Psychological Aspects of Pain: Implications for Neural Blockade
Joel Katz
Ronald Melzack
Pain is a personal, subjective experience influenced by cultural learning, the meaning of the situation, attention, and other psychological variables. Pain processes do not begin with the stimulation of receptors. Rather, injury or disease produces neural signals that enter an active nervous system that (in the adult organism) is the substrate of past experience, culture, anxiety, and so forth. These brain processes actively participate in the selection, abstraction, and synthesis of information from the total sensory input. Pain, then, is not simply the end product of a linear sensory transmission system; it is a dynamic process that involves continuous interactions among complex ascending and descending systems.
This chapter focuses upon four areas of interest to anesthesiologists and psychologists: (a) the major psychological contributions to pain; (b) theories of pain, which are based on psychological assumptions of the nature of perception, including the gate control and neuromatrix models; (c) the measurement of pain; and (d) labor pain, which is influenced by anesthetic blocks as well as by manipulating psychological variables. Several topics surveyed in this chapter, because of their broad scope and special importance to pain management, are explored in greater detail within dedicated chapters elsewhere in this volume. Examples of the latter include social and cultural influences upon pain and disability (Chapter 29) and the placebo effect (Chapter 36).
Psychological Contributions to Pain
When compared with vision or hearing, the perception of pain seems simple, urgent, and primitive. We expect the nerve signals evoked by injury to “get through,” unless we are unconscious or anesthetized. But experiments and clinical observations show that pain is much more variable and modifiable than many people have believed in the past. Pain differs from person to person and from culture to culture. Stimuli that produce intolerable pain in one person may be tolerated without a whimper by another. Pain perception, then, cannot be defined simply in terms of particular kinds of stimuli. Rather, it is a highly personal experience that depends in part on psychological factors that are unique to each individual (1).
Cultural Determinants
It is often asserted that variations in pain experience from person to person are due to different “pain thresholds”; however, several thresholds are related to pain, and it is important to distinguish among them. Typically, thresholds are measured by applying a stimulus, such as electric shock or radiant heat, to a small area of skin and gradually increasing the intensity. Four thresholds can be measured by this technique: (a) sensation threshold (or lower threshold)—the lowest stimulus value at which a sensation such as tingling or warmth is first reported; (b) pain perception threshold—the lowest stimulus value at which the person reports that the stimulation feels painful; (c) pain tolerance (or upper threshold)—the lowest stimulus level at which the subject withdraws or asks to have the stimulation stopped; and (d) encouraged pain tolerance—the same as (c), but the person is encouraged to tolerate higher levels of stimulation.
Evidence now suggests that that all people, regardless of cultural background, have a uniform sensation threshold. Sternbach and Tursky (2) made careful measurements of sensation threshold, using electric shock as the stimulus, in American-born women belonging to four different ethnic groups: Italian, Jewish, Irish, and Old American. They found no differences among the groups in the level of shock that was first reported as producing a detectable sensation. The sensory conducting apparatus, in other words, appears to be essentially similar in all people, so that a given critical level of input always elicits a sensation.
Cultural background, however, has a powerful effect on the pain perception threshold. For example, levels of radiant heat that are reported as painful by people of Mediterranean origin (such as Italians and Jews) are described merely as warmth by Northern Europeans (3). Similarly, Nepalese porters on a climbing expedition are much more stoic than the Occidental visitors for whom they work. Even though both groups are equally sensitive to changes in electric shock, the Nepalese porters require much higher intensities before they call them painful (4).
The most striking effect of cultural background, however, is on pain tolerance levels. Sternbach and Tursky (2) report that the levels at which subjects refuse to tolerate electric shock, even when they are encouraged by the experimenters, depend
in part on the ethnic origin of the subject. Women of Italian descent tolerate less shock than women of Old American or Jewish origin. In a similar experiment (5) in which Jewish and Protestant women served as subjects, the Jewish, but not the Protestant, women increased their tolerance level after they were told that their religious group tolerated pain more poorly than others.
in part on the ethnic origin of the subject. Women of Italian descent tolerate less shock than women of Old American or Jewish origin. In a similar experiment (5) in which Jewish and Protestant women served as subjects, the Jewish, but not the Protestant, women increased their tolerance level after they were told that their religious group tolerated pain more poorly than others.
