The national DBS brain tissue network pilot study: need for more tissue and more standardization

ORIGINAL PAPER

The national DBS brain tissue network pilot study: needfor more tissue and more standardization

V. Vedam-Mai • N. Krock • M. Ullman • K. D. Foote • W. Shain •

K. Smith • A. T. Yachnis • D. Steindler • B. Reynolds • S. Merritt •

F. Pagan • J. Marjama-Lyons • P. Hogarth • A. S. Resnick • P. Zeilman •

M. S. Okun

Received: 26 January 2010 / Accepted: 11 June 2010 / Published online: 30 June 2010

� Springer Science+Business Media B.V. 2010

Abstract Over 70,000 DBS devices have been

implanted worldwide; however, there remains a pau-

city of well-characterized post-mortem DBS brains

available to researchers. We propose that the overall

understanding of DBS can be improved through the

establishment of a Deep Brain Stimulation-Brain

Tissue Network (DBS-BTN), which will further our

understanding of DBS and brain function. The objec-

tives of the tissue bank are twofold: (a) to provide a

complete (clinical, imaging and pathological) database

for DBS brain tissue samples, and (b) to make available

DBS tissue samples to researchers, which will help our

understanding of disease and underlying brain cir-

cuitry. Standard operating procedures for processing

DBS brains were developed as part of the pilot project.

Complete data files were created for individual patients

and included demographic information, clinical infor-

mation, imaging data, pathology, and DBS lead

locations/settings. 19 DBS brains were collected from

11 geographically dispersed centers from across the

U.S. The average age at the time of death was

69.3 years (51–92, with a standard deviation or SD of

10.13). The male:female ratio was almost 3:1. Average

post-mortem interval from death to brain collection

was 10.6 h (SD of 7.17). The DBS targets included:

subthalamic nucleus, globus pallidus interna, and

ventralis intermedius nucleus of the thalamus. In

16.7% of cases the clinical diagnosis failed to match

the pathological diagnosis. We provide neuropatho-

logical findings from the cohort, and perilead responses

to DBS. One of the most important observations made

V. Vedam-Mai � K. D. Foote � B. Reynolds �M. S. Okun (&)

Department of Neurosurgery, University of Florida,

100 S. Newell Drive, Room L3-100, P.O Box 100236,

Gainesville, FL 32610, USA

e-mail: [email protected]

N. Krock � M. Ullman � S. Merritt � A. S. Resnick �P. Zeilman � M. S. Okun

Department of Neurology, University of Florida,

Gainesville, FL, USA

W. Shain � K. Smith

Wadsworth Center, Albany, NY, USA

D. Steindler

Department of Neuroscience, University of Florida,

Gainesville, FL, USA

A. T. Yachnis

Department of Pathology, University of Florida,

Gainesville, FL, USA

F. Pagan

Georgetown University, Washington,

DC, USA

J. Marjama-Lyons

Department of Neurology, University of New Mexico,

Albuquerque, NM, USA

P. Hogarth

Department of Molecular and Medical Genetics,

University of Oregon, Eugene, OR, USA

123

Cell Tissue Bank (2011) 12:219–231

DOI 10.1007/s10561-010-9189-1

in this pilot study was the missing data, which was

approximately 25% of all available data fields.

Preliminary results demonstrated the feasibility and

utility of creating a National DBS-BTN resource for

the scientific community. We plan to improve our

techniques to remedy omitted clinical/research data,

and expand the Network to include a larger donor pool.

We will enhance sample preparation to facilitate

advanced molecular studies and progenitor cell

retrieval.

Keywords Pathology � Neuromodulation �Devices � Safety � Adverse events �Deep brain stimulation � Parkinson’s disease �Essential tremor � Dystonia

Introduction

A dedicated brain repository with a defined mission for

providing deep brain stimulation (DBS) tissue does not

currently exist. The main objective of such a brain

tissue banking resource would be to facilitate the

availability of high quality samples and data for

research purposes (Alafuzoff and Winblad 1993).

Brain tissue studies provide the important framework

for understanding neurological diseases, as well as for

providing a platform and infrastructure for develop-

ment of interventions (Haroutunian 2007). Presently,

there are only a handful of brain banks that harvest

Parkinson’s disease (PD) tissue, and none have a

specific focus on DBS (e.g. the AZ brain consor-

tium (http://www.azpd.org/), the ‘‘Penn Brain Bank’’;

(http://www.med.upenn.edu/cndr/donatingbrain.shtml),

and the Netherlands Brain Bank to name a few

(http://www.brainbank.nl/)). This critical limitation has

narrowed the scope as well as the overall number of

DBS tissue studies that could potentially address the

question of what determines success and failure in this

procedure.

DBS is a neurosurgical procedure where a small

wire with four lead contacts is inserted into one or two

sides of the brain to address various neurological and

neuropsychiatric disorders (e.g. PD, essential tremor,

dystonia, Tourette syndrome, obsessive compulsive

disease, depression, etc.) that may prove refractory to

standard treatments. DBS acts as a neuromodulatory

therapy, however, its exact mechanisms of action

remain unknown. DBS is a symptomatic treatment,

and its positive and negative effects are com-

pletely reversible (e.g. leads can be removed) and

programmable.

A very important aspect of any tissue resource is

the collection of detailed clinical information. This

information must be paired with high quality neuro-

pathological samples. In the case of DBS, the tissue

samples must also be accompanied by detailed

imaging, and even potentially by intra-operative

neurophysiological data. Additionally, examination

of the tissue studies available on PD-DBS has

revealed major discrepancies between the clinical

and pathological diagnoses (Aalten et al. 2006;

Roulson et al. 2005; Hughes et al. 1992). These

mismatches in clinical versus pathological diagnosis

will be important to better understand, especially

when attempting to evaluate DBS outcome and

mechanism of action.

