ORIGINAL ARTICLE

Electrodiagnostic Study of Peripheral Nerves in

High-Voltage Electrical Injury
Ki Han Kwon, MD,* Se Hoon Kim, MD, PhD,† Yang Ki Minn, MD, PhD*

It is well known that peripheral nerves are very vulnerable to electricity. However, only a
small portion of individuals who have had high-voltage electrical injury exhibit peripheral
nerve damage. The aim of this study was to investigate peripheral nerve damage in high-
voltage electrical injury, which often occurs in the industrial field. The authors reviewed
the medical records of patients who were admitted to their hospital from January 2009 to
December 2011, because of electrical injuries. The results of nerve conduction studies (NCSs)
were reviewed retrospectively. NCS data of the injured site were compared with those of the
opposite noninjured site and follow-up data. Thirty-seven extremities were reviewed. The
authors found that 18 of 33 median nerves (48.6%) showed abnormalities in at least one
parameter and 15 of 36 ulnar nerves (41.7%) exhibited abnormalities. There was no evidence
of demyelination. Eight patients had undergone NCS on the opposite normal extremities. The
compound muscle action potential and nerve conduction velocity were higher at the normal
site. Follow-up NCS were performed in 14 patients: the compound muscle action potential
and nerve conduction velocity values of all patients were improved. High-voltage electricity
damaged peripheral nerves by causing axonal injury rather than demyelinating injury. Hence,
even if NCSs yield normal findings, peripheral nerves may be damaged. F/U studies and
opposite examinations are required for the exact evaluation of peripheral nerve damage.
(J Burn Care Res 2014;35:e230–e233)

