Cortical nociceptive responses and behavioral correlates in the monkey

Brain Research, 397 (1986) 47-60 Elsevier

47

BRE 12134

Cortical Nociceptive Responses and Behavioral Correlates in the Monkey ERIC H. CHUDLER, WILLIE K. DONG and YORIKO KAWAKAMI

Departments of Anesthesiology, Psychology and Multidisciplinary Pain Center, University of Washington School of Medicine, Seattle, WA 98195 (U.S.A.) (Accepted 15 April 1986)

Key words: Second somatosensory cortex - - Tooth pulp - - Evoked potential - - Nociception - - Escape behavior - - A6 Nerve fiber

Experiments were performed to characterize cerebral cortical activity and pain behavior elicited by electrical stimulation of the tooth pulp in unanesthetized monkeys. Four monkeys were trained on two different operant paradigms: two on a simple escape task and two on an appetitive tolerance-escape task. All monkeys were implanted with bipolar stimulating electrodes in the right maxillary canine tooth and subdural recording electrodes over the left primary (SI) and/or secondary (SII) somatosensory cortices. Subdural tooth pulp-evoked potentials (TPEPs) recorded over the SII consisted of components P1 (27.5 ms), N1 (40.3 ms), P2 (84.0 ms), N2 (163.5 ms), P3 (295.3 ms), and N3 (468.0 ms). The long latency component (P3-N3) was found exclusively over the SII and was elicited by high intensity stimulation. The appearance of component P3-N3 required the recruitment of A6 nerve fibers into the maxillary nerve compound action potential and was correlated with high frequencies of escape. Administration of morphine sulfate (4 mg/kg, i.m.) caused a contemporaneous reduction in escape frequency and in the amplitude of P3-N3 recorded over the SII. The relationships between TPEP amplitude, escape behavior and A6 nerve fiber activity strongly suggest that the SII is involved with nociception and pain behavior. INTRODUCTION Cortical mechanisms that receive and integrate nociceptive information are the least u n d e r s t o o d and studied in pain research. O n e a p p r o a c h to investigate cortical pain mechanisms has been to correlate neurophysiological responses (e.g. e v o k e d potentials, e l e c t r o e n c e p h a l o g r a m ) with subjective pain report in human subjects. M a n y of these studies have correlated the amplitude of specific e v o k e d potential components r e c o r d e d from vertex or o t h e r scalp locations with pain r e p o r t (for review, see C h u d l e r and Dongl4). H o w e v e r , dissociations between e v o k e d potential a m p l i t u d e and pain intensity have been rep o r t e d 1°. Conclusions that can be drawn from studies using this a p p r o a c h are weak for a n u m b e r of reasons. First, the anatomical and physiological bases of electrical events r e c o r d e d from the scalp surface in humans, especially at the vertex locus, are not fully understood. F e w researchers have directly r e c o r d e d from the cortex to d e t e r m i n e the origins of pain-re-

lated e v o k e d potentials. Second, cortical responses elicited by p e r i p h e r a l stimulation may be contaminated by low threshold input. The same vertex potential may be elicited by both non-painful and painful stimuli 19. In these cases, the o b s e r v e d electrophysiological response may not be specifically related to nociceptive events. Therefore, correlations between electrophysiological responses and pain sensation in human studies may be difficult to interpret. While most animal studies of the somatosensory cortex have emphasized non-nociceptive mechanisms, responses to nociceptive stimuli have been reported. In studies using cutaneous stimuli, neurons with nociceptive-specific properties or a wide dynamic response range have been located in the primary (SI) and secondary (SII) s o m a t o s e n s o r y cortices of rats 29'3°, cats 9 and m o n k e y s 28'36'39'54. E v o k e d potential and single unit responses to tooth pulp stimulation have also been r e p o r t e d in rats 26'47'48, cats 16"34"41"42"53and m o n k e y s 4'15'5°. Despite the importance of results from these studies in elucidating non-

Correspondence: W.K. Dong, Department of Anesthesiology, RN-10, University of Washington School of Medicine, Seattle, WA 98195, U.S.A. 0006-8993/86/$03.50 (~) 1986 Elsevier Science Publishers B.V. (Biomedical Division)

48 nociceptive and nociceptive cortical mechanisms, the relevance of the results to pain sensation is weakened by several methodological limitations. Anesthesia in most of these studies probably reduced the responsiveness of many nociceptive units. It is likely that the number of cortical neurons activated by noxious stimuli has been underestimated. However, two studies have examined cortical neuronal responses to nociceptive stimuli in unanesthetized monkeys. Whitsel et al. 54 found in awake, paralyzed monkeys a population of neurons in the SII that responded to noxious cutaneous stimuli. Since stress induced by immobilization can produce analgesia 2'27, the number of neurons responsive to noxious stimuli might have been reduced. Robinson and Burton 3s-4° examined in awake and untrained monkeys the somatotopy and response properties of neurons in the SII, area 7b, retroinsular, postauditory and granular insular cortical areas. They found a small percentage of neurons that responded to 'nociceptive' stimuli (>2 g applied force, heated probe). These 'nociceptive' stimuli generally evoked a slight withdrawal reflex. It is possible that the number of nociceptive neurons may have been overestimated because reflex and avoidance responses to non-nociceptive stimuli were not excluded in their behavioral design. Therefore, the lack of adequate behavioral tests and the use of anesthetics in previous animal experiments have not permitted the study of relationships between cortical neuronal responses and pain behavior. The purpose of the present experiments was to characterize pain behavior and cerebral cortical activity elicited by innocuous and noxious stimulation of the tooth pulp in unanesthetized monkeys trained on operant escape tasks. Tooth pulp-evoked potentials were recorded from the SII, correlated to nerve compound action potentials recorded from the maxillary nerve, and correlated to pain behavior. The data from these experiments support the assumption that the second somatosensory cortex is involved with the processing of nociceptive input and pain perception. MATERIALS AND METHODS

Subjects Four male monkeys (Macaca fascicularis) weighing between 4.5 and 5.6 kg were used in these experiments. The monkeys were housed in individual

cages. On a moved from chair and put experimental

daily schedule, each monkey was reits home cage, placed in a restraining in an isolation chamber for training and recording sessions.

Surgical preparation Using aseptic surgical techniques and halothane anesthesia in each monkey, bipolar stimulating electrodes consisting of multistranded, Teflon-coated stainless steel wires (250 ,urn diameter) were implanted in the dentine of the right maxillary canine tooth as described in an earlier study 15. Stimulating electrode wires were led subcutaneously from the tooth to a multipin connector on the skull. The connector was secured by dental acrylic to vitallium screws and a T-screw assembly on the skull. The impedance (n = 4, 17.0 + 5.6 kff2) across each pair of electrodes was unchanged throughout the entire experimental period. One week after the stimulating electrodes were implanted, subdural recording electrodes were implanted on the cortical surface. Under halothane anesthesia, 4 Teflon-coated, stainless steel or platinumiridium wires with flat, blunt tips were led through an opening in the dura and were placed across the primary and/or second somatosensory cortices. Similar wires inserted through holes in the skull at the occiput served as epidural reference and ground electrodes. All recording electrodes were connected to a multipin connector and secured to the skull with dental acrylic.

