- Email: [email protected]
Neurophysiology of Pain and Pain Modulation
Neurophysiology of Pain and Pain Modulation
An endogenous central nervous system pain-modulating network, with links in the mid brain, medulla, and spinal cord, has recently been discovered. This system produces analgesia by interfering with afferent transmission of neural messages produced by intense stimuli. Although other neurotransmitters are involved, the analgesia produced by this system depends on the release of endogenous opioid substances, generically referred to as endorphins. The system is set in motion by clinically significant pain-such as that resulting from bony fractures or postoperative pain. The analgesia network monitors the pain and controls it at the level of the spinal cord. Complex psychologic factors play an important role in the variability of perceived pain, partly because of their ability to trigger this pain-suppressing system. For example, this system contributes to the analgesic potency of placebo administration and is also activated by stress. Knowledge of this analgesia system has greatly expanded our understanding of the mechanisms underlying pain management. Opiates, like morphine and meperidine, produce analgesia by mimicking the action of endorphins in the brain. Tricyclic drugs may produce analgesia by enhancing the nonendorphin links of the same system. Future research on this system will provide new insights and, consequently, new approaches to the management of pain.
HOWARD L. FIELDS, M.D. San Francisco, California
From the Departments of Neurology and Physiology, University of California School of Medicine, San Francisco, California. Requests for reprints should be addressed to Dr. Howard L. Fields, University of California School of Medicine, Department of Neurology, San Francisco, California 94143.
2
September 10, 1984
Within the last decade or so, remarkable progress has been made in understanding the anatomy and physiology of pain. The two most important recent advances in thi~ field have been (1) the discovery of a highly organized central nervous system network for the monitoring and modulation of pain, and (2) the isolation and description of the endorphins, which are synthesized by nerve cells and have the pharmacologic properties of morphine. The mechanisms by which this endogenous neural system produces analgesia will be reviewed. Before describing the pain-suppression system in detail, it will be useful to briefly outline the neural basis of pain sensation. Pain functions as a warning system. The neural pathways that transmit pain messages signal that tissue damage is occurring or about to occur. The pain-signaling system is required to protect the body, and it works well most of the time. On the other hand, the characteristic unpleasantness of pain has made analgesia one of the most important things that physicians can provide for their patients. As documented in this symposium, progress has been made in refining our tools for pain management. Part of this progress is due to an improved understanding of the neural mechanisms that underlie the perception of pain.
The American Journal of Medicine
PAIN MANAGEMENT SYMPOSIUM-FIELDS
Normally, intense stimuli or tissue damage sets in motion processes that lead to activity in primary afferent nociceptors. Although these processes are not fully understood, they are similar, if not identical, to those that underlie inflammation. These are fully discussed by Dr. Galin elsewhere in this symposium issue. The peripherally acting analgesics, such as aspirin, acetaminophen, and the nonsteroidal anti-inflammatory agents, probably act on these processes. Once the primary afferent nociceptor ("pain receptor") is activated, its message is transmitted over small diameter axons to the spinal cord. This message is subjected to two important modifying influences at this stage: one is the action of the large, myelinated afferent fibers, and the other is the influence of the sympathetic nervous system. The large, myelinated afferent fibers have no specific response to intense stimuli; in fact, they act to inhibit pain transmission at the level of the spinal cord. This phenomenon is well established and probably forms the basis of certain local therapies, such as massage, acupuncture, and transcutaneous electrical nerve stimulation (TENS). That the sympathetic nervous system can produce pain is dramatically illustrated by the syndrome of causalgia, the severe burning pairi that sometimes follows nerve injury. This pain is immediately and completely relieved by sympathetic block. When the pain signal arrives at the spinal cord, it activates cells there that, in turn, relay the pain signal to the brainstem and/or the thalamus. From the thalamus, the message is relayed to the cortex; it is there, presumably, that the activity occurs that underlies the conscious experience of pain. Before leaving the subject of pain transmission, it would be worthwhile to clarify the differehce between the sensory aspect of pain and the unpleasantness that usually accompanies pain. The sensory aspect involves the identification of the sensation as painful. The suffering aspect has more to do with the desire to escape the pain. Whereas the sensory aspect is fairly similar among individual persons and in the same person over time, the suffering or tolerance aspect is variable and may depend on personality, situational, and cultural factors. For example, most of us will either ignore a headache or treat ourselves with a mild analgesic, but if we were told that the headache was due to a brain tumor, the anguish and suffering would be immeasurable. This suffering aspect is an important source of the variability of clinical pain. It can be dealt with by a variety of approaches, discussed by Addison and Posner in their reports in this symposium issue. STIMULATION-PRODUCED ANALGESIA
The first evidence that this analgesia system existed was the observation that, in rats, electrical stimulation of discrete areas of the brain-primarily the mid-brain periaqueductal gray-profoundly inhibited responses to painful stimuli [1-3]. This analgesic effect was found to be
strikingly selective. Although the animals remained alert, active, and normally responsive to innocuous stimuli, noxious stimuli did not produce the expected vocalization, biting, or escape behavior. These findings were confirmed in patients with intractable pain who had stimulating electrodes implanted at sites near those in the periaqueductal gray from which this stimulation-produced analgeSia was elicited in animals (4,5]. As with the laboratory models, no consistent sensory or motor effects other than relief of pain were associated with the procedure, indicating that in human beings, as in animals, the system activated iS specifically designed for pain control. The induction of analgesia, separated from any other effect, is of fundamental importance in establishing pain modulation as a distinct physiologic function of the central nervous system. OPIATE ANALGESIA
At about the time this pain-modulation system was described physiologically and clinically, there were findings of equal significance in the field of neuropharmacology. First was the discovery of receptors ih the brain to which opiate compounds attach specificaliy [6-8]. The importance of these binding sites on receptors was established by showing that the potency of narcotic drugs for relief of human clinical pain correlates directly with their binding affinity for the receptor (8]. The discovery of these opiate receptors triggered a search for a substance produced within the central nerve ous system that would bind to the receptor. The first fruit of this search was the isolation from the brain of two endogenous pentapeptides, leucine and methionine enkephalin. These substances have morphine-like activity that is blocked by haloxone, a specific antagonist to morphine [9]. Subsequent research uncovered other endogenous, morphine-like compounds that are now classified, categorically, as endorphins. Several distinct endorphins with different peptide sequences are present in different cell populations of the brain; pituitary, adrenal, gut, and sympathetic nervous system. For example, beta endorphin is present in the pituitary gland, and enkephalins are present in the adrenal gland. Both have morphine-like actions. Knowledge of endorphins has provided considerable insight into the manner in which narcotic analgesics relieve pain. Drugs like morphine presumably work by mimicking the action of endorphins at synapses in the painmodulation network. In addition, endorphins have provided both an anatomic and chemical explanation for the phenomenon of stimulation-produced analgesia. That stimulation-produced analgesia and opiate analgesia share a common neural pathway is documented by a number of discoveries, such as: • Opiate receptors and endorphins are normally present in relatively high concentrations in areas such as the peri-
September 10, 1984
The American Journal of Medicine
3
PAIN MANAGEMENT SYMPOSIUM-FI ELDS
Figure 1. The endogenous pain-modulation system. A = mid-brain level; B = medullary level; C = spinal level; DLF = dorsolateral funiculus; E = endorphin; SHT = serotoinin; NE = norepinephrine; NRM = nuc/ei.Js raphe magnus; PAG = periaqueductal gray; R = reticular formation; T = pain transmission neuron (adapted with permission from [14].
