INVESTIGATIONS
Clinical neurophysiology
Robin P Kennett
Sidra Aurangzeb
Abstract
Clinical neurophysiology uses measurement of electrical activity in
the nervous system to diagnose and monitor disease. Electro-
encephalography (EEG) and nerve conduction studies (NCS) with elec-
tromyography (EMG) are the tests most used in general medical pa-
tients. Specialized investigations for neurological diseases and
operative or long-term monitoring are available, but are beyond the
scope of this article.
The EEG is a simple, non-invasive test of brain function used
mostly to diagnose and classify epilepsy. Bilateral three per second
spike and slow-wave complexes are seen in primary generalized ep-
ilepsy, and localized spike discharges in focal seizures. Prolonged
recordings of seizures, allowing a clinical and electrographic correla-
tion, are sometimes needed for precise diagnosis. Characteristic
EEG waveforms (slow waves, triphasic or repetitive complexes) can
help in the management and prognosis of other disorders such as
encephalitis and encephalopathy. Nerve conduction and needle
EMG studies investigate peripheral neuromuscular disorders, espe-
cially to confirm and assess the severity of peripheral nerve disease.
Clinical neurophysiologists are often asked to investigate patients
with undiagnosed neurological symptoms, and these tests can pro-
vide objective information as an extension of the clinical examination.
Measurement of sensory nerve action potentials helps to locate le-
sions in sensory pathways. A combination of motor nerve conduction
and needle EMG studies is particularly useful when investigating pa-
tients with weakness.
Keywords Electroencephalogram; electromyography; epilepsy;
motor neuron disease; nerve conduction; peripheral neuropathy
Electroencephalography (EEG)
Electrodes attached to the scalp with conductive paste record
electrical potentials originating in the cerebral cortex. The signals
vary in frequency and amplitude in different regions; the most
consistently observed activity is at 8e12 cycles/second over the
occipital region with the eyes closed (a-rhythm). The mechanism
of EEG rhythms is not fully understood but probably involves
diffusely projecting neuronal circuits between the thalamus and
Robin P Kennett BSc MD FRCP(Lon) FRCP(Ed) is Consultant in Clinical
Neurophysiology at the John Radcliffe Hospital, Oxford, UK. His
interests include diagnosis of neuromuscular disorders and
indications for surgical treatment of epilepsy. Competing interests:
none declared.
Sidra Aurangzeb MB BS (Pakistan) MRCP (UK) FCPS Neurology (Pakistan) is
Specialist Registrar in Neurology and Clinical Neurophysiology at the
John Radcliffe Hospital, Oxford, UK. Competing interests: none
declared.
MEDICINE 44:8
Key points
C Scalp electroencephalography (EEG) is non-invasive, available
and can be specific in epilepsy, but has limited sensitivity. A
normal standard EEG does not exclude the diagnosis of
epilepsy
C Long-term EEG and video-telemetry (simultaneous video and
EEG recordings) can be diagnostic in patients with unexplained
recurrent symptoms, including non-epileptic attack disorder
C EEG is important in the management of status epilepticus,
especially non-convulsive status, and in the assessment of
patients with hypoxic-ischaemic encephalopathy
C Nerve conduction studies and electromyography provide
objective evidence of peripheral nerve disease and aid man-
agement of the common entrapment neuropathies
C Nerve conduction studies differentiate demyelinating from
axonal generalized peripheral neuropathy. In acute neuropa-
thies such as acute inflammatory demyelinating poly-
neuropathy (GuillaneBarre syndrome), tests can be normal in
the first few weeks
C Peripheral neurophysiological studies are helpful in the diag-
nosis of patients presenting with weakness and normal
sensation, such as in motor neuron disease and myasthenia
gravis
cerebral cortex. The normal EEG changes with levels of alertness
and with development from birth to adulthood. A wide range of
patterns is seen in healthy individuals. Diagnosis can be difficult,
but the abnormalities shown in Figure 1 (slow waves, spike and
slow waves, repetitive complexes) are strongly associated with
disorders of cortical function.
EEG in epilepsy
A single 20e60-minute EEG reveals interictal epileptic discharges
(IEDs) in about 50% of patients with epilepsy. Sensitivity in-
creases to more than 90% with repeated recordings and activa-
tion procedures (photic stimulation and over-breathing enhance
electrical abnormalities in primary generalized epilepsy, and
sleep or sleep deprivation can trigger IEDs in most types of epi-
lepsy). In many patients with epilepsy, standard EEGs can be
used for diagnosis and syndromic classification, but, because of
the low sensitivity, a normal standard EEG cannot exclude a
diagnosis of epilepsy. False-positive EEGs showing spike and
slow-wave complexes occur in less than 0.5% of adults without
epilepsy.1
Standard EEGs are usually recorded with a simultaneous pa-
tient video to help identify artefacts and allow correlation be-
tween electrical abnormalities and symptoms. When doubt
remains, prolonged recording can be helpful, using either
ambulatory outpatient recording or inpatient video-telemetry, in
which time-locked EEG and video recordings allowing precise
499 Ó 2016 Published by Elsevier Ltd.
 INVESTIGATIONS