Behavioral training The behavioral paradigms used in this study conform to both established ethical guidelines and methodological guidelines for assessing pain behavior in animals (see Appendix). Simple escape task (Fig. 1). Two monkeys were trained to press a response lever to terminate electrical stimulation of the tooth (escape). The methods of ascending and descending limits were used to determine threshold. Unsignaled electrical pulses (250 /~A) were delivered to the maxillary canine tooth at a rate of 1/s in a train of 7 shocks (a trial) and an intertrial interval of 5 s. If the lever was not pressed in response to this stimulus train, current intensity was increased in steps of 250/~A until a lever press occurred. An escape response immediately terminated

49 No Escape

III!1

III

ElcCt~l Slimuli

Escape

LIII

El~t ti¢=l Slimult

Bar prell

k_

T i ~ Oul

I]lllllll 5

10

20

25

30

35

S

Fig. 1. Temporal sequence of events during the simple escape task. Top of the figure shows events that would occur without an escape. The experimenter initiates the delivery of electrical stimuli to the monkey's tooth by depressing a switch. If no response is made by the monkey after 7 shocks, the stimulation is terminated for 5 s. Bottom of the figure illustrates the events associated with an escape. In this example, a bar press by the monkey after 4 shocks terminates the stimuli and delays the succeeding stimuli by 30 s. electrical tooth stimulation for a period of 30 s. Next, a descending series of stimulus trains was presented. A stimulus train was initially presented at an intensity that was known to elicit escape in the ascending stimulus series. Stimuli were terminated for 30 s following each escape. Subsequent stimulus trains were presented at lower current intensities and reduced in 250 /~A steps until the monkey no longer escaped. By repeating the ascending and descending series, an averaged threshold (1.0 T) was obtained for each monkey. Stimulus current intensities are expressed as multiples or fractions of this averaged escape threshold. The possibility of bias and avoidance behavior was reduced by using both the methods of ascending and descending limits to determine escape threshold 32 and by initiating the ascending series at different stimulus intensities. Each animal was considered to be adequately trained when the escape threshold and latency to escape were stable for several days. Current intensities above, at or near, and below escape threshold (1.0 T) were used during the recording sessions. Appetitive tolerance-escape task (Fig. 2). Two additional monkeys were shaped by the method of successive approximations to depress a response lever for 4 s to receive a fruit sauce reward (0.3 ml). Monkeys were fed daily only after an experimental ses-

sion to increase their motivation to obtain food reward by lever pressing. After this task was consistently performed for several days, the maxillary canine tooth was stimulated at a low current intensity (e.g. 250/~A, l/s) during the lever press. The duration of the lever press was accompanied by a low frequency tone (auditory cue) and a trial completion (acceptance of 4 shocks) was followed by a brief high frequency tone. After completion of a trial, the monkey was reinforced with fruit sauce and the current intensity was increased in steps of 250/~A until the monkey released the lever (an escape). An escape was not followed by a reward. After an escape, the stimulation intensity returned to the lowest level (250/~A) and then ascended in 250/~A steps after each completion. This method of ascending limits determined the tolerance-escape threshold (1.0 T). Current intensities above, at or near, and below this threshold were used during the recording sessions. Use of an appetitive component in this pain assessment task may reduce the tendency to avoid higher stimulus intensities. Data from both the appetitive tolerance-escape task and the simple escape task were used to validate the relationships between tooth pulp-evoked potentials and escape behavior (see below).

Data acquisition Evoked potentials and escape behavior were observed at stimulation intensities that were below, at or near, and above the predetermined escape thresholds established using the ascending and descending method of limits. These intensities were preset on a specially designed stimulator that was equipped with an intensity level stepper/coder. Each escape or trial completion changed the stimulus intensity in the following ascending and descending order: A, B, C, D, C, B, A . . . . (where A and D = stimulus intensity below escape threshold, B = stimulus intensity at or near threshold, C = stimulus intensity above escape threshold). This stimulation sequence provided the presentation of an equal number of trials at intensities below, at or near, and above escape threshold. The stimulator-coder device labeled the intensity and order of shocks (events) within a train on magnetic tape. Cortical evoked potentials were led to high impedance probes that were coupled to preamplifiers. The amplified signals were stored on a magnetic tape re-

50 I n i t i a t i o n with C o m p l e t i o n

Bar Press

J

I

lal0~l#01011411D01OOlllllEIIl#llllJ4DllJID010gl0#10110Qe0il011H0

Required Task Time i Actual Response Time"' ...................... Electrical Stimuli

]

i

I

]

' ..................................................................

I

Penalty Time Out

J

Food Delivery Time

!

Initiation without Completion

Bar Press

]

I

~000|*|oaoo--,,,,u--oo0000ooeoioo|.llo00o0on0100ao0aooo0o0|

Required Task Time

~

Electrical Stimuli

I

Actual Response Time '""'"'"'"'"'

1

I

" ;'""'""'"'"'""'""'""""'""'"'"'"'"'"'"'"'"'"'"'""

I l

Penalty Time Out

[

Food Delivery Time

0

I

1.0

2.0

! I

3.0

4.0

5.0

I

6.0

7.0

8.0

9.0

s Fig. 2. Temporal sequence of events during the appetitive tolerance-escape paradigm. Top of the figure shows the events that would occur if a monkey completed the required task. Electrical stimulation of the tooth is initiated when the monkey presses the bar. If the monkey depresses the bar for the preset (required) task time, it is rewarded with 0.3 ml of fruit sauce. Bottom of the figure illustrates the events that would occur if the monkey failed to complete the required task. Electrical stimulation of the tooth is again initiated by the monkey. However, in this case, the monkey releases the bar (escapes) before the required task time is completed. An escape immediately terminates tooth stimulation, initiates a penalty time during which a bar press will not initiate the task and prevents the delivery of a fruit sauce reward.

corder along with coded stimulation intensities and events. The frequency response of the preamplifiers used to record evoked potentials was from 1 Hz to 1 or 3 kHz (3 dB down points, 6 dB/octave slope). In experimental sessions using the appetitive tolerance-escape task, the following behavioral data were recorded. (1) The number of trial completions, (2) the number of escapes, (3) the current intensity associated with an escape and (4) the duration of an experimental session. These data permitted comparison of the ratios of escapes to trial initiations for various stimulus intensities. In experimental sessions

using the simple escape task, the number of escapes, the percent of escape and latency to escape at each stimulus intensity were determined. In two monkeys, the effects of morphine sulfate (4.0 mg/kg, i.m.) on the behavioral and evoked potential responses to tooth pulp stimulation were assessed. To minimize the development of drug tolerance, injections of morphine were given at minimum intervals of 10 days. In one monkey, an attempt was made to reverse the effects of morphine on the evoked potential amplitude and escape behavior by administration of naloxone hydrochloride (400/~g,

