Effects of deep brain stimulation on driving in Parkinson’s disease - a prospective controlled study

doi:10.21203/rs.3.rs-2108605/v1

Download PDF

Article

Effects of deep brain stimulation on driving in Parkinson’s disease - a prospective controlled study

https://doi.org/10.21203/rs.3.rs-2108605/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

About 50% of patients with Parkinson’s disease (PD) treated with deep brain stimulation (DBS) actively drive a car. In this controlled, three-armed study, non-demented active drivers with PD and DBS were recruited to prospectively examine their driving simulator performance before (Baseline) and 6–12 months after brain surgery (Follow-up). Data was compared to that of clinically matched PD patients, eligible for but without undergoing DBS and to age-matched healthy controls. Clinical Baseline characteristics and driving performance of PD patients with (n = 23) and without DBS (n = 29) did not differ significantly but driving safety in PD was worse compared to controls (n = 33). At Follow-up, driving performance was still comparable between both PD groups. Disease duration was associated with poorer driving performance. This first prospective study on driving safety in PD after DBS implantation indicates that DBS does not influence driving negatively, but also does not improve driving safety as hypothesized based on previous studies.

Patients with Parkinson’s disease frequently depend on car driving due to impairments of mobility and gait. About 60% of all Parkinson’s disease patients and 50% of those with deep brain stimulation (DBS) actively drive a car.¹ Generally, Parkinson’s disease patients drive less safely^2,3 and quit driving more often than controls.⁴ Age and cognitive impairment are the main risk factors for impaired driving⁵ while slight to moderate motor impairment plays a minor role (reviews:^5–7). Previously, we illustrated in a controlled cross-sectional driving simulator study that Parkinson’s disease patients with subthalamic DBS (STN-DBS) drive safer than patients without DBS despite comparable age but even higher disease severity according to Hoehn & Yahr. Furthermore, DBS patients drove better in the test condition “stimulation on” than with “medication on” alone, despite a comparable positive effect of both conditions on motor disability.⁸ This suggests potential beneficial effects of subthalamic DBS on driving ability besides motor but potentially due to non-motor driving relevant aspects.

However, there is no investigation yet assessing the influence of DBS on driving in Parkinson’s disease patients in a controlled and prospective manner. Due to the increasing number of DBS-treated Parkinson’s disease patients it is clinically relevant to know whether DBS influences driving safety. As motor impairment is not crucial for driving ability in both Parkinson’s disease patients with⁸ and without DBS,^2,9 other factors found to improve after DBS, such as implicit procedural learning, sequence learning, goal-directed action selection or decision-learning (review:¹⁰) might be more relevant skills when driving a car. On the other hand, STN-DBS can induce cognitive impairment with a decline of executive functions¹¹ or altered impulse control,¹² both of which can interfere with driving ability.

Here, we applied an evidence class II study design to prospectively evaluate driving performance in Parkinson’s disease patients before and six to twelve months after STN-DBS implantation (depending on the moment of reaching a stable postoperative stimulation and medication adjustment). Driving performance was also compared to a group of DBS-eligible but non-operated Parkinson’s disease patients under best medical treatment that was frequency-matched for age, cognition, and disease severity. Assessment of driving performance in controls at one time point should allow to estimate the Parkinson’s disease patients’ general level of driving ability. We hypothesized that STN-DBS does not affect driving negatively but might even improve driving safety.

Participants

Participants were recruited in our Movement Disorders Center (October 2013 to November 2017). Two groups of patients with Parkinson’s disease basically eligible for DBS^13,14 were investigated and frequency-matched for age, gender and cognition: 1) patients willing to undergo STN-DBS implantation (“PD-DBS”, implantations were performed in-house in all patients except one) and 2) patients who preferred to continue best medical treatment without DBS (“PD-nDBS”). Furthermore, a healthy control group of frequency-matched age without any neurologic diseases was recruited (“controls”) to estimate the Parkinson’s disease patients’ driving level in general. Further inclusion criteria were the possession of a valid driver license and driving actively during the last three years. Exclusion criteria were the presence of aspects resulting in ineligibility for DBS, especially dementia or another relevant neurologic, psychiatric or cardiologic disease that might interfere with driving or result in consciousness disturbances (e.g. stroke, epilepsy etc.), an already known driving inability e.g. due to immobility or sleep attacks, a known higher order visual impairment or severe visual problems, the presence of hallucinations, unwillingness or inability to drive an automatic vehicle and orthopedic symptoms interfering with handling of gas or brake pedals.

To exclude relevant cognitive impairment the Mini-Mental State Examination (MMSE¹⁵) was applied as screening tool with a cut-off score < 24/30.¹⁶ Additionally, we used the more sensitive Parkinson Neuropsychometric Dementia Assessment (PANDA¹⁷) to detect mild cognitive impairment¹⁸ and to evaluate subtle cognitive changes at Follow-up before and after DBS.

All participants were informed about the study verbally and by letter. Written informed consent was obtained from all participants. The study was approved by the local ethics committee of the Medical Council Hamburg (amendment to trial number PV3557).

Data acquisition

First, demographic and clinical data of all participants were acquired including the patients’ MMSE and levodopa equivalent dosage (LED).¹⁹ All participants were asked about the date they had received their driver’s license, the mileage (in kilometers) they drove and number of accidents they had during the last three years and about a subjective rating on their individual driving safety.

Both Parkinson’s disease groups (PD-DBS and PD-nDBS) were evaluated clinically and within the driving simulator in the clinical “on” state about one hour after taking their regular medication at two different time points (Fig. 1A): at Baseline and six to twelve months later, i.e. postoperatively in one group (Follow-up). Baseline and Follow-up investigations included two separate examinations each on two different days with interval no larger than three days (= four driving examinations in total).

Motor scores (Unified Parkinson's Disease Rating Scale part III (UPDRS III)²⁰ and Hoehn & Yahr Score (H&Y)²¹ were applied at all examination appointments. Participants’ cognition according to the PANDA, level of depression (Beck Depression Inventory; BDI)²² and quality of life (Parkinson’s Disease Questionnaire; PDQ-39)²³ were tested once at Baseline and once at Follow-up. Controls were only evaluated once (Baseline). All participants were told to wear their regular vision aids best suitable for filling-out questionnaires and/or driving (near-vision or distance glasses, respectively).

[Please insert Fig. 1 here]

Driving simulator

We used a driving simulator FT-SR 200 (Fig. 1B; Fa. SimuTech GmbH, Bremen, Germany, http://www.simutech.de/index.htm), which is composed of an LCD panel and original VW Golf parts (2007) with automatic transmission comprising a detailed replication of the gearshift and pedals and a robust electronic engine to recreate steering forces. It complies with the European guideline 2003/59/EG and Driver Qualification Act (“Kraftfahrerqualifikationsgesetz”). Driving was simulated applying the software “3D-Simulator”, version 5.1 (Besier 3D-Edutainment Wiesbaden, Germany, (c) Harro Besier 2001–2020, http://www.3dfahrschule.de/, Fig. 1C).

