Lumbar Chain Research Articles

4. Painful diabetic polyneuropathy.

Pain as a symptom of diabetic polyneuropathy (DPN) significantly lowers quality of life, increases mortality and is the main reason for patients with diabetes to seek medical attention. The number of people suffering from painful diabetic polyneuropathy (PDPN) has increased significantly over the past decades. The literature on the diagnosis and treatment of diabetic polyneuropathy was retrieved and summarized. The etiology of PDPN is complex, with primary damage to peripheral nociceptors and altered spinal and supra-spinal modulation. To achieve better patient outcomes, the mode of diagnosis and treatment of PDPN evolves toward more precise pain-phenotyping and genotyping based on patient-specific characteristics, new diagnostic tools, and prior response to pharmacological treatments. According to the Toronto Diabetic Neuropathy Expert Group, a presumptive diagnosis of "probable PDPN" is sufficient to initiate treatment. Proper control of plasma glucose levels, and prevention of risk factors are essential in the treatment of PDPN. Mechanism-based pharmacological treatment should be initiated as early as possible. If symptomatic pharmacologic treatment fails, spinal cord stimulation (SCS) should be considered. In isolated cases, where symptomatic pharmacologic treatment and SCS are unsuccessful or cannot be used, sympathetic lumbar chain neurolysis and/or radiofrequency ablation (SLCN/SLCRF), dorsal root ganglion stimulation (DRGs) or posterior tibial nerve stimulation (PTNS) may be considered. However, it is recommended that these treatments be applied only in a study setting in a center of expertise. The diagnosis of PDPN evolves toward pheno-and genotyping and treatment should be mechanism-based.

Isolation and Electrophysiology of Murine Sympathetic Postganglionic Neurons in the Thoracic Paravertebral Ganglia.

The thoracic paravertebral sympathetic chain postganglionic neurons (tSPNs) represent the predominant sympathetic control of vascular function in the trunk and upper extremities. tSPNs cluster to form ganglia linked by an interganglionic nerve and receive multisegmental convergent and divergent synaptic input from cholinergic sympathetic preganglionic neurons of the spinal cord (Blackman and Purves, 1969; Lichtman et al., 1980 ). Studies in the past have focused on cervical and lumbar chain ganglia in multiple species, but few have examined the thoracic chain ganglia, whose location and diminutive size make them less conducive to experimentation. Seminal studies on the integrative properties of preganglionic axonal projections onto tSPNs were performed in guinea pig (Blackman and Purves, 1969; Lichtman et al., 1980 ), but as mice have become the accepted mammalian genetic model organism, there is need to reproduce and expand on these studies in this smaller model. We describe an ex vivo approach that enables electrophysiological, calcium imaging, and optogenetic assessment of convergence, divergence, and studies on pre- to postganglionic synaptic transmission, as well as whole-cell recordings from individual tSPNs. Preganglionic axonal connections from intact ventral roots and interganglionic nerves across multiple segments can be stimulated to evoke compound action potential responses in individual thoracic ganglia as recorded with suction electrodes. Chemical block of synaptic transmission differentiates spiking of preganglionic axons from synaptically-recruited tSPNs. Further dissection, including removal of the sympathetic chain, enables whole-cell patch clamp recordings from individual tSPNs for characterization of cellular and synaptic properties.

Early experience with endoscopic lumbar sympathectomy for plantar hyperhidrosis.

We describe our endoscopic lumbar sympathectomy technique and our early experience using it to treat plantar hyperhidrosis. We reviewed 20 lumbar sympathectomies performed in our vascular unit for plantar hyperhidrosis in 10 patients from 2011 and 2014. Demographics and outcomes were analyzed and a review of the literature conducted. All procedures were carried out endoscopically with no intraoperative or postoperative morbidity. Plantar anhidrosis was achieved in all the patients, although two patients (20%) suffered a relapse. Unwanted side-effects occurred in the form of compensatory sweating in three patients (30%) and post-sympathectomy neuralgia in two patients (20%). None of the patients experienced sexual dysfunction. Management of plantar hyperhidrosis may be based upon a therapeutic ladder starting with conservative measures and working up to surgery depending on the severity of the disease. Minimally invasive (endoscopic) sympathectomy for the thoracic chain is well established, but minimally invasive sympathectomy for the lumbar chain is a relatively new technique. Endoscopic lumbar sympathectomy provides an effective, minimally invasive method of surgical management, but long-term data are lacking.

