With Leslie J Crofford MD, Robert M Bennett MD FRCP, Don L Goldenberg MD, and John E Ware Jr PhD
Leslie J Crofford
I'd like to welcome the participants. My name is Leslie Crofford, I am an associate professor of medicine in the Division of rheumatology at the University of Michigan in Ann Harbor. The title of our program today is Fibromyalgia update: health-related quality of life and treatment strategies. I'd like to welcome all our speakers.
The first speaker today will be Don Goldenberg who is the Chief of rheumatology at Newton-Wellesley Hospital and Professor of medicine at Tufts University School of Medicine. Dr Goldenberg will be speaking on the diagnosis of fibromyalgia (FM) and will cover recent development in the pathophysiology of FM pain and in the relationship to clinical science and symptoms.
Our second speaker today will be John Ware PhD. He is the CEO and Chief Science Officer of QualityMetric, Inc in Lincoln, Rhode Island. He is also the executive director of the health assessment laboratory in Boston Massachusetts. His topic for today is Health-related quality of life. He will discuss what this means and what is its importance. He will also discuss the burden of FM on health-related quality of life and the clinically meaningful relationship between pain and health-related quality of life.
Robert Bennett is a Professor of medicine and former head of Division of arthritis and rheumatic diseases at the Oregon Health Sciences University. Dr Bennett will speak on the treatment of FM. He will specifically speak about a clinical trial summarizing the effects of tramadol and acetaminophen on health-related quality of life. He will also provide comments on optimizing pain management strategies for patients with FM.
We will follow those presentations with a group discussion involving myself with Professors Goldenberg and Bennett.
| Fibromyalgia syndrome: diagnosis and pathophysiology |
Don Goldenberg
Thank you for the introduction, Dr Crawford. I'm going to be talking both about the diagnosis of the FM syndrome, as well as the pathology and physiology that is involved.
FM is a disorder of widespread pain. Actually, about 30% of the population has chronic, generalized pain at any one point, which is really a symptom. The majority of those people don't have a defined disease. Certainly osteoarthritis, or similar chronic pain disorders, can also present with widespread pain, but most people who do have FM or symptoms of FM don't have a defined structural disease to account for the pain. Using the current criteria, established by the American College of Rheumatology, about 3-7% of everyone in the population would meet the criteria for fibromyalgia syndrome (FMS).
These criteria that were set up in 1990, have helped established some uniformity to the clinical diagnosis of FM, and have been critical to studies involving both epidemiology as well as the pathology and the physiology of this widespread disorder. The criteria involve one symptom, which is a history of chronic widespread pain, involving both the right and left sides of the body, above and below the waist, and the axial skeleton, and, the physical finding of having tenderness in at least 11 of 18 sites. These are palpated by digitally pressure of 4 kg, and these tender areas are usually at the end of a muscle where it attaches to a bone or a tendon.
 |
 |
This depicts where these 9 pair of tender points are. It should be pointed out that these are sites that are commonly involved in people with pain, such as the area over the trochanter versa of the hip, or over the lateral epicondyle of the arm, the so-called tennis elbow location. And although these 9 pairs are the most common areas that are involved in FM, the tender points represent exaggerated pain perception, and many other regions are also excessively tender, but these were chosen for the American College of Rheumatology criteria. |
The other interesting thing about FM is the diverse nature of people's symptoms. So although the widespread pain and the tender points are the cardinal features of the condition, almost everyone with FM reports widespread… other pain issues such as irritability of the bladder or bowel, fatigue, sleep disturbances, headaches, numbness, tingling and burning, as well as psychological and cognitive disturbances.
And therefore, the conditions that often overlap with FM include chronic fatigue syndrome, irritable bowel syndrome, muscle and migraine headaches, irritable bladder, as well as depression and anxiety.
FM usually occurs in women. It's 4-8 times more common in women than in man. It can occur at any age, but peaks between the ages of 40 and 60-65 at onset, but actually will increase in the population, as we get older, to reach 7% of the population, in females, at the age of 75. Most people who get FM are currently otherwise quite healthy. However, FM can occur, along with other chronic illnesses, particularly rheumatoid arthritis and osteoarthritis. It also seems to more often follow infectious diseases, such as a chronic viral illness as well as Lyme disease.
 |
|
This depicts where these 9 pair of tender points are. It should be pointed out that these are sites that are commonly involved in people with pain, such as the area over the trochanter versa of the hip, or over the lateral epicondyle of the arm, the so-called tennis elbow location. And although these 9 pairs are the most common areas that are involved in FM, the tender points represent exaggerated pain perception, and many other regions are also excessively tender, but these were chosen for the American College of Rheumatology criteria. |
This depicts the prevalence of FM in the general population. Note the striking prevalence of being higher in women than in men, at all ages, and how the percentage increases as the population ages up to, as mentioned, 7% in women between 70 and 80 years of age.
This brings us to research in regard to the etiology of FM. In summarizing this, I would state that there are 3 important factors. First of all, there is no single cause of FM, but rather a number of factors, including both physiologic, psychological, and environmental, that are important in why a single person will develop this chronic pain. Secondly, the muscle pain, muscle dysfunction, and tender points are epiphenomena. There is no evidence of a direct, pathologic, inflammatory, or immunologic role in the musculoskeletal system. Third, it's felt that various central nervous system abnormalities have been demonstrated and are operative, but none of them currently would be sufficient to explain all aspects of FM.
Investigators currently believe that the pain in FM is primarily central nervous system rather than in the muscles, as far as its origin. We believe this because there's clearly no structural damage in the muscle, and that people have a widespread decreased pain threshold rather than, as mentioned, the pain directly in the tender points, indicating any pathology. Furthermore, there's a strong association, as noted, with chronic fatigue syndrome, migraine headaches, irritable bowel syndrome, sleep disturbances, and mood disturbances, all linking the central nervous system. Finally, the symptoms of FM are very similar to those that occur in glucocorticoid withdrawal.
