Biofeedback and Headache Seymour Diamond, M.D. Assistant Professor of Neurology, University of Chicago School of Medicine, Chicago, Illinois. Director, Diamond Headache Clinic, Chicago, Illinois. Reprint requests to: Seymour Diamond, M.D., Diamond Headache Clinic Ltd., 5252 No. Western Avenue, Chicago, Illinois 60625. Biofeedback uses instrumentation to teach a person to bring previously unused or involuntary bodily functions under voluntary control. Norbert Weiner, the mathematician, coined the word feedback and defined it as a method of controlling the system by reinserting into it the results of our past performance. Biofeedback is a biological system where the feedback is artifically mediated by man-made detection, amplification, and display instrumentation, rather than by an inborn feedback loop within the biological system. The psychological basis for the effectiveness of biofeedback can be demonstrated by the difference between traditional Pavlovian classical conditioning and instrumental or operant conditioning. Biofeedback, in principle, is consistent with instrumental learning. This can be traced to Ivan Pavlov, the Russian scientist, and his classical conditioning theory. Pavlovian classical conditioning begins with an unconditioned stimulus. The unconditioned stimulus is paired with a conditioned stimulus. Eventually, the conditioned stimulus will produce what was formerly an unconditioned response (Fig. 1). From this emerged the theory of operant conditioning, forming the basis of the biofeedback technique. A conditioned stimulus is provided along with an opportunity to respond in various ways. The desired response is then reinforced or rewarded. After several reinforcements, the stimulus serves as a signal for performing the learned response (Fig. 2). Biofeedback borrows from past psychological experimentation, with the most ancient being the Hindu practice of yoga. Yoga teaches mental discipline over bodily functions to achieve tranquility with the universe. The disciple chants a series of mantras to help achieve inner discipline, with the master acting merely as a teacher. In Germany, Johannes Schultz developed a mind-body training system called autogenic training. In Schultz's system, the patient directed himself, repeating a series of
phrases to help him relax. The physician again acts solely as a teacher; thus the technique falls into the realm of self-regulation. Similarly, Edmond Jacobson1 devised a system of training known as progressive relaxation. These techniques form the background for the work of Dr. Joe Kamiya,2 in which he taught subjects to control their brain waves. With electronic circuitry a tone was produced whenever a subject produced an alpha brain wave; as a result, most subjects learned to produce alpha waves at will. Dr. Neal Miller3 and his associates have demonstrated that many autonomic responses can be induced in animals by operant conditioning. After administering curare and maintaining artificial breathing in the animals, to rule out the possibility that autonomic responses were controlled by skeletal muscles under voluntary control, animals were taught to increase or decrease blood pressure and heart rate, produce localized blood flow to an ear, and vary urinary secretions or gastric and intestinal contractions. When the animals showed the desired responses, they were rewarded with electrical stimulation to the pleasure centers in the hypothalamus. There has also been success in controlling blood pressure temporarily in humans with biofeedback techniques. The use of biofeedback techniques in the treatment of migraine headache problems was first suggested by Dr. Elmer Green* after a serendipitous finding at Menninger Clinic. A subject being monitored for relaxation therapy aborted her migraine headache and observed an accompanying increase in hand temperature. This led to further research at Menninger Clinic in which Sargent, Green, and Walters4 experimented with autogenic feedback training. As an aid in training patients to raise hand temperature, the autogenic phrases of Johannes Schultz were used, since in previous reports Schultz had shown that some subjects were able to alleviate migraine attacks while practicing autogenic training exercises. The combination of thermal feedback with autogenic training involved simultaneous management of mental and somatic functions. "Passive concentration" was used to induce specific physiological changes, such as hand warmth, by focusing on visual, auditory, and somatic imagery. Each training session consisted of the practice of two groups of autogenic phrases; the first group aimed at achieving passive concentration and relaxation of the entire body, and the second group focused on achieving warm hands, followed by the visualization of images while using a temperature trainer. This was accompanied by daily practice and accurate record keeping at home. They found that temperature change in the hands is directly related to blood flow in the area as measured by a plethysmograph, and an increase in skin temperature in the hands is used as an index of voluntary control of the sympathetic section of the autonomic nervous system. Subjects were seen weekly until they showed a consistent positive response. The training phase was later limited to one month, as it was found that most subjects were able to acquire the techniques rapidly. Follow-up sessions were held every one to three months, with the follow-up period lasting at least a year. Baseline data, consisting of headache intensity and the use of medication, were collected for the month prior to initiation of instruction. In the first pilot study, 63% of the 20 migraine sufferers were evaluated as improved,4 and in the second pilot study, 74% of the 63 migraine sufferers improved.5 The criteria for improvement were based on a decrease in headache severity and on the type and dose of analgesic used. In a follow-up evaluation, Solbach and Sargent6 found that 55 subjects, 74% of those who had completed 270 days of training and follow-up