The sural nerve was studied orthodromically using the near-nerve tech-

nique in 273 normal subjects (155 females, 118 males) aged 5 to 90 years.
The sensory action potential (SAP), evoked at the dorsum of the foot, was
recorded at the lateral malleolus and midcalf, and at the midcalf when
evoked at the lateral malleolus. In addition, the SAP was recorded at inter-
mediate distal sites and at proximal sites at the popliteal fossa, the gluteal
fold, and the S-1 root. The amplitude of the SAP recorded at midcalf was
32% higher in females than in males. This was probably due to volume-
conduction properties, as differences between genders were less notice-
able at more distal recording sites. The amplitude decreased steeply and
exponentially with age. Conduction distance had a strong influence on the
amplitude of the SAP, which decreased with increasing distance following a
power relationshipwith an exponent of 1.4 to 1.7. This decrease was due to
temporal dispersion with decreased summation and increased phase can-
cellation. The conduction velocity was slightly lower along the very distal
course of the nerve than along more proximal segments.
Key words: sural nerve action potential amplitude conduction velocity.
gender
MUSCLE & NERVE 15:374383 1992

CONDUCTION STUDIES OF THE

NORMAL SURAL NERVE
STEVEN H. HOROWITZ, MD, and CHRISTIAN KRARUP, MD

Because of its accessibility for morphological
studies,' the sural nerve is the sensory nerve in the
lower extremity most often studied electrophysio-
logically in patients with peripheral nerve disor-
ders, allowing correlation between pathological
and pathophysiological findings.".' *
Using near-nerve e l e ~ t r o d e ,our
~ purpose was
to update values in normal subjects, and to inves-
tigate the effects of stimulation and recording
from different sites on the amplitude, duration,
and conduction velocity. The sural nerve was stim-
ulated and evoked compound sensory action po-
From the Laboratories of Clinical Neurophysiology, Department of Neu-
rology, Long Island Jewish Medical Center, Long Island Campus for the
Albert Einstein College of Medicine, New York (Dr. Horowitz), and De-
partment of Medicine (Neurology), Brigham and Women's Hospital, and
the Department of Neurology, Harvard Medical School, Boston, Massa-
chusetts, and the Department of Clinical Neurophysiology, Rigshospita-
iet, Copenhagen, Denmark (Dr. Krarup)
Presented in part at the Vlllth International Congress of Electromyogra-
phy and Related Clinical Neurophysiology, Sorrento, Italy, May, 1987.
Acknowledgments: We are indebted to F Buchthal, MD and W. Troja-
borg, MD for their comments and criticism. We thank Professor A. Rosen-
falck for allowing us to include recordings performed in collaboration with
Alborg University.
Address reprint requests to C. Krarup, MD, Department of Clinical Neu-
rophysiology, Rigshospitalet, Blegdamsvej 9, 2100 Copenhagen, Den-
mark.
Accepted for publication June 1, 1991
CCC 0148-639)(/92/030374- 10 $04.00
0 1992 John Wiley & Sons, Inc.
tentials (SAP) were recorded at various sites from
the distal foot to the first sacral root. The advan-
tages of the technique have been delineated5,6:
these are the precise determinations of the sites of
stimulation and recording along the nerve and de-
tailed resolution of evoked responses.
MATERIALS AND METHODS
Subjects. The sural nerve was studied in 273
subjects: 155 females (F), aged 5 to 80 years; 118
males (M), aged 8 to 90 years. The age distribu-
tion of females (average 42.9 years) was not signif-
icantly different from that of males (average 40.9
years) (Mann- Whitney U test). The subjects were
either paid volunteers or patients without symp-
toms or signs of generalized neuropathy. We in-
cluded patients referred for mononeuropaties in
the upper extremities and lower back pain without
evidence of radiculopathy. All patients had clinical
evaluation of sensory function and motor func-
tion, tendon reflexes, and supplementary EMG
and nerve conduction studies appropriate for
their complaints. Only those patients without clin-
ical and electrophysiological signs of neuropathy
or radiculopathy were included in the study.
Electrophysiological Methods. Orthodromic con-
duction along various segments of the sural nerve

