Clinical Neurophysiology 114 (2003) 1419–1422

www.elsevier.com/locate/clinph

Early diagnosis of diabetic neuropathy using double-shock

stimulation of peripheral nerves
Meliha Tana,*, Uner Tanb
a
Department of Neurology, Adana Research Hospital, Baskent University, Adana, Turkey
b
Department of Physiology, Medical School, Cukurova University, Adana, Turkey
Accepted 6 May 2003

Abstract
Objective: The purpose of this study was to determine the changes in the amplitudes of a sensory nerve action potential (NAP) to a
conditioning stimulus given prior to a test stimulus at 2 – 8 ms intervals in healthy subjects and patients with diabetes mellitus with no clinical
signs of neuropathy and normal nerve conduction velocities (NCVs), to be able to diagnose peripheral neuropathy at its very early stages.
Methods: NAPs in the superficial branch of the radial nerve were recorded in healthy subjects (28 women and 7 men) and type II diabetes
patients without neuropathy (22 women and 12 men). Radial nerve was first stimulated with a single shock and then with double shocks at
intervals of 2, 3, 4, 5, 6, 7, and 8 ms; NAP amplitudes and NAP1/NAP2 ratios were calculated in normals and diabetics. NCVs were within
the normal ranges (.50 m/s) in all subjects.
Results: Of the independent variables—group (control, patient), sex (male, female), and hand (right, left)—only group significantly
influenced NAP amplitude; mean NAP amplitude (single shock) was significantly lower in patients than controls. NAP1/NAP2 ratios were
slightly below one (facilitation) in controls; it was above one at 1 – 8 ms stimulus intervals (inhibition) in diabetics, which was strongest at
smallest intervals, gradually decreasing, and almost disappearing as the stimulus interval approached 8 ms.
Conclusions: Using double-shock stimuli, an early diagnosis of peripheral neuropathy would be possible in diabetics without clinical signs
of peripheral neuropathy and exhibiting no slowing in NCV.
q 2003 International Federation of Clinical Neurophysiology. Published by Elsevier Science Ireland Ltd. All rights reserved.
Keywords: Diabetic neuropathy; Nerve action potential; Refractory period; Peripheral nerve; Excitation; Inhibition

1. Introduction 2000). Neural excitability in diabetic neuropathy has

The most frequent complication of diabetes mellitus is
diabetic neuropathy, which is associated with a decrease in
the speed of neural transmission, a decrease in the amplitude
of nerve and compound muscle action potentials, and
electromyographic signs of denervation. It is important to
identify subclinical neuropathy before the clinical findings
appear; the patients may respond better to an early treatment
if the condition is diagnosed early enough (Braune, 1999).
In a number of peripheral nerve diseases, including
diabetic neuropathy, sensory findings are clearer than motor
findings, and thus, attempts have been made to develop
methods that will provide information on the excitability of
sensory nerve fibers (see Burke et al., 2001; Mogyoros et al.,
* Corresponding author.
E-mail addresses: melihatan100@netscape.net (M. Tan), unertan@cu.
edu.tr (U. Tan).
1388-2457/03/$30.00 q 2003 International Federation of Clinical Neurophysiology. Published by Elsevier Science Ireland Ltd. All rights reserved.
doi:10.1016/S1388-2457(03)00154-8
been investigated by the threshold electrotonus method
(Quasthoff, 1998). Horn et al. (1996) reported that diabetic
neuropathy may be the result of a disorder in inward
rectification in the peripheral nerves. On the other hand,
changes in neural excitability may also occur in the
refractory period. Accordingly, an increased refractory
period was reported in diabetic patients (Lowitzsch et al.,
1973; Tackmann et al., 1974).
The above-mentioned studies suggested that increase in
sensory nerve refractory period may occur earlier than
diminishing of the sensory nerve conduction velocity
(NCV). Despite this important finding, there have been
only a few studies on this subject (Schutt et al., 1983;
Ruijten et al., 1994; Braune, 1999), probably because of this
time consuming technique. Ruijten et al. (1994) reported
that motor transmission speed was 46% sensitive in
determining diabetic neuropathy, while this rate increased
1420 M. Tan, U. Tan / Clinical Neurophysiology 114 (2003) 1419–1422
to 73% with refractory period measurements. Braune (1999)
measured the sural and radial nerve’s relative refractory
periods with double shocks and found pathological
reduction of sural NCV in 4 of 30 patients, two of them
with additional reduction of radial NCV. The NCVs were
below normal in these patients. Therefore, the author has
suggested that “the measurements should be performed if n.
suralis NCV is normal”.