51 s.c.). Control experiments using the morphine vehicle (sterile water) as a placebo were also performed. Upon completion of all chronic recording sessions, nerve compound action potentials (CAPs) from the maxillary nerve were recorded in each monkey under barbiturate anesthesia (sodium thiopental). The head of each monkey was placed in a stereotaxic apparatus to perform a craniotomy and hemispherectomy (ipsilateral to the stimulated tooth) that exposed the maxillary nerve immediately distal to the trigeminal ganglion. The CAPs were recorded with bipolar or monopolar platinum-iridium macro-electrodes. The amplified CAPs were led into a high impedance probe and preamplifier with a frequency response of 1 Hz to 3 kHz. These recordings were then led to a Nicolet 1174 computer for signal averaging. Several control experiments were performed and analyzed. To investigate the possibility of contamination of tooth pulp-evoked potentials by muscle-related activity, evoked potentials were recorded in anesthetized monkeys after muscular paralysis with gallamine triethiodide. No alteration in the amplitude or latency of the evoked potential was assumed to indicate the absence of muscle-related artifacts in the recordings. Experiments were also performed to insure that stimulus currents used in chronic experiments were confined to intradental tissues. During the acute experiment and final recording session, CAPs were observed before and after the pulp was removed from the tooth that was stimulated during the chronic recording sessions. Immediately following the pulpectomy, saline-soaked cotton was inserted into the pulp chamber. The absence of CAPs following the pulpectomy indicated that the stimulus did not activate receptors in extradental tissues. The possibility of contaminating tooth pulp-evoked potentials by cortical responses related to auditory cues presented during the appetitive tolerance-escape task was tested by recording evoked potentials and behavioral responses in the absence of tooth stimulation. Two findings were expected in the absence of tooth stimulation. (1) If escapes were made, they should be randomly distributed throughout the experimental session and (2) evoked potentials should be absent from the recordings.

Data analysis Evoked potentials stored on analog magnetic tape

were retrieved and sorted according to stimulus intensity and event. Each averaged record represented a summation of 128 or 256 successive trials and contained 1024 data points (900 or 1112/~s/point sampling rate). For the simple escape task, the percent escape at a given stimulus intensity was calculated using the following formula: % escape =

number of escapes

× 100

number of trials For the appetitive tolerance-escape paradigm, the ratio of escapes to trials was normalized in the two monkeys by the following formula: Ex escape index =

number of trials x Eh number of trials h

where E x = number of escapes at a given stimulus intensity; number of trials x = number of trials at a given stimulus intensity; E h = number of escapes at the highest preset stimulus intensity; number of trials h = number of trials at the highest preset stimulus intensity. Therefore, the escape index for the highest stimulus intensity was equal to a value of 1.00. Escape responses elicited during both the appetitive tolerance-escape paradigm (escape index) and simple escape task (% escape) were combined for comparison with evoked potential data. Escape responses generated above, at or near, or below escape threshold were designated, respectively, as high, medium or low escape frequency. The differences in evoked potential amplitude for different levels of escape behavior were analyzed using the non-parametric Combined S-test 3~. The relationships between the activation of peripheral nerve fiber groups and evoked potential amplitude or escape behavior were qualitatively determined. The same stimulus intensities that elicited various evoked potential components and escape frequencies were used in an acute experiment and final recording session to elicit CAPs from the maxillary nerve. Therefore, inferences could be made about the specific nerve fiber groups that were activated during the chronic recording sessions.

52 TABLE I Peak latencies o f subdural tooth pulp-evoked potentials recorded over the contralateral secondary somatosensory cortex in anesthetized and unanesthetized monkeys

Mean peak latencies (ms) and standard deviations (+ S.D.) of TPEP components were determined from 4 monkeys anesthetized by sodium thiopental and from 4 monkeys used in chronic behavioral experiments. Data from anesthetized monkeys taken from Chudler et al. 15. There was no significant (P > 0.05) main effect between groups (anesthetized and unanesthetized monkeys) for components P1-N3 as determined by repeated measures analysis of variance techniques. However, a significant (P < 0.01) main effect for component latency within the anesthetized and unanesthetized groups was observed. P1

Anesthetized monkeys (n = 4)

23.1 +4.7 Unanesthetized monkeys (n = 4) 27.5 •+3.0

N1

P2

N2

P3

N3

P4

N4

44.3 +16.4 40.3 __+ 6.7

71.8 _+4.2 84.0 +15.0

160.7 +45.7 163.5 +13.0

280.4 +66.3 295.3 +60.3

419.9 + 56.2 468.0 +133.3

560.9 +74.9

661.7 -+74.8 -

On completion of data collection, electrolytic lesions (3 m A , 2 min) were made during acute experimental sessions to mark the positions of the subdurai recording electrodes. Subsequent removal of the recording connector and underlying cranial bone and dura permitted the identification of chronic recording sites on the surface of the cortex. 0

[]

[]

RESULTS In 4 chronically prepared monkeys, tooth pulpevoked potentials (TPEPs) were subdurally recorded over the primary (SI) and second (SII) somatosensory cortices. Shown in Fig. 3 are the locations of subdural sites at which T P E P components were of the largest amplitude. Subdural TPEPs recorded over the SII and elicited by high intensity stimulation were characterized by waveforms consisting of components P1, N1, P2, N2, P3 and N3*. Each c o m p o n e n t was identified by its polarity and average peak latency. Onset latencies of all T P E P components were difficult to define. Mean peak latencies of subdural TPEPs recorded over the SII in 4 anesthetized (data from Chudler et al.15) and 4 unanesthetized monkeys are given in Table I. Although there was some variability in the peak latency of components P 1 - N 3 among animals, no significant (P > 0.05) main effect for iatencies was observed between anesthetized and unanesthetized groups. Subdural TPEPs recorded over the SI were studied in one unanesthetized monkey. Only components P1, N1, P2 and N2 were observed over the SI; components P3 and N3 were not

.4L M 2 - M •

M3-R M4-T

Fig. 3. Subdural recording sites over the primary and second somatosensory cortices in two monkeys trained on the simple escape task (M2-M, M3-R) and two monkeys trained on the appetitive tolerance-escape task (M1-M, M4-T). The large, solid symbols represent recording locations where tooth-pulp evoked potentials were of the largest amplitude for each monkey. The small, open symbols represent additional recording sites where TPEPs were smaller or absent. CS, central sulcus; IS, intraparietal sulcus; LS, lateral sulcus; STS, superior temporal sulcus.

* Alphanumeric symbols indicate the order of major positive (P) or negative (N) waves.