This chart depicts a study demonstrating the increased muscle pain in FM versus normal, healthy controls. And, as the pressure is increased, patients with FM have a persistently increased and linear relationship of widespread pain. The normal, healthy controls don't have striking pain until very high levels of pressure intensity are exerted.
The current thinking is reported pictorially in this slide, which represents what's called the "pain-brain-pain loop". In other words, both central sensitization, involved in the spinal cord and the brain, and peripheral tissues, which are peripheral pain generators, are consistent loops of pain in people with FM. Central sensitization includes amplification of pain, or hyperalgesia, as well as allodynia, which is a transformation of nonpain sensations to pain. Furthermore, the limbic system, involving psychological aspects of pain, including avoidance and attention, are clearly important in people with central sensitization. Pain research has involved in particular wind-up, which includes pain augmentation at the dorsal horn neuron, related to NMDA and neurokinin activation, and descending pain inhibition, particularly the role of norepinephrine and spinal norepinephrine, creating chronic hyperalgesia. Much of the research involves studies on elevated cerebral spinal fluid levels of substance P and neuroimaging studies. This next slide depicts the wind-up phenomena, and has been clearly proven, in a number of different investigations, as playing an important role in the exaggerated muscle tenderness in FM.
The changes in substance P have been demonstrated particularly in the cerebral spinal fluid. Substance P is a neuromodulator that sensitizes neurons to excitatory neurotransmitters. There's been a moderate clinical correlation of this finding with FM symptoms. There's also an inverse relationship to serotonin, since that decreases the release of substance P. These were the initial study results, demonstrating the differences in the cerebral spinal fluid levels of substances P in FM patients being much higher than in healthy controls. These could all be tied in also with neuroendocrine and the autonomic nervous system research in FM. For example, a number of groups have found alterations in cortico-releasing hormone, so-called CRH, in FM. Other investigators have concentrated on the autonomic nervous system, demonstrating the exaggerated blood pressure and heart responses to postural changes, so-called orthostatic intolerance. Serotonin, in FM, has been suggested as being an important player primarily because of clinical improvement with antidepressants such as tricyclics, and serotonin-reuptake inhibitors. Furthermore, genetic studies, particularly with the S/S genotype, are interesting, in particular because serotonin, as mentioned, inhibits descending pain pathways. Probably the most promising and instructive area in research in pain management in FM has been neuroimaging studies. First, the group in Alabama demonstrated hypoperfusion of the thalamus and the head of the caudate nucleus, and this was confirmed by SPECT scans. More recently, functional MRIs have demonstrated an abnormal subjective pain response in the cerebral cortex, highly consistent with the theory of central sensitization. This demonstrates the MRI and SPECT cerebral blood flow changes in FM versus control, with the attenuated blood flow in the FM pain areas of the caudate nucleus and thalamus.
In summary, FM is a disorder of widespread pain. As depicted here, this is a continuum. Many people in the population do have localized pain, and the reaching of these specific symptoms of FM, with the number of tender points, probably is related not only to pain factors but other factors, including fatigue and sleep disturbances, as well as psychological distress. The intriguing studies in the last year or two, and certainly, particularly those with imaging studies, in the last 5-10 years, give us an idea that the central nervous system, and particularly central sensitization, will be the common denominator for understanding this chronic pain disorder.
Thank you.
| Health-related quality of life |
John Ware
Thank you, Dr Crawford, for that introduction. I'm John Ware, and I'm going to be talking about health-related quality of life. A good place to begin this discussion is in terms of the traditional clinical definitions that we use of health status and outcomes. And, of course, that definition begins with bodily structure and function, as shown here. And, in these pictures, you can see the many different things that people do that give us the meaning of health-related quality of life. At the risk of over-simplifying the concept, I would say we can reduce this to three basic things. The first is what we're able to do, and we call that functioning. The second is how we feel, and this subjective notion includes both distress and well being. And, finally, no matter what you're able to do or how you feel, the third leg on the health-related quality of life stool is your personal evaluation of those things. So, all of these together are about what we would call human functioning and well-being, as opposed to the functioning of a particular organ of our body.
In this diagram, you can see the clinical continuum of disease-specific and generic measures. And over on the far left are the very familiar clinical markers of the functioning of an organ in the body, and over on the far right are the more generic indicators of health-related quality of life. The advantage of the generic measures, in theory, is that they allow us to compare the burden of different diseases and conditions, and to compare the benefits of different treatments. Over on the far left, the clinical markers give us information about whether we're having the desired, specific effects of treatment. And, what I'm going to do now is walk us through a specific example.
This particular example comes from a condition called peripheral vascular disease, or intermittent claudication disorder. And I picked this example because it's very well documented in the literature, and I give you several references there to articles in The New England Journal, and The Journal of the American Geriatric Society that talk about this in much greater detail. But, as you can see in this picture, one clinical marker for peripheral vascular disease is called the ankle-brachial index, which is a measurement of blood pressure in one of the limbs of your body compared with another limb in your body.
The second box is pain assessment. And, pain is very important in this disease. And one very objective way to assess pain clinically, as shown here in this picture, is how far can you walk on a treadmill before pain causes you to have to stop. We can also measure that specific symptom of pain, as shown here, in terms of what we call a categorical rating scale or self-assessment, such as the question that asks you how much did your pain limit your usual activities, or your enjoyment of everyday life. And, we might give you five choices, raging from, 'it doesn't limit me at all', or 'it limits me extremely'. So that's the second box in the clinical continuum.
