| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
The Journal of Neurophysiology Vol. 80 No. 2 August 1998, pp. 628-637
Copyright ©1998 by the American Physiological Society
Department of Physiology, University of Oklahoma Health Sciences Center, Oklahoma City, Oklahoma 73190
 |
ABSTRACT |
|---|
 Abstract
 Introduction
Methods
Results
Discussion
References
|
|---|
Chandler, Margaret J., Jianhua Zhang, and Robert D. Foreman. Cardiopulmonary sympathetic input excites primate cuneothalamic neurons: comparison with spinothalamic tract neurons. J. Neurophysiol. 80: 628-637, 1998. Stimulation of cardiopulmonary sympathetic afferent fibers excites thoracic and cervical spinothalamic tract (STT) cells that respond primarily to noxious somatic stimuli. Neurons in dorsal column nuclei respond primarily to innocuous somatic inputs, but noxious stimulation of pelvic viscera activates gracile neurons. The purpose of this study was to compare effects of thoracic visceral input on cuneothalamic and STT neurons. Stellate ganglia of 17 anesthetized monkeys (Macaca fascicularis) were stimulated electrically to activate cardiopulmonary sympathetic afferent fibers. Somatic receptive fields were manipulated with brush, tap, and pinch stimuli. Extracellular discharge rate was recorded for neurons antidromically activated from ventroposterolateral (VPL) thalamus. Stimulation of the ipsilateral stellate ganglion increased activity of 17 of 38 cuneothalamic neurons and of 1 gracilothalamic neuron with an upper body somatic field. Spinal cord transections showed that cardiopulmonary input to cuneothalamic neurons traveled in ipsilateral dorsal column and probably in dorsolateral funiculus. One of eight gracilothalamic neurons with lower body fields was inhibited by cardiopulmonary input, and none were excited. Stimulation of the ipsilateral stellate ganglion increased activity in 10 of 10 T3-T4 STT neurons. Evoked discharge rates, latencies to activation and durations of peristimulus histogram peaks were significantly less for cuneothalamic neurons compared with STT neurons. Furthermore, additional long latency peaks of activity developed in histograms for 6 of 10 STT neurons but never for cuneothalamic neurons. Contralateral cardiopulmonary sympathetic input did not excite cuneothalamic neurons but increased activity of 7 of 10 T3-T4 STT neurons. Most cuneothalamic neurons (24 of 31 cells tested) responded primarily to innocuous somatic stimuli, whereas STT neurons responded primarily or solely to noxious pinch of somatic fields. Neurons that responded to cardiopulmonary input most often had somatic fields located on proximal arm and chest. Results of this study showed that cardiopulmonary input was transmitted in dorsal pathways to cuneate nucleus and then to VPL thalamus and confirmed that STT neurons transmit nociceptive cardiopulmonary input to VPL thalamus. Differences in neuronal responses to noxious stimulation of cardiopulmonary sympathetic afferent fibers suggest that dorsal and ventrolateral pathways to VPL thalamus play different roles in the transmission and integration of nociceptive cardiac information.
Studies in primates show that stimulation of A Experiments were performed on 17 male monkeys (Macaca fascicularis) weighing between 4.3 and 6.9 kg. These animals also were used to examine other hypotheses not addressed in this study. Protocols were approved by the Institutional Animal Care and Use Committee and followed guidelines of the American Physiological Society and the International Association for the Study of Pain. Monkeys were tranquilized with ketamine (10-20 mg/kg im), and catheters were inserted in the right femoral artery to measure blood pressure and in the right femoral vein to infuse drugs and fluids. Anesthesia was induced with Neurons in dorsal column nuclei
Extracellular unit recordings were made of 50 neurons located in dorsal column nuclei (Fig. 1A). Cuneothalamic or gracilothalamic neurons were recorded from both the right and left side in 8 of 12 monkeys. All neurons recorded in these experiments were antidromically activated from contralateral VPL or lateral posterior thalamus (Fig. 1B). Individual neurons recorded in monkeys with two stimulating electrodes placed 2-mm apart in the coronal plane were antidromically activated from either one or both thalamic electrodes; stimulus threshold for cuneothalamic neurons was lowest from the medial electrode, whereas stimulus threshold for gracilothalamic neurons was lowest from the lateral electrode.
