Base-peak assessment of levodopa response and detection of fluctuating patients in Parkinson's disease.

Abstract

This study aims to evaluate the base-peak difference in levodopa response for detecting patients with motor fluctuations in Parkinson's disease (PD). Two independent PD samples were evaluated at baseline and 2 h after the administration of the usual morning dose of levodopa using the Unified Parkinson's Disease Rating Scale part III (UPDRS-III). The presence of motor fluctuations was defined by the UPDRS-IV. We quantified the magnitude of motor variation as absolute (Δ) and percentage (Δ%) change in UPDRS-IIIscores. Optimal cut-offs for each index distinguishing patients with or without fluctuations were calculated on the exploratory population. The accuracy of the identified cut-offs was then verified in a testing population. In the exploratory population (N = 26), the optimal cut-off for detecting fluctuations was a Δ of 6 points and a Δ% of 18.4%. When we applied the identified thresholds to the testing population (N = 139), we found a sensitivity of 93.8% (95% CI: 89.7 to 97.8) and a specificity of 91.2% (95% CI: 86.5 to 95.9) for Δ, 83.3% (95% CI: 77.1 to 89.5) and 86.8% (95% CI: 81.2 to 92.4) for Δ%, respectively. The assessment of levodopa usual morning dose response through the base-peak evaluation represents an accurate method for detecting parkinsonian patients with motor fluctuations, and for defining the Minimal Important Difference (MID) in levodopa response suggesting the presence of motor fluctuations in PD patients demanding further therapeutic interventions.