The above studies indicated that measurements of the
relative refractory period are more sensitive in detection of
neuropathy than NCV alone. In light of the above-
mentioned studies, the response characteristics of a sensory
nerve to double-shock stimulations seem to be very
important in diagnosis of peripheral neuropathy at its very
early stages. In doing so, the therapeutic interventions
would be more successful than those at the irreversible
stages of neuropathy. Therefore, we have studied the
response characteristics of a sensory nerve to a conditioning
stimulus applied at relatively short intervals, to be able to
diagnose neurodegenerative changes in diabetic patients
exhibiting no clinical signs of neuropathy and having
normal NCVs. Sex and hand preference of the subjects were
also considered, since these factors have been reported to
influence the peripheral NCVs (see for instance, Tan and
Tan, 1995, 1998a,b).
2. Methods
Thirty-four type II diabetes mellitus patients (22 women
and 12 men, average age: 55 ^ 4.8) were included in the
study. The average duration of diabetes was 2 years.
Subjects with histories of trauma and cervical disk
herniation were excluded. Patients had no clinical findings
for diabetic neuropathy with respect to neurological and
electrophysiological examinations. The control group
comprised of 35 healthy subjects (28 women and 7 men,
average age: 52 ^ 6.2) with no history of neurological
disease. On the morning of electrophysiological examin-
ation, the subjects had breakfast and, if required, took their
antidiabetic medication. In addition to general physical
examination, a detailed neurological examination was
carried out. Standard electrophysiological tests were
performed on all patients, and patients with polyneuropathy
were excluded from the study. Care was taken to maintain
extremity temperature above 33 8C, with heating as
necessary. In order to record sensory nerve action potentials
(NAP), the recording electrodes were attached to the thumb
and stimulus electrodes were placed over the radial nerve on
the radius. The distance between the recording and stimulus
electrodes was 100 mm. The stimulus intensity just for the
maximal NAP was determined. Then, at the same stimulus
intensity, the sensory radial nerve (ramus superficialis) was
stimulated first with a single shock followed by double
shocks at intervals of 2, 3, 4, 5, 6, 7, and 8 ms, during which
the NAPs were recorded. There were 2 s intervals between
consequent trials. In this study, we chose the superficial
radial nerve, since it is a pure sensory nerve and easy to
access. The amplitudes of the NAPs were measured from
baseline to peak; the ratio of the amplitudes for the first and
second NAPs (NAP1/NAP2) was calculated. We have taken
subjects only with normal NCVs (. 50 m/s).
The statistical package SPSS for Windows (Version
10.0) was used for statistical analysis. ANOVA was used to
determine whether there were any significant differences
between the NAP1/NAP2 ratios for the right and left hands
in healthy subjects and diabetic patients in men and women.
So, the following factors were considered: sex, hand, and
groups (normal and diabetic); the dependent variable was
NAP1/NAP2. The traditional level of significance ðp ¼
0:05Þ was also accepted for the present study.
3. Results
In single-shock stimulations, sex was not a significant
factor influencing the amplitudes of the NAPs from the right
and left hands (F ¼ 0:06, p . 0:05). There was also no
significant difference between NAPs from the right and left
radial nerves, that is, the difference between sides was also
insignificant (F ¼ 0:03, p . 0:05). Therefore, the data
obtained from the right and left sides were combined.
At single-shock stimulations, there was a significant
difference between the mean amplitudes of the NAPs in the
control and diabetic groups (independent variables): the
mean amplitude of the NAPs was significantly smaller in
diabetics than controls (F ¼ 11:5, p , 0:001) and the mean
amplitudes of the radial NAPs were 9.4 ^ 2.5 and 7.4 ^ 2.5
mV for the healthy and diabetic groups, respectively. The
box-and-whisker plots in Fig. 1 illustrates the median values
(line across the boxes), interquartile range (50% of values),
minimum, and maximum values of the NAPs (caps) for the
control and diabetic groups.