sessions, had a 26% or greater reduction of headache activity lasting for at least two years after the conclusion of the study. Recent work has found that the resting frontalis EMG level is higher in muscle contraction headache patients than in normals, and that the headaches are usually due to a sustained contraction of the scalp and neck muscles. Thus, Drs. Thomas Budzynski, Johann Stoyva, and Charles Adler,7 of the University of Colorado Medical Center, have studied the use of electromyograph feedback to relieve these headaches. They used two control groups: a "pseudofeedback" condition and a "no treatment" condition, in addition to the experimental group, to rule out placebo or suggestion effects. Baseline data on headache intensity was collected two weeks before training began. The 18 patients diagnosed as suffering from muscle contraction headaches were assigned to one of the three groups. The treatment period involved 16 sessions, ideally two per week, with a three-month follow-up session. During the sessions, the patient reclined on a couch in a dimly lit, electrically shielded room. Three EMG leads were placed across the patient's forehead over the frontalis muscle. Patients in the feedback condition heard a tone whose frequency matched the EMG level of the frontalis muscle. Patients were also encouraged to practice at home for two 15 to 20-minute periods per day without the instrument. Budzynski et al7 found training for relaxation of the forehead muscles was effective in eliminating muscle contraction or tension headaches. There was a significant reduction in headache activity in the EMG feedback group as compared with the two control groups, as 75% of the patients showed significant declines in headache activity. Patients were able to learn to decrease their resting electromyograph levels by 50% to 70% in three to six 20-minute feedback sessions. The three-month follow-up sessions showed that patients retained the ability to produce low forehead EMG levels. At the Diamond Headache Clinic8 a combination of
hand-warming and EMG feedback techniques are used, since many patients suffer from more than one type of headache. The use of both techniques helps to achieve a greater reduction of both vascular and muscle contraction headaches. Since stress can provoke migraine headaches, EMG feedback training plus thermal feedback can benefit these patients. In the initial training session an explanation of biofeedback is given to the patient and the goals of therapy are discussed. The importance of home practice is also stressed. Sessions are held in quiet, dimly lit rooms, with patients sitting in comfortable, reclining chairs. The Clinic sessions are composed of three stages. The first stage is skin temperature feedback with autogenic phrases (Table 1). The autogenic phrases were adopted from the work of Johannes Schultz. Luthe and Schultz9 found therapeutic results with Table 1 Autogenic Phrases I feel quite quiet . . . I am beginning to feel quite relaxed . . . my feet feel heavy and relaxed . . . my ankles, my knees, and my hips, feel heavy, relaxed, and comfortable . . . my solar plexus, and the whole central portion of my body, feel relaxed and quiet . . . my hands, my arms, and my shoulders, feel heavy, relaxed, and comfortable . . . my neck, my jaws, and my forehead feel relaxed . . . they feel comfortable and smooth . . . my whole body feels quiet, heavy, comfortable and relaxed. I am quite relaxed . . . my arms and hands are heavy and warm . . . I feel quite quiet . . . my whole body is relaxed and my hands are warm, relaxed and warm . . . my hands are warm . . . warmth is flowing into my hands, they are warm . . . warm.
autogenic therapy in a majority of patients, showing a decrease in the frequency and intensity of migraine attacks. The phrases are autosuggestive in nature, focusing on feelings of warmth and relaxation, and are practiced before initiating skin temperature feedback. In addition to these phrases, patients are encouraged to focus on warm and relaxing images. Then the patient practices raising hand temperature with a thermal feedback instrument for ten minutes. The second stage of the Clinic session involves progressive relaxation exercises (Table 2), adapted from the work of Joseph Wolpe10 of Temple University, in which the patients practice tensing and relaxing various muscles. The third stage of the session is electromyograph feedback. Three electrodes are placed across the frontalis muscle. The patients receive EMG feedback for 20-minute sessions, during which time they try to learn to identify certain tensor points in the facial, neck, and shoulder areas. The EMG monitors are equipped with various sensitivity levels so that as the patient reduces tension at one level, the monitor is made increasingly sensitive. The importance of home practice is stressed. Patients are expected to practice at home for at least two 15-minute sessions (including muscle relaxation exercises and temperature feedback). Patients lease a temperature trainer for home use for the first month of training. This facilitates the gradual weaning of the patient from the biofeedback monitors. There are two types of training periods: (1) Patients from the Chicago area have four weeks of training with two sessions per week at the Clinic, totalling eight sessions, and two daily sessions at home for four weeks; and (2) patients from out of town have intensive two-week training periods, with two sessions per day at the Clinic, for a total of 22 sessions, and two daily sessions at home for four weeks. Follow-up sessions are dependent upon the success of each individual patient. Diamond and Franklin11 used autogenic training with both electromyograph and temperature feedback in treating 32 children under the age of 18 diagnosed as having migraine headaches. The importance of daily home practice was stressed and records of home sessions were kept. Twenty-six children responded with good results, denoted by a decrease in both the frequency and the severity of migraine headaches; three