Gracilothalamic neurons
The major purpose of this study was to determine responses of cuneothalamic neurons to cardiopulmonary sympathetic afferent input. However, lesion sites for 12 antidromically activated neurons were located in the gracile nuclei (Fig. 1A). No somatic fields were found for 3 of 12 gracilothalamic neurons, and these neurons did not respond to cardiopulmonary input. Eight of 12 gracilothalamic neurons had somatic fields located on proximal lower body regions, such as the groin, hip, or thigh. Stimulation of ipsilateral and contralateral cardiopulmonary afferents inhibited one cell and did not affect the activity of seven of eight cells with lower body receptive fields. One gracilothalamic neuron bordering the left cuneate nucleus was included in the analyses for cuneothalamic neurons because it had a somatic field located on the left upper arm/thorax and was excited by stimulation of the left stellate ganglion.
Cuneothalamic neurons
A total of 38 neurons were recorded in or above the cuneate nucleus. Lesion sites were located histologically for 32 cells; cell locations for six neurons were extrapolated from distance parameters recorded in relation to lesion sites of other neurons (Fig. 1A). Electrical stimulation of ipsilateral cardiopulmonary sympathetic afferent fibers increased activity of 17 of 39 cells, including 1 cell that was recorded in the gracile nucleus (see Fig. 1A, EXCITATORY RESPONSES RECORDED AS PERISTIMULUS HISTOGRAMS.
Peristimulus histograms (50 sweeps) were generated at 1 Hz for 14 of 17 cuneothalamic neurons that were excited by ipsilateral cardiopulmonary sympathetic input; a short latency (
RESPONSES AT CHARACTERISTICS OF SOMATIC FIELDS.
Excitatory somatic receptive fields were located usually on the ipsilateral hand, arm, or chest; one cell was excited by brushing the lower jaw. Many somatic fields were small; however, some fields encompassed a large part of the arm, and one neuron that responded to stimulation of the ipsilateral stellate ganglion was excited by brushing and pinching the entire ipsilateral thorax and arm. Figure 4A shows representative somatic fields for cuneothalamic neurons that were excited by cardiopulmonary sympathetic input, and Fig. 4B shows fields for cells that did not respond. Neurons tested for somatic fields (n = 31) were grouped in Table 2 according to their responses to stimulating the ipsilateral stellate ganglion and the distal or proximal components of their somatic fields. Neurons excited by cardiopulmonary sympathetic input were significantly more likely (P < 0.01, Fisher's exact test) to have somatic fields that included proximal regions (upper arm, chest, jaw), whereas unresponsive neurons were more likely to have somatic fields located exclusively in distal regions (hand, forearm). The majority of cuneothalamic neurons responded primarily to innocuous somatic inputs; most were classified as LT or LTtap neurons and three cells responded maximally or exclusively to gentle flexion or extension of a finger joint (Table 3). Five of seven neurons that responded maximally or exclusively to noxious pinch (WDR or HT) were excited by stimulation of the ipsilateral stellate ganglion. However, no significant correlations were found between the type of somatic input that increased discharge rate and the response of a neuron to stimulating cardiopulmonary sympathetic fibers.
SPINAL CORD TRANSECTION.
Spinal cord transections were made at C3-C6 to identify pathways that transmitted cardiopulmonary sympathetic input to five cuneothalamic neurons. All cells required transection of lateral fibers in the ipsilateral dorsal column before their responses to stimulating the stellate ganglion were eliminated. Portions of the ipsilateral dorsal funiculus were included in all lesions, varying from interruption of fibers bordering the dorsal column to transection of the entire dorsal funiculus. Medial dorsal column fibers remained intact in four of five lesions. Figure 5 shows a representative example of the effects of ipsilateral spinal cord transection. Interruption of fibers in lateral dorsal column and dorsal funiculus at C5 (Fig. 5D) eliminated responses to cardiopulmonary input (Fig. 5, A and B). Effects of somatic input to this LT cuneothalamic neuron were attenuated markedly (Fig. 5C) after the spinal cord lesion.
Spinothalamic tract neurons
For direct comparison with effects on cuneothalamic neurons, and to establish that findings in this study were the same as past results (Ammons et al. 1983 RESPONSES TO IPSILATERAL CARDIOPULMONARY INPUT.