The third box begins to get at health-related quality of life. And here we use the example of the Walking Impairment Questionnaire, which was used in these studies to get specific information about how peripheral vascular disease was affecting a patient's ability to walk. Finally, over on the far right, we have, for example, the SF-36 Health Survey and the scales from that survey and the summary measures from that survey, as a generic measure of health-related quality of life. Of course, treatment can have an effect at any of these levels. We can improve arterial blood flow, we can treat symptoms directly, we can use various devices to improve walking, and there are other things that we can do to in the environment to improve health-related quality of life more generically.
The most frequently used measure, throughout the world, for the generic aspect of health-related quality of life is the SF-36 Health Survey, SF standing for Short Form, and it's Short Form 36, because is has 36 questionnaire items. As you can see, in this diagram, these items are grouped into eight categories such as physical functioning, body pain, vitality, and mental health.
Those eight scales are further summarized into psychometrically based summary measures, including the Physical Component Summary, and the Mental Component Summary. What we mean when we say psychometrically based summary measures is that psychometric methods have been used to pool the physical variants from the eight measures into one summary, and to pool the mental variants from the eight measures into another summary. And, those two summaries together capture about 80-85% of the reliable variants about health-related quality of life that's in all of the items and all of the scales. The advantages of these summaries include they give us two outcomes instead of eight, so, in that sense, they are easier to interpret, and the two summaries cover a much wider range, and cover these concepts more comprehensively than would any one of the eight scales.
Let's look at an example of how one of these summaries, the Physical Component Summary, has helped us to understand the health of the well population and the burden of specific diseases.
Here we see the distribution of physical functioning scores from the Physical Component Summary in the general US population. We see that the average adult has a score of about 50. The average well adult, and that would be defined as somebody who has no chronic disease, scores about 5 points higher, a score of about 55. That's half a standard deviation. The standard deviation for this measure is 10. Now, we add some other diseases to this diagram, for example, asthma, chronic lung disease, congestive heart failure (CHF). And you can see that CHF, way over there on the left, is about 16 points below the average, that's about 1.6 standard deviations. So by standardizing the metric for this generic outcome, we can get a much better understanding of the burden of different diseases and conditions. At last count, there was published information in the medical literature for more than 200 different diseases on this particular physical and mental summary measure. Let's add one more disease to the diagram, from a clinical trial that I'm going to discuss in just a moment. We can see that the average patient with FM in that clinical trial had a score of 30, even worse than the score for the average patient with congestive heart failure. In other words, FM is a condition that has substantial impact on the physical performance and capacity of the average person with that disease.
Here we see the consequences of differences in health-related quality of life. And, you can see that the health-related quality of life boxes, box 3 and box 4 from that diagram that I just showed you, have considerable impact and consequence from a social, from an economic, and from a clinical point of view. Shown here on the far right, we can see the differences in health-related quality of life, as measured by a disease-specific impact measure, or a more generic tool, like the SF-36, are predictive of future health, inpatient, and outpatient medical expenditures, job loss, who's going to respond to treatment, return to work, work productivity among those who are able to continue a job, and outcomes as important as mortality. So, from a predictive point of view, differences in health-related quality of life are valid, in terms of these very important consequences. The information shown in this diagram also gives us very useful information that we can use to interpret the clinical, social, and economic importance of differences caused by disease or treatment in these generic measures, and we will see an example of that in just a minute.
In a recent clinical trial, comparing treatment for FM with a placebo, Bennett and colleagues used the design shown in this diagram.
After a washout period, they followed patients who were randomly assigned to medical treatment or to a placebo, and they followed them for a little over 90 days, as shown in this particular diagram.
Well, because they administered the SF-36 Health Survey at baseline and during follow-up, the first thing we get from this trial is a picture of the burden of FM in relation to national norms for the SF-36 scales and summary measures. As you can see here, focusing on the scales over on the left - those are the SF-36 scales that respond most to the physical component of health status - we can see that the burden of FM is greatest for those scales. And, when we look over on the far right and we look at the Physical Summary and the Mental Component Summary, again we see that the burden of this condition is primarily concentrated in the Physical Component, as we would expect for a condition that causes substantial physical morbidity. But, it's important to notice, in this chart, that all eight of the scales have a star, which means all are significantly below the norm for the average adult. And both the Physical and the Mental Component Summaries were significantly below the norm at baseline in this particular trial. It's just that the differences were greatest with a Physical Component Summary.
This slide compares the treated group with the placebo group in terms of changes in the SF-36 health profile, and the physical and mental summary measures. As you can see, the three scales over on the left, three of the scales that were most affected by this disease, all improved significantly more when treated with the drug therapy in comparison with the placebo group. And as you would expect, when you look over at the Physical Component Summary on the far right, again we see that that summary is improved a significantly higher amount for those that got the active agent in comparison with those that were assigned to the placebo group. None of the other differences were statistically significant although, as you can see here, some of them are noteworthy.
The 64 000$ question with these new measures of health-related quality of life is 'what do the numbers mean?' And, we use a couple of different strategies, which I'll explain very briefly, to interpret differences like the ones we saw in this particular clinical trial. One of the approaches is called content-based interpretation. In other words, we actually go and look at the content of the items that patients answered differently before and after treatment, and we look at the things that they said differently that caused them to get higher or lower scores. The other thing we can do, as listed here, is look at external criteria, such as what is the relationship between improvements in quality of life and performance at work, or medical expenditures, or mortality.
Let's consider a specific example by looking at one of the items in the physical functioning scale of the SF-36. Let's look at the item to the left on this particular diagram. As you can see here, the percentage reduction in limitations in climbing a flight of stairs was 24% for those that received the active agent in comparison with only 11% for those that were assigned to the placebo group. In other words, underlying the improvement that I showed you previously was an improvement rate on that particular performance that was twice as good for the treatment group in comparison with the placebo. Other examples are shown on this particular slide.