showed fair results, decreasing in either frequency or severity; two showed no response, and one patient was lost to follow-up. It Table 2 Relaxation Exercises - Wolpe Relaxation of facial area with neck, shoulders and upper back. Time: 4-5 minutes. Let all your muscles go loose and heavy. Just settle back quietly and comfortable. Wrinkle up your forehead now; wrinkle and smooth it out. Picture the entire forehead and scalp becoming smoother as the relaxation increases . . . now frown and crease your brows and study the tension . . . let go of the tension again. Smooth out the forehead once more . . . now, close your eyes tighter and tighter. Feel the tension . . . and relax your eyes. Keep your eyes closed, gently, comfortably, and notice the relaxation . . . now clench your jaws, bite your teeth together; study the tension throughout the jaws . . . relax your jaws now. Let your lips part slightly . . . appreciate the relaxation. . . now press your tongue hard against the roof of your mouth. Look for the tension. . . all right, let your tongue return to a comfortable and relaxed position . . . now purse your lips, press your lips together tighter and tighter . . . relax the lips. Note the contrast between tension and relaxation, feel the relaxation all over your face, all over your forehead and scalp, eyes, jaws, lips, tongue, and your neck muscles. Press your head back as far as it can go and feel the tension in the neck; roll it to the right and feel the tension shift; now roll it to the left. Straighten your head and bring it forward and press your chin against your chest. Let your head return to a comfortable position, and study the relaxation. Let the relaxation develop . . . shrug your shoulders right up. Hold the tension . . . drop your shoulders and feel the relaxation. Neck and shoulders relaxed . . . shrug your shoulders again and move them around. Bring your shoulders up and forward and back. Feel the tension in your shoulders and in your upper back . . . drop your shoulders once more and relax. Let the relaxation spread deep into the shoulders, right into your back muscles; relax your neck and throat, and your jaws and other facial areas as the pure relaxation takes over and grows deeper. . . deeper . . . ever deeper.
was concluded that children are ideal candidates for biofeedback therapy. Drs. Diamond and Medina12 retrospectively studied patients instructed in both electromyograph and temperature feedback by mailing a questionnaire to all 556 patients trained within the preceding five-year period. Patients were asked for their opinions about the effectiveness of their therapy. Four hundred thirteen patients answered the questionnaire; their ages ranged from 9 to 71 years. One hundred fifteen patients were diagnosed as having migraine headaches; 15 had muscle contraction headaches; and 283 had mixed migraine and muscle contraction headaches. Their training periods had ended from 3 to 62 months prior to the date when the questionnaires were sent. Ninety percent of the patients responded that the biofeedback techniques had helped them to relax; 160 patients (39%) believed that biofeedback had helped their headaches permanently, from 4 to 52 months; 133 patients (32%) found temporary relief, lasting from 1 to 36 months; and 120 patients (29%) did not feel that biofeedback had helped their headaches. Of the 160 patients permanently helped, 56 were diagnosed as having migraine headache, 97 had mixed muscle contraction and migraine headaches, and 7 had muscle contraction headaches alone. Of those patients who felt that biofeedback helped their headaches, 82% responded that the improvement consisted of a reduction in both frequency and severity of the headaches, 13% found only the severity reduced, and 5% responded that the frequency alone had decreased. Since many of the patients who were put on biofeedback were "rejects" from various other therapies, we feel that the results have great significance. Support for the effectiveness of biofeedback can be found indirectly from the work of Wilbert Fordyce13 on chronic pain. He supports the theory that there are two sets of factors influencing chronic pain: the organic and the learned. The learned factors can promote and maintain a pain "habit." Pain behavior may come under the control of learning factors in the form of environmental consequences. Pain of this type is called operant pain. Operants followed by reinforcing consequences increase the rate of that operant. The environmental consequences of pain behavior may maintain the pain even after the original organic factor is gone. There are two ways to decrease a learned behavior: (1) withdraw the positive reinforcers, or (2) increase the rate of an incompatible behavior. It is my opinion that biofeedback combined with operant conditioning is a viable therapy for many headache patients. We are now engaged in a controlled study to further substantiate this point. REFERENCES 1.
Jacobson E: Progressive Relaxation, Chicago, University of Chicago Press, 1938.
2.
Kamiya J: Operant control of the EEG alpha rhythm and some of its reported effects on consciousness, in Altered States of Consciousness, New York, John S. Wiley, 1969.
3.
Miller N: Neal E. Miller: Selected Papers. Chicago, Aldine-Atherton, Inc., 1971.
4.
Sargent JD, Green EE, Walters ED: The use of autogenic feedback training in a pilot study of migraine and tension headaches. Headache 12(3): 120-124, 1972.
5.
Sargent JD, Green EE, Walters ED: Preliminary report on the use of autogenic feedback training in the treatment of migraine and tension headaches. Psychosom Med 35(2):129-135, 1973.
6.
Solbach P, Sargent JD: A follow-up evaluation of the Menninger pilot migraine study using thermal training. Biofeedback Society of America Annual Meeting, 1977.
7.
Budzynski TH, Stoyva JM, Adler CS, Multaney DJ: EMG biofeedback and tension headache: a controlled study. Psychosom Med 35:6, 1973.
8.
Diamond S, Dalessio DJ: The Practicing Physician's Approach to Headache, 2nd ed, Baltimore, Williams & Wilkins, 1978, pp 136-137.