374 Sural Nerve Conduction Studies MUSCLE 8, NERVE March 1992

 I The temperature of the skin over the sural
I
c S 1 root
nerve was kept above 35°C by an infrared, auto-
I matically controlled heating element.
I
I Recording Parameters. The latency of the SAP
I was measured to its first positive peak, its ampli-
I tude peak-to-peak, and the duration of its nega-
t Gluteal fold. GF
I tive phase between the first and last positive peaks.
I The conduction velocity (CV) was calculated ei-
I ther from the latency between the point of stimu-
I
I lation and the point of recording (one recording
I point method, 1R-method), or from the differ-
I
Popliteal fossa, PF ence in latencies between 2 recording points (2-
recording point method, 2R-method).

#
Upper calf. UC
Midcalf. MC
Intermediate calf, IC
Low calf. LC
I
Malleolus Lateralis. ML
__-'
Dorsum pedis. DP
FIGURE 1. Schematic representation of sites of recording and
stimulation of the sural nerve between the dorsum of the foot
and the S1 root. The distance from DP to ML was 80 to 100 mm,
from ML to LC 70 mrn, from LC to IC 30 mm, from 1C to MC 30
mm, and from MC to UC 30 to 50 mrn. Distances proximal to MC
varied among subjects.
was studied (Fig. 1). The basic principles of the
near-nerve technique have been d e ~ c r i b e d . ~
Stimulation and Recording Electrodes. For stimu-
lation and recording, a 0.7-mm diameter insulated
needle with a 3-mm bared tip was placed close to
the nerve. The anode or reference electrode was
another needle with a 5-mm bared tip placed at a
medial transverse distance of 25 to 35 mm. At the
dorsum of the foot, the stimulating electrodes were
placed at a longitudinal distance of 10 to 15 mm.
The stimulus was a 0.2-ms rectangular pulse
from a constant current stimulator. The cathode
was placed close to the nerve by means of a
threshold of <1 mA. The stimulus strength was
then increased to 10% above maximal (4 to 8 mA).
The position of the near-nerve electrode for re-
cording of the sensory response was adjusted to
obtain the maximal amplitude of the sensory ac-
tion potential (SAP). To reduce impedance, all
electrodes were current treated in saline at 90"C.5
The response was recorded via a low-noise Dantec
15C02 amplifier with a frequency band of 200 to
4000 Hz (3 dB down) and averaged 8 to 128 times
using a sampling interval of 20 or 40 ps (Dantec
1500 averager, Nicolet 1074).
Sites of Stimulation and Recording. The sural
nerve was studied from the dorsum of the foot
(DP) to the lateral malleolus (ML) (distal segment),
from ML to midcalf (MC) (intermediate segment),
and proximal to MC (proximal segment) (Fig. 1).
The intermediate and proximal segments were
further subdivided in some subjects (Fig. 1).
With stimulation at DP, the distal segment was
studied by recording the SAP at ML in 182 sub-
jects (103 F, 79M) at low calf (LC) in 65, at inter-
mediate calf (IC) in 5,and at MC in 102.
With stimulation at ML, the intermediate seg-
ment was studied by recording the SAP at LC in
86 subjects, IC in 13 subjects, and at MC in all 273
subjects. In order to further investigate the effect
of conduction distance on the SAP, the proximal
segment was studied by recording at: upper calf
(UC) in 13, at popliteal fossa (PF) in 14, from the
sciatic nerve at the gluteal fold (GF) in 12, and
from the first sacral root (S-1) in 6 subjects (Fig. 1).
Statistical Evaluation.Nonparametric tests and
ANOVA were used to compare values in different
groups. In normal subjects, the lower 95% confi-
dence limits of the amplitudes of the SAP in dif-
ferent age groups were estimated after probability
transformation. When individual amplitudes and
lower 95% confidence limits were plotted against
age after logarithmic transformation, regression
lines were calculated using the least square
method.
Following determination of the mean CV in
each age group, lower 95% confidence limits were
calculated from the mean regression line against
age minus 2 x the estimate of the error (SEy,x).
RESULTS
We examined 109 (SH) and 164 (CK) cases. There
was no difference (Mann- Whitney U ) in age dis-