There was a significant difference between the diabetic
and control groups in terms of NAP1/NAP2 ratios for 2 ms
stimulus intervals (F ¼ 30:6, p , 0:001): the average ratios
were found to be 1.36 ^ 0.35 mV for the diabetic group and
0.99 ^ 0.26 mV for the control group. The significant
difference between the healthy and diabetic groups
continued at 3 and 4 ms intervals ðp , 0:000Þ, but at 5
and 6 ms intervals, the significance began to decrease
(p , 0:001 and p , 0:004 for 5 and 6 ms, respectively).
There was no significant difference between the control and
patient groups at 7 ms stimulus interval (F ¼ 3:5,
p . 0:05).
Fig. 2 illustrates the relationships between stimulus
intervals and NAP1/NAP2 ratios in controls (closed circles,
straight line) and diabetics (open circles, dashed line). As
seen in Fig. 2, the mean ratios in the control group were
consistently below one, while those for the diabetic group
were above one. Interestingly, although there are no striking
differences between the ratios in the control group, the ratios
 M. Tan, U. Tan / Clinical Neurophysiology 114 (2003) 1419–1422
Fig. 1. Box-and-whiskers plots showing the median (across the boxes),
interquartile range (top and bottom of boxes), and the largest and smallest
amplitudes of NAPs observed (whiskers) in control subjects (CONTROLS)
and patients (DIABETICS).
in the diabetic group were high at the smallest intervals,
gradually decreasing as the stimulus intervals increased,
approaching one at intervals of 7– 8 ms.
The box-and-whisker plots in Fig. 3 show the NAP1/
NAP2 ratios with interquartile range containing 50% of
values (boxes), as well as minimum and maximum ratios
(whiskers) in the control (A) and diabetic groups (B) at
various stimulus intervals ranging from 2 to 8 ms. While
these average values were below one in the control group,
they were above one in the diabetic group. There were,
Fig. 2. The relationships between double-shock intervals (abscissa) and NAP1/NAP2 ratios (ordinate) for controls (closed circles, straight line) and diabetics
(open circles, dashed line).
1421
however, some overlaps among minimum and maximum
values of ratios.
4. Discussion
Using single-shock stimuli, we have found that the mean
amplitude of the radial NAPs recorded from the superficial
branch was significantly smaller in the diabetic patients than
the healthy controls. This decrease suggests that some thick
fibers disappear before neuropathic findings appear. The
fact that the NAP1/NAP2 ratio did not change depending on
the double-stimulus interval or even showed slight facili-
tation in healthy controls and the NAP1/NAP2 ratio
increased towards inhibition in the diabetic group may be
interpreted as an early symptom of peripheral nerve
degeneration. The NAP1/NAP2 ratio may be considered
as an index for the number of fibers responding to the second
stimulus. If the normal refractory period increases in
diabetics, the number of fibers responding to the second
stimulus will decrease and the amplitude of NAP2 will be
smaller than that of NAP1, and, as a result, the NAP1/NAP2
ratio will increase in diabetics. The fact that the refractory
period increases in diabetics suggests that the excitability of
the sensory peripheral nerves decreases in this disease.
The results of the present work are consistent with those
reporting an increase in the sensory nerve refractory period
in diabetics with and without neuropathy (see above).
Moreover, we have observed that these changes can occur in
diabetics without clinical signs of neuropathy and even with
a normal NCV. Braune (1999) has also reported refractory
period changes in some diabetics without clinical signs of
neuropathy but only in 3 of 30 patients (superficial radial
nerve). He studied the double-shock intervals from 4.6 ms
backwards. We used much longer interstimulus intervals up
1422 M. Tan, U. Tan / Clinical Neurophysiology 114 (2003) 1419–1422
Fig. 3. Box-and-whiskers plots for NAP1/NAP2 ratios in control (A) and
diabetic groups (B) for various interstimulus intervals. Boxes, interquartile
range which contains 50% of values; whiskers, largest and smallest values
observed.
to 8 ms. Additionally, we performed our study in patients
with normal NCVs. We have usually observed facilitatory
influences in double-shock experiments in healthy subjects
and inhibitory influences in diabetic subjects. These original
results suggest that the double-stimulus technique may be of
utmost importance in detecting the very early changes in
diabetic nerves before the appearance of any clinical signs
of neuropathy in patients even with normal NCVs. This
conclusion cannot, however, be applied to all individual
cases, since there were overlappings between both groups as
Fig. 3 clearly shows.