These differences in pain tolerance reflect different ethnic attitudes toward pain. Zborowski (6) found that Old Americans have an accepting, matter-of-fact attitude toward pain and pain expression. They tend to withdraw when the pain is intense, and cry out or moan only when they are alone. Jews and Italians, on the other hand, tend to be vociferous in their complaints and openly seek support and sympathy. The underlying attitudes of the two groups, however, appear to be different. Jews tend to be concerned about the meaning and implications of the pain, whereas Italians usually express a desire for immediate pain relief.
The main findings of these early studies on ethnic and cultural differences have been supported by more recent empirical work using laboratory-induced pain in a variety ethnic groups, including African Americans, White Americans, White British, and South Asians. Pain thresholds do not differ by ethnicity (7). In contrast, heat pain tolerance levels for intensity and unpleasantness ratings show differences between ethnic groups (8,9,10). Differences in pain tolerance may be partially accounted for by other factors, such as hypervigilance and daily levels of pain (8,9), but even after statistically controlling for these variables, ethnicity remains a significant factor.
Meaning of the Pain-Producing Situation
Considerable evidence shows that people attach variable meaning to pain-producing situations and that the meaning greatly influences the degree and quality of pain they feel. Beecher (11) observed that soldiers wounded in battle rarely complained of pain, whereas civilians with similar surgical wounds usually claimed that they were in severe pain. Beecher (11) concluded the following from his study:
The common belief that wounds are inevitably associated with pain, and that the more extensive the wound the worse the pain, was not supported by observations made as carefully as possible in the combat zone. … The data state in numerical terms what is known to all thoughtful clinical observers: there is no simple direct relationship between the wound per se and the pain experienced. The pain is in very large part determined by other factors, and of great importance here is the significance of the wound. … In the wounded soldier [the response to injury] was relief, thankfulness at his escape alive from the battlefield, even euphoria; to the civilian, his major surgery was a depressing, calamitous event.
A similar study (12) of Israeli soldiers with traumatic amputations after the Yom Kippur War provided similar observations. Most of the wounded men spoke of their initial injury as painless and used neutral terms such as “bang,” “thump,” or “blow” to describe their first sensation. They often volunteered their surprise that the injury did not hurt.
Melzack, Wall, and Ty (13) examined the features of acute pain in patients at an emergency clinic. Patients who had severe, life-threatening injuries or who were agitated, drunk, or “in shock” were excluded from the study. Of 138 patients who were alert, rational, and coherent, 51 (37%) stated that they did not feel pain at the time of injury. The majority of these patients reported onset of pain within an hour of injury, although the delays were as long as 9 hours or more in some patients. The predominant emotions of the patients were embarrassment at appearing careless or worry about loss of wages.
The occurrence of delays in pain onset was related to the nature of the injury. Of 46 patients whose injuries were limited to skin (lacerations, cuts, abrasions, burns), 53% had a pain-free period. Of 86 patients with deep-tissue injuries (fractures, sprains, bruises, amputation of a finger, stabs, and crushes), 28% had a pain-free period. The results indicate that the relation between injury and pain is highly variable and complex.
Attention Diversion and Distraction
If a person’s attention is focused on a potentially painful experience, he will tend to perceive pain more intensely than he would normally. Hall and Stride (14) found that the simple appearance of the word “pain” in a set of instructions made anxious subjects more likely to report a given level of electric shock as painful; the same level of shock was rarely reported to be painful when the word was absent from the instructions. Thus, the mere anticipation of pain is sufficient to raise the level of anxiety and thereby the intensity of perceived pain. Similarly, Hill, Kornetsky and associates (15,16) have shown that if anxiety is dispelled (by reassuring the subject that he has control over the pain-producing stimulus), a given level of electric shock or burning heat is perceived as significantly less painful than the same stimulus under conditions of high anxiety.
In contrast to the effects of attention on pain, it is well known that distraction of attention away from pain can diminish or abolish it. Distraction of attention may partly explain why boxers, football players, and other athletes sometimes sustain severe injuries during the excitement of the sport without being aware that they have been hurt.
Distraction of attention, however, is usually effective only when the pain is steady or rises slowly in intensity (17). If radiant heat is focused on the skin, for example, the pain may rise so suddenly and sharply that subjects are unable to control it by distraction. But when the pain rises slowly, people may use various stratagems to distract their attention from it. They often find that the pain actually levels off or decreases before it reaches the anticipated intolerable level. Distraction stratagems are used effectively by some people to control pain produced by dental drilling and extraction (18).
