The innervation of the viscera is very sparse compared to the somatic system. On the other hand, most visceral afferents converge with neurons that receive input from superficial and deep somatic tissue as well as other viscera. The diffuse and widespread innervation of the spinal cord may explain the unpleasant nature of visceral pain.
Although other pathways such as vagus are involved, visceral pain is mainly conveyed to the spinal cord together with the sympathetic nerves. These are passing through the dorsal root ganglion and project to distinct laminae of the spinal cord dorsal horn (mainly laminae I and V, and occasionally to laminae V and X). These afferent projections are organized in a somewhat segmental manner, but distributed over several spinal segments in both rostral and caudal directions (6). This diffuse spinal termination pattern may explain the poor localization of visceral sensations often seen in clinical practice . From the spinal cord, pain transmits to the brain through diverse pathways. Most afferents travel in the spinothalamic tract to the thalamus. From the thalamus projections to the insula, hypothalamus, amygdala as well as to higher cortical levels such as cingulate and prefrontal cortices have been described. Insula has an important function for integrating the visceral sensory and motor activity together with limbic integration and is particularly important in pain perception from the gut . The anterior cingulate cortices and prefrontal cortices are a part of the medial pain system, which mediates the affective, emotional and cognitive components of pain experience . In addition to the spinothalamic tract, some afferents ascend in the spinoreticular tract mediating arousal and autonomic responses through interaction with the reticular formation. Finally, a population of afferents ascends in the spinomesencephalic tract, which relates to a complex neuronal network including the periaqueductal grey, rostroventral medulla and the dorsolateral pontine tegmentum. This network comprise the structural basis of descending pain control and possess modulation of spinal pain processing through so called on- and off-cells in the rostroventral medulla, which are pro-nociceptive or anti-nociceptive respectively .
Mechanisms Behind Referred Pain to Somatic and Visceral Structures
The fact that each segment in the spinal cord receives afferent fibers from visceral as well as somatic structures causes another phenomenon known as referred pain. Although simplified, referred pain originates due to convergence between visceral and somatic fibers on the same second order neuron (for details see [12,13]). Since the brain cannot localise the precise origin of the visceral pain stimulus, it may therefore interpret the pain as originating from a somatic structure with the same segmental innervation.
Viscero-visceral hyperalgesia is a complex form of hypersensitivity probably explained by more than one mechanism. Since this phenomenon takes place between visceral organs sharing their central afferent termination, it is plausible that central sensitization play an important role . Animal experiments as well as human experimental studies have shown that convergence to the same second order neuron of afferents from different organs exist [15, 16]. Also, our group has shown that esophageal perfusion with acid and capsaicin increased the sensitivity in the rectum – and that the referred pain areas to duodenal stimulation were increased [17,18]. Besides changes at the spinal level, changes in the cortical processing of pain may be involved in these mechanisms . Viscero-visceral hyperalgesia may explain the epidemiological findings that several clinical conditions with organic diseases show evidence of increased pain from other organs. This was recently investigated in patients suffering among others from coronary artery disease and gallbladder stones or inflammatory bowel disease and dysmenorrhea . In these studies co-existing visceral pain conditions with share of spinal projections between organs increased the symptoms generated from the other pain condition. Besides, effective treatment of one condition significantly improved symptoms from the other.
Like in other tissues visceral nerves are able to sensitize with major changes in the central nervous system. The peripheral nociceptor sensitization underlies the hyperalgesia that develops immediately around an injury site. Analogous to mechanisms documented in the somatic system, visceral afferent fibers may become sensitized by endogenous chemicals. This results in an increase in their responsiveness to a given stimulus and/or an increase in the spontaneous activity. In contrast to the cutaneous system, where only nociceptors sensitize, both low and high threshold fibers in the viscera can undergo sensitization [21,22]. This reduces the perception threshold of primary afferents and recruit previously silent nociceptors again leading to increased afferent activity to the spinal cord and exacerbation of the pain. Enhanced spinal input can activate intracellular signaling cascades within the spinal dorsal horn neurons. This results in an increased synaptic efficacy and is known as central sensitization. Visceral pain input to the spinal cord is more potent than cutaneous pain in the induction of central sensitization . Simplified, the input leads to e.g., activation of the N-methyl-D-aspartic acid (NMDA) receptor and results in changes of the resting potential of the second order neuron. Blocking the NMDA receptor has been shown to prevent experimentally acid-induced central sensitization from esophageal afferents . This effect is used clinically, where antagonism of the receptor during surgery leads to less postoperative pain . Sensitization of the viscera also affects the brain processing of pain as discussed later in this chapter.
CENTRAL NERVOUS SYSTEM EFFECTS TO EXPERIMENTAL PAIN IN HEALTHY VOLUNTEERS
Our knowledge about human pain is, to a major degree, based on clinical studies. However, pain in patients is often blurred by other symptoms such as anxiety and cognitive consequences of the pain. Furthermore, sedative properties of some analgesics make evaluation difficult. Experimental methods to evoke and assess pain under controlled circumstances are advantageous as they encompass many of these problems and offer a unique opportunity to investigate analgesic effects on different pain modalities arising from different tissues as well as peripheral and central pain mechanisms, for details see [26,27].
