Abstract

Five to fifteen per cent of de novo patients recruited for recent clinical trials of anti-parkinsonian drugs had no evidence of nigrostriatal dopamine denervation on functional imaging. These diagnostically challenging subjects without evidence of dopaminergic deficit (SWEDDs) had tremulous syndromes other than Parkinson's disease (PD). Our objective was to analyse the accuracy of spirography in distinguishing cases of tremor dominant PD from tremulous SWEDDs cases. Analyses were carried out by observers blinded to the clinical data and supplied with just spiral drawings, from which tremor severity, 3-turn spiral diameter and spiral density were measured. A spiral coefficient, averaged for the spirals drawn by each hand, was derived from these three indices. A cut off of <4 in the coefficient was taken to indicate PD. Of the 65 cases analysed, the data were felt to be of insufficient quality in 6. Of the remaining 59 cases, the sensitivity and specificity for differentiating tremor dominant PD from tremulous SWEDDs was 65.2% and 61.1% respectively. An analysis was also performed looking at the individual spiral components. This showed that the sensitivity and specificity for tremor severity were 62.5% and 74.3%, 3-turn diameter 75% and 77.8% and spiral density 28% and 67.3% respectively for predicting PD. The simple 3-turn spiral diameter has similar sensitivity and specificity for distinguishing PD from SWEDDs as reported for two blind PD experts assessing these patients from standardised videotapes.