aqueductal gray [1 0,11 ], from which stimulation-produced analgesia is elicited. • A precise anatomic correlation exists among opiate receptors, endorphin distribution, and the nuclei from which analgesia can be elicited by electrical stimulation or microinjection of opiates [12]. • Naloxone, a specific opiate antagonist, partially blocks , stimulation-produced analgesia [4,5]. • Lesions of the periaqueductal gray block analgesia from systemic opiates [13]. ANATOMIC PATHWAY
A great deal of effort has been put into mapping the anatomic connections of this pain-modulation system. Present evidence indicates that it is a descending system and that it controls pain at the level of the spinal cord by blocking neural activity produced by intense (noxious) stimuli. The first evidence that this system produces analgesia at
4
September 10, 1984
The American Journal of Medicine
spinal cord levels was the observation that discrete lesions of certain descending spinal pathways block both stimulation-produced analgesia [14, 15] and the analgesia produced by systemic injection of morphine [16]. Where ar'3 the cells of origin of this descending pathway for pain suppression? A substantial number lie in two major brainstem groups [16-18]. The one of most interest to us lies near the midline of the medulla (raphe nuclei) and includes many serotonin-containing cells [18]. There is an extensive body of evidence implicating these serotonergic neurons in pain control. For example, serotonin depletion counteracts stimulation-produced analgesia [19,20] and antagonizes opiate analgesia as well [21 ,22]. The nerve cells in this region of medulla send their axons in a well-defined pathway to terminate in the spinal cord. The termination pattern is fascinating and revealing in its specificity. The terminals of these brain stem painmodulating cells are directly upon the very spinal cord cells that respond to pain-producing stimuli. Furthermore, stimulation of the pain-modulating cells in the brainstem profoundly inhibits the spinal pain-responsive cells. Thus, the pain-suppression system blocks the pain message in the spinal cord so that it does not reach higher centers. There is very extensive anatomic overlap of the spinal terminals of pain-modulating cells, opiate receptors, enkephalin, and the location of spinal neurons that respond to noxious stimuli. This close correlation suggests that the mechanism of analgesia from brainstem stimulation involves the endorphins [14]. The reversal of stimulationproduced analgesia by the opiate antagonist naloxone [23] is consistent with this view. Figure 1 summarizes the anatomic and the physiologic aspects of the endogenous pain-modulation system. At the mid-brain level, you find the periaqueductal gray, an important locus for stimulation-produced analgesia. This area is rich in endorphins and opiate receptors, although the anatomic details of the endorphin connections are not known. Microinjection of small amounts of opiates into the periaqueductal gray also produces analgesia. At the medullary level, serotonin-containing cells of the nucleus raphe magnus and the adjacent reticular formation receive excitatory input from periaqueductal gray and, in turn, send impulses along efferent fibers to the spinal cord. At the spinal level, efferent fibers from the medulla travel in a discrete pathway in the dorsolateral funiculus to terminate among pain-transmission cells concentrated in the dorsal horn, where they exert an inhibitory effect specifically on pain-transmission neurons. Part of this effect is probably exerted via endorphin-containing interneurons. Pain-transmission neurons project to supraspinal sites and indirectly contact the cells of the descending analgesia system in the periaqueductal gray and nucleus raphe magnus, thus establishing a negative feedback loop that we will discuss in greater detail shortly.
PAIN MANAGEMENT SYMPOSIUM-FIELDS
Four Major Operating Characteristics of Endogenous Pain-Modulation System
OPERATING CHARACTERISTICS OF PAINMODULATION PATHWAY
TABLE I
There are four major operating characteristics of the endogenous pain-modulation system (Table 1): First, the system is at least partly endorphin-mediated. Second, it involves serotonergic neurons. Third, it depends on activation of descending pathways from the brain stem that inhibit pain-transmission neurons at the spinal cord. Finally, it is effectively activated by stress and pain-producing stimuli. Figure 2 illustrates the essential features of the intrinsic analgesia system. The basic element of the endogenous pain-modulation system is a negative feedback loop that monitors the output of the pain-transmission system and controls its input. The loop is set in motion when a potentially painful stimulus excites the pain-transmission system. When the activity in these neurons is sufficient to reach the level of the pain threshold, pain is perceived. Simultaneously, the activity of the pain-transmission neurons activates the pain-suppression system. The pain-suppression system, in turn, feeds back upon, and thus controls, the spinal pain-transmission neurons. The effectiveness of the pain-suppression system can be enhanced by administering analgesics, such as morphine or codeine, or by stimulating the analgesia network electrically. This reduces the perceived pain intensity. Conversely, disruption of the loop, either by opiate or serotonin antagonists, will increase the perceived intensity of the pain. Although the simplified schema of Figure 2 accounts for many of the aspects of pain transmission and modulation, it is far from complete, implying only one transmission and one modulation system. There is considerable evidence to indicate that there are at least two, and possibly multiple, ascending pain-transmission systems and several descending pain-modulation systems. The descending feedback loops may involve either naloxone-sensitive (endorphin-mediated) or naloxone-insensitive (nonendorphinergic) pathways. [14]. Pain modulation is, in part, a consequence of pain-producing stimuli. It follows, then, that pain perceived in re-
At least partly endorphin-mediated Involves serotonergic mechanisms Depends on activation of descending efferent pathways that inhibit pain-transmission neurons Effectively activated by noxious stimulation
sponse to a sustained stimulus should reach a peak intensity and, having activated the analgesia system, decline to a steady level. In other words, adaptation to noxious stimulation would be expected to result in part from activation of the pain-suppression system; the level of pain perceived would be less than that produced by pain transmission in the absence of the feedback modulation. The fact that postoperative pain is made worse by the administration of naloxone, and that naloxone is demonstrably worse than placebo in managing clinical pain, is consistent with this view [24,25]. NALOXONE, PLACEBO, AND ENDORPHIN-MEDIATED ANALGESIA
Studies with naloxone, often in conjunction with placebo studies, have been a bulwark for the existence of an endorphin-mediated analgesia system. What have these studies shown? To begin with, brief, superficial pains do not seem to cause endorphin release [26,27]. Noxious stimuli must be significantly above the pai~ threshold and have a prolonged duration in order to produce analgesic actions that are blocked by naloxone [28]. This indicates that endorphins are released only when the pain level is significantly above threshold intensity, perhaps only when a requisite degree of stress has been produced. In order to understand more fully the actions of endorphins on pain, one must realize that many, if not most, painful conditions remit without specific treatment. Common examples of this are headache, tic douloureux, dysmenorrhea, and herpetic pain. In these cases, patients go from a pain-free state to experiencing severe pain and then back to having either less severe pain or no pain. We
Pain Transmission System
Noxious Input
Figure 2. Essentials of the endorphinmediated negative feedback loop of the endogenous pain-modulation system (adapted with permission from [14]).