Driving sessions

Participants performed a 5-minute-training session comprising driving on a virtual parking area to accustom to the simulator system and clarify questions regarding its handling. Afterwards, they drove a predefined route without time limits and with identical challenges for everyone including sceneries in and out of town and realistic everyday life situations with oncoming traffic, crossroads, roundabouts, turning left and right, overtaking, crossing pedestrians and hazardous situations (e.g., animals crossing). Participants were asked to drive “as usual” by following common traffic regulations. Recording started on the first acceleration and ended after completing the route indicated by target flags on the screen.

Driving analysis

For recording and analysis of driving performance, two different software components were used and synchronized with each other:

The software DataView® (http://qualistar.chauvin-arnoux.com/de/dataview) recorded and analyzed videos of the main driving sessions providing speed information.
Other driving performance aspects (e.g. driving errors) were directly recorded by the simulator software “INTERACT”™ (https://www.mangold-international.com/de/produkte/software/verhalten-erforschen-mit-interact) that served as a control for personal rater analysis described below in cases of potential uncertainty of error definition.

As our gold standard, videos of every driving session were analyzed simultaneously by two raters (EB and CS) who evaluated driving performance by counting the participants driving errors. We used a rater-based error counting additionally to INTERACT™ as we analyzed a broader spectrum of errors that were not included in the software, and we defined some errors slightly differently to allow for error severity classification (e.g., speed definitions; Table 1). Furthermore, we recognized some rarely appearing software problems during test runs (e.g., randomly error-logfiles were not exactly synchronized with error events) which could only be controlled manually. Two raters were chosen to increase the detection rate and decrease the missing rate of errors.

Table 1

Catalogue of driving errors
Error category	Severity (factor)	In town (speed limit 50km/h)	Out of town (speed limit 100km/h)
Speed errors	Slight (1)	DRIVING too SLOW: <40km/h for 10s or more	- DRIVING too SLOW: <50km/h for 10s or more (Exception: slowing down due to curves in the road is allowed) - DRIVING IN FOG: >50km/h for 5s or more - DRIVING too FAST: ≥10km/h above speed limit for 10s or more
	Moderate (2)	DRIVING too FAST: >10km/h above speed limit for 10s or more	DRIVING too FAST: ≥20km/h above speed limit for 10s or more
	Severe (4)	- DRIVING too FAST: >20km/h above speed limit for 10s or more - DECELERATING (to 0km/h) without any reason - RUNNING PAST A BUS: >10km/h	DECELERATING (to 0km/h) without any reason
	Very severe (8)	DRIVING too FAST: >30km/h above speed limit for 10s or more	DRIVING too FAST: >30km/h above speed limit for 10s or more
Distance keeping errors	Slight (1)	Distance < 5m	Distance < 1/2 speedometer* for 10s or more
	Moderate (2)	Distance < 3m	Distance < 1/4 speedometer*
	Severe (4)	Distance < 1m	Distance < 1/5 speedometer*
	Very severe (8)	Briefly very close	Briefly very close
Lane keeping errors	Slight (1)	- One-lane roads: TO SWERVE ABOUT A ROAD within the road edge marking - Multi-lane roads: TO SWERVE ABOUT A ROAD while crossing the road edge marking left or right
	Moderate (2)	Touching the road edge marking (= if lane reaches the medial third of the screen)
	Severe (4)	- CROSSING the road edge marking on the right side - LEAVING the lane completely
	Very severe (8)	CROSSING the road edge marking on the left side
Traffic sign errors	Slight (1)	GREEN LIGHT flashing for > 3s (delayed starting)
	Moderate (2)	YELLOW LIGHT
	Severe (4)	- RED LIGHT flashing for < 1s - NO STOP AT CROSSWALK (although pedestrians approached)
	Very severe (8)	RED LIGHT flashing for > 1s
Indicator errors	Slight (1)	- MISSED TURN INDICATOR: to turn off - MISSED TURN INDICATOR: after passing another car on the right side
Indicator errors	Moderate (2)	- MISSED TURN INDICATOR: after passing another car on the left side (when passing another car it is expected to use the turn indicator twice: when passing another car and when returning into the initial lane) - MISSED TURN INDICATOR: when making a lane change
Accidents	Severe (4)	With inanimate OBJECTS
Accidents	Very severe (8)	With PEOPLE or ANIMALS

Driving errors were classified into six “error categories” considering (1) speed, (2) keeping distance, (3) keeping lane, (4) following traffic signs, (5) indicator errors and (6) accident. Errors were ranked according to their severity as slight, moderate, severe or very severe (Table 1).

[Please insert Table 1 here]

Total number of errors (error rate), driving time (mean duration of driving sessions), error rate per error category and error severity were recorded. Moreover, a “driving safety score” (DSS) was defined to reflect overall driving safety. The DSS displays that error severity is more safety relevant than error frequency and was calculated as follows: Every error counted (across all error categories) was multiplied by a “severity factor (SF)” ranking the errors according to their severity as geometric sequence with a power of two due to safety risk (SF 1 for slight errors, SF 2 moderate, SF 4 severe and SF 8 very severe and/or fatal errors (accident with injury to people or extremely dangerous errors with high safety-critical relevance)). All ranked errors were added up to build the DSS. Due to participants’ or technical problems (see results) not all patients completed the scheduled two examinations each at Baseline and Follow-up. As consequence we either used the mean error numbers at Baseline and/or Follow-up (two examinations available) or the total number of errors (only one examination available). The same was applied to UPDRS III and H&Y scores.

Statistical analysis

Data was collected and pre-analyzed using IBM software SPSS version 18 (https://www.ibm.com/analytics/spss-statistics-software). Final analysis was performed with the open-source software R version 4.0.2 (2020-06-22, R Core Team, https://www.R-project.org).

Baseline and Follow-up clinical characteristics were summarized (separately) as number of patients (and %) for categorical variables and as mean, standard deviation and range for continuous variables and were compared between groups using Chi-square tests, t-test (two-sided) and one-way ANOVA F-tests, respectively. Group comparisons of within-group differences from Baseline to Follow-up were performed with ANCOVA (= difference between the group-specific mean changes from Baseline, adjusted for Baseline).

Driving parameters were compared between groups (PD-DBS vs. PD-nDBS vs. healthy controls) at Baseline and Follow-up and within Parkinson’s disease groups between Baseline and Follow-up using a single generalized linear mixed model with negative binomial link for count models and normal link for scores.

Spearman correlations between Baseline DSS and Baseline age, gender, disease duration, LED, H&Y, UPDRS III, PANDA, MMSE and BDI were estimated for both Parkinson’s disease groups combined to assess associations between clinical characteristics and participants' general driving performance.

Data availability

The data that support the findings of this study are available from the corresponding author, upon reasonable request.

A total of 128 subjects were screened. 85 participants (23 PD-DBS, 29 PD-nDBS, 33 controls) were included into the study with complete data sets (two driving sessions each at Baseline and Follow-up in 60 study participants and at least one driving session at Baseline and Follow-up in 25 participants).

