Sidtis

Dysarthria in Parkinson’s Disease and AtaxiaASHA, 2011

John J. Sidtis

Director,Brain and Behavior LaboratoryThe Nathan Kline Institute for Psychiatric Research

Professor,NYU School of Medicine

[email protected]

Dysarthria in Parkinson’s Disease and Ataxia

The ability to speak clearly involves a complex brain system that is not fully understood.

Parkinson's disease and the ataxias interfere with this ability.


The ability to speak clearly involves a complex brain system that is not fully understood.

Parkinson's Disease and the ataxiasinterfere with this ability.

Our Laboratory explores this speech control system with a combination of studies on the effects of speech tasks and functional brain imaging with PET.

We have been particularly interested in the role ofcortical-subcortical interactions during speech.


How do we begin to understand this complex brain system?

Clinically informed behavioral studies:perceptual ratingsintelligibilityacoustic analysis




Clinically informed functional imaging:disease relevant tasksperformance-based analysis




Clinically informed functional imaging:clinically relevant tasksperformance-based analysis

Take advantage of advances in neurobiology,genetics, and neuroscience.

Taking advantage of advances in neurobiology,genetics, and neuroscience

SCA1 mixed ataxia chromosome 6 CAG repeat

SCA5 “pure” ataxia chromosome 11 spectrin mutation SCA6

“pure” ataxia chromosome 19 CAG repeat

Normal cerebellum

The genetics of hereditary, spino-cerebellar ataxia (SCA)

The genetics of SCA

Sidtis JJ, Ahn JS, Gomez C, Sidtis D. Speech characteristic associated with three genotypes of ataxia. Journal of Communication Disorders 44: 478-492, 2011.

Using perceptual ratings, are there genotypic differencesin the speech produced by subjects with SCA1, SCA5, and SCA6?

Speech samples were ratedon the following primarydimensions: articulation,rate, rhythm, and prosody.

When a primary dimension was ratedabnormal, secondary dimensionswere rated as well. These includedarticulatory and voice dimensions.

There were significanttask differences.

Diadochokinesis producedthe most consistent ratingsacross genotypes.

There were also significantdimension differences.

Articulation was the mostImpaired primary dimension.

Primary Dimensions


The genetics of SCA

Secondary Dimensions


The genetics of SCA

Picture description was most effective in eliciting abnormal secondarydimensions.

Voice dimensions showedgreater differences across SCA types than articulation.

The genetics of SCA

Articulation is the most impaired primary dimension across SCA types.

This corresponds to a “core dimension” described by Zeplin and Kent.

Syllable repetition was the most effective task for the primary features of ataxic dysarthria (Kent et al, 1997; Ziegler, 2002).

Voice subgroups have been reported in the ataxic dysarthrias(Grémy et al. 1967; Joanette & Dudley, 1980).

Spontaneous speech (picture description) is more effective than repetitionin characterizing problems in normal communicative settings (Kempler & Van Lancker, 2002; D. Sidtis et al., 2010).

Zeplin & Kent (1996). In Robin, Yorkston, Beukelman, (Eds.), Disorders of motor speech. Baltimore: Brookes.

Kent et al. (1997). Folia Phoniatrica et Logopaedica, 49, 63-82.

Ziegler (2002). Brain and Language, 80, 556-575.

Grémy et al. (1967). Revue Neurologique (Paris), 116(5), 401-426.

Joanette & Dudley (1980). Brain and Language, 10, 39–50.

Kempler & Van Lancker (2002). Brain and Language, 80, 449-464.

D. Sidtis et al., (2010) JSLHR, 53, 1167-1177.

http://content.karger.com/ProdukteDB/produkte.asp?Aktion=Ausgabe&ProduktNr=224177&Ausgabe=225815�

The ability to map brain activity with functional imaging.

Taking advantage of advances in neurobiology,genetics, and neuroscience.

fMRI measures the signals produced by nuclear particles as they respond to magnetic pulses. The BOLD signal is based on the differences in signal produced by oxygenated and de-oxygenated blood.