Perhaps the simplest method of coping with chronic pain is by means of cognitive and/or behavioral strategies that divert the patient’s attention away from pain to some other activity, object, or event (19). The scope of distracting and attention-diverting strategies is extensive, covering direct intentional efforts to reduce pain awareness through techniques such as relaxation and transformational imagery, as well as indirect approaches that accomplish similar goals without distraction being the primary objective (e.g., reading, painting, socializing). Although direct and indirect approaches appear equally effective in reducing awareness of pain, they may result in important differences in patients’ attributions and self-efficacy beliefs regarding pain control.
Distraction and attention-diverting strategies are effective for many reasons (20). For example, relaxation and related activities that require sustained, focused attention may actually reduce pain intensity, as well as pain awareness, by decreasing sympathetic nervous system activity (21). Engaging in distracting, social interactions with others may alter the frequency and use of maladaptive strategies, such as catastrophizing and avoidance, in part because of response-incompatibility and in part because of improved mood. At the same time, socializing may also lead to a reduced emphasis on the importance of pain.
The effectiveness of distraction appears to approach a limit as pain intensity increases (20). This is due to an inherent limitation of most distractors; they cannot compete successfully with pain for attentional resources. Simply put, pain may be too intense, unpleasant, and threatening. Recent technological advances have made it possible to employ three-dimensional (3-D) virtual reality (VR) audiovisual devices as adjuncts in the management of painful medical procedures. Virtual reality devices typically consist of a head-mounted display or helmet that provides an immersive 3-D experience. These devices are considered ideal distractors since they create an immersive, interactive experience involving multiple modalities, including vision, proprioception, and audition (22,23).
During invasive medical procedures, VR devices not only distract the patient by redirecting attention to the novel immersive environment, but, by excluding visual and auditory inputs from the surroundings outside the VR device itself, they also distract the patient away from pain and anxiety (24,25). Experimental and clinical studies of VR immersion show significant distraction effects (23). Pain unpleasantness scores and tolerance time to ischemic arm pain were significantly lower among healthy volunteer subjects receiving VR immersion compared with a control group (25).
Hoffman and colleagues have shown significant benefits of VR distraction in children and adults undergoing painful treatment for burn injury (24,26). Not only are pain intensity and unpleasantness reduced significantly by VR distraction, but the duration of time spent thinking about pain during the procedure was reduced by approximately 45% for experimental pain (27) and 75% for clinical pain (26). Moreover, experimental subjects undergoing heat pain stimulation reported an increase in “fun” ratings of 79% when VR and non-VR distraction were compared (27). Virtual reality–induced pain relief is accompanied by reduced activity in five brain regions typically activated by pain, including anterior cingulate cortex, primary and secondary somatosensory cortex, insula, and thalamus (27,28). These results suggest that VR distraction is a promising, new nonpharmacologic therapy for managing pain associated with a variety of painful medical procedures.
Fear of and Avoidance of Pain
Avoidance behavior is usually motivated by fear. Many patients with pain are afraid that movement will cause reinjury and pain (29,30). For example, in the days and weeks after surgery, it is common for pain to be exacerbated by movement. Depending on the location of the incision, deep breathing, coughing, laughing, and getting in and out of bed all may substantially increase pain. Many patients appropriately fear, and therefore avoid, moving about.
Understanding the personal meaning of the fear is important. Patients may fear that in sitting up or walking their stitches will break and the wound will split open. Or, they may fear these activities will cause internal damage. Other patients simply fear the increase in pain associated with activity; they may feel helpless in the presence of intense pain, or they may feel dependent on the nursing staff for pain relief. Apprehension about moving about after surgery is based on first-hand experience that activity causes increased pain. The misinterpretation of activity and pain as harmful engenders avoidance behaviors that may set the stage for decreased activity and increased pain and disability.
Although avoidance by social withdrawal and inactivity may reduce ongoing pain, particularly in the early recovery phase, the long-term consequence of these behaviors is reinforcement of the belief that avoidance prevents further pain. Such fear-avoidance beliefs are expectations about the consequences of certain actions such as physical activity or work upon pain (31). Avoidance behaviors (e.g., guarding, limping, social withdrawal) may be part of coping with ongoing pain or anticipated increases in pain. However, some of these behaviors may also represent maladaptive ways of coping with concurrent life difficulties. Unpleasant work obligations, aversive household chores, marital strife, and interpersonal difficulties may thus be avoided by the person in pain. Pain behaviors that serve a dual purpose of avoiding not only expected flare-ups in pain, but also other unpleasant life events are at high risk of becoming entrenched because of the multiple sources of negative reinforcement (32).