Supplementary Table 1 informs on drop-out reasons of the excluded 43 of 128 screened subjects, Supplementary Table 2 describes reasons of the 25 included patients with less than 4 driving sessions and Supplementary Table 3 presents completion rates per session and group. In the PD-DBS group, Baseline examinations were performed in, on average, 76.3 days (+/- SD 62.3 [2-231]) prior to DBS implantation. Follow-up sessions of both patient groups were performed about 9.15 months (+/- SD 2.7 [4–18]) after Baseline (Supplementary Table 4 shows the distribution of Follow-up dates). Time to Follow-up differed between the PD-DBS (10.4 months +/- SD 3.0 [7–18]) and PD-nDBS (8.2 months +/- SD 2.0 [4–11], p = 0.003) group.

Clinical scores and questionnaires

Baseline clinical characteristics of all groups are given in Table 2. All three groups did not differ significantly regarding age, gender, disease duration, driving experience, and cognition according to MMSE. Healthy subjects scored higher in the PANDA and lower in the BDI and PDQ-39 compared to both patient groups (all p < 0.005) indicating less cognitive impairment and depressed mood and better quality of life. Importantly, clinical characteristics, especially motor scores (H&Y and UPDRS III), cognition (MMSE and PANDA) and LED did not differ significantly between PD-DBS and PD-nDBS group at Baseline (Table 2).

Table 2

Baseline clinical parameters of all study participants
Clinical parameter	PD-DBS (n = 23) Mean +/- SD, Min-Max	PD-nDBS (n = 29) Mean +/- SD, Min-Max	Controls (n = 33) Mean +/- SD, Min-Max	Statistics
Age (years)	60.1 (+/- SD 8.1; 45–72)	61.4 (+/- SD 7.6; 49–73)	57.7 (+/- SD 11.3; 29–77)	F(2,82) = 1.230, p = 0.298
Sex (n; female/male)	7 /16	5/24	12/21	χ²(2) = 2.875, p = 0.238^a
Disease duration (years)	11.0 (+/- SD 5.0; 5–27)^f	9.0 (+/- SD 4.8; 2–24)	n.a.	F(1,49) = 2.017, p = 0.162
Ownership of driver license (years)	41.3 (+/- SD 10.2; 20–58)	41.1 (+/- SD 7.8; 26–55)	37.7 (+/- SD 11.7; 11–58)	F(2,82) = 1.225, p = 0.299
Kilometers driven during the last 3 years (km)	31,822 (+/- SD 25,232; 300 − 80,000)	34,293(+/- SD 59,797; 1,000–30,000)	41,424 (+/- SD 42,5672; 2,000–20,000)	F(2,82) = 0.343, p = 0.710
Subjective Driving Safety (n)	- safe: 11 - average: 10 - unsafe: 2	- safe: 15 - average: 11 - unsafe: 3	- safe: 20 - average: 11 - unsafe: 2	χ²(4) = 1.179, p = 0.882^a
Accidents during last 3 years (n)	− 1 accident: 3 - >1 accident: 0	− 1 accident: 8 - >1 accident: 0	− 1 accident: 3 - >1 accident: 2	χ²(4) = 7.006, p = 0.136
Levodopa Equivalent Dose (mg)	1,197.1 (+/- SD 626.28; 174-2,840)	1,133.04 (+/- SD 465.32; 260-1,814)	n.a.	F(1,50) = 0.179, p = 0.674
Dopamine Agonists (n)	19	28	n.a.	χ²(1) = 2.869, p = 0.090^a
H&Y	2.2 (+/- SD 0.8; 0–4)	2.3 (+/- SD 0.6; 1–3)	n.a.	F(1,50)=-0.225, p = 0.823
UPDRS III	16.8 (+/- SD 11.4; 1.5–43)	16.4 (+/- SD 8.5; 3.5–39.5)	n.a.	F(1,50) = 0.170, p = 0.866
MMSE (screening exclusion criterion)	28.2 (+/- SD 3.2; 15–30)^b	28.7 (+/- SD 1.9; 24–30)	28.8 (+/- SD 1.5; 24–30)	F(2,82) = 6.435, p = 0.514
PANDA	20.4 (+/- SD 6.5; 9–30)^c	20.2 (+/- SD 6.0; 5–28)	24.8 (+/- SD 4.2; 14–30)	F(2,81) = 6.624, p = 0.002**
BDI	9.4 (+/- SD 6.8; 1–30)	7.0 (+/- SD 5.2; 1–25)	2.2 (+/- SD 2.8; 0–13)^d	F(2,81) = 15.232, p < 0.001***
PDQ-39	30.4 (+/- SD 18.1; 6.4–82.0)	25.1 (+/- SD 11.8; 5.1–53.2)	2.2 (+/- SD 3.5; 0-12.8)^e	F(2,80) = 44.244, p < 0.001***

[Please insert Table 2 here]

At Follow-up (Table 3) lower (better) H&Y scores in PD-DBS compared to PD-nDBS participants (p = 0.022) and a trend towards lower (better) UPDRS III scores in the PD-DBS compared to the PD-nDBS group (p = 0.064) was found. LED and number of patients with dopamine agonist intake were lower in the PD-DBS compared to the PD-nDBS group (p = 0.001 and p = 0.032, respectively). Other clinical parameters at Follow-up did not differ significantly.

Table 3

Follow-up clinical parameters of study participants
Clinical parameter	PD-DBS (n = 23) Mean ± SD [Min-Max]	Number of patients with available data (PD-DBS)	PD-nDBS (n = 29) Mean ± SD [Min-Max]	Number of patients with available data (PD-nDBS)	Statistics
Levodopa Equivalent Dose (mg)	699.9 ± 352.7 [0-1399.6]	n = 23	1176.2 ± 533.5 [75-1998.8]	n = 24	t(45)=-3.595, p = 0.001**
Dopamine agonists (n)	15	n = 23	22	n = 24	χ²(1) = 4.600, p = 0.032*^a
H&Y	2.1 ± 0.5 [1–3]	n = 23	2.4 ± 0.4 [1.25-3]	n = 29	t(50)=-2.369, p = 0.022*
UPDRS III	14.9 ± 10.3 [4–47]	n = 23	20.2 ± 9.7 [8–48]	n = 29	t(50)=-1.894, p = 0.064
PANDA	20.7 ± 6.0 [9–29]	n = 23	22.7 ± 4.6 [11–28]	n = 29	t(50)=-1.328, p = 0.190
BDI	6.8 ± 4.1 [1–16]	n = 20	6.6 ± 4.9 [0–20]	n = 22	t(40) = 0.185, p = 0.854
PDQ_39	27.3 ± 12.7 [4.2–46.5]	n = 17	24.8 ± 13.4 [10.9–52.8]	n = 21	t(36) = 0.587, p = 0.561
Time from Baseline to Follow-up (month)	10.4 ± 3.0 [7–18]	n = 23	8.2 ± 2.0 [4–11]	n = 29	t(50) = 3.090, p = 0.003**

[Please insert Table 3 here]

Group comparisons of clinical changes from Baseline to Follow-up

In the PD-DBS group, LED and H&Y score decreased from Baseline to Follow-up compared to the PD-nDBS group (p < 0.001 and p = 0.019), indicating less need for medication and better clinical condition after DBS-surgery. All other clinical scores (UPDRS III, MMSE, PANDA, BDI, PDQ-39) did not show any significant group differences when comparing Baseline vs. Follow-up (differences; Supplementary Table 5).

