Table 1.
Summary of class I and II evidence for duodenal levodopa infusion and deep brain stimulation in Parkinson’s disease patients with motor fluctuations and dyskinesia
| Study | Therapy versus comparator | Study type | Patients n; age ± SD (range) (years); disease duration ± SD (range) (years) | Follow-up | Key study outcomes | |
|---|---|---|---|---|---|---|
| Efficacy | Safety | |||||
| Class I studies | ||||||
| Deep brain stimulation (subthalamic nucleus or globus pallidus) | ||||||
| Follett [5] | Bilateral STN–DBS versus bilateral GPi–DBS | Long-term results of an observer-blind RCT [7] | 299; 61.9 ± 8.7 (STN), 61.8 ± 8.7 (GPi); 11.1 ± 5.0 (STN), 11.5 ± 5.4 (GPi) | 24 months |
No differences in efficacy between therapies. Patients undergoing STN–DBS required lower dose of dopaminergic agents versus GPi–DBS (p = 0.02) |
Visuomotor component of processing speed declined more after STN–DBS versus GPi–DBS (p = 0.03) Depression worsened after STN–DBS, improved after GPi–DBS (p = 0.02) 335 serious AEs in 83 STN–DBS patients and 77 GPi–DBS patients; no significant between-group differences at 24 m |
| Okun [6] | Unilateral STN–DBS versus unilateral GPi–DBS | Double-blind RCT | 45; 60 ± 8.2; 12.9 ± 3.8 | 7 months |
No significant differences in primary outcome measures (mood, cognition) between STN–DBS and GPi–DBS in the optimal DBS state Similar motor improvement observed with STN–DBS and GPi–DBS |
Adverse mood effects occurred ventrally in both targets Worsening of letter verbal fluency with STN–DBS Persistence of deterioration in verbal fluency in ‘off’ STN–DBS state suggestive of surgical rather than stimulation-induced effect |
| Weaver [7] | DBS (STN or GPi) versus BMT | Observer-blind RCT | 255; 62.4 ± 8.9 (37–83); 12.4 ± 5.8 | 6 months | With STN– or GPi–DBS versus BMT, significantly greater improvements in: |
Neurocognitive testing suggested small decrements in some areas of information processing with DBS versus BMT DBS associated with increased risk of serious AEs versus BMT (49 vs. 15 patients, respectively; p < 0.001); 99 % resolved by 6 months Rate of non-serious AEs higher in older versus younger patients, but rate of serious AEs was comparable |
| ‘On’ time (without troubling dyskinesia) (p < 0.001) | ||||||
| Quality of life (p < 0.001) | ||||||
| Similar benefits observed in younger and older patients | ||||||
| Anderson [8] | Bilateral STN–DBS versus bilateral GPi–DBS | Extension of a double-blind RCT [8] | 23; 61 ± 9 (STN), 54 ± 12 (GPi); 15.6 ± 5 (STN), 10.3 ± 2 (GPi) | 12 months | With STN–DBS versus GPi–DBS: | Tendency towards more cognitive and behavioral changes with STN–DBS versus GPi–DBS, mostly mild and transient |
| Tendency towards greater levodopa dose reduction | ||||||
| Tendency towards greater improvement in bradykinesia | ||||||
| Similar improvement in off-medication UPDRS motor scores | ||||||
| Similar improvement in dyskinesia | ||||||
| No improvement in on-medication function in either group | ||||||
| Smeding [9] | Bilateral STN–DBS versus unilateral GPi–DBS | Substudy of an observer-blind RCT [10] | 34; 59.2 ± 8.6 (STN), 62.1 ± 8.1 (GPi); 13 (3–50) (STN), 11 (7–20) (GPi) | 12 months | With STN–DBS versus GPi–DBS: | One STN–DBS patient showed severe confusion and cognitive decline after surgery |
| Significantly smaller improvement in two tests of executive function (stroop color word; trailmaking) at 6 months (p < 0.05) | ||||||
| No significant differences at 12 months | ||||||
| Esselink [10] | Bilateral STN–DBS versus unilateral pallidotomy | Observer-blind RCT | 34, 61 (range 55–66) (STN), 62 (range 57–68) (pallitodomy) | 6 months | With STN–DBS versus pallitodomy, significantly greater improvements in: |
More pallitodomy patients experienced AEs versus STN–DBS patients One major AE in each group |
| Off UPDRS motor scores (p = 0.002) | ||||||
| On UPDRS motor (p = 0.02) and duration of dyskinesia (p = 0.004) | ||||||
| Reduction of anti-PD drugs (p = 0.02) | ||||||