Electrical stimulation of the ipsilateral stellate ganglion at 1 Hz (50 sweeps) produced at least one peak of evoked impulses in peristimulus histograms for 10 of 10 STT neurons. The second row in Table 1 shows mean values for number of impulses/stimulus, threshold stimulus intensity, latency to first bin of evoked impulses, and duration of the first peak of contiguous evoked impulses (33 V, n = 9; 9 V, n = 1). The average number of impulses/stimulus was significantly greater (P < 0.05), and both latency and duration for the first peak of evoked activity were longer (P < 0.01) for T3-T4 STT neurons compared with responses of cuneothalamic neurons. Effects of graded stimulus intensities, and the mean of the responses (heavy line), are shown in Fig. 3B for the first peak of evoked activity in seven STT neurons. Increasing stimulus intensity generally increased STT cell activity.
RESPONSES TO CONTRALATERAL CARDIOPULMONARY INPUT.
In contrast to cuneothalamic neurons, which were not excited by contralateral cardiopulmonary input, stimulation of the contralateral stellate ganglion at 1 Hz increased cell discharges of seven STT neurons and did not affect activity of three cells; stimulus intensity was 33 V except for one cell that was excited with 13 V stimulus. Mean increase in discharge rate, however, was significantly less (P < 0.001) in response to contralateral cardiopulmonary input (1.9 ± 0.3 imp/stimulus) compared with ipsilateral cardiopulmonary input (4.6 ± 0.8 imp/stimulus) for the same STT neurons (n = 7). Furthermore, latency to first bin of evoked activity was significantly longer (P < 0.05) in response to stimulating the contralateral stellate ganglion compared with stimulating the ipsilateral stellate ganglion (6.1 ± 0.8 ms and 3.1 ± 0.3 ms, respectively, n = 7), and threshold stimulus intensity was significantly higher (P < 0.05) for contralateral input than for ipsilateral input (6.9 ± 2.1 V and 2.9 ± 1.3 V, respectively, n = 6).
CHARACTERISTICS OF SOMATIC FIELDS FOR STT NEURONS.
Seven of 10 STT neurons were examined for characteristics of excitatory somatic receptive fields. Five STT neurons were excited only by noxious pinch (HT/HTi) and two cells were WDR. The excitatory somatic receptive fields for six of seven STT cells encompassed the chest and upper arm (see Fig. 6D). The somatic field for one WDR cell was located primarily on the forearm and did not include the chest. Location of somatic fields and effects of somatic stimuli on T3-T4 STT neurons agreed with previous data from this laboratory (Ammons et al. 1983 Responses of cuneothalamic and gracilothalamic cells to ipsilateral cardiopulmonary input
Stimulation of ipsilateral cardiopulmonary sympathetic afferents increased activity of cuneothalamic neurons. Lesion sites for most neurons that responded to cardiopulmonary input were near the outer edges of the cuneate nucleus and many were in the dorsomedial region. This finding is consistent with anatomic studies in rats and primates. Cardiopulmonary afferent fibers enter upper thoracic segments of the spinal cord (Vance and Bowker 1983 Responses of STT cells to ipsilateral cardiopulmonary input
Ipsilateral cardiopulmonary input increased activity of 10 of 10 T3-T4 STT neurons. Responses of this small population of STT neurons were similar to responses recorded in previous studies from this laboratory using either tungsten or glass microelectrodes (Ammons et al. 1983 Comparison of cuneothalamic and STT responses to cardiopulmonary input
Ipsilateral cardiopulmonary sympathetic input excited about 50% of cuneothalamic neurons and 100% of T2-T4 STT neurons. However, this apparent difference diminished when somatic field locations for these neurons were considered. About 65% of cuneothalamic neurons examined had somatic fields confined to the hand or distal forearm, and these cells were significantly less likely to respond to stimulating the stellate ganglion than cells with proximal somatic fields. This organization of viscerosomatic convergence is similar to that described for C3-T6 STT neurons (Hobbs et al. 1992 Potential implications for cardiac nociception
Referred pain associated with ischemic episodes of the heart has been explained as convergence of nociceptive cardiac and somatic inputs on the same STT neurons (Foreman 1993
INTRODUCTION
Abstract
Introduction
Methods
Results
Discussion
References
- and C-fiber cardiopulmonary sympathetic afferents excites thoracic and cervical spinothalamic tract (STT) neurons that receive convergent input from noxious stimulation of proximal somatic fields (Blair et al. 1981
; Chandler et al. 1996a; Hobbs et al. 1992). These findings support the classic convergence-projection theory for a neural mechanism underlying referred visceral pain (Ruch 1961) and are consistent with the idea that nociceptive visceral spinal inputs travel to the thalamus via the STT.