PET measuresthe concentrationof isotope pairedwith a biologicallyactive substance(e.g., water, drug).

Currently, the most common approach is to use fMRI to Identify areas of “activation” in the BOLD signal.

Activation is a significant signal increase when two or moreconditions are contrasted.

We have argued that alternative approaches may be moresuitable for a systems approach to studying the neurological systems for speech motor control.

The ability to map brain activity with functional imaging

is an alternative to activation approaches.

does not require contrasting two or more conditions.

does not remove brain areas that do not activate from further analysis.

simply seeks to determine if there is a linear combination of brain areas in which activity predicts performance on the task done during scanning (e.g., repetition of /pa-ta-ka/).

Performance-based analysis:

The ability to map brain activity with functional imaging

Although speech and language are strongly lateralized, functional images typically show bilateral signals.

Performance-based analysis using multiple linear regressionallows us to predict syllable rate during the repetition of /pa-ta-ka/.

Although the images are bilaterally symmetrical, the performance-based analysis identifies an inverse relationship between the leftinferior frontal region (Broca’s area) and the right caudate nucleus.

Sidtis JJ, Strother SC, Rottenberg DA. Predicting performance from functional imaging data: Methods matter. NeuroImage 20(2): 615-624, 2003.

Sidtis JJ. Some problems for representations of brain organization based on activation. Brain and Language 102(2): 130-140, 2007.

Mapping brain activity with performance-based analysis.

/pa,ta,ka/ repetition

INFERIOR FRONTAL REGION

LEFT RIGHT1.0

1.2

1.4

1.6

vnrC

BF

ns

CAUDATE NUCLEUS

LEFT RIGHT1.2

1.4

1.6

1.8

vnrC

BF

nsPerformanceBasedAnalysis

SPEECH RATE PREDICTORS:NORMALS

-5.0

-2.5

0.0

2.5

5.0

RIGHT CAUDATE

LEFT IFG

Reg

ress

ion

Wei

ght

Predicting normal rate



/pa,ta,ka/ repetition

PerformanceBasedAnalysis


LEFT RIGHT0.00

0.02

0.04

0.06

0.08

0.10

vnrC

BF (s

peec

h-re

st)

p = 0.002

CAUDATE NUCLEUS

-0.05

-0.03

-0.01

0.01

0.03

0.05

LEFT RIGHT

ns

vnrC

BF (s

peec

h-re

st)

No Solution



Performance-based connectivity analysis.

The original performance-based analysis establishes primaryrelationships between brain regions and performance (e.g., rate).

This was expanded to explore a larger system of functional connectivity using partial correlation.

This expanded analysis determines the relationship between eachprimary predictor region and other brain regions, controlling for the correlation between the primary predictor and its contralateral homologous region (e.g., relationships with the left inferior frontal region are determined controlling for the influence of the rightinferior frontal region).

Sidtis JJ. Performance-Based Connectivity Analysis: A Path to Convergence with Clinical Studies. NeuroImage 2011, doi:10.1016/j.neuroimage.2011.09.037.


L RR L

PRIMARY RELATIONSHIPSWITH RATE

PRIMARY AND SECONDARYRELATIONSHIPS WITH RATE

GREEN = POSITIVE ASSOCIATION RED = NEGATIVE ASSOCIATION

Increased speech rate is predicted by a linear combination of increasedcontribution by the left inferior frontal region and decrease by the right caudate. Secondary relationships present a pattern consistent with clinical evidence.

Sidtis JJ. Performance-Based Connectivity Analysis: A Path to Convergence with Clinical Studies. NeuroImage (2011), doi:10.1016/j.neuroimage.2011.09.037.


Performance-based analysis captures:

left hemisphere motor control;

thalamic involvement in speech rate;

auditory suppression during speech.

Sidtis JJ. Performance-Based Connectivity Analysis: A Path to Convergence with Clinical Studies. NeuroImage (2011), doi:10.1016/j.neuroimage.2011.09.037.


Performance-based analysis captures:

left hemisphere motor control;

thalamic involvement in speech rate;

auditory suppression during speech.

Important to note:

both positive and negative relationships are significant;

regions that do not activate (e.g., the caudate) play a significant rolein predicting speech rate.

Mapping the ataxic brain during speech.

How does brain function in ataxia compare to normal brainfunction during speech?

Ataxic mean flow values

PerformanceBasedAnalysis

SPEECH RATE PREDICTORS:ATAXIA

-5.0

-2.5

0.0

2.5

5.0

RIGHT CAUDATE

LEFT IFG

LEFT TRANSVERSE TEMPORAL

RIGHT INFERIORCEREBELLUM

Brain Regions

Reg

ress

ion

Wei

ght


LEFT RIGHT1.0

1.2

1.4

1.6

vnrC

BF ns


Predicting ataxic rate

Sidtis JJ, Gomez C, Naoum A, Strother SC, Rottenberg DA. Mapping cerebral blood flow during speech production in hereditary ataxia. NeuroImage 31: 246-254, 2006.



In a combined group of SCAs, performance-based analysis:

replicated the normal inverse cortical-subcortical relationshipwith speech rate;

identified a role for the right cerebellum;

identified a role for an auditory area in the left temporal lobe.



In a combined group of SCAs, performance-based analysis:

replicated the normal inverse cortical-subcortical relationshipwith speech rate;

identified a role for the right cerebellum;

identified a role for an auditory area in the left temporal lobe.

Can genotypic differences be identified?

SCA 5SCA 1 SCA 6

INFERIORFRONTALREGION

CAUDATENUCLEUS

Genotypic Differences in Functional Connectivity of Predictor Regions

GREEN = POSITIVE ASSOCIATION RED = NEGATIVE ASSOCIATION

Performance-based connectivity analysis identified genotypic differences among the ataxias

SCA5, with least affected speech, has generalized positiverelationships with speech rate associated with the left inferiorfrontal region and negative relationships associated with the right caudate nucleus;

SCA6, with the greatest speech impairment, had no identifiable secondary associations with the right caudate;

The two “pure” ataxias did not have similar functional connectivity;

The two trinucleotide repeat ataxias appeared more similar toeach other than to the spectrin based pathology;

Molecular pathophysiology may be more important than gross pathology

These results, together with the results of listening studies, suggest that neurobiological advances will contribute to a better understanding of the complex systems underlying articulatory and vocal control.


The ability to study the effects of stimulating basal gangliastructures in the treatment of Parkinson’s Disease (Deep brain stimulation or DBS)

High frequency, repetitive electricalstimulation of certain brain areasimproves some of the symptoms ofmovement disorders, includingParkinson’s Disease (PD).

Originally, DBS was conceived of asa reversible lesion.

DBS is now seen as changing firingpatterns of nuclei in the basal ganglia.

Side effects are produced by stimulationof adjacent pyramidal tracts.


MRI showing the placement ofbilateral stimulating electrodesin the STN (Alterman).

The subthalamic nucleus (STN), part ofthe basal ganglia, is the most commontarget in PD.

STN-DBS is effective at controlling tremorand rigidity.

Continuous STN-DBS reduces the on-offeffects of PD medication.

STN-DBS also allows a reduction in PDmedication.

Our studies are investigating how DBS affects speech and brain activity as a function of speaking task.


Sidtis D, RogersT, Godier V, Tagliati M, Sidtis JJ. Voice and fluency changes as a function of speech task and deep brain stimulation. JSLHR 53(5): 1167-77, 2010.

Sidtis, D, Cameron K, Bonura L, Sidtis JJ. Speech intelligibility by listening in Parkinson speech with and without deep brain stimulation: Task effects. Journal of Neurolinguistics, In press.

Dysfluencies are greater duringconversation compared to conversation-repetition.

There is a tendency for greaterdysfluency with DBS on duringconversation.




HNR improved with DBSduring conversation.

The DBS effect on HNRis comparable to the effectof repetition.




Voice abnormalities are reducedwith DBS during conversation.

The effect of repetition on voicequality is greater than the effectsof DBS.


During conversational speech, DBS improves voice but tends to reducearticulatory performance.

The effects of DBS are comparable to, or slightly less effective than the effects of repetition.

Providing an external model (repetition) appears to reduce the burdenon the basal ganglia during conversational speech.

Can we learn more about these effects using functional imaging?

Performance-based analysisidentified an inverse cortical-subcorticalrelationship with speech rate.



Performance-based analysisidentified an inverse cortical-subcorticalrelationship with speech rate.

In PD, these same brain regionspredicted speech rate, but theweighting for the inferior frontalregion is inverted.

For the PD analyses, data forthree repetition tasks were used.


With DBS, the inverse cortical-subcorticalrelationship with rate observed innormal and ataxic speakers is restored,but is now bilateral.

LEFT RIGHT


With DBS off, the inferior frontalregion is lost and the right caudateis inverted. Short term cessationof DBS does not reflect PD, andmay reflect a temporally unstable state.

LEFT RIGHT

Take advantage of advances in neurobiology,genetics, and neuroscience.The study of the effects of DBS on different speech tasks promisesto provide insights into basal ganglia function and the ways in which these structures interact with the rest of the brain.

As with molecular genetics in the spino-cerebellar ataxias, the DBSmanipulation is biologically complex and not fully understood.

Using DBS to better understand motor speech control, we will alsohave to better understand the neurobiology of DBS.

For example, in studying cerebral blood flow during speech in DBS,we discovered a previously unknown effect: a significant increase inglobal blood flow (Sidtis et al., 2011).

In spite of the unknowns, the advances in neuroscience will lead to a more sophisticated understanding of motor speech control.

Sidtis JJ et al. Therapeutic high frequency stimulation of the subthalamic nucleus in Parkinson’s Disease produces global increases cerebral blood flow. J Cerebral Blood Flow and Metabolism, doi:10.1038/jcbfm.2011.135

When advances in neurobiology, genetics, and neuroscience areincorporated into research on the dysarthrias, do we replicatewhat we already know (or think we know)?

Can we use these advances to learn more about the neurology ofspeech motor control?

Advances in neurobiology, genetics, and neuroscience should beviewed as providing potential tools to pursue questions raised by clinical experience, and never as replacements for that experience.


Regards from Diana Sidtis,Associate Director, whowas unable to attend afterbreaking her arm at theGerman Aphasia Conference.

AcknowledgmentsNIH R01 NS37211

Danielle AcernoNicole AcquafreddaJi Sook AhnAmy AlkenBettina ArmstrongMaria AndersonKatie BarnesSara-Jean BartkyPelinsu BelutKathy BihLisa BonuraKelly BridgesTarun CalidasKrista Cameron*Claudia CerulliLiz Dovlatyan*Violette Godier*

Cerebellar Ataxia

Christopher Gomez, University of ChicagoDavid Rottenberg, University of MinnesotaStephen Strother, University of Toronto

Parkinson’s Disease

Michele Tagliati, Ron Alterman, Cathy ChoFiona Gupta, Tyler ChungMount Sinai Medical SchoolDavid Eidelberg, Vijay DhawanFeinstein Institute,North Shore University Hospital

Image Processing

Babak Ardekani, Ali Tabash, Khadija FigarskyNathan Kline Institute for Psychiatric Research

Brain and Behavior Laboratory

Diana Sidtis, Associate DirectorStudents:

Ariana GluckAly HoffmanHae Su KangDora KatsnelsonKathy KougentakisNina LisitsaJennifer MelgarejoRaz MeltzerJeon MoonHolly PralgeverTiffany RogersRobert SidtisLiz Sweeting*Elana WintersTheresa YangLisa YeungJudy YuenVictoria Zeldin

* Served as Laboratory Manager