Neuroscience Letters 316 (2001) 7982

www.elsevier.com/locate/neulet

Contact heat evoked potentials as a valid means to study

nociceptive pathways in human subjects
Andrew C.N. Chen*, David M. Niddam, Lars Arendt-Nielsen
Human Brain Mapping and Cortical Imaging Laboratory, The International Doctoral School in Biomedical Sciences and Engineering,
Centre for SensoryMotor Interaction, Aalborg University, Fredrik Bajer Vej 7, D-3, Aalborg 9220, Denmark
Received 9 August 2001; received in revised form 10 October 2001; accepted 10 October 2001

Abstract
Contact heat evoked potentials (CHEPs) have been difficult to elicit due to slow temperature rise times. A recently
developed heat-foil technology was used to elicit pain and CHEPs. Two groups of subjects were separately stimulated at
the left arm with contact heat via one fast-acting (708C/s) heat-foil thermode. A set of CHEPs was recorded, each at three
subjective intensities: warm; slight; and moderate pain. In CHEPs, the 3D topography exhibited four components: T3T4/
N450; Cz/N550; Cz/P750; and Pz/P1000. A vertex topography map was observed in the late Cz/N550Cz/P750 and parietal
topography in the very-late Pz/P1000 components. Consistent statistical values in the peak latencies and amplitudes
were noted between consecutive investigations. The correlation between the pain intensity ratings and the major Cz/
P750 amplitudes was highly significant in each study. Our validity tests suggested CHEPs to be useful for research and
clinical applications in studying human pain activation related to thermal and nociceptive pathways. q 2001 Elsevier
Science Ireland Ltd. All rights reserved.
Keywords: Contact heat evoked potentials; Pain perception; Late/very-late components; Consistency; Thin-fibre afferent

Although heat pain examined by laser evoked potentials
(LEPs) has been studied for 25 years [2,6], the coherent
beam and its small stimulation area (,5 mm diameter)
may not constitute a natural activation as in contact heat.
In addition, laser stimuli are too brief to simulate real life
experience of thermal pain, especially the nagging quality of
deep pain associated with natural thermal injury. LEPs often
consist of late [3], and even ultra-late, vertex potentials [1].
Ultra-late LEPs, first found in a nerve block study [4], can
be obtained by a specially designed apparatus of tiny beam
[1], rectified cohesive beam [13] and a thin hole in metal
film [16]. Nevertheless, it is quite difficult to directly record
the brain potentials of C-fibre thin afferents by the available
laser stimulation unless specific adjustment is implemented.
One further drawback is that lasers often cause superficial
burns lasting several days with hyper-pigmented spots. The
laser apparatus is also expensive, and difficult to calibrate
and operate.
Contact heat can generally evoke both fast and slow pain.
However, due to its slow rising and falling times, the aver-
* Corresponding author. Tel.: 145-9635-9826; fax: 145-9815-
4008.
E-mail address: ac@smi.auc.dk (A.C.N. Chen).
aged evoked potentials to contact heat have rarely been
studied. Here, we report a newly developed heat-foil with
a rapid rising time at 708C/s to elicit contact heat evoked
potentials (CHEPs). The aims of this study were to: (a),
examine the 3D spatio-temporal dynamic topography of
CHEPs; (b), extract the major components in CHEPs; (c),
relate these components to heat pain perception; and finally
(d), investigate the consistency of CHEPs for potential use
in clinical applications.
CHEPs were recorded from two separate groups of 13
(Study-I: age, 25.5 ^ 4.1) and ten (Study-II: age,
26.8 ^ 5.2) males studied independently 2 months apart;
informed consent was obtained from each subject. Contact
heat was delivered via one 2 cm circular (3.14 cm 2), fast-
acting (708C/s), heat-foil thermode (Minco Products, Inc.,
USA), which is a resistive heating element. Pulsed thermal
stimuli were delivered by a computer-controlled stimulator
via one thermode [12]. Heat pulses were sent from the base-
line (adaptive temperature, 31.58C) using an adjustable
power for 300 ms at three intensity levels (Fig. 1).
The stimuli were applied within the volar surface of the
forearm in the left hand. The subject could withdraw his
hand if the stimulus was intolerable. The subject was
instructed to move the arm and area of thermode contact