In conclusion, we have studied the effects of conditioning
stimulus (NAP1) on the test stimulus (NAP2) with stimulus
intervals of 2– 8 ms in healthy subjects and diabetic patients
exhibiting normal NCVs and no clinical signs of neuro-
pathy. In normal subjects, the NAP1 usually evoked
facilitation in NAP2. Interestingly, NAP1 caused an
inhibition in NAP2 in diabetics. The results suggested that
the double-shock stimulation might be used in very early
detection of pathologic changes in diabetic nerves with
normal NCVs. This method may be used to detect the very
early changes in sensory nerves of diabetic patients before
the irreversible changes occur; this may be of utmost
importance for early treatment of diabetic neuropathy. Both
structural and functional abnormalities seen in diabetic
neuropathy were indeed reported to be completely corrected
with an early insulin therapy (Brismar et al., 1987;
Quasthoff, 1998).
Acknowledgements
We cordially thank Dr Memet Ozmenoglu, Chief of the
Department of Neurology, BlackSea Technical University
(Trabzon, Turkey) for allowing to work in the EMG
Laboratory and Dr Cavit Boz for technical help.
References
Braune HJ. Testing of refractory period in sensory nerve fibres is the most
sensitive method to assess beginning polyneuropathy in diabetics.
Electroenceph clin Neurophysiol 1999;39:355–9.
Brismar T, Sima AA, Greene DA. Reversible and irreversible nodal
dysfunction in diabetic neuropathy. Ann Neurol 1987;21:504 –7.
Burke D, Kiernan MC, Bostock H. Excitability of human axons. Clin
Neurophysiol 2001;112:1575–85.
Horn S, Quasthoff S, Grafe P, Bostock H, Renner R, Schrank B. Abnormal
axonal inward rectification in diabetic neuropathy. Muscle Nerve 1996;
19:1268– 75.
Lowitzsch K, Hoff HC, Schlegel HJ. Conduction of two or more impulses in
relation to the fibre spectrum in the mixed human peripheral nerve. In:
Desmedt JE, editor. New developments in electromyography and
clinical neurophysiology, vol. 3. Basel: Karger; 1973. p. 272–8.
Mogyoros I, Lin C, Dowla S, Grosskreutz J, Burke D. Reproducibility of
indices of axonal excitability in human subjects. Clin Neurophysiol
2000;111:23–8.
Quasthoff S. The role of axonal ion conductances in diabetic neuropathy: a
review. Muscle Nerve 1998;21:1246 –55.
Ruijten MW, De Haan GJ, Michels RP, Pruijs-Brands A, Ongerboer de
Visser BW, Verberk MM. Motor nerve refractory period distribution
assessed by two techniques in diabetic polyneuropathy. Electroenceph
clin Neurophysiol 1994;93:306–11.
Schutt P, Muche H, Lehmann HJ. Refractory period impairment in sural
nerves of diabetics. J Neurol 1983;229:113–9.
Tackmann W, Lehmann HJ, Ullerich D. Refractory period in human
sensory nerve fibres—transmission of frequent impulse series in human
sensory nerve fibres. Eur Neurol 1974;12:277–92.
Tan M, Tan U. The motor conduction velocities of the median and ulnar
nerves in relation to the carpal tunnel diameters in the male and female
controls and carpet weavers. Int J Neurosci 1998a;94:223–32.
Tan M, Tan U. Correlation of carpal tunnel size and conduction velocity of
the sensory median and ulnar nerves of male and female controls and
carpet weavers. Percept Mot Skills 1998b;87:1195–201.
Tan M, Tan U. Possible lateralization of peripheral nerve conduction
associated with gender. Percept Mot Skills 1995;81:939–43.