9.
Luthe W, Schultz JH: Autogenic Therapy, Vol. 2, Medical Applications, New York and London, Grune & Stratton, 1969.
10.
Wolpe J: The Practice of Behavior Therapy, New York, Pergamon, 1969.
11.
Diamond S, Franklin M: Autogenic training with biofeedback in children with migraine, in Luthe W, Antonelli F (eds.): Therapy in Psychosomatic Medicine, Proceedings of the 3rd Congress of the International College of Psychosomatic Medicine, Rome, 1975.
12.
Diamond S, Medina JL: Value of biofeedback in the treatment of chronic headache: the patients' opinions, in Friedman AP, Granger ME (eds.): Research and Clinical Studies in Headache. Vol. 6, Basel, S. Karger, 1978, pp 155-159.
13.
Fordyce WE: Pain viewed as learned behavior. In Bonica JJ (ed.): Advances in Neurology, Vol. 4, New York, Raven Press, 1974. DISCUSSION
David Rothner, M.D., Cleveland, Ohio: It's always a pleasure to listen to Dr. Diamond, especially when he's talking about biofeedback. Dr. Diamond, how old were the children you treated? Were they mostly between 12 and 18, or did you deal with a significant number of patients under the age of 12? Dr. Diamond: I had three or four under the age of 12 in the group. Dr. Rothner: Was biofeedback as successful in the younger patient? Dr. Diamond: More so - a very cooperative group of patients, really. Dr. Rothner: Also, what is the value of using family counseling or psychotherapy along with the biofeedback? It's been my experience that the families who participate in family counseling, or at least in some form of therapy in which they can talk about their problems as well as receive biofeedback, do better than the ones depending on biofeedback alone. Do you use concomitant psychotherapy? Is your biofeedback technician a trained psychologist who can talk to the patient before, during, or after the session? Dr. Diamond: I often use concomitant psychotherapy, but this particular group did not have such
psychotherapy. The diagnosis in these children was migraine; I don't think any were of a mixed headache type. My technologists have B.S. degrees in psychology. Jerry E. Wesch, Ph.D., Indianapolis, Indiana: I'd like to comment on your follow-up statistics. Given the context and the short duration of your training, those are excellent statistics. But I think that they underrepresent what can be done using biofeedback in a comprehensive, whole-person approach. Your model, by and large, is a physiologic model, in which you're retraining the end organs or some other physiologic substrata, to eliminate a response. But that does not deal with the triggering mechanism, whereas the concomitant use of psychotherapy, social learning techniques, self-assertion training, family counseling, behavior modification, and even meditation, as well as medication and competent physical medicine, come into play in order to give maximum results. Dr. Diamond: I agree. We are doing a retrospective study on patients who were cured of their headaches with biofeedback, and we are finding that a certain number are developing other symptoms. We've had some who have developed asthma and others who have developed ulcers. Jack J. Pinsky, M.D., Duarte, California: With regard to Will Fordyce's definition of pain as a learned behavior, this refers only to the behavioral aspects of pain and not to what pain is. We should not be fooled into thinking that we have a definition of pain because we're seeing some observable behaviors that we congregate, and call pain behavior. Learning, of course, applies in all of life, but learning from behavior that is reinforced is only one way in which people learn. Dr. Diamond, you did a wonderful job in describing the whole historical perspective. There have been many lyric leaps made in the literature, from the visceral athletes like Harry Houdini to the man in the street, yet we have completely ignored the tremendous individual potentiality these modalities have; to be used and learned. And again, people in the field continue to write about somebody's modification of Jacobson's progressive muscular relaxation, or Schultz' autogenic training. Most studies, if done carefully, show very little difference between people who learn to relieve symptoms with autogenic training and/or those using progressive muscular relaxation with or without biofeedback training. A great debate has been raised in the literature, and it's far from being resolved. A particular point is specificity. For years, we used frontalis muscles mostly for doing EMG training and biofeedback. The careful work in the literature shows that what happens in the frontalis muscle is absolutely not transferable to muscles in the rest of the body. We did one blind study with regard to alpha enhancement with chronic pain patients. Using consecutive patients, we computerized the tape that triggered the light source for feedback. These patients did not know that they were getting false feedback; the technician knew that patients were getting false feedback but did not know to which group they belonged. All were exposed to the same total time of light over 15 training sessions. Patients whose lights stayed on for longer durations, over a period of time, achieved what we called a learning curve. These patients had a success curve, compared to those who ended up where they started. Eight of the ten who got a learning curve felt they were helped by alpha enhancement. Two of 11 who didn't get a learning curve said they were helped. When patients examined their own indigenous alpha output, there was absolutely no difference in the indigenous alpha output with regard to their subjective response in either group. Arthur Scherbel, M.D., Cleveland, Ohio: Dr. Diamond, in my specialty, we usually look at 50% improvement as placebo improvement. You showed an overall figure of 48% to 50%. I'd like to know what were your criteria for improvement. Was it fewer headaches, less severity, less frequency? How long have you followed these patients? Dr. Diamond: Your 50% placebo effect rate strikes me as rather high. The rate in our population is generally less than 40%. Since placebo effects are usually time limited, I do not consider study results for at least one month. In the present study, the "improved" patients had been improved for at least a year and some for as long as four or five years. We determined improvement by a decrease in frequency, severity, or both. We also collect data by a questionnaire. When these patients spontaneously report that they're better, I think it indicates that we're getting results. Does that answer your questions, Dr. Scherbel? Dr. Scherbel: I understand what you're doing but I still don't see objective criteria by which to classify improvement. Now if you give patients tender loving care throughout this whole study, there are people who are very suggestible and they will feel better. Dr. Kudrow: Dr. Diamond, you mentioned something about symptom substitution. One of our Ph.D. candidates, who just completed a controlled study where psychometric testing was obtained on migraine subjects, pre- and post-treatment, found two things: (1) there was no symptom substitution after pain was significantly relieved; and (2) psychological and emotional aspects improved after they had obtained significant headache relief.