Driving simulator performance

Error rates

Considering mean absolute error rates, we found two relevant outliers in the PD-DBS group (male and female; Fig. 2, red box) with substantially higher error rates (and DSS) in the post-operative Follow-up session compared to Baseline (88 vs. 37 and 70.5 vs. 47, respectively) and about four-fold higher error rates compared to mean rates of other PD-DBS subjects (18.4 at Baseline and 18.0 at Follow-up). Additionally, both outliers also belonged to the “worst 5” participants with the highest overall error rates at Baseline (Supplementary Table 6). We excluded these two patients from main analysis to avoid statistical bias but included them into a post-hoc analysis, first to explore reasons or explaining factors for the strikingly unsafe driving at Follow-up and second, to reflect “real life”.

[Please insert Fig. 2 here]

Table 4 shows the detailed measurements of driving performance at Baseline and Follow-up investigation for both PD patient groups (outliers excluded). Statistic values are given for between-group comparisons at Baseline and Follow-up (left part of the table) as well as for within-group comparisons from Baseline to Follow-up (right part).

Table 4

Driving performance parameters (two outliers excluded)
		Between-group comparisons								Within-group comparisons
Parameter	Time point	Healthy controls		PD-DBS	PD-nDBS	PD vs Ctrl	p-value	DBS vs nDBS	p-value	FU vs BL: DBS	p-value	FU vs BL: nDBS	p-value
Driving time (in s; estimated residuals, LMM)	Baseline	835.0 (784.7, 885.3)		876.2 (813.2, 939.3)	874.2 (820.6, 927.9)	40.24 (-24.89, 105.37)	0.224	2.00 (-80.78, 84.78)	0.962	-	-	-	-
Driving time (in s; estimated residuals, LMM)	Follow-up	-		809.8 (746.7, 872.8)	840.4 (784.8, 896.0)	-	-	-30.64 (-114.70, 53.41)	0.472	-66.47 (-155.63, 22.68)	0.143	-33.83 (-111.09, 43.43)	0.388
Absolute error rate (mean, estimated residuals, LMM)	Baseline	16.0 (13.9, 18.5)		18.4 (15.5, 22.0)	20.3 (17.6, 23.5)	1.21 (1.01, 1.45)	0.041	0.91 (0.72, 1.14)	0.396	-	-	-	-
Absolute error rate (mean, estimated residuals, LMM)	Follow-up	-		18.0 (15.1, 21.4)	17.5 (15.0, 20.3)	-	-	1.03 (0.82, 1.29)	0.811	0.97 (0.80, 1.18)	0.775	0.86 (0.73, 1.01)	0.061
Error severity (quotients, gLMM)
Slight	Baseline		9.5 (8.2, 11.1)	9.9 (8.2, 11.9)	10.6 (9.1, 12.3)	1.07 (0.89, 1.30)	0.467	0.93 (0.74, 1.18)	0.566	-	-	-	-
Slight	Follow-up		-	9.5 (7.9, 11.4)	9.1 (7.8, 10.7)	-	-	1.04 (0.82, 1.33)	0.753	0.96 (0.75, 1.23)	0.767	0.87 (0.70, 1.07)	0.171
Moderate	Baseline		3.7 (2.9, 4.8)	5.6 (4.2, 7.6)	5.8 (4.5, 7.5)	1.54 (1.11, 2.12)	0.009	0.96 (0.65, 1.42)	0.847	-	-	-	-
Moderate	Follow-up		-	5.6 (4.2, 7.6)	5.4 (4.2, 7.0)	-	-	1.05 (0.71, 1.55)	0.806	1.00 (0.74, 1.36)	0.976	0.92 (0.71, 1.19)	0.529
Severe	Baseline		1.8 (1.4, 2.4)	1.5 (1.1, 2.2)	2.0 (1.5, 2.6)	0.94 (0.68, 1.32)	0.735	0.76 (0.49, 1.18)	0.219	-	-	-	-
Severe	Follow-up		-	1.2 (0.8, 1.8)	1.6 (1.2, 2.2)	-	-	0.76 (0.47, 1.24)	0.273	0.81 (0.48, 1.37)	0.433	0.81 (0.55, 1.19)	0.286
Very severe	Baseline		0.7 (0.5, 1.2)	1.2 (0.8, 2.0)	1.5 (1.0, 2.3)	1.87 (1.10, 3.17)	0.022	0.79 (0.44, 1.42)	0.427	-	-	-	-
Very severe	Follow-up		-	1.2 (0.8, 2.0)	0.9 (0.6, 1.4)	-	-	1.32 (0.70, 2.48)	0.387	1.01 (0.59, 1.73)	0.980	0.60 (0.37, 0.97)	0.037
Error classes (quotients, gLMM)
Distance	Baseline		0.2 (0.0, 0.5)	0.2 (0.0, 0.7)	0.2 (0.1, 0.7)	1.42 (0.33, 6.03)	0.636	0.79 (0.14, 4.52)	0.789	-	-	-	-
Distance	Follow-up		-	0.2 (0.1, 0.9)	0.2 (0.1, 0.6)	-	-	1.38 (0.24, 7.97)	0.716	1.25 (0.19, 8.14)	0.814	0.71 (0.14, 3.59)	0.681
Indicator	Baseline		2.9 (2.3, 3.6)	4.1 (3.2, 5.2)	3.8 (3.1, 4.7)	1.36 (1.04, 1.79)	0.024	1.07 (0.78, 1.47)	0.665	-	-	-	-
Indicator	Follow-up		-	3.1 (2.4, 4.1)	3.3 (2.6, 4.1)	-	-	0.96 (0.68, 1.36)	0.819	0.76 (0.56, 1.05)	0.097	0.85 (0.65, 1.12)	0.247
Lane keeping	Baseline		4.2 (3.2, 5.7)	5.8 (4.1, 8.1)	6.3 (4.7, 8.5)	1.42 (0.98, 2.05)	0.061	0.91 (0.58, 1.43)	0.673	-	-	-	-
Lane keeping	Follow-up		-	6.3 (4.5, 8.9)	5.0 (3.7, 6.8)	-	-	1.26 (0.80, 1.98)	0.318	1.10 (0.77, 1.55)	0.600	0.79 (0.58, 1.07)	0.128
Accident	Baseline		0.5 (0.3, 0.8)	0.6 (0.4, 1.1)	0.7 (0.5, 1.2)	1.38 (0.76, 2.51)	0.288	0.88 (0.44, 1.76)	0.712	-	-	-	-
Accident	Follow-up		-	0.5 (0.2, 0.9)	0.5 (0.3, 0.8)	-	-	0.99 (0.43, 2.27)	0.973	0.71 (0.31, 1.62)	0.418	0.64 (0.32, 1.25)	0.189
Velocity	Baseline		7.0 (5.8, 8.4)	6.2 (4.9, 8.0)	6.6 (5.4, 8.1)	0.93 (0.73, 1.18)	0.527	0.94 (0.69, 1.29)	0.702	-	-	-	-
Velocity	Follow-up		-	6.1 (4.8, 7.8)	7.0 (5.7, 8.5)	-	-	0.88 (0.64, 1.20)	0.411	0.98 (0.72, 1.33)	0.902	1.05 (0.82, 1.36)	0.697
Traffic sign	Baseline		0.8 (0.6, 1.2)	1.0 (0.7, 1.6)	1.7 (1.2, 2.2)	1.55 (0.99, 2.44)	0.057	0.63 (0.38, 1.05)	0.078	-	-	-	-
Traffic sign	Follow-up		-	1.0 (0.7, 1.6)	0.9 (0.6, 1.3)	-	-	1.17 (0.66, 2.07)	0.592	1.00 (0.55, 1.82)	1.000	0.54 (0.33, 0.88)	0.013
Driving safety score (DSS, quotients, gLMM)	Baseline		31.9 (24.6, 39.2)	40.0 (30.8, 49.2)	46.7 (38.9, 54.5)	11.41 (1.91, 20.92)	0.019	-6.74 (-18.82, 5.35)	0.272	-	-	-	-
Driving safety score (DSS, quotients, gLMM)	Follow-up		-	39.0 (29.8, 48.2)	37.2 (29.3, 45.0)	-	-	1.83 (-10.25, 13.91)	0.765	-0.95 (-10.92, 9.02)	0.850	-9.52 (-18.00, -1.04)	0.028