53

Current ThrRihnld~xT)

TPEP (SIt)

IntAnmitv(mA~l

0.1

0.1T

~

1.7

0.TT

"

0.11 ~

}

IVI~ P2~ 13.6

cape behavior were studied in all 4 trained monkeys. Figs. 4 and 5 illustrate these relationships obtained from one monkey using the appetitive tolerance-escape task and another monkey using the simple escape task. As expected, stimuli of higher intensity elicited more frequent escape on both behavioral tasks. Both the percentage of escape and the escape index increased with increasing stimulus intensity. In the simple escape task, escape latency was inversely related to stimulus intensity in one of the two monkeys. However, the order of shocks (events) within a stimulus train at which the monkeys escaped was not correlated to stimulus intensity. Figs. 4 and 5 also show the changes in the TPEP morphology as stimulus intensity was increased. Low stimulus intensities (0.3 _+ 0.2 T) which produced only a low frequency of escape, elicited a TPEP comprised of components P1-N1 and P2-N2. At stimulus intensities at or near escape threshold (0.9 + 0.2 T), the amplitudes of components P1-N1 and P2-N2 were increased and component P3-N3 was recruited. Stimulus intensities above escape threshold (2.3 + 1.8 T) increased the amplitude of component P3-N3. Longer-latency components following the P3-N3 component were not observed at any stimulus intensity. The nerve fiber groups that contributed to eliciting

Escan~ Index

o

/

0.22

y

v

[+

1.4"1"

1.00 I

0

J

I

'

I

'

I

'

I

'

I

'

200 400 600 800 1000 ml

Fig. 4. Relationships between stimulus intensity, subdural tooth pulp-evoked potentials (TPEPs) recorded from the second somatosensory (SII) cortex and escape behavior (appetitive tolerance-escape task). Each evoked potential represents an average of 256 trials displayed in 1146.9 ms sweep lengths. Stimulus onset is preceded by 40.0 ms of baseline activity. Escape index values represent the number of escapes elicited by a given stimulus intensity relative to the number of escapes elicited by the highest stimulus intensity (see Materials and Methods). Stimulus current intensity is represented in milliamperes (mA) and as a fraction or multiple of the average escape threshold (xT).

observed at any stimulus intensity. These findings are consistent with the cortical surface distribution of TPEPs in lightly anesthetized monkeys observed by Chudler et al. 15. The relationships between stimulus current intensity, TPEP amplitude recorded over the SII and esCurrent Intensitv(mA)

1.5

% Escape

Threshql~(xT)

TPEP

I

(SII)

(÷ escapes x 10o)

" 25 trials

lo

. . . . . . . .

0.2T

(~AP

/l

P1

!~ ¢1

P2

P3

4-

@

I 0

[~ 10pV

~ ~

-

t.

PII 100

'

I 200

i

I

'

300

I 400

ms

i

I 500

L

I 600

'

F

0

'

I 2.0

,

f 4.0

i

l 6.0

'

l

'

8.0

ms

Fig. 5. Relationships between stimulus intensity, subdural tooth pulp-evoked potentials (TPEPs) recorded from the second somatosensory (SII) cortex, compound action potentials (CAP) recorded from the maxillary nerve near the trigeminal ganglion, and escape behavior (simple escape task). Each evoked potential represents an average of 128 trials displayed in 921.6 ms sweep lengths. Stimulus onset is preceded by 40.0 ms of baseline activity. Percent escape values represent an average of 25 trials. Each compound action potential recording represents an average of 64 trials displayed in 10.24 ms sweep lengths. Stimulus onset is preceded by 1.0 ms of baseline activity. Stimulus current intensity is represented in milliamperes (mA) and as a fraction or multiple of the average escape threshold (xT). When stimulation intensity was at and above 0.7 T, note the higher percentage of escape behavior, the appearance of the P3-N3 component in the TPEP and the A6 nerve fiber response.

54 different T P E P components and escape frequencies were identified by acute recording experiments (Fig. 5). Stimulus intensities that were used during chronic experiments were again used in acute studies to elicit compound action potentials from the maxillary nerve. As shown in Fig. 5, stimulus intensities that elicited the least number of escapes and the P 1 - N 1 component activated only the large myelinated A~ nerve fibers. The average conduction velocity of the Aft nerve fibers measured in 3 of the 4 monkeys was 42.4 m/s (+ 6.8 m/s). Stimulus intensities (above 0.7 T) that produced a greater frequency of escape and the P 3 - N 3 component recruited the small myelinated A 6 nerve fibers into the maxillary nerve compound action potential. The average conduction velocity of the A6 nerve fibers measured in 3 monkeys tested was 21.4 m/s ( + 2.0 m/s). In all cases, the P 3 - N 3 component was only observed if A6 nerve fibers were activated. Since an increase in stimulus intensity produced an increase in both T P E P amplitude and escape frequency, it was important to establish a relationship between T P E P amplitude and escape frequency. Thus, behavioral data from all 4 monkeys were com-

100-

[~

PI-N1

SII

~iiiiiiiiiiliiii.:.:-:. !i!i!i!i!ii::::::: !i! i!!iiii!i::::::: i !i!i!!iiiiiiiii::.:.:-: iiii!iiiiiili}i~:.:.:.: i:i:i:i:~:~:~:i: ~.~.:."

P2-N2

80-

~

p3-.3

+:.:+:+: ..... ::::::::::::::: ...... ~i!i! i!!!i!!~

]

::::.:::::::::: .-.....

60-

!ii!i:i~i!iii!i _ ..... .....":':':" ::..:::::::::: ....-., .._..... ..... :.:.:.:.::-:. ................... iii'i'i'x :::.:.::.: ....- • •

~i!i!iiiiii:i:":':':' •:::::::::::: ....... • :::::.:.: ....,.,

!ili!i;ii!~!~:.:.:.: .. v. • •..........

20

:.:::.:: ..........

'''

. . . . .....

....... ......

0

!!ili~i~i:iii:i.:.:.:. :zlzi~i~!:i:!~!~ '.','.' !:!:!:.::ii:i:i:':':Ji LOW

MEDIUM

HIGH

Escape Behavior

Fig. 6. Relationship between amplitudes of subdural tooth pulp-evoked potentials (TPEP) recorded over SII and escape behavior. Data from 4 animals, 2 on the simple escape task and 2 on the appetitive tolerance-escape task, have been combined. TPEP amplitudes have been normalized with respect to the amplitude generated by the highest stimulus intensity utilized. Bars represent mean (+ S.D.) peak-to-peak amplitude of TPEP components recorded over the second somatosensory cortex (SII) in relation to different degrees of escape. Significant differences (P < 0.05) in TPEP amplitude were determined for different degrees of escape using the Combined Stest.

