Table 2.
Interventional TMS studies to improve LID in patients with Parkinson’s disease
| First author, year Sample size Brain target |
TMS protocol | Main findings | Clinical dyskinesia score | Score before TMS | Score after sham TMS | Score after real TMS | Pairwise testing (P-value) | ANOVA (P-value) | |
|---|---|---|---|---|---|---|---|---|---|
|
8 patients Target: SMA |
Single rTMS session at 1 or 5 Hz for 15 min with continuous apomorphine infusion | 1-Hz rTMS session reduced AIMS for 15 min after rTMS; 5-Hz rTMS session had no effect | AIMS |
6.75 ± 2.0 6.00 ± 1.8 |
6.35 ± 1.9 6.35 ± 1.9 |
2.25 ± 1.8 7.38 ± 2.7 |
P < 0.01a |
P < 0.00b P < 0.12b |
|
|
10 patients Target: SMA |
Single and multiple 1-Hz rTMS sessions for 15 min after levodopa intake | Single 1-Hz rTMS session reduced AIMS for 15 min after rTMS; Multiple 1-Hz rTMS sessions (1/day, 5 days) had same effect | AIMS |
0.0 ± 0.0c 0.0 ± 0.0c |
6.62 ± 1.0 6.62 ± 1.0 |
1.37 ± 1.5 1.45 ± 1.4 |
P < 0.001b P < 0.001b |
||
|
Lohse et al., 2020 (data reported in this publication) 17 patients Target: PreSMA |
Single 1-Hz rTMS session for 30 min before levodopa intake | Single-session 1 Hz reduced and delayed UDysRS |
UDysRS Time to LID onset |
12d 24.5 min |
9 36.5 min |
P = 0.032e P = 0.019e |
|||
|
9 patients (single) 6 patients (multiple) Target: M1 |
Single bilateral 1-Hz rTMS session for 16 min after levodopa intake; Multiple bilateral 1-Hz rTMS sessions for 32 min during continued levodopa treatment | Single 1-Hz rTMS session did not produce an anti-dyskinetic effect Multiple sessions (2/day, 5 days) did not produce an anti-dyskinetic effect |
AIMS AIMS UPDRSIV PDYS-26 |
0 ± 0 12 ± 5 4.8 ± 0.8 22 ± 11 |
8 ± 5 11 ± 5 4.7 ± 0.5 22 ± 12 (Day 5) |
10 ± 5 8 ± 5 4.8 ± 0.8 22 ± 11 (Day 5) |
P = 0.11f P = 0.23f P = 0.84f P = 0.40f |
||
|
6 patients Target: M1 |
Multiple 1-Hz rTMS sessions for 15 min after levodopa intake | Multiple sessions (1/day, 10 days in 2 weeks) reduced CAPSIT for 1 day after the last stimulation | CAPSIT-PD | 5.6 ± 2.8 | No sham | 3.5 ± 1.6 (1 day after last rTMS) | P = 0.03g | P < 0.06h | |
|
10 patients Target: M1 |
Multiple 1-Hz rTMS sessions for 32 min during continued levodopa treatment | Multiple sessions (1/day, 4 days) did not reduce CDRS on direct comparison | CDRS | 22.1 ± 8.2 | 20.8 ± 6.7 | 20.4 ± 7.9 | P = 0.22i | P < 0.043j | |
|
17 patients Target: SMA |
Multiple 1-Hz rTMS sessions for 30 min during continued levodopa treatmentk | Multiple sessions (1/day, 10 days) reduced AIMS for 24 h after stimulation | AIMS | 16.61 ± 6.23 | 9.56 ± 4.06l | 13.72 ± 7.28 | P = 0.024m | ||
|
10 patients (single) 20 patients (multiple) Target: Cerebellum |
Single cTBS sessions after levodopa intake; Multiple cTBS sessions after levodopa intakek | Single cTBS session reduced CAPSIT-PD; Multiple cTBS sessions (1/day, 10 days, in 2 weeks) reduced CAPSIT for up to 4 weeks after stimulations | CAPSIT-PD |
∼ 6.5n |
∼5.8n ∼5.8n (3 days after last rTMS) |
∼3.8n ∼3.8n (3 days after last rTMS) |
P = 0.011o P = 0.008p |
P < 0.007o P < 0.0001p |
|
|
11 patients Targets: right IFC and M1 |
Single cTBS session after levodopa intake | Single cTBS session of right IFG reduced AIMS for 30 min after stimulation; No effect of single cTBS session over M1 | AIMS |
6.33 ± 1.4 ∼5.0q |
4.50 ± 1.3 ∼5.0q |
P = 0.04r | |||
|
10 patients Target: right IFC |
Single cTBS session after levodopa intake | Single-session cTBS reduced AIMS relative to sham cTBS | AIMS | ∼3.0c,s | ∼7.0s | ∼4.0s | P = 0.02t | ||
Data are presented as mean ± standard deviation. Studies are sorted by TMS protocol and thereafter by year. AIMS = Abnormal Involuntary Movement Scale; CAPSIT-PD = The Core Assessment Program for Surgical Interventional Therapies in Parkinson’s Disease; cTBS = continuous theta burst transcranial magnetic stimulation; IFC = inferior frontal cortex; M1 = primary motor cortex; PDYS-26 = 26-item Parkinson Disease Dyskinesia Scale; preSMA = pre-supplementary motor area; rTMS = repetitive transcranial magnetic stimulation; SMA = supplementary motor area; UDysRS = Unified Dyskinesia Rating Scale; UPDRS = Unified Parkinson’s Disease Rating Scale.
Wilcoxon test of ‘score after real TMS’ versus ‘pre-TMS score’.
Friedman ANOVA for repeated measures with time as the main effect for the entire 1-Hz TMS condition with multiple time points, not only the score reported in this table.
The pre-score was measured immediately after levodopa intake in patients withdrawn from therapy since the night before.
These are median UDysRS scores.
One-tailed Wilcoxon test for ‘score after real TMS’ versus ‘score after sham TMS’.
Two-way repeated measure ANOVA of interaction between stimulation (sham or real) as between-subject factor and time as within-subjects factor.
Paired t-test. Scores are measured at peak-of-dose dyskinesia.
Repeated measures ANOVA.
One-tailed directional Wilcoxon-matched pairs test for ‘score after real TMS’ versus ‘score after sham TMS’ change from ‘pre-TMS score’.
Friedman ANOVA for ‘pre-TMS score’ versus ‘score after real TMS’ versus ‘score after sham TMS’.
This study used a parallel group design for the multi-session part of the study. All others used a one group cross-over design.
The ‘pre-TMS score’ for the sham group was 8.31 ± 3.52.
Wilcoxon test for ‘score pre-rTMS’ versus ‘score after real-rTMS’ both at 90 min after levodopa.
Scores were estimated taken from Fig. 4.
Respectively a Wilcoxon test and Friedman ANOVA for ‘score after real TMS’ versus ‘score after sham TMS’ both at 30 min after levodopa.
Respectively a Wilcoxon test and Friedman ANOVA for the ‘score after real TMS’ versus the ‘pre-TMS score’ both at 30 min after levodopa.
Scores were estimated taken from Fig. 4.
Wilcoxon test for ‘score after real TMS’ versus ‘score after sham TMS’ both at 30 min after levodopa.
Scores were estimated taken from Fig. 1D.
Wilcoxon t-test for ‘score after real TMS’ versus ‘score after sham TMS’.