1------------.. . --+ Medullary Center
Pain
~Exogenous Opiates
Pain Suppression System
September 10, 1984
The American Journal of Medicine
5
PAIN MANAGEMENT SYMPOSIUM-FIELDS
Natural history
A
Improvement Severity
Placebo effect B
Severity Observed
l Placebo
do not understand this cycling, but clinical experience has established that this is the usual course of pain in most patients (Figure 3). Obviously, under these circumstances, all treatments will be followed by clinical improvement. This improvement is due to neither medication nor a placebo effect but is indicative of a reduction in disease activity. Whether the endorphins play a role in this "natural" cycling, observed in certain painful conditions, is still an open question. Figure 3 illustrates what is meant by placebo response. As discussed, the placebo response must be distinguished from the tendency of some painful conditions to remit spontaneously. Evidence that the placebo effect involves the release of endorphins has been seen in patients with postoperative dental pain; naloxone produces significant hyperalgesia in patients who respond to a placebo but has little effect in those who do not [25]. Painful stimuli that are stressful, inescapable, or associated with anxiety have been shown to be particularly effective in producing naloxone-reversible, and thus presumably endorphin-mediated, analgesia [29]. Likewise, anxiety has been correlated positively with effectiveness of placebo response [30]. In a recent study, a placebo was found to be more frequently effective for clinical than for experimentally induced pain, a result the investigators attributed to the increased stress of the clinical situation [31].
6
September 10, 1984
The American Journal of Medicine
Figure 3. The placebo response (adapted with permission from [34]).
Several studies have shown that there do not seem to be any specific personality characteristics that have a bearing on the ability to mobilize a placebo response (and hence, endorphins). In one such study, involving a group of patients with cancer pain who received a placebo several times, 90 percent reported a significant decrease in pain at least once [32], indicating that virtually everyone is capable of responding to a placebo. In another study [33], a group of women on an obstetric service were given a placebo once for relieving labor pain, once for postpartum pain, and once for experimental ischemic pain. If a woman showed a positive response to the placebo in one situation, the probability that she would show a positive response in a second situation was not greater than that expected by chance. In other words, situational factors were more important than personality traits in determining responsiveness to a placebo. The findings of these two studies are very intriguing because they suggest that, if we could create the appropriate environment, we might be able to produce placebo analgesia consistently in most people. One fact we do know in regard to creating an "appropriate" environment is that the interaction between what the physician communicates and what the patient expects is critical for the production of a placebo response. Of course, what a physician says and how it is said will de-
PAIN MANAGEMENT SYMPOSIUM-FIELDS
pend, in turn, on what the physician believes about the treatment being administered. The analgesic response to a placebo is enhanced by positive suggestion; conversely, even effective analgesic agents may have their overall efficacy diminished if the physician communicates doubt [34]. Thus, the outlines of an endogenous pain-modulation system, activated in the brain stem and functioning via descending connections to control afferent pain transmission, have been presented. Evidence suggests that the system relies on the release of endorphins even though other neurotransmitters are also involved. The pain-modulation system acts like a negative feedback loop that is set in motion by stress and noxious stimuli and that limits pain perception by interfering with afferent input at the level of the spinal cord. CLINICAL IMPLICATIONS
Knowledge of the pain-modulation systems and the endorphins provides a better basis for rational pain management. First of all, because the pain-transmission and painmodulation systems are separate entities and operate by distinct mechanisms, there are at least two general classes of approaches to pain management. One set of approaches is to block the transmission system; the other is to turn on or enhance the suppression system. Drugs like aspirin, acetaminophen, or the nonsteroidal anti-inflammatory agents block transmission at sites in peripheral tissues and/or at nonopioid receptors in the central nervous system. In contrast, the opiates, such as morphine or codeine, act at specific sites in the brain to activate the pain-suppression system. Because the drugs act at different sites by distinct mechanisms, their use in combination is rational, if neither class alone is sufficient for pain control. One new development in the use of opiates has been to apply them locally, at the level of the spinal cord. Remem-
ber that both endorphins and opiate receptors are present in the spinal cord. This leads to the prediction that the application of morphine (or any narcotic analgesic) directly to the cord-either in the epidural or intrathecal space-will produce local analgesia. In fact, this has been verified in a variety of painful conditions. The analgesia produced is powerful, long-lasting, and reversible with opiate antagonists. This method is being used clinically for lower body pains in labor, for postoperative pain, and for cancer patients. Some cancer patients have cannulas placed in their epidural space for weeks at a time, with an implanted reservoir to deliver drug continuously. The fact that the pain-suppression system has nonendorphin synapses raises the possibility that nonopiate drugs may be potent centrally acting analgesic agents. For example, serotonin plays a major role producing analgesia, and it is well established that tricyclic antidepressant drugs, such as amitriptyline, are effective in certain chronic painful conditions (for example, migraine, postherpetic neuralgia). The tricyclics block the removal of serotonin from the synaptic cleft and thus prolong and increase its action. In animal pain models, tricyclics have a direct analgesic action and potentiate the analgesic action of opiates [35]. The tricyclics represent a totally new class of centrally acting, nonaddicting pain-killers. Finally, it should be stressed that almost any therapeutic manipulation has the potential to elicit placebo analgesia and that this effect is at least partly mediated by endorphins. Thus, the common practice of using a placebo to diagnose psychogenic pain is based on erroneous assumptions. In fact, there is evidence to suggest that more intense pain or stress is associated with greater activation of the endorphin-mediated analgesia system; therefore, up to a point, the more intense the pain, the more likely the response to a placebo. The future of pain management will include closer attention to psychologic and physical methods for activating the analgesia system.
REFERENCES 1. 2.
Mayer DJ, Liebeskind JC: Pain reduction by focal electrical stimulation of the brain: an anatomical and behavioral analysis. Brain Res 1974; 68: 73-93. Mayer DJ, Wolfe TL, Akil H, et al: Analgesia from electrical stimulation in the brain stem of the rat. Science 1971; 174: 1351-
7.
8. 9.
1354. 3.
Reynolds DV: Surgery in the rat during electrical analgesia induced by focal brain stimulation. Science 1969; 164: 444-
10.
445. 4. 5. 6.
Adams JE: Naloxone reversal of analgesia produced by brain stimulation in the human. Pain 1976; 2: 161-166. Hosobuchi Y, Adams JR, Linchitz R: Pain relief by electrical stimulation of central gray matter in humans and its reversal by naloxone. Science 1977; 197: 183-186. Goldstein A: Opioid peptides in pituitary and brain. Science
1976; 193: 1091-1096.
11.
12.
Pert CB, Aposhian D, Snyder SH: Phylogenetic distribution of opiate receptor binding. Brain Res 1974; 75: 356-361. Snyder SH, Matthyse S, eds: Neurosci Res Program Bull 1975;
13: 166.