So in summary, what have we learned about health-related quality of life measurement and treatment from this particular clinical trial? First of all, we've learned that standardized measures reveal significant information about the health-related quality of life burden associated with FM. And by using the same generic measures in population surveys in surveys of people with other diseases and surveys of patients with FM, we can actually make comparisons like the one we just looked at. The other thing we've learned from this is that drug therapy significantly improves health-related quality of life and decreases pain, as shown on the previous slides. When we compare the endpoints, after treatments, with baseline, those improvements were observed in all eight of the SF-36 scales. However, the improvements were statistically significant and greatest for the Physical Component Summary, and of course for the three scales that get the highest weights, when we score those summaries. In the particular trial we looked at, that would be the physical functioning scale, the role physical scale, and the body pain scale.
So, what's going to happen next? Well, actually, we're at a very exciting time in the development of this field, and I'd like to briefly conclude with some information about improvements in health-related quality of life assessments that are coming. By the way, in the cartoon here, you can see the crane is constructing an improved ruler, and we construct health measures much like the ruler that's being built in this cartoon.
 |
|
We need to put the marks on the ruler where we need them for the things that we want to measure. So, we want to make sure that our measures provide information at the levels where patients are scoring today, and also measure the levels to which treatments might improve them in the days to come. |
There are two technological advances that are going to make it much easier to do what I just said. One of these comes from what is called modern psychometric methods. And, these are methods that we're using to take the questionnaire items from widely used measures, and calibrate them all on the same ruler. For example, in this chart on the left, you can see that measures of very poor physical functioning levels, such as those from activities of daily living measures, or the sickness impact profile, or the functional independence measure, those items from the ruler on the left are being combined with measures of physical functioning at much higher levels, such as those items from the SF-36 and the SF-12. The resultant combined ruler covers a much wider range and gives us more precision for monitoring outcomes than we would have if we used any one of these tools alone. And the cross-calibration of, for example, an ADL measure and an SF-36 measure, is much like the cross-calibration we did, a few hundred years ago, for the Celsius scale and the Fahrenheit scale. And we know how to convert, for example, the boiling point of water or the freezing point of water, or the average temperature of human blood -we know the cross-calibration, across the different thermometers, for those different temperatures. And, we're doing the same thing for our measures of physical functioning, our measures of vitality, and our measures of mental health, for example.
Now, as shown here, it's very important that we measure those very low levels. We've administered the SF-36 to more than 1 000 000 seniors, individuals 65 and older, who receive Medicare in the United Sates, and about 3% of them scored below the lowest level on the SF-36 Health Survey. So it's very important, when someone scores that low that we have information from the other measures, the ADL measures, as shown in this particular diagram; 3% isn't very many people, but those individuals are very expensive to take care of, and we need very good information about whether we're helping them with the treatments that are available.
The other major technological advance is shown in the chart on the right. For example, again using the metaphor as a ruler - and that's a good one, because what we're trying to do with the health status measure is the same thing we do with a ruler: we're trying to get quantitative information about what people are able to do, and how they feel. So, once we know, for example, that a patient scores between 6 and 7 on this particular ruler, we don't keep asking them the hundreds of questions that are printed in a questionnaire booklet that may or may not be relevant to them. Rather, we ask the questions that are relevant to a score between 6 and 7. This computerized dynamic approach to health assessment makes it possible for us to ask only the relevant questions, and even though you and I may answer different sets of questions, we can compare our results because they are all scored on the same metric. This computerized dynamic approach to doing health surveys is going to make it possible for us to gather data much more quickly from patients, to get data that will meet the clinical standard of precision, and it will cost a lot less than the way we've been collecting data up until now.
So, what are the advantages of dynamic health assessment? Well, with these new dynamic surveys we'll have a much more accurate way of screening individual patients in terms of their pain or their mental health. We will get score estimates that are reliable enough for clinicians to use those at the individual patient level. So, we can monitor outcomes accurately. Further, by only asking the relevant questions, research to date suggests that these more precise measures will have the brevity of a very short form. We will be able to reduce the number of questions by 90% for most patients. Another advantage is that we will eliminate what we call ceiling and floor effects. In other words, having a ruler that doesn't measure high enough or doesn't measure low enough in terms of functioning and well being. By being able to go to, for example, items from sports medicine for those patients that, you know, run every day, to measures that measure very basic activities of daily living for those who are performing at that level, we can measure anywhere on the ruler without the burden of making everyone answer the same questions. We can use these new dynamic surveys on the basis of various data collection technologies, as shown in the pictures here. We can do traditional interviews. We can do telephone interviews with, for example, speech recognition, where the interviewer can be using the computerized dynamic methodology, and the computer tells her what questions to ask the patient. Also, as shown here, individuals can answer questions on the internet by themselves, or using a computer, in a doctor's office. And finally, as shown here, we now have hand-held devices that have the power of a computer. So a patient, in the privacy of a waiting room, can use one of these hand-helds to do a dynamic assessment. The bottom line is that we can markedly reduce data collection cost by reducing the number of items and reducing the cost of processing the data.
I'd like to conclude by giving you some internet addresses where you can look at demos, and get more information about the SF-36 and other surveys that I have discussed, and other information about the advances in health-outcomes assessment that I've talked about in this brief presentation.
