Functional Abdominal Pain: The Basic Science

Primary causes of abdominal pain of digestive tract origin are distension and excessively strong contractions of the musculature. Other stimuli (eg, pinching and burning of the mucosa) do not evoke pain. Hypersensitivity of the mechanoreceptors, which detect stretch (ie, distension) and contractile force, is a pain factor in functional gastrointestinal disorders (FGIDs). Hypersensitivity to distension is found in a substantial subset of patients that meet Rome III symptom-based criteria for diagnosis of irritable bowel syndrome (IBS) (1–4). Inflation of a balloon placed in the rectosigmoid region evokes sensations of discomfort and pain at lower distending volumes for patients with IBS than for healthy subjects. Hypersensitivity to distension in these cases is not restricted to the distal large intestine. Patients with functional dyspepsia also experience discomfort and pain at lower distending volumes in the stomach that healthy individuals (4–6) and hypersensitivity to distension is present in the esophagus of patients with noncardiac chest pain (7). Patients with IBS experience the same kind of hypersensitivity to electrical stimulation applied to the mucosa of the rectosigmoid region as they do during balloon distension in the rectosigmoid region (8). The hypersensitivity to direct electrical stimulation of the intramural sensory innervation implicates abnormal sensory neurophysiology rather than mechanical factors (eg, wall tension, compliance) as underlying the decreased sensory threshold in patients with IBS. Precise identification of the sensory defect in IBS is an area of active investigation and not yet satisfactorily established. In fact, the lower sensory threshold in IBS may reflect sensitization of intramural mechanoreceptors, sensitization of neurotransmission at synapses in the spinal cord, abnormal processing of the sensory information when it reaches the brain, and combinations of each of these possibilities. PERIPHERAL NEUROPATHOLOGICAL FACTORS Mechanosensitive primary sensory afferents in the walls of the specialized organs of the digestive tract detect and signal the strength and rate of change of contraction and distension of the musculature. The mechanosensitive structures expressed at the afferent terminals behave as if they are attached in series with the long axes of the smooth muscle fibers. This arrangement accounts for activation of the sensors by either muscle contraction or distension. The cell bodies of the neurons that give rise to gastrointestinal sensory afferents are located in vagal nodose ganglia and dorsal root spinal ganglia. Mechanosensitive information is transmitted along spinal afferents to the spinal cord by way of dorsal root ganglia and along vagal sensory afferents to the brainstem by way of the nodose ganglia and synaptic relays in the nucleus tractus solitarius. Sensations of pain and discomfort of digestive tract origin reflect transmission in spinal afferents and information processing in the spinal cord and brain. Sensory information transmitted by the vagus nerves appears not to reach the level of conscious sensation because patients with high spinal cord transections and intact vagal pathways experience little or no sensations of digestive tract origin. Conscious perception of sensations from the digestive tract remains after a surgical vagotomy and, presumably, reflects transmission over spinal afferents to the spinal cord and onward to conscious centers in the brain. The streams of mechanosensory information, which are transmitted from the small and large intestine to the central nervous system, are generated by 3 kinds of sensory afferents identified as low-threshold, high-threshold or silent afferents (9). Low-threshold afferents, which code for small changes in wall tension, are believed to transmit the minute-to-minute information required for functional autonomic negative-feedback control during contractile events. To be activated, high-threshold afferents require stronger changes in wall tension, which may be produced by distension of the lumen or strong contraction of the musculature. High-threshold afferents are postulated to be responsible for the range of sensations from mild to severe pain that are associated with excessive distension or exceptionally strong muscle contraction. Silent afferents cannot be activated by distension when the bowel is in its normal state. They acquire pathophysiological significance by becoming active and highly sensitive to stimuli during inflammatory states (10,11). POSTINFECTIOUS IBS A significant percentage of individuals develop IBS-like symptoms after an acute bout of infectious enteritis (12–15). The issue of whether the association between acute infectious enteritis and IBS reflects low-level inflammation (eg, microscopic enteritis) and chronic exposure of the neural and glial elements of the enteric nervous system to elevated levels of serotonin, histamine, or other inflammatory mediators remains to be fully resolved. SEROTONIN RECEPTORS ON SENSORY AFFERENTS Application of serotonin in animal studies evokes increased firing in sensory fibers leaving the stomach and small and large intestine and this is mediated by the serotonergic 5-HT3 receptor subtype. Selective 5-HT3 receptor antagonists (eg, alosetron) block this action (16,17). Intramural terminals of both spinal and vagal sensory nerves express 5-HT3 receptors. Efficacy of alosetron in the treatment of abdominal pain and discomfort in the diarrhea-predominant form of IBS in women suggests that their symptoms may reflect overactive endogenous release and elevated levels of serotonin (18,19). Persistence of serotonin at its receptors, due to weakened uptake by the serotonin transporter, is a second possibility for hyperstimulation of intramural serotoninergic receptors. Active uptake mediated by the serotonin transporter, which is expressed by enteric neurons and mucosal epithelial cells, normally restricts accumulation and action of serotonin at its receptors following its release (20). Downregulation of gene expression for the serotonin transporter is reported to be present in mucosal biopsies taken from the large