The mechanisms of electrical injury are 1) direct-
contact injury, 2) thermal injury caused by an
electric arc, 3) flame burn from cloth or the envi-
ronment, and 4) falling injury. Unlike electric arc or
flame burns, direct-contact injuries cause deep ther-
mal damage to small and restricted input and output
areas.1,2 The source of tissue injury resulting from
direct contact with electricity is Joule heating (Q = I2
× R × t; Q, heating value; I, current; R, resistance; t,
time) and subsequent secondary injury.3 Because of
the high resistance of the skin, especially in the kera-
tinized areas of hands, most of the electrical energy is
concentrated in input and output areas during a con-
tact injury. With the exception of secondary damage,
From the *Department of Neurology, Hallym University College of
Medicine, Seoul, Korea; and †Department of Pathology, Yonsei
University College of Medicine, Seoul, Korea.
S.H. Kim and Y.K. Minn contributed equally to this work.
Address correspondence to Yang Ki Minn, MD, PhD, Department
of Neurology, Kangnam Sacred Heart Hospital, Hallym
University College of Medicine, Shin-Gil Ro 1, Yeoungdeungpo
gu, Seoul 150-950, Korea.
Copyright © 2013 by the American Burn Association
1559-047X/2014
DOI: 10.1097/BCR.0b013e31829b39b9
e230
the intermediate part between the input and output
areas suffers minimal damage. Because most electri-
cians work in elevated places, such as poles or roofs,
electrical burn injuries may be accompanied by fall-
ing injuries.
Animal studies have demonstrated that nerves can
be damaged by an electrical current without heat:
the flow of the electrical current creates pores on
the membranes of nerves and damages nerve cells.4
However, it is unknown whether electrical current
itself damages peripheral nerves between the input
and output areas, especially in injuries caused by
high-voltage electricity.5 Butler and Gant3 reported
that only a small portion (9.3%) of individuals with
this type of injury showed peripheral nerve damage.
The results of our previous study revealed that about
half of the individuals who had electrical injuries and
underwent an electrodiagnostic examination showed
abnormal findings, even in case of high-voltage
electrical injuries.6 Some authors have claimed that
peripheral nerve damage from electrical injuries is a
secondary injury.7
Electrodiagnostic examination, such as nerve con-
duction studies (NCSs), are the most objective and
Journal of Burn Care & Research
Volume 35, Number 4 Kwon, Kim, and Minn   e231
only method to check the condition of peripheral
nerves. The aim of this study was to evaluate the
effect of electricity on peripheral nerves by analyzing
NCS findings after high-voltage electrical injury.
METHODS
We reviewed the medical records of patients who
were admitted to our hospital from January 2009 to
December 2011, because of electrical injuries. Only
patients who had direct contact between their upper
extremities and high-voltage currents (22,900  V)
were selected. Patients with thermal burns caused
by electric arcs, flashover, electrostatic induction, or
clothes that caught fire were excluded from the study.
Patients who had undergone electrodiagnostic stud-
ies (NCSs) were selected for the analysis. The median
nerve and ulnar nerves were analyzed. Motor NCS
data were recorded on the abductor policis brevis
muscle (for median nerve) and abductor digiti quinti
muscle (for ulnar nerve). For motor nerve analy-
sis, compound muscle action potentials (CMAPs)
on elbow stimulation (lower normal limits: median
nerve, 5.4 mV and ulnar nerve, 4 mV) and nerve
conduction velocity (NCV) on the wrist–elbow seg-
ment (lower normal limits: 50 m/s) were analyzed.
In the ulnar nerve, the stimulation site was distal to
the ulnar groove. Nerves that exhibited no CMAP
during the test were excluded from the analysis.
Data Analysis
Data were analyzed using the paired t-test analysis in
the Statistical Package for the Social Sciences (SPSS;
Windows version 18; SPSS Inc., Chicago, IL). The
level of statistical significance was set at 0.05 for
validity testing.
Ethics Statement
The study protocol was reviewed and approved by
the institutional review board of the Hangang Sacred
Heart Hospital (No. 2012–183). Informed consent
was waived by the board.
Table 1. Nerve conduction study findings in high-voltage electrical injury
Mean
Median nerve (N = 33)
 CMAP (N = 33) 5.5 mV
 NCV (N = 33) 47.7 m/s
Ulnar nerve (N = 36)
 CMAP (N = 36) 6.3 mV
 NCV (N = 36) 53.2 m/s
CMAP, compound muscle action potential; NCV, nerve conduction velocity.
RESULTS
A total of 346 patients visited our hospital because
of electrical injuries during the period examined.
Thirty-seven extremities from 28 patients were
selected. All patients were men, and their mean age
was 45.0 years. The mean duration between acci-
dents and NCsS was 197 days.
Eighteen of 33 median nerves (48.6%) showed
abnormalities in at least one parameter. In addition,
15 of the 36 ulnar nerves (41.7%) showed abnor-
malities. The mean CMAP amplitudes were 5.5 and
6.3 mV in the median and ulnar nerves and 17 of the
33 median nerves (51.5%), and 11 of the 36 ulnar
nerves (30.6%) exhibited abnormal CMAP findings.
The mean NCV in the median and ulnar nerves was
47.7 and 53.2 m/s, respectively. Thirteen of 33
median nerves (39.4%) and eight of 36 ulnar nerves
(22.2%) exhibited NCV abnormalities (Table 1).
Two patients with median nerve injury and two
patients with ulnar nerve injury showed an NCV that
was 70% below the normal limit. All these patients
had low CMAP (<2 mV). No patients exhibited a
proximal–distal amplitude ratio lower than 0.7.
Eight patients underwent NCS for the uninjured
extremities. CMAP and NCV values were higher at
the normal site than at the injured site (Table 2).
Fourteen patents had follow-up NCS. The mean
duration between the first and second NCS was 139
days. All CMAP and NCV values were improved on
follow-up (Table 3).
DISCUSSION
Electricity that enters the body forms a serial circuit
(power source → input area of the skin → periph-
eral nerve → output area of the skin → ground; Fig-
ure 1). The total heat that is produced by this circuit
can be calculated using Joule’s law: Q total = I 2 × Rtotal
× t. Because this circuit is a serial circuit, the total heat
can represent the sum of the heat of the fractions:
Q total = Q Input skin + Q Peripheral nerve + Q Output skin = I 2 ×
RInput skin × t + I 2 × RPeripheral nerve × t + I 2 × ROutput skin × t).
Lower Normal Limit Normal Abnormal
15 18
>5.4 mV 16 17
>50 m/s 20 13
21 15
>4 mV 22 14
>50 m/s 28 8
 Journal of Burn Care & Research
e232   Kwon, Kim, and Minn July/August 2014

Table 2. Electrodiagnostic findings at the injury site and

Heat
opposite normal site
Injury Site Normal Site

Median nerve
 CMAP (N = 8) 7.7 mV 12.6 mV‡ Input Skin Peripheral Output Skin
 NCV (N = 8) 53 m/s 56.4 m/s (R1) nerve (R2) (R3)
Ulnar nerve
 CMAP (N = 7) 5.3 mV 10.2 mV‡
 NCV (N = 7) 53 m/s 56.2 m/s

CMAP, compound muscle action potential; NCV, nerve conduction velocity.