bined to determine whether T P E P amplitude changes recorded over the SII were significantly different for various frequencies of escape (Fig. 6). The degrees of escape were categorized as low, medium or high according to the escape frequency elicited, respectively, by stimulus intensities below, at or near, or above escape threshold. For each of the 4 monkeys tested, the degree of escape was based on the following escape frequencies: low escape behavi o r - 10% and 16% escape for trials on the simple escape task and a 0.11 and 0.14 escape index for the appetitive tolerance-escape task; medium escape behavior - - 28% and 66% escape for trials on the simple escape task and a 0.22 and 0.43 escape index for the appetitive tolerance-escape task; high escape behavior - - 84% and 90% escape for trials on the simple escape task and a 1.0 escape index for the appetitive tolerance-escape task. As illustrated in Fig. 6, the amplitudes of components P 1 - N 1 , P 2 - N 2 and P 3 - N 3 all increased with higher degrees of escape. Component P 3 - N 3 was never observed when the degree of escape was low. The Combined S-test 31 indicated significant differences (P < 0.05) in the amplitude of each T P E P component for different degrees of escape. Control experiments were performed on all trained monkeys to insure that subdural TPEPs were not produced by auditory stimuli, muscle-related artifacts or by stimulation of extradental tissue. Since auditory cues were used during the appetitive tolerance-escape task, it was possible that auditory evoked potentials could have been recorded from some of the cortical locations. However, when animals performed the appetitive tolerance-escape task without tooth pulp stimulation, no evoked potentials were recorded (Fig. 7). This indicated that the TPEPs were not contaminated by auditory artifacts. Moreover, escapes were randomly distributed throughout the recording session. Evidence also indicated that the TPEPs recorded from awake monkeys were of neural and not muscular origin. TPEPs that were recorded in all awake monkeys were similar in latency and waveform to TPEPs recorded from monkeys anesthetized with sodium thiopental and paralyzed with gallamine triethiodide. A pulpectomy of the canine tooth stimulated during the chronic recording sessions eliminated all TPEPs and maxillary nerve compound action potentials. Therefore, stimu-

55 P2 /~

A

Ip1 N1

t'

0

li that were used to generate TPEPs in the awake animals were confined to intradental tissues. The effects of morphine sulfate on escape behavior and on the amplitudes of subdural TPEPs recorded over SII were examined in two monkeys trained on the simple escape task. As seen in the left panels of Fig. 8, morphine (4 mg/kg, i.m.) injected after baseline recording did not affect P1-N1 amplitude, but did reduce the amplitudes of P2-N2 and P3-N3. Reduction in escape frequency was contemporaneous with the attenuation of P2-N2 and P3-N3 amplitudes. After morphine injection, each monkey appeared disoriented. Subsequent experiments in the same animal 10 days later (right side of Fig. 8) demonstrated similar reductions in escape frequency and TPEP amplitudes after morphine injection. Administration of naloxone hydrochloride (400 pg, s.c.), an opiate antagonist, 33 min after morphine injection, produced an increase in the amplitudes of P2-N2 and

2°~vE +

P3 N2

'l'

100 200

N3

300

'1400

500 600

1'1' 700

ms

800 M3-R

'

Fig. 7. Comparison of activity recorded over the contralateral SII with and without tooth pulp stimulation. Pairs of overlapping traces are displayed in 921.6 ms sweep lengths. Stimulus onset is preceded by 40.0 ms of baseline activity. Each trace represents an average of 128 trials. A: tooth pulp-evoked potentials recorded over the second somatosensory cortex (SII) during the appetitive tolerance-escape task with auditory cues. B: activity observed with auditory cues but without tooth pulp stimulation. SII

SII

120

100

- -..-4

PI-N1

100

.

="

~

~ o

P3-N3

80' ,1¢

-N2

o~.

e0

P-

0

60"

:" •

=

0

40-

i

20-

20

40

60

80

100

3 - d s y post

20

rain

40

60

j 80

100

3 - d a y post

min

100

100

80

0;×

J

80 ¸

: 8o v

4O

40"

20

20-

o=

'

20

40

60 rain

80

100

i 3 - d a y post

20

3 - d a y pOSt rain

Fig. 8. Effects of morphine sulfate (4 mg/kg, i.m.) and naloxone hydrochloride (400 pg) on subdural TPEP amplitudes recorded over SII and escape behavior (simple escape task). Values are represented as the percent of control peak-to-peak TPEP amplitude (top panels) and percent escape (bottom panels) elicited immediately before morphine injection. Left panels illustrate the effects of morphine alone on TPEP amplitude and escape frequency. Note that the decrease in escape (%) after injection is associated with the decrease in amplitude of components P2-N2 and P3-N3. Component P1-N1 remains unchanged after injection. Right panels illustrate the effects of morphine and naloxone. Components P2-N2 and P3-N3 and escape frequency are reduced after morphine administration, but all are increased after subsequent naloxone injections. Solid circles, P1-N1 amplitude; open circles, P2-N2 amplitude; asterisks, P3-N3 amplitude.

56

SII 140

140

120

120

~ 100

~

80-

~ °

_

~100

~

ao

• 8o-

6o

40

40

20

20 20

40

60 rain

80

100

3-day

post

100

I 20

I 40

I 60 rain

I 80

40

60 mm

80

•

~ 1O0

~

41

q 3-day post

lOO

80 -

/

"3-day pOSt

80

~

40

w

2

20

mln

100

3-day post M4-T

Fig. 9. Effects of morphine and sterile water on subdural TPEP amplitudes recorded over SII and escape behavior (simple escape task). Values are represented as the percent of control peak-to-peak TPEP amplitude (top panels) and percent escape (bottom panels) elicited immediately before morphine injection. Left panels illustrate the effects of morphine (4 mg/kg) and sterile water (1.0 ml) on TPEP amplitude and escape frequency. Note that after morphine injection, the amplitudes of components P2-N2 and P3-N3 and the percentage of escapes were reduced. In the right panels, injection of 1.0 ml sterile water instead of naloxone at 32 min after morphine injection had no effect on either the TPEP amplitude or frequency of escape. After an initial rise in the amplitude of the TPEP components, all amplitudes returned to near baseline levels. No effects on escape behavior were observed following water injection. Solid circles, P1-N1 amplitude; open circles, P2-N2 amplitude; asterisks, P3-N3 amplitude.

P 3 - N 3 and a simultaneous increase in escape frequency (from 0% to 52%). The amplitude of P 2 - N 2 remained at baseline levels for the r e m a i n d e r of the experiment, but P 3 - N 3 amplitude and the frequency of escapes returned to low levels. A n additional injection of naloxone (400/~g, s.c.), 76 min after morphine administration, p r o d u c e d a n o t h e r increase in P 3 - N 3 amplitude and in the percentage of escapes. Two placebo experiments were p e r f o r m e d to insure that the effects of morphine and naloxone on escape behavior and the T P E P s were not related to stress p r o d u c e d by the h y p o d e r m i c injection. In the first placebo e x p e r i m e n t (Fig. 9, left side), morphine was injected as in previous experiments. Escape frequency and T P E P amplitude were again reduced.