Two-second epochs of electroencephalography recordings

1 second
1
5 1
2
6 2 3
4

7 3 5
6
8 4
7
Right
8

a Normal b Generalized slow-wave activity in

a patient with HIV encephalopathy

c Left hemisphere slow-waves in d Bilateral spike and slow-wave e Focal spike and slow-waves in left
a patient with cerebral vasculitis complexes in primary generalized mesial temporal epilepsy
epilepsy (e.g. absence seizures)

f Repetitive generalized slow-wave g Repetitive generalized slow-wave h Lateralized slow wave complexes
complexes in sporadic Creutzfeldt– complexes in severe hypoxic- in herpes simplex encephalitis
Jakob disease ischaemic encephalopathy

Figure 1

correlation. This technique is most useful in preoperative
assessment of potential candidates for epilepsy surgery and in
non-epileptic attack disorders.2
Spike and slow-wave complexes: these are the hallmark of
epilepsy, seen during seizures (ictal discharges) or, more often,
subclinically between IEDs. Generalized spike and slow-wave
complexes in primary generalized epilepsy have a frequency of
three per second and are bilaterally synchronous and maximal
frontally. In photosensitive patients, these discharges can be
provoked by light flickering at about 18 flashes/second. Local-
ized spike and slow-wave discharges are seen in focal epilepsy,
usually over the relevant cortical region (mostly the temporal or
frontal lobes).

MEDICINE 44:8 500 Ó 2016 Published by Elsevier Ltd.

 INVESTIGATIONS

EEG in encephalopathy: generalized excess slow-wave discharges second. Periodic lateralized epileptiform discharges, similar to
(<8 Hz) indicate a diffuse disorder, such as metabolic, endocrine or IEDs, occur in disorders causing acute cortical necrosis, such as
ischaemic encephalopathy, or neurodegenerative dementia. When infarction or herpes simplex encephalitis. Symmetric triphasic
localized, they indicate focal or multifocal structural cortical dis- waveform complexes are seen in sporadic CreutzfeldteJakob
ease. These findings are non-specific and further tests such as disease and hepatic encephalopathy.
neuroimaging and lumbar puncture may be required.
EEG in intensive treatment unit (ITU) monitoring and
Complex discharges: these have diverse waveforms, typically status epilepticus
with a rhythmic periodic discharge at about one or two per Computer technology using a full set of scalp electrodes facili-
tates management of patients with impaired consciousness from
status epilepticus or encephalopathy, following head injury or
Normal sensory and motor nerve conduction
during administration of anaesthetic drugs. EEG is particularly
important for non-convulsive status epilepticus, which cannot be
detected clinically but, if untreated, leads to a poorer outcome.
10
µV Combined neurophysiological, clinical and imaging findings are
5 used to predict prognosis following cardiac arrest.3
milliseconds
Conduction time Nerve conduction studies and needle electromyography
(EMG)
Record Stimulate Peripheral nerve conduction studies are most easily performed
Conduction by cutaneous electrical stimulation sufficient to excite all axons
distance and applied to a nerve trunk. Propagated activity is recorded at a
distance, either along the nerve or from a target muscle. The
response recorded from the skin is the sum of individual nerve or
muscle fibre action potentials. Only small (5e50 microvolt)
Record Conduction
Stimulate distance Stimulate
1 2 Electromyography recordings
a
5 200 µV
1 F F
mV
5
milliseconds
PSW