The effect of short‐term exercise training on the skeletal muscle vascular response to sympathetic stimulation

The effect of 4 weeks of exercise training (ET) on skeletal muscle sympathetic vasoconstriction was investigated. It was hypothesized that vascular responsiveness to sympathetic stimulation would be increased following ET. Sprague-dawley rats were randomly assigned to sedentary (S: n=12), mild- (M: 20m.min−15% grade; n=13) or heavy-intensity (H: 40m.min−15% grade, n=8) ET groups. Rats were trained 5d.week−1 for 4 weeks with total training volume matched between ET groups. Rats were anesthetised and instrumented with indwelling catheters and a femoral artery flow probe. The vascular response to lumbar sympathetic chain stimulation delivered continuously at 2Hz or in patterns at 20 or 40Hz was determined. The change in femoral vascular resistance (FVR) was calculated and expressed in arbitrary resistance units (RU). Basal blood flow (S: 4±1, M: 3±1, H: 4±1 mL·min−1), MAP (S: 100±14, M: 90±14, H: 89±15 mmHg) and FVR (S: 33±15, M: 27±5, H: 25±7 mmHg.mL·min−1) were similar between groups (p>0.05). Lumbar chain stimulations produced a smaller (p<0.05) increase in FVR at all frequencies in S (2Hz: 7±3; 20Hz: 9±4; 40Hz: 11±5 RU) compared to M (2Hz: 14±4; 20Hz: 18±4; 40Hz: 18±6 RU) and H (2Hz: 21±4; 20Hz: 27±6; 40Hz: 28±4 RU). These data demonstrate that following ET, the vascular response to sympathetic stimulation was up-regulated in a manner dependent on the intensity of training stimulus. NSERC Canada.

A decerebrate, artificially-perfused in situ preparation of rat: Utility for the study of autonomic and nociceptive processing

By extending established technology, we have developed a relatively simple in situ preparation from juvenile rat that overcomes some technical restrictions present in vivo (e.g. need for anaesthesia, mechanical instability for intracellular recording and control over the extracellular milieu). The in situ preparation is decerebrate and artificially perfused via the left ventricle with a colloid containing solution. It exhibits an eupneic pattern of respiratory motor activity and demonstrates numerous somatic and visceral reflexes including those evoked by stimulation of the tail, hindlimbs, bladder, baroreceptors and peripheral chemoreceptors. We have employed this preparation to allow recordings from multiple sympathetic motor outflows such as thoracic and lumbar chain, inferior cardiac, splanchnic, renal and adrenal nerves. We show that the sympathetic motor discharge shows strong respiratory modulation and exhibits pronounced reflex modulation indicating intact communication between the periphery, the brainstem and the spinal cord. Further, we have made extracellular and whole cell recordings from neurones in the spinal cord, demonstrating good mechanical stability. The decerebrate, artificially-perfused rat (DAPR) provides a powerful methodology with which to study peripheral and central control of the autonomic nervous system with many of the benefits of an in vitro environment.