Hughes J, Smith TW, Kosterliz HW, et al: Identification of two related pentapeptides from the brain with potent opiate agonist activity. Nature 1975; 258: 577-579. Atweh SF, Kuhar MK: Autoradiographic localization of opiate receptors in rat brain. I. Spinal cord and lower medulla. Brain Res 1977; 124: 53-67. Hiller JM, Pearson J, Simon EJ: Distribution of stereospecific binding of the potent narcotic analgesic etorphine in the human brain: predominance in the limbic system. Res Commun Chern Pathol Pharmacal 1973; 6: 1052-1062. Fields HL, Basbaum AI: Brain stem control of spinal pain-transmission neurons. Ann Rev Physiol 1978; 40: 217-248.
September 10, 1984
The American Journal of Medicine
7
PAIN MANAGEMENT SYMPOSIUM-FIELDS
13. 14. 15. 16. 17.
18.
19. 20.
21. 22. 23.
Dostrovsky JO, Deakin JFW: Periaqueductal gray lesions reduce morphine analgesia in the rat. Neurosci Lett 1977; 4: 99-103. Basbaum AI, Fields HL: Endogenous pain control mechanisms: review and hypothesis. Ann Neurol 1978; 4: 451-462. Basbaum AI, Clanton CF, Fields HL: Opiate and stimulus-produced analgesia: functional anatomy of a medulospinal pathway. Proc Natl Acad Sci USA 1976; 73: 4685-4688. Basbaum AI, Marley NJE, O'Keefe J, et al: Reversal of morphine and stimulus-produced analgesia by subtotal spinal cord lesions. Pain 1977; 3: 43-56. Murfin R, Bennett J, Mayer OJ: The effect of dorsolateral spinal cord (DLF) lesions on analgesia from morphine microinjected into the periaqueductal gray matter (PAG) of the rat. Neurosci Abstr 1976; 2: 946. Fields HL: An endorphin-mediated analgesia system: experimental and clinical observations. In: Martin JB, Reichlin S, Bick KL, eds. Neurosecretion and brain peptides. New York: Raven Press, 1981; 199-212. Akil H, Mayer OJ: Antagonism of stimulation-produced analgesia by p-CPA, a serotonin synthesis inhibitor. Brain Res 1972; 44: 692-697. Hayes RL, Newlon PG, Rosecrans JA, et al: Reduction of stimulation-produced analgesia by lysergic acid diethylamide, a depressor of serotonergic neural activity. Brain Res 1977; 122: 367-372. Vogt M: The effect of lowering the 5-hydroxytryptamine content of the rat spinal cord on analgesia produced by morphine. J Physiol (Lond) 1974; 236: 483-498. Tenen SS: Antagonism of the analgesic effect of morphine and other drugs by p-chlorophenylalanine, a serotonin depletor. Psychopharmacologia 1968; 12: 278-285. Oliveras JL, Hosobuchi Y, Redjemi F, et al: Opiate antagonist,
24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35.
naloxone, strongly reduces analgesia induced by stimulation of a raphe nucleus (centralis inferior). Brain Res 1977; 120: 221-229. Lasagna L: Drug interaction in the field of analgesic drugs. Proc R Soc Med 1965; 58: 978-983. Levine JD, Gordon NC, Jones RT, et al: The narcotic antagonist naloxone enhances clinical pain. Nature 1978; 272:826-827. Lewis JW, Cannon JT, Liebeskind JC: Opioid and non-opioid mechanisms of stress analgesia. Science 1980; 208: 623625. EI-Sobky A, Dostrovsky JO, Wall PO: Lack of effect of naloxone on pain perception in humans. Nature 1976; 263: 783-784. Levine JD, Gordon NC, Bornstein JC, et al: The role of pain in placebo analgesia. Proc Natl Acad Sci USA 1979; 76: 35283531. Watkins LR, Mayer OJ: Organization of endogenous opiate and nonopiate pain control systems. Science 1982; 216: 11851192. Evans FJ: The placebo response in pain reduction. In: Bonica JJ, ed. Advances in neurology, vol 4. New York: Oxford University Press, 1974; 289. Beecher HK: Measurement of subjective responses. New York: Oxford University Press, 1950. Houde RW, Wallenstein MS, Rogers A: Clinical pharmacology of analgesics. Clin Pharmacal Ther 1966; 1: 163. Liberman R: An experimental study of the placebo response under three different situations of pain. J Psychiatr Res 1964; 2: 233-246. Fields HL, Levine JD: Biology of placebo analgesia. Am J Med 1981; 70: 745-746. Botney M, Fields HL: Amitriptyline potentiates morphine analgesia by a direct action on the central nervous system. Ann Neurol 1983; 13: 160-164.
DISCUSSION Dr. Stephen Silberstein: Is there a theoretic reason for not using mixed agonist-antagonist narcotics in clinic settings? Dr. Howard Fields: Yes. In some patients with postoperative pain, these drugs will at best be ineffective and may possibly increase the pain. In general, in large studies of postoperative and cancer pain, a minority of patients seem to have a significant placebo analgesia-approximately 25 to 45 percent. Administration of an adequate dose of an opioid, such as morphine, can increase the percentage of those who get substantial pain relief to 95 percent. So I still think there are good grounds for using the opioids. Dr. William T Beaver: I agree. Antagonist analgesics, those agents used for their analgesic effects, tend to be less potent antagonists. When you administer pure antagonists, such as naloxone, hyperalgesia may occur. Dr. Peter Amadio Jr.: To what extent does the euphoria of the narcotic analgesics affect the patient's perception of pain? Dr. Charles lnturrisi: If by "euphoria" you mean that patients are "high," that is a very uncommon phenomenon. But, if you are speaking of the change in mood that accompanies pain relief, I think patients who experience pain relief· with narcotic analgesics certainly do have accompanying mood changes.
8
September 10, 1984
The American Journal of Medicine
Dr. Beaver: Postoperatively, particularly in cancer patients, you see concomitant improvement in mood and relief of tension and anxiety, but euphoria in the sense of "getting high" is rare. It is our experience that less than one of every 100 patients actually feels "high" as a side effect after narcotic administration. Dr. lnturrisi: Dr. Fields, I would like to ask you a question about what I see as the challenge of the 1980s and maybe the 1990s. How can we learn to maximize the interactions of the opioids with the endogenous pain-modulating system so that we can maximize analgesia and minimize adverse effects? Dr. Fields: A study by Houde Wallenstein and Rogers [32] showed that, with multiple doses of placebo, up to 90 percent of the patients will show a positive response. This suggests that it is not so much the person's personality, anatomy, neuroanatomy, or chemistry that affects pain relief as the way the patient interprets the particular situation in which he or she finds himself or herself. I think we need to go a long way toward the dissection of the psyche before we will be able to pursue pain relief strategies on the basis of the individual patient. Giving somebody a sugar pill may not be as effective a placebo as an injection, and one injection might not be as potent as a series of injections. An operation may be the most potent placebo of all.