0304-3940/01/$ - see front matter q 2001 Elsevier Science Ireland Ltd. All rights reserved.
PII: S03 04 - 394 0( 0 1) 02 37 4- 6
80 A.C.N. Chen et al. / Neuroscience Letters 316 (2001) 7982

facts, and rejected if any were found. Valid epochs of the 40

repeated sweeps were averaged within each stimulus inten-
sity before grand mean averaging.
For each averaged CHEP, global field power [10] analysis
was conducted to measure the time window where signifi-
cant brain activation could be isolated. The electrode site
closest to the identified peak extreme was selected for peak
detection within each of the isolated windows. CHEPs from
all 31 electrodes were displayed as topographic maps using
a double-spline interpolation (ASA Software, ANT A/S, the
Netherlands). Peak latencies and amplitudes were subjected
to repeated measure analysis of variance (ANOVA) within
the group and mixed measure ANOVA for between group
Fig. 1. Dynamics of the temperature ramp measured in the ther- comparison.
mode at three intensities corresponding to pain levels at: I-2, A systematic effect between stimulus intensities and pain
non-painful, warm; I-4, slight pain; and I-6, moderate pain. The
ratings was observed. The two groups showed nearly iden-
peak temperature was reached 360 ms after the offset of stimu-
lus pulse (300 ms duration). A fast-acting temperature ramp at tical rating patterns (Table 1). In CHEPs, the prominent
708C/s is the unique feature in inducing synchronized neural activation resided at the vertex and its vicinity. Similar
activation for recording sharp CHEPs. vertex waveforms were observed across both groups for
non-painful and painful intensities (Fig. 2).
slightly between stimuli to avoid potential sensitization of At the low stimulus intensity, only a shallow, very-late
the skin or receptor fatigue. In five ascending and descend- positive wave was observed at the vertex Cz site. In contrast,
ing trials, each of three pre-fixed stimulus levels was rated four clear peaks were identified and isolated at painful
on a 010 point scale (defined below). These thermal levels levels. These were T3/N450, Cz/N550, Cz/P750 and Pz/
(temperatures: 41.0, 49.5 and 54.58C) were then used for P1000. There were no statistical differences between
recording of CHEPs. Between each series of CHEP record- Study-I and Study-II in peak latencies and amplitudes for
ings at a constant stimulus temperature, the subject had a 5 the isolated components (Table 1). The mean and deviation
min break. Forty repeated trials of thermal stimuli, with a values were very similar between these two independent
mean inter-stimulus interval of 10 s (random 812 s), at the groups.
same stimulus intensity were presented to the subjects for For the grand averages in Study-I and Study-II, Fig. 3
recording of CHEPs. The subject rated each stimulation illustrates the 3D topographic maps at the moderate pain
verbally 5 s after each trial of the stimulation on a 010 (I-6) intensity. Comparing these two independent groups,
scale. The corresponding intensities were: 0, no pain; 1,
slight intense; 2, mild intense; 3, moderate intense; 4, slight Table 1
Pain ratings and peak CHEP amplitudes: consistency in two
pain (pain threshold); 5, mild pain; 6, moderate pain; 7,
groups a
moderatestrong pain; 8, strong pain; 9, severe pain; 10,
unbearable pain. I-2 I-4 I-6
Electrodes were placed in accordance with the extended
Pain rating b
1020 International System, and two extra electrodes (C3 0 Study-I 0.8 ^ 0.5 2.7 ^ 0.9 5.9 ^ 1.0
and C4 0 ) were placed about 1 cm posterior to C3 and C4, Study-II 2.2 ^ 0.2 2.7 ^ 1.5 5.5 ^ 1.3
respectively. The EEGs were recorded from 32 surface elec-
trodes (including one electro-oculogram) mounted on the CHEP peak amplitudes c
scalp using a standard EEG-cap and referenced to bilateral T3/N450
Study-I 2 0.2 ^ 1.4 2 0.4 ^ 1.9 2 0.2 ^ 1.5
earlobes (A1/A2). The impedance of the electrodes was kept Study-II 2 0.3 ^ 2.2 2 0.1 ^ 2.1 2 1.2 ^ 2.3
below 5 kV. The EEG signals were sampled at 250 Hz with Cz/N550
bandpass (0.1550 Hz). EEG activity was sampled for 512 Study-I 2 0.4 ^ 2.3 2 0.5 ^ 2.1 2 6.0 ^ 5.1
ms before and 2048 ms following the onset of each series of Study-II 2 0.5 ^ 2.1 2 1.4 ^ 1.1 2 7.4 ^ 3.2
stimuli. Baselines were computed for the time interval 50 Cz/P750
Study-I 1.3 ^ 2.7 3.2 ^ 2.9 13.3 ^ 4.5
ms before stimulation and subtracted prior to averaging. Study-II 0.7 ^ 2.2 2.1 ^ 2.0 15.1 ^ 6.2
However, the CHEPs were triggered by stimulus offset to Pz/P1000
exclude the occasional stimulus artefacts generated from Study-I 2.6 ^ 2.7 4.1 ^ 3.2 6.2 ^ 3.8
amplifier saturation. Also, the use of stimulus offset was Study-II 1.7 ^ 2.8 3.6 ^ 2.2 7.0 ^ 4.1
deemed necessary due to the 300 ms stimulus pulse duration a
Stimulus: I-2, non-pain, warm; I-4, slight pain; I-6, moderate
required to induce noxious sensation. Nevertheless, peak pain.
b
latencies were adjusted to the stimulus onset for reporting. Pain rating (010 scale): mean ^ SD.
c
Each epoch was checked for eye-movement and blink arte- Amplitude (mV): mean ^ SD.
 A.C.N. Chen et al. / Neuroscience Letters 316 (2001) 7982 81