In human experimental pain models, the evoked sensations can be assessed with subjective methods quantitatively (e.g. by using a visual analogue scale) and qualitatively (e.g. by using the McGill Pain Questionnaire), and stimulus-response relationships can be investigated. Objective, physiological responses for the pain can also be recorded with e.g. the nociceptive reflex, cerebral evoked potentials and imaging reflecting central changes.
Referred Pain and Viscera-Visceral Hyperalgesia
Several studies have confirmed animal experiments showing that sensitization of the human gut such as with acid perfusion of the esophagus results in CNS alterations reflected in e.g., an increase in the referred pain area as well as a lowering of the pain thresholds to stimulation of other organs. We previously showed that perfusion of the distal esophagus induced an increase in the referred pain are to subsequent stimulations of the sensitized area . Furthermore, FrØkjær et al. showed that acidification of the distal esophagus resulted in hypersensitivity to stimulation of very remote organs such as the rectum . Such changes thought to be mainly spinal may be modality specific and influenced by descending modulation. Hence, in a subsequent study where an more intense sensitization model was used (perfusion of the esophageal with both acid and capsaicin) there was rectal hyperalgesia to heat and mechanical stimulations, whereas there was hypoalgesia to electro-stimulation of both the esophagus and the sigmoid colon . The sensitization may lead also lead to profound changes in the autonomous systems and thereby influence the brains so-called pain matrix .
The Nociceptive Reflex
The connection from the primary afferents to the motor neurons is a polysynaptic spinal pathway, which can be modulated by other afferent input, spinal neuronal excitability, and activity in descending control systems. In the studies by Bouhassira et al. [32,33], tonic distension of the stomach and rectum resulted in inhibition of the reflex. In other studies of somatic tissues, however, painful stimuli resulted in either a decrease or an increase in reflex excitability depending on the conditioning site . In a study where we distended the esophagus, no change in reflex size was present before sensitization was induced . Therefore, the different stimulation sites and experimental situations may explain the discrepancies between the experiments. In our experiments of the esophagus sensitization to acid, perfusion resulted in a significant increase in the baseline reflex excitability followed by a gradual inhibition during the visceral stimulus. The initial increased excitability may be explained by the chemical stimulation resulting in increased visceral input to the spinal cord.
Imaging Studies (and their Limitations)
Most previous reviews have focused on imaging methods as these have many advantages. For example has functional magnetic resonance imaging (fMRI) an excellent spatial resolution (2–5 mm), especially in the more superficial layers. However as the temporal resolution ranges from a 300 ms theoretical value to a more realistic 1–3 s there will invariably be many bias relating to non-specific activation and activation of centers and networks not related directly to pain. Therefore in this chapter the focus will be on findings from electroencephalographic (EEG) studies although the most important imaging studies will also be highlighted. For a more comprehensive overview the reader is referred to e.g., [36,37]. Neuroimaging has mainly been based on tools such as positron emission tomography (PET), fMRI and magnetoencephalography. Results of these studies are, despite some emerging consistencies, fairly heterogeneous. Differences are likely due to methodological differences in stimulation method (fixed stimulus intensity or titrated to individual perception threshold), imaging modality (PET vs fMRI) and/or image processing and analysis . Nevertheless, the studies have greatly improved our knowledge of the brain regions involved in visceral sensory processing by permitting the identification of the ‘visceral pain neuromatrix’, including brainstem regions, thalamus, insular cortex, primary and secondary somatosensory cortex (SI/ SII), cingulate cortex subregions (mainly anterior and midcingulate cortex (ACC, MCC)), prefrontal cortex (PFC) subregions, and cerebellar areas.
Even in healthy volunteers, psychosocial context and normal psychological processes exert a profound influence on pain processing through complex reciprocal interactions between emotional and cognitive pain modulatory brain networks . For example in a study where the esophagus was distended it was showed that a non-painful visceral stimulus is associated with higher activation of the pain-processing areas in anterior MCC and insula when perceived in a negative emotional context . In more recent studies Coen et al. found that the reduction in pain ratings during distraction from a painful esophageal stimulus was paralleled by a reduction in neural activity in the right MCC and the PFC . Experimentally induced sadness was also associated with higher activity in the right anterior MCC, insula and PFC during painful esophageal distension, but not with higher pain scores . Apart from the discussion of pain specificity, the common interpretation that the signal reflects synaptic activity is also disputed. Critics refer to alternative neurobiological accounts of the hemodynamic response, among these the removal of lactate, adjustment of the tissues acid-base or ionic balance or temperature regulation. Furthermore there are many technical limitations with fMRI that will not be discussed in the current paper, for review the reader is referred to [41–43]. Therefore there is a constant search to improve the methods and try new modalities more specific for pain such as localizing networks with resting state MRI and arterial spin labeling. Structural assessment of the brain with cortex volumetric methods and diffusion tensor imaging has also improved our understanding of the pain system (43). For example a combination of diffusion tensor imaging and fMRI were used to investigate the anatomical relationships between areas involved in sensations to rectal distension in healthy subjects, giving insight into the neural network of visceral sensory perception with direct connections between insula, and the ACC, thalamus, SI, SII and PFC  (Fig. 2).