An endogenous central nervous system pain-modulating network, with links in the mid brain, medulla, and spinal cord, has recently been discovered. This system produces analgesia by interfering with afferent transmission of neural messages produced by intense stimuli. Although other neurotransmitters are involved, the analgesia produced by this system depends on the release of endogenous opioid substances, generically referred to as endorphins. The system is set in motion by clinically significant pain-such as that resulting from bony fractures or postoperative pain. The analgesia network monitors the pain and controls it at the level of the spinal cord. Complex psychologic factors play an important role in the variability of perceived pain, partly because of their ability to trigger this pain-suppressing system. For example, this system contributes to the analgesic potency of placebo administration and is also activated by stress. Knowledge of this analgesia system has greatly expanded our understanding of the mechanisms underlying pain management. Opiates, like morphine and meperidine, produce analgesia by mimicking the action of endorphins in the brain. Tricyclic drugs may produce analgesia by enhancing the nonendorphin links of the same system. Future research on this system will provide new insights and, consequently, new approaches to the management of pain.
HOWARD L. FIELDS, M.D. San Francisco, California
From the Departments of Neurology and Physiology, University of California School of Medicine, San Francisco, California. Requests for reprints should be addressed to Dr. Howard L. Fields, University of California School of Medicine, Department of Neurology, San Francisco, California 94143.
2
September 10, 1984
Within the last decade or so, remarkable progress has been made in understanding the anatomy and physiology of pain. The two most important recent advances in thi~ field have been (1) the discovery of a highly organized central nervous system network for the monitoring and modulation of pain, and (2) the isolation and description of the endorphins, which are synthesized by nerve cells and have the pharmacologic properties of morphine. The mechanisms by which this endogenous neural system produces analgesia will be reviewed. Before describing the pain-suppression system in detail, it will be useful to briefly outline the neural basis of pain sensation. Pain functions as a warning system. The neural pathways that transmit pain messages signal that tissue damage is occurring or about to occur. The pain-signaling system is required to protect the body, and it works well most of the time. On the other hand, the characteristic unpleasantness of pain has made analgesia one of the most important things that physicians can provide for their patients. As documented in this symposium, progress has been made in refining our tools for pain management. Part of this progress is due to an improved understanding of the neural mechanisms that underlie the perception of pain.
The American Journal of Medicine
PAIN MANAGEMENT SYMPOSIUM-FIELDS
Normally, intense stimuli or tissue damage sets in motion processes that lead to activity in primary afferent nociceptors. Although these processes are not fully understood, they are similar, if not identical, to those that underlie inflammation. These are fully discussed by Dr. Galin elsewhere in this symposium issue. The peripherally acting analgesics, such as aspirin, acetaminophen, and the nonsteroidal anti-inflammatory agents, probably act on these processes. Once the primary afferent nociceptor ("pain receptor") is activated, its message is transmitted over small diameter axons to the spinal cord. This message is subjected to two important modifying influences at this stage: one is the action of the large, myelinated afferent fibers, and the other is the influence of the sympathetic nervous system. The large, myelinated afferent fibers have no specific response to intense stimuli; in fact, they act to inhibit pain transmission at the level of the spinal cord. This phenomenon is well established and probably forms the basis of certain local therapies, such as massage, acupuncture, and transcutaneous electrical nerve stimulation (TENS). That the sympathetic nervous system can produce pain is dramatically illustrated by the syndrome of causalgia, the severe burning pairi that sometimes follows nerve injury. This pain is immediately and completely relieved by sympathetic block. When the pain signal arrives at the spinal cord, it activates cells there that, in turn, relay the pain signal to the brainstem and/or the thalamus. From the thalamus, the message is relayed to the cortex; it is there, presumably, that the activity occurs that underlies the conscious experience of pain. Before leaving the subject of pain transmission, it would be worthwhile to clarify the differehce between the sensory aspect of pain and the unpleasantness that usually accompanies pain. The sensory aspect involves the identification of the sensation as painful. The suffering aspect has more to do with the desire to escape the pain. Whereas the sensory aspect is fairly similar among individual persons and in the same person over time, the suffering or tolerance aspect is variable and may depend on personality, situational, and cultural factors. For example, most of us will either ignore a headache or treat ourselves with a mild analgesic, but if we were told that the headache was due to a brain tumor, the anguish and suffering would be immeasurable. This suffering aspect is an important source of the variability of clinical pain. It can be dealt with by a variety of approaches, discussed by Addison and Posner in their reports in this symposium issue. STIMULATION-PRODUCED ANALGESIA
The first evidence that this analgesia system existed was the observation that, in rats, electrical stimulation of discrete areas of the brain-primarily the mid-brain periaqueductal gray-profoundly inhibited responses to painful stimuli [1-3]. This analgesic effect was found to be
strikingly selective. Although the animals remained alert, active, and normally responsive to innocuous stimuli, noxious stimuli did not produce the expected vocalization, biting, or escape behavior. These findings were confirmed in patients with intractable pain who had stimulating electrodes implanted at sites near those in the periaqueductal gray from which this stimulation-produced analgeSia was elicited in animals (4,5]. As with the laboratory models, no consistent sensory or motor effects other than relief of pain were associated with the procedure, indicating that in human beings, as in animals, the system activated iS specifically designed for pain control. The induction of analgesia, separated from any other effect, is of fundamental importance in establishing pain modulation as a distinct physiologic function of the central nervous system. OPIATE ANALGESIA
At about the time this pain-modulation system was described physiologically and clinically, there were findings of equal significance in the field of neuropharmacology. First was the discovery of receptors ih the brain to which opiate compounds attach specificaliy [6-8]. The importance of these binding sites on receptors was established by showing that the potency of narcotic drugs for relief of human clinical pain correlates directly with their binding affinity for the receptor (8]. The discovery of these opiate receptors triggered a search for a substance produced within the central nerve ous system that would bind to the receptor. The first fruit of this search was the isolation from the brain of two endogenous pentapeptides, leucine and methionine enkephalin. These substances have morphine-like activity that is blocked by haloxone, a specific antagonist to morphine [9]. Subsequent research uncovered other endogenous, morphine-like compounds that are now classified, categorically, as endorphins. Several distinct endorphins with different peptide sequences are present in different cell populations of the brain; pituitary, adrenal, gut, and sympathetic nervous system. For example, beta endorphin is present in the pituitary gland, and enkephalins are present in the adrenal gland. Both have morphine-like actions. Knowledge of endorphins has provided considerable insight into the manner in which narcotic analgesics relieve pain. Drugs like morphine presumably work by mimicking the action of endorphins at synapses in the painmodulation network. In addition, endorphins have provided both an anatomic and chemical explanation for the phenomenon of stimulation-produced analgesia. That stimulation-produced analgesia and opiate analgesia share a common neural pathway is documented by a number of discoveries, such as: • Opiate receptors and endorphins are normally present in relatively high concentrations in areas such as the peri-
September 10, 1984
The American Journal of Medicine
3
PAIN MANAGEMENT SYMPOSIUM-FI ELDS
Figure 1. The endogenous pain-modulation system. A = mid-brain level; B = medullary level; C = spinal level; DLF = dorsolateral funiculus; E = endorphin; SHT = serotoinin; NE = norepinephrine; NRM = nuc/ei.Js raphe magnus; PAG = periaqueductal gray; R = reticular formation; T = pain transmission neuron (adapted with permission from [14].