[9]. In addition, the validity of CHEPs is strengthened by

our findings of a close association between the pain ratings
and the brain responses at the vertex Cz/P750 component
and its proceeding Cz/N550 component. The high correla-
tion of the late/very-late component and pain perception is
reminiscent of that in LEP also [5].
A recent publication on CHEPs using a 2000 ms duration
of thermal pulse showed recognizable vertex brain poten-
tials [7]. Limited topographic waveform distribution was
shown in that paper. In our CHEPs to painful stimulation,
we observed four major components in 3D topography: T3/
N450; Cz/N550; Cz/P750; and Pz/P1000 (Fig. 3). Our
CHEPs at the non-painful intensity produced no specific
peak component, which is similar to the absence of the
first positive peak from the recent report [8]. This was attrib-
uted to a lack of synchronized activation of warm and/or
low threshold mechanothermal afferents at the adapting
temperature intensity of 398C. At painful intensities, we
observed a temporal T3/N450 component, followed by a
vertex Cz/N550 component. This vertex N550 component
may correspond to the reported first peak of 500 ms in

Fig. 2. Vertex waveforms in relation to stimulus heat energy

levels (I-2, I-4, I-6). Systematic increase in relation to increase
of energy levels was noted, and the results also demonstrated
the consistency between Study-I (dark) and Study-II (grey).
Prominent vertex components are marked: Cz/N550; Cz/P750;
and Cz/P1000 at I-6 (moderate pain level).

