- Email: [email protected]
Neurohormonal perturbations in fibromyalgia
9 Neurohormonal perturbations in fibromyalgia L E S L I E J. C R O F F O R D N. C A R Y E N G L E B E R G M A R K A. D E M I T R A C K
Fibromyalgia (FM) is a poorly understood syndrome characterized by chronic, widespread musculoskeletal pain and the presence of tender points on physical examination (Wolfe et al, 1990). Other clinical manifestations include fatigue, sleep disturbance, headache, and neuropsychological complaints (Yunus et al, 1981). The concept that disorders such as FM might by associated with 'subtle and undetectable' disturbances in the central nervous system was introduced as early as 1869 by Beard in his description of neurasthenia, a syndrome characterized by marked somatic and constitutional complaints, but few physical findings (Beard, 1869). Although the aetiology and pathogenesis of FM remain elusive, recent years have seen great progress toward defining neurochemical abnormalities in FM. A clue to the pathophysiology in FM may be found in the clinical observation that patients with FM frequently report the onset of their illness following a significant physical or psychological stress (Crofford et al, 1994; Demitrack and Crofford, 1995). In addition, patients with FM report that the course of their syndrome is significantly influenced by stress (Yunus et al, 1981). Finally, patients with FM report a high level of perceived daily stress (Dailey et al, 1990). For these reasons FM falls into the spectrum of what might be termed 'stress-related illness'. Other clinical syndromes, including chronic fatigue syndrome (CFS), irritable bowel syndrome, headache syndromes, irritable bladder (female urethral syndrome, interstitial cystitis), temporomandibular joint syndrome, and dysmenorrhoeic syndromes, share substantial symptomatic overlap and cooccur with FM (Hudson et al, 1992). These syndromes are also associated with high levels of stress (Buchwald et al, 1987; Goldenberg et al, 1990; Hudson et al, 1992). It is well known that there is a stereotypic response to endogenous or exogenous stress, mediated primarily through activation of the hypothalamic-pituitary-adrenal (HPA) axis and the sympathetic nervous system (Chrousos and Gold, 1992). These physiological responses are designed to respond to any internal or external disturbance to homeostatic equilibrium, Bailli~re's Clinical Rheumatology-365 Vol. 10, No. 2, May 1996 1SBN 0-7020--2182-2 0950-3579/96/020365 + 14 $12,00/00
Copyright 9 1996, by Baillirre Tindall All rights of reproduction in any form reserved
366
L . J . CROFFORD ET AL
which is the way we will define stress. A stressor may be metabolic, physiological, traumatic, inflammatory/infectious, psychological, or emotional in nature. Patients with FM were shown to have abnormalities of both HPA axis and sympathetic stress-response systems (Griep et al, 1993; Crofford et al, 1994). It is important to emphasize, however, that the primary stress-response systems share important connections with other neuroendocrine and neurochemical systems that may also be disturbed in patients with FM (Chrousos and Gold, 1992). In this chapter, we will review the HPA axis, sympathetic nervous system, and other neuroendocrine systems in patients with FM and related disorders. We will also present a coherent framework to link these biological findings to susceptibility and clinical manifestations of FM. THE HPA AXIS Consistent with the concept that FM is a stress-related illness, we and others have demonstrated perturbations of the HPA axis (Table 1). The first abnormality described was a low 24-hour urine-free cortisol compared with either normal subjects or patients with rheumatoid arthritis (McCain and Tilbe, 1989; Crofford et al, 1994). FM patients, however, exhibited normal morning (peak) and elevated evening (trough) cortisol levels resulting in a loss of the normal diurnal cortisol fluctuation (McCain and Tilbe, 1989; Crofford et al, 1994). Abnormal cortisol measurements were most prominent in patients with a longer duration (> 2 years) of disease (McCain and Tilbe, 1989). No differences in levels of cortisol binding globulin have been detected in FM patients (Crofford et al, 1994). Potential reasons for these appparently disparate findings become clearer when one considers the detailed circadian dynamics of the HPA axis. For example, there are normally eight to nine peaks of cortisol over a 24-hour period, presumably in response to basal surges of corticotrophin-releasing hormone (CRH) and adrenocorticotrophic hormone (ACTH). The majority of these cortisol surges occur during the night-time circadian peak. In FM, the amplitude of the cortisol peaks might be normal or even elevated, but the frequency of the peaks could be decreased resulting in a low cortisol output when examined over 24 hours. Further study of the basal circadian HPA axis dynamic in FM is clearly indicated. Table 1. HPA axis perturbations in FM. Basal Decreased 24-hour urine-free cortisol Normal peak, elevated trough total and free plasma cortisol levels Blunted circadian change in plasma cortisol levels Normal plasma cortisol binding globulin levels Stress Enhanced ACTH release after exogenous CRH and insulin-induced hypoglycaemia Blunted cortisol response after exogenous CRH and insulin-induced hypoglycaemia, even in the face of exaggerated ACTH release Blunted cortisol response to exercise
NEUROHORMONAL PERTURBATIONS IN FM
367
The HPA axis response to administration of exogenous CRH has been used as an index of the stress-induced activation of the axis, whereas insulin-induced hypoglycaemia tests the integrity of the stress-associated increase in endogenous hypothalamic CRH as a stimulus to pituitary-adrenal hormone secretion. CRH-stimulation tests performed in the evening, when FM patients exhibit elevated basal trough cortisol levels compared with control patients, revealed no statistically significant differences in CRH-stimulated ACTH levels between patients with FM and controls. However, a non-significant trend toward elevated mean basal and CRH-stimulated ACTH levels was noted (Crofford et al, 1994). Griep et al (1993) performed stimulation testing of HPA axis function using exogenous CRH and insulin-induced hypoglycaemia. They demonstrated significantly elevated ACTH responses to both stimuli in patients with FM. Their studies were performed in the morning at the circadian peak, when there are no differences in basal cortisol levels compared with controls. Elevated trough cortisol levels in patients with FM as compared with normal subjects may have blunted the effect of CRH on ACTH in the evening by negative feedback mechanisms. Both studies demonstrated a relatively blunted adrenal cortisol response to stimulated ACTH release (Griep et al, 1993; Crofford et al, 1994). Further evidence for perturbed HPA axis function is demonstrated by significantly lower cortisol levels after exercise in FM patients as compared to control subjects (van Denderen et al, 1992). The reasons for this response are unclear, but could result from a relatively hyporesponsive adrenal gland, could be due to intrinsic hypoactivity of the adrenal cortex itself, or could develop in the context of chronic understimulation due to deficiencies of CRH or ACTH (Sternberg, 1993). HPA axis function has also been examined in patients with the related chronic fatigue syndrome (CFS) (Demitrack et al, 1991). Comparison of HPA axis function in FM with patients with CFS reveals differences despite the significant clinical overlap between these patients. Low 24-hour urinefree cortisol excretion, similar to that seen in patients with FM, was demonstrated in patients with CFS (Demitrack et al, 1991). However, in contrast to FM, in CFS, evening plasma ACTH levels were elevated and cortisol levels were reduced compared to control subjects (Demitrack et al, 1991). CRH-induced stimulation of pituitary-adrenal secretion revealed an attenuated, rather than an exaggerated ACTH response in CFS patients. Additionally, stimulation of the adrenal gland revealed low maximal cortisol responses to ACTH, compatible with secondary adrenal atrophy. We postulated that these data suggested central CRH insufficiency in CFS patients. While preliminary studies performed by Beam et al (1995) failed to confirm changes in ACTH or cortisol in response to hypoglycaemia or serotonergic stimulation with d-fenfluramine in the morning in patients with CFS, further studies by the same group demonstrated decreased plasma cortisol responses to d-fenfluramine in CFS patients compared with normal and depressed controls (Cleare et al, 1995). Although there are differences in dynamic function of the HPA axis in FM and CFS patients, both patient groups share low 24-hour urine cortisol levels. The possibility exists that differences in provocative tests between
368
L.J.
CROFFORD
ET A L
FM and CFS could be explained on the basis of differences in other ACTH secretagogues. For example, these syndromes could represent different forms of insufficient stimulation of the HPA axis, with both syndromes expressing low hypothalamic CRH, but FM characterized by increased exposure of the corticotrophs to arginine vasopressin (AVP) and CFS patients with decreased AVP levels. In fact, there is preliminary experimental data that support this view. Our group detected a trend towards increased release of AVP after postural change in patients with FM (Crofford et al, 1994). Five of 12 FM patients achieved AVP levels more than two standard deviations above the highest control value. In contrast, Bakheit et al (1993) evaluated a group of patients with postviral fatigue syndrome and demonstrated significantly lower basal AVP levels. CFS patients also demonstrated a lack of correlation between serum and urine osmolality with plasma AVR These data, however, should be interpreted with caution since plasma AVP reflects release of AVP from the magnocellular nucleus of the hypothalamus, not the parvocellular nucleus that releases AVP to the hypophyseal venous circulation. THE SYMPATHETIC NERVOUS SYSTEM The HPA axis and sympathetic nervous system are intimately linked as the predominant mediators of a co-ordinated stress-response system (Chrousos and Gold, 1992). Functional studies of sympathetic nervous system function in FM, as measured by changes in skin microcirculation, suggest dysfunction of the adrenergic component of peripheral sympathetic activity (Vaer0y et al, 1989; Qiao et al, 1991). In studies of muscle sympathetic activity, there was no difference between FM patients and controls at rest. However, after stimulation of muscle sympathetic activity by static hand grip, contraction of the jaw muscle, or mental stress, there was a lack of exaggerated sympathetic activity and, in fact, a tendency to lower muscle sympathetic activity (Elam et al, 1992). The observation that some patients with chronic fatigue syndromes may have identifiable neurally-mediated hypotension (Streeten and Anderson, 1992; Rowe et al, 1995) has prompted an evaluation of autonomic dysfunction as measured by tilt-table testing in patients with FM. Clauw et al (1995a) demonstrated that patients with FM, who were not chosen on the basis of symptomatology, displayed significantly smaller pulse pressures in the upright position and higher blood pressures on return to the supine position. Further studies will be needed to clarify the utility of functional tests of autonomic function in patients with both FM and CFS. Early studies of sympathetic nervous system hormones in FM patients either failed to show difference in plasma or urine catecholamine levels (Yunus et al, 1992), or demonstrated a subgroup of patients with elevated urinary norepinephrine levels (Russell, 1989). In CFS, patients were shown to have a significant reduction in basal plasma levels of the monoamine metabolite 3-methoxy-4-hydroxyphenylglycol (Demitrack et al, 1992). Neuropeptide Y (NPY) co-localizes with norepinephrine in the sympathetic
N E U R O H O R M O N A L PERTURBATIONS IN FM
369
nervous system, and plasma NPY can serve as a correlate of sympathoadrenal output (Lundberg et al, 1990). In human subjects, elevated plasma NPY levels are seen with heavy physical exercise or other situations of strong sympathetic activation (Lundberg et al, 1990). Our group demonstrated that patients with FM had significantly lower plasma NPY levels than matched control subjects (Crofford et al, 1994). These findings have been confirmed by Clauw et al (1995b) who demonstrated significantly lower basal NPY levels in FM patients, who also met criteria for CFS, compared with controls. While reduced NPY levels could result from a decrease in the level of physical activity in these patient groups, this is unlikely to be the sole explanation since the study by Clauw et al was performed on patients that were ambulatory and performing regular exercise. Centrally, NPY increases the concentration of CRH in the hypothalamus, and is present in high concentrations in the portal-hypophyseal venous circulation where it acts synergistically with CRH to stimulate release of ACTH (Haas and George, 1987; Koenig, 1990). NPY has also been reported to have ACTH-like activities at the level of the adrenal cortex (Kamilaris et al, 1989). These findings provide mechanisms by which NPY interacts with HPA axis function. The significance of low plasma NPY levels in patients with FM is unclear. Nevertheless, low NPY levels may represent another measure of hypofunction or depletion of the sympathetic stress axis. The relationship between autonomic dysfunction and neuroendocrine abnormalities in FM has not yet been fully defined. SEROTONIN
Serotonin influences the circadian fluctuation of HPA axis products (Krieger and Rizzo, 1969; Scapagnini et al, 1971). Numerous studies have shown that serotonin and serotonin agonists stimulate the pituitary-adrenal system, probably through stimulation of CRH release from the hypothalamus, and that density of serotonin receptors and serotonin levels parallel HPA axis activity (Jones et al, 1976; Buckingham and Hodges, 1979; Holmes et al, 1982; Burnett et al, 1992). There is continuing controversy about the role of serotonin during stress-induced stimulation of the HPA axis. However, serotonin enhances the HPA axis response during insulin-induced hypoglycaemia, perhaps due to a direct effect of insulin on the availability of tryptophan to the central nervous system (Yehuda and Meyer, 1984). It is not yet known whether abnormalities in serotonin pathways could result in the HPA axis perturbations as observed in FM patients. The metabolism of serotonin and its precursor, tryptophan, are abnormal in FM patients (Russell, 1989). The serum concentration of serotonin is lower in patients with FM than matched controls, and patients with FM have an increased number of serotonin re-uptake receptors on platelets (Russell et al, 1992). There is an inverse correlation between the level of free serum tryptophan, a precursor of serotonin, and severity of pain in FM patients, although tryptophan levels were in the normal range (Moldofsky and Warsh, 1978). In CFS, we have demonstrated significant increases in
370
L. J. C R O F F O R D
ET A L
plasma levels of the serotonin metabolite, 5-hydroxyindoleacetic acid (5HIAA) (Demitrack et al, 1992). In addition, serotonin neurotransmission, as assessed by prolactin responses to the serotonin receptor agonist, buspirone, is increased in CFS relative to healthy and depressed control groups (Bakheit et al, 1992). Taken together with the d-fenfluramine challenge abnormalities in CFS, these studies suggests that serotonergic systems may be disturbed in CFS patients as well (Cleare et al, 1995).
THE GROWTH HORMONE AXIS Patients with FM have been noted to have low basal levels of growth hormone (Griep et al, 1994) and insulin-like growth factor-1 (IGF-1 or somatomedin C), which reflects integrated growth hormone stimulation (Bennett et al, 1992). Analogous to the exaggerated release of ACTH after insulin-induced hypoglycaemic stress, exaggerated release of growth hormone is seen after the same stimulus in patients with FM (Griep et al, 1994). Bennett et al (1995a) reported blunted growth hormone secretion from the pituitary after administration of clonidine. This dichotomy suggests that growth hormone axis hypofunction most probably occurs at the level of the hypothalamus or other higher brain regions. The causes and consequences of these changes in growth hormone physiology are not completely understood. However, since secretion of growth hormone is maximum during sleep, it has been proposed that the sleep disturbance may be responsible f o r perturbations in the growth hormone axis in FM. It should also be noted that HPA axis hormones have profound effects on the growth hormone axis, through direct effects of CRH to stimulate somatostatin secretion, and suppressive effects of glucocorticoids on secretion of growth hormone itself from the pituitary (Chrousos and Gold, 1992). Mechanisms by which both cortisol and growth hormone/IGF-1 are low in patients with FM remain to be determined.
THE THYROID HORMONE AXIS Symptoms of FM are remarkably similar to patients with hypothyroidism (Becker et al, 1963; Carette and Lefran~ois, 1988). Although basal levels of thyroid hormones are normal, two independent groups have reported blunted thyroid-stimulating hormone (TSH), but exaggerated prolactin response to administration of thyroid hormone releasing hormone (Ferraccioli et al, 1990; Neeck and Riedel, 1992). Similar to the growth hormone axis, CRH stimulates somatostatin, which exerts inhibitory effects on the release of TSH from the pituitary. In addition, glucocorticoids antagonize release of TSH and actions of thyroid hormone in the periphery (Chrousos and Gold, 1992). Again, further studies of the interactions between the HPA and thyroid axis hormones in FM are needed.
NEUROHORMONAL
P E R T U R B A T I O N S IN FM
371
THE H P G AXIS
The incidence of FM is dramatically higher in women, and the onset of disease often correlates with the onset of menopause (Russell, 1989). This has prompted the hypothesis that changes in levels of the gonadal steroids may modulate the development of FM (Wasman and Zatzkis, 1986). Carette et al (1992) compared levels of HPG axis hormones in 10 patients with FM compared with normal control subjects. They were unable to demonstrate significant differences in oestrogens, androgens, or gonadotrophins, with the exception that testosterone levels were decreased in FM patients. Nevertheless, gonadal steroids play a role in the regulation of the HPA axis. There is in vitro evidence that oestrogen directly increases CRH expression (Vamvakopoulos and Chrousos, 1993). Studies have also demonstrated an effect of gonadal steroids and glucocorticoid negative feedback, with oestrogen leading to impaired shut-off of ACTH and corticosterone secretion following stress (Young, 1996). In general, both oestrogen and progesterone appear to antagonize glucocorticoid negative feedback, but some effects of oestrogen may be reversed by progesterone (Young, 1996). Hormonal status may be important for development of HPA axis perturbation. For example, virtually all post-menopausal women develop HPA axis dysregulation during depression, while pre-menopausal women are relatively resistant to changes in HPA axis function in depressive states (Young et al, 1993). Finally, it should be noted that HPA axis hormones exert reciprocal effect on the reproductive hormone axis, leading to decreased overall activity of the axis (Chrousos and Gold, 1992). N E U R O H O R M O N A L SYSTEMS AND THE AETIOPATHOGENESIS OF FM
The observed neuroendocrine abnormalities in FM patients could be a contributing cause to the aetiology or pathogenesis of FM, a consequence of the illness, or an association with other factors causal for FM. In support of the hypothesis that neuroendocrine dysfunction may be important to susceptibility to FM or to clinical features of the syndrome (Figure 1), it is important to note that the symptoms of FM are similar to many neuroendocrine deficiency states. For example, it is of considerable interest to note the clinical consequences of inadequate HPA axis activation. Subtle grades of HPA axis insufficiency may occur due to abnormalities at the level of the hypothalamus or higher central nervous system centres, the pituitary, the adrenal, or even at the level of target tissue responsiveness to cortisol. By way of example, cortisol resistance due to a genetic abnormality of the glucocorticoid receptor was reported to present with fatigue as the only symptom (Br6nnegard et al, 1986). It has been suggested that a number of syndromes that share clinical features with FM, including CFS, the depressed phase of seasonal affective disorder or other atypical depressive syndromes, and hypothyroidism may share a functional deficit of hypothalamic CRH (Chrousos and Gold, 1992; Sternberg, 1993).
372
L. J, CROFFORD ET AL
Vulnerability to Stress Disorders 9 Genetic 9 Hormonal 9 Environmental
?Mechanism: 9SNS contribution 9 SerotonergicJ system ~F" input
Syndrome Triggers 9 Trauma 9 Infection 9 Emotional stress
Perpetuating factors: Physical deconditioning
Psychological response to illness
9 Stress
Adrenal FM:
~CRH
"I'AV___PP CFS: ~,AVP
FM: 'I~ACTH
Cortisol
CFS:~ACTH
" Central Sequelae 9 Sleep disorder 9 Cognitive dysfunction ? Fatigue .i
eripheral Sequela= 9 Pain syndromes
Figure 1. Neurohormonal hypothesis of the pathogenesis of flbromyalgia (FM). Potential relationships between the HPA axis and aetiopathogenic mechanisms of disease in FM. We hypothesize that low levels of CRH and cortisol may contribute to both central and peripheral disease manifestations, and that factors which perpetuate disease may be integrated through HPA axis activity.
CRH serves not only as a principal stimulus to the HPA axis, but also as a behaviourally active neurohormone whose central administration to animals and non-human primates induces signs of physiological and behavioural arousal including activation of the sympathetic nervous system, hyper-responsiveness to sensory stimuli, and increased locomotion (Sutton et al, 1982; Swerdlow et al, 1982). Hence, a relative or absolute deficiency of hypothalamic CRH could contribute to the profound lethargy and fatigue that are characteristic of these syndromes (Chrousos and Gold, 1992; Stemberg, 1993). Although this is an attractive model, it should be emphasized that, to date, the assessment of CRH activity in these patient
NEUROHORMONAL
P E R T U R B A T I O N S IN FM
373
groups is inferential, based primarily on peripheral pituitary-adrenal responses to hormonal challenge. Even direct measurement of CRH in the cerebrospinal fluid may merely reflect cortical or spinal sources of CRH, and not functional CRH activity in the paraventricular nucleus (PVN). Moreover, since the behavioural effects of CRH are produced at disparate sites within the central nervous system, there is no reason to presume that a functional deficit of CRH in the PVN is associated with similar reductions of CRH activity in limbic or cortical locations. The musculoskeletal pain that is the dominant clinical feature of FM could also be influenced by HPA axis abnormalities. Myalgia, arthralgia, and muscle weakness are features of steroid withdrawal syndromes, and FM had been documented after hypophysectomy for Cushing's disease (Disdier et al, 1991). Both CRH and AVP neurons, as well as sympathetic neurons, have connections to brain pro-opiomelanocortin containing neurons that produce opioids (Nikolarakis et al, 1986). Descending opioid peptidergic pathways may be important in influencing the biochemical determinants of pain at brain stem and spinal cord levels, such as substance R which is elevated in patients with FM (Vaer0y et al, 1988; Russell, 1995). Other neuroendocrine or neurochemical systems may influence HPA axis function in FM. The high prevalence of FM in women, as well as the observed increase in incidence at the time of menopause, could be the result of sexual dimorphism in the development of stress-response systems and/or the influence of gonadal steroids on stress-response systems. For example, oestrogen impairs glucocorticoid negative feedback at the level of the hippocampus leading to a functional increase of HPA axis reactivity. Decreased oestrogenic stimulation at the time of menopause could be permissive for relative hypofunction of the HPA axis, contributing to the development of FM or increased severity of FM symptoms. The role of sympathetic nervous abnormalities as a contributing factor to the pathogenesis and persistence of symptoms in FM and related syndromes, such as CFS, remains to be fully elucidated. Ongoing studies will assess autonomic function in larger groups of patients with FM and other fatigue states, and future controlled treatment trials should be performed in patients demonstrating abnormalities (Streeten and Anderson, 1992; van Denderen et al, 1992; Clauw et al, 1995a; Rowe et al, 1995). Abnormalities in the tryptophan/serotonin metabolic pathway were among the first neurochemical abnormalities to be demonstrated in patients with FM (Russell et al, 1989; Russell et al, 1992; Hrycaj et al, 1993; Russell, 1995). The HPA axis abnormalities observed in patients with FM could be related to alterations in serotonin receptor activation (Russell, 1995). Serotonin abnormalities could be related to the sleep disturbance seen in FM patients, and which may contribute to other symptoms of pain and fatigue. In addition, serotonin may contribute to aberrant pain perception in FM, via interactions with substance P at the level of the spinal cord (Russell, 1995). Finally, the growth hormone-IGF-1 axis is an important anabolic stimulus for musculoskeletal tissues (Bennett et al, 1992). Hence, reductions in
374
L.J. CROFFORD ET AL
circulating levels of these substances, as seen in FM, may be related to the diffuse myalgias which are characteristic of FM. IMPLICATIONS OF NEUROENDOCRINE PERTURBATIONS FOR FM TREATMENT STRATEGIES
One of the most compelling reasons for considering HPA axis and sympathetic system abnormalities as the key to understanding FM, is that many therapeutic strategies useful in FM either modulate patient perceptions of stressors (cognitive behavioural therapy) or directly influence function of stress-response systems (aerobic exercise, tricyclic antidepressants) (Bradley, 1989; McCain, 1989; Carette, 1995; Sharpe, 1995). These observations support the hypothesis that central determinants of HPA axis function, not primary adrenal insufficiency, are operative in FM. This is particularly true since cortisol replacement has never been shown to be useful as a treatment (Clark et al, 1985). Further investigation into the nature of the HPA axis abnormalities and the role of other central nervous system hormones in modulating HPA axis function in patients with FM may lead to improved treatment strategies. Direct testing of the hypothesis that low growth hormone and IGF-1 levels play a role in the symptoms of FM are currently underway. Bennett and co-workers have undertaken a double-blind placebo controlled trial of growth hormone replacement therapy in FM patients with low IGF-1 levels (Bennett et al, 1995). There was significant improvement within the group of patients treated with growth hormone over a 9-month period, and growth hormone resulted in significant improvement compared with placebo (Bennett et al, 1995). Further studies comparing growth hormone with other treatment modalities may clarify its role in the overall treatment of patients with FM. A therapeutic goal in patients with FM may be to restore physiological neuroendocrine and neurochemical function. Since it is likely that the FM syndrome represents a final common pathway for a number of aetiopathogenic mechanisms, therapy may need to be directed towards a number of different central nervous system pathways. SUMMARY
Fibromyalgia (FM) falls into the spectrum of what might be termed 'stressassociated syndromes' by virtue of frequent onset after acute or chronic stressors and apparent exacerbation of symptoms during periods of physical or emotional stress. Patients with FM exhibit disturbances of the major stress-response systems, the HPA axis and the sympathetic nervous system. Integrated basal cortisol levels measured by 24-hour urine-free cortisol are low. FM patients display a unique pattern of HPA axis perturbation characterized by exaggerated ACTH response to exogenous CRH or to endogenous activators of CRH such as insulin-induced hypo-
NEUROHORMONAL PERTURBATIONS IN FM
375
glycaemia. The cortisol response to increased ACTH in these stress paradigms is blunted, as is the cortisol response to exercise. Functional analysis suggests that FM patients may also exhibit disturbed autonomic system activity. For example, plasma NPY, a peptide co-localized with norepinephrine in the sympathetic nervous system, is low in patients with FM. Abnormalities of related neuronal systems, particularly decreased serotonergic activity, may contribute to the observed neuroendocrine perturbations in FM. Finally, other neuroendocrine systems, including the growth hormone axis, are also abnormal in FM patients. Many clinical features of FM and related disorders, such as widespread pain and fatigue, could be related to the observed neuroendocrine perturbations. This hypothesis is supported by the observation that many useful treatments for FM affect the function of these central nervous system centres. Further clarification of the role of neuroendocrine abnormalities in patients with FM, and the relationship of these disturbances with particular symptoms, may lead to improved therapeutic strategies.
REFERENCES Bakheit AMO, Behan PO, Dinan TG et al (1992) Possible upregulation of hypothalamic 5-hydroxytryptamine receptors in patients with postviral fatigue syndrome. British Medical Journal 304: 1010-1012. Bakheit AMO, Behan PO, Watson WA & Morton JJ (1993) Abnormal arginine-vasopressin secretion and water metabolism in patients with postviral fatigue syndrome. Acta Neurologica Scandinaviea 87: 234-238. Beard G (1869) Neurasthenia, or nervous exhaustion. Boston Medical and Surgical Journal III: 217-221. Beam J, Allain T, Coskeran Pet al (1995) Nearoendocrine responses to d-fenfluramine and insulininduced hypoglycemia in chronic fatigue syndrome. Biological Psychiatry 37: 245-252. Becker KL, Ferguson RH & McConahey WM (1963) The connective tissue disease and symptoms associated with Hashimoto's thyroiditis. New England Journal of Medicine 268: 275-280. Bennett RM, Clark SR, Burckhardt CS & Cook D (1995a) 1GF-1 assays and other GH tests in 500 fibromyalgia patients. Journal of Musculoskeletal Pain 3 (supplement 1): 109(abstract). Bennett RM, Clark SR, Burckhardt CS & Walczyk J (1995b) A double blind placebo controlled study of growth hormone therapy in fibromyalgia. Journal of Musculoskeletal Pain 3 (supplement 1): 110(abstrac0. Bennett RM, Clark SR, Campbell SM & Burckhardt CS (1992) Low levels of somatomedin C in patients with the fibromyalgia syndrome. Arthritis and Rheumatism 35:1113-1116. Bradley LA (1989) Cognitive-behavioral therapy for primary fibromyalgia. Journal of Rheumatology 16 (supplement 19): 131-136. Br6nnegard M, Wemer S & Gustafsson J-A (1986) Primary cortisol resistance associated with a thermolabile glucoeorticoid receptor in a patient with fatigue as the only symptom. Journal of Clinical Investigation 78: 1270-1278. Buchwald D, Goldenberg DL, Sullivan JL & KomaroffAL (1987) The 'chronic active Epstein-Barr virus infection' syndrome and primary fibromyalgia. Arthritis and Rheumatism 30:1132-1136. Buckingham C & Hodges JR (1979) Hypothalamic receptors influencing the secretion of corticotrophin releasing hormone in the rat. Journal of Physiology 290:421-431. Bumett PWJ, Mefford IN, Smith CC et al (1992) Hippocampal 8-[3H]hydroxy-2-(di-n-propylamino)tetralin binding site densities, serotonin receptor (5HT1A) messenger ribonucleic acid abundance, and serotonin levels parallel the activity of the hypothalamopituitary-adrenal axis in rat. Journal of Neurochemistry 59: 1062-1070. Carette S (1995) What have clinical trials taught us about the treatment of fibromyalgia? Journal of Masculoskeleml Pain 3: 133-140.
376
L . J . CROFFORD ET AL
Carette S & Lefrangois L (1988) Fibrositis and primary hypothyroidism. Journal of Rheumatology 15: 1418-1421. Carette S, Dessureault M & Btlanger A (1992) Fibromyalgia and sex hormones. Journal of Rheumatology 19:831 (letter). Chrousos GP & Gold PW (1992) The concepts of stress and stress system disorders: Overview of physical and behavioral homeostasis. Journal of the American Medical Association 267: 1244-1252. Clark S, Tindall E & Bennett RM (1985) A double blind crossover trial of prednisone versus placebo in the treatment of fibrositis. Journal of Rheumatology 12- 980-983. Clauw DJ, Radulovic D, Katz P e t al (1995a) Tilt table testing as a measure of dysantonomia in fibromyalgia. Journal of Musculoskeletal Pain 3 (supplement 1): 10 (abstract). Clauw DJ, Sabol M, Radulovic D et al (1995b) Serum neuropeptides in patients with both fibromyalgia (FM) and chronic fatigue syndrome (CFS). