Transcript Neuroimaging: from image to Inference

Transcranial Direct Current Stimulation
Chris Rorden
www.mricro.com
www.cabiatl.com
Method
Designs
Safety
1
tDCS vs TMS
Transcranial magnetic stimulation
– Relatively expensive (~$50,000).
– Moderate sized effects (e.g. mild speech arrest).
– Safe, but there are reports of inducing seizures
when high amplitude and frequency are combined.
– Causes resting neurons to fire.
Very brief pulse stops interrupts processing for ~30ms,
can be used repetitively.
Depending on frequency, sustained TMS
can induce excitability reduction (long-term
depression) or enhancements (long-term
potentiation) that can persist for hours or days.
2
tDCS vs TMS
Transcranial direct current stimulation
– Very inexpensive (~$250 for iontophoresis unit).
– Believed to be exceptionally safe.
– Does not cause resting neurons to fire (Purpura and
McMurtry, 1965; Terzuolo and Bullock,1956).
– Believed to modulate the firing rate of active
neurons.
Depending on polarity, tDCS can induce cortical
excitability reduction or enhancement can
persists for hours.
3
tDCS vs TENS
Transcutaneous Electrical Nerve Stimulation
systems are used to treat pain.
TENS pulsed 2-160Hz, 5-80 mA.
At slow frequency and
high amplitude TENS
induces muscle
contraction.
In contrast, tDCS uses
constant 1-2mA.
4
History of tDCS
’50-60s exposed cortex
of animals: diminish
(cathodal) or enhance
(anodal) cortical
excitability and activity.
Lippold & Redfearn
(1964) report scalp
tDCS relieves
depression in humans.
+Anodal
-Cathodal
Bindman et al. (1964)
5
Why a revival?
 New methods provide converging
support
– Confirmed using consistent behavioral
measures : corticospinal excitability,
measured with TMS; TENS (Nitsche
2000; Ardolino 2005).
– Confirmed using imaging: e.g. one
sees less task related activation
following cathodal stimulation
(Baudewig et al., 2001)
– Mechanism: change in membrane
potential, NMDA receptor efficacy for
longer duration effects (Nitsche, 2004).
Ardolino (2005)
Baseline
0min
60min
Baseline
After -tDCS
Baudewig et al. (2003) 6
Effects persist
Effects of tDCS persist after stimulation ends.
Longer stimulation, slower return to baseline.
Duration
5min
7min
9min
Nitsche et al. (2003)
7
Typical design
 Convention is to conduct behavioral task during and/or
immediately after stimulation.
 E.G. Dockery reports that prefrontal tDCS polarity
influences learning of Tower of London task – with effects
seen 6-12 months later.
Dockery et al. (2009)
8
Scientific concerns
Current is very small (1-2mA)
– So tiny, many doubt neural effects are real.
Behavioral effects typically very small
– ‘File drawer problem’ most null results not counted.
– Electrode placement crucial.
– Controlling for experimenter demand crucial.
9
Where to stimulate
Null result if stimulated region not
involved with task.
Our Visor neuronavigation system
allows you to identify regions
based on fMRI or MRI data.
10
Where to stimulate
Sadleir et al. (2010)
suggest effects will be
diffuse.
Datta (2009) suggest
high density electrode
placement could provide
more specificity.
11
Where to stimulate
Stimulation region not well
focused.
Must create electrical circuit: both
anode and cathode.
– If both on scalp, are effects due to
facilitation or inhibition?
– If one electrode on shoulder/limbs
(Baker, 2010), perhaps spinal
influence.
– One option is large, diffuse electrode
over mastoid (Elmer, 2009).
+
_
_
+
12
Clever Hans (1907)
 Can a horse perform arithmetic?
 Actually, animal was responding to body language of
human observers.
 tDCS effects are small.
 Small effects vulnerable
to experimenter demand.
 Double-blind rare but
crucial.
 I personally remain sceptical
of many findings: we need
scientific rigor.
13
Our tDCS units
Our tDCS units designed
for iontophoresis.
Can deliver up to 4mA:
contemporary studies do
not exceed 2mA.
Disposable sponge
electrodes.
Optional USB system can
ensure double blind
research.
14
Theoretical safety concerns
 Potential side effects with tDCS
 electrode-tissue interface could lead to skin irritation
and damage.
 Stimulations could lead to excitotoxic firing rates.
 Tissue damage due to heating.
 Rat studies suggest injury only when
current density is several orders of
magnitude beyond those used in humans
(Liebetanz et al. 2009).
 Standard doses in humans does not
appear to alter serum neuron specific
enolase (NSE), a sensitive marker of
neuronal damage (Nitsche et al, 2003).
 Datta (2009) heating in humans is
negligible.
15
Practical safety concerns
Subtle but common side effects
Nitsche et at. (2003) reports that in more than 500
participants the only side effects are initial scalp tingling
or sensation of a light flash.
Some studies suggest that higher current densities can
lead to skin irritation.
If cognitive effects are prolonged, perhaps we should
warn participants about driving or other hazardous tasks
after a treatment session.
• Koenigs (2009) note one neurologically healthy participant
reported a couple hours dysphoria following cathodal tDCS.
16