Effect of 5-hydroxytryptamine on neurogenic vasoconstriction in the isolated, autoperfused hindquarters of the rat

1. In the present study, we analysed the effect of different doses of 5-hydroxytryptamine (5-HT; intravenous infusions of 0.001-40 microg/kg per min) in the autoperfused hindquarters of the rat subjected to electrical stimulation (frequencies of 0.5-20 Hz) of the lumbar chains, investigating the relationship between the adrenergic and serotonergic systems in this vascular bed. 2. Because we observed that 5-HT inhibited the increases in perfusion pressure induced by electrical stimulation of the lumbar chains, we used different agonists and antagonists to analyse the mechanism of action of 5-HT. 3. The effect of 5-HT was inhibited by methiothepin (a non-specific 5-HT receptor antagonist), but not by ritanserin (a selective 5-HT2 receptor antagonist). The effects of 5-HT were mimicked by 5-carboxamidotryptamine (a 5-HT1 receptor agonist) and L-694 247 (a selective 5-HT1D receptor agonist), but not by 8-hydroxy-2-dipropylaminotetralin (a 5-HT1A receptor agonist), CGS-12066B (a 5-HT1B receptor agonist), alpha-methyl-5-HT (a 5-HT2 receptor agonist), 1-(3-chlorophenyl) piperazine (a 5-HT2C receptor agonist) or 1-phenylbiguanide (a 5-HT3 receptor agonist). The selective 5-HT1D/1B receptor antagonist BRL 15572 inhibited the effect of the agonist L-694 247. 4. Our data suggest that 5-HT inhibits the increases in perfusion pressure induced by the electrical stimulation of the lumbar chains, acting on presynaptic 5-HT1D receptors and decreasing the release of noradrenaline from the sympathetic nerves in the hindquarter vascular bed of the rat.

Variations in the number and position of human lumbar sympathetic ganglia and rami communicantes.

The purpose of this study is to provide precise anatomical and statistical information about the number and location of lumbar sympathetic ganglia and the number and length of the related rami communicantes, and to consider the neurological pathway for nociception from the low back. Three-hundred ninety-three ganglia and 719 rami communicantes from 50 human cadavers were identified by gross dissection. The number of ganglia in a single lumbar chain ranged from 2 to 6, the mean being 3.9. The mean lengths of rami connected to the 1st and 5th lumbar spinal nerves, respectively, were significantly longer and shorter than those connected to the 2nd, 3rd, and 4th nerves. The lumbar sympathetic ganglia and rami communicantes were not distributed segmentally. The present results may assist in understanding the nociceptive pathway from the low back.

Neurokinin-1 receptor localisation in guinea pig autonomic ganglia

We have used multiple-labelling immunohistochemistry and confocal microscopy to determine the distribution of immunoreactivity to the tachykinin neurokinin-1 (NK(1)) receptors in guinea pig sympathetic ganglia. Although nerve fibres containing immunoreactivity to substance P were common in all ganglia except the superior cervical ganglia, most neurons expressing NK(1) receptor immunoreactivity were not closely surrounded by pericellular baskets of substance P-immunoreactive boutons. Conversely, many neurons surrounded by baskets of substance P-immunoreactive boutons lacked NK(1) immunoreactivity. In the coeliac and inferior mesenteric ganglia, NK(1) receptor expression was restricted almost entirely to noradrenergic neurons that contained somatostatin immunoreactivity and projected to the enteric plexuses. In the lumbar chain and paracervical ganglia, NK(1) immunoreactivity was expressed by nonnoradrenergic vasodilator neurons containing immunoreactivity to vasoactive intestinal peptide. Taken together, our results show that sympathetic neurons in different functional pathways express NK(1) receptor immunoreactivity. However, the neurons that could respond to endogenously released substance P through NK(1) receptors may be distant from presynaptic release sites. These observations suggest that, in sympathetic ganglia, substance P may modulate ganglionic transmission through heterosynaptic actions on NK(1) receptors.