Thirty-two min after m o r p h i n e injection, 1.0 ml of sterile water (the morphine sulfate vehicle) was injected instead of naloxone. As seen in Fig. 9, both the escape frequency and T P E P amplitudes r e m a i n e d attenuated. In the second control experiment, 1.0 ml of sterile water was injected instead of morphine after completion of baseline recording (Fig. 9, right side). A f t e r a short period in which the T P E P amplitudes were elevated, the T P E P c o m p o n e n t s returned and remained at p r e t r e a t m e n t levels. The frequency of escape was unaffected by water injection. DISCUSSION Electrical tooth pulp stimulation elicited short la-

57 tency, tooth pulp-evoked potentials (TPEPs) from the primary (SI) and secondary (SII) somatosensory cortices of all unanesthetized, trained monkeys. Subdural rtmpping studies in anesthetized monkeys have shown that these short latency TPEPs (P1-N1) are maximum in amplitude on the postcentral gyrus near the tip of the intraparietal sulcus 15'5°. Tooth pulpdriven neurons have also been identified in the SI of anesthetized monkeys and baboons 4. Intracortical recordings within the postcentral gyrus have revealed the locations of dipole current source generators. Wave polarity inversions of the P1-N1 component occur in cytoarchitectonic area 4 of MI and area 3b of SI but not within the SI115. These observations suggest that the short latency, P 1 - N I component observed in the unanesthetized monkeys was volumeconducted from the SI to the SII. Recent anatomical, physiological and behavioral data suggest that short latency, tooth pulp responses recorded from the SI may not be related to nociception. The tooth pulp contains some large myelinated (Aft) nerve fibers 2°'24 and the extradental conduction velocities of some tooth pulp afferents are in the Aft nerve fiber range sA6'17'33. Physiologically identified intradental afferents with extradental conduction velocities in the Aft nerve fiber range are activated by non-nociceptive mechanical transients applied to the intact enamel of the cat canine tooth 17. Short latency TPEP components recorded from anesthetized monkeys 15 and unanesthetized monkeys in the present study were elicited by low intensity electrical stimulation that activated only Aft nerve fibers. These stimulus intensities elicited escape in less than 20% of the trials. Electrical stimulation of the tooth pulp below pain threshold also produces non-painful sensations in human subjects 3"7'12'35'37'45. The short latency TPEPs recorded from SI are minimally affected by anesthetics 15. Moreover, the present study showed that morphine sulfate has minimal effects on the P1-N1 component. A dissociation between the amplitude of P1-N1 and escape behavior was evident after morphine injection: the percentage of escape was greatly reduced while the amplitude of P1-N1 remained unchanged. Nevertheless, a number of neurons in the SI of anesthetized monkeys has been shown to respond to nociceptive stimuli z8.36. A paucity of nociceptive neurons in the SI may preclude the observation of a surface- or intracortically-recorded

evoked potential related to noxious stimulation. The importance of SI to nociception will require further study of the response properties of individual SI neurons to natural and electrical nociceptive stimulation in unanesthetized monkeys. Subdural TPEPs recorded over the SII of unanesthetized, trained monkeys contained a long-latency component (P3-N3) that was not observed over the SI. Wave polarity reversal of this response within the superior bank of the lateral sulcus indicates a current generator source in the SII ~5. Component P3-N3 was elicited by stimulus currents many times greater than that necessary to evoke components P1-N1 and P2-N2 in unanesthetized monkeys. The P3-N3 component was elicited only when the stimulus intensity was sufficient to activate small myelinated, A6 nerve fibers. Moreover, the P 3 - N 3 component was not observed during either the simple escape task or the appetitive tolerance-escape task when escape frequency was low. These observations collectively suggest that the SII may have an important role in nociception. Alpsan 1 also found an evoked potential component over the SII in anesthetized cats that was only elicited if A6 nerve fibers were activated. Bromm and Scharein 6 found an evoked potential component (330 ms latency) in humans that was observed only when pain was reported. In their study, mechanical or electrical stimuli were applied to the middle finger and evoked potentials were recorded from vertex (Cz). Bromm et al. 5 have also observed a late evoked potential component (1260 ms latency) that was correlated to C-nerve fiber activity. Unfortunately, since vertex was the only recording location for these studies, the underlying generator sources for these evoked potentials remain unknown. Evoked potential recording from a small number of recording sites in humans suggested that the parietal cortex is activated by dental stimulation associated with pain H. Recent mapping studies utilizing a superconducting quantum interference device (SQUID) magnetometer in human subjects, have attempted to localize cortical areas activated by nociceptive stimulation. Hari et al. 23 concluded that the magnetic field pattern elicited by painful dental stimulation was due to a current source in the upper bank of the anterior Sylvian sulcus (SII). Noxious stimulation of the nasal mucosa with humidified carbon dioxide also elicited magnetic fields indicative of a current dipole near SI122.

58 The long-latency, P4-N4 component recorded over the SII of anesthetized monkeys 15 was not observed in the current experiments with unanesthetized monkeys. Two factors may account for this discrepancy. First, high stimulus intensities used in experiments with anesthetized monkeys may not have been tolerated by trained, unanesthetized monkeys. Therefore, a smaller number of high threshold nerve fibers may have been activated in awake, behaving monkeys. Second, monkeys were lightly anesthetized with sodium thiopental in previous acute experiments 15. While large doses of barbiturates commonly reduce the discharge frequency of many neurons throughout the central nervous system, small doses of barbiturates have been shown to increase the nociceptive activity of spinal cord neurons in monkeys and c a t s 25'49 and enhance the activity of neurons in the reticular formation 46, diencephalon 43 and cortex 2~. TPEPs recorded over the SII in unanesthetized, paralyzed rats have been shown to increase in amplitude after barbiturate administration 44. Furthermore, small doses of barbiturates may increase pain sensitivity in humans TM. The effects of morphine sulfate, an opiate analgesic, on the late TPEPs also suggest that the SII has a role in nociception. Morphine was found to attenuate the amplitude of the P2-N2 and P3-N3 components and reduce the number of escapes in the simple escape paradigm. Naloxone hydrochloride, an opiate antagonist, partially reversed the effects of morphine. Placebo experiments using sterile water indicated that the effects observed with morphine and naloxone were not related to stress produced by the drug administration. Unfortunately, high morphine dosages produce analgesic, as well as motor and sedative effects52, that may confound interpretation of the present results. Nevertheless, amplitude changes in components P2-N2 and P3-N3 recorded over SII co-varied with the frequency of escape. Shigenaga and Inoki 44 found that morphine reduced the amplitude of SI and SII TPEPs elicited by contralateral tooth pulp stimulation in the paralyzed, unanesthetized rat. The dose-response relationship between morphine and SII TPEPs led them to suggest that SII is more closely related to the analgesia due to morphine than is SI. Chin and Domino 13found that morphine sulfate (2 mg/kg, i.v.) did not alter the latency or amplitude of TPEPs recorded over the contralat-