phrases to help him relax. The physician again acts solely as a teacher; thus the technique falls into the realm of self-regulation. Similarly, Edmond Jacobson1 devised a system of training known as progressive relaxation. These techniques form the background for the work of Dr. Joe Kamiya,2 in which he taught subjects to control their brain waves. With electronic circuitry a tone was produced whenever a subject produced an alpha brain wave; as a result, most subjects learned to produce alpha waves at will. Dr. Neal Miller3 and his associates have demonstrated that many autonomic responses can be induced in animals by operant conditioning. After administering curare and maintaining artificial breathing in the animals, to rule out the possibility that autonomic responses were controlled by skeletal muscles under voluntary control, animals were taught to increase or decrease blood pressure and heart rate, produce localized blood flow to an ear, and vary urinary secretions or gastric and intestinal contractions. When the animals showed the desired responses, they were rewarded with electrical stimulation to the pleasure centers in the hypothalamus. There has also been success in controlling blood pressure temporarily in humans with biofeedback techniques. The use of biofeedback techniques in the treatment of migraine headache problems was first suggested by Dr. Elmer Green* after a serendipitous finding at Menninger Clinic. A subject being monitored for relaxation therapy aborted her migraine headache and observed an accompanying increase in hand temperature. This led to further research at Menninger Clinic in which Sargent, Green, and Walters4 experimented with autogenic feedback training. As an aid in training patients to raise hand temperature, the autogenic phrases of Johannes Schultz were used, since in previous reports Schultz had shown that some subjects were able to alleviate migraine attacks while practicing autogenic training exercises. The combination of thermal feedback with autogenic training involved simultaneous management of mental and somatic functions. "Passive concentration" was used to induce specific physiological changes, such as hand warmth, by focusing on visual, auditory, and somatic imagery. Each training session consisted of the practice of two groups of autogenic phrases; the first group aimed at achieving passive concentration and relaxation of the entire body, and the second group focused on achieving warm hands, followed by the visualization of images while using a temperature trainer. This was accompanied by daily practice and accurate record keeping at home. They found that temperature change in the hands is directly related to blood flow in the area as measured by a plethysmograph, and an increase in skin temperature in the hands is used as an index of voluntary control of the sympathetic section of the autonomic nervous system. Subjects were seen weekly until they showed a consistent positive response. The training phase was later limited to one month, as it was found that most subjects were able to acquire the techniques rapidly. Follow-up sessions were held every one to three months, with the follow-up period lasting at least a year. Baseline data, consisting of headache intensity and the use of medication, were collected for the month prior to initiation of instruction. In the first pilot study, 63% of the 20 migraine sufferers were evaluated as improved,4 and in the second pilot study, 74% of the 63 migraine sufferers improved.5 The criteria for improvement were based on a decrease in headache severity and on the type and dose of analgesic used. In a follow-up evaluation, Solbach and Sargent6 found that 55 subjects, 74% of those who had completed 270 days of training and follow-up sessions, had a 26% or greater reduction of headache activity lasting for at least two years after the conclusion of the study. Recent work has found that the resting frontalis EMG level is higher in muscle contraction headache patients than in normals, and that the headaches are usually due to a sustained contraction of the scalp and neck muscles. Thus, Drs. Thomas Budzynski, Johann Stoyva, and Charles Adler,7 of the University of Colorado Medical Center, have studied the use of electromyograph feedback to relieve these headaches. They used two control groups: a "pseudofeedback" condition and a "no treatment" condition, in addition to the experimental group, to rule out placebo or suggestion effects. Baseline data on headache intensity was collected two weeks before training began. The 18 patients diagnosed as suffering from muscle contraction headaches were assigned to one of the three groups. The treatment period involved 16 sessions, ideally two per week, with a three-month follow-up session. During the sessions, the patient reclined on a couch in a dimly lit, electrically shielded room. Three EMG leads were placed across the patient's forehead over the frontalis muscle. Patients in the feedback condition heard a tone whose frequency matched the EMG level of the frontalis muscle. Patients were also encouraged to practice at home for two 15 to 20-minute periods per day without the instrument. Budzynski et al7 found training for relaxation of the forehead muscles was effective in eliminating muscle contraction or tension headaches. There was a significant reduction in headache activity in the EMG feedback group as compared with the two control groups, as 75% of the patients showed significant declines in headache activity. Patients were able to learn to decrease their resting electromyograph levels by 50% to 70% in three to six 20-minute feedback sessions. The three-month follow-up sessions showed that patients retained the ability to produce low forehead EMG levels. At the Diamond Headache Clinic8 a combination of
