Ancillary investigations to diagnose parkinsonism: a prospective clinical study.

Various ancillary investigations can assist clinicians in the differential diagnosis of patients with parkinsonism. It is unknown which test offers greatest diagnostic value in clinical practice. We included 156 consecutive patients with parkinsonism, but with an initially uncertain diagnosis. At baseline, all patients underwent extensive clinical testing and the following ancillary investigations: brain magnetic resonance imaging (MRI); (123)I-iodobenzamide single photon-emission computed tomography (IBZM-SPECT); analysis of cerebrospinal fluid (CSF); and anal sphincter electromyography (EMG). The final diagnosis was established after 3-year follow-up by two movement disorder specialists, according to international consensus criteria. We determined the diagnostic value by comparing the baseline clinical parameters and ancillary studies with the final diagnosis. Out of a potential 138 parameters, univariate analysis identified 35 parameters that discriminated Parkinson's disease (PD, n = 62) and atypical parkinsonism (AP, n = 94), with AUC of 0.55-0.81. Stepwise logistic regression showed that the combination of tandem gait, axial UPDRS subscore, slow saccadic eye movements and dysphagia yielded an AUC of 0.93, adjusted for optimism. The combination of tandem gait and axial UDPRS subscore yielded an AUC of 0.90. None of the ancillary investigations alone or in combination with clinical testing improved this clinically based diagnostic accuracy, not even in a subgroup of patients with the greatest diagnostic uncertainty at baseline. Our study demonstrates that a comprehensive set of clinical tests provides good accuracy to differentiate PD from AP. Our results also suggest that routine MRI, IBZM-SPECT, CSF analysis and anal sphincter EMG do not improve this diagnostic accuracy. Future work should evaluate the possible diagnostic value of more advanced diagnostic tests.