,b, 1997; Apkarian et al. 1995; Berkley and Hubscher 1995). In female rats, gentle or noxious stimulation of reproductive organs either excites or inhibits nearly one-half of gracile nucleus neurons examined; spinal lesions show that excitatory visceral input to gracile neurons travels in dorsal columns or dorsolateral funiculi (Berkley and Hubscher 1995). Noxious colorectal distension also excites rat gracile neurons; lesions of the medial dorsal column and blockade of lumbosacral synapses provide evidence that pelvic nerve input reaches gracile neurons via the postsynaptic dorsal column pathway (Al-Chaer et al. 1996a).
), influenced us to address the relative roles of neurons in the dorsal column/medial lemniscal system and the STT system for transmission of inputs originating from thoracic organs. Cardiopulmonary afferent fibers enter the spinal cord via dorsal root ganglia of T2-T6 segments (Vance and Bowker 1983), and cutaneous inputs from upper thoracic dorsal roots terminate primarily in the cuneate nucleus (Shriver et al. 1968). Therefore our objective was to examine responses of cuneothalamic neurons. Some gracilothalamic neurons also were included to determine if their activity was affected by stimulation of cardiopulmonary afferents. Effects of cardiopulmonary sympathetic afferent input on upper thoracic STT neurons have been reported extensively from this laboratory (Ammons et al. 1983-1985; Blair et al. 1981; Hobbs et al. 1992). However, we considered it important to determine responses of a small group of T3-T4 STT neurons in this study instead of relying exclusively on past results for comparisons.
).
METHODS
Abstract
Introduction
Methods
Results
Discussion
References
-chloralose (40-60 mg/kg iv). Monkeys were ventilated artificially via a tracheal cannula and paralyzed with pancuronium bromide (0.08-0.1 mg/kg iv). Anesthesia and muscle paralysis were maintained during the experiments with constant infusion of pentobarbital sodium (2-4 mg·kg
1·h1) and pancuronium (0.15-0.2 mg·kg1·h1). Anesthesia level was regulated by observing blood pressure and pupil diameter. End expiratory CO2 was maintained at 4-5%, and core body temperature was maintained at 37 ± 1°C with a servo-controlled heat lamp.
).
25 discharges were evoked; threshold stimulus intensity was determined for most neurons. To calculate evoked discharge rate, control activity was measured for the same number of bins that comprised the histogram peak of evoked impulses and was subtracted from total impulses of the histogram peak; evoked discharge rate was divided by the number of sweeps to determine impulses/stimulus. When a rate histogram program (1-s bins) was used, a neuron was considered excited if discharge rate increased 20% above control activity at 10- or 20-Hz stimulus (Hobbs et al. 1992). Changes in cell activity (imp/s) were calculated by subtracting the mean of 10 s of control activity from the mean of 10 s of activity recorded during stimulation.
2 or Fisher's exact test were used to determine if characteristics of a cell were related. Statistical significance was established as P < 0.05.
RESULTS
Abstract
Introduction
Methods
Results
Discussion
References
View larger version (30K):
[in a new window]
FIG. 1.
Diagrams of recording and stimulation lesion sites (drawings based on Biedenbach 1972 and Szabo and Cowan 1984, respectively). A: 
, neurons that were excited by stimulating the ipsilateral stellate ganglion; 
, neurons that did not respond to cardiopulmonary input; 
, recording sites that were extrapolated from reference to another lesioned site; 
, neuron inhibited by ipsilateral cardiopulmonary input. C, cuneate nucleus; E, external cuneate nucleus; G, gracilis nucleus. B: and , antidromic stimulation sites in left and right thalamus, respectively, for cuneothalamic or gracilothalamic neurons; 
, and , antidromic stimulation sites in left and right thalamus, respectively, for STT neurons. CL, nucleus centralis lateralis; CM, centrum medianum; Hb, habenula; LP, lateral posterior nucleus; MD dorsomedial nucleus; Pf, parafascicular nucleus; Pul.o, oral pulvinar nucleus; VLc, caudal ventrolateral nucleus; VLps, ventrolateral nucleus, pars postrema; VPI ventroposteroinferior nucleus; VPLc, caudal ventroposterolateral nucleus; VPLo, oral ventroposterolateral nucleus; VPM, ventroposteromedial nucleus; VPMpc, parvocellular ventroposteromedial nucleus.