hand-warming and EMG feedback techniques are used, since many patients suffer from more than one type of headache. The use of both techniques helps to achieve a greater reduction of both vascular and muscle contraction headaches. Since stress can provoke migraine headaches, EMG feedback training plus thermal feedback can benefit these patients. In the initial training session an explanation of biofeedback is given to the patient and the goals of therapy are discussed. The importance of home practice is also stressed. Sessions are held in quiet, dimly lit rooms, with patients sitting in comfortable, reclining chairs. The Clinic sessions are composed of three stages. The first stage is skin temperature feedback with autogenic phrases (Table 1). The autogenic phrases were adopted from the work of Johannes Schultz. Luthe and Schultz9 found therapeutic results with Table 1 Autogenic Phrases I feel quite quiet . . . I am beginning to feel quite relaxed . . . my feet feel heavy and relaxed . . . my ankles, my knees, and my hips, feel heavy, relaxed, and comfortable . . . my solar plexus, and the whole central portion of my body, feel relaxed and quiet . . . my hands, my arms, and my shoulders, feel heavy, relaxed, and comfortable . . . my neck, my jaws, and my forehead feel relaxed . . . they feel comfortable and smooth . . . my whole body feels quiet, heavy, comfortable and relaxed. I am quite relaxed . . . my arms and hands are heavy and warm . . . I feel quite quiet . . . my whole body is relaxed and my hands are warm, relaxed and warm . . . my hands are warm . . . warmth is flowing into my hands, they are warm . . . warm.
autogenic therapy in a majority of patients, showing a decrease in the frequency and intensity of migraine attacks. The phrases are autosuggestive in nature, focusing on feelings of warmth and relaxation, and are practiced before initiating skin temperature feedback. In addition to these phrases, patients are encouraged to focus on warm and relaxing images. Then the patient practices raising hand temperature with a thermal feedback instrument for ten minutes. The second stage of the Clinic session involves progressive relaxation exercises (Table 2), adapted from the work of Joseph Wolpe10 of Temple University, in which the patients practice tensing and relaxing various muscles. The third stage of the session is electromyograph feedback. Three electrodes are placed across the frontalis muscle. The patients receive EMG feedback for 20-minute sessions, during which time they try to learn to identify certain tensor points in the facial, neck, and shoulder areas. The EMG monitors are equipped with various sensitivity levels so that as the patient reduces tension at one level, the monitor is made increasingly sensitive. The importance of home practice is stressed. Patients are expected to practice at home for at least two 15-minute sessions (including muscle relaxation exercises and temperature feedback). Patients lease a temperature trainer for home use for the first month of training. This facilitates the gradual weaning of the patient from the biofeedback monitors. There are two types of training periods: (1) Patients from the Chicago area have four weeks of training with two sessions per week at the Clinic, totalling eight sessions, and two daily sessions at home for four weeks; and (2) patients from out of town have intensive two-week training periods, with two sessions per day at the Clinic, for a total of 22 sessions, and two daily sessions at home for four weeks. Follow-up sessions are dependent upon the success of each individual patient. Diamond and Franklin11 used autogenic training with both electromyograph and temperature feedback in treating 32 children under the age of 18 diagnosed as having migraine headaches. The importance of daily home practice was stressed and records of home sessions were kept. Twenty-six children responded with good results, denoted by a decrease in both the frequency and the severity of migraine headaches; three showed fair results, decreasing in either frequency or severity; two showed no response, and one patient was lost to follow-up. It Table 2 Relaxation Exercises - Wolpe Relaxation of facial area with neck, shoulders and upper back. Time: 4-5 minutes. Let all your muscles go loose and heavy. Just settle back quietly and comfortable. Wrinkle up your forehead now; wrinkle and smooth it out. Picture the entire forehead and scalp becoming smoother as the relaxation increases . . . now frown and crease your brows and study the tension . . . let go of the tension again. Smooth out the forehead once more . . . now, close your eyes tighter and tighter. Feel the tension . . . and relax your eyes. Keep your eyes closed, gently, comfortably, and notice the relaxation . . . now clench your jaws, bite your teeth together; study the tension throughout the jaws . . . relax your jaws now. Let your lips part slightly . . . appreciate the relaxation. . . now press your tongue hard against the roof of your mouth. Look for the tension. . . all right, let your tongue return to a comfortable and relaxed position . . . now purse your lips, press your lips together tighter and tighter . . . relax the lips. Note the contrast between tension and relaxation, feel the relaxation all over your face, all over your forehead and scalp, eyes, jaws, lips, tongue, and your neck muscles. Press your head back as far as it can go and feel the tension in the neck; roll it to the right and feel the tension shift; now roll it to the left. Straighten your head and bring it forward and press your chin against your chest. Let your head return to a comfortable position, and study the relaxation. Let the relaxation develop . . . shrug your shoulders right up. Hold the tension . . . drop your shoulders and feel the relaxation. Neck and shoulders relaxed . . . shrug your shoulders again and move them around. Bring your shoulders up and forward and back. Feel the tension in your shoulders and in your upper back . . . drop your shoulders once more and relax. Let the relaxation spread deep into the shoulders, right into your back muscles; relax your neck and throat, and your jaws and other facial areas as the pure relaxation takes over and grows deeper. . . deeper . . . ever deeper.
