Assessment of imbalance in idiopathic Parkinson's disease (IPD) usually relies on semiquantitative ratings. Posturography has been proposed as an objective means to assess imbalance but its relationship to clinical disequilibrium is questionable. In this study static and dynamic posturography was performed in 58 patients with IPD and 29 healthy controls. In patients, posturography was related to performance in established clinical tests (pull-test and tandem gait). Posturography did not differentiate between controls and patients with impaired pull-test (IPDimb, n 5 28). Patients with normal pull-test (IPDstab n 5 30) had lower sway than controls in static (P 5 0.042) and dynamic posturography (P 5 0.001) and also differed from patients with impaired pull-test in static (P 5 0.007) and dynamic (P < 0.001) conditions. In patients with side-steps in tandem gait (n 5 21), sway in static and dynamic posturography was increased. Sway measures did not differentiate between patients with pull-test scores 1 and 2 or one and >1 side step in tandem-gait, respectively. Results of ANOVA showed that variance of static posturography was related to performance in tandem-gait (P < 0.0001) but not to pulltest performance (P 5 0.91). In contrast, dynamic posturography was related to both, tandem-gait (P 5 0.012) and pull-test (P 5 0.03). Posturographic sway is increased in patients with IPD with disturbance of tandem gait and pull-test. Posturographic measures did not distinguish between different degrees of deficits in clinical tests. V