Differential distribution of substance P binding sites in guinea-pig sympathetic ganglia

We have used a combination of autoradiographic and immunohistochemical techniques to investigate the distribution of binding sites for substance P in relation to the distribution of substance P-immunoreactive nerve fibres and specific functional populations of neurons in the sympathetic ganglia of guinea-pigs. There was considerable heterogeneity in the density of binding sites for Bolton Hunter labelled 125 I−substance P (BHSP). Binding sites were more dense in the prevertebral ganglia, such as the coeliac and inferior mesenteric ganglia, than in the paravertebral ganglia, such as the superior cervical or lumbar chain ganglia. The binding sites tended to be clumped within the ganglia. Within the prevertebral ganglia, they were associated predominantly with neurons projecting to the enteric plexuses. Many of these neurons contained somatostatin immunoreactivity. In the lumbar sympathetic chain ganglia, there was a weak association of binding sites with neurons containing immunoreactivity to vasoactive intestinal peptide. Overall, the density of binding sites matched the density of nerve fibres containing immunoreactivity to substance P in different ganglia. However, within particular ganglia, there was little, if any, correlation between the distribution of binding sites and nerve fibres containing substance P. Most of the binding sites in the ganglia had the pharmacological characteristics of NK 1 receptors. Our results show that there is considerable heterogeneity in the expression of NK 1 receptors in the sympathetic ganglia of guinea-pigs. However, given the relatively poor spatial correlation between the distribution of binding sites and potential sites of substance P release from intraganglionic nerve fibres, we suggest that substance P may diffuse for relatively large distances through the ganglia, with actions only on those neurons selectively expressing NK 1 receptors.

Nociceptin and its receptor in guinea-pig sympathetic ganglia

The occurrence and distribution of nociceptin, a 17-amino acid peptide with structural similarity to dynorphin A, and its receptor, opioid receptor-like-1 (ORL1) receptor, were investigated in the guinea-pig sympathetic nervous system by means of immunohistochemistry and reverse transcription-polymerase chain reaction (RT-PCR), respectively. Immunofluorescence revealed varicose nociceptin-immunoreactive axons and some paraganglionic cells in prevertebral (coeliaco-superior mesenteric, inferior mesenteric), but not in paravertebral (superior cervical, stellate, lumbar chain) sympathetic ganglia. Messenger RNA for the ORL1 receptor, however, was detected by RT-PCR in both para- and prevertebral ganglia. The findings suggest participation of the nociceptin/ORL1 receptor signalling pathway in processing of information within prevertebral ganglia, and a general responsiveness of sympathetic neurons to nociceptin.

Neurogenically derived nitrosyl factors mediate sympathetic vasodilation in the hindlimb of the rat.

Skeletal muscle vasculature of the hindlimb is innervated by a sympathetic noncholinergic vasodilator system. The aim of this study was to determine whether this vasodilator system may represent postganglionic lumbar sympathetic neurons that synthesize and release nitric oxide (NO) or related NO-containing factors. We examined whether NO synthase (NOS)-positive postganglionic lumbar nerves innervate the hindlimb vasculature of the rat and whether the hindlimb vasodilation produced by electrical stimulation of the lumbar sympathetic chain of anesthetized rats is reduced after the systemic administration of the specific inhibitor of neuronal NOS 7-nitroindazole (7-NI). Subpopulations of lumbar sympathetic cell bodies stained intensely for NOS. Postganglionic fibers and varicosities within the iliac and femoral arteries also stained for NOS. Double ligation of the lumbar chain demonstrated that NOS was transported from the cell bodies toward the peripheral terminals. Low-intensity electrical stimulation of the lumbar chain produced a pronounced hindlimb vasodilation that was markedly diminished by pretreatment with 7-NI (45 mg/kg i.v.). In contrast, the vasodilator potency of acetylcholine and S-nitrosocysteine were augmented by 7-NI. These results suggest that postganglionic lumbar sympathetic neurons may synthesize and release NO-containing factors.

Locations and chemistries of sympathetic nerve cells that project to the gastrointestinal tract and spleen.

