We thank Morrish for his comments 1 about our study, 2 and would like to reply as follows.
A high positive predictive value for diagnosing idiopathic Parkinson's disease (IPD) at the last premortem visit in patients with mean disease duration of 13 years 3 is unhelpful when we are considering early or de novo cases. We and others consistently have found that between 4 and 14% of cases thought to be IPD have normal presynaptic imaging using single photon emission computed tomography (SPECT) or positron emission tomography (PET). Because of the long presymptomatic phase of IPD, the theoretical overlap between normal and abnormal imaging results that Morrish estimates was not seen even in the early, diagnostically uncertain population (see dot plot of FP-CIT imaging results 4 ).
We absolutely disagree that a normal scan in a diagnostically difficult case "simply adds confusion." The 11% normal PET scans in the REAL-PET study remained normal at 2 years, 5 and the authors provisionally suggest essential tremor, but we await their full report. A false-negative baseline test, at least in a subset of cases, would become abnormal when the test was repeated (partly for statistical reasons and partly because further dopamine depletion occurs in 2 years). This finding was not seen, so the Morrish "overlap" hypothesis between normal and abnormal dopamine activity is probably incorrect in this clinical setting.
It is unhelpful of Morrish to confuse the issues over differential diagnosis of parkinsonism and tremor by shifting the frame of his comments to akinetic-rigid syndrome. We agree that there are potential differences in the cost-effectiveness calculations for functional diagnostic imaging for tremor-positive and tremor-negative movement disorders, and this is an area we are currently evaluating. We also value the opinion of our renowned colleagues whose majority view is that presynaptic dopaminergic imaging makes a significant contribution to the diagnosis of movement disorder patients.