intestine of patients with IBS (21). Enhanced propulsive motility and watery stools occurred in mice with a deletion in the serotonin transporter gene, and Chen et al (22) reported that these mice sometimes alternate between diarrhea and constipation in ways reminiscent of the subset of patients with IBS, who are classified as alternators. RECEPTORS FOR INFLAMMATORY MEDIATORS ON SENSORY AFFERENTS Receptors for bradykinin, adenosine triphophate, adenosine, prostaglandins, leukotrienes, histamine and mast cell proteases, like 5-HT3 receptors, are expressed on intramural spinal sensory nerve terminals (9,23). Any of these inflammatory cell-derived or ischemia-related mediators has potential for elevating the sensitivity of intestinal sensory nerves, especially in the disordered conditions of inflammation or ischemia. This likelihood is reinforced by findings that a reduced threshold for painful responses to balloon distension in the large intestine is associated with degranulation of enteric mast cells in animal models. Treatment with mast cell stabilizing drugs prevents the lowering of the pain threshold, which occurs during mucosal inflammation in the animal models (24), and suggests that mast cell stabilization may be an efficacious treatment in human IBS (25,26). CENTRAL NEUROPATHOLOGICAL FACTORS The evidence is consistent with altered peripheral sensory transduction as the underlying factor for the exaggerated sensitivity to distension found in patients with IBS (27). In this scenario, mechanosensitive primary afferents in the gut wall become hypersensitive to mechanical stimuli and as a result transmit at elevated frequencies of firing, which is then interpreted in the central nervous system as nociception. In an alternative scenario, normally functioning mechanoreceptors transmit accurate information, which is then misinterpreted in processing circuits in the spinal cord and/or brain to evoke conscious perceptions of disordered sensations from the gut. ASCENDING SENSORY PATHWAYS IN THE SPINAL CORD Ascending spinal pathways involved in the transmission of nociceptive signals from the digestive tract are the spinothalamic, spinohypothalamic, spinosolitary, spinoreticular, and spinoparabrachial tracts. Visceral pain information is also transmitted to higher processing centers in the brain along pathways in the dorsal spinal columns (28–30). Spinal afferents, which transmit nociceptive information, connect synaptically with second-order neurons in the dorsal column nuclei (ie, nucleus gracilis and nucleus cuneatus). Second-order neurons in the spinal dorsal horn also provide input to the dorsal column nuclei. The pain signals are transmitted cephalad by way of the ipsilateral dorsal column nuclei to the contralateral ventroposterolateral nucleus of the thalamus (28,30,31). A midline myelotomy, which severs axons in the human dorsal columns, attenuates visceral pain that is otherwise intractable (31,32). Electrical stimulation of the dorsal columns in patients with severe IBS evokes an immediate increase in the intensity of their abdominal pain (33). These observations in humans are consistent with experimental results in animals (29,34). SENSORY PROCESSING IN THE CEREBRAL CORTEX Methods of brain imaging have emerged as a useful investigative tool for addressing questions related to abnormal processing of sensory information in the cerebral cortex of patients with FGID (27,35–37). Functional magnetic resonance imaging (fMRI) and positron emission tomography (PET) are methods commonly used to study information processing in the higher brain centers that underlie an individual's conscious experience of gut-related sensations. Results from imaging studies suggest that unlike somatic sensation, which has a main homuncular representation in the primary somatosensory cortex, visceral sensation is mainly represented in the secondary somatosensory cortex (37,38). Differences in processing in these specialized regions may account for the imprecise nature of an individual's ability to localize visceral sensation in relation to somatic sensation. Beyond the sensory cortices, fMRI and PET images show representation of both somatic and visceral sensation to be similar in the limbic and paralimbic regions of the cortex (eg, anterior insular, anterior and posterior cingulate, prefrontal and orbitofrontal cortices (35,36). Activity in these areas underlies the individual's motivational and emotional mood and the cognitive aspects of visceral sensations. Sex differences, reminiscent of the female predominance in IBS, are seen in the cortical representation when a balloon is distended in the rectosigmoid region in healthy subjects of both sexes (39). On the one hand, activation in the sensory/motor and parieto-occipital areas of the cortex does not differ for males and females. On the other hand, activation in the anterior cingulate and prefrontal cortices is more extensive in females than in males. Amounts of evoked cortical activity in females is greater than in males for perception levels in the range from the urge to defecate to sensations of fullness, mild discomfort, and pain. Significance of these sex differences is unclear; nevertheless, they are reminiscent of the greater perceptual responses reported in female patients with FGIDs (eg, IBS) (40). In healthy subjects, painful sensations evoked by distension of a balloon in the rectosigmoid region, as well as the anticipation of a potentially painful distension, are associated with increased blood flow in the anterior cingulate cortex in PET. In patients with IBS, activation of the anterior cingulate cortex occurs in response to painful distension or the anticipation of painful distension (35,41). A different study with fMRI found that patients with IBS showed enhanced activation of the mid-cingulate cortex in response to rectal distension (42). The cingulate cortex has become a focus of attention because it is generally thought to be an integrative center for both emotional experience and pain-specific sensory information that may account for the well-known linkage between pain and the individual's emotional state. The so-called affective areas of the cingulate cortex have extensive