‡P < .05

The resistance of peripheral nerves is very small

compared with that of keratinized skin; thus, most
heat is produced in the input and output skin areas.
This is consistent with the results of previous clinical
reports.1,2 To exclude the local effect of burns, the
wrist–elbow segment was chosen for analysis. As sen-
sory NCSs are more vulnerable to skin conditions,
only motor NCS was performed.
Among patients with abnormal NCS findings,
abnormal CMAP findings were more common than
abnormal NCV findings Only four patients exhib-
ited slow NCV (>70% of the lower normal limit).
All these patients had a very low CMAP (>0.1 mV).
Although distal latency and temporal dispersion
were not known, demyelinating criteria were not
fully met,8 thus, we concluded that the main pathol-
ogy of nerve injury in high-voltage electrical burns
is axonal injury rather than demyelinating injury.
These findings were consistent with a previous the-
ory that postulates that long axis cells exposed to an
electrical current are more vulnerable to electricity-
induced pore formation. Myelinating Schwann cells
have much shorter length for the electrical current
axis than do nerve axons.
In our previous report, we claimed that the mech-
anism of nerve damage in high-voltage injury was
Joule heat because the most proximal segment of
peripheral nerves exhibited normal findings.6 In this
Table 3. Electrodiagnostic findings of the initial and
follow-up studies
Initial Follow-up
Median nerve
 CMAP (N = 12) 5.3 mV 9.4 mV‡
 NCV (N = 11) 51.0 m/s 52.8 m/s
Ulnar nerve
 CMAP (N = 7) 6.0 mV 9.4 mV‡
 NCV (N = 7) 54.6 m/s 51.7 m/s
CMAP, compound muscle action potential; NCV, nerve conduction velocity.
‡P < .05.
Figure 1.  Electric circuit in high-voltage electrical nerve
injury. The circuit of power source → input skin area →
peripheral nerve → output skin area → ground consists of
a serial circuit. Because input (R1) and output (R2) skin
resistances are much higher than that of peripheral nerves
(R2), most heat develops at input and output skin areas.
study, only half or less of the nerves studied exhib-
ited values outside the normal range. However, a
comparison between these values and those of the
opposite noninjured site revealed that the CMAPs of
the injured site were significantly lower than those of
the noninjured site. A follow-up study showed that
CMAPs were improved significantly. These findings
represented evidence that nerves were damaged by
electricity, although their NCS findings were within
the normal range.
CMAP, and not NCV, was significantly lower in
injured vs noninjured sites and exhibited improve-
ment at follow-up. These findings constituted addi-
tional evidence of axonal injury vs demyelinating
injury.
The results presented here suggest that motor
nerve damage caused by electrical injury was com-
mon, although mild and reversible. Mean CMAPs
were higher than the lower normal limit. However,
this conclusion was extracted from subjects with
mild damage (with measurable CMAP findings and
without amputation). In clinical field, many patient
reported prolonged sensory symptoms such as tin-
gling sensation. We think that peripheral nerve dam-
age is reversible but not completely.
Heat is proportional not only to resistance but also
to time (Q = I2 × R × t). The heat that develops at
input and output skin sites leads to the rapid evapora-
tion of water content in tissues. Thus, tissues become
nonconductors rapidly, and this interval of time is
Journal of Burn Care & Research
Volume 35, Number 4 Kwon, Kim, and Minn   e233
inversely proportional to the voltage of the shock:
it occurs within 20 s at 125 V, within 8 s at 250 V,
within 5 s at 500 V, and within 2.5 s at 1,000 V.9
According to this ratio, tissues may become noncon-
ductors within 100 μs at 22,900 V. In fact, the flow-
ing time of electricity is thought to be very short.
Therefore, peripheral nerves might not be damaged
severely in very high-voltage electrical injuries.
In a clinical setting, it is difficult to assess periph-
eral nerve damage. Rhabdomyolysis and vascular
damage commonly accompany high-voltage elec-
trical injuries.7 Kim and Kim10 reported that 82.6%
of high-voltage electrical injury patients showed