Survey Demos (SF-36, SF-12 and SF-8)
www.sf-36.com
Headache Impact Test (HIT)
www.headachetest.com
www.amIhealthy.com
Internet-based Assessments
www.qualitymetric.com
www.amIhealthy.com
Understanding Health Outcomes Educational Series
www.healthstatprod.com
I hope you find these websites interesting and stimulating. Thank you.
| The rational management of fibromyalgia patients |
Robert Bennett
Thank you, Dr Crawford, for your introduction. I would like to talk about the rational management of FM patients. I think we now know quite a lot about FM, and it is possible to talk about the rational management. I think there are many physicians who are reluctant to treat FM patients, because they consider it to be such a complex problem.
However, complex problems can be broken down. And here is a list of the major features that I think we're thinking about when you see a FM patient. You certainly can't deal with all of these in one visit. There are a lot of misunderstandings about FM, so education is important. Pain is the critical aspect of FM. We know now that it's mainly a central pain, but we have to think of those peripheral pain generators, and central pain, and its management. People with chronic pain and fatigue often become unfit and deconditioned; that is a very critical aspect of management. And, many studies have shown that low-grade exercise is beneficial in FM. People with conditions that basically cause problems in their perception of themselves often develop psychological distress, sometimes major problems such as depression and anxiety, and this needs to be recognized and treated. And, it's been known since the middle 1970s that patients with FM have a sleep disorder; it's called alpha-delta sleep. So, the management of this problem is important. Last, many FM patients, have a host of other problems that seem to be commonly association with FM. These would include things like irritable bowel, irritable bladder, postural hypotension, cognitive dysfunction, etc.
Now let's discuss the educational aspects. I think these are some of the major education points. First of all, FM is not purely a psychological diagnosis. There are abnormalities in pain perception, and it's important that patients try to understand this. It doesn't mean to say that the psychological factors are not important; they are in all chronic conditions, but patients need to know this is not a psychological diagnosis. I think a useful piece of information is that it's not a destructive disorder. It's not like, say, rheumatoid arthritis or multiple sclerosis. I think this is something that's positive, but this doesn't mean to say that chronic pain is a benign condition, either. I think one doesn't want to belittle the diagnosis of FM, but not being a destructive disorder is a positive thing that one can talk about. As in many conditions that rheumatologists deal with, we can't always cure FM, but it can be managed, if one goes about it in a constructive way. And, as with many other chronic diseases, patients need to focus on being well, and need to take an active role in their treatment.
So now, let's talk more specifically about pain, and we're going to divide this into peripheral pain and central pain. The major physiological abnormality that is being found in FM to date is called central sensitization, which basically means that impulses coming into the central system are amplified. However, those impulses have to come from somewhere, and they come from the periphery, and we call those peripheral pain generators.
So, it's important to evaluate the peripheral pain generators in all FM patients. If you're a rheumatologist you'll see a lot of osteoarthritis and inflammatory arthritis. If you're neurologist you'll probably see neuropathies. We all see injuries, and disc disorders. Visceral pain, particularly irritable bowel, irritable bladder, are common peripheral pain generators. Migraine headaches, upper vascular headaches, are much more common in FM patients. They obviously can be managed with medications. Spinal stenosis, as people get older, I find, is a common peripheral pain generator. In younger people who have manual jobs, repetitive strain may be an important peripheral pain generator. We have been aware that many young women with endometriosis, who have pelvic and abdominal pain, develop FM. And, it's quite interesting that a recent study by Arendt-Neilsen has shown that endometriosis is a potent condition in causing central sensitization. And lastly and probably most importantly, all FM patients have a lot of myofascial trigger points, and these need to be evaluated, and managed with physical therapy, and usually trigger point injections.
So, now, let's talk a little bit about the management of central pain. Basically, you're talking about the use of analgesic, NSAIDs, tramadol, and tramadol combinations, schedule-3 narcotics and the combinations, such as VicodinÒ, with hydrocodone, that is, and other similar medications with codeine; and then lastly, the schedule-2 opioids. Let's look at NSAIDs first.
I think the critical message that these are not very effective in treating central sensitization. However, many of the peripheral pain generators, particularly osteoarthritis and similar things, are certainly very susceptible to NSAIDs. So they do have an important use, but mainly in the management of peripheral pain generators. There is no reason to think that the newer, selective COX-2/NSAIDs are going to any better than the nonselective ones, because they don't seem to have a different profile as regards analgesia.
The opioids-these, at least the schedule 2-are not commonly used in FM. We find they're used in about 15% of the patients in our FM clinic. And there are certain issues that people are always concerned about. Dependency; I think you have to realize that this always occurs. Dependency really means that if someone is on an opioid for more than a few weeks and they stop it, suddenly they'll get a withdrawal phenomenon. Tolerance, that occurs, and basically there's a need for dose escalation over time. Addiction is actually rare in chronic pain patients; probably occurs in 1-2%, and needs to be distinguished from pseudo-addiction, which is basically, when physicians get fed up with prescribing opioids, and start to give negative appraisals of the patient, like they're narcotic seeking. Sometimes, the distinction between addiction and pseudo-addiction is quite difficult.
And then there is the issue, which has just recently come to light, that as you increase opioid dosage, in some patients, they start to have more pain, and this is called opioid hyperalgesia. In fact this mechanism in part is due to activation of NMDA receptors. And the only way of knowing if opioid hyperalgesia has occurred is to halve the dose and to see whether there is some improvement. Basically, there's been no control study of opioids in FM apart from tramadol. There's been a study in tramadol itself, and another study in tramadol added to acetominophen, so, with UltracetÒ, and I'll now describe that study.