Electro- and Magnetoencephalography
Although EEG methods also have many limitations – some of them shared with fMRI – they are generally highly available, relatively easy to use and low-cost [41,42]. The main advantage is the high temporal resolution, which allows assessment of the primary pain processing, including sequential activation and analysis of coherence and cross talk between brain centers. However, EEG has a relative poor spatial resolution even though various inverse modeling algorithms and signal decomposition procedures (see below) have overcome this limitation to some extent. Even though multichannel EEG, cerebral evoked potentials (EPs) and inverse modeling of the brain sources offers a non- invasive approach to study brain activity with time resolution on millisecond scale, it must not be overlooked that the position of the calculated dipolar sources does not represent the accurate position but rather the “centre of gravity” of brain activity. The resting EEG is typically used to study the pathophysiology of pain in chronic pain patients, while the EPs are used to study the nociceptive pain response including the sequential brain activation following visceral stimulations . Taken together, spontaneous EEG and EPs provide complimentary information about the modulation of CNS after painful stimuli.
In spite of the chaotic nature of the signals, it has been shown that the resting stage EEG oscillates in certain frequency bands associated with specific brain functions such as pain perception and attention to pain. EEG changes to experimental pain have been described in several studies of somatic tissues, but no studies were done in volunteers to experimental visceral pain. In contrast to the spontaneous EEG, the EP presents the time-locked response to an external stimulus. As this response is highly influenced by the sequential activation of distinct brain centers, the morphology of the EP is different from the spontaneous EEG (Fig. 2). The EP is characterized by several peaks of both negative and positive polarity, and may be quantified by the peak amplitudes and latencies. Although simplified the amplitudes represent the number of synchronous activated neurons, while the latency represents the delay in activation due to cortico-cortical connections.
Multi-channel EEG recordings combined with inverse modeling can determine the location of brain centers underlying EPs. This method possesses the opportunity to study pain specific cortical activation dynamically, as it reflects the sequential activation of neuronal pain networks underlying the EPs. There are a number of commercial software and freeware available for inverse modeling (Fig. 3). However, one of the major limitations with inverse modeling has been the instability of algorithms to model several simultaneously active sources and sources in deep brain structures such as thalamus or brainstem. To bypass these problems, signal decomposition methods have been developed in order to separate the signal into a sum of waveforms, whereby signals corresponding to specific evoked brain activity can be separated from artifacts and noise. Some of the most common approaches for signal decomposition are blind source separation algorithms such as independent component analysis and second order blind identification. Drewes et al. have successfully used signal decomposition with independent component analysis to study sequential brain activation and cross talk between brain centers following electrical stimulation of the esophagus in healthy controls . Recently, multichannel matching pursuit was introduced, which decomposes the EP data into a sum of waveforms (usually termed atoms), each of them being defined in time, frequency and space. Inverse modeling on these atoms is superior to inverse modeling on instantaneous EP data and other signal decomposition methods . For a short review the reader is referred to .
Evoked potentials have been used to explore the pain matrix following experimental visceral pain in healthy volunteers. For example we modeled the brain sources of EPs to electrical stimulation of the upper gut and sigmoid colon and these were explained by bilateral brain sources in the SII, insular cortices and a single dipole in the ACC . Interestingly, a viscerotopic organization of the different gut segments was seen, thus revealing a “visceral homunculus” mimicking that seen for the somatic sensory system. Acid sensitization of the esophagus has been shown to cause neuroplastic changes at the spinal cord level reflected in increased referred pain areas and amplitude of the nociceptive withdrawal reflex [28,35]. Sarker et al. used stimulation of a proximal region of the esophagus in healthy volunteers (apart from the distal acid exposure) and found a decrease in latency of the N1 and P2 components in the vertex recordings . In a subsequent experiment where we stimulated the acid perfused area in the lower esophagus we also found a reduction of P2 latency and using inverse modeling a posterior shift and latency reduction of the ACC dipole was seen . Short-term sensitization of the esophagus therefore results in central neuroplastic changes involving e.g., the cingulate gyrus, which has in many studies in functional diseases of the gut shown pathological activation, thus reflecting the importance of this region in visceral pain and hyperalgesia. Hence, experimental pain and sensitization in healthy volunteers is able to induce rather profound changes in the way pain in sequentially processed in the brain. The stability and temporal resolution of the EPs and source modeling validates their use in patient studies, however, advanced inverse modeling is still restricted to expert use.