similar 3D topographic maps were observed in these late

and very-late components (Fig. 3). The correlation between
the pain ratings and the Cz/P750 amplitudes over three
intensities were found to be highly significant (P , 0:001)
in both groups (r 0:71; r 0:80), as were the pain ratings
and Cz/N550 amplitudes (negative correlation, r 20:83;
r 20:81), but not others. In addition, similar rate effects
(slopes) were noted for the amplitude/pain relations.
Contact heat exhibits several characteristics of unique
quality for pain research. It activates thin-fibre noxious ther-
mal afferents, in the C-fibre range, in addition to mechano-
thermal A-delta fibres. It elicits diffused nagging pain at a
sufficient intensity. Sometimes, it can produce double pain
sensations: first, pain of sharp quality; and second, pain of Fig. 3. 3D topography of CHEPs at the moderate pain level (I-6)
dull quality [11,14]. Our consistent reports from two inde- for both Study-I and Study-II. The head model is oriented in a
superioranterior view of the right hemisphere (nose down)
pendent groups/studies strongly confirm the reliability in
perspectives. Discernible focal maxima identified are: T3/N450;
elicitation of quantitative CHEPs, and thus show applicabil- Cz/N550; Cz/P750; and Pz/P1000. Similar topographies are
ity for routine clinical use. The reliability of the CHEPs is largely demonstrated in both Study-I and Study-II. (Lower
comparable with that reported in repeated measures of LEPs panel) EEG montage labels.
82 A.C.N. Chen et al. / Neuroscience Letters 316 (2001) 7982
CHEPs induced at a temperature of 458C [7]. At a painful
intensity of 528C, a second peak was also reported around
1000 ms [7], as seen in our Pz/P1000 component. In
contrast, only a very-late component at 830 ms was reported
following stimulus onset to arm stimulation for the contact
heat evoked response [8]. This very-late positive vertex
peak may correspond to our Cz/P750 component. The
small discrepancy of peak latencies observed might be
due to the difference in study apparatus and stimulus para-
meters.
Possibly due to a lack of synchronization from peripheral
volley in receptor activation or cortical responses, no
primary somatosensory activation would be noted. The
first discernible volley can be observed at the temporal
area, which is likely to reflect the initiation of the secondary
somatosensory (SII) activation, and is then followed 100 ms
later by a probable bilateral SII activation with a focal
vertex maxima as the Cz/N550 component. A further
processing, 200 ms later, reflected as the large vertex posi-
tive wave may index the deep generator from the posterior
cingulus cortex as the scalp Cz/P750 component. Finally,
the very-late component shown as the scalp Pz/P1000 wave
may reflect a late processing of the parietal cortex in
noxious somatosensory information. The neuropsychologi-
cal sequence in CHEPs is similar to that of LEPs [6].
The late Cz/N550 component may be in association with
A-delta fibre activation since its conduction velocity has
been estimated at 10 m/s [7]. The very-late Pz/N1000
component at 8001000 ms may be in association with C-
fibre activation with the conduction velocity estimated at 2
3 m/s for the ultra-late component at 10001500 ms [13,16].
Thus, the isolation of late Cz/N550 and very-late Pz/P1000
components may allow to infer the integrity of A-delta and
C-fibre peripheral afferents and could be useful for further
evaluation of the clinical pathophysiology associated with
thin-fibre alteration in patients and neuropathic pain [15].
This study was supported by funding from the Danish
National Research Foundation.
[1] Bragard, D., Chen, A.C.N. and Plaghki, L., Direct isolation of
ultra-late (C-fibre) evoked brain potentials by CO2 laser
stimulation of tiny cutaneous surface areas in man,
Neurosci. Lett., 209 (1996) 8184.
[2] Bromm, B. and Lorenz, J., Neurophysiological evaluation of
pain, Electroenceph. clin. Neurophysiol., 107 (1998) 227
253.
[3] Bromm, B. and Treede, R.D., Laser-evoked cerebral poten-
tials in the assessment of cutaneous pain sensitivity in
normal subjects and patients, Rev. Neurol., 147 (1991)
625643.
[4] Bromm, B., Neitzel, H., Tecklenburg, A. and Treede, R.D.,
Evoked cerebral potential correlates of C-fibre activity in
man, Neurosci. Lett., 43 (1983) 109114.
[5] Carmon, A., Dotan, Y. and Sarne, Y., Correlation of subjec-
tive pain experience with cerebral evoked responses to
noxious thermal stimulations, Exp. Brain Res., 33 (1978)
445453.
[6] Chen, A.C.N., Arendt-Nielsen, L. and Plaghki, L., Laser-
evoked potentials in human pain: II. Cerebral generators,
Pain Forum, 7 (1998) 201211.
[7] Harkins, S.W., Price, D.D., Roy, A., Itskovich, V.V. and Fei,
D.Y., Somatosensory evoked potentials associated with
thermal activation of type II Adelta mechanoheat nocicep-
tive afferents, Int. J. Neurosci., 104 (2000) 93111.
[8] Itskovich, V.V., Fei, D.Y. and Harkins, S.W., Psychophysio-
logical and psychophysical responses to experimental pain
induced by two types of cutaneous thermal stimuli, Int. J.
Neurosci., 105 (2000) 6375.
[9] Kazarians, H., Scharein, E. and Bromm, B., Laser evoked
brain potentials in response to painful trigeminal nerve acti-
vation, Int. J. Neurosci., 81 (1995) 111122.
[10] Lehmann, D. and Skrandies, W., Spatial analysis of evoked
potentials in man a review, Prog. Neurobiol., 23 (1984)
227250.
[11] Magerl, W., Ali, Z., Ellrich, J., Meyer, R.A. and Treede, R.D.,
C- and A delta-fiber components of heat-evoked cerebral
potentials in healthy human subjects, Pain, 82 (1999) 127
137.
[12] Nielsen, J., Evaluation of Temporal and Spatial Parameters
Involved in Summation of Heat Pain, Aalborg University
Press, Aalborg, 1997 Ph.D. Thesis.
[13] Opsommer, E., Weiss, T., Plaghki, L. and Miltner, W.H.,
Dipole analysis of ultralate (C-fibres) evoked potentials
after laser stimulation of tiny cutaneous surface areas in
humans, Neurosci. Lett., 298 (2001) 4144.
[14] Price, D.D., Selective activation of A-delta and C nociceptive
afferents by different parameters of nociceptive heat stimu-
lation: a tool for analysis of central mechanisms of pain,
Pain, 68 (1996) 13.
[15] Santiago, S., Ferrer, T. and Espinosa, M.L., Neurophysiolo-
gical studies of thin myelinated (A delta) and unmyelinated
(C) fibers: application to peripheral neuropathies, Neuro-
physiol. Clin., 30 (2000) 2742.
[16] Tran, T.D., Lam, K., Hoshiyama, M. and Kakigi, R., A new
method for measuring the conduction velocities of Abeta-,
Adelta- and C-fibers following electric and CO(2) laser
stimulation in humans, Neurosci. Lett., 301 (2001) 87190.