[Please insert Table 4 here]

At Baseline, overall error rates did not differ between both Parkinson’s disease patient groups (18.4 vs. 20.3; p = 0.396), but patients in general showed more errors compared to controls (16.0, p = 0.041). At Follow-up, error rates of both patient groups also were similar (18 vs. 17.5, p = 0.811). Within-group analysis showed comparable error rates between Baseline and Follow-up in both groups - with by trend lower error rates in the PD-nDBS group at Follow-up compared to Baseline (p = 0.061).

Error severity

At both time points, most errors of all three participant groups belonged to the “slight error” severity and lowest rates were found for “very severe” errors (Table 4). Compared to controls, both patient groups showed higher rates of “moderate” and “very severe” errors (p = 0.009 and p = 0.022, respectively), but without between-patient group differences for all error severities.

At Follow-up, between-patient group analysis showed similar error rates for each error severity and within-group comparisons revealed a lower rate only for “very severe” errors in the PD-nDBS group compared to Baseline (0.9 vs. 1.5; p = 0.037).

Error category

In all three groups, in descending order, “speed”, “lane keeping” and “indicator” errors were most common (Table 4). Considering both, error rate and severity, “lane keeping” was the prominent error category in all groups (Fig. 3).

At Baseline, patients with Parkinson’s disease made more “indicator” errors than controls (p = 0.024) and by trend more “lane keeping” and “speed” errors (p = 0.061 and p = 0.057, respectively). Performance regarding all error categories was comparable in both patient groups. Within-group comparisons revealed lower error rates only for the category “traffic sign” in the PD-nDBS group when comparing the Follow-up session to Baseline (0.9 vs. 1.7; p = 0.013).

[Please insert Fig. 3 here]

Driving time and Driving safety score (DSS)

Driving time at Baseline and Follow-up did not differ between patients with Parkinson’s disease and controls as well as between both patient groups even regarding within-group comparison.

Baseline between-group comparisons revealed that patients with and without DBS scored higher than controls, indicating poorer driving performance (40.0 and 46.7 vs. 31.9; p = 0.019). However, DSS was similar in both patient groups not only at Baseline but also at Follow-up (39.0 and 37.2, respectively).

Within-group comparison showed a lower DSS (better performance) only in the PD-nDBS groups’ Follow-up session compared to Baseline (p = 0.028) but not in the PD-DBS group (p = 0.850).

Post-hoc analysis (outliers included)

To assess reasons or explaining factors for outstanding unsafe driving at Follow-up and to reflect real life, additional post-hoc analysis of driving performance was done including the outliers mentioned above in the PD-DBS group. Supplementary Table 7 shows variations from main analysis marked in red for synopsis. These two outliers would have biased the results as follows (as the outliers lead to more instable modelling and distorted effects, we here only present numerical data without addressing significances (p-values) of the results):

Results for absolute error rates and driving time remained comparable to main analysis and patients with Parkinson’ disease still had higher DSS scores (poorer performance) compared to controls (without reaching significance), but of note, PD-DBS patients now showed a significantly poorer performance at Follow-up compared to PD-nDBS patients (DSS: 64.7 vs. 37.2). Furthermore, the difference of DSS at Baseline compared to Follow-up now becomes significant in the PD-DBS group (45.4 and 64.7), indicating less safe driving after DBS surgery. Moreover, compared to the PD-nDBS group, the PD-DBS group now showed slightly more “very severe” mean errors at Baseline (1.3) and more prominently at Follow-up (2.0), with the latter now being significantly higher. While in the main analysis the categories “lane keeping” and “traffic sign” errors differed only by trend between both patient groups, PD-DBS patients now showed significantly higher errors in these categories at both time points.

The other way round, when performing a post-hoc analysis of clinical scores without the two outliers, results were similar (Supplementary Table 8 and Supplementary Table 9) compared to main analysis of clinical patient characteristics (Tables 2 and 3). To address whether it would have been possible at Baseline to identify these two retrospectively at Follow-up strikingly unsafe and incompetently driving outliers, we reviewed their individual clinical characteristics (Supplementary Table 10), but except a suspicious PANDA (< 15; but normal MMSE) in one of the two outliners at Baseline, none of our clinical parameters could have “predicted” poor driving performance of these two outliers at study inclusion.

Correlations of clinical characteristics with DSS

Spearman correlation analyses of clinical parameters and DSS reflecting overall driving safety for both patient groups combined revealed only weak correlations (all r < 0.40, Supplementary Table 11) with the strongest correlation of DSS with disease duration at Baseline (r = 0.353, CI [0.080, 0.577]) and with DSS at Follow-up (r = 0.391, CI [0.126,0.603]), indicating that longer disease duration is associated with poorer driving performance and that participants who performed worse at Baseline also drove unsafe at Follow-up.

We evaluated driving-simulator performance in patients with Parkinson’s disease before and after DBS in a prospective and controlled three-armed design. Our data supports the hypothesis that patients with Parkinson’s disease do not drive less safely after DBS implantation but did not indicate that DBS might have beneficial effects on driving performance. Furthermore, our data implies that in DBS-eligible, non-demented patients younger than 75 years, unsafe driving prior to DBS operation, rather than any specific clinical patient characteristic, might be a risk factor for deterioration of driving safety postoperatively.