Sural Nerve Conduction Studies MUSCLE & NERVE March 1992 375
 Relationship between the amplitude of
the SAP and age in females and males
1000

AMP(ma) = 56.6 x exp(-0.021 x age), r = -0.6450

100

10

1
A 0 20 40 60 80 100

Relationship between the conduction velocity

of the SAP and age in females and males
70
CV(all) = 52.8 - 0.05 x age, r = -0.1789 t
i

-----._____

z
0
I-
0
3
n
z
0
0

0 20 40 60 80 100
B AGE (years)

FIGURE 2. The effect of age on amplitude (A) and conduction velocity (B) of the sural sensory nerve
action potential evoked at ML (see text) and recorded at MC in 273 normal subjects. Filled symbols
represent findings in females (155) and open symbols findings in males (1 18). (A) The amplitude is
shown at logarithmic scale. Circles represent individual values, and squares represent the low 95%
confidence limit calculated after probability transformation at each decade. Horizontal bars at the
squares indicate the SEM of the mean age at each decade. Full lines with different elevations were
used to describe the effect of age in females and males. The dashed line show the lower 95% confi-
dence limit as function of age. (B) The conduction velocity is shown on a linear scale. All values have
been pooled to calculate the regression curve (full line). The dashed curves represent the 95% con-
fidence limits calculated from 2 x the SE of the estimate (SEY..J.The least square method was used
to calculate regression curves.

376 Sural Nerve Conduction Studies MUSCLE & NERVE March 1992
tribution (43.0 f 1.7 and 41.4 t 1.3 years, respec-
tively; mean * standard error of the mean, SEM).
The mean amplitude of the SAP evoked by stimu-
lation at ML and recorded at MC in the 2 groups
was the same (33.8 2 2.1 compared with 32.7 t
1.4 pV), indicating comparable electrode place-
ment in the 2 laboratories.
Amplitude of the Compound Sensory Nerve Action
Potential (SAP)
Intermediate Segment. The amplitude of the
SAP at ML (stimulation site, MC) was 32% larger
in females (37.0 k 1.7 p+V)than in males (28.0 t
1.5 pV) (P < 0.0001, ANOVA) in the presence of
similar conduction distances (males, 135 t 1 mm,
females, 132 f 1 mm, not significant, NS). With
stimulation at ML, the SAP recorded at LC was
17% larger in females (77.0 5 5.6 pV) than in
males (66.0 f 4.3 pV). This trend was, however,
not significant (P = 0.12, ANOVA, NS).
In both genders, there was an exponential de-
cline in amplitude with age. This decline was steep
and, at age 80, the amplitude was 87% lower than
at age 10 (Fig. 2, Table 1). The slope of the de-
crease in amplitude averaged 18% to 19% per de-
cade, and was similar in males and females (t =
0.2959, NS), whereas the regression line was sig-
nificantly elevated (t = 4.9939, P < 0.001) in fe-
57m/s
59
I 61A
62
s 4 8 12
FIGURE 3. Sural nerve action potentials evoked at the lateral rnalleolus and recorded at low calf,
rnidcalf, politeal fossa, gluteal fold, and S-1 root. The CV is indicated between the traces.
Sural Nerve Conduction Studies
males compared with males. In contrast to mean
values, the lower 95% limits were similar in both
genders and were, therefore, pooled. T h e slope of
the pooled lower 95% confidence limit as a func-
tion of age was not statistically significant ( P < 0.2)
from the mean regression line.
The duration of the negative phase of the SAP
was similar in males (1.17 f 0.03 ms) and females
(1.10 ? 0.03 ms) (Mann-Whitney U test). The du-
ration increased linearly with age by about 5% per
decade ( P < 0.001), as was also found in record-
ings of the SAP from the median nerve.5
Studies of the Distal Segment. The amplitude of
the SAP at ML (stimulation site, DP) was 24.4 f
1.5 pV in females, which was 12% larger than in
males (21.7 * 2.0 pV); and at MC, 6.0 2 0.5 p V
in females, which was 9% larger than the 5.5 2
0.6 pV in males. The amplitude differences at ML
and MC in females and males were not significant
(ANOVA). At both recording sites, the amplitudes
decreased exponentially with age with a slope of
15% to 17% per decade.
The Amplitude and Duration of the SAP as a Func-
tion cf Conduction Distance. The relationship be-
tween amplitude of the SAP and conduction dis-
tance was best described by a power function
where the amplitude decreased from the site of
stimulation, sharply at short distances, and to a
Mall. lat. - lower c a l f
[+;opv
- midcalf
iI1 O
- popliteal fossa
L3
- gluteal f o l d
L2
- S1 root
L-2
16 20 24 28 32 36ms
MUSCLE & NERVE March 1992 377
lesser degree at longer distances (Fig. 4). For ex-
ample, when the nerve was stimulated at ML and
the action potential recorded at different sites
along the calf (Fig. l), the amplitude decreased by
50 2 2% when the conduction distance was in-
creased from 70 to 130 mm. A similar relationship
was present when the nerve was stimulated at DP
and the SAP recorded at ML and different sites
along the calf. When the responses evoked at LC,
ML, and DP were recorded at the same site, MC, a
similar relationship was present between ampli-
tude and conduction distance.
In 6 to 14 additional subjects, electrodes were
placed at more proximal sites; the popliteal fossa,
gluteal fold, and S-1 root (Fig. 1). At these levels,
the needle was first used to stimulate the nerve,
and a muscle action potential was recorded from
the gastrocnemius muscle. The position of the