aqueductal gray [1 0,11 ], from which stimulation-produced analgesia is elicited. • A precise anatomic correlation exists among opiate receptors, endorphin distribution, and the nuclei from which analgesia can be elicited by electrical stimulation or microinjection of opiates [12]. • Naloxone, a specific opiate antagonist, partially blocks , stimulation-produced analgesia [4,5]. • Lesions of the periaqueductal gray block analgesia from systemic opiates [13]. ANATOMIC PATHWAY
A great deal of effort has been put into mapping the anatomic connections of this pain-modulation system. Present evidence indicates that it is a descending system and that it controls pain at the level of the spinal cord by blocking neural activity produced by intense (noxious) stimuli. The first evidence that this system produces analgesia at
4
September 10, 1984
The American Journal of Medicine
spinal cord levels was the observation that discrete lesions of certain descending spinal pathways block both stimulation-produced analgesia [14, 15] and the analgesia produced by systemic injection of morphine [16]. Where ar'3 the cells of origin of this descending pathway for pain suppression? A substantial number lie in two major brainstem groups [16-18]. The one of most interest to us lies near the midline of the medulla (raphe nuclei) and includes many serotonin-containing cells [18]. There is an extensive body of evidence implicating these serotonergic neurons in pain control. For example, serotonin depletion counteracts stimulation-produced analgesia [19,20] and antagonizes opiate analgesia as well [21 ,22]. The nerve cells in this region of medulla send their axons in a well-defined pathway to terminate in the spinal cord. The termination pattern is fascinating and revealing in its specificity. The terminals of these brain stem painmodulating cells are directly upon the very spinal cord cells that respond to pain-producing stimuli. Furthermore, stimulation of the pain-modulating cells in the brainstem profoundly inhibits the spinal pain-responsive cells. Thus, the pain-suppression system blocks the pain message in the spinal cord so that it does not reach higher centers. There is very extensive anatomic overlap of the spinal terminals of pain-modulating cells, opiate receptors, enkephalin, and the location of spinal neurons that respond to noxious stimuli. This close correlation suggests that the mechanism of analgesia from brainstem stimulation involves the endorphins [14]. The reversal of stimulationproduced analgesia by the opiate antagonist naloxone [23] is consistent with this view. Figure 1 summarizes the anatomic and the physiologic aspects of the endogenous pain-modulation system. At the mid-brain level, you find the periaqueductal gray, an important locus for stimulation-produced analgesia. This area is rich in endorphins and opiate receptors, although the anatomic details of the endorphin connections are not known. Microinjection of small amounts of opiates into the periaqueductal gray also produces analgesia. At the medullary level, serotonin-containing cells of the nucleus raphe magnus and the adjacent reticular formation receive excitatory input from periaqueductal gray and, in turn, send impulses along efferent fibers to the spinal cord. At the spinal level, efferent fibers from the medulla travel in a discrete pathway in the dorsolateral funiculus to terminate among pain-transmission cells concentrated in the dorsal horn, where they exert an inhibitory effect specifically on pain-transmission neurons. Part of this effect is probably exerted via endorphin-containing interneurons. Pain-transmission neurons project to supraspinal sites and indirectly contact the cells of the descending analgesia system in the periaqueductal gray and nucleus raphe magnus, thus establishing a negative feedback loop that we will discuss in greater detail shortly.
PAIN MANAGEMENT SYMPOSIUM-FIELDS
Four Major Operating Characteristics of Endogenous Pain-Modulation System
OPERATING CHARACTERISTICS OF PAINMODULATION PATHWAY
TABLE I
There are four major operating characteristics of the endogenous pain-modulation system (Table 1): First, the system is at least partly endorphin-mediated. Second, it involves serotonergic neurons. Third, it depends on activation of descending pathways from the brain stem that inhibit pain-transmission neurons at the spinal cord. Finally, it is effectively activated by stress and pain-producing stimuli. Figure 2 illustrates the essential features of the intrinsic analgesia system. The basic element of the endogenous pain-modulation system is a negative feedback loop that monitors the output of the pain-transmission system and controls its input. The loop is set in motion when a potentially painful stimulus excites the pain-transmission system. When the activity in these neurons is sufficient to reach the level of the pain threshold, pain is perceived. Simultaneously, the activity of the pain-transmission neurons activates the pain-suppression system. The pain-suppression system, in turn, feeds back upon, and thus controls, the spinal pain-transmission neurons. The effectiveness of the pain-suppression system can be enhanced by administering analgesics, such as morphine or codeine, or by stimulating the analgesia network electrically. This reduces the perceived pain intensity. Conversely, disruption of the loop, either by opiate or serotonin antagonists, will increase the perceived intensity of the pain. Although the simplified schema of Figure 2 accounts for many of the aspects of pain transmission and modulation, it is far from complete, implying only one transmission and one modulation system. There is considerable evidence to indicate that there are at least two, and possibly multiple, ascending pain-transmission systems and several descending pain-modulation systems. The descending feedback loops may involve either naloxone-sensitive (endorphin-mediated) or naloxone-insensitive (nonendorphinergic) pathways. [14]. Pain modulation is, in part, a consequence of pain-producing stimuli. It follows, then, that pain perceived in re-
At least partly endorphin-mediated Involves serotonergic mechanisms Depends on activation of descending efferent pathways that inhibit pain-transmission neurons Effectively activated by noxious stimulation
sponse to a sustained stimulus should reach a peak intensity and, having activated the analgesia system, decline to a steady level. In other words, adaptation to noxious stimulation would be expected to result in part from activation of the pain-suppression system; the level of pain perceived would be less than that produced by pain transmission in the absence of the feedback modulation. The fact that postoperative pain is made worse by the administration of naloxone, and that naloxone is demonstrably worse than placebo in managing clinical pain, is consistent with this view [24,25]. NALOXONE, PLACEBO, AND ENDORPHIN-MEDIATED ANALGESIA
Studies with naloxone, often in conjunction with placebo studies, have been a bulwark for the existence of an endorphin-mediated analgesia system. What have these studies shown? To begin with, brief, superficial pains do not seem to cause endorphin release [26,27]. Noxious stimuli must be significantly above the pai~ threshold and have a prolonged duration in order to produce analgesic actions that are blocked by naloxone [28]. This indicates that endorphins are released only when the pain level is significantly above threshold intensity, perhaps only when a requisite degree of stress has been produced. In order to understand more fully the actions of endorphins on pain, one must realize that many, if not most, painful conditions remit without specific treatment. Common examples of this are headache, tic douloureux, dysmenorrhea, and herpetic pain. In these cases, patients go from a pain-free state to experiencing severe pain and then back to having either less severe pain or no pain. We
Pain Transmission System
Noxious Input
Figure 2. Essentials of the endorphinmediated negative feedback loop of the endogenous pain-modulation system (adapted with permission from [14]).
