating movement limitations especially in children with longstanding spasticity of the upper extremity. © 2014 Elsevier Ltd.