AMP = 110815.4 x lIDIST"1.7,

70;
, !
r = -0.9940
.

0 200 400 600 800 1000

V .

0 200 400 600 800 1000

0.5 DUR = 0.675t 0.0035 x DIST,

r = 0.9970

0 200 400 600 800 1000

CONDUCTION DISTANCE (mm)

FIGURE 4. Relationship between mean amplitude (A), latency (B), and duration (C) and conduction
distance in 6 to 14 subjects in whom the SAP evoked at ML was recorded at LC, IC, MC, UC, FP, GF,
and S-1 root. The regression curves were calculated using the least square method in (A) after loga-
rithmic transformation of the axes to yield the power relationship. The vertical and horizontal bars in-
dicate the SEM.

378 Sural Nerve Conduction Studies MUSCLE & NERVE March 1992
 ~~

Table 1. Normal sural nerve conduction Stimulation at the lateral malleous-recording of the compound sensory action
potential at the midcalf (temperature 235°C)

Amplitude (kV) Conduction velocity (m/s)

Females and Females and males

Females males, Lo. 95%
mean (n)* Males, mean (n) conf. limt Mean Lo. 95% conf. 1im.S
5-14 86 (4) 55 (6) 29 52 44
15-24 40 (12) 35 (1 1) 22 52 43
25-34 46 (37) 31 (30) 16 51 43
35-44 36 (39) 30 (28) 12 51 42
45-54 38 (22) 22 (21) 9 50 42
55-64 31 (20) 19 (9) 7 50 42
65-74 19 (16) 19 (9) 5 49 41
75- 11 (5) 6 (4) 4 49 40
*n = Number of subjects.
f95% lower confidence limit calculated from probability distribution for pooled males and females.
#Lower limit calculated from regression of conduction velocity versus age for males and females minus 2 x SD, X.