1------------.. . --+ Medullary Center
Pain
~Exogenous Opiates
Pain Suppression System
September 10, 1984
The American Journal of Medicine
5
PAIN MANAGEMENT SYMPOSIUM-FIELDS
Natural history
A
Improvement Severity
Placebo effect B
Severity Observed
l Placebo
do not understand this cycling, but clinical experience has established that this is the usual course of pain in most patients (Figure 3). Obviously, under these circumstances, all treatments will be followed by clinical improvement. This improvement is due to neither medication nor a placebo effect but is indicative of a reduction in disease activity. Whether the endorphins play a role in this "natural" cycling, observed in certain painful conditions, is still an open question. Figure 3 illustrates what is meant by placebo response. As discussed, the placebo response must be distinguished from the tendency of some painful conditions to remit spontaneously. Evidence that the placebo effect involves the release of endorphins has been seen in patients with postoperative dental pain; naloxone produces significant hyperalgesia in patients who respond to a placebo but has little effect in those who do not [25]. Painful stimuli that are stressful, inescapable, or associated with anxiety have been shown to be particularly effective in producing naloxone-reversible, and thus presumably endorphin-mediated, analgesia [29]. Likewise, anxiety has been correlated positively with effectiveness of placebo response [30]. In a recent study, a placebo was found to be more frequently effective for clinical than for experimentally induced pain, a result the investigators attributed to the increased stress of the clinical situation [31].
6
September 10, 1984
The American Journal of Medicine
Figure 3. The placebo response (adapted with permission from [34]).
Several studies have shown that there do not seem to be any specific personality characteristics that have a bearing on the ability to mobilize a placebo response (and hence, endorphins). In one such study, involving a group of patients with cancer pain who received a placebo several times, 90 percent reported a significant decrease in pain at least once [32], indicating that virtually everyone is capable of responding to a placebo. In another study [33], a group of women on an obstetric service were given a placebo once for relieving labor pain, once for postpartum pain, and once for experimental ischemic pain. If a woman showed a positive response to the placebo in one situation, the probability that she would show a positive response in a second situation was not greater than that expected by chance. In other words, situational factors were more important than personality traits in determining responsiveness to a placebo. The findings of these two studies are very intriguing because they suggest that, if we could create the appropriate environment, we might be able to produce placebo analgesia consistently in most people. One fact we do know in regard to creating an "appropriate" environment is that the interaction between what the physician communicates and what the patient expects is critical for the production of a placebo response. Of course, what a physician says and how it is said will de-
PAIN MANAGEMENT SYMPOSIUM-FIELDS
pend, in turn, on what the physician believes about the treatment being administered. The analgesic response to a placebo is enhanced by positive suggestion; conversely, even effective analgesic agents may have their overall efficacy diminished if the physician communicates doubt [34]. Thus, the outlines of an endogenous pain-modulation system, activated in the brain stem and functioning via descending connections to control afferent pain transmission, have been presented. Evidence suggests that the system relies on the release of endorphins even though other neurotransmitters are also involved. The pain-modulation system acts like a negative feedback loop that is set in motion by stress and noxious stimuli and that limits pain perception by interfering with afferent input at the level of the spinal cord. CLINICAL IMPLICATIONS
Knowledge of the pain-modulation systems and the endorphins provides a better basis for rational pain management. First of all, because the pain-transmission and painmodulation systems are separate entities and operate by distinct mechanisms, there are at least two general classes of approaches to pain management. One set of approaches is to block the transmission system; the other is to turn on or enhance the suppression system. Drugs like aspirin, acetaminophen, or the nonsteroidal anti-inflammatory agents block transmission at sites in peripheral tissues and/or at nonopioid receptors in the central nervous system. In contrast, the opiates, such as morphine or codeine, act at specific sites in the brain to activate the pain-suppression system. Because the drugs act at different sites by distinct mechanisms, their use in combination is rational, if neither class alone is sufficient for pain control. One new development in the use of opiates has been to apply them locally, at the level of the spinal cord. Remem-
ber that both endorphins and opiate receptors are present in the spinal cord. This leads to the prediction that the application of morphine (or any narcotic analgesic) directly to the cord-either in the epidural or intrathecal space-will produce local analgesia. In fact, this has been verified in a variety of painful conditions. The analgesia produced is powerful, long-lasting, and reversible with opiate antagonists. This method is being used clinically for lower body pains in labor, for postoperative pain, and for cancer patients. Some cancer patients have cannulas placed in their epidural space for weeks at a time, with an implanted reservoir to deliver drug continuously. The fact that the pain-suppression system has nonendorphin synapses raises the possibility that nonopiate drugs may be potent centrally acting analgesic agents. For example, serotonin plays a major role producing analgesia, and it is well established that tricyclic antidepressant drugs, such as amitriptyline, are effective in certain chronic painful conditions (for example, migraine, postherpetic neuralgia). The tricyclics block the removal of serotonin from the synaptic cleft and thus prolong and increase its action. In animal pain models, tricyclics have a direct analgesic action and potentiate the analgesic action of opiates [35]. The tricyclics represent a totally new class of centrally acting, nonaddicting pain-killers. Finally, it should be stressed that almost any therapeutic manipulation has the potential to elicit placebo analgesia and that this effect is at least partly mediated by endorphins. Thus, the common practice of using a placebo to diagnose psychogenic pain is based on erroneous assumptions. In fact, there is evidence to suggest that more intense pain or stress is associated with greater activation of the endorphin-mediated analgesia system; therefore, up to a point, the more intense the pain, the more likely the response to a placebo. The future of pain management will include closer attention to psychologic and physical methods for activating the analgesia system.
REFERENCES 1. 2.
Mayer DJ, Liebeskind JC: Pain reduction by focal electrical stimulation of the brain: an anatomical and behavioral analysis. Brain Res 1974; 68: 73-93. Mayer DJ, Wolfe TL, Akil H, et al: Analgesia from electrical stimulation in the brain stem of the rat. Science 1971; 174: 1351-
7.
8. 9.
1354. 3.
Reynolds DV: Surgery in the rat during electrical analgesia induced by focal brain stimulation. Science 1969; 164: 444-
10.
445. 4. 5. 6.
Adams JE: Naloxone reversal of analgesia produced by brain stimulation in the human. Pain 1976; 2: 161-166. Hosobuchi Y, Adams JR, Linchitz R: Pain relief by electrical stimulation of central gray matter in humans and its reversal by naloxone. Science 1977; 197: 183-186. Goldstein A: Opioid peptides in pituitary and brain. Science
1976; 193: 1091-1096.
11.
12.
Pert CB, Aposhian D, Snyder SH: Phylogenetic distribution of opiate receptor binding. Brain Res 1974; 75: 356-361. Snyder SH, Matthyse S, eds: Neurosci Res Program Bull 1975;
13: 166.