Patients’ fear-avoidance beliefs and negative expectancies about the consequences of activity are meant to reduce anticipated increases in pain through avoidance. Avoidance behaviors tend to increase in frequency as pain becomes chronic, so that pain behaviors and pain intensity become increasingly decoupled (33). This desynchrony sets the stage for reduced self-efficacy beliefs and further avoidance (34). Thus, in attempts to gain control over the pain through avoidant coping, patients report progressively less control over the pain. Self-management programs focused on behavioral exposure and nonavoidance lead to improved self-efficacy and a reduction in preoccupation with pain, because patients acquire increasingly realistic appraisals of the relationship between pain and behavior (34).
Although use of avoidant cognitive coping strategies for pain may be beneficial when pain is acute, long-term use of these strategies is associated with impaired psychosocial functioning, increased pain and disability, and loss of employment. Avoidant coping strategies are adaptive in the early period following an injury because they minimize ongoing pain, reduce the risk of exacerbation through further injury, and thus promote healing. However, once healing has occurred, these strategies are maladaptive because they promote continued isolation, inactivity, and faulty reality testing.
Recent controlled case reports show that pain-related fear can be effectively treated by in vivo exposure, in which patients are exposed to fear-eliciting and hierarchically ordered physical movements. Concomitant reductions in catastrophic thinking, pain intensity, and pain disability were observed (35,36).
Feelings of Control over Pain
It is now clear that the severity of postsurgical pain is significantly reduced when patients are taught how to cope with their pain. In a classic study by Egbert and associates, patients scheduled to undergo major surgery to remove the gallbladder, uterus, or portions of the digestive tract were given detailed information about the pain they would feel after the operation and how they could best cope with it. They were told where they would feel pain, how severe it could be, how long it could last, and that such pain is normal after an operation. They were also shown how to relax by using breathing and relaxation stratagems. Finally, they were told that total relaxation is difficult to achieve and that they should request medication if they were uncomfortable. The results showed that patients who received these instructions reported significantly less pain, asked for much less medication during recovery, and spent less time in hospital than a similar group of patients who received no instructions (37).
It was originally thought that providing information alone is sufficient psychological preparation to reduce the uncertainty and anxiety associated with major surgery. In this case,
however, knowledge may increase anxiety because of the certain expectation of pain and other discomforts. It is essential to provide the patient with skills to cope with the pain and anxiety—at the very least, to provide the patient with a sense of control. Recent studies have shown that simply giving patients information about their pain tends to make them focus on the discomforting aspects of the experience, thereby magnifying rather than reducing their pain. However, when patients are taught skills to cope with their pain, such as relaxation or distraction strategies, their pain is less severe (38). Other studies have shown that postsurgical pain intensity is directly proportional to the amount of anxiety perceived by the patient (39). Achieving a sense of control, then, appears to diminish both anxiety and pain.
however, knowledge may increase anxiety because of the certain expectation of pain and other discomforts. It is essential to provide the patient with skills to cope with the pain and anxiety—at the very least, to provide the patient with a sense of control. Recent studies have shown that simply giving patients information about their pain tends to make them focus on the discomforting aspects of the experience, thereby magnifying rather than reducing their pain. However, when patients are taught skills to cope with their pain, such as relaxation or distraction strategies, their pain is less severe (38). Other studies have shown that postsurgical pain intensity is directly proportional to the amount of anxiety perceived by the patient (39). Achieving a sense of control, then, appears to diminish both anxiety and pain.
Suggestion and Placebos
The influence of suggestion on the intensity of perceived pain is clearly demonstrated by studies of the effectiveness of placebos. Chapter 36 by Finniss and Benedetti surveys these studies in detail and discusses their implications for neural blockade, so the present description is deliberately concise. As Beecher carefully documented in the 1950s (11), severe pain, such as postsurgical pain, can be relieved in some patients by giving them a placebo in place of morphine or other analgesic drugs. About 35% of the patients report marked relief of pain after being given a placebo. This is a strikingly high proportion because morphine, even in large doses, relieves severe pain in only about 75% of patients.