and ). No cuneothalamic neurons were inhibited, and 22 cuneothalamic neurons did not respond to ipsilateral cardiopulmonary input. Effects of stimulating the contralateral stellate ganglion were examined on 28 cells in nine monkeys. One cell, which was excited by stimulation of ipsilateral cardiopulmonary input, was inhibited by stimulation of the contralateral stellate ganglion. Activity of the other cuneothalamic cells tested was not affected by contralateral cardiopulmonary input.
3.2 ms) peak of evoked impulses developed for 13 of these cells. Discharge rate for one cell increased with 1-Hz stimulus, but a stimulus-locked peak of evoked impulses did not develop. Mean values for number of impulses/stimulus, threshold stimulus intensity, latency to the first bin of evoked impulses, and duration of the contiguous impulses evoked by stimulating the stellate ganglion at 1 Hz, 0.1 ms, 33 V (n = 10) or 17-28 V (n = 3) are shown in the first row of Table 1. Figure 2 shows typical responses for a cuneothalamic neuron. Collision of the antidromic spike stimulated from VPL thalamus with the orthodromic impulses evoked by stimulating the stellate ganglion are shown in Fig. 2A. Figure 2B shows the histogram with 1-ms bins. Because the latency to the evoked activity peak was so short, a histogram also was generated with 0.1-ms bins (Fig. 2C); this histogram revealed accumulation of two separate impulses that also were observed with the single trace (Fig. 2A). Responses to stimulating the excitatory somatic field for this neuron (Fig. 2E) are shown in Fig. 2D. Effects of varying stimulus intensities at 1 Hz are shown in Fig. 3A for seven individual cuneothalamic neurons. Mean response (heavy line) increased as stimulus intensity increased.
View this table:
TABLE 1.
Comparison of excitatory responses to cardiopulmonary sympathetic input on cells
projecting to VPL thalamus
View larger version (18K):
[in a new window]
FIG. 2.
Responses of a cuneothalamic neuron to cardiopulmonary and somatic stimulation. A, top: single sweep of 2 impulses ( 
) evoked by stimulating left stellate ganglion (1 Hz, 33 V, 0.1 ms) and antidromic potential (*) produced by stimulating right VPL thalamus. Bottom: collision of the antidromic potential with an orthodromic potential. B: peristimulus histogram (50 sweeps, 1-ms bins) of response to stimulation of the stellate ganglion. , stimulus artifact. C: same as B except 0.1-ms bins. , stimulus artifact. D: effects of stimulating somatic receptive field. BR, brush hair; PI, pinch skin and underlying muscle. E: black area shows location of wide dynamic range (WDR) somatic field on left hand.
View larger version (19K):
[in a new window]
FIG. 3.
A: responses to stimulating ipsilateral stellate ganglion at varying stimulus intensities for 7 cuneothalamic neurons. B: responses to stimulating ipsilateral stellate ganglion at varying stimulus intensities for 7 STT neurons.
10-Hz STIMULUS FREQUENCY.
Effects of stimulating the ipsilateral stellate ganglion at 10 Hz were examined on 34 cuneothalamic neurons. All neurons that were classified as not responding to cardiopulmonary input (n = 22) were tested at 10- or 20-Hz stimulus frequency and maximal (33 V) stimulus intensity. Three of 12 neurons that were excited at 10-Hz frequency stimulus were not examined at 1 Hz, including one cell that was damaged before computer documentation of the increased discharge rate that was observed on the oscilloscope. Nine of 12 cuneothalamic neurons that increased discharge rate to 10-Hz stimulus of cardiopulmonary input also were excited at 1 Hz (see preceding text; Table 1). In all, 10-Hz (n = 10) or 20-Hz (n = 1) stimulation of the stellate ganglion increased cell discharges from mean control activity of 8.3 ± 3.0 imp/s to 29.9 ± 9.0 imp/s.
View larger version (16K):
[in a new window]
FIG. 4.