| Continuous duodenal levodopa infusion | ||||||
| Nyholm [71] | Levodopa/carbidopa gel infusion versus conventional treatment | Observer-blind, crossover RCT | 24; median 68 (51–79) (oral/infus), median 64 (50–75) (infus/oral); NS | 3 + 3 weeks | With DLI versus conventional treatment, significantly greater improvements in: | AEs (e.g., dyskinesia/hyperkinesia, constipation, depression, etc.) similar for DLI versus conventional treatment |
| Functional ‘on’ time (p < 0.01) | ||||||
| Off time (p < 0.01) | ||||||
| Median UPDRS score (p < 0.05) | ||||||
| Quality of life (p < 0.01) | ||||||
| Dyskinesia unchanged versus baseline | ||||||
| Class II studies | ||||||
| Deep brain stimulation (subthalamic nucleus or globus pallidus) | ||||||
| Williams [12] | DBS (STN or GPi) versus BMT | Open label, RCT | 366; 59 (37–79) (DBS), 59 (36–75) (BMT); 11.5 (2.0–32.2) (DBS), 11.2 (1.0–30.0) (BMT) | 12 months |
At 1 year, 75 DBS versus 21 BMT patients reported no waking day dyskinesia (p < 0.0001) and 45 DBS versus 5 BMT patients reported no off time (p < 0.0001) Compared with baseline, mean improvement in PDQ-39 was 5.0 points with DBS versus 0.3 points with BMT (p = 0.001) Difference between DBS versus BMT in mean change in PDQ-39 mobility domain was −8.9 (p = 0.0004), activities of daily living domain was −12.4 (p < 0.0001), bodily discomfort domain was −7.5 (p = 0.004) |
In total, 19 % DBS patients had serious surgery-related AEs; there were no suicides but there was one procedure-related death Twenty patients in the DBS group and 13 in the BMT group had serious AEs related to PD and drug treatment |
| Witt [11] | Bilateral STN–DBS versus BMT | Open label, RCT (ancillary study to Deuschl [14]) | 123; 60.2 ± 7.9 (DBS), 59.4 ± 7.5 (BMT); 13.8 ± 6.3 (DBS), 14.0 ± 6.1 (BMT) | 6 months |
Overall, STN–DBS did not reduce cognition and affectivity With STN–DBS versus BMT: |
Severe psychiatric AEs observed in ten STN–DBS patients and eight BMT patients |
| No significant difference in scores for overall cognition and affectivity | ||||||
| Significantly greater reduction in anxiety (p < 0.0001) | ||||||
| Adverse changes in verbal fluency and performance in the Stroop test not associated with changes in psychiatric scales and did not affect improvements in QoL | ||||||
| Schüpbach [13] | Bilateral STN–DBS versus BMT | Open label, RCT | 20; 48.4 ± 3.3 (DBS), 48.5 ± 3.0 (BMT); 7.2 ± 1.2 (DBS), 6.4 ± 1.1 (BMT) | 18 months |
QoL improved to a greater extent with STN–DBS versus BMT (p < 0.05) After 18 months, severity of parkinsonian motor signs in medication-off conditions, levodopa-induced motor complications, and daily levodopa dose reduced with STN–DBS versus baseline and increased with BMT versus baseline |
AEs were mild or transient Overall psychiatric morbidity and anxiety improved with DBS |
| Deuschl [14] | Bilateral STN–DBS + BMT versus BMT alone | Open label, paired RCT | 156; 60.5 ± 7.4 (DBS), 60.8 ± 7.8 (BMT); NS | 6 months |
DBS + BMT had greater beneficial effect than BMT alone on: QoL (PDQ-39; p = 0.02) Severity of symptoms without medication (UPDRS-III; p < 0.001) |
Serious AEs more common with DBS + BMT (10 events) versus BMT alone (3 events) (p < 0.04), but majority resolved without permanent complications Frequency of AEs higher with BMT alone versus DBS + BMT (p = 0.08) |
| Continuous duodenal levodopa infusion | ||||||
| Nyholm [73] | Oral sustained-release levodopa versus levodopa/carbidopa gel infusion | Open label, crossover RCT | 12; 61.2 ± 11.0 (39–76); NS | 3 + 3 weeks |
With DLI versus oral levodopa: Significantly lower average intra-individual coefficient of variation for plasma levodopa concentration (p < 0.01) Significantly greater increase in on time and decrease in off time and dyskinesia (p < 0.01) |
No major complications or serious AEs with either therapy |
AEs adverse events, BMT best medical therapy, except infusions, DBS deep brain stimulation, NS not specified, QoL quality of life, RCT randomized controlled trial