Another interesting discovery about placebos is that their effectiveness is of the order of half that of the drug with which they are compared, even in double-blind experiments (40); that is, if the drug is a mild analgesic such as aspirin, the pain relief produced by the placebo is half that of the aspirin. If it is a powerful drug such as morphine, the placebo has greater pain-relieving properties, again about half that of morphine. This indicates that even though the “double-blind” is maintained, the therapist’s enthusiasm is conveyed to the patient. However, in a recent analysis of controlled trials, 38% of patients obtained more than 10% of maximum possible pain relief after placebo, whereas 16% obtained greater than 50% relief (50% of maximum with a potent agent). Thus, there was considerable variation among patients in placebo response, within a given study (41).
There are large individual differences in susceptibility to placebos, and studies have been done to determine some of the factors involved (42). These studies have revealed that placebos are more effective for severe pain than for mild pain, and more effective in patients experiencing great stress and anxiety than in those who are not. McGlashan and co-workers (43) have shown that placebo-induced analgesia is not significantly related to suggestibility, hypnotic susceptibility, or anxiety induced specifically by pain or the therapeutic situation (termed state-anxiety); however, placebo effects occur more powerfully in people who have chronic, generalized anxiety (personality-trait anxiety). Recent work has documented differences in the neurobiology associated with placebo responders and nonresponders. Interestingly, responders and nonresponders differ in the brain areas activated by the ultrashort-acting opioid remifentanil (44). Placebo responders but not nonresponders show activation of the rostral part of the anterior cingulate cortex, a structure associated with attentional processing of pain experience (45), suggesting that placebo responders may have a more effective endogenous opioid system than nonresponders (46).
The placebo response is affected by other fascinating factors. Two placebo capsules, for example, are more effective than one capsule, and large capsules are better than small ones. A placebo is more effective when injected than when given by mouth, and is more potent when accompanied by a strong suggestion that a powerful analgesic has been given. In short, the greater the implicit and explicit suggestion that pain will be relieved, the greater is the relief obtained by the patient. Unfortunately, however, patients tend to get progressively less relief during repeated administration of placebos. We know very little about the extent to which placebo effects actually contribute to pain relief in the clinical setting (47).
The placebo effect is no longer viewed as a nuisance factor or artifact to be controlled, parcelled out, or discounted, but a phenomenon worthy of study in its own right (48). This interest has increased in part because we now know that the opioid antagonist naloxone reverses placebo analgesia, implying that endogenous opioids mediate placebo analgesia (49,50). Moreover, the neurobiology of the placebo response is now better understood as a result of modern imaging techniques (46,51). Using positron emission tomography (PET) scans, Petrovic and co-workers (44) showed that both placebo analgesia and the rapidly acting opioid remifentanil activate the rostral anterior cingulate cortex, suggesting a common underlying neural network. Consistent with this suggestion, a recent functional magnetic resonance imaging study of patients with irritable bowel syndrome found that decreases in the activity of pain-related brain regions, including thalamus, somatosensory cortex, insula, and anterior cingulate cortex accompanied placebo analgesia induced by verbal suggestion (52).
The precise mechanisms underlying the placebo response are not fully understood (47,48), but we know that two nonmutually exclusive factors play important roles in mediating placebo analgesia; namely, expectation and conditioning. Expectation usually is generated by verbal suggestions, but may also involve implicit, nonverbal cues. Conditioning effects operate by classical or Pavlovian conditioning and typically involve the pairing of an active drug with a specific context or contextual cues. Under most circumstances, placebo responses that arise from expectation induced by verbal suggestions appear to be opioid-mediated, since they are naloxone-reversible. The sensitivity to naloxone of the placebo response following conditioning depends, in part, on the opioidergic status of the unconditioned stimulus (drug). For example, naloxone-reversible placebo responses develop after conditioning with the μ-opioid agonist morphine but not the nonsteroidal anti-inflammatory drug ketorolac (49). The neurobiology of the placebo effect has progressed to the point at which we now know that some drugs act specifically to enhance placebo analgesia. This fascinating finding was first shown by Benedetti and co-workers (53) in a randomized, double-blind controlled trial of patients following posterolateral thoracotomy. After patients had recovered from the general anesthetic, they were randomly assigned to receive a hidden or open injection of saline or proglumide, a cholecystokinin antagonist. Patients in the open injection group received the injection by a physician in full view of the infusion pump and were told that it was a potent analgesic. Patients receiving the hidden injection (by way of an infusion pump that was out of sight) were unaware that an injection had been administered.