200 µV
Conduction
time
2
2 mV

(a) Concentric needle electrode recordings from a denervated

Maximal conduction velocity Conduction distance (mm)
=
(m/second) Conduction time (milliseconds)
Sensory nerve action potential (upper tracing), and compound
muscle action potentials recorded from a distal muscle following
nerve stimulation at two points (lower tracings). Note the
difference in amplitude between the different types of recording
and method for calculating the maximal conduction velocity.
sharp waves (PSW). (b) Examples of single motor unit action
potentials (MUAPs), recorded from a normal muscle (upper
trace) and a chronic neurogenic disorder (bottom trace). Note
the change of sensitivity for the neurogenic potential, which is
much larger. Both the neurogenic and myopathic MUAPs have
a more complex shape than the normal unit.

Figure 2 Figure 3

MEDICINE 44:8 501 Ó 2016 Published by Elsevier Ltd.

 INVESTIGATIONS

recordings are obtained from sensory nerves; necessitating
electronic averaging to separate the nerve potential from back-
ground noise. Compound muscle action potentials are larger (10
e25 mV in the hand). Measurement of conduction distance al-
lows calculation of maximal motor or sensory conduction ve-
locities (Figure 2), usually 50e60 metres/second in the arm and
40e50 metres/second in the leg. The most reliably tested nerves
are the median, ulnar and radial in the arm, and peroneal, tibial
and sural in the leg. Examination of proximal nerves is techni-
cally more difficult, and it is often impossible to calculate con-
duction velocity.
EMG is usually performed using a specially designed needle
electrode that records the electrical potential generated by mus-
cle fibres near its tip. In normal muscle, there is no electrical
activity at rest. With voluntary contraction, progressively more
motor neurons discharge, and summated electrical responses
from muscle fibres (motor unit action potentials (MUAPs)) can
be recorded. In strong contractions, this recruitment pattern of
overlapping MUAPs forms an ‘interference pattern’ that fills the
recording screen.
With degeneration of motor neurons (e.g. axonal peripheral
neuropathy, motor neuron disease), muscle fibres losing their
nerve supply depolarize spontaneously. These small potentials
recorded from muscles at rest are termed ‘fibrillations’ and ‘posi-
tive sharp waves’ (Figure 3a). The number of motor units and the
density of the interference pattern on maximal contraction
decrease, proportionally to the extent of denervation. During re-
covery by reinnervation, the shape of the MUAPs changes because
the number of muscle fibres in the motor unit increases above
normal, producing complex, large-amplitude MUAPs (Figure 3b).
Fasciculations are spontaneous discharges of motor units that
produce a visible twitch and an EMG signal looking like a MUAP.
Uses of nerve conduction and electromyography
Table 1 shows the conditions that can be assessed using a
combination of nerve conduction studies and EMG, along with
the expected findings.
Peripheral neuropathy: nerve conduction studies distinguish
between treatable demyelinating peripheral neuropathy (e.g.

Indications for nerve conduction studies and electromyography (EMG)

Clinical presentation Differential diagnosis Neurophysiological findings

Localized weakness Peripheral nerve lesion Nerve conduction and EMG changes limited to
and sensory the distribution of one nerve
symptoms Radiculopathy Sensory nerve action potential retained in the
area of numbness; EMG changes, if any,
confined to one myotome
Generalized Peripheral neuropathy Slow motor and sensory conduction velocities
weakness and in demyelinating neuropathy
sensory symptoms Reduced response amplitudes and EMG signs
of denervation in axonal neuropathy
Combination of the above often found
Polyradiculopathy and spinal cord lesions Normal motor and sensory nerve conduction
Weakness without Motor neuron disease Retained sensory nerve conduction,
sensory symptoms widespread fasciculation potentials and EMG
signs of denervation/reinnervation
Motor neuropathies Slowing of motor conduction velocity; may
exhibit conduction block, retained sensory
responses
Neuromuscular transmission disorders Myasthenia gravis e may exhibit muscle
action potential decrement on repetitive nerve
stimulation, abnormal ‘jitter’ on single-fibre
EMG
LamberteEaton myasthenic syndrome e low
resting compound muscle action potential
amplitude, increasing after maximal voluntary
activation or tetanic nerve stimulation
Primary myopathy Usually normal nerve conduction; EMG may
show fibrillation potentials caused by muscle
fibre necrosis, myotonia, low-amplitude
polyphasic motor unit action potentials, full
interference pattern of electrical activity
despite weak contraction

Table 1

MEDICINE 44:8 502 Ó 2016 Published by Elsevier Ltd.

