Copyrigh
Nerv d
nerv
Chris
Introduction
Conventionally peripheral nerve disorders are classified
as due primarily to axonal degeneration or to demyelina-
tion. Electrophysiological studies are important in the
distinction of these main categories though overlapping
findings
diagnos
criteria
Though
acquired
between
compart
the clos
clinical
secondary event in peripheral neuropathy characterized
by axonal loss [6] and axonal loss is a major cause of
disability in patients with demyelinating neuropathy [7].
Furthermore nerve conduction velocity varies consider-
ably in neuropathies due to myelin involvement both in
patients with similar mutations [8,9] or with variable
Departmen
and the Ins
University o
Correspond
of Clinical
Blegdamsv
Tel: +45 3
Current O
ing
t ha
log
die
f p
By measuring responses in nerve that are related to nodal function (strength–duration
time constant, rheobase and recovery cycle) and internodal function (threshold
electrotonus, current–threshold (I/V) relationship) it is possible to assess the function of
transient and persistent Naþ, fast and slow Kþ and HCN inward rectifying channels as
well as ion pumps. This has allowed insight into normal axon physiology and normal
fluctuations of electrolyte concentrations. Studies of different metabolic neuropathies
have assessed the influence of uremia, diabetes and ischemia, and the use of these
methods in toxic neuropathies has allowed pinpointing damaging factors. Various
mutations in ion channels associated with central nervous system disorders have been
shown to have counterparts in the peripheral nervous system, in some instances without
peripheral nervous system symptoms. Both hereditary and acquired demyelinating
neuropathies have been studied and the effects on nerve pathophysiology have been
compared with degeneration and regeneration of axons.
Summary
Excitability testing holds promise for further understanding of peripheral nerve
pathophysiology but is as yet not universally available. Interpretation may be challenging
as changes in parameters may have different explanations, and modeling has been
helpful in the use of the methods in clinical neurophysiology.
Keywords
axonal degeneration, compound muscle action potential, compound sensory action
potential, demyelination, ion channel dysfunction, membrane depolarization, nerve
conduction studies, nerve excitability
Curr Opin Neurol 22:460–466
 2009 Wolters Kluwer Health | Lippincott Williams & Wilkins
1350-7540
1350-754pathological conditions that affect these two ancemay present difficulties, and to increase the
tic certainty nerve conduction study (NCS)
of demyelination have been proposed [1–3].
the therapeutic consequences of inferring
demyelination are important, a sharp separation
mutations within the same molecule [4], and convention-
al NCS may be of limited help to probe underlying
complex pathophysiology. NCS in renal failure [10,11]
could show changes consistent with axonal loss, which,
however, were reversible, indicating functional disturb-
s that may not be studied by conventional NCS. It ise conduction and excitability stu
e disorders
tian Krarup and Mihai Moldovan
t of Clinical Neurophysiology, Rigshospitalet
titute of Neuroscience and Pharmacology,
f Copenhagen, Copenhagen, Denmark
ence to Christian Krarup, MD, Department
Neurophysiology 3063, Rigshospitalet,
ej 9, DK-2100 Copenhagen, Denmark
545 3060; e-mail: ckrarup@rh.dk
pinion in Neurology 2009, 22:460–466
Purpose of review
The review is aimed at provid
peripheral nerve disorders. I
peripheral nerve pathophysio
limited. Nerve excitability stu
important role in the study o
Recent findingst © Lippincott Williams & Wilkins. Unauthorized re
ments [4] is an oversimplification considering
e interaction between glia and axon [5], and in
conditions, abnormalities in myelin occur as a
the pur
nerve e
siology
0  2009 Wolters Kluwer Health | Lippincott Williams & Wilkinsies in peripheral
information about the role of nerve excitability studies in
s been known for many years that the insight into
y provided by conventional nerve conduction studies is
s are relatively novel but are acquiring an increasingly
eripheral nerves.production of this article is prohibited.
pose in this review to discuss the potential of
xcitability testing in the exploration of pathophy-
of peripheral nerve disorders.
DOI:10.1097/WCO.0b013e3283304c9d

Copyrig
Nerve
excitab
Supram
conduct
(CMAP
measure
large m
about f
fibers i
refracto
function
that rel
during
limitatio
aspects
of patho
thresho
amplitu
valuable
using th
myelina
conduct
arization
depolar
Kþ chan
observa
biophys
ipheral
tracking
elicit a
The cl
thresho
of the Q
physiolo
set-up i
stimulat
whereas
placed a
is carrie
controll
Owing
100ms)
able (typ
is carri
Stimula
lator. It
excitabi
meters
respons
where t
target (
adjust t
Multipl
[24] inc
ste
ier
ion
ery
to
y
erv
nel
rd r
ess
na
tab
al
s i
in
in
ar
ral
qu
e is
cr
en
l d
est
ien
od
e c
be
res
ar-
en
red
nt
ent
re
oto
17
ve
uct
elocity on the control side (Fig. 1a).