Retrograde tracing was used to determine the locations of sympathetic nerve cells whose axons project to the stomach, small intestine, caecum, proximal colon, distal colon and spleen of the guinea-pig. Projections from prevertebral ganglia were organotopically arranged within and between ganglia. The cranially located coeliac ganglion provided the major input to proximal gut regions; the distal gut received more caudal input, from superior and inferior mesenteric and the hypogastric nerve ganglia. Nevertheless, minor proportions of the innervation of some target organs arose from other than the closest ganglion and the caecum had input from each of the coeliac, superior mesenteric and inferior mesenteric ganglia. Topography within a ganglion was best defined in the coeliac, in which nerve cells whose axons projected to the spleen, stomach and duodenum were preferentially laterally located, whereas most of those projecting to the proximal colon were medial. Fewer neurons projected from paravertebral--compared with prevertebral--ganglia to abdominal viscera. Projections to the stomach came from all thoracic chain ganglia, those to the duodenum and spleen from lower thoracic ganglia and those to the large intestine from lumbar chain ganglia. It is suggested that the previously reported chemical topography of nerve cells in sympathetic ganglia might be secondary to their organotopic organization.

Comparison of locations and peptide content of postganglionic neurons innervating veins and arteries of the rat hindlimb

The ganglionic location of hindlimb vasoconstrictor sympathetic neurons in several species is known but the locations of neurons innervating limb arteries or veins, specifically, have not been compared and neurochemical differences between them have not been examined in detail. This study was designed to determine whether neurons innervating arteries and veins are organized as distinct populations and whether neurons innervating arteries, veins or footpads contain the same peptides. Retrograde transport of fluorescent dyes was used to identify, separately, paravertebral postganglionic neurons in the 13th thoracic to 6th lumbar (T13—L6) chain ganglia that innervate the femoral arteries, femoral veins and footpads of the rat hindlimb. The proportions of venous and arterial vasomotor neurons and footpad neurons containing neuropeptide Y- and vasoactive intestinal polypeptide-immunoreactivity (NPY-Ir, VIP-Ir) were then compared. Venous vasomotor neurons were found mostly (62%) in the L1 and L2 ganglia. The majority of arterial vasomotor neurons (81%) were distributed slightly more caudally in L1–L3. Veins and arteries were not innervated by the same cells. Footpad neurons were located mostly in L4–L6. NPY-Ir was identified in 17% of the venomotor neurons, in 94% of arterial neurons and in 24% of footpad neurons. VIP-Ir was found in 3% of the venomotor neurons, 8% of the arterial neurons and in 44% of the footpad neurons. In conclusion, hindlimb venous and arterial vasomotor neurons are anatomically distinct, are mixed randomly within the chain ganglia and differ markedly in their content of NPY, consistent with reported differences in neuromuscular transmission to arteries and veins. The most likely hindlimb postganglionic neurons to contain VIP were those innervating footpads, probably controlling sweat gland function.

Preganglionic and sensory axons in developing bullfrog sympathetic ganglia express three neuropeptides during early tadpole stages

Retrograde tracing in fixed tissue with the fluorescent carbocyanine dye, Dil, was used to identify neurons that project to paravertebral sympathetic ganglia 9 and 10 in bullfrog tadpoles. Applying Dil to ganglion 9 at stage II labelled spinal preganglionic neurons and sensory neurons in dorsal root ganglia. When examined in a stage V tadpole, the segmental boundaries of the preganglionic cell column which supply the lumbar chain ganglia were identical to that in the adult. Using immunocytochemistry, luteinizing hormone releasing hormone-like immunoreactivity, calcitonin gene-related peptide-like immunoreactivity, and substance P-like immunoreactivity were localized at stage III in axons within sympathetic ganglia 9 and 10. During subsequent stages, the density of fibers containing these peptides increased and it became easier to identify synaptic boutons in contact with postganglionic neurons. These observations demonstrate that projections to the lumbar sympathetic ganglia are already formed by the earliest tadpole stages, they are consistent with the previous physiological observation of nicotinic synapses in stage III ganglia, and they suggest that neuropeptide function may also begin during early stages.

Histamine in the developing sympathoadrenal system

The ontogenetic distribution of histamine in correlation with catecholamines in the developing rat sympathoadrenal system was analyzed by using an indirect immunohistochemical method and a specific rabbit anti-histamine antiserum. Tyrosine hydroxylase (TH) immunoreactivity was used as a marker of catecholamine synthesis. TH immunoreactivity appeared in retroperitoneal sympathetic tissues on embryonic day 12.5 (E 12.5) when it was found in cells of lumbar chain ganglia. In preaortic sympathetic tissue. TH immunoreactivity was observed on day E 13.5 and in adrenal medullae on day E 14.5. Histamine immunoreactivity was expressed in all of these tissues beginning from day E 14.5. First it was found mainly in nerve fibers, but also in some cells. During the embryonic development the number of histamine-immunoreactive cells increased in all sympathetic tissues studied. In newborn rats, histamine immunoreactivity was restricted to a subpopulation of sympathetic cells, i.e. small intensely fluorescent (SIF) cells of sympathetic ganglia, paraganglion-type cells and some adrenaline-synthesizing cells of the adrenal medulla.

Effects of axotomy on synaptic transmission and structure in frog sympathetic ganglia.

Axotomy was carried out on frog sympathetic neurons of the two last lumbar chain ganglia. At different times thereafter, synaptic transmission was analysed electrophysiologically by intracellular microelectrodes and compared with synaptic density, measured by electron microscopy in the same ganglia. For this purpose, modifications in synaptic transmission were estimated first, by the numbers of B and C sympathetic neurons exhibiting subthreshold excitatory postsynaptic potentials in response to 10 Hz orthodromic stimulation of preganglionic fibres, and second, by the amplitude and number of excitatory postsynaptic potentials occurring over 5-10 s periods of 10 Hz stimulation. By distinguishing two types of morphological relationships between the pre- and postsynaptic elements, two contact indices were defined: a synaptic index (ratio of the number of synapses encountered to the number of perikarya explored) and a simple contact index corresponding to the same type of contacts, but without any membrane differentiation. Both the electrophysiological and morphological results showed that the first effects were detectable 4 days after axotomy, and that the main alterations in synaptic transmission and density occurred at 2 weeks. In addition, while in normal ganglia the excitatory postsynaptic potentials of B and C neurons reached the threshold for action potential generation in response to 10 Hz stimulation, about 29% of the axotomized neurons had subthreshold excitatory postsynaptic potentials 1 week after section. At 2 weeks, this proportion reached 65%, and the synaptic and simple contact indices, at 90% and 60% respectively, were significantly lower than the control ganglion indices. At longer times after axotomy, there was a discrepancy between the morphological and electrophysiological results: at 1 month, the synaptic index seemed to rise as the decline in the efficacy of synaptic transmission became more marked. The amplitude of the subthreshold excitatory postsynaptic potentials recorded in B neurons was 5.5 +/- 2.8 mV (mean +/- SD, n = 18); this value was significantly lower by about 50% than that measured 1 week after axotomy. In addition, the number of excitatory postsynaptic potentials in B neurons reached an average maximum of 83 +/- 29 for 100 stimuli applied at 10 Hz. Similar results were obtained for C neurons. Two months after axotomy, the physiological and morphological parameters of synaptic efficacy began to recover and return to normal values, but had not reached them by 4 months. These observations show that some synaptic transmission remains possible, even with a much reduced number of synaptic complexes. It is suggested that after axotomy, simple contacts also might be involved in synaptic transmission.

Lumbar sympathetic ganglia in man: an electrophysiological study in vitro

A method has been developed for the removal, preservation and electrophysiological study ‘in vitro’ of sympathetic lumbar chains (L 1-L 3) from subjects undergoing lumbar ganglionectomy in the treatment of peripheral vascular diseases. Extracellular recordings from interganglionic trunks, and intracellular recordings from single sympathetic neurons, were performed. The extracellular experiments substantiated the concept, hitherto deduced from animal experiments, that the preganglionic fibres in the sympathetic lumbar chain are mainly of a descending nature. In fact, stimulation of the interganglionic trunk cranial to ganglia is always much more effective in driving ganglion neurons to fire than stimulation of the interganglionic trunk caudal to ganglia. The intracellular experiments produced a good definition of the main electrical characteristics of human sympathetic neurons. The results can be summarized as follows: the resting membrane potential ranged from 50 to 75 mV (63.4 ± 9.2 mV; 21 neurons); action potential amplitude from 62 to 93 mV (74.3 ± 8.1 mV; 27 neurons); membrane input resistance was 42.3 ± 12.6 mΩ (8 neurons) and total membrane capacitance 83.7 ± 15.3 pF (8 neurons).

Chapter 6 Differential regulation of sympathetic nerve activity by lateral and medial subregions of the rostral ventral medulla

Microinjections of lidocaine were used to examine the contributions of two subregions of the RVM, RVLM (2 mm lateral to midline) and RVMM (1 mm lateral to midline) to the maintenance of AP and SNA in urethane-anesthetized rats. Lidocaine microinjected into either site reduced AP to similar levels. Blockade of RVLM and RVMM produced a small further reduction in AP and essentially abolished neurogenic maintenance of AP. Blockade of either RVLM or RVMM elicited similar falls in RSNA. In contrast, inactivation of RVLM elicited larger falls in lumbar chain (LSNA) and splanchnic (SSNA) SNA than did inactivation of the RVMM. Combined blockade of RVLM and RVMM essentially eliminated LSNA, while RSNA and SSNA were reduced only 60%. From these data we conclude that (1) RVLM and RVMM contribute equally to the neurogenic maintenance of AP; (2) RVLM and RVMM differentially control the activity of individual sympathetic nerves; and (3) a substantial portion of RSNA and SSNA originates outside the RVM and may not be involved in vasomotor control.

Blockade of spinal pathways decreases pre- and postganglionic discharge differentially.

Although discharge of many sympathetic nerves decreases substantially on transection of the spinal cord in cats, firing of splenic and mesenteric postganglionic nerves continues unabated (R. L. Meckler and L. C. Weaver, Brain Res. 338: 123-135, 1985; R. D. Stein and L. C. Weaver. J. Physiol. Lond. 396: 155-172, 1988). Therefore, ongoing sympathetic outflow directed to the splanchnic circulation was proposed to be less dependent on supraspinal excitatory drive than that directed to other vascular beds. Blockade or transection of cervical spinal pathways in chloralose-anesthetized cats significantly decreased firing of gastric, hepatic, adrenal, and lumbar chain sympathetic nerves. Discharge of gastric and hepatic nerves decreased as much as that of adrenal and lumbar chain nerves; therefore, sustained discharge in the absence of bulbospinal excitation is not a characteristic of all sympathetic nerves innervating the splanchnic circulation. In contrast, discharge of the preganglionic greater splanchnic nerves was not decreased after spinal transection, although it changed from a rhythmic to an asynchronous pattern. This provocative finding suggests that rhythmicity of preganglionic discharge may be important for effective synaptic transmission to some postganglionic neurons, since postganglionic gastric, hepatic, and adrenal nerve firing decreased in the presence of sustained but asynchronous preganglionic input.

Effect of experimental heart failure on peripheral sympathetic vasoconstriction.

To investigate whether heart failure impairs peripheral sympathetic vasoconstriction, hindlimb vascular responses to lumbar chain stimulation (0.5-20 Hz) were studied in normal dogs and in dogs with chronic heart failure produced by rapid ventricular pacing. At lumbar chain stimulation rates of 0.5-3 Hz, hindlimb vascular responses were comparable in both groups. However, at stimulation rates of 5-20 Hz, vascular responses were significantly attenuated in the dogs with heart failure. Vascular responses to norepinephrine (0.1, 1, and 10 micrograms/min) were not altered. These findings suggest that chronic heart failure results in impaired sympathetic vasoconstriction, probably because of reduced neurotransmitter release. This abnormality may interfere with the capacity of the failing circulation to compensate for a low cardiac output and thereby intensify the severity of heart failure.