was concluded that children are ideal candidates for biofeedback therapy. Drs. Diamond and Medina12 retrospectively studied patients instructed in both electromyograph and temperature feedback by mailing a questionnaire to all 556 patients trained within the preceding five-year period. Patients were asked for their opinions about the effectiveness of their therapy. Four hundred thirteen patients answered the questionnaire; their ages ranged from 9 to 71 years. One hundred fifteen patients were diagnosed as having migraine headaches; 15 had muscle contraction headaches; and 283 had mixed migraine and muscle contraction headaches. Their training periods had ended from 3 to 62 months prior to the date when the questionnaires were sent. Ninety percent of the patients responded that the biofeedback techniques had helped them to relax; 160 patients (39%) believed that biofeedback had helped their headaches permanently, from 4 to 52 months; 133 patients (32%) found temporary relief, lasting from 1 to 36 months; and 120 patients (29%) did not feel that biofeedback had helped their headaches. Of the 160 patients permanently helped, 56 were diagnosed as having migraine headache, 97 had mixed muscle contraction and migraine headaches, and 7 had muscle contraction headaches alone. Of those patients who felt that biofeedback helped their headaches, 82% responded that the improvement consisted of a reduction in both frequency and severity of the headaches, 13% found only the severity reduced, and 5% responded that the frequency alone had decreased. Since many of the patients who were put on biofeedback were "rejects" from various other therapies, we feel that the results have great significance. Support for the effectiveness of biofeedback can be found indirectly from the work of Wilbert Fordyce13 on chronic pain. He supports the theory that there are two sets of factors influencing chronic pain: the organic and the learned. The learned factors can promote and maintain a pain "habit." Pain behavior may come under the control of learning factors in the form of environmental consequences. Pain of this type is called operant pain. Operants followed by reinforcing consequences increase the rate of that operant. The environmental consequences of pain behavior may maintain the pain even after the original organic factor is gone. There are two ways to decrease a learned behavior: (1) withdraw the positive reinforcers, or (2) increase the rate of an incompatible behavior. It is my opinion that biofeedback combined with operant conditioning is a viable therapy for many headache patients. We are now engaged in a controlled study to further substantiate this point. REFERENCES 1.
Jacobson E: Progressive Relaxation, Chicago, University of Chicago Press, 1938.
2.
Kamiya J: Operant control of the EEG alpha rhythm and some of its reported effects on consciousness, in Altered States of Consciousness, New York, John S. Wiley, 1969.
3.
Miller N: Neal E. Miller: Selected Papers. Chicago, Aldine-Atherton, Inc., 1971.
4.
Sargent JD, Green EE, Walters ED: The use of autogenic feedback training in a pilot study of migraine and tension headaches. Headache 12(3): 120-124, 1972.
5.
Sargent JD, Green EE, Walters ED: Preliminary report on the use of autogenic feedback training in the treatment of migraine and tension headaches. Psychosom Med 35(2):129-135, 1973.
6.
Solbach P, Sargent JD: A follow-up evaluation of the Menninger pilot migraine study using thermal training. Biofeedback Society of America Annual Meeting, 1977.
7.
Budzynski TH, Stoyva JM, Adler CS, Multaney DJ: EMG biofeedback and tension headache: a controlled study. Psychosom Med 35:6, 1973.
8.
Diamond S, Dalessio DJ: The Practicing Physician's Approach to Headache, 2nd ed, Baltimore, Williams & Wilkins, 1978, pp 136-137.
9.
Luthe W, Schultz JH: Autogenic Therapy, Vol. 2, Medical Applications, New York and London, Grune & Stratton, 1969.
10.
Wolpe J: The Practice of Behavior Therapy, New York, Pergamon, 1969.
11.
Diamond S, Franklin M: Autogenic training with biofeedback in children with migraine, in Luthe W, Antonelli F (eds.): Therapy in Psychosomatic Medicine, Proceedings of the 3rd Congress of the International College of Psychosomatic Medicine, Rome, 1975.
12.
Diamond S, Medina JL: Value of biofeedback in the treatment of chronic headache: the patients' opinions, in Friedman AP, Granger ME (eds.): Research and Clinical Studies in Headache. Vol. 6, Basel, S. Karger, 1978, pp 155-159.
13.