Our methodological approach using a driving simulator assessment permits faithful replication of the experimental road conditions across participants, in contrast to on road settings.²⁴ Simulator performance reflects real life driving ability²⁵ and is suitable to monitor on-road driving impairments in patients with Parkinson’s disease.² Assessing speed and driving errors in key safety traffic situations has been approved to reflect safety-related parameters in traffic simulation models.²⁶ Our application of a self-developed safety score (DSS) weighs for the “quality” of driving errors and accounts for high frequency/low severity and low frequency/high-severity errors. This is meaningful for determining crash risk in Parkinson’s disease, because especially the latter errors lead to car crashes.²⁴ Our patient groups were very comparable due to selection regarding principle eligibility for DBS. For detailed matching we focused on age and cognition considering that higher age and cognitive impairment are main risk factors for unsafe driving.⁵ The controlled prospective study design allows comparison of driving between both patient groups irrespective of the disease progression. On the other hand, Follow-up investigation on average nine months later is likely short enough to rate disease progression as irrelevant anyway but is long enough to reach a stable clinical condition of DBS and/or medication adjustments.²⁷ The baseline comparison with age-matched healthy controls enables quantifying the driving level of patients with Parkinson’s disease in general. In the healthy controls, we abstained from Follow-up investigations as no relevant change in driving performance was assumed within nine month and without intervention. Despite a mean Follow-up investigation at nine months, time points varied from four to 18 months due to clinical or logistic reasons, resulting in a significant between-patient group difference at Follow-up of about two months. This short period should not bias results as it likely does not relevantly affect disease progression between groups.

As a general result, we found healthy controls to drive safer (DSS) than both patient groups. This is in line with former studies showing patients with Parkinson’s disease driving at least somewhat insecurely compared to controls.^2,25 Longer disease duration and higher disease severity (according to H&Y) influenced driving safety negatively, which corresponds to higher accident rates found to be related to disease burden in Parkinson’s disease.^25,26 Driving experience (expressed as the driven mileage during the last three years) had no relevant influence on driving safety, which is also in line with former findings.²⁸ Age and cognitive impairment correlated very weakly with unsafe driving in our study but strongly in others.^2,26,29 This is likely related to the fact that we only included DBS-eligible patients without relevant cognitive impairment and of comparatively “young” age.

Own previous studies suggested that DBS might influence driving in Parkinson’s disease positively: In a survey, patients who had undergone DBS reported three times more subjective improvement than deterioration of driving after DBS operation and postoperatively twice as many patients restarted than quit driving, independently of their H&Y stage.¹ In a previous, controlled, cross-sectional, class IV driving-simulator study we found a positive effect of DBS on driving when comparing two groups of patients, with and without DBS, and comparing the therapeutic “stimulation on” and the “medication (levodopa) only” condition in the DBS group. Because motor aspects could not explain results, we hypothesized that DBS might improve driving due to non-motor driving relevant cognitive aspects⁸ such as improvements of implicit procedural³⁰ and/or sequence learning³¹, goal-directed action selection³² or decision-learning;³³ all crucial cognitive skills for car driving.

However, results of the present study suggest that DBS intervention does not have a clinically relevant effect on driving. This includes single driving parameters such as driving time, error rates, error severity and error category and – most important - overall driving safety weighting for error severity and frequency. This conclusion is based on the analysis after exclusion of two prominent outliers in the DBS group to avoid bias and not to confound general study results. These patients were found to drive badly already at Baseline (among the “worst five” of unsafe drivers) but comparatively outstandingly worse after surgery. Importantly, at study inclusion and retrospectively there were no recognizable reasons to exclude these patients at Baseline from further analysis. Because these two patients in fact actively drove a car, we included both subjects in an additional post-hoc analysis to reflect the “real-life” situation. As a result, now the patient group that has undergone surgery drives more unsafely at Follow-up than the patient group that was not operated, mainly due to a higher rate of very severe errors.

Therefore, DBS might lead to deterioration of driving in certain patients with Parkinson’s disease that, however, do not present with obvious “predicting” characteristics prior to surgery. Noteworthy, the two outliers belonged to the “worst five” of bad drivers at study inclusion. Hence, unsafe driving prior to surgery might be a risk factor for postoperative deterioration. It seems advisable to recommend such patients particularly urgently to have their driving ability checked not only before but especially after DBS surgery. Alarmingly, in retrospective none of the assessed clinical parameters such as age, cognition, UPDRS III, LED etc. indicated these patients’ bad driving or worsening of driving after DBS. To date, there is lack of a single parameter or a test battery reliably indicating insufficient driving ability.⁵ Recent own findings suggest a comprehensive test battery including the Montreal Cognitive Assessment (MoCA), TAP-M “flexibility” subscore (“Testbatterie für Aufmerksamkeitsprüfung”), Trail Making Test part A and Driving Behavior Questionnaire (DBQ) subscore “errors“ as potential screening tools to detect PD patients at risk for driving.³⁴ However, this is not practicable in daily routine. Furthermore, as a selective decrease in frontal cognitive functions after STN-DBS has been described.³⁵ other frontal assessments might be a potential predictor. However, we did not find the Frontal Assessment Battery (FAB) predictive for bad driving in patients with Parkinson’s disease in a previous study.³⁴

We therefore suggest to advise every patient with aspects of unsafe driving, for example based on information from a third party such as relatives, to have their ability to drive checked in general ⁵ and especially after DBS surgery (optimally on-road by a professional driving instructor).

Our study has some limitations. First, we faced some rarely, randomly occurring technical problems in some patients regarding synchronization between INTERACT™ and DataView®. Therefore, two investigators checked all logfiles manually and single missing data had to be interpolated. Randomly short stutters of the screen presenting the driving scenery occurred in some trials due to reduced storage capacity which might have confused some participants. Second, we initially planned two Baseline and two Follow-up investigations with identical driving routes and with averaged scores of both examinations to reduce bias due to a possible individual bad daily condition, especially in fluctuating patients. However, several participants performed only one session at Baseline and/or at Follow-up (reasons and distribution described in Supplementary Table 2 and Supplementary Table 3). In cases with two sessions mean scores were calculated. Despite a gap of a few days between investigations, a learning effect cannot be excluded. However, numbers of DBS and nDBS patients with either two or only one examination at a time point were similar (Baseline PD-DBS vs. PD-nDBS: 5 vs. 7, Follow-up: 10 vs. 9). Future studies should provide randomly designed but in total identical challenges during the driving course. Third, comprehensive visual testing was not done but patients’ vision history had to be unsuspicious, and participants wore their appropriate visual aids.

Subthalamic DBS does not seem to alter driving ability in patients with Parkinson’s disease. However, unsafe driving prior to DBS surgery might be a risk factor for deterioration of driving safety postoperatively.

Acknowledgements

We thank Thomas Wriedt for his technical and Fred Rikkers† and Thomas Rikkers† for their logistic support. We especially thank all PD patients and controls for study participation.

Funding

Acquisition of the driving simulator was supported by the Georg & Jürgen Rickertsen Stiftung Hamburg.

Author contributions

O. Fründt (OF), T. Mainka (TM), E. Vettorazzi (EV), E. Baltaci (EB), C. Schwarz (CS), M. Südmeyer (MS), C. Gerloff (CG), W.H. Zangemeister (WHZ), M. Poetter-Nerger (MPN), U. Hidding (UH), W. Hamel (WH), C.K.E. Moll (CKEM), C. Buhmann (CB).