needle was adjusted such that the threshold of the
muscle action potential was <1 mA. The ampli-
tude of the SAP evoked at ML decreased by about
98% over a distance of 1 m and could be de-
scribed by a power function, which indicated that
the amplitude would decrease only slightly at dis-
tances longer than 400 mm (Figs. 3 and 4).
The duration of the negative phase of the SAP
increased linearly with conduction distance. Over
a distance of 800 mm, the duration increased
more than 5-fold due to marked dispersion (Fig.
4). This dispersion precluded precise measure-
ment of the duration of the polyphasic response at
even longer distances. At distances where dura-
tion could be accurately measured (< 800 mm),
plotting relative amplitude as a function of rela-
tive duration (Fig. 5) revealed marked amplitude
decrements with small changes in duration.
Conduction Velocity of the SAP.
The Intermediate Segment. T h e mean conduc-
tion velocity from ML to MC was 51.0 lr: 0.4 rnls in
females and 50.6 *
0.4 m/s in males (Mann-
Whitney U test, NS) and all data were therefore
pooled (Table 1, Fig. 3). The conduction velocity
decreased linearly by 0.5 rnls per decade ( P <
0.01). The lower 95% confidence limits are indi-
cated in Figure 2 and in Table 1.
The Distal Segment. T h e average conduction
velocity from DP to ML was 46.2 0.4 mls, and*
decreased with age by 0.5 rnls per decade. In 99
subjects, the SAP evoked at DP was recorded at
both ML and MC. The conduction velocity be-
tween the 2 recording electrodes (2R-method) was
55.5 & 0.5 mls.
Conduction Velocities Calculated by the 1R- and the
2R-Method. The conduction velocity between ML
and MC was 9% faster (P < 0.0001, paired t-test)
when calculated from the 2R-method (55.5 0.5 *
m/s) than from the 1R-method (51.5 -+ 0.4 m/s).
This difference indicates that conduction velocities
along different nerve segments should be com-
pared using the same methods of calculation. The
effect of the method of calculation on the conduc-
tion velocity is being analyzed further (Krarup,
Horowitz, and Dahl, in preparation).
Changes in Conduction Velocityfrom Distal to Prox-
imal Along the Neme. When 2 segments of the
sural nerve from distal to proximal were com-
pared using the 1R-method, the conduction velocity
from ML to MC (50.8 lr: 0.3 m/s) was 9 2 1%
greater (P < 0.0001) than that from DP to ML
(46.2 2 0.4 rnls). Similarly, using the 2R-method of
responses evoked at DP, the conduction velocity
*
from ML to LC was 53.1 0.9 m/s, which was 8 +-
2% lower (P < 0.0005, paired t-test) than the
58.2 5 1.0 m/s from LC to MC.
This finding indicated that there was an in-
crease in velocity from distal to proximal segments
of the nerve. In order to determine whether this
increase in velocity continued along very proximal
segments, the 2R-method was used in a small
number of subjects (6 to 14) along different seg-
ments up to the S-1 root. Using ANOVA for com-
parison of segments localized distal to the knee,
along the thigh, and along the lumbosacral plexus,
the conduction velocity distal to the knee was
lower ( P < 0.05) than proximal to the knee,
whereas there was no difference between the 2
segments proximal to the knee. The differences

Sural Nerve Conduction Studies MUSCLE & NERVE March 1992 379
 W
n
3
t
A
n
I
a
W
L
I-
a
A
W
K

0.5 1 .o 1.5 2.0 2.5 3.0

RELATIVE DURATION
FIGURE 5. Mean relative amplitude as a function of mean relative duration of the SAP. Amplitude
and duration were set at 1.0 for the SAP evoked at ML and recored at MC, or evoked at DP and
recorded at ML. The regression curve was calculated using the least square method after logarithmic
transformation of the axes. The vertical and the horizontal bars indicate the SEM.