This is a study that was just recently published, in the American Journal of Medicine. It was a 3-month, double-blind, placebo-controlled study. There were 158 patients on the tramadol/acetaminophen combination and 157 on placebo. There was a 3-week washout phase. However, importantly, patients in this study were allowed to continue on short-acting hypnotics such as zolpidem, and also on low dosages of selective serotonin-re-uptake inhibitors. On day 1 of the study they started on 1 tablet. By 10 days they were on 4 tablets, and they could escalate up to 2 tablets, 4 times a day. The major outcome measure was time to discontinuation due to lack of efficacy.
Here you see a Kaplan-Meier analysis of the probability of continuation, and at the end of the study, in 90 days, 72% of patients on the tramadol/APAP were still in the study compared to only 49% of the patients on placebo. This is, as you see, statistically significant.
The secondary outcome measure was the FM impact questionnaire, which consists of 10 items, each of which is scored 0-10. This gives a total score for the FM impact questionnaire of 100. And you can see the statistically significant changes there, in physical impairment, feeling good, doing their job, and also in the pain. Fatigue didn't improve, and maybe that wouldn't be expected, because these patients were allowed to take their short-acting hypnotic.
This is the continuation of the FM impact questionnaire. The stiffness improved, and anxiety improved, again, depression didn't improve, but they were allowed to continue to take their antidepressant. So the total change was very significant, a p value of 0.008. Side effects in this study: the major ones were some nausea, some dizziness, and constipation, which occurred in 19%, 12% and 10% of patients, respectively.
Antidepressants – these are usually the first line, for most physicians, particularly the tricyclic antidepressants such as amitriptyline, and, in those patients who have significant depression, of course, we use the selective serotonin re-uptake inhibitors, and more recently the selective norepinephrine-re-uptake inhibitors such as venlafaxine, and several newer ones that are coming along, which seem to have more promise in managing fibromyalgia pain. This shows a meta-analysis of tricyclic antidepressants in FM. And as you can see there's a moderate improvement in many aspects of fatigue, pain, sleep well-being, and trigger points; not incredibly impressive. And, it has been found that, at least for amitriptyline, in the studies that have been done, that the effects tend to wear off after about 3-4 months.
Very often, in FM, in the more difficult patients, you're left with someone who basically has refractory pain, and you wonder whether you should be going on the schedule-2 narcotics.
Before you do that I would suggest using adjunctive pain medications, maybe added to a codeine or a tramadol or a hydrocodone. And these are the major ones that are being used: the membrane stabilizers, which are basically anti-epileptics, like NeurontinÒ, pregabilin coming along, and topiramate. The alpha-2 adrenergic agonists: we really only have one that is available, and that is tizanidine, and that has been found to be a very useful medication as an adjunctive pain medication. It is used just at night. Use is starting at 2 mg, increasing by 2 mg every week, to a total dose of about 10 mg. And that is a muscle relaxant, but it also activates the descending pain system and modulates pain. In Europe, 5-HT3 antagonists are being used, particularly tropisetron, but that is not available in this country yet, and we await with interest the long-term studies on this. And then, lastly, there are the NMDA receptor antagonists. Ketamine is sometime used by pain doctors, but seldom by rheumatologists. Dextromethorphan has now become available as a single medication, 30 mg, and that is a weak NMDA receptor antagonist, and can be added to a low-dose opioid. And also one should remember that methadone, of course, which is a schedule-2, is also a weak NMDA receptor antagonist, which makes it a particularly useful schedule-2 when those have to be used.
Now, let's move on to the conditioning. People who hurt and are fatigued often become unfit, and many, many studies has shown that becoming more physically fit is beneficial for FM patients. I think one should realize this helps but doesn't cure FM, because of problems of tolerance and long-term compliance.
The general guidelines are don't start exercise immediately when you see a patient. Maybe after you got them sleeping better, and they have some success with pain medications, and you're getting to know them better. Start with really low-impact exercise, try to avoid excentric muscle contractions, and increase the intensity very slowly. And, I think that sort of a useful, philosophical point is that both the physician and patient should consider exercise as a drug. Once patients get this into their mind and realize the benefits, they want to continue.
Psychological distress. A lot of patients with FM become depressed-up to 60% in many studies. Anxiety, and PTSD, are increasingly recognized. But basically, having a condition that affects your quality of life, as FM does, produces an existential crisis. In fact, one of the major causes of psychological distress in many patients is the fact that their doctor doesn't really believe that they hurt. And we might call that iatrogenic psychological distress. The treatment of psychological distress is the same as the treatment of depression and anxiety in any other patient. It's also worthwhile considering cognitive behavioral therapy.
This has been used in many chronic conditions. It basically helps people to look on a positive side. And so the psychologist who is administering this really has to do a careful analysis and identify what we might call maladaptive thoughts and behaviors. And again, the studies that have been done in FM have shown that cognitive behavioral therapy, particularly when it's combined with exercise, are two useful, non-pharmacological remedies.
Non-restorative sleep. This was the first abnormality really ever described in FM, the alpha-delta sleep. And, one should analyze sleep. Is there a primary sleep disorder? Some patients do have a sleep apnea. And, that's a major cause of fatigue, and obviously, can be treated. Nearly all FM patients, at least, in our clinic, up to 60% have restless leg syndrome, and that certainly is something that can be easily managed with dopamine-agonists. In many patients you need to discuss the basic sleep hygiene measures; going to bed at the same time, in a cool, dark room. Not having caffeine etc. just before going to bed. To minimize nocturnal pain, you can often give high doses of analgesics, at nighttime. Minimize psychological distress; we've already discussed that. And ask the patient about exercise. If people exercise within a couple of hours of going to bed, that sort of stimulates them and it's difficult for them to get to sleep. If they don't exercise at all, they're not going to get tired during the day. So, exercise is important.