Drafting/revising the manuscript for content, including medical writing for content: OF, TM, MPN, WH, WHZ, MS, CKEM, CG, CB

Study concept or design: EV, CB

Analysis or interpretation of data: OF, TM, EV, CB

Acquisition of data: TM, EB, CS, TR, MPN, UH, CB

Statistical analysis: OF, EV, CB

Study supervision or coordination: CB

Competing interests

The authors have no conflict of interest to report regarding this manuscript.

Disclosures are as follows:

Prof. Dr. C. Buhmann and Prof. W.H. Zangemeister received study research support from the Georg & Jürgen Rickertsen Stiftung Hamburg. Prof. C. Buhmann served on scientific advisory boards and/or received honoraria for lectures from Abbvie, Bial, Desitin, Kyowa Kirin, Merz, Stada, TAD Pharma, UCB and Zambon. Prof. C. Gerloff is supported by the German Research Foundation (DFG Ge 844/4-1, SFB 936) and the EU (FP7-HEALTH-2011-278276) and he received honoraria as speaker and/or scientific advisory board member from Actitor Biotech,

Amgen, Bayer Healthcare, Boehringer Ingelheim, Prediction Biosciences and Sanofi Aventis. Dr. O. Fründt is supported by the Klaus Thiemann Foundation and received honoria for lectures from Daiichi-Sankyo and support for congress participation and trainings from Abbvie and St. Jude/Abbott. PD Dr. M. Pötter-Nerger is supported by DFG (SFB 936, C8), EU (MSC-ITN-MULTI), and received lecture fees from Abbott, Boston Scientific, Abbvie and Licher, study-related fees from Boston scientific, Zambon, Abbott and served as consultant for Boston Scientific and Abbvie. Prof. Dr. W. Hamel received lecture fees and honoraria for serving on advisory boards, travel grants and educational grants from Abbott, Aleva, Boston Scientific, and Medtronic. Dr. T. Mainka has received royalties from Elsevier in Urban & Fischer, speaker fees from Stadapharm and has served in an advisory board for Biomarin Pharmaceutical. She is supported by the BIH-Charité Clinician Scientist Program of the Charité-Universitätsmedizin Berlin and the Berlin Institute of Health. Prof. Dr. M. Südmeyer received grants and/or honoraria for consultancy by TEVA Pharma, Medtronic, Abbott/St. Jude Medical, Novartis, Desitin, Meda Pharma, AbbVie as well as grants from the Helmholtz Society, “Stiftung für Altersforschung” and “Forschungskommission”, Heinrich-Heine-University, Düsseldorf.

PD Dr. C.K.E. Moll, E. Vettorazzi, Dr. U. Hidding, E. Baspinar and C. Schwarz report no disclosures.

Buhmann, C. et al. Impact of Deep Brain Stimulation on Daily Routine Driving Practice in Patients with Parkinson's Disease. Parkinsons Dis 2015, 608961, doi:10.1155/2015/608961 (2015).
Uc, E. Y. et al. Road safety in drivers with Parkinson disease. Neurology 73, 2112–2119, doi:10.1212/WNL.0b013e3181c67b77 (2009).
Wood, J. M., Worringham, C., Kerr, G., Mallon, K. & Silburn, P. Quantitative assessment of driving performance in Parkinson's disease. J Neurol Neurosurg Psychiatry 76, 176–180, doi:10.1136/jnnp.2004.047118 (2005).
Uc, E. Y. et al. Real-life driving outcomes in Parkinson disease. Neurology 76, 1894–1902, doi:10.1212/WNL.0b013e31821d74fa (2011).
Buhmann, C. & Gerloff, C. Autofahren bei Morbus Parkinson - Driving with Parkinson´s Disease Aktuelle Neurologie 40, 315–320 (2013).
Apolinario, D. et al. Cognitive impairment and driving: A review of the literature. Dement Neuropsychol 3, 283–290, doi:10.1590/S1980-57642009DN30400004 (2009).
Buhmann, C., Vesper, J. & Oelsner, H. [Driving ability in Parkinson's disease]. Fortschr Neurol Psychiatr 86, 43–48, doi:10.1055/s-0043-110051 (2018).
Buhmann, C. et al. Effect of subthalamic nucleus deep brain stimulation on driving in Parkinson disease. Neurology 82, 32–40, doi:10.1212/01.wnl.0000438223.17976.fb (2014).
Uc, E. Y. et al. Impaired navigation in drivers with Parkinson's disease. Brain 130, 2433–2440, doi:10.1093/brain/awm178 (2007).
Buhmann, C. & Gerloff, C. Could deep brain stimulation help with driving for patients with Parkinson's? Expert Rev Med Devices 11, 427–429, doi:10.1586/17434440.2014.929495 (2014).
Parsons, T. D., Rogers, S. A., Braaten, A. J., Woods, S. P. & Troster, A. I. Cognitive sequelae of subthalamic nucleus deep brain stimulation in Parkinson's disease: a meta-analysis. Lancet Neurol 5, 578–588, doi:10.1016/S1474-4422(06)70475-6 (2006).
Cavanagh, J. F. et al. Subthalamic nucleus stimulation reverses mediofrontal influence over decision threshold. Nat Neurosci 14, 1462–1467, doi:10.1038/nn.2925 (2011).
Deuschl, G. et al. A randomized trial of deep-brain stimulation for Parkinson's disease. N Engl J Med 355, 896–908, doi:10.1056/NEJMoa060281 (2006).
Schuepbach, W. M. et al. Neurostimulation for Parkinson's disease with early motor complications. N Engl J Med 368, 610–622, doi:10.1056/NEJMoa1205158 (2013).
Folstein, M. F., Folstein, S. E. & McHugh, P. R. "Mini-mental state". A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 12, 189–198, doi:10.1016/0022-3956(75)90026-6 (1975).
Kochhann, R., Varela, J. S., Lisboa, C. S. M. & Chaves, M. L. F. The Mini Mental State Examination: Review of cutoff points adjusted for schooling in a large Southern Brazilian sample. Dement Neuropsychol 4, 35–41, doi:10.1590/S1980-57642010DN40100006 (2010).
Kalbe, E. et al. Screening for cognitive deficits in Parkinson's disease with the Parkinson neuropsychometric dementia assessment (PANDA) instrument. Parkinsonism Relat Disord 14, 93–101, doi:10.1016/j.parkreldis.2007.06.008 (2008).
Kessler, J., Calabrese, P., Kohn, N. & Kalbe, E. [PANDA versus MMST: Sensitivität und Spezifität zweier Screenings in der Demenzdiagnostik bei Parkinson.]. Aktuelle Neurologie doi:10.1055/s-2007-987941 (2007).
Tomlinson, C. L. et al. Systematic review of levodopa dose equivalency reporting in Parkinson's disease. Mov Disord 25, 2649–2653, doi:10.1002/mds.23429 (2010).
Fahn, S., Elton, R. and Members of the UPDRS Development Committee in Recent Developments in Parkinson's Disease, Vol. 2 (ed Marsden CD Fahn S, Calne DB, Goldstein M, eds.) 153–163, 293–304 (Macmillan Health Care Information, 1987).
Hoehn, M. M. & Yahr, M. D. Parkinsonism: onset, progression and mortality. Neurology 17, 427–442, doi:10.1212/wnl.17.5.427 (1967).
Beck, A. T., Ward, C. H., Mendelson, M., Mock, J. & Erbaugh, J. An inventory for measuring depression. Arch Gen Psychiatry 4, 561–571, doi:10.1001/archpsyc.1961.01710120031004 (1961).
Jenkinson, C., Fitzpatrick, R., Peto, V., Greenhall, R. & Hyman, N. The Parkinson's Disease Questionnaire (PDQ-39): development and validation of a Parkinson's disease summary index score. Age Ageing 26, 353–357, doi:10.1093/ageing/26.5.353 (1997).
Rizzo, M., Uc, E. Y., Dawson, J., Anderson, S. & Rodnitzky, R. Driving difficulties in Parkinson's disease. Mov Disord 25 Suppl 1, S136-140, doi:10.1002/mds.22791 (2010).
Madeley, P., Hulley, J. L., Wildgust, H. & Mindham, R. H. Parkinson's disease and driving ability. J Neurol Neurosurg Psychiatry 53, 580–582, doi:10.1136/jnnp.53.7.580 (1990).
Dubinsky, R. M. et al. Driving in Parkinson's disease. Neurology 41, 517–520, doi:10.1212/wnl.41.4.517 (1991).
Kremer, N. I. et al. Deep Brain Stimulation for Tremor: Update on Long-Term Outcomes, Target Considerations and Future Directions. J Clin Med 10, doi:10.3390/jcm10163468 (2021).
Singh, R., Pentland, B., Hunter, J. & Provan, F. Parkinson's disease and driving ability. J Neurol Neurosurg Psychiatry 78, 363–366, doi:10.1136/jnnp.2006.103440 (2007).
Worringham, C. J., Wood, J. M., Kerr, G. K. & Silburn, P. A. Predictors of driving assessment outcome in Parkinson's disease. Mov Disord 21, 230–235, doi:10.1002/mds.20709 (2006).
Wilkinson, L., Beigi, M., Lagnado, D. A. & Jahanshahi, M. Deep brain stimulation of the subthalamic nucleus selectively improves learning of weakly associated cue combinations during probabilistic classification learning in Parkinson's disease. Neuropsychology 25, 286–294, doi:10.1037/a0021753 (2011).
Mure, H. et al. Improved sequence learning with subthalamic nucleus deep brain stimulation: evidence for treatment-specific network modulation. J Neurosci 32, 2804–2813, doi:10.1523/JNEUROSCI.4331-11.2012 (2012).
Wylie, S. A. et al. Subthalamic nucleus stimulation influences expression and suppression of impulsive behaviour in Parkinson's disease. Brain 133, 3611–3624, doi:10.1093/brain/awq239 (2010).
van Wouwe, N. C. et al. Deep brain stimulation of the subthalamic nucleus improves reward-based decision-learning in Parkinson's disease. Front Hum Neurosci 5, 30, doi:10.3389/fnhum.2011.00030 (2011).
Fründt, O. et al. Do impulse control disorders impair car driving performance in patients with Parkinson's disease? Journal of Parkinson’s Disease (JDP) In press (2022).
Witt, K. et al. Neuropsychological and psychiatric changes after deep brain stimulation for Parkinson's disease: a randomised, multicentre study. Lancet Neurol 7, 605–614, doi:10.1016/S1474-4422(08)70114-5 (2008).

