Microstructure predicts motor outcomes following Deep Brain Stimulation in Parkinson’s disease

doi:10.21203/rs.3.rs-5093901/v1

Download PDF

Research Article

Microstructure predicts motor outcomes following Deep Brain Stimulation in Parkinson’s disease

https://doi.org/10.21203/rs.3.rs-5093901/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Deep brain stimulation of the subthalamic nucleus (STN-DBS) is an established treatment for both motor and non-motor symptoms in advanced Parkinson's disease (PD). Outcomes, however, can vary significantly among patients. Therefore, neuroimaging markers, such as neurite orientation dispersion and density imaging (NODDI), a biophysical model-based MRI technique, have been proposed to predict clinical outcomes and aid in preoperative counselling. In this prospective open-label study, we aimed to identify microstructural metrics associated with short-term motor outcomes following STN-DBS in PD. Thirty-five patients underwent diffusion-weighted MRI and comprehensive clinical assessments before surgery and at a six-month follow-up postoperatively. Using a whole-brain voxel-wise approach, we explored associations between microstructural metrics and motor outcomes, adjusting for multiple comparisons employing permutation-based methods. Intact microstructure within the bilateral putamen, bilateral insula, and left pallidum were associated with a higher reduction in postoperative motor symptom burden. Furthermore, preserved microstructure in pre- and postcentral gyrus and right precuneus were associated with increases in time with good mobility and no troublesome dyskinesia as well as decreases in time with poor mobility. These findings suggest that diffusion weighted MRI can be a valuable tool in preoperative counselling by identifying patients who are likely to experience above- or below-average motor responses.

Computational Neuroscience

Neurology

motor symptoms

Parkinson’s disease

deep brain stimulation

diffusion imaging

NODDI

microstructure

Deep brain stimulation of the subthalamic nucleus (STN-DBS) is an effective treatment for motor and non-motor symptoms in Parkinson’s disease (PD).^1–5 Besides its local effects on the micro- and mesoscale, it is believed that its effectiveness is based on the modulation of distributed neurophysiological networks.^6,7 Therefore, the microstructural integrity of the structures comprising these networks appears to be crucial to the effectiveness of DBS.⁸ In this regard, we recently demonstrated that beneficial outcomes of non-motor symptoms depend on intact microstructure within specific areas of the basal-ganglia-thalamo-cortical circuit.⁸ These findings support the notion that quantitative MRI techniques, such as neurite orientation dispersion and density imaging (NODDI), can support preoperative counselling.^8,9 This concept has gained increasing attention in recent years.¹⁰ NODDI is an advanced, multi-compartmental diffusion-weighted MRI technique, which is used to investigate the microstructure of brain tissue.¹¹ The model provides two voxelwise metrics of neurite morphology: the neurite density index (NDI), which reflects the density of axons and dendrites within a voxel, and the orientation dispersion index (ODI), which captures the degree of neurite dispersion, reflecting their alignment and dendritic arborization.¹¹ Notably, the connection between NODDI metrics and the underlying tissue properties has been validated through histological analysis.^12,13 Furthermore, NODDI models have a higher sensitivity than regular diffusion tensor imaging (DTI) models, underlining the prospect of implementing it as a clinical tool.¹⁴ Additionally, NODDI as supplementary DTI technique seems to be rather robust to the choice of acquisition, whereby NODDI protocols have a clinically feasible data acquisition time.¹⁴ Previous studies using NODDI in PD have shown that this technique identifies disease-related pathology, such as retrograde degeneration of the nigrostriatal pathway, and its metrics have been linked to bimanual motor control, disease severity, and disease duration.^14,15 Combining NODDI with conventional DTI metrics in a multi-parametric analysis therefore provides complementary information on microstructural properties and could serve as an imaging-based marker for predicting treatment outcomes. Thus, in the present study, we aimed to demonstrate that this approach can detect microstructural properties of brain areas that predict motor outcomes following STN-DBS in PD. Specifically, we focused on identifying regions whose microstructural metrics were associated with (1) changes in the overall burden of motor symptoms and (2) changes in time with good mobility and no troublesome dyskinesia (ON) as well as time with poor mobility (OFF). The findings of this study could help to guide preoperative patient counselling by identifying brain microstructure that predict above- or below-average motor response to STN-DBS.

Clinical outcomes

Thirty-five patients with PD (26 males, mean age 58.7 ± 7.4 years) were analysed. At the six-month follow-up, the following primary outcomes improved: Unified Parkinson's Disease Rating Scale (UPDRS) part III (p < .001, Cohen’s d = 1.02), time with good mobility and no troublesome dyskinesia (p < .001, Cohen’s d=-.99), and time with poor mobility (p < .001, Cohen’s d = .75). Furthermore, the following scales improved: UPDRS total score (p < .001, Cohen’s d = .94), UPDRS part I (p = .011, Cohen’s d = .48), UPDRS part IV (p < .001, Cohen’s d = .76), Parkinson's Disease Questionnaire (PDQ)-8 summary index (SI) (p < .001, Cohen’s d = .69), LEDD (p < .001, Cohen’s d = 1.3), and LEDD of dopamine agonists (p < .001, Cohen’s d = .7). Analysis of movement records additionally revealed beneficial effects of STN-DBS on time with good mobility and troublesome dyskinesia (p = .001, Cohen’s d = .55). Longitudinal changes of clinical outcomes are reported in Table 1 and Fig. 1.

Table 1

Baseline characteristics and outcomes at baseline and 6-month follow-up
	n	M	SD
Age [y]	35	58.7	7.4
Disease duration [y]	35	8.9	4.5
Sex (female/male) [%]	35	9/26	[25.7/74.3]
	Baseline			6-MFU			Baseline vs. 6-MFU
	n	M	SD	n	M	SD	p	Cohen’s d
UPDRS total score	34	40.7	14.7	35	28.3	11.7	< .001	.94
UPDRS part I	34	2.9	2.1	35	1.9	1.8	.011	.48
UPDRS part II	34	8.3	5.4	35	8.1	5.3	.562	.04
UPDRS part III	35	24.9	11.9	35	14.5	8.0	< .001	1.02
UPDRS part IV	34	6.0	3.1	35	3.8	2.6	< .001	.76
Movement Record
ON [%]	35	37.1	18.6	35	54.5	16.4	< .001	− .99
OFF [%]	35	23.9	15.9	35	11.8	16.4	< .001	.75
SLEEP [%]	35	31.9	11.1	35	32.0	7.9	.97	− .07
ON TD [%]	35	7.0	12.65	35	1.7	5.6	.001	.55
PDQ-8 SI	35	32.8	15.3	35	22.6	14.4	< .001	.69
LEDD [mg]	35	964.9	408.3	35	531.2	267.5	< .001	1.3
LEDD DA [mg]	35	266.5	130.7	35	176.9	125.3	< .001	.7

Abbreviations: 6-MFU = 6-month follow-up; LEDD = Levodopa equivalent daily dose; LEDD-DA = LEDD of Dopamine Agonists; PDQ-8 SI = 8-item Parkinson’s Disease Questionnaire summary index; SD = standard deviation; UPDRS = Unified Parkinson’s Disease Rating Scale.

Interaction between Fractional Anisotropy and postoperative change in motor symptom burden

We assessed the relationship between DTI metrics and postoperative motor symptom change using a generalised linear model corrected for multiple comparisons using a permutation-based approach. Lower regional mean fractional anisotropy (FA) values in the anterior division of the right cingulate gyrus, right inferior fronto-occipital fasciculus, and the left superior longitudinal fasciculus were associated with greater postoperative change in motor symptom burden, as measured by change in UPDRS part III values (negative clusters N1, N2, and N7 cluster-wise p-value (CWP): <.001-.035). No clusters with a positive association between FA and postoperative change in motor symptom burden were found. The associations of regional mean FA and motor symptom burden are depicted in Supplementary Table 1 and Supplementary Fig. 1.

Interaction between NODDI-parameters and postoperative change in motor symptom burden

Whole brain analysis of NODDI parameters showed that both, ODI and NDI, were associated with changes in postoperative UPDRS part III scores. These results are depicted in Supplementary Tables 2 and 3 as well as Figs. 2 and 3 (ODI) and Supplementary Figs. 3 and 4 (NDI-values). Higher ODI-values in the left pallidum, bilateral putamen, and bilateral insular cortex were associated with greater change in postoperative motor symptom burden (P1-6, CWP: <.001-.038, Fig. 2). Negative associations, however, were found exclusively in white matter (WM) regions adjacent to the left middle frontal gyrus, the left cingulate gyrus, and the left precuneus (N1-3, CWP: <.001-.019, Fig. 3). Furthermore, higher ODI- and NDI-values in the bilateral insular cortex were related to a greater reduction in postoperative motor symptoms (ODI: P3, P5, CWP: <.001-.003; NDI: P1-P3, CWP: <.001-.032). On the other hand, higher NDI-values in the left putamen were associated with a greater reduction in postoperative motor symptoms (P1, CWP: <.001). Negative associations between NDI and postoperative motor symptom burden were found in WM regions adjacent to the left middle frontal gyrus, left precentral gyrus, right precuneus, and left paracingulate gyrus, corresponding to the negative ODI-clusters mentioned before (N1-3, CWP: <.001-.019).

Interaction between microstructural metrics and time with good/poor mobility

Regional mean values of microstructural metrics associated with significant postoperative changes in time spent in ON and OFF are detailed in Supplementary Tables 4 and 5 as well as Supplementary Figs. 3 and 4. Lower values of microstructural metrics within gray matter (GM) regions of the right pre- and postcentral gyrus and the right precuneus were related to increased time spent in the ON-state (ODI: N3, N6, N7, CWP: <.001-.002). No corresponding negative NDI clusters were found. With regard to the time spent in OFF, associations of vast GM regions were found. Specifically, higher ODI-values in the right cingulate gyrus, left pre- and postcentral gyrus, and the right precuneus were related to greater postoperative change in time spent in OFF (ODI: P1-3, CWP: <.001). Furthermore, higher NDI-values in the left middle frontal gyrus and the left precentral gyrus were related to greater postoperative change in time spent in OFF (NDI: P1, P4, CWP: <.001-.01).

In the present study, we employed NODDI-DTI, a novel method for analyzing diffusion-weighted imaging (DWI) data, to investigate the association between cerebral microstructure and motor symptom changes following neurostimulation in PD. There are three key findings. First, we demonstrate that intact microstructure within the bilateral putamen, bilateral insula, and left pallidum are associated with a higher reduction in postoperative motor symptom burden. Second, we determine the white matter tracts whose selective degeneration is associated with below-average improvements in postoperative motor symptom burden. Third, we identify the structures whose preserved microstructure is linked to postoperative increases in time with good mobility and no troublesome dyskinesia as well as decreases in time with poor mobility.

Microstructure is associated with beneficial motor outcomes

PD is characterized by the cardinal symptoms akinesia, rigidity, and tremor which severely affect the quality of life of patients.^16,17 Clinically, these debilitating symptoms are assessed using the UPDRS to gauge the overall motor burden.¹⁸ The UPDRS part III was implemented as the standard surrogate parameter for overall mobility in PD and used as a primary outcome parameter in numerous clinical trials.^18–20 In the present study, whole brain analysis of NODDI parameters identified an association between higher ODI-values within bilateral putamen, bilateral insula, and left pallidum and a higher reduction in postoperative UPDRS part III values. As ODI is high in gray matter,²¹ this finding supports the notion that intact sprawling of dendritic processes in these areas is important for beneficial postoperative motor outcomes. Given the intricate relationship between the underlying disease mechanisms and the physiological impacts of the intervention, several aspects have to be considered when interpreting the present findings. First, the pathological accumulation of alpha-synuclein in intraneuronal Lewy inclusions results in neuronal degeneration and significant morphological changes in dendrites.^22,23 These tissue alterations are not detectable with conventional MRI, whereas NODDI is sensitive to neurite morphology. Previous research has demonstrated reduced putaminal ODI-values in PD patients compared to healthy controls, interpreted as decreased dendrite length and loss of spines in striatal medium spiny neurons, which are the primary targets of dopaminergic nigrostriatal projections. Therefore, it can be hypothesized that low ODI-values in the aforementioned structures reflect the deteriorated neurite morphology described by Kamagata as well as Schröder and colleagues.^23,24 Second, given the close topographical relationship between the STN, the putamen, and the pallidum as well as the integration of STN in basal ganglia-thalamo-cortical loops, it is essential to examine the putative mechanisms of action of DBS. Besides its effects on the micro- and mesoscale, the high-frequency electrical pulses from DBS electrodes impact interregional networks on a macroscale.²⁵ This network modulation has been shown to predict postoperative outcomes across motor as well as various non-motor symptoms.²⁶ Therefore, it can be hypothesized that the positive association between ODI and postoperative motor outcomes in the present study reflects the dependency of DBS on intact tissue structure to exert its network effects. In this regard, Hermann et al. highlighted the importance of intact microstructure in motor regions such as STN, substantia nigra (SN), and putamen for the efficacy of STN-DBS.²⁷ Using free interstitial fluid (V-CSF) as a marker of microstructure in a region of interest based analysis, the authors found a correlation between V-CSF-values in their region of interest (ROI) and postoperative improvements in UPDRS part III. Despite notable effect sizes, these results did not reach statistical significance after adjusting for multiple comparisons. In the present study, we overcome the limitations of ROI-based as well as correlation analyses by combining a whole brain approach with generalized linear modelling and partially confirm the findings named above (i.e. the association between intact putaminal microstructure and improvements in motor burden). Taken together, these results underline the potential of microstructural imaging to assess postoperative outcomes of neurostimulation in PD.

Besides associations with areas of the basal ganglia-thalamo-cortical loop, ODI showed a positive association with post-operative motor burden in the bilateral insular cortex. Historically, the insular cortex has been primarily considered a visceral-somatic region within the limbic system²⁸, being one of the major target areas of thalamocortical projections.^29,30 Studies suggested a role in integrating autonomic, cognitive-affective, and somatosensory information and therefore an involvement in mediating control of non-motor symptoms in PD.^31,32 In this regard, the authors specifically point to the pathological accumulation of alpha-synuclein³³ as a cause of altered insular function subsequently leading to non-motor symptoms such as aberrant sensory processing and urinary symptoms. On the other hand, recent studies have also demonstrated an influence of the insular cortex on gating mechanisms in motor control. Wu and colleagues found decreased functional connectivity from the pre-supplementary motor area to the mid-anterior insula in patients with PD.³⁴ The authors suggested that these disrupted connections indicate a lack of readiness for movement and may partly contribute to the difficulty in initiating movements in PD. Extending this hypothesis, Tinaz et al. proposed a circuit model that highlights the interaction between the insula and the dorsomedial frontal cortex (harboring supplementary motor area (SMA) and pre-SMA) for the generation of intentional movements.³⁵ In this model, the insula processes afferent viscerosensory and somatosensory information from the body and integrates them with the motivational and emotional context. This information is conveyed to the frontal cortex, which is involved in various higher-level cognitive functions,^36–38 and is supposed to generate the impetus to move. The authors conclude that altered insular function in PD patients is an important contributor to the observed difficulty of PD patients to internally generate intentional movements and to maintain the speed, size, and vigor of movements.³⁵ With respect to the effects of STN-DBS on the insular cortex, investigated the process of sensory gating of urinary bladder afferents following STN-DBS.³⁹ The authors demonstrated a modulation of the activity in the insular cortex and the thalamus by STN-DBS. These findings lead to the assumption that partially restoring basal ganglia function via DBS is beneficial for sensory gating and urinary outcomes in PD. Integrating our findings with these studies, we suggest that the positive association between ODI and postoperative motor outcomes in the present study reflects the dependency of DBS on intact tissue structure within the bilateral insula to exert beneficial effects on gating processes important for movement control.

Whole brain analysis of NODDI parameters identified an association between lower FA- and NDI-values in several WM tracts, including the cingulum and a detrimental motor response to STN-DBS. In healthy WM tissue, NDI is usually high, as fibers are highly coherent to another whereby low values of NDI represent axonal loss and degeneration.²¹ On the other hand, ODI in WM is typically low. As most of the negative FA- and NDI-clusters were overlapping with clusters in which ODI showed a negative association with postoperative motor symptom burden and lay within regions of high fiber crossing and dispersion, the selective degeneration of crossing fibers might underlie the observed relationship. The results, therefore, support the hypothesis that intact microstructure of the cingulum and its crossing fibers is important for beneficial postoperative changes in motor symptoms.

Microstructure is associated with beneficial outcomes of motor states

In PD, fluctuations in overall motor performance related to treatment are commonly referred to as ON and OFF states. Here, ON describes the time with good mobility and no troublesome dyskinesia, whereas OFF refers to the time with poor mobility.⁴⁰ In the present study, whole brain analysis of NODDI parameters identified an association between beneficial outcomes of motor states and intact microstructure within bilateral pre- and postcentral gyrus as well as right precuneus and cingulum. In particular, increased time with good mobility was associated with intact microstructure, i.e. ODI, within the right pre- and postcentral gyrus and the right precuneus. Furthermore, reduced time with poor mobility after STN-DBS was associated with intact microstructure within the right cingulate gyrus, left pre- and postcentral gyrus, and the right precuneus. Numerous functional MRI (fMRI) studies exploring the influence of stimulation on brain networks in PD found significant effects of neurostimulation on primary motor cortex connectivity (for a review see Miao et al. and Li et al.).^41,42 These studies consistently showed that STN-DBS significantly impacts the resting-state functional connectome by strengthening the coupling within the direct pathway while reducing the coupling in the hyperdirect pathway. Furthermore, connectivity between DBS electrodes and a distributed network of brain regions including primary motor cortex predicted clinical response to STN-DBS.⁴³ Integrating these studies with the fact that the pre- and postcentral gyrus harbor the structures essential for voluntary motor control and processing somatosensory information, it is conceivable that intact microstructure within these areas is essential for the effectiveness of DBS.

Limitations

This research has its limitations. First, while the NODDI model has been histopathologically validated and is widely used in PD research, there are no studies validating this model in PD. Second, the underlying assumptions of the NODDI model may be an oversimplification, potentially leading to reduced specificity. Third, the resolution of the DTI scan in our study is limited to 2.0 x 2.0 x 2.0 mm³. This resolution, however, is similar to previous NODDI studies^44–46 and was chosen to find a compromise between scanning time, image resolution, and signal to noise ratio, as longer durations in the scanner lead to more noticeable motion artefacts, a crucial consideration in PD. Furthermore, the general limitations of open-label studies are inherent to this study as well. Despite several advantages of this study design, open-label studies have several limitations including placebo effects that can significantly influence subjective measures and contribute to overall patient well-being. Furthermore, the inclusion of a control group is not practicable in open-label DBS studies. Nevertheless, the use of clinician rated forms, as employed in this study, can help to overcome the challenges named above and to minimize bias.

In conclusion, we identified a distinct spatial profile of microstructural alterations associated with outcomes of motor outcomes following neurostimulation in PD. In particular, intact microstructure within the bilateral putamen, bilateral insula, and left pallidum were associated with a higher reduction in postoperative motor symptom burden. Furthermore, preserved microstructure in pre- and postcentral gyrus and right precuneus and cingulum was associated with increases in time spent in ON as well as decreases in time spent in OFF. Our results highlight the possibility of preoperatively assessing microstructural alterations to support patient counselling and treatment planning by identifying patients likely to experience above- or below-average motor responses.

Participants

In this ongoing, prospective, and observational study, 37 PD patients were enrolled upon written informed consent. Inclusion criteria comprised indication for DBS surgery because of advanced PD following international criteria.⁴⁷ Patients with pathological MRI, concurrent neurological or psychiatric conditions, or impaired visual or auditory function were excluded from the study. Two patients had to be excluded due to missing follow-up data. The study was carried out following the Declaration of Helsinki and approved by the Philipps-University of Marburg ethics committee (study-number: 155/17).

Clinical Assessment

Patients were assessed at the preoperative baseline and six months after DBS surgery. Both study visits were conducted in the ON-medication state, while follow-up assessments were conducted in the ON-medication/ON-stimulation state (MedON/StimON). Standardized case report forms were used to collect demographic and clinical data, including the UPDRS, the PDQ-8, and movement records from the 36 hours preceding the study visit. Here, time with good mobility and no troublesome dyskinesia (ON), time with poor mobility (OFF), time with good mobility and troublesome dyskinesia (ON TD), and time spent sleeping (SLEEP) were recorded in a patient diary. The LEDD was calculated based on the method described by Jost et al.⁴⁸ Demographic and clinical characteristics are outlined in Table 1 and Fig. 1.

MRI Data Acquisition and Processing

At preoperative baseline, PD patients were scanned at the University of Marburg’s Core Unit Brain Imaging using a 3-Tesla Trio scanner (Siemens, Erlangen, Germany). The detailed acquisition protocol is reported in the supplementary material. At the time of image acquisition, images were investigated to be free of motion or ghosting and high frequency or wrap-around artefacts.

Image Processing

Analysis of structural images was performed within the FreeSurfer image analysis suite 7.1.1 (http://surfer.nmr.mgh.harvard.edu), as reported previously by our group.¹⁵ In short, the processing included skull stripping, automated Talairach transformation, cortical and subcortical segmentation, intensity normalization, tessellation of the gray/white matter boundary, automated topology correction, and surface deformation based on intensity gradients.

Analysis of DWI data was performed within the FMRIB Software Library (FSL) analysis suite 6.0.5.2 (https://fsl.fmrib.ox.ac.uk/fsl). Raw DWI volumes were registered and resampled to the first b0 volume to correct for eddy-current distortions and involuntary movements. Subsequently, the diffusion tensor for each voxel was estimated using linear regression and FA was derived from the diffusion tensor. Additionally, NODDI-DTI, a modification of NODDI, was employed to obtain NDI and ODI from the DTI data. Visual inspection of the b0-images confirmed that no changes beyond those in the tissue structure contributed to the observed effects. Regional analyses were performed by linearly registering the first b0 image of each scan to the structural T1-weighted image using a boundary-based method and resulting in an affine matrix. Using the inverse of this matrix, the T1-derived segmentations and brain masks were transformed to the diffusion space. Using linear and nonlinear transformations, FA-maps were first co-registered to the MNI152 space.⁴⁹ Subsequently, masked FA-, NDI-, and ODI-maps were registered to the MNI152 space using the transformation from the previous step and voxels of brain tissue not present in every subject were excluded from the analysis.

Statistical analysis

Statistical analysis of clinical outcomes was performed using MATLAB R2020b (The MathWorks, Inc.). Mean changes of clinical scores between baseline and 6-month follow-up (6-MFU) were assessed using Wilcoxon signed-rank- or t-tests, with the use of the Shapiro-Wilk test. We controlled the false discovery rate by employing the Benjamini-Hochberg method and calculated effect sizes according to Cohen.^50,51 Reported p-values are two-sided and were accepted as significant where p < .05.

Statistical voxelwise analysis of image data was performed using the tool mri_glmfit of FreeSurfer. Here, clinical outcomes were represented as percentage difference between baseline and follow-up measures of UPDRS part III, time with good mobility and no troublesome dyskinesia, and time with poor mobility. Associations between microstructure and time with good mobility and troublesome dyskinesia (ON TD) were not performed due to the small number of patients experiencing dyskinesia (n = 14).

Associations between metrics of microstructure and changes of clinical outcomes were carried out for the whole brain using a generalized linear model.⁵² A permutation-based approach based on the Analysis of Functional NeuroImages (AFNI) null-z simulator was used to correct for multiple comparisons employing 12,000 simulations under the null hypothesis.⁵³ Clusters were formed using a threshold of p < .01 and a clusterwise p-value was calculated. Results were accepted as significant with clusterwise p < .05.

Acknowledgement

The authors would like to thank the participants for their active engagement in this study.

Data and Code Availability

The data that support the findings of this study is available on request from the corresponding author (PAL). The data is not publicly available due to privacy or ethical restrictions. All tools used for the analysis of MRI data are based on FreeSurfer Version 7.1 (http://surfer.nmr.mgh.harvard.edu/) and FSL 6.0.5.2 (http://www.fmrib.ox.ac.uk/fsl) packages, which are freely available. Scripts for automation were written in tcshell and parts of the statistics were written in Python using the packages numpy, pandas, seaborn, matplotlib, nibabel and scipy, which are also freely available. Python program code for the analysis of NODDI-DTI is available from https://github.com/dicemt/DTI-NODDI.

Contributorship

PAL: study concept and design, data acquisition, data analysis, drafting of the manuscript

JF: data acquisition, drafting of the manuscript

MBo: data acquisition, surgical intervention, critical revision of the manuscript

AC: data acquisition, critical revision of the manuscript

HSD: study design, critical revision of the manuscript

JW: data acquisition, critical revision of the manuscript

AS data acquisition, critical revision of the manuscript

CA: data acquisition, critical revision of the manuscript

SK: critical revision of the manuscript

CN: data acquisition, surgical intervention, critical revision of the manuscript

LT: study design, critical revision of the manuscript

MBe: study concept and design, data acquisition, data analysis, drafting of the manuscript

DJP: study concept and design, data acquisition, data analysis, drafting of the manuscript

Financial disclosure/Conflicts of Interest

PAL was supported by the SUCCESS-Program of the Philipps-University of Marburg, the ‘Stiftung zur Förderung junger Neurowissenschaftler’, and the Professor Klaus Thiemann Foundation in the German Society of Neurology. JF reports no financial disclosures. MBo is a scientific consultant for Brainlab. AC has participated in a training course which was industry funded by Stada Arzneimittel AG. HSD was funded by the EU Joint Programme – Neurodegenerative Disease Research (JPND), the Prof. Klaus Thiemann Foundation in the German Society of Neurology, the Felgenhauer Foundation, the KoelnFortune program of the Medical Faculty of the University of Cologne and has received honoraria by Everpharma, Kyowa Kirin, Bial, Oruen, and Stadapharm. JW reports no financial disclosures. AS reports no financial disclosures. CA reports no financial disclosures. SK reports no financial disclosures. CN is a scientific consultant for Brainlab. LT received payments as a consultant for Medtronic Inc. and Boston Scientific and received honoraria as a speaker on symposia sponsored by Bial, Zambon Pharma, UCB Schwarz Pharma, Desitin Pharma, Medtronic, Boston Scientific, and Abbott. The institution of LT, not LT personally, received funding by the German Research Foundation, the German Ministry of Education and Research, and Deutsche Parkinson Vereinigung. MBe reports no financial disclosures. DJP has received honoraria for speaking at symposia sponsored by Boston Scientific Corp, Medtronic, AbbVie Inc, Zambon and Esteve Pharmaceuticals GmbH. He has received honoraria as a consultant for Boston Scientific Corp and Bayer, and he has received a grant from Boston Scientific Corp for a project entitled "Sensor-based optimisation of Deep Brain Stimulation settings in Parkinson's disease" (COMPARE-DBS). The institution of DJP, not DJP personally, has received funding from the German Research Foundation, the German Ministry of Education and Research, the International Parkinson Foundation, the Horizon 2020 programme of the EU Commission and the Pohl Foundation in Marburg. Finally, DJP has received travel grants to attend congresses from Esteve Pharmaceuticals GmbH and Boston Scientific Corp.

Deuschl G et al (2006) A randomized trial of deep-brain stimulation for Parkinson's disease. N Engl J Med 355:896–908. 10.1056/NEJMoa060281
Jost ST et al (2023) Non-motor effects of deep brain stimulation in Parkinson's disease motor subtypes. Parkinsonism Relat Disord 105318. 10.1016/j.parkreldis.2023.105318
Jost ST et al (2020) A prospective, controlled study of non-motor effects of subthalamic stimulation in Parkinson's disease: results at the 36-month follow-up. J Neurol Neurosurg Psychiatry 91:687–694. 10.1136/jnnp-2019-322614
Gronostay A et al (2023) Stratifying quality of life outcome in subthalamic stimulation for Parkinson's disease. J Neurol Neurosurg Psychiatry. 10.1136/jnnp-2023-332272
Sauerbier A et al (2024) Predictors of short-term anxiety outcome in subthalamic stimulation for Parkinson’s disease. NPJ Parkinson's disease 10:114. 10.1038/s41531-024-00701-6
Albano L et al (2023) Altered Functional Connectivity of the Subthalamic Nucleus in Parkinson's Disease: Focus on Candidates for Deep Brain Stimulation. J Parkinsons Dis 13:797–809. 10.3233/JPD-230005
Petry-Schmelzer JN et al (2019) Non-motor outcomes depend on location of neurostimulation in Parkinson's disease. Brain 142:3592–3604. 10.1093/brain/awz285
Loehrer PA et al (2024) Microstructure predicts non-motor outcomes following deep brain stimulation in Parkinson’s disease. NPJ Parkinson's disease 10. 10.1038/s41531-024-00717-y
Loehrer PA et al (2024) No evidence for an association of voxel-based morphometry with short-term non-motor outcomes in deep brain stimulation for Parkinson’s disease. NPJ Parkinson's disease 10:91. 10.1038/s41531-024-00695-1
Kamiya K, Hori M, Aoki S (2020) NODDI in clinical research. J Neurosci Methods 346. 10.1016/j.jneumeth.2020.108908
Zhang H, Schneider T, Wheeler-Kingshott CA, Alexander DC (2012) NODDI: practical in vivo neurite orientation dispersion and density imaging of the human brain. NeuroImage 61:1000–1016. 10.1016/j.neuroimage.2012.03.072
Grussu F et al (2017) Neurite dispersion: a new marker of multiple sclerosis spinal cord pathology? Ann Clin Transl Neurol 4:663–679. 10.1002/acn3.445
Sato K et al (2017) Understanding microstructure of the brain by comparison of neurite orientation dispersion and density imaging (NODDI) with transparent mouse brain. Acta Radiol Open 6:2058460117703816. 10.1177/2058460117703816
Kamiya K, Hori M, Aoki S (2020) NODDI in clinical research. J Neurosci Methods 346:108908. https://doi.org/10.1016/j.jneumeth.2020.108908
Loehrer PA et al (2022) Microstructural alterations predict impaired bimanual control in Parkinson's disease. Brain Commun 4:fcac137. 10.1093/braincomms/fcac137
Chaudhuri KR, Fung VSC (2016) Fast facts: Parkinson's disease. Health Press Limited
Calvano A, Timmermann L, Loehrer PA, Oehrn CR, Weber I (2023) Binaural acoustic stimulation in patients with Parkinson's disease. Front Neurol 14. 10.3389/fneur.2023.1167006
Goetz CG et al (2008) Movement Disorder Society-sponsored revision of the Unified Parkinson's Disease Rating Scale (MDS-UPDRS): Scale presentation and clinimetric testing results. Mov Disord 23:2129–2170. https://doi.org/10.1002/mds.22340
Jost ST et al (2022) Gender gap in deep brain stimulation for Parkinson’s disease. NPJ Parkinson's disease 8. 10.1038/s41531-022-00305-y
Sauerbier A et al (2021) The New Satisfaction with Life and Treatment Scale (SLTS-7) in Patients with Parkinson's Disease. J Parkinsons Dis. 10.3233/jpd-212823
Zhang H, Schneider T, Wheeler-Kingshott CA, Alexander DC (2012) NODDI: Practical in vivo neurite orientation dispersion and density imaging of the human brain. NeuroImage 61:1000–1016. https://doi.org/10.1016/j.neuroimage.2012.03.072
Halliday GM, Leverenz JB, Schneider JS, Adler CH (2014) The neurobiological basis of cognitive impairment in Parkinson's disease. Mov disorders: official J Mov Disorder Soc 29:634–650. 10.1002/mds.25857
Kamagata K et al (2016) Neurite orientation dispersion and density imaging in the substantia nigra in idiopathic Parkinson disease. Eur Radiol 26:2567–2577. 10.1007/s00330-015-4066-8
Schroter N et al (2022) Disentangling nigral and putaminal contribution to motor impairment and levodopa response in Parkinson's disease. NPJ Parkinson's disease 8:132. 10.1038/s41531-022-00401-z
Horn A (2022) Connectomic deep brain stimulation
Chu C et al (2022) Subthalamic and pallidal stimulation in Parkinson's disease induce distinct brain topological reconstruction. NeuroImage 255:119196. 10.1016/j.neuroimage.2022.119196
Hermann MG et al (2024) The connection of motor improvement after deep brain stimulation in Parkinson's disease and microstructural integrity of the substantia nigra and subthalamic nucleus. NeuroImage Clin 42:103607. 10.1016/j.nicl.2024.103607
Uddin LQ, Nomi JS, Hebert-Seropian B, Ghaziri J, Boucher O (2017) Structure and Function of the Human Insula. J Clin neurophysiology: official publication Am Electroencephalographic Soc 34:300–306. 10.1097/WNP.0000000000000377
Allen GV, Saper CB, Hurley KM, Cechetto DF (1991) Organization of visceral and limbic connections in the insular cortex of the rat. J Comp Neurol 311:1–16. 10.1002/cne.903110102
Clasca F, Llamas A, Reinoso-Suarez F (1997) Insular cortex and neighboring fields in the cat: a redefinition based on cortical microarchitecture and connections with the thalamus. J Comp Neurol 384:456–482. 10.1002/(sici)1096-9861(19970804)384:3<456::aid-cne10>3.0.co;2-h
Christopher L, Koshimori Y, Lang AE, Criaud M, Strafella AP (2014) Uncovering the role of the insula in non-motor symptoms of Parkinson's disease. Brain 137:2143–2154. 10.1093/brain/awu084
Sauerbier A et al (2021) Predictors of short-term impulsive and compulsive behaviour after subthalamic stimulation in Parkinson disease. J Neurol Neurosurg Psychiatry 92:1313–1318. 10.1136/jnnp-2021-326131
Braak H, Ghebremedhin E, Rüb U, Bratzke H (2004) Del Tredici, K. Stages in the development of Parkinson's disease-related pathology. Cell Tissue Res 318:121–134. 10.1007/s00441-004-0956-9
Wu T et al (2011) Functional connectivity of cortical motor areas in the resting state in Parkinson's disease. Hum Brain Mapp 32:1443–1457. 10.1002/hbm.21118
Tinaz S et al (2018) Insula as the interface between body awareness and movement: A neurofeedback-guided kinesthetic motor imagery study in Parkinson’s disease. 12:400054. 10.3389/FNHUM.2018.00496/BIBTEX
Loehrer PA et al (2016) Ageing changes effective connectivity of motor networks during bimanual finger coordination. NeuroImage 143:325–342. 10.1016/j.neuroimage.2016.09.014
Loehrer PA et al (2021) Increased prefrontal top-down control in older adults predicts motor performance and age-group association. NeuroImage 240:118383. 10.1016/j.neuroimage.2021.118383
Nettersheim FS et al (2019) Dopamine substitution alters effective connectivity of cortical prefrontal, premotor, and motor regions during complex bimanual finger movements in Parkinson's disease. NeuroImage 190:118–132. 10.1016/j.neuroimage.2018.04.030
Herzog J et al (2008) Improved sensory gating of urinary bladder afferents in Parkinson's disease following subthalamic stimulation. Brain 131:132–145. 10.1093/brain/awm254
Marsden CD, Parkes JD (1976) On-off effects in patients with Parkinson's disease on chronic levodopa therapy. Lancet (London England) 1:292–296. 10.1016/s0140-6736(76)91416-1
Li Z et al (2023) BOLD frequency-dependent alterations in resting-state functional connectivity by pallidal deep brain stimulation in patients with Parkinson's disease. J Neurosurg 139:1354–1365. 10.3171/2023.1.JNS221858
Miao J et al (2022) Use of Functional MRI in Deep Brain Stimulation in Parkinson's Diseases: A Systematic Review. Front Neurol 13:849918. 10.3389/fneur.2022.849918
Horn A et al (2017) Connectivity Predicts deep brain stimulation outcome in Parkinson disease. Ann Neurol 82:67–78. 10.1002/ana.24974
Kamagata K et al (2017) Gray Matter Abnormalities in Idiopathic Parkinson's Disease: Evaluation by Diffusional Kurtosis Imaging and Neurite Orientation Dispersion and Density Imaging. Hum Brain Mapp 38:3704–3722. 10.1002/hbm.23628
Mitchell T et al (2019) Neurite orientation dispersion and density imaging (NODDI) and free-water imaging in Parkinsonism. Hum Brain Mapp 40:5094–5107. 10.1002/hbm.24760
Ogawa T et al (2021) White matter and nigral alterations in multiple system atrophy-parkinsonian type. NPJ Parkinson's disease 7:96. 10.1038/s41531-021-00236-0
Postuma RB et al (2016) The new definition and diagnostic criteria of Parkinson's disease. Lancet Neurol 15:546–548. 10.1016/s1474-4422(16)00116-2
Jost ST et al (2023) Levodopa Dose Equivalency in Parkinson's Disease: Updated Systematic Review and Proposals. Mov disorders: official J Mov Disorder Soc. 10.1002/mds.29410
Andersson JLR, Jenkinson M, Smith (2007) S
Benjamini Y, Hochberg Y (1995) Controlling the false discovery rate: a practical and powerful approach to multiple testing. J Roy Stat Soc: Ser B (Methodol) 57:289–300
Cohen J (1988) Statistical power analysis for the behavioral sciences. L. Erlbaum Associates
Belke M et al (2010) Diffusion Tensor Imaging (DTI) in idiopathic REM sleep behaviour disorder (iRBD). Klinische Neurophysiologie 41:ID136
Nichols TE, Holmes AP (2002) Nonparametric permutation tests for functional neuroimaging: A primer with examples. Hum Brain Mapp 15:1–25. 10.1002/hbm.1058

The authors declare no competing interests.

supplementarymaterial.docx
supplementary material

Download PDF

Version 1

posted

You are reading this latest preprint version

Microstructure predicts motor outcomes following Deep Brain Stimulation in Parkinson’s disease

Status:

Version 1

Abstract

Figures

Introduction

Results

Clinical outcomes

Interaction between Fractional Anisotropy and postoperative change in motor symptom burden

Interaction between NODDI-parameters and postoperative change in motor symptom burden

Interaction between microstructural metrics and time with good/poor mobility

Discussion

Microstructure is associated with beneficial motor outcomes

Microstructure is associated with beneficial outcomes of motor states

Limitations

Conclusion

Methods

Participants

Clinical Assessment

MRI Data Acquisition and Processing

Image Processing

Statistical analysis

Declarations

Acknowledgement

Data and Code Availability

Contributorship

Financial disclosure/Conflicts of Interest

References

Additional Declarations

Supplementary Files

Status:

Version 1