The medications to sleep, the commonest ones, the tricyclics, particularly amitriptyline. I think there's increasing use of trazodone, as I talk to doctors, around the country. Many patients prefer the short-acting hypnotics, particularly zolpidem, and the muscle relaxants, FlexerilÒ, and this usually really low dose, 5 mg, 2.5 mg. And tizanidine again, which we mentioned as a muscle relaxant that helps sleep.
And lastly there are the associated syndromes, and there are many of those. The major ones-the irritable bowel, irritable bladder-which have very specific treatments; and there are some newer treatments for irritable bowel available now. The restless leg is very common, and we usually start out with L-dopa, or carbidopa, that's SinemetÒ, and if that doesn't work, or the effects wear off, go to a dopamine agonist, such as miropex or ropinirole. And lastly, consider the possibility of postural hypotension. Neurally mediated hypotension is said to occur in about 30% of FM patients, and can only be really diagnosed by a tilt-table test, but these patients have a lot of fatigue when they exercise. When they stand up they feel light-headed and dizzy, and this could be helped just by increasing the salt and fluid intake and in some other cases one needs to give plasma expanders such as florinef.
So, in conclusion, FM patients differ a lot and a thorough individual evaluation is important. Some people require more complex management program than others, and in these, a comprehensive, multi-disciplinary approach is needed. And one has to employ both pharmacological and nonpharmacological interventions. Thank you.
Leslie J Crofford
Rob, I'd like to thank you so much for your presentation and we'd now like to move into a group discussion of FM and treatment of FMS. I'd like to start by asking both Don Goldenberg and Rob Bennett their opinions of the strength of the evidence that FMS is a central pain syndrome. Could we start with Don?
Don Goldenberg
I think the evidence is overwhelming in many different ways. Intuitively, because we certainly know that there is no major pathology in the muscles, the tendons, and the periphery of the body. It makes logic that this is not a peripheral pain disorder. The association with numerous other symptoms - headaches, chronic fatigue, mood disturbances, irritable bowel syndrome - would lead us to believe that. And all the research that we have been discussing already about imaging studies, about wind up, about central pain mechanisms, and even the responsiveness to treatment, point clearly the central mechanisms.
Leslie J Crofford
Rob, do you have anything you'd like to add?
Robert Bennett
I think I certainly agree with Don in that respect. Basically these people are tender all over, we've known that and that's part of the definition, and yet there doesn't seem to be a peripheral cause for their pain. Of course the very first abnormality that suggested a central change was the finding of elevated levels of substance P in the CSF which was found I think in 1989 by Vaeroy, and I think there have been at least four other studies that basically substantiated that more recently. Of course substance P is a neurotransmitter involved in pain. More recently elevated levels of nerve growth factors have been found - up to 5 fold. So these do suggest there is something central going on. And then of course there's the very nice work on functional MRI showing that FM patients basically have more activity in the cerebral cortex and so in the central areas, and also to a certain extent in the deeper areas, the areas involved in emotional responses, to any given level of stimulus. So that, and some of the things that Don mentioned, like wind up and temporal summation; there is about 5 or 6 abnormalities that have now been described. But I think overwhelmingly they indicate the part of the problem, in FM at lease, is an abnormality of central processing of stimuli.
Leslie J Crofford
So I think that we all agree then that by and large we would want to be using therapies that were addressed at these central mechanisms rather than the typical treatments that we would apply to a peripherally generated nociceptive pain. So in terms of the implications for treatment maybe Rob I could start with you: your opinion about how the mechanism of pain impacts on the type of treatments that one might use for FM.
Robert Bennett
First of all I think it is important to realize that pain is very complex and is made more complex in humans by the fact that we have large cerebral cortex so we interpret pains in different ways, and obviously many of the experimental animals are lacking that aspect. But apart from the changes that we've already described we have to look at this from the point of view of maybe changing maybe the transmission of the level of the spinal cord and also we have to look at other pharmacological and nonpharmacological ways in which we can help patients cope with pain. So if we discuss, first of all, trying to dampen pain at the level of the spinal cord - this is where, typically, analgesics work. Drugs such as Tylenol seem to have some central effects. Others see the opioids, both the schedule 3 and the schedule 2, work to a certain extent at this level. So I think we're having to use medications that target central pain. Drugs that seem to help peripheral pain such as the NSAIDs don't seem to be particularly effective in the management of FM. Although I should say that many FM patients who do have what we call peripheral pain generators like OA are helped by NSAIDs, so it is not that we aren't using them, but we aren't using them for the central pain.
So the drug we will be using in these patients are drugs like opioids, particularly the ones that are not scheduled such as tramadol and tramadol/acetaminophen combinations, codeine, hydrocodone, and then, in about 15% of the patients we see, they are actually on schedule-2 narcotics, but most of our patients are not on those and that's because we often use what we would call adjunctive medications - things like membrane stabilizers that are typically drugs that are used in the treatment of epilepsy, and NMDA [N-methyl-D-aspartate] receptor antagonist particularly dextromethorphan. These are the drugs that we will commonly use. Another one that we commonly use actually, which activates the descending pain pathways and inhibits pain at the level of the spinal cord is tizanidine, which is an alpha-2 adrenergic agonist. So that's a sort a summary of the medications that we commonly use to target the spinal cord level.
Leslie J Crofford
Don, do you want to comment on other drugs that may be effective when we are targeting particularly descending inhibition?