There is a conflict of interest The authors have no conflict of interest to report regarding this manuscript. General disclosures are as follows: Prof. Dr. C. Buhmann and Prof. W.H. Zangemeister received study research support from the Georg & Jürgen Rickertsen Stiftung Hamburg. Prof. C. Buhmann served on scientific advisory boards and/or received honoraria for lectures from Abbvie, Bial, Desitin, Kyowa Kirin, Merz, Stada, TAD Pharma, UCB and Zambon. Prof. C. Gerloff is supported by the German Research Foundation (DFG Ge 844/4-1, SFB 936) and the EU (FP7-HEALTH-2011-278276) and he received honoraria as speaker and/or scientific advisory board member from Actitor Biotech, Amgen, Bayer Healthcare, Boehringer Ingelheim, Prediction Biosciences and Sanofi Aventis. Dr. O. Fründt is supported by the Klaus Thiemann Foundation and received honoria for lectures from Daiichi-Sankyo and support for congress participation and trainings from Abbvie and St. Jude/Abbott. PD Dr. M. Pötter-Nerger is supported by DFG (SFB 936, C8), EU (MSC-ITN-MULTI), and received lecture fees from Abbott, Boston Scientific, Abbvie and Licher, study-related fees from Boston scientific, Zambon, Abbott and served as consultant for Boston Scientific and Abbvie. Prof. Dr. W. Hamel received lecture fees and honoraria for serving on advisory boards, travel grants and educational grants from Abbott, Aleva, Boston Scientific, and Medtronic. Dr. T. Mainka has received royalties from Elsevier in Urban & Fischer, speaker fees from Stadapharm and has served in an advisory board for Biomarin Pharmaceutical. She is supported by the BIH-Charité Clinician Scientist Program of the Charité-Universitätsmedizin Berlin and the Berlin Institute of Health. Prof. Dr. M. Südmeyer received grants and/or honoraria for consultancy by TEVA Pharma, Medtronic, Abbott/St. Jude Medical, Novartis, Desitin, Meda Pharma, AbbVie as well as grants from the Helmholtz Society, “Stiftung für Altersforschung” and “Forschungskommission”, Heinrich-Heine-University, Düsseldorf. PD Dr. C.K.E. Moll, E. Vettorazzi, Dr. U. Hidding, E. Baspinar and C. Schwarz report no disclosures.

SupplementarymaterialnpjPD27092022.docx

Download PDF

Editorial decision: revise
14 Nov, 2022
Review #2 received at journal
11 Nov, 2022
Review #1 received at journal
14 Oct, 2022
Reviewer #2 agreed at journal
09 Oct, 2022
Reviewer #1 agreed at journal
04 Oct, 2022
Reviewers invited by journal
02 Oct, 2022
Submission checks completed at journal
29 Sep, 2022
Editor assigned by journal
27 Sep, 2022
First submitted to journal
27 Sep, 2022

You are reading this latest preprint version

Effects of deep brain stimulation on driving in Parkinson’s disease - a prospective controlled study

Status:

Version 1

Abstract

Figures

Background

Materials And Methods

Participants

Data acquisition

Driving simulator

Driving sessions

Driving analysis

Statistical analysis

Data availability

Results

Clinical scores and questionnaires

Group comparisons of clinical changes from Baseline to Follow-up

Driving simulator performance

Error rates

Error severity

Error category

Driving time and Driving safety score (DSS)

Post-hoc analysis (outliers included)

Correlations of clinical characteristics with DSS

Discussion

Conclusion

Declarations

Acknowledgements

Funding

Author contributions

Competing interests

References

Additional Declarations

Supplementary Files

Status:

Version 1