 INVESTIGATIONS
acute inflammatory demyelinating polyneuropathy (AIDP),
chronic inflammatory demyelinating polyneuropathy, multifocal
motor neuropathy with conduction block) and untreatable
axonal neuropathy (most metabolic, nutritional and toxic
neuropathies).
Peripheral nerve demyelination impairs saltatory conduction
and reduces maximal conduction velocity, and there may be a
failure of electrical impulse transmission, producing ‘conduction
block’. In hereditary demyelinating neuropathy, conduction
block and dispersion are less than in acquired disease, despite
the very low conduction velocity. When demyelination is
confined to proximal nerve segments, as in AIDP, the only
abnormal finding may be delay of the F-response e a late
response recorded from muscles because a proportion of nerve
impulses travel to the motor neuron cell body and back again.
Criteria based on conduction velocity, conduction block and
prolongation of distal motor and F-response latencies have been
established for the diagnosis of demyelinating neuropathies.4
In contrast, axonal peripheral neuropathy causes a reduction
in the number of excitable nerve fibres, resulting in small com-
pound muscle and sensory action potentials but with normal
conduction velocities in the surviving axons.
Diabetic distal symmetrical neuropathy is length-dependent,
affecting the legs more than the arms, and the nerve conduc-
tion abnormality is similar to axonal neuropathy with the
exception that conduction velocity can be reduced.
Entrapment neuropathy: carpal tunnel syndrome, ulnar neu-
ropathy at the elbow and peroneal palsy at the knee typically
cause selective demyelination with focal slowing of conduction
velocity or block, which localizes the lesion. The degree of the
conduction abnormality can indicate the severity and longevity
of the nerve lesion and can predict prognosis following
treatment.5
Motor neuron disease: sensory nerve conduction studies stay in
the normal range. Muscle responses evoked by electrical stimu-
lation are small, but motor conduction velocity is normal. Needle
EMG shows denervation in a wide anatomical distribution,
including the paraspinal muscles, often with fasciculations.
MEDICINE 44:8
Multifocal motor neuropathy with conduction block can produce
a similar clinical picture, but nerve conduction studies show the
conduction block, and fibrillations are more restricted than in
motor neuron disease.
Myasthenia gravis: the most commonly used test for myasthenia
is reduction of compound muscle action potential amplitude with
repetitive nerve stimulation at three per second (decrement).
EMG is used to exclude other diseases of the motor unit that can
mimic myasthenia. Specialized computer-aided single-fibre EMG
showing jitter (increased variability) is the most sensitive elec-
trodiagnostic test of abnormal neuromuscular transmission.
Primary myopathy: muscle diseases often cause shrinkage of
muscle fibres. When the process is uniform, MUAPs have normal
outlines but are smaller than normal; when non-uniform, MUAPs
become small but more complex (Figure 3b). The number of
motor neurons is usually normal, so maximal contraction, albeit
weak, produces a full interference pattern. Muscular dystrophy
and other diseases with muscle fibre necrosis such as myositis
show fibrillation potentials, and collateral reinnervation with
compensatory muscle fibre hypertrophy produces long-duration,
polyphasic MUAPs. A
KEY REFERENCES
1 Smith SJM. EEG in the diagnosis, classification, and management
of patients with epilepsy. J Neurol Neurosurg Psychiatry 2005;
76(suppl 11): ii2e7.
2 Kennett RP. Modern electroencephalography. J Neurol 2012; 259:
783e9.
3 Fugate JE, Wijdicks EFM, Mandrekar J, et al. Predictors of neuro-
logic outcome in hypothermia after cardiac arrest. Ann Neurol
2010; 68: 907e14.
4 Van den Bergh PYK, Hadden RDM, Bouche P, et al. European
Federation of Neurological Societies/Peripheral Nerve Society
guideline on management of chronic inflammatory demyelinating
polyradiculoneuropathy. Euro J Neurol 2010; 17: 356e63.
5 Bland JD. A neurophysiological grading scale for carpal tunnel
syndrome. Muscle Nerve 2000; 23: 1280e3.
503 Ó 2016 Published by Elsevier Ltd.