were, at most, about 20% ( P < 0.02) along the
most distal segment from LC to MC (55.4 1.0
m/s) and the most proximal segment from GF to
S-1 (67.9 k 3.3 m/s). Because the conduction
velocity was lower only along the most distal
segments, this was not reflected in the latency
as function of the conduction distance, which
was described by a linear function (r = 1.000,
Fig. 4).
DISCUSSION
The Effect of Gender and Age on the Amplitude of
the Sensory Potential. Using needle electrodes
for recording, the average amplitude of the sen-
sory potential was 30% to 40% larger in females
than in males, not due to age or distance, and con-
firming findings by Moon et a1." in a smaller
group of subjects. T h e amplitude of the main
component of the sensory potential is determined
mainly by the number of fibers larger than 9 pm
in diameter,3 by volume conduction properties at
the site of by myelinated fiber den-
sity," and by the proximity of the recording elec-
* trode to the nerve.5
There is no histological data to suggest that the
number and density of nerve fibers in the sural
nerve differ in the 2 genders, and it seems un-
likely that the recording electrode was placed sys-
tematically closer to the nerve in females than in
males. It is, however, likely that the subcutaneous
fat layer is thicker at MC in females than in males
in parallel with their generally higher body fat
content. A thicker subcutaneous fat layer could al-
ter the amplitude of the response by increasing
impedance around the nerve, thereby reducing
shunting of the action current into surrounding
tissue. This factor is of particular importance in
chidren 4 years16 or less in whom the larger am-
plitude, however, is also influenced by the greater
density of fibersl3 and by the shorter distance be-
tween the cathode and recording electrode. That
the difference in amplitude between females and
males became smaller, and statistically nonsignifi-
cant, at more distal recording sites may relate to
less subcutaneous fat closer to the ankle.

380 Sural Nerve Conduction Studies MUSCLE & NERVE March 1992
The Influence of Age. At all recording sites, there
was a strong inverse relationship between the am-
plitude of the sensory potential and the age of the
subject. T h e lower 95% confidence limits were not
significantly different in age-matched males and
females indicating that the dispersion of ampli-
tudes was greater in females than in males. The
coefficient of variation was 57% in both males and
females, but in absolute terms, the SD was 33%
larger in females (21.2 pV) than in males (15.9
PV).
The decline in amplitude of the sensory poten-
tial with age is well established. It has been shown
that the number7 and density13 of myelinated fi-
bers in the sural nerve decrease with age, and the
change in both reduces the amplitude of the sen-
sory potential. Quantitatively, the decline in am-
plitude is, however, larger than would be expected
from the reduction in fiber ~ o n t e n t suggesting
,~
that additional factors, such as the subcutaneous
tissue atrophy seen with aging, may contribute to
the reduction in amplitude.
The Effect of Conduction Distance on the Sensory
Potential. The striking effect of conduction dis-
tance on the amplitude of the sensory potential
emphasizes the need for standardized electrode
placement in studies of normal subjects and pa-
tients with peripheral nerve disorders. Lack of
standardization could compromise the diagnosis
and characterization of peripheral nerve disor-
ders, particularly when axonal loss may be
present.
The reduction in amplitude of the SAP as a
function of conduction distance is well estab-
l i ~ h e d ~ ~ ' ~ ~ 'is~ -considered
-it a consequence of
temporal dispersion (Fig. 5) with decreasing sum-
mation and increasing cancellation of phases and
increasing potential duration due to increasing
asynchrony at longer conduction distance.8 The
decline in amplitude of the compound sensory po-
tential is much more pronounced than that of the
compound muscle action potential when the same
segment of nerve is considered." This difference
is due to the triphasic configuration and short du-
ration (1 to 2 ms) of the single fiber nerve action
potential, in contrast to the diphasic shape and
longer duration (5 to 15 ms) of the motor unit po-
tentiak8 This results in pronounced phase cancel-
lation and decreased summation of the compound
sensory action potential with even small increases
in asynchrony.
In our findings, the decline in amplitude was
steep at short distances, but leveled off at dis-
Sural Nerve Conduction Studies
tances of about 400 mm and longer. Kimura et
a1.' found an increase in amplitude and area un-
der the waveform following initial declines, proba-
bly due to differing interactions between unit dis-
charges (phase summation and cancellation). Such
an increase was not seen in the recordings from
the sural nerve nor in a previous study from the
median nerve.") It is, however, likely that the ab-
sence of further decreases in amplitude at long
conduction distance is the result of a balance be-
tween reductions of phase summation and cancel-
lation.
The relationship between the amplitude of the
SAP and conduction distance has been considered
exponential in nature. 12,15 In this study, the am-
plitude along the calf could be described by expo-
nential as well as power functions. Hoyever, at
our longest conduction distances, including re-
cording of the SAP up to the S-1 root, a single ex-
ponential relationship could not adequately de-
scribe our findings. In Figure 6 , it can be seen that
2 exponential functions with widely different
slopes were necessary to describe the amplitudes
throughout the length of the nerve, suggesting
different volume conduction properties at the re-
cording sites along the calf than along the thigh
and the lumbosacral plexus. On the other hand, a
single power function adequately described the
amplitude as a function of distance. Therefore, we
have adopted this relationship throughout this ar-
ticle.
A consequence of the interaction between sin-
gle nerve fiber potentials is that the decline in am-
plitude with increasing distance may be influenced
by the number of fibers in the nerve. I f the num-
ber of fibers is pathologically reduced, and the
SAP polyphasic due to reduced summation, then
the reduction in amplitude as a function of dis-
tance may be smaller than seen in normal subjects.
Thus, the presence of conduction block may be
unnoticed in the SAP if a reduction in the number
of fibers has a diminishing effect on the summa-
tion of the single fiber potentials.
Branching is a factor which may contribute to
the decline in amplitude of the SAP at different
sites along the nerve. Branching of the sural nerve
occurs distal to the ankle, along the upper calf
(e.g., the sural communicating fibers of the sural
nerve to the superficial peroneal nerve), and at
the level of the root, e.g., projection of the sural
nerve to the L-5 as well as the S-1 root (Liquori,
Trojaborg, Krarup, unpublished). If substantial
numbers of fibers deviate from the sural nerve in
these branches, a noticeable decline in the sensory
MUSCLE & NERVE March 1992 381
 2’5jf @
’
REL AMP = 3548.1 x lIDIST”1.7,
r = -0.9945
2.0: :!
s
I
8