Hughes J, Smith TW, Kosterliz HW, et al: Identification of two related pentapeptides from the brain with potent opiate agonist activity. Nature 1975; 258: 577-579. Atweh SF, Kuhar MK: Autoradiographic localization of opiate receptors in rat brain. I. Spinal cord and lower medulla. Brain Res 1977; 124: 53-67. Hiller JM, Pearson J, Simon EJ: Distribution of stereospecific binding of the potent narcotic analgesic etorphine in the human brain: predominance in the limbic system. Res Commun Chern Pathol Pharmacal 1973; 6: 1052-1062. Fields HL, Basbaum AI: Brain stem control of spinal pain-transmission neurons. Ann Rev Physiol 1978; 40: 217-248.
September 10, 1984
The American Journal of Medicine
7
PAIN MANAGEMENT SYMPOSIUM-FIELDS
13. 14. 15. 16. 17.
18.
19. 20.
21. 22. 23.
Dostrovsky JO, Deakin JFW: Periaqueductal gray lesions reduce morphine analgesia in the rat. Neurosci Lett 1977; 4: 99-103. Basbaum AI, Fields HL: Endogenous pain control mechanisms: review and hypothesis. Ann Neurol 1978; 4: 451-462. Basbaum AI, Clanton CF, Fields HL: Opiate and stimulus-produced analgesia: functional anatomy of a medulospinal pathway. Proc Natl Acad Sci USA 1976; 73: 4685-4688. Basbaum AI, Marley NJE, O'Keefe J, et al: Reversal of morphine and stimulus-produced analgesia by subtotal spinal cord lesions. Pain 1977; 3: 43-56. Murfin R, Bennett J, Mayer OJ: The effect of dorsolateral spinal cord (DLF) lesions on analgesia from morphine microinjected into the periaqueductal gray matter (PAG) of the rat. Neurosci Abstr 1976; 2: 946. Fields HL: An endorphin-mediated analgesia system: experimental and clinical observations. In: Martin JB, Reichlin S, Bick KL, eds. Neurosecretion and brain peptides. New York: Raven Press, 1981; 199-212. Akil H, Mayer OJ: Antagonism of stimulation-produced analgesia by p-CPA, a serotonin synthesis inhibitor. Brain Res 1972; 44: 692-697. Hayes RL, Newlon PG, Rosecrans JA, et al: Reduction of stimulation-produced analgesia by lysergic acid diethylamide, a depressor of serotonergic neural activity. Brain Res 1977; 122: 367-372. Vogt M: The effect of lowering the 5-hydroxytryptamine content of the rat spinal cord on analgesia produced by morphine. J Physiol (Lond) 1974; 236: 483-498. Tenen SS: Antagonism of the analgesic effect of morphine and other drugs by p-chlorophenylalanine, a serotonin depletor. Psychopharmacologia 1968; 12: 278-285. Oliveras JL, Hosobuchi Y, Redjemi F, et al: Opiate antagonist,
24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35.
naloxone, strongly reduces analgesia induced by stimulation of a raphe nucleus (centralis inferior). Brain Res 1977; 120: 221-229. Lasagna L: Drug interaction in the field of analgesic drugs. Proc R Soc Med 1965; 58: 978-983. Levine JD, Gordon NC, Jones RT, et al: The narcotic antagonist naloxone enhances clinical pain. Nature 1978; 272:826-827. Lewis JW, Cannon JT, Liebeskind JC: Opioid and non-opioid mechanisms of stress analgesia. Science 1980; 208: 623625. EI-Sobky A, Dostrovsky JO, Wall PO: Lack of effect of naloxone on pain perception in humans. Nature 1976; 263: 783-784. Levine JD, Gordon NC, Bornstein JC, et al: The role of pain in placebo analgesia. Proc Natl Acad Sci USA 1979; 76: 35283531. Watkins LR, Mayer OJ: Organization of endogenous opiate and nonopiate pain control systems. Science 1982; 216: 11851192. Evans FJ: The placebo response in pain reduction. In: Bonica JJ, ed. Advances in neurology, vol 4. New York: Oxford University Press, 1974; 289. Beecher HK: Measurement of subjective responses. New York: Oxford University Press, 1950. Houde RW, Wallenstein MS, Rogers A: Clinical pharmacology of analgesics. Clin Pharmacal Ther 1966; 1: 163. Liberman R: An experimental study of the placebo response under three different situations of pain. J Psychiatr Res 1964; 2: 233-246. Fields HL, Levine JD: Biology of placebo analgesia. Am J Med 1981; 70: 745-746. Botney M, Fields HL: Amitriptyline potentiates morphine analgesia by a direct action on the central nervous system. Ann Neurol 1983; 13: 160-164.
DISCUSSION Dr. Stephen Silberstein: Is there a theoretic reason for not using mixed agonist-antagonist narcotics in clinic settings? Dr. Howard Fields: Yes. In some patients with postoperative pain, these drugs will at best be ineffective and may possibly increase the pain. In general, in large studies of postoperative and cancer pain, a minority of patients seem to have a significant placebo analgesia-approximately 25 to 45 percent. Administration of an adequate dose of an opioid, such as morphine, can increase the percentage of those who get substantial pain relief to 95 percent. So I still think there are good grounds for using the opioids. Dr. William T Beaver: I agree. Antagonist analgesics, those agents used for their analgesic effects, tend to be less potent antagonists. When you administer pure antagonists, such as naloxone, hyperalgesia may occur. Dr. Peter Amadio Jr.: To what extent does the euphoria of the narcotic analgesics affect the patient's perception of pain? Dr. Charles lnturrisi: If by "euphoria" you mean that patients are "high," that is a very uncommon phenomenon. But, if you are speaking of the change in mood that accompanies pain relief, I think patients who experience pain relief· with narcotic analgesics certainly do have accompanying mood changes.
8
September 10, 1984
The American Journal of Medicine
Dr. Beaver: Postoperatively, particularly in cancer patients, you see concomitant improvement in mood and relief of tension and anxiety, but euphoria in the sense of "getting high" is rare. It is our experience that less than one of every 100 patients actually feels "high" as a side effect after narcotic administration. Dr. lnturrisi: Dr. Fields, I would like to ask you a question about what I see as the challenge of the 1980s and maybe the 1990s. How can we learn to maximize the interactions of the opioids with the endogenous pain-modulating system so that we can maximize analgesia and minimize adverse effects? Dr. Fields: A study by Houde Wallenstein and Rogers [32] showed that, with multiple doses of placebo, up to 90 percent of the patients will show a positive response. This suggests that it is not so much the person's personality, anatomy, neuroanatomy, or chemistry that affects pain relief as the way the patient interprets the particular situation in which he or she finds himself or herself. I think we need to go a long way toward the dissection of the psyche before we will be able to pursue pain relief strategies on the basis of the individual patient. Giving somebody a sugar pill may not be as effective a placebo as an injection, and one injection might not be as potent as a series of injections. An operation may be the most potent placebo of all.