Not surprisingly, the results showed that patients who received the hidden injection of saline reported no pain relief whatsoever. In comparison, patients who received the open injection of saline showed significantly more pain relief (i.e., a placebo response due to expectation), and patients who received the open injection of proglumide showed the most
pain relief of all (i.e., an apparent analgesic response plus a placebo response due to expectation). These findings appear to be straightforward and suggest that proglumide is an effective analgesic since it produces more pain relief than both a placebo injection and no treatment (i.e., the hidden saline injection). However, this interpretation is clearly incorrect since patients receiving the hidden injection of proglumide showed a total lack of analgesic effect, equal to that of the control group that received the hidden injection of saline. These striking findings suggest that proglumide produces pain relief only in the context of a placebo procedure (i.e., in response to verbal suggestions of pain relief) and that proglumide exerts its effects on an “expectation pathway” to enhance the placebo analgesic response (48).
pain relief of all (i.e., an apparent analgesic response plus a placebo response due to expectation). These findings appear to be straightforward and suggest that proglumide is an effective analgesic since it produces more pain relief than both a placebo injection and no treatment (i.e., the hidden saline injection). However, this interpretation is clearly incorrect since patients receiving the hidden injection of proglumide showed a total lack of analgesic effect, equal to that of the control group that received the hidden injection of saline. These striking findings suggest that proglumide produces pain relief only in the context of a placebo procedure (i.e., in response to verbal suggestions of pain relief) and that proglumide exerts its effects on an “expectation pathway” to enhance the placebo analgesic response (48).
Context plays an extremely important role in the expression of the placebo response. By context we mean the environmental cues and situations that co-occur with a treatment, along with concurrent words, and past and present meanings attributed to these words. One of the most profound contextual cues known to produce a placebo response consists of the words spoken by a physician to her patient, and in particular, the degree of certainty with which the message is conveyed (54,55,56). The power of the placebo response generated by a physician’s word is best illustrated by Pollo and colleagues (55), who administered a 72-hour continuous intravenous (IV) infusion of saline to three groups of patients after posterolateral thoracotomy. In addition to the background saline infusion, patients in all three groups received 0.15 mg IV buprenorphine on demand for pain relief. Patients in group 1 (natural history) were not told anything about the analgesic action of the saline infusion and believed it was a rehydrating solution. Patients in group 2 (double-blind placebo) were told that the infusion was either buprenorphine or saline and thus were uncertain as to the contents of the infusion. Patients in group 3 (deceptive placebo) were told the infusion contained a powerful painkiller. From their perspective, group 3 patients were certain they were receiving a potent analgesic agent.
The results showed that the number of doses of buprenorphine received by the three groups differed significantly as a function of the instructions they were given and, by implication, the certainty with which patients believed they were receiving an active drug infusion. Patients who were told that the saline was a powerful drug (deceptive placebo) demonstrated the largest placebo effect, a 33.8% reduction in buprenorphine intake compared with the natural history control group. Being less certain about the analgesic properties of the saline infusion (double-blind placebo) produced a 20.8% reduction in buprenorphine use compared with the natural history control group. Since pain scores did not differ significantly among the groups because all were able to self-medicate with buprenorphine upon demand, the magnitude of the placebo effects are indicated by the different doses of buprenorphine taken by the three groups of patients (55).
Not all placebos result in pain reduction. Under certain circumstances, it has been observed that verbal suggestion, conditioning, or expectation can increase pain (57). The term nocebo hyperalgesia has been coined to describe the increase in pain that occurs after administration of an inert substance (58). The nocebo response appears to be as robust as the placebo response, but it is not as well researched. Nocebo hyperalgesia is blocked by the cholecystokinin (CCK) antagonist proglumide and is unaffected by naloxone, suggesting that nocebo-induced hyperalgesia is mediated by an opioid independent, CCK-activated increase in anxiety (58). This suggestion is supported by a recent study showing that nocebo hyperalgesia is greatest among individuals high in pain anxiety (59).