Don Goldenberg
Obviously I would say the drugs that most physicians pick first in treating people with this condition are antidepressants, and it is intriguing how they work in these conditions, and how they work in all pain disorders. Amitriptyline for example, where there have been more randomized clinical trials showing evidence of at least some efficacy in all the domains of FM including pain, has been using in many different pain disorders. So we don't know if it's working possibly on central pain, descending pain pathways, if it partly is related to the cognition and the mood improvement. I always tell my own patients that some day I hope that as research gets more specific, using the analogy that as central mechanisms, for example, in understanding pain and migraine got more specific, scientists were able to come up with more targeted sites and generation of medicines like the triptans were used specifically for that type of pain. And I can see as these imaging studies and more biologic studies become even better in the next decade that we're going to be able to discover more specific targeted areas of pain pathways. I must admit that I am always very much intrigued by the fact that FM of course is not just a pain disorder and the multiple aspects - fatigue, depression, irritable bowel syndrome, and headaches - mean that in thinking about management, pain management or global management, we need to be quite universal.
Leslie J Crofford
I think that that is an interesting point specifically from the standpoint of the acetaminophen/tramadol study, Rob, that you reported, and in that particular study there seems to be a bit of dissociation between the response of the pain symptom compared with the fatigue symptom. And this is something that I think we oftentimes see with analgesics-type therapies. Do you have any comments about the treatment issues having to do with using combination treatments to address symptom domains?
Robert Bennett
This study actually was unusual in respect to FM studies in that patients were allowed to take low doses of short active hypnotics, particularly zolpidem, and also still allowed to take the SSRIs [selective serotonin re-uptake inhibitors]. So fatigue, you might not have expected it to improve very much because they were already getting fairly restful sleep, and also depression didn't improve. I've often wondered whether that's why we didn't see those changes, at least on the FIQ scores, for this. But as regards the general problems with fatigue, obviously this is an extremely important problem in FM, and I think the assumption has been very often that the very first abnormality ever described in FM was an abnormality of deep sleep, stage-3 and -4 sleep, so called alpha-delta sleep disturbance, in 1976 by Dr Moldofsky. And if you ask the question "How many days a month do you wake up feeling refreshed?" which I think it is a useful question, many of these patients will say zero, or just two or three. In other words they have very nonrestorative sleep. And we assume this is related to sleep disturbance. But sometimes medication can cause fatigue, and we know fatigue is a common component of depressive illness. And another cause of fatigue in some FM patients is neurally-mediated hypotension, particularly fatigue when they are standing or post-exercise. So there are several things to think about in managing the fatigue of FM.
Leslie J Crofford
Don, what is your approach to treating the multiple symptoms of FM? Do you try to start with medication that addresses multiple symptoms and then move to more specifics? Or can you give us a little bit more of your strategy for treating?
Don Goldenberg
I'd be happy to. My overall idea would be first to sit with a person and educate them and their family about what FM is. I think there is so much confusion and misinformation out there that either we as rheumatologists, primary care physicians, specialists in physical medicine and rehabilitation, we need to give these people insight into the multiple problems and as you mentioned the multiple domains that occur with FM. I would usually begin medications before anything else. My own choice usually would be a low dose of something to help their sleep, and I usually favor the tricyclic medications in low dose at night time although is somebody is also very fatigued might add a SSRI. I would do those actually before I would address more the specific pain approaches that Dr Bennett was talking about with tramadol and other medications. At the same time I work very hard in starting an exercise program with the patient explaining the importance of that, maybe working with allied health professionals in setting that up, but really setting the stage where they understand that regular exercise and stretching and some slow incremental cardiovascular exercise is critical, and again there is a lot of evidence that that kind of therapy is probably as important as medicinal approaches, but I think if we can start the medications first, we have a bit more ability to keep people complying with their exercise.
Leslie J Crofford
That's a nice segue and I'd like to finish with asking Rob if he can provide any insight as to how the important treatment of exercise and cognitive modalities may work in FMS.
Robert Bennett
Yes. I would like to comment on something Dr Goldenberg said. I think it is very important. These patients have a lot of problems that are complex. We've got the pain, the psychological distress, we've got people who are deconditioned, we have irritable bowel and bladder and all these things. And you have to start in some order and I agree entirely with Don that the two most important things to start with are first of all education and trying to alleviate psychological distress. Now of course these people are often depressed, they are often anxious, but another cause of psychological distress is what I have called iatrogenic. Some doctors don't believe in FM, or don't want to treat these people, and so sometimes that is very critical, that one establishes a good relationship with these patients and start on antidepressants if it is necessary before getting on to other components of the program. For instance we don't usually tell people to start exercise until we have their confidence, and that they feel comfortable with us. So this might be one, two, three months down the line. But all the studies that have been done in FM, nearly all of them have shown some benefit from low-grade exercise. And you might ask how does this help, because I think it is a good question, because undoubtedly FM patients when they start to exercise hurt more. But as they get into an exercise program and persist with it they start to feel somewhat improved. I think there are probably several reasons why they benefit from exercise. We know that exercise stimulates endorphins. But I think even more important, it is probably a cognitive behavior aspect to exercise. Many of these people think that they cannot do these things. But when they start to realize that they can do things it helps their self-efficacy. And self-efficacy improvement has been shown to be important in many chronic diseases and not just FM. And another aspect of exercise is if you do it at the right time of day, you can feel tired enough to get a good night's sleep. So I think there are many ways in which exercise is probably important in FM. I mean it has also been shown to be of some benefit in depressive illness as well. And the same probably goes for cognitive behavioral therapy, which has been used a lot in FM as another non-pharmacological type of treatment that seems to be having some benefit: it improves self-efficacy.
Leslie J Crofford
And certainly there is some idea that central pain syndromes may be what they call in animal physiology restorative pain rather than danger pain so that the brain in and of itself has a strong drive towards quite quiescence and that we actually need to help patients to get over this almost normal response to central pain problem.
Robert Bennett
Yes, there is a very interesting article in Pain this month showing that some sort of activity has an effect on the appreciation of pain. I mean I guess maybe that's just a diverting effect, but again I think that is important.
Leslie J Crofford