DBS has largely supplanted lesion therapy (e.g.

pallidotomy, thalamotomy) in most regions of the

world, and has over the last two decades been shown to

be highly effective for addressing medication refrac-

tory movement disorders. DBS in the setting of PD for

example, has proven to be a powerful treatment for

addressing disabling motor fluctuations, and also for

increasing the number of daily hours of quality ‘‘on’’

time for individual sufferers (Weaver et al. 2005).

The most common diagnostic indication for DBS

has been PD, and thus it naturally was the focus of

our pilot DBS brain tissue repository. Although the

cause of PD is poorly understood, it is likely that a

combination of genetic and environmental factors is

involved (Weiner 2008; Hughes et al. 1992). Addi-

tionally, PD is thought not to be a single disease, with

most experts now subscribing to the view that PD is a

syndrome, which in most cases has a largely favor-

able response to dopaminergic therapy. In many

cases, it is believed that an abnormal protein aggre-

gation (e.g. the Lewy Body) feeds a degenerative

cascade which may result in cell dysfunction, and

ultimately cell death within very specific neuronal

populations (e.g. dopaminergic neurons of the sub-

stantia nigra, or a family of parallel segregated basal

ganglia circuits; Delong et al. 1984; Chu et al. 2009).

The diagnosis of PD remains a clinical bedside

endeavor, and it rests largely on the presence

of characteristic features including bradykinesia,

rigidity, gait/postural abnormalities, and a resting

220 Cell Tissue Bank (2011) 12:219–231

123

tremor, although in 20% of cases tremor may be

absent. Post-mortem studies have consistently

revealed that up to a quarter of patients clinically

diagnosed with PD actually had an alternative

pathology (Davie 2008; Zijlmans et al. 2004).

Although over 70,000 DBS devices have been

implanted worldwide (http://www.medtronic.com/

our-therapies/neurostimulators-movement-disorders/

index.htm), there remains a paucity of reports on the

associated pathology. An example of the limited

amount of available data is demonstrated in Table 1.

Mild and non-specific gliosis seem to be the cardinal

themes of all available publications, and most reports

have a common shortcoming—missing detailed clin-

ical information (e.g. standardized scales, pre-opera-

tive and post-operative imaging, and detailed

microscopic analysis). Moreover, there is an over-

riding assumption that DBS leads are well placed, and

that there is only one ‘‘ideal’’ place in each target that

may result in clinical benefit(s).

We are establishing the first comprehensive DBS-

Brain Tissue Network (DBS-BTN), which will make

two essential contributions to the field. First, the

DBS-BTN will provide a database to include critical

clinical, imaging, physiological, and pathological

information. The acquired information will include

basic imaging (which also has the benefit of revealing

any underlying gross brain pathology) as well as a

clear description of the DBS lead track (Fig. 1).

Second, the DBS-BTN will create a catalogue and

storage system to provide opportunities for research-

ers to describe tissue responses (peri-lead, up/down-

stream), in order to more accurately assess

pathological changes associated with the DBS lead(s)

(Fig. 2). The availability of these tissues should allow

the researcher to investigate in greater detail than the

currently available conventional histopathology. Use

of DBS-BTN tissue samples will enable detailed

neuroanatomical studies using various imaging meth-

ods (e.g. MRI, light/confocal microscopy, and elec-

tron microscopy), as well as new investigations of

genomics, proteomics, and progenitor cells from

brain samples (at DBS lead sites as well as in other

brain regions). Thus, the DBS-BTN will enable many

new research initiatives. The DBS-BTN database will

include these ‘‘research data’’ and thus provide a

unique infrastructure to investigate correlations

within a comprehensive clinical database. The pur-

pose of the DBS-BTN was to (1) identify gaps in

information available on DBS brains, (2) establish

standard operating procedures for collection/storage,

and (3) establish a resource for high quality material

in order to facilitate availability of tissue for a clinical

and research network interested in studying DBS. We

present the results from the first 19 brains acquired in

the pilot DBS-BTN, and we highlight lessons learned,

as well as the areas of improvement/cooperation that

will be necessary to make this a higher quality

resource (e.g. a fully functioning DBS-BTN) for the

DBS research community.

Methods

Patient recruitment and the brain donor program

The DBS-BTN pilot project was launched in 2006 by

an interdisciplinary team with the stated intention to

collect post-mortem brains of DBS patients from

within the University of Florida Shands Hospital

System (UF/Shands), as well as from outside DBS

implantation sites. Subjects/donors were prospec-

tively enrolled (IRB approved study) when still

living if possible, but in most cases (particularly

tissue collected outside of the UF/Shands system)

consent was obtained post-mortem.

Tissue acquisition

The first step in tissue acquisition was the notification

of death of a donor. This notification was documented

by several potential sources: family members, doc-

tors, allied health professionals, funeral home direc-

tors, tissue coordinators, hospice, etc. Upon arrival to

the pilot DBS-BTN, each brain received an accession

number, and a general information page was gener-

ated. This information was linked to the accession

number after removal of all personal information to

comply with HIPPA regulations, and this information

was later linked to a database. The post mortem

interval (PMI) for brain tissue removal depended

primarily on the donor’s location (within-town, or

out-of-town), but an attempt was made to keep brain

removal to within 24 h. Appropriate paperwork for

removal had to be verified and signed by family

members/next of kin (similar to the protocol of the

donor program established by Waldvogel et al., at the

University of Auckland, NZ; Waldvogel et al. 2008).

