Prediction of the Levodopa Challenge Test in Parkinson's Disease Using Data from a Wrist-Worn Sensor

Abstract

The response to levodopa (LR) is important for managing Parkinson's Disease and is measured with clinical scales prior to (OFF) and after (ON) levodopa. The aim of this study was to ascertain whether an ambulatory wearable device could predict the LR from the response to the first morning dose. The ON and OFF scores were sorted into six categories of severity so that separating Parkinson's Kinetigraph (PKG) features corresponding to the ON and OFF scores became a multi-class classification problem according to whether they fell below or above the threshold for each class. Candidate features were extracted from the PKG data and matched to the class labels. Several linear and non-linear candidate statistical models were examined and compared to classify the six categories of severity. The resulting model predicted a clinically significant LR with an area under the receiver operator curve of 0.92. This study shows that ambulatory data could be used to identify a clinically significant response to levodopa. This study has also identified practical steps that would enhance the reliability of this test in future studies.

Keywords: Parkinson’s Disease; ambulatory systems; levodopa challenge test; levodopa response; machine learning; wearable devices.

Figure 1

Stylized representation of one day of the Parkinson’s Kinetigraph (PKG) recording. The Y axis shows the PKG’s bradykinesia score in bradykinesia score (BKS) units and the X axis is time in minutes, before after the first reminder of the morning (red vertical dotted line at “0” time). The acknowledgement that the dose was consumed is shown as a red diamond. The dots represent individual BKS for each two-minute epoch: green dots represent epochs that lie in the “inactive” range and red within the active range. The green line represents the smoothed time series from all 6 days of recording, with the heavy line being from 46 min to 90 min after the first acknowledgement of the reminder (red diamond at ~6 mins). The apricot shading area shows the ten minutes (5 epochs, circled) used to establish the bradykinesia scores defined in Section 2.3 at the time of the dose (DT) and around the time of the peak (ET).

Figure 2

(A) shows the range of total Levodopa Equivalent dose (LED) and the dose of levodopa (L-Dopa). (B) shows the LED from the first dose of levodopa (1st dose) and from D2 agonists over the course of the day. (C) shows the percentage of the LED contributed to by D2 agonists.

Figure 3

This shows the relationship between %∆UPDRS (Y axis) and abs∆UPDRS (X axis). The grey circles represent all PwP. Gray dots with a green center are subjects with disease duration of 4 years or less and those with a red center are PwP with disease duration of 10 or more years. The vertical apricot shaded region shows the “uncertain” zone (see Section 3.1.2). To the right of this shade area are cases where the abs∆UPDRS was considered to be a clinical meaningful increase (see text for criteria) whereas to the left, abs∆UPDRS was not clinically meaningful. The 3 horizontal lines indicate the three commonly used %∆UPDRS, showing that a region of clinical uncertainty also exists.

Figure 4

These two plots are box and whiskers plots of the distribution of Class 0 and Class 1 in Table 8 (A) and 10 (B) plotted according the 4 corresponding column groups in that table: “U” indicates “uncertain”, “O” indicates “Already ON” and “V” indicates “Variable”. The small, pink shaded “box and whiskers” plot, between Case 0 and Case 1 in (A) in the Exclude U group shows the distribution of abs∆PKG of the uncertain cases. The boxes are the median and quartiles with the “whiskers” showing the 90th and 10th percentile.

Figure 5

(A,B) show the change in LR according to duration of disease (in years). (A) shows the abs∆_PKG and 5B shows the abs∆_UPDRS. (note that 1 unit on the Y axis of (A) approximates 12 UPDRS III units shown on the Y axis of (B)). (C) shows abs∆_PKG before and after deep brain stimulation (DBS). (D) shows the same data, with the difference in abs∆_PKG before and after DBS (X axis) plotted against the abs∆_PKG before DBS. The grey region marked NSD (no significant response) in (D) indicates the absolute delta before DBS that is not significant. In (A–C) the boxes are the median and quartiles with the “whiskers” showing the 90th and 10th percentile.