Sizes and locations of excitatory somatic fields for cuneothalamic neurons. A: representative somatic fields for cuneothalamic neurons that were excited by stimulating the ipsilateral stellate ganglion. B: representative somatic fields for cuneothalamic neurons that did not respond to stimulating the ipsilateral stellate ganglion.
View this table:
TABLE 2.
Relationship of responses to cardiopulmonary sympathetic afferent input and locations of somatic fields
View this table:
TABLE 3.
Relationship of responses to cardiopulmonary sympathetic afferent input and somatic receptive field characteristics
View larger version (15K):
[in a new window]
FIG. 5.
Effect of dorsal transection at C5 on responses to cardiopulmonary and somatic stimulation. A: rate histograms (1-s bins) of responses to stimulating stellate ganglion (10 Hz, 9 V, 0.1 ms) before and after cord cut. CPSA, cardiopulmonary sympathetic afferent. B: peristimulus histograms (50 sweeps, 0.1-ms bins) of responses to stimulating stellate ganglion (1 Hz, 9 V, 0.1 ms) before and after cord cut. Arrows mark time of stimulus. C: rate histograms (1-s bins) of responses to brushing upper arm/chest before and after cord cut. D: black area shows location of ipsilateral spinal cord lesion at C5.
-1985; Blair et al. 1981; Hobbs et al. 1992), extracellular unit activity was recorded for 10 STT neurons located in right (n = 5) or left (n = 5) T3-T4 spinal segments. Neuronal responses observed in the current study agreed with previous reports from this laboratory. All STT neurons were antidromically activated from contralateral VPL or lateral posterior thalamus (Fig. 1B). Lesion sites for two STT neurons were in lamina I, two were in lamina IV, five were in lamina V, and one cell location was not lesioned.
View larger version (19K):
[in a new window]
FIG. 6.
Comparison of responses to cardiopulmonary sympathetic input and somatic fields. A: histogram (50 sweeps, 1-ms bins) of response to stimulating ipsilateral stellate ganglion (1 Hz, 33 V, 0.1 ms) for a cuneothalamic neuron. Arrow, stimulus artifact. B: histogram of response of a T3 STT neuron to stellate ganglion stimulation (1 Hz, 33 V, 0.1 ms). Arrow, stimulus artifact. C: black area shows excitatory somatic receptive field (LT) for the cuneothalamic neuron. D: somatic receptive field (HTi) for the STT neuron. Inhibitory response to brush (striped and black areas) included larger region than excitatory response to pinch (black area).
, 1984a; Blair et al. 1981).
DISCUSSION
Abstract
Introduction
Methods
Results
Discussion
References
), and most termination sites from upper thoracic dorsal roots are in dorsomedial regions of the cuneate nucleus, whereas primary afferent fibers from the hand project primarily to the central core (Florence et al. 1989; Shriver et al. 1968). Moreover, postsynaptic dorsal column fibers labeled from thoracic segments are found in medial cuneate and lateral gracile nuclei, whereas fibers labeled from the cervical enlargement are found throughout the cuneate nucleus (Cliffer and Geisler 1989; Cliffer and Willis 1994).
report that lesions of the dorsolateral funiculus are needed to eliminate inputs from the cervix to rat gracile neurons, although dorsal column lesions eliminate or alter uterine inputs. Al-Chaer el al. (1996a,b) report that midline dorsal column lesions, most likely of postsynaptic dorsal column fibers, eliminate effects of colorectal distension on gracile and VPL thalamic neurons. Midline lesions were not needed to eliminate effects of cardiopulmonary input on cuneothalamic neurons; moreover, a midline lesion for one cell did not alter responses. Somatotopic organization of dorsal column fibers (Brodal 1981) supports our conclusion that cardiopulmonary input travels in lateral dorsal pathways.
). Furthermore, inhibitory and excitatory effects of visceral inputs are about equal for cells in dorsal column nuclei of rats and cats (Berkley and Hubscher 1995; Blair and Thompson 1995). Species differences could be a factor. Another possibility is that some gracile or cuneate cells in other studies were either interneurons or projected to motor nuclei (Berkley et al. 1986), whereas all primate neurons in the current study projected to contralateral VPL thalamus and thus might have a more uniform response to visceral inputs.
-1985; Blair et al. 1981; Hobbs et al. 1992). For example, 32 of 32 T3-T5 STT neurons examined by Blair et al. (1981) and 85 of 85 T1-T5 STT neurons examined by Ammons et al. (1984a) were excited by stimulation of A- and C-fiber afferents coursing in the stellate ganglion. In those studies, 40% of cells received both A- and C-fiber inputs; Blair et al. (1981) also found that 10% of cells received C-fiber input only. In the present study, 60% of cells were activated at latencies corresponding to both A- and C-fiber inputs. Somatic receptive fields were found primarily on the upper arm and chest in this and previous studies of upper thoracic STT cells. Classification of somatic inputs were similar. In this study, 29% of cells were WDR and 71% were HT/HTi. Other studies classified 26-41% of cells as WDR and 58-74% as HT/HTi (Ammons et al. 1984a; Blair et al. 1981). Laminar locations of recording sites and VPL locations of antidromic stimulating electrodes in the current study were similar to those in previous studies (Ammons et al. 1983, 1984a; Blair et al. 1981; Hobbs et al. 1992).
). Regression analyses show that excitatory effects of stimulating cardiopulmonary afferent fibers are positively correlated with occurrence of somatic fields on the chest and proximal arm. Conversely, STT neurons with somatic fields located on the hand and forearm are less likely to receive convergent cardiopulmonary input.
; Blair et al. 1981). Furthermore, the pattern of evoked impulses observed in peristimulus histograms was markedly different for cuneothalamic neurons compared with STT neurons. Blair et al. (1981) notes that STT cells respond to stimulation of cardiopulmonary sympathetic afferents with a burst discharge rather than with a single spike, and this observation was confirmed in subsequent studies. In contrast, the duration of evoked impulses for cuneothalamic neurons was significantly shorter than for the first peak of impulses evoked in STT neurons.
). This difference cannot be due to the distance from spinal entry of cardiopulmonary afferent fibers to cell bodies, which is much longer for dorsal column nuclei compared with thoracic spinal cord. It is possible that more interneurons were located between primary afferent fibers and STT cells. Evidence in rats indicates that most pelvic inputs to the gracile nucleus are carried in the postsynaptic dorsal column (Al-Chaer et al. 1996a), but this pathway could exist with just one other synapse between primary afferents and neurons in dorsal column nuclei. Another possibility is that large A- fibers, which have faster conduction velocities, activated cuneothalamic neurons and STT cells were activated by small A and C fibers.
). Unilateral activation of cuneothalamic neurons, combined with short latency to response, is consistent with rapid transport of information to cuneate cell bodies and then to thalamus.
). In the present study, 9 of 24 (38%) LT cuneothalamic neurons were excited by stimulating the stellate ganglion compared with 5 of 7 (71%) WDR/HT neurons. This difference in selective convergence was not significant, but LT cuneothalamic neurons likely send a different code of activity to VPL thalamus than STT cells because few STT cells are LT (Ammons et al. 1984a, 1985; Blair et al. 1981). Differences in discharge pattern of cardiopulmonary input to cells of these two pathways support our speculation that visceral information received in VPL thalamus from LT cuneothalamic neurons provides a different message than information received from STT cells.
). This study in primates confirmed previous reports supporting this hypothesis (Ammons et al. 1985; Blair et al. 1981; Hobbs et al. 1992) and also demonstrated that cardiopulmonary sympathetic input excites cuneothalamic neurons. Average threshold stimulus intensities were not different, and increasing stimulus intensity generally increased responses in each population of neurons. These similarities indicate that both ventrolateral and dorsal pathways transmitted nociceptive cardiopulmonary information. Important differences in responses (see preceding text) indicate that integration of this information was not identical for cuneothalamic and STT neurons.
,b, 1997; Apkarian et al. 1995). Based on lesions of spinal pathways or the gracilis nucleus in rats, Al-Chaer et al. (1996b, 1997) concluded that most colorectal information travels in dorsal column pathways to gracilis nucleus and then to VPL thalamus. Apkarian et al. (1995) proposed that nociceptive visceral inputs travel in dorsal column or vagal pathways. This conclusion is based on their finding that distension of several visceral organs excites or inhibits widely distributed LT neurons in monkey thalamus (Brüggemann et al. 1994). We find that urinary bladder input to monkey VPL neurons with proximal somatic fields on the lower body is similar to STT organization, i.e., cells excited by urinary bladder distension are excited by nociceptive somatic input (Chandler et al. 1992). Data in monkey thalamus appear contradictory, but Brüggemann et al. (1994) also reported that VPL neurons with nociceptive somatic input from the lower body have the same probability of receiving excitatory urinary bladder input as neurons in our study (Chandler et al. 1992).
proposed that sustained noxious stimulation of a visceral organ sensitizes STT pathways to activate contiguous VPL neurons that also respond to nociceptive inputs from specific somatic regions. Lack of C-fiber input to cuneothalamic neurons supports this idea. Blair et al. (1984) showed that STT neurons with C-fiber input are more likely to respond to occlusion of the left coronary artery than STT neurons that receive input from A fibers only. Thus cuneothalamic neurons might be less likely than STT neurons to be activated during severe coronary ischemia, a relevant clinical stimulus that often produces referred pain. One possibility is that early and/or moderately noxious stimuli activate A-fibers that reach dorsal column pathways. This rapid input perhaps has an arousal function in the thalamus, similar to that proposed for upper cervical STT neurons (Smith et al. 1991). Extended and/or intensive noxious stimuli might then activate A- and C-fiber inputs that reach STT neurons and result in visceral pain referred to specific somatic regions.
also proposed that visceral nociception depends on balance of information from various spinal and central pathways. The concept that nociceptive information from visceral organs is transmitted in multiple pathways, rather than a single pathway, provides logical explanations for inconsistent results obtained clinically with spinal lesions (Berkley and Hubscher 1995). For example, unilateral anterolateral cordotomy has been used successfully to treat visceral pain (White and Sweet 1969; White et al. 1950) but is considered ineffective for visceral pain by others (Ranson and Clark 1959). Recent clinical cases provide evidence that dorsal pathways transmit nociceptive pelvic inputs; in several patients with colon cancer, midline dorsal column lesions alleviated intractable pain (Hirshberg et al. 1996).
- and C-fiber input to STT neurons also produced rapid A-fiber input to cuneothalamic neurons. Multiple ascending pathways conduct visceral information, but the content of coded visceral messages and the somatic information that each pathway contributes to processing in the thalamus must be evaluated to understand their respective roles in sculpting visceral sensation. In addition, modulatory descending influences on these pathways (Ammons et al. 1984b, 1986; Jundi et al. 1982) are likely factors that contribute to the transmission and integration of nociceptive information that ultimately results in the sensation of referred cardiac pain.
| |
ACKNOWLEDGEMENTS |
|---|
The authors thank D. Holston for excellent technical assistance and preparation of the figures.
This work was supported by National Heart, Lung, and Blood Institute Grants HL-22732 and HL-52986.
| |
FOOTNOTES |
|---|
Address for reprint requests: M. J. Chandler, Dept. of Physiology, University of Oklahoma HSC, PO Box 26901, Oklahoma City, OK 73190.
Received 18 September 1997; accepted in final form 28 April 1998.
| |
REFERENCES |
|---|
|
Abstract
Introduction
Methods
Results
Discussion
References
|
|---|
This article has been cited by other articles:
|
|
 |
|
  M. A. Gray, P. Taggart, P. M. Sutton, D. Groves, D. R. Holdright, D. Bradbury, D. Brull, and H. D. Critchley A cortical potential reflecting cardiac function PNAS, April 17, 2007; 104(16): 6818 - 6823. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H.-L. Pan and S.-R. Chen Myocardial Ischemia Recruits Mechanically Insensitive Cardiac Sympathetic Afferents in Cats J Neurophysiol, February 1, 2002; 87(2): 660 - 668. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Moore and M. Chester Neuromodulation for chronic refractory angina Br. Med. Bull., October 1, 2001; 59(1): 269 - 278. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. D. Al-Chaer, Y. Feng, and W. D. Willis Comparative Study of Viscerosomatic Input Onto Postsynaptic Dorsal Column and Spinothalamic Tract Neurons in the Primate J Neurophysiol, October 1, 1999; 82(4): 1876 - 1882. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H.-L. Pan, J. C Longhurst, J. C Eisenach, and S.-R. Chen Role of protons in activation of cardiac sympathetic C-fibre afferents during ischaemia in cats J. Physiol., August 1, 1999; 518(3): 857 - 866. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