Research into the placebo/nocebo effect clearly demonstrates that a physician’s choice of words profoundly affects her patients’ pain intensity (56) and demand for analgesics after surgery (55,57). Controlled studies designed to better understand the placebo response may require patients to be deceived, providing that backup “rescue” medication is immediately available. However, this clearly is not always feasible in clinical practice. Deceptive administration of a placebo in clinical practice is therefore unethical, since doing so denies patients their legal and ethical rights to informed consent and to refuse treatment (60). Nevertheless, it may be possible to leverage the desirable effects of the placebo response without deception, by emphasizing in a truthfully worded statement the benefits of a procedure or drug whose analgesic efficacy is well documented (47,54).
Hypnosis
The manipulation of attention together with strong suggestion is part of the phenomenon of hypnosis. The hypnotic state eludes precise definition. Loosely speaking, hypnosis is a trance state in which the subject’s attention is focused intensely on the hypnotist while attention to other stimuli is markedly diminished. After people are hypnotized they can, with appropriate suggestion, receive normally noxious tactile or thermal stimuli yet report no pain (61,62). They may say that they felt a sharp tactile sensation or strong heat, but they maintain that the sensations never welled up into pain. A small percentage of people can be hypnotized deeply enough to undergo major surgery entirely without anesthesia. For a larger number of people, hypnosis reduces the amount of pain-killing drug required to produce successful analgesia.
Despite the long history of hypnotism, which has been used for hundreds of years under different names such as animal magnetism and mesmerism, very little is known about its mechanisms. Still worse, most of its major features are highly controversial. For example, there is a vigorous debate about its fundamental nature. Is it a special state of consciousness known as a “trance state,” or is it merely a trait of responsiveness to strong suggestion? There is no resolution yet to this question (61).
Nevertheless, anyone who has observed the behavior of people who have been hypnotized realizes that this is an especially interesting phenomenon. Under hypnosis, people tolerate pain, during demonstrations or experiments, from stimuli that would normally cause them to cry out and withdraw. Countless accounts of such observations are supplemented by reports that hypnosis is effective in relieving severe clinical pains, such as phantom limb pain. Although excellent studies of hypnotic analgesia have been carried out with experimentally induced pains (63), there are as yet no convincing studies, using the necessary control groups, of clinical pain. The evidence thus far is observational or “anecdotal.”
It is known, however, that not all people can be hypnotized. About 30% of people can reach a state of deep hypnosis, 30% reach a moderate state, 30% achieve a drowsy state, and 10% of people are not susceptible at all. These figures resemble the proportions of placebo reactors and nonreactors. There is, however, strong evidence that hyporesponsiveness to pain in hypnotized subjects is not simply a placebo effect. An elegantly designed experiment (43) has shown that pain perception threshold and pain tolerance level are both strikingly increased during hypnosis but that only pain perception threshold is raised after administration of a placebo. The same study demonstrated that the hypnotic procedure itself has not only
a placebo effect, but also an additional effect that raises pain threshold and tolerance still further.
a placebo effect, but also an additional effect that raises pain threshold and tolerance still further.
Posttraumatic Stress Disorder and Pain
Posttraumatic stress disorder (PTSD) usually occurs after exposure to a situation or event that is perceived to be threatening to the physical or emotional integrity of an individual. The diagnostic criteria (64) for PTSD comprise three symptom clusters, including (a) reexperiencing the traumatic event (e.g., “flashbacks” and nightmares); (b) emotional numbing (e.g., feeling detached from others) and avoidance of thoughts, feelings, and activities associated with the trauma; and (c) increased arousal (e.g., insomnia, hypervigilance, exaggerated startle reflex). The lifetime prevalence for PTSD ranges from 1% to 14% but may be as high as 58% for at-risk individuals (e.g., soldiers or victims of violent crime) (64). Many more individuals who do not meet all criteria for PTSD may suffer with partial or subthreshold PTSD (65).
Both PTSD and chronic pain are highly prevalent and often difficult to manage effectively. Recent evidence suggests that PTSD and chronic pain occur as comorbidities more frequently than would be expected by chance alone. The prevalence of chronic pain in patients with PTSD ranges from 20% to 80% (66,67,68). Similarly, the prevalence of PTSD in patients with chronic pain ranges from 10% to 50% (69,70). The high comorbidity has led to a call for routine assessment of both chronic pain and PTSD when either one is diagnosed (71).