eral coronal gyrus (SI) in paralyzed, unanesthetized dogs. The relationships between TPEP amplitude, escape behavior and Ad nerve fiber activity strongly suggest that the SII is involved with nociception and pain perception. Investigations of single neuronal response properties in the SII of trained, unanesthetized animals should further clarify the role of the SII as an integrative site for nociceptive input. A better understanding of the operation of SII n0ciceptive mechanisms may be necessary to formulate appropriate methodological strategies tb study the next order of sensory integration and eventually, the neural substrates that subserve pain perception. APPENDIX The operant behavioral paradigms used in this study conform to suggested guidelines for the ethical treatment of laboratory animals 55 and for assessing pain behavior 5~. Ethical guidelines proposed by the Committee for Research and Ethical Issues of the International Association for the Study of Pain 55 have been satisfied in the current experiments. As suggested in these guidelines, the proposed methods and potential benefits have been reviewed through the National Institutes of Health granting process. Second, the experimenters have experienced painful electrical tooth stimulation and qualitatively evaluated their sensations. Third, careful monitoring of behavioral reactions to noxious stimulation was performed throughout the experiment. Fourth, the stimulus intensities and durations used were the minimum required to produce the escape behaviors and electrophysiological responses in each animal. Fifth, neuromuscular blocking agents were never administered without general anesthetics. Vierck and Cooper 51 have established several criteria for assessing pain reactions in animals. (1) The intensities and locations of stimuli should be known, (2) the animals' reactions should occur at latencies that correspond to the expected onset of pain sensations, (3) the reactions should occur preferentially or exclusively in response to nociceptive stimuli, (4) some of the monitored reactions should not be reflexive, (5) the reactions should be graded in proportion to stimulus intensities within the pain-sensitivity range and (6) multiple reactions should be assessed and should be

59 modulated in the same direction in individual animals by variations of stimulus intensity. Attempts to satis-

capes to evaluate pain behavior.

fy these criteria in the current experiments have been made respectively by (1) using calibrated, bipolar

ACKNOWLEDGEMENTS

electrical stimulation of the tooth and performing control experiments to insure that electrical current

This research was supported by the N I D R ( D E 05130) and in part by the N I N C D S (NS 16329). Dr. Charles J. Vierck is gratefully acknowledged for con-

was confined to intradental tissues, (2) measuring escape behaviors during the presentation of electrical stimuli, (3) using a range of stimulus intensities above and below escape thresholds, (4) using two operant behavioral tasks, an appetitive t o l e r a n c e - e s c a p e and simple escape paradigm, (5) using a range of stimulus intensities in the noxious range and (6) observing escape latencies, trial initiations, and percentage of es-

REFERENCES 1 Alpsan, D., The effect of the selective activation of different peripheral nerve fiber groups on the somatosensory evoked potentials in the cat, Electroencephalogr. Clin. Neurophysiol., 51 (1981) 589-598. 2 Amir, S. and Amit, Z., Endogenous opioid ligands may mediate stress-induced changes in the affective properties of pain related behavior in rats, Life Sci., 23 (1978) 1143-1152.

3 Azerad, J. and Woda, A., Sensation evoked by bipolar intrapulpal stimulation, Pain, 4 (1977) 145-152. 4 Biedenbach, M.A., Van Hassel, H.J. and Brown, A.C., Tooth pulp-driven neurons in somatosensory cortex of primates: role in pain mechanisms including a review of the literature, Pain, 7 (1979) 31-50. 5 Bromm, B., Neitzel, H., Tecklenburg, A. and Treede, R., Evoked cerebral potential correlates of C-fibre activity in man, Neurosci. Lett., 43 (1983) 109-114. 6 Bromm, B. and Scharein, E., Principal component analysis of pain-related cerebral potentials to mechanical and electrical stimulation in man, Electroencephalogr. Clin. Neurophysiol., 53 (1982) 94-103. 7 Brown, A.C., Beeler, W.J., Kloka, A.C. and Fields, R.W., Spatial summation of pre-pain and pain in human teeth, Pain, 21 (1985) 1-16. 8 Cadden, S.W., Lisney, S.J.W. and Matthews, B., Thresholds to electrical stimulation of nerves in cat canine toothpulp with Aft-, A6- and C-fibre conduction velocities, Brain Research, 261 (1983) 31-41. 9 Carreras; M. and Andersson, S,A., Functional properties of neurons of the anterior ectosylvian gyrus of the cat, J. Neurophysiol., 26 (1963) 100-126. 10 Chapman, C.R., Cotpins, Y.M., Mayeno, J.K. and Gagliardi, G.J., Rate of stimulus repetition changes evoked potential amplitude: dental and auditory modalities compared, Exp. Brain Res., 43 (1981) 246-252. 11 Chatrian, G.E., Canfield, R.C., Knauss, T.A. and Lettich, E., Cerebral responses to electrical tooth pulp stimulation in man, Neurology, 25 (1975) 745-757. 12 Chatrian, G.E., Fernandes De Lima, V.M., Lettich, E., Canfield, R.C., Miller, R.C. and Soso, M.J., Electrical

sultations related to our behavioral paradigms. We thank Dr. Richard F. Martin for his participation in the early stages of this study. We are indebted to Mr. Paul R. D e t m e r and Mr. Thomas J. Chang for their design and construction of electronic e q u i p m e n t used in our data analysis. The secretarial assistance of Ms. G w e n Bell was greatly appreciated.

stimulation of tooth pulp in humans. II. Qualities of sensations, Pain, 14 (1982) 233-246. 13 Chin, J.H. and Domino, E.F., Effects of morphine on brain potentials evoked by stimulation of the tooth pulp of the dog, J. Pharm. Exp. Ther., 132 (1961) 74-86. 14 Chudler, E.H. and Dong, W.K., The assessment of pain by cerebral evoked potentials, Pain, 16 (1983) 221-244. 15 Chudler, E.H., Dong, W.K. and Kawakami, Y., Tooth pulp-evoked potentials in the monkey: cortical surface and intracortical distribution, Pain, 22 (1985) 221-233. 16 Dong, W.K. and Chudler, E.H., Origins of tooth pulpevoked far-field and early near-field potentials in the cat, J. Neurophysiol., 51 (1984)859-889. 17 Dong, W.K., Chudler, E.H. and Martin, R.F., Physiological properties of intradental mechanoreceptors, Brain Research, 334 (1985) 389-395. 18 Dundee, J.W., Alterations in response to somatic pain associated with anaesthesia. II. The effect of thiopentone and pentobarbitone, Br. J. Anaesthesiol., 32 (1960) 407-414. 19 Fernandes de Lima, V.M., Chatrian, G.E., Lettich, E., Canfield, R.C., Miller. R.C. and Soso, M.J., Electrical stimulation of tooth pulp in humans. I. Relationships among physical stimulus intensities, psychological magnitude estimates and cerebral evoked potentials, Pain, 14 (1982) 207- 232. 20 Fried, K. and Hildebrand, C., Pulpal axons in developing, mature, and aging feline permanent incisors. A study by electron microscopy, J. Comp. NeuroL, 203 (1981) 23-26. 21 Hargbarth, K.E. and Hojeberg, S., Evidence for subcortical regulation of the afferent discharge to the somatic sensory cortex in man, Nature (London), 179 (1957) 526-527. 22 Hari, R. and Kaukoranta, E., Neuromagnetic studies of somatosensory system: principles and examples, Prog. Neurobiol., 24 (1985) 233-256. 23 Hari, R., Kaukoranta, E., Reinidainen, K., Huopaniemie, T. and Mauno, J., Neuromagnetic localization of cortical activity evoked by painful dental stimulation in man, Neurosci. Lett., 42 (1983) 77-82. 24 Holland, G.R. and Robinson, P.P., The number and size of axons at the apex of the cat's canine tooth, Anat. Rec., 205 (1983) 215-222. 25 Hori, Y., Lee, K.H., Chung, J.M., Endo, K. and Willis,

