Are medium and long latency reflexes a screening tool for early Parkinson's disease?

We have studied whether assessment of medium latency (ML) and long latency (LL) reflex amplitudes may serve as a marker for early Parkinson's disease. Twenty-three patients with idiopathic Parkinson's disease (Hoehn and Yahr stage 1 to 4) and 24 controls received 20 4 degrees toe-up rotations of a platform upon which they were standing. All antiparkinsonian medication was withheld for at least 12 h before the study. ML reflexes in the stretched gastrocnemius muscle and LL reflexes in the shortened tibialis anterior muscle were recorded from both legs. ML responses were significantly enhanced in patients compared to controls. In contrast to previous studies which studied patients who continued their usual treatment, we observed that LL responses were significantly reduced in patients compared to controls. For the purpose of individual analysis, we subsequently determined the optimal specificity and sensitivity using various criteria for abnormality. The presence of either enhanced ML responses or reduced LL responses (or both) in at least one leg yielded a maximum sensitivity of 65.2% with a specificity of 75.0% (positive likelihood ratio 2.6; negative likelihood ratio 0.5). Abnormal reflexes were almost exclusively present in patients with advanced and long-standing Parkinson's disease. These results show abnormalities of ML and LL responses in advanced Parkinson's disease, but render it unlikely that these abnormalities are a suitable screening tool for early stages of the disease. The fact that LL responses were reduced in patients taken off antiparkinsonian medication raises the possibility that this reflex is under supraspinal dopaminergic control.