n
rot
rot
esi
ar t
ear
ion
rm
e wconduction studies and nerve
ility
aximal electrical stimulation and measurement of
ion velocity and amplitudes of compound motor
) and sensory (SNAP) responses provide
s of the number and conduction velocities of
yelinated fibers. Supplementary information
unctional properties of still conducting nerve
n peripheral neuropathy may be obtained by
ry period measurements that address ion-channel
[12,13]. These measurements used stimulation
ied on the stimulus being supramaximal even
periods of lowered excitability. In spite of this
n [14], studies that address these fundamental
of nerve physiology may improve understanding
physiology. Studies that used measurements of
ld in humans to obtain a response of a given
de (for example a single motor unit) provided a
alternative [15,16]. Experimental observations
reshold measurements to assess excitability of
ted nerve fibers indicated that activity dependent
ion block in demyelination was due to hyperpol-
[17], and showed that accommodation during
ization was due to activation of different types of
nels in rats and human nerves [18–20]. These
tions indicated that ion-channel function and
ical properties of myelinated fibers within a per-
nerve in vivo could be accessed by automatic
of threshold, defined as the current required to
submaximal CMAP and SNAP.
inical availability of excitability measures by
ld tracking was made possible by the development
TRAC software (for reviews of methodology and
gical and clinical applications see [21–23]). The
s similar to conventional NCS, placing a surface-
ing electrode over the nerve being studied,
the reference stimulating surface electrode is
t some distance from the nerve [24]. Stimulation
d out by a linear constant current stimulus isolator
ed by the software via an analog–digital convertor.
to the use of long stimulation pulses (more than
, the stimulation electrodes should be nonpolariz-
icallyAg/AgCl).Recording of theCMAPorSNAP
ed out using conventional surface electrodes.
tion is carried out via direct current linear stimu-
should be emphasized that the study of nerve
lity measures physiological and biophysical para-
directly under the stimulating electrode. The
e is digitized and recorded in a personal computer
he size of the recorded response in relation to a set
conventionally set to 40% of maximal) serves to
persi
Ranv
durat
recov
refrac
ity). B
the n
chan
inwa
is ass
Axo
exci
Axon
cesse
myel
rema
50 ye
in pa
conse
Littl
the in
comp
noda
sugg
defic
paran
In th
only
we p
35-ye
accid
repai
patie
conv
plete
of m
only
howe
cond
the v
Whe
elect
elect
the l
simil
31/2 y
funct
abno
phasht © Lippincott Williams & Wilkins. Unauthorized r
he stimulus strength such that the target is met.
emeasures of excitability in theTRONDprotocol
lude assessment of ion channels (transient and
by an
(Fig. 1a
axons (tested at the wrist, threshold changes during
onus were markedly abnormal, whereas threshold
onus of the same axons at the elbow (proximal to
on) was normal (Fig. 1a). These changes were
o a previous recording in the same patient at only
s after the lesion [29], strongly indicating that the
of regenerated internodes remains persistently
al. In regenerated axons the deviations during S1
ere increased, however, this was compensated fornt Naþ channels, slow Kþ channels) at nodes of
by computing stimulus response curves, strength
time constant (chronaxie), rheobase, and the
cycle after passage of an action potential (relative
ry periods, supernormality and late subexcitabil-
applying long polarizing currents (electrotonus) to
e the influence of voltage on voltage gated ion
s under the myelin (fast and slow Kþ channels,
ectifying current (Ih)mediatedbyHCNchannels)
ed.
l properties that can be explored by
ility testing
degeneration and regeneration are common pro-
n peripheral nerve disorders. As a consequence of
remodeling, the internodal length is known to
persistently short [25]. As studies more than
s ago showed that conduction velocity recovers
lel with the fiber diameter [26], the functional
ences of the short internodes remained obscure.
known about how peripheral neurons cope with
eased number of nodes except that there may be a
satory increase in Naþ channels so that the inter-
ensity is restored [27]. Nevertheless, it was
ed that regenerated axons might be functionally
t as the access to the Kþ channel under the
al myelin may be increased [28].
linical setting, the function of the internode can
explored by excitability studies. For illustration
ent the conduction and excitability studies in a
old healthy man who 13 years previously had an
tal section of the left median nerve at the elbow
by suture 2 days after the accident (Fig. 1). This
was particularly suitable for excitability testing;
ional motor conduction studies revealed a com-
covery of CMAP amplitude (even if the number
r units counted by the incremental method were
versus 115 on the healthy side). The CSAP was
r of reduced amplitude. Both motor and sensory
ion velocities were reduced, approaching 75% of
Peripheral nerve disorders Krarup and Moldovan 461eproduction of this article is prohibited.
increased accommodation during S2 phase
). These changes were similar in the sensory
Fig. 1c). Threshold electrotonus alone cannot