1.5;

.
W
P
3
k 1.0- ?
J
n 2
l-
E a
a J
W
K
0.5 -
0.0I.....-”.l.--mm 1 . . . ..--I---+. . ’. .. .rl.rq
I

0 200 400 600 800 1000 0 200 400 600 800 1000

100
8o
70
AMP = 21.24 x exp(-0.003 x DIST), . ->
I
10 Y

w
0 40
3
c
i 30
n 1 n
z
a 5 1 1
t

.1
0 200 400 600 800 1000 0 200 400 600 800 1000
CONDUCTION DISTANCE (mm)

FIGURE 6. Average amplitude of the SAP as a function of conduction distance using power (A and B) or exponential (C) functions, or
both (D). Absolute amplitudes were used in (A), (C), and (D), and the relative amplitude in (6). In (C), the logarithm of amplitude is
shown as a function of distance. Two exponential functions were used to fit the data at short (0) and long (A) distances. Vertical and
horizontal bars indicate the SEM. The least square method was used to calculate the regression curves.

potential would occur. However, the effect of con-
duction distance on the amplitude was striking
when the recordings were confined to the lower
calf where branching does not occur.
Variation in Conduction Velocity at Different Proxi-
mal Conduction Distance. Tapering, i.e., a grad-
ual decrease in diameter of nerve fibers along
their distal course, is a popular notion for which
there is little direct experimental evidence. In ani-
mal studies, a reduction in diameter is considered
to occur only when branching 0ccu1-s.~In cat
nerve, conduction velocities along hindlimb nerves
were almost constant from the root level to the
very distal end of the tibia1 nerve, and morpho-
metric measurements showed a small decrease of
the diameter of the largest fibers in the plantar
nerve.g In human median nerve, the conduction
velocity was constant proximal to the wrist and de-
creased distally along the palm and digits, indicat-
ing that a reduction in diameter of the largest fi-
bers only occurred distally and not gradually over
the length of the nerve.”
In agreement with this consideration, the con-
duction velocities were about 10% lower along the
segment distal to the ankle compared with that
proximal to the ankle, and there was a further
slight increase along the lower calf. At more prox-
imal levels, the velocity remained constant. Behse’
found that the diameter of the largest fibers was 1
to 2 p.m less at the dorsum of the foot than at the
lateral malleolus, which corresonds to the 4.0 k
0.4 m/s (n = 182) slower velocity along the distal
than the more proximal segment in our study,
considering a conversion factor of 4.3 m/s per mi-
~rometer.~