60

26

27

28

29

30

31 32

33

34

35

36

37

38

39

40

W.D., The effects of small doses of barbiturate on the activity of primate nociceptive tract cells, Brain Research, 307 (1984) 9-15. lriki, A. and Toda, K., Morphine and electroacupuncture: comparison of the effects on the cortical evoked responses after tooth pulp stimulation in rats, Eur. J. Pharmacol., 68 (1980) 83-87. Jorgensen, H.A., Fasmer, O.B., Berge, O.G., Tveiten, L. and Hole, K., Immobilization-induced analgesia: possible involvement of a non-opioid circulating substance, Pharmacol. Biochem. Behav., 20 (1984) 289-292. Kenshalo, D.R., Jr. and Isensee, O., Responses of primate SI cortical neurons to noxious stimuli, J. Neurophysiol., 50 (1983) 1479-1496. Lamour, Y., Wilier, J.C. and Guilbaud, G., Neuronal responses to noxious stimulation in rat somatosensory cortex, Neurosci. Lett., 29 (1982) 35-40. Lamour, Y., Wilier, J.C. and Guilbaud, G., Rat somatosensory (SmI) cortex: I. Characteristics of neuronal responses to noxious stimulation and comparison with responses to non-noxious stimulation, Exp. Brain Res., 49 (1983) 35-45. Leach, C., Introduction to Statistics, John Wiley, Chichester, 1979. Lineberry, C.G., Laboratory animals in pain research. In W.I. Gay (Ed.), Methods in Animal Experimentation, Vol. 6, Academic Press, New York, pp. 237-311. Lisney, S.J.W., Some anatomical and electrophysiological properties of tooth-pulp afferents in the cat, J. Physiol. (London), 284 (1978) 19-36. Matsumoto, N., Functional difference of tooth pulp-driven neurons in oral and facial areas of the somatosensory cortex of the cat, Exp. Neurol., 85 (1984) 437-451. McGrath, P.A., Gracely, R.H., Dubner, R. and Heft, M.W., Non-pain and pain sensations evoked by tooth pulp stimulation, Pain, 15 (1983) 377-388. Mountcastle, V.B. and Powell, T.P.S., Neural mechanisms subserving cutaneous sensibility with special reference to the role of afferent inhibition in sensory perception and discrimination, Bull. Johns Hopkins Hosp., 105 (1959) 201-232. Mumford, J.M. and Stanley, S.J., Sensations on stimulating the pulps of human teeth, thresholds and tolerance ratio, Pain, 10 (1981) 391-398. Robinson, C.J. and Burton, H., Organization of somatosensory receptive fields in cortical areas 7b, retroinsular, postauditory and granular insula of M. fascicularis, J. Comp. Neurol., 192 (1980) 69-92. Robinson, C.J. and Burton, H., Somatic submodality distribution within the second somatosensory (SII), 7b, retroinsular, postauditory and granular insular cortical areas of M. fascicularis, J. Comp. Neurol., 192 (1980) 93-108. Robinson, C.J. and Burton, H., Somatotopographic organization in the second somatosensory areas of M, fascicularis, J. Comp. Neurol., 192 (1980) 43-67.

41 Roos, A., Rydenhag, B. and Andersson, S.A., Cortical responses evoked by tooth pulp stimulation in the cat. Surface and intracortical responses, Pain, 14 (1982) 247-265. 42 Roos, A., Rydenhag, B. and Andersson, S.A., Activity in cortical cells after stimulation of tooth pulp afferents in the cat. Extracellular analysis, Pain, 16 (1983) 61-72. 43 Schlag, J., A study of the action of nembutal on diencephalic and mesencephalic unit activity, Arch. Int. Physiol. Biochem., 54 (1956) 470-488. 44 Shigenaga, Y. and Inoki, R., Effects of morphine and barbiturate on the SI and SII potentials evoked by tooth pulp stimulation of rats, Eur. J. Pharmacol., 36 (1976) 347-353. 45 Shimizu, T., Tooth pre-pain sensation elicited by electrical stimulation, J. Dent. Res., 43 (1964) 467-475. 46 Shimoji, K. and Bickford, R.G., Differential effects of anesthetics on mesencephalic reticular neurons: II. Responses to repetitive somatosensory electrical stimulation, Anesthesiology, 35 (1971) 76-80. 47 Toda, K., Ichioka, M, and Iriki, A., Quantitative relations between noxious stimulus intensity and magnitude of central afferent evoked responses in rat, Neurosci. Lett., 17 (1980) 313-316. 48 Toda, K., Iriki, A. and Tanaka, H., Electroacupuncture suppresses the cortical evoked responses in somatosensory I and II areas after tooth pulp stimulation in the rat, Jap. J. Physiol., 30 (1980) 487-490. 49 Tomemori, N., Shingu, K., Komatsu, T., Urabe, N. and Mori, K., Antianalgesic action of thiamylal sodium in cats, Acta Anesthesiol. Scand., 25 (1981)523-525. 50 Van Hassel, H.J., Biedenbach, M.A. and Brown, A.C., Cortical potentials evoked by tooth pulp stimulation in rhesus monkeys, Arch. Oral Biol., 17 (1972) 1059-1066. 51 Vierck, C.J., Jr. and Cooper, B.Y., Guidelines for assessing pain reactions and pain modulation in laboratory animal subjects. In L. Kruger and J.C. Liebeskind (Eds.), Advances in Pain Research and Therapy, Vol. 6, Raven Press, New York, 1984, pp. 305-322. 52 Vierck, C.J., Jr., Cooper, B.Y., Franzen, O., Ritz, L.A. and Greenspan, J.D,, Behavioral analysis of CNS pathways and transmitter systems involved in conduction and inhibition of pain sensations and reactions in primates. In J. Sprague and A. Epstein (Eds.), Progress in Psychobiology and Physiological Psychology, Vol. 10, Academic Press, New York, 1983, pp. 113-165. 53 Vyklicky, L., Keller, O., Brozek, G. and Butkhuzi, S.M., Cortical potentials evoked by stimulation of tooth pulp afferents in the cat, Brain Research, 41 (1972) 211-213. 54 Whitsel, B.L., Petrucelli, L.M. and Werner, G., Symmetry and connectivity in the map of the body surface in somatosensory area II of primates, J. Neurophysiol., 32 (1969) 170-183. 55 Zimmermann, M., Ethical guidelines for investigations of experimental pain in conscious animals, Pain. 16 (1983) 109-110.