increased creatine kinease, 34% showed myoglobi-
emia, and 27.1% had their extremities amputated.
Electricity flows through low-resistance materi-
als. However, high-voltage electricity passes through
tissues as if they are a homogeneous single con-
ductor.7,9 An animal study showed that peripheral
nerves are very vulnerable to electrical current: 40
mA/3  mm of nerve diameter led to irreversible
nerve damage.3 This is equivalent to 5.7 mA/mm2.
Assuming that the skin resistance at the input and
output areas is 50,000 Ω,7 and that the diameter
of the forearm is 8 cm, the electrical density of the
forearm is only 0.46 mA/mm2, which is too low to
generate nerve damage. We think that this is why
some authors claim that electrical nerve damage is a
secondary damage resulting from heat injury. How-
ever, our study showed that high-voltage electrical
current truly damaged peripheral nerves directly.
We think that high-voltage electricity does not pass
through tissues as if they are a homogeneous single
conductor. We see many patients in whom gross
muscle damage is not homogenous, and often either
flexor or extensor muscles are necrotic, with only
mild damage visible to adjacent muscles. We think
this is another evidence that high-voltage electricity
does not pass tissue homogeneously. Only 28 among
346 patients were selected for this study. However,
we did not select patients arbitrarily. We included
all patients who had undergone an NCS. This study
had a retrograde chart-review design. Because
NCSs were performed through the skin, patients
with severe damage, including those with amputa-
tion, were excluded from the study (as they had not
undergone NCS).
NCS is the only method that can evaluate periph-
eral nerves objectively. However, it remains a very
crude assessment of nerve function. Most laborato-
ries that perform NCSs consider findings to be abnor-
mal if the test values are different from the mean by
more than 2 SD. Abnormal NCS findings mean a
peripheral nerve is abnormal; in contrast, normal
NCS findings do not mean that a peripheral nerve is
normal. Our data showed that more than half of the
nerves tested showed normal NCS findings. How-
ever, comparison with the opposite extremity or the
follow-up study showed that the nerves exhibiting
normal NCS findings were not truly normal. This is
an important result because the presence of periph-
eral nerve damage is crucial for the compensation of
individuals who have had industrial accidents.
In conclusion, high-voltage electricity damaged
peripheral nerves by causing axonal injury rather
than demyelinating injury. Even if NCSs yield nor-
mal findings, peripheral nerves may be damaged.
Follow-up studies and opposite-site examinations
are required for the exact evaluation of peripheral
nerve damage
ACKNOWLEDGMENT
We give special thanks to Dr. K. Jang.
REFERENCES
1. Rosenberg DB, Nelson M. Rehabilitation concerns in elec-
trical burn patients: a review of the literature. J Trauma
1988;28:808–12.
2. Jang K, Hwang J, Jang Y, Kim D. The characteristic of neu-
ropathy in electrical burned patients. J Korean Burn Soc
2001;4:109–16.
3. Butler ED, Gant TD. Electrical injuries, with special refer-
ence to the upper extremities. A review of 182 cases. Am J
Surg 1977;134:95–101.
4. Lee RC, Gaylor DC, Bhatt D, Israel DA. Role of cell mem-
brane rupture in the pathogenesis of electrical trauma. J Surg
Res 1988;44:709–19.
5. Ritenour AE, Morton MJ, McManus JG, Barillo DJ, Cancio
LC. Lightning injury: a review. Burns 2008;34:585–94.
6. Minn YK, Cho SJ, Kwon KH, et al. Features of the pripheral
nerve injury in patients with high voltage electrical burn. J
Korean Neurol Assoc 2004;22:604–8.
7. Koumbourlis AC. Electrical injuries. Crit Care Med
2002;30(11 Suppl):S424–30.
8. Alam TA, Chaudhry V, Cornblath DR. Electrophysiological
studies in the Guillain-Barré syndrome: distinguishing sub-
types by published criteria. Muscle Nerve 1998;21:1275–9.
9. Hunt JL, Mason AD Jr, Masterson TS, Pruitt BA Jr.
The pathophysiology of acute electric injuries. J Trauma
1976;16:335–40.
10. Kim D, Kim H. A clinical investigation of tissue injuries in
electric burns- difference between low- and high- tension
injuries. J Korean Burn Soc 2006;9:16–32.