382 Sural Nerve Conduction Studies MUSCLE & NERVE March 1992
REFERENCES
1 . Behse F: Morphometric studies on the human sural nerve.
Acta Neurol Scand 1990;82(suppl):5-38.
2. Behse F, Buchtahal F: Normal sensory conduction in the
nerves of the leg in man. J Neurol Neurosurg Psychiatry
197 1 ;34:404-4 14.
3. Behse F, Buchthal F: Sensory action potentials and biopsy
of the sural nerve in neuropathy. Brain 1978;101:473-
493.
4. Berthold C-H: Morphology of normal peripheral axons, in
Waxman SG (ed), Physiology and Pathology of Axom. New
York, Raven Press, 1978, pp 3-63.
5. Buchthal F, Rosenfalck A: Evoked action potentials and
conduction velocity in human sensory nerves. Brain Res
1966;3:1- 122.
6. Buchthal F, Rosenfalck A, Behse F: Sensory potentials of
normal and diseased nerves, in Dyck PJ, Thomas PK, Lam-
bert EH, Bunge R (eds), Peripheral Neuropathy (2 ed). Phil-
adelphia, WB Saunders, 1984, pp 981-1015.
7. Jacobs JM, Love S: Qualitative and quantitative morphol-
ogy of human sural nerve at different ages. Brain
1985;108:897-924.
8. Kimura J, Sakimura S, Machida M, Fuchigami Y, Ishida T,
Claus D, Kameyama S , Nakazumi Y, Wang J, Yamada T:
Effect of desynchronized inputs on compound sensory
and muscle action potentials. Muscle Nerve 1988;ll:
694-702.
9. Krarup C, Loeb GE: Conduction studies in peripheral cat
Sural Nerve Conduction Studies
nerve using implanted electrodes: I. Methods and findings
in controls. Mwcle Nerve 1988;11:922-932.
10. Krarup C, Stewart JD, Surnner AJ, Pestronk A, Lipton SA:
A syndrome of asymmetric limb weakness with motor con-
duction block. Neurology 1990;40:118- 127.
11. Lambert EH, Dyck PJ: Compound action potentials of
sural nerve in vitro in peripheral neuropathy, in Dyck PJ,
Thomas PK, Lambert EH, Bunge R (eds), Peripheral Neu-
ropathy (2 ed). Philadelphia, WB Saunders, 1984, pp 1030-
1044.
12. Moon A, Reid JC, Trojaborg W: Sural nerve conduction as
a function of age, sex, height and skin temperature. Elec-
troencephalogr Clin Neurophysiol 1985;61 :S 1 1 1 .
13. Ouvrier RA, McLeod JG, Conchin T: Morphometric stud-
ies of sural nerve in childhood. Muscle Nerve 1987;10:47-
53.
14. Schoonhoven R, Schellens RLLA, Stegeman DF, Gabreels-
Festen AAWM: Sensory potentials and sural nerve biopsy:
a model evaluation. Muscle Nerue 1987;10:246-262.
15. Stegeman DF, de Weerd JPC: Modelling compound action
potentials of peripheral nerves in situ. I. Model descrip-
tion; evidence for a non-linear relation between fibre di-
ameter and velocity. Electroencephalogr Clin Neurophysiol
1982;54:436-448.
16. Wagner AL, Buchthal F: Motor and sensory conduction in
infancy and childhood: Reappraisal. Dev Med Child Neurol
1972;14: 189-216.
MUSCLE & NERVE March 1992 383