connections with the amygdala and periaqueductal gray matter and with autonomic nuclei in the brainstem (43). Activity in these connections integrates recall of emotional experiences with neural autonomic and endocrine control functions at the level of the gut. Cognition is believed to reside in the caudal portion of the anterior cingulate cortex where the microcircuits are delegated to premotor functions and processing of nociceptive information. The functional neuroanatomy of the cingulate cortex offers an explanation for the well-documented association of psychosocial disturbances (eg, negative life events) with more severe cases of IBS (40). Rectal distension in people with histories of severe sexual and/or physical abuse has been reported to selectively activate the perigenual region of the anterior cingulate cortex. A case report described a middle-age female whose low pain threshold for rectal distension and diarrhea-predominant IBS improved after she was extricated from an abusive psychosocial situation. Brain imaging with fMRI in this individual showed activation of the mid-cingulate cortex during rectal distension before treatment and resolution of activation associated with improvement in the patient's psychosocial situation (44). CENTRAL SENSITIZATION (“WIND UP”) Elevated excitability in nociceptive dorsal horn neurons underlies spinally mediated hyperalgesia, which is called central sensitization to distinguish it from sensitization that occurs at nociceptive terminals in the periphery. In conditions of severe tissue injury and persistent injury, nociceptive C fibers fire repetitively and the excitability of the second-order neurons in the dorsal horn increases progressively in response to the elevated synaptic input. This effect is sometimes called wind-up and reflects the synaptic release of glutamate from the incoming C-fibers and activation of N-methyl-D-aspartate-type glutamate receptors expressed by the second-order neurons. These long-lasting changes in the excitability of dorsal horn neurons are like a memory imprint of the nociceptive input. Accumulating evidence suggests that the spinal wind up phenomenon is operational for the intestinal tract and may be an underlying factor in the hypersensitivity associated with IBS. Suggestive evidence for wind up hypersensitivity is the finding that second-order neurons in the dorsal horn show induction of the early gene c-fos in response to noxious balloon distension of the colon in rats (45). Changes in gene expression in this case are expected to underlie postsynaptic excitability changes in the second-order nociceptive neurons. A related phenomenon occurs in the rat spinal cord where excitability of second-order nociceptive neurons becomes sensitized to distension-evoked input following inflammation of the colon. The same phenomenon is evident in dorsal column nuclei where firing of second-order neurons to colorectal distension becomes sensitized following inflammation of the colonic mucosa (46). In a study, which is reminiscent of the connection between sexual and physical abuse documented for early childhood and IBS in adult humans, Al-Chaer et al (47) reported that central sensitization may also be induced in animal models in the absence of inflammation. Neonatal rats in their study were subjected daily to noxious colorectal distension or intracolonic injection of mustard oil beginning 8 days postpartum and lasting for 21 days. When tested in adulthood, the rats that were “abused” as neonates were hypersensitive to colorectal distension as reflected by a lower threshold and elevated intensity of reflex responses indicative of abdominal pain. Single-unit electrophysiological recording from dorsal horn neurons in the lumbar and sacral regions of the spinal cord of the adult animals detected significantly higher background firing frequencies in the animals that were abused as neonates and enhanced firing frequencies in response to colorectal distension when compared with nonabused controls. Histological examination uncovered no evidence of an inflammatory state in the large intestine in either the adult abused animals or their controls in these studies. Limited evidence for central sensitization to distension of the large bowel has been obtained for humans who meet Rome II or III diagnostic criteria for IBS (3). In this group of patients, repetitive inflation of a balloon in the sigmoid colon to noxious levels of stimulation altered the processing of afferent information entering the spinal cord from the rectum. Altered central processing was present in the patients with IBS and not in healthy controls. INTESTINAL MOTILITY AND ABDOMINAL PAIN Strong contractions of the intestinal circular muscle coat during intestinal power propulsion underlie the sensation of cramping abdominal pain (48,49). Power propulsion occurs more frequently in patients with IBS than in healthy subjects, and the circular muscle contractions are significantly stronger than normal in patients with IBS (50). Moreover, postprandial power propulsion is more prevalent in the colons of patients with IBS than in normal individuals and power propulsion in the colon is often associated with their diarrheal state (51). The pain and discomfort in patients with IBS during the powerful contractions of power propulsion may be explained in 3 ways, either separately or in combination. One explanation is for the exceptionally powerful circular muscle contractions to activate high-threshold mechanoreceptors that transmit the information centrally, where it is processed and projected to consciousness as the perception of pain and discomfort. A second is for the mechanoreceptors to become sensitized in the patients with IBS (eg, by inflammatory mediators or other paracrine signals) and send erroneously coded information to processing centers in the spinal cord and brain. A third explanation is for accurately coded sensory information carried by spinal afferents to be misinterpreted as it is decoded in the spinal cord and central processing centers of the brain. Acknowledgment The author thanks the National Institutes of Health NIDDK Institute for support of work on enteric neurobiology since 1973.