Fordyce WE: Pain viewed as learned behavior. In Bonica JJ (ed.): Advances in Neurology, Vol. 4, New York, Raven Press, 1974. DISCUSSION
David Rothner, M.D., Cleveland, Ohio: It's always a pleasure to listen to Dr. Diamond, especially when he's talking about biofeedback. Dr. Diamond, how old were the children you treated? Were they mostly between 12 and 18, or did you deal with a significant number of patients under the age of 12? Dr. Diamond: I had three or four under the age of 12 in the group. Dr. Rothner: Was biofeedback as successful in the younger patient? Dr. Diamond: More so - a very cooperative group of patients, really. Dr. Rothner: Also, what is the value of using family counseling or psychotherapy along with the biofeedback? It's been my experience that the families who participate in family counseling, or at least in some form of therapy in which they can talk about their problems as well as receive biofeedback, do better than the ones depending on biofeedback alone. Do you use concomitant psychotherapy? Is your biofeedback technician a trained psychologist who can talk to the patient before, during, or after the session? Dr. Diamond: I often use concomitant psychotherapy, but this particular group did not have such
psychotherapy. The diagnosis in these children was migraine; I don't think any were of a mixed headache type. My technologists have B.S. degrees in psychology. Jerry E. Wesch, Ph.D., Indianapolis, Indiana: I'd like to comment on your follow-up statistics. Given the context and the short duration of your training, those are excellent statistics. But I think that they underrepresent what can be done using biofeedback in a comprehensive, whole-person approach. Your model, by and large, is a physiologic model, in which you're retraining the end organs or some other physiologic substrata, to eliminate a response. But that does not deal with the triggering mechanism, whereas the concomitant use of psychotherapy, social learning techniques, self-assertion training, family counseling, behavior modification, and even meditation, as well as medication and competent physical medicine, come into play in order to give maximum results. Dr. Diamond: I agree. We are doing a retrospective study on patients who were cured of their headaches with biofeedback, and we are finding that a certain number are developing other symptoms. We've had some who have developed asthma and others who have developed ulcers. Jack J. Pinsky, M.D., Duarte, California: With regard to Will Fordyce's definition of pain as a learned behavior, this refers only to the behavioral aspects of pain and not to what pain is. We should not be fooled into thinking that we have a definition of pain because we're seeing some observable behaviors that we congregate, and call pain behavior. Learning, of course, applies in all of life, but learning from behavior that is reinforced is only one way in which people learn. Dr. Diamond, you did a wonderful job in describing the whole historical perspective. There have been many lyric leaps made in the literature, from the visceral athletes like Harry Houdini to the man in the street, yet we have completely ignored the tremendous individual potentiality these modalities have; to be used and learned. And again, people in the field continue to write about somebody's modification of Jacobson's progressive muscular relaxation, or Schultz' autogenic training. Most studies, if done carefully, show very little difference between people who learn to relieve symptoms with autogenic training and/or those using progressive muscular relaxation with or without biofeedback training. A great debate has been raised in the literature, and it's far from being resolved. A particular point is specificity. For years, we used frontalis muscles mostly for doing EMG training and biofeedback. The careful work in the literature shows that what happens in the frontalis muscle is absolutely not transferable to muscles in the rest of the body. We did one blind study with regard to alpha enhancement with chronic pain patients. Using consecutive patients, we computerized the tape that triggered the light source for feedback. These patients did not know that they were getting false feedback; the technician knew that patients were getting false feedback but did not know to which group they belonged. All were exposed to the same total time of light over 15 training sessions. Patients whose lights stayed on for longer durations, over a period of time, achieved what we called a learning curve. These patients had a success curve, compared to those who ended up where they started. Eight of the ten who got a learning curve felt they were helped by alpha enhancement. Two of 11 who didn't get a learning curve said they were helped. When patients examined their own indigenous alpha output, there was absolutely no difference in the indigenous alpha output with regard to their subjective response in either group. Arthur Scherbel, M.D., Cleveland, Ohio: Dr. Diamond, in my specialty, we usually look at 50% improvement as placebo improvement. You showed an overall figure of 48% to 50%. I'd like to know what were your criteria for improvement. Was it fewer headaches, less severity, less frequency? How long have you followed these patients? Dr. Diamond: Your 50% placebo effect rate strikes me as rather high. The rate in our population is generally less than 40%. Since placebo effects are usually time limited, I do not consider study results for at least one month. In the present study, the "improved" patients had been improved for at least a year and some for as long as four or five years. We determined improvement by a decrease in frequency, severity, or both. We also collect data by a questionnaire. When these patients spontaneously report that they're better, I think it indicates that we're getting results. Does that answer your questions, Dr. Scherbel? Dr. Scherbel: I understand what you're doing but I still don't see objective criteria by which to classify improvement. Now if you give patients tender loving care throughout this whole study, there are people who are very suggestible and they will feel better. Dr. Kudrow: Dr. Diamond, you mentioned something about symptom substitution. One of our Ph.D. candidates, who just completed a controlled study where psychometric testing was obtained on migraine subjects, pre- and post-treatment, found two things: (1) there was no symptom substitution after pain was significantly relieved; and (2) psychological and emotional aspects improved after they had obtained significant headache relief.