Cell Tissue Bank (2011) 12:219–231 221

123

Table 1 Studies reporting pathological response to DBS electrode

Author Year na Diagnosis Target Laterality Pathology

Boockvar 2000 1 ETb VIMc Bilateral Both stimulators terminated in Vim

Minor reactive changes noted from chronic DBS confined

to electrode tracks

Haberler 2000 8 PDd VIM(6),

STN(2)eBilateral Well-preserved neural parenchyma

Mild gliosis around lead track

Conclusion: Chronic DBS does not cause damage to adjacent brain tissue

Burbaud 2003 1 Chorea VIM Bilateral Minimal tissue damage near electrode tip

Conclusion: DBS has small impact on surrounding tissues

Chou 2004 1 MSAf STN Bilateral Neuron loss in SN and and basal ganglia

Numerous a-synuclein-positive glial cytoplasmic inclusions

in subcortical nuclei, cerebellum, and brainstem

Atypical and robust inflammatory reaction, numerous

glial cytoplasmic inclusions surrounding electrode termination sites

Lezcano 2004 1 MSA STN Bilateral Data not available on PubMed

Talmant 2006 1 MSA STN Bilateral Severe neuronal depletion in SN, but no Lewy bodies

Argyrophilic glial cytoplasmic inclusions positive for a-synuclein and

ubiquitin in the STN, putamen, globus pallidus, pontine nuclei and

cerebellar white matter, significant of MSA

Conclusions: DBS is not recommended for MSA as improvement

is time limited

Sun 2008 1 PD STN/ZIg Bilateral The neural tissue surrounding active and nonactive contacts responds

similarly, with a thin glial capsule and foreign-body giant cell reaction

surrounding the leads as well as piloid gliosis, hemosiderin-laden

macrophages, scattered lymphocytes, and Rosenthal fibers; Separate

tracts in adjacent tissue for intraoperative microelectrode and

semimacroelectrode passes together with reactive gliosis, microcystic

degeneration, and scattered hemosiderin deposition;

Active bilateral contacts used effectively for 6 years lie in the ZI

Guehl 2008 1 PD STN/ZI Bilateral Position of the right electrode track more posterior and less deep

than the left electrode track

Glial reaction due to electrode track visible slightly lateral to the

anterior part of the STN within the ZI & reached external part of the SN

No trace was found through the STN on the preceding or subsequent

histologic sections

The left electrode track passed along the dorsal border of the STN,

continued more medially within the ZI, and stopped in the SN

No trace was found through the STN

Pilitsis 2008 1 Epilepsy Anterior

nucleus

Bilateral Minimal tissue damage, mild astrocytosis, and mild inflammation

surrounding the electrode termination site

Compared to control tissue, no significant difference other than mild

inflammation along the lead track

a Numberb Essential tremorc Ncl ventralis intermedius thalamid Parkinson’s diseasee Subthalamic nucleusf Multiple system’s atrophyg Zona Incerta

222 Cell Tissue Bank (2011) 12:219–231

123

Pathological methods

Fixation and blocking

A Standard Operating Procedure (SOP) was devel-

oped for the DBS-BTN pilot study to standardize the

harvest, collection, and processing of tissue. Per this

SOP, the brain harvest procedure was performed in

the autopsy suite at UF/Shands and the McKnight

Brain Institute if the death occurred in town, or

nearby. If death of a subject occurred out of town, the

brains were removed by outside neuropathologists

following the DBS-BTN protocol. Prior to removal of

the subject’s brain, the DBS lead wires were cut

enabling the DBS leads to be left in the brains with

minimal disruption. Each brain was weighed, and

gross examination was performed by a Board certi-

fied neuropathologist (ATY). Each brain was placed

in a 10% Zn-buffered formalin fixative solution.

After the tissue was completely fixed, usually in

1–2 weeks, the brain was prepared for dissection.

The cerebellum and brainstem were separated at

the level of the caudal brainstem, and the brain was

sectioned in the coronal plane into 3-cm slabs. The

goal of the initial blocking was to obtain a 3 9 3 cm

column of CNS tissue surrounding the lead track.

This ensured that adequate tissue was collected to

study changes in cell and tissue morphology and the

organization around the lead. This procedure allowed

for a clear description of brain tissue changes as a

function of the distance from the lead. We used

available MRIs or CTs to identify the DBS target

(lead tip) and to describe the device trajectory. As

part of the pilot study we obtained post-mortem

imaging information on four brains to identify the

general pathology, and the location/trajectory of the

DBS lead (Fig. 1). The information from imaging in

these cases demonstrated the potential utility of using

these data for planning the brain cutting and obtain-

ing useful samples. Once the tissue columns were

dissected, a series of 1 cm thick tissue blocks were

collected from the surface of the brain until the end of

the lead cavity. All steps of the dissection process

were photographed to ensure accurate tissue block

identification.

Fig. 1 Panel a MRI imaging showing the right DBS lead

terminates in thalamus superior to the intended target of the

STN. Panel b MRI imaging showing that the left sided lead

terminates deeper in the STN/brainstem region. The two slices

reveal that the trajectories were different, and that the antero-

posterior difference in lead placement could not be captured in

the same coronal sections without using multiple MRI slices.

This type of preliminary data has been critical in guiding the

brain cutting procedures. This scan was obtained using a 3T

magnet with overnight imaging (12 h). We can now obtain

similar information using a 1 h inverted MRI T1 sequence for

1 h or less of scan time

Fig. 2 An example of a traditional histopathology section,

showing the DBS lead track within the subthalamic nucleus in

a brain from the UF DBS-BTN pilot

Cell Tissue Bank (2011) 12:219–231 223

123

The tissue block containing the terminal ends of

the lead including the distal most lead contact was

processed by conventional histopathological tech-

niques (Fig. 2). Associated astrocytic gliosis was

assessed by immunohistochemical study for glial

fibrillary acidic protein (GFAP) and in occasional

cases a Masson’s Trichrome stain was performed to

evaluate collagen accumulation near the lead tip.

Representative areas of the brain were sampled, and

embedded in paraffin blocks for pathological analy-

ses. The remaining tissue was stored in the fixative

solution in a covered plastic storage box. The fixative

solution was replaced at least every 6 months.

Immunohistochemistry and pathological methods

Data were recorded for each donor and included the

date, time, and cause of death, age, sex, PMI and DBS

characteristics. For standard pathological examina-

tion, brains received by the DBS-BTN were routinely

dissected from select and standard regions of the brain,

as described previously by the Columbia Brain Bank

Protocol (Vonsattel et al. 2008). These tissue regions

were subject to standard tissue processing, and paraffin

embedding. Paraffin blocks were sectioned at 5 lm,

and were stained with the standard hematoxylin-eosin

(H&E) tissue staining method. For neuropathological

confirmation of neurodegenerative disease diagnosis,

standard immunohistochemistry was performed on

select tissue sections, using antibodies against all, or a

subset of the following proteins: b-amyloid peptide

(1:15 dilution, DAKO M0872), phosphorylated tau

(1:2000 dilution, DAKO A0024), a-synuclein (1:100,

Novocastra NCL-ASYN), and ubiquitin (1:100 dilu-

tion DAKO Z0458) (others employed on a case by case

basis; Spillantini et al. 1997; Schubert et al. 1991;

Perry et al. 1987). Briefly, antigen retrieval was

performed for only a-synuclein by incubating slides

in Trilogy solution (Cell Marque) at 92�C for 25 min.

The histopathological findings were described by a

neuropathologist (ATY) and were based on the

standard, widely accepted criteria for neurodegenera-

tive diseases (Waldvogel et al. 2008; Braak et al. 2001,

2004; McKeith et al. 1996; Schiesling et al. 2008;

Beach et al. 2009). Any additional tissue regions that

were required for the specific diagnosis of a neurode-

generative disorder were by protocol sampled at the

discretion of the trained neuropathologist (Waldvogel

et al. 2006). Tissue samples can be processed for

additional immunohistochemistry (eg: GFAP, Iba1)

and confocal imaging to further our knowledge on

the cellular reactivity produced by lead-tissue

interactions.

Database

Two standardized reports were generated and stored

for each patient based on the results of the gross/

microscopic examinations and other available out-

come data. All medical records were requested for

each patient, and pre-defined fields were populated in

a database. A designated clinical study coordina-

tor collected clinical data. The specific fields were

defined by a group of expert movement disorders

neurologists who examined the plethora of informa-

tion collected on DBS brains, and narrowed the fields

to the most critical for research purposes.

Results

General characteristics of the DBS-BTN pilot

study

Between 2006 and 2009, 19 subjects were enrolled

into the pilot study. Ten of the brains were from

outside the UF system. Brains were logged into the

system and institutional identity was protected during

the data processing procedures. The UF site was the

only site prospectively enrolling potential donors for

this pilot study (other sites retrospectively consented

subjects). Over 60 months, 19 DBS brains were

collected from eleven total sites (including the UF

system).

The age range for the collected brains was 51–

92 years, and the mean age was 69. The average age

for females was 70.5 years (SD = 12.2), and for

males it was 68.9 (SD = 9.9). Males were more

represented (14 vs. 5).

Identification of missing data points

The overarching goal of the pilot DBS-BTN was to

assess needs for a larger DBS-BTN and to specifically

and objectively assess all missing data points (clinical

and pathological) that may have been important for

quantifying outcome. Table 2 summarizes all of the

clinical, imaging, and pathological data points that

224 Cell Tissue Bank (2011) 12:219–231

123

were collected. The table reveals the number of points

available, the number of points missing, and the

percentage of completed items. Strikingly, the data

missing was that from the recording of objective scale

based scores. Predominantly, mini-mental status

examination (MMSE) post-op scores for almost 68%

and MMSE pre-op scores for 27% of the patients were

not recorded. Also Mattis Dementia Rating Scale

(DRS) pre-op scores for 48% and DRS post-op scores

for 74% of the subjects enrolled were unknown.

Further, early and late cognitive and mood related

issues were not known for 16 and 48% of these

patients, respectively. The lack of available clinical

information was significant, and appeared in multiple

domains (see Table 2). Other potentially important

information that was missing included specific details

of lead location and stimulation parameters (almost

45% of this data section was not available), the angle

of the intended lead trajectory, microelectrode and

macroelectrode intraoperative physiology data, and

actual measurements of lead location.

DBS characteristics

The brain areas targeted included the subthalamic

nucleus or STN (n = 19 leads in 12 patients), the

Globus pallidus internus or GPi (n = 9 leads in 6

patients), and the ncl ventralis intermedius thalami or

Vim thalamus (n = 2 leads in 2 patients). Two

patients had two different brain targets implanted

(n = 2 STN, n = 1 Vim thalamus, n = 1 GPi). The

number of months of stimulation varied from 1 to

102, with the average being 35.4 months for the left

side (SD = 22.5), and 41.5 months for the right side

(SD = 29.0).

The lead placement based on available information

was grossly judged (from the combination of avail-

able imaging, clinical response, and pathology) to be

likely accurate in 64.7% of subjects, possibly accu-

rate in 23.5% of the subjects, and inaccurate in 11.8%

of subjects. This determination was made based a

gross determination that the lead reasonably reached

the intended target and also from available clinical

data. Though the exact determination of x, y, z

imaging and pathological correlation of lead loca-

tions was not precisely available on all cases, enough

information was available to make preliminary and

very general determinations of lead locations in all

cases.

Clinical features

Primary clinical diagnosis

The brains were collected from 11 geographically

dispersed centers from across the United States. The

average post-mortem interval for brain removal was

10.8 h. The primary clinical diagnosis was PD

(n = 16; 88.8%), but other diagnoses included multi-

ple system atrophy (n = 1; 5.5%), and essential

tremor (n = 1; 5.5%). The mean disease duration of

the cohort was 18.1 years (SD = 13.5).

Post-operative clinical symptoms

Post DBS surgery, there was a reported improvement

of PD symptoms as perceived by the clinician and by

the patients regardless of diagnosis, or lead place-

ment. This improvement was reported both early

(within the first 6 months) as well as late (chronic). It

is interesting to note that although there was much

clinical discrepancy with the lead placement (12%

inaccurate, and 23.5% possibly accurate), both the

patient and the clinician perceived a significant

improvement following DBS in all cases. Some of

the early (within the first 6 months) cognitive, mood

and behavioral issues encountered in patients

included hallucinations (n = 5), dementia (n = 5),

and impulsivity/compulsions (n = 3). Such early, as

well as late cognitive issues (after the first 6 months)

were poorly documented (47% of data was unavail-

able). These data raise important questions as to the

validity of clinician and patient derived general

outcomes.

Objective scale based scores

A mini-mental status examination (Folstein et al.

1975) is commonly used as a test to screen for

dementia. Usually, a score of greater than or equal to

25 points (Folstein et al. 1975) is considered accept-

able, and anything below may be indicative of

developing or possibly even frank dementia (how-

ever, these cut-offs frequently are oversimplified). In

our pilot study, the pre-operative MMSE data was

available for 74% of patients, however, the post-op

MMSE data was only available for 32%. The DRS

(Mattis dementia rating scale), another test designed

to screen and track the mental state of adults was

Cell Tissue Bank (2011) 12:219–231 225

123

Table 2 Patient documentation statistics

Number

available

Number not

available

Total number

of patients

Percent

available

(%)

Percent not

available (%)

Number not

performed

Percent not

performed (%)

General characteristics

1� Clinical diagnosis 19 0 19 100.00 0.00

PMI (hours) 18 1 19 94.74 5.26

Cause of death 16 3 19 84.21 15.79

Pathological diagnosis 17 2 19 89.47 10.53

2� Neurological diagnosis 19 0 19 100.00 0.00

2� Non-neurological diagnosis 19 0 19 100.00 0.00

Family history (neurological) 17 2 19 89.47 10.53

City, state of original

implantation

18 1 19 94.74 5.26

DBS characteristics

Age 19 0 19 100.00 0.00

Disease duration of 1�neurological diagnosis

16 3 19 84.21 15.79

Side/target/lead replaced?

(first lead)

19 0 19 100.00 0.00

# Mo’s of stimulation (first lead) 18 1 19 94.74 5.26

Side/target/lead replaced?

(second lead)

19 0 19 100.00 0.00

# Mo’s of stimulation

(second lead)

18 1 19 94.74 5.26

Pre-operative clinical symptoms

Tremor 17 2 19 89.47 10.53

Gait/postural stability 16 3 19 84.21 15.79

On/off fluctuations dyskinesias 16 3 19 84.21 15.79

Dystonia 15 4 19 78.95 21.05

Non-motor side effects 14 5 19 73.68 26.32

Autonomic symptoms 15 4 19 78.95 21.05

Falls 16 3 19 84.21 15.79

Frontal release signs 15 4 19 78.95 21.05 3 15.79

Swallowing problems 17 2 19 89.47 10.53

Post-operative clinical symptoms

Patient perceived dbs improved

symptoms 1st 6 months

18 1 19 94.74 5.26

Clinician perceived dbs

improved symptoms 1st

6 months

16 3 19 84.21 15.79

Patient perceived DBS improved

symptoms late/chronic

16 3 19 84.21 15.79

Clinician perceived DBS

improved symptoms

late/chronic

17 2 19 89.47 10.53

Early cognitive/mood/behavioral

issues (1st 6 months)

17 2 19 89.47 10.53

Late cognitive/mood/behavioral

issues (after first 6 months)

11 8 19 57.89 42.11

226 Cell Tissue Bank (2011) 12:219–231

123

Table 2 continued

Number

available

Number not

available

Total number

of patients

Percent

available

(%)

Percent not

available (%)

Number not

performed

Percent not

performed (%)

Objective scale based scores

Pre-Op DRS 10 9 19 52.63 47.37 1 5.26

Post-Op DRS 5 14 19 26.32 73.68 2 10.53

Pre-Op MMSE 14 5 19 73.68 26.32 1 5.26

Post-Op MMSE 6 13 19 31.58 68.42 1 5.26

Pre-Op DSM IV 4 15 19 21.05 78.95 9 47.37

Post-Op DSM IV 4 15 19 21.05 78.95 9 47.37

UPDRS baseline OFF 15 4 19 78.95 21.05

UPDRS baseline ON 13 6 19 68.42 31.58

UPDRS baseline off/on med

improvement

12 7 19 63.16 36.84

UPDRS baseline off/longest

follow-up off meds/On DBS

13 6 19 68.42 31.58

UPDRS baseline off/longest

follow-up on med/On DBS

13 6 19 68.42 31.58

TRS baseline 1 1 2 50.00 50.00 1 50.00

TRS baseline/longest follow-up

off DBS

1 1 2 50.00 50.00 1 50.00

TRS baseline/longest follow-up

on DBS

1 1 2 50.00 50.00 1 50.00

Lead location and stimulation parameters #1

Side-target 19 0 19 100.00 0.00

Stimulation parameters 19 0 19 100.00 0.00

AC-PC angle 13 6 19 68.42 31.58

Microelectrode passes 10 9 19 52.63 47.37

Macroelectrode passes 10 9 19 52.63 47.37

Measured lead tip location 12 7 19 63.16 36.84

0 Contact location 14 5 19 73.68 26.32

1 Contact location 13 6 19 68.42 31.58

2 Contact location 13 6 19 68.42 31.58

3 Contact location 13 6 19 68.42 31.58

Lead location and stimulation parameters #2

Side-target 12 0 12 100.00 0.00

Stimulation parameters 12 0 12 100.00 0.00

AC-PC angle 6 6 12 50.00 50.00

Microelectrode passes 2 10 12 16.67 83.33

Macroelectrode passes 3 9 12 25.00 75.00

Measured lead tip location 7 5 12 58.33 41.67

0 Contact location 8 4 12 66.67 33.33

1 Contact location 6 6 12 50.00 50.00

2 Contact location 6 6 12 50.00 50.00

3 Contact location 6 6 12 50.00 50.00

Cell Tissue Bank (2011) 12:219–231 227

123

available for only part of the cohort. A score below

130 is noteworthy. The DRS scores were available

for only 53%/73% of the subjects with respect to pre-

and post-operative screenings.

Pathological features

Pathological assessment (clinical and pathological

mismatches)

The majority of the patients were diagnosed clinically

with idiopathic PD (Braak et al. 2003). The patho-

logical diagnosis failed to correlate with the clinical

diagnosis in 3/19 cases (15.78%). The clinical

diagnosis for one of the patients was determined to

be multiple system atrophy, however, upon patho-

logical examination it was shown that the subject

actually had idiopathic PD with diffuse Lewy body

deposition (McKeith et al. 1996). Of the other two

subjects, both had multiple system atrophy at post-

mortem examination.

One patient had a severe loss of nigral cells, but

did not exhibit any Lewy Bodies or neurofibrillary

pathology. This patient had a family history of PD

and we suspect carried one of the PD genes (e.g.

Parkin, or a LRRK2 variant; Schiesling et al. 2008).

This was a significant observation as the patient was

clinically diagnosed with PD, and had a tremendous

response to levodopa and to DBS therapy.

The primary cause of death of the subjects enrolled

in our pilot study was variable: myocardial infarction

(n = 4, 21%), respiratory arrest (n = 3, 15.8%), organ

failure (n = 2, 10.5%), cancer (n = 2, 10.5%), anoxic

brain injury (n = 1, 5.2%), drowning (n = 1, 5.2%),

car accident (n = 1, 5.2%), and stroke (n = 1, 5.2%).

Forty-two percent lacked a family history of any

neurological disorder. The most common ailment for

those with a positive family history was PD (26.3%).

Other neurological comorbidities included essential

tremor (10.5%), depression (5.2%), stroke (5.2%), and

dementia (5.2%).

Pathological and peri-lead responses

Neuropathologically the cohort of brains revealed the

following: cortical Lewy Bodies (n = 5), changes

indictative of Alzheimer’s (n = 1), changes indica-

tive of hypoxic-ischemic encephalopathy (n = 2),

Fig. 3 A representative sample of the cellular reactive

responses seen at a DBS lead site. A confocal image taken

from a 100 lm thick tissue section. The image is from the STN

containing the DBS lead site (DBS SITE). Immunohistochem-

istry shows astrocytes (GFAP = magenta), microglia (Iba-

1 = cyan), and all cell nuclei (CyQuant = yellow). The

microglia (white arrows) seen throughout the image are in an

amoeboid reactive state. The astrocytes (white arrow heads)

are also reactive with extended and enlarged processes.

The astrocytes and microglia can be seen in layers around

the DBS site. The field denoted by asterisks shows a large

blood vessel coursing into the area near the lead site.

Calibration bar = 100 lm

Table 2 continued

Number

available

Number not

available

Total number

of patients

Percent

available

(%)

Percent not

available (%)

Number not

performed

Percent not

performed (%)

Pathological response around DBS lead

Pathological response around

DBS lead

17 2 19 89.47 10.53

Lead location clinical 13 6 19 68.42 31.58

228 Cell Tissue Bank (2011) 12:219–231

123

nigrostriatal degeneration with oligodendroglial

inclusions (2), and one PD brain with nigral degen-

eration but no Lewy Bodies (n = 1). An example of

the results of the techniques employed can be seen in

Fig. 3, showing both reactive astrocytes and reactive

microglia relative to the DBS lead site. These data

reveal that the reactive gliosis (astrocytes and microg-

lia) is oriented in layers surrounding the DBS lead

site. Post-mortem histological analysis revealed that

in all the subjects except for one, the peri-lead

pathological response was minimal, and was primarily

gliosis. This tissue specimen showed the most

dramatic response to the DBS lead. A thick fibro-

collagenous wall was discovered around the DBS

lead, which was associated with chronic inflamma-

tion. This observation can be further confirmed by

techniques such as Masson’s trichrome staining. The

peri-lead pathological response was also evaluated

and the following characteristics were uncovered:

gliotic cuffing (n = 3), an organized cavity with a

mild astrocytic response (mainly microglia; n = 8;

Fig. 3), mild hemorrhage at the lead site (n = 3), peri-

lead stroke (n = 1), organizing cortical necrosis at the

lead insertion site (probable prior hemorrhage;

n = 3), and the formation of a fibrocollagenous wall

of chronic inflammation (as confirmed by Masson’s

trichrome stain; n = 1).

Discussion

DBS is becoming an increasingly important therapy

for a wide variety of movement disorders. This report

presents the initial pilot findings from 19 brain

samples utilized to establish a National Pilot DBS-

BTN. The preliminary results demonstrated the fea-

sibility of such a tissue bank for the preparation, and

storage of DBS brain specimens, but also revealed

shortcomings particularly in obtaining complete clin-

ical datasets. The data revealed the post-mortem

cohort on average was close to age 70, and there was a

larger than expected preponderance of men (2:1 is the

usual male to female disease ratio, and in our study it

was almost 3:1. The most common clinical diagnosis

was PD (89%), and post-mortem examination

revealed (similar to non-DBS studies) diagnostic

mismatch in three cases (18%). The primary brain

target utilized was STN (n = 19 leads) and lead

placement was grossly inaccurate in 12% of the

cohort, despite patient and clinician reports of signif-

icant benefit. In general, our data supports the results

of previous post-mortem studies, which have shown

that a glial response (e.g. mild inflammation) is

consistently generated to DBS leads (Sun et al. 2008;

Nielsen et al. 2007; Haberler et al. 2000; Henderson

et al. 2002). Interestingly we were able to observe a

few cases of peri-lead inflammatory cuffing, as well as

cortical and peri-lead hemorrhage/stroke.

This pilot study revealed some important suc-

cesses and failures that will need to be addressed in

the future. Perhaps the most important observation

was the missing data rate which of almost 25% of all

data fields sought by investigators (Table 2). In total,

the relatively small dataset sought by the pilot

investigators proved extremely hard to obtain across

11 centers. The future success of the DBS-BTN will

therefore be highly dependent on limiting the lack of

missing data, and also by introducing standardization

of collection instruments and techniques.

There were several important successes derived

from this project. There was an establishment of a DBS

SOP for the preparation of brain tissue. Establishing

such a standardized protocol enabled the generation of

high quality tissue, which was essential for the

maintenance of a systematic histological review of

brain regions. In order to ensure that our samples were

suitable for further use, the collection, dissection, and

processing of samples were carried out only by an

experienced and qualified multidisciplinary team.

Additionally, we developed a clinico-pathological-

imaging post-mortem DBS database, a protocol to

preserve tissue for proteomics/genomics, and a proto-

col to preserve and grow post-mortem cells from

various brain regions (both cells and progenitor/stem

cells). The pilot DBS-BTN was also successful in the

collection of 19 post-mortem DBS brains from 11

different geographical locations/implanting centers

throughout the United States. The study defined a

SOP for brain fixation, and generation of basic

histological profiles, and it developed a quick and

simple 3T MRI sequence to examine post-mortem

DBS brains. The imaging development was important

for describing the lead trajectory, as well as for

describing the position of the DBS lead tips prior to the

brain cutting sessions. This information allowed us to

refine our brain cutting technique and facilitated the

availability of a larger amount of useful tissue. To our

knowledge there have been no post mortem MRI

Cell Tissue Bank (2011) 12:219–231 229

123

studies of human brain confirming lead position. These

studies can be hampered by artifacts and by fixation

issues once the brain has been removed from the skull.

Usually histological sectioning approaches are utilized

for localization, however, these techniques in isolation

often fail to provide an accurate picture of lead

location. Although we had some CT and MRI images

pre-mortem, post-mortem imaging, histology, and

information on clinical outcome, it is not clear that

this was enough to justify the classification of lead

placement/misplacement we used for this paper. This

component of the repository will need to be addressed

in the future in lieu of establishing a correlation

between lead placement and patient’s/clinician’s per-

ception of the DBS procedure.

There were several important failures that can be

improved upon. One failure was in systematically

keeping the PMI’s as short as possible. Some

biomarkers have relatively short half-lives, and

specific antigenicity that can be lost over time.

Typically, advanced studies requiring fresh tissue

involve RNA (such as gene expression studies), and it

is crucial that the PMI be as short as possible

(typically \ 24 h). The range of studies that a

specific tissue donation can support differs from

one sample to another (Haroutunian and Pickett

2007), and we need to attempt to maximize this

range. The DBS BTN utilizes a system where we

attempt to pre-morbidly consent subjects, although

when not possible or feasible we utilize post-mortem

consenting.

Forming a more integrated network and adopting

common institutional protocols/procedures as well as

collecting a common dataset will be critical for the

future of the DBS-BTN. New molecular research

techniques for gene and protein expression studies

will require the use of snap-frozen tissue, and we will

need to begin to routinely collect this material. Even

though the use of frozen tissue has limitations (such

as the need for a shorter PMI, proper freezing

techniques and appropriate storage and handling), the

studies of gene and corresponding protein expression

can best be performed with appropriately frozen

specimens (Schmitt et al. 2007). Immunohistochem-

ical methods using specific antibodies usually work

best on tissue that has been formalin-fixed/paraffin

embedded material. For our future tissue collections

and SOPs, we plan to broaden our procedures to

include both fixed as well as frozen materials.

It is our hope that in the future we will be able to

offer other researchers our tissue resources as well as

clinical databases in a collaborative manner, in order

to further the understanding of DBS, PD and other

diseases potentially addressed by neuromodulation.

Acknowledgments This work was supported by the Adelaide

Lackner Professorship to MSO, The University of Florida

Foundation Funds, The Eric and Jennifer Scott Fund, the

National Parkinson Foundation Center of Excellence, and NIH

(R01-NS044287 to WS). We would also like to thank the

members of the NPF DBS Working Group for their contributions

and the Wadsworth Center Advanced Microscopy and Image

Analysis Core facility for their contributions. This study has

Institutional Review Board (UF-IRB) approval for use and the

IRB# is: 130-2008.

References

Aalten CM, Samson MM, Jansen PA (2006) Diagnostic errors;

the need to have autopsies. Neth J Med 64(6):186–190

Alafuzoff I, Winblad B (1993) How to run a brain bank: poten-

tials and pitfalls in the use of human post-mortem brain

material in research. J Neural Transm Suppl 39:235–243

Beach TG, Adler CH, Lue L, Sue LI, Bachalakuri J, Henry-

Watson J, Sasse J, Boyer S, Shirohi S, Brooks R, Esc-

hbacher J, White CL 3rd, Akiyama H, Caviness J, Shill

HA, Connor DJ, Sabbagh MN, Walker DG (2009) Unified

staging system for Lewy body disorders: correlation with

nigrostriatal degeneration, cognitive impairment and

motor dysfunction. Acta Neuropathol 117:613–634

Braak E, Sandmann-Keil D, Rub U, Gai WP, de Vos RA, Steur

EN, Arai K, Braak H (2001) Alpha-synuclein immuno-

positive Parkinson’s disease-related inclusion bodies in

lower brain stem nuclei. Acta Neuropathol 101(3):195–201

Braak H, Del Tredici K, Rub U, de Vos RA, Jansen Steur EN,

Braak E (2003) Staging of brain pathology related to

sporadic Parkinson’s disease. Neurobiol Aging 24:197–

211

Braak H, Ghebremedhin E, Rub U, Bratzke H, Del Tredici K

(2004) Stages in the development of Parkinson’s disease-

related pathology. Cell Tissue Res 318(1):121–134

Chu Y, Dodiya H, Aebischer P, Olanow CW, Kordower JH

(2009) Alterations in lysosomal and proteasomal markers

in Parkinson’s disease: relationship to alpha-synuclein

inclusions. Neurobiol Dis 35:385–398

Davie CA (2008) A review of Parkinson’s disease. Br Med

Bull 86:109–127

Delong MR, Georgopoulos AP, Crutcher MD, Mitchell SJ,

Richardson RT, Alexander GE (1984) Functional orga-

nization of the basal ganglia: contributions of single-cell

recording studies. Ciba Found Symp 107:64–82

Folstein MF, Folstein SE, McHugh PR (1975) Mini-mental

state. A practical method for grading the cognitive state of

patients for the clinician. J Psychiatr Res 12:189–198

Haberler C, Alesch F, Mazal PR, Pilz P, Jellinger K, Pinter

MM, Hainfellner JA, Budka H (2000) No tissue damage

230 Cell Tissue Bank (2011) 12:219–231

123

by chronic deep brain stimulation in Parkinson’s disease.

Ann Neurol 48:372–376

Haroutunian V, Pickett J (2007) Autism brain tissue banking.

Brain Pathol 17:412–421

Henderson JM, Pell M, O’Sullivan DJ, McCusker EA, Fung

VS, Hedges P, Halliday GM (2002) Postmortem analysis

of bilateral subthalamic electrode implants in Parkinson’s

disease. Mov Disord 17:133–137

Hughes AJ, Daniel SE, Kilford L, Lees AJ (1992) Accuracy of

clinical diagnosis of idiopathic Parkinson’s disease: a

clinico-pathological study of 100 cases. J Neurol Neuro-

surg Psychiatry 55:181–184

McKeith IG et al (1996) Consensus guidelines for the clinical

and pathologic diagnosis of dementia with Lewy bodies

(DLB): report of the consortium on DLB international

workshop. Neurology 47:1113–1124

Nielsen MS, Bjarkam CR, Sorensen JC, Bojsen-Moller M,

Sunde NA, Ostergaard K (2007) Chronic subthalamic

high-frequency deep brain stimulation in Parkinson’s

disease—a histopathological study. Eur J Neurol 14:

132–138

Perry G, Friedman R, Shaw G, Chau V (1987) Ubiquitin is

detected in neurofibrillary tangles and senile plaque neu-

rites of Alzheimer disease brains. Proc Natl Acad Sci

USA 84:3033–3036

Roulson J, Benbow EW, Hasleton PS (2005) Discrepancies

between clinical and autopsy diagnosis and the value of

post mortem histology; a meta-analysis and review. His-

topathology 47(6):551–559

Schiesling C, Kieper N, Seidel K, Kruger R (2008) Review:

familial Parkinson’s disease—genetics, clinical phenotype

and neuropathology in relation to the common sporadic

form of the disease. Neuropathol Appl Neurobiol 34:

255–271

Schmitt A et al (2007) How a neuropsychiatric brain bank

should be run: a consensus paper of Brainnet Europe II. J

Neural Transm 114:527–537

Schubert W, Prior R, Weidemann A, Dircksen H, Multhaup G,

Masters CL, Beyreuther K (1991) Localization of Alz-

heimer beta A4 amyloid precursor protein at central and

peripheral synaptic sites. Brain Res 563:184–194

Spillantini MG, Schmidt ML, Lee VM, Trojanowski JQ, Jakes

R, Goedert M (1997) Alpha-synuclein in Lewy bodies.

Nature 388:839–840

Sun DA, Yu H, Spooner J, Tatsas AD, Davis T, Abel TW, Kao

C, Konrad PE (2008) Postmortem analysis following

71 months of deep brain stimulation of the subthalamic

nucleus for Parkinson disease. J Neurosurg 109:325–329

Vonsattel JP, Amaya Mdel P, Cortes EP, Mancevska K, Keller

CE (2008) Twenty-first century brain banking: practical

prerequisites and lessons from the past: the experience of

New York Brain Bank, Taub Institute, Columbia Uni-

versity. Cell Tissue Bank 9:247–258

Waldvogel HJ, Curtis MA, Baer K, Rees MI, Faull RL (2006)

Immunohistochemical staining of post-mortem adult

human brain sections. Nat Protoc 1:2719–2732

Waldvogel HJ, Bullock JY, Synek BJ, Curtis MA, van Roon-

Mom WM, Faull RL (2008) The collection and processing

of human brain tissue for research. Cell Tissue Bank

9:169–179

Weaver F, Follett K, Hur K, Ippolito D, Stern M (2005) Deep

brain stimulation in Parkinson disease: a metaanalysis of

patient outcomes. J Neurosurg 103:956–967

Weiner WJ (2008) There is no Parkinson disease. Arch Neurol

65:705–708

Zijlmans JC, Daniel SE, Hughes AJ, Revesz T, Lees AJ (2004)

Clinicopathological investigation of vascular parkinson-

ism, including clinical criteria for diagnosis. Mov Disord

19:630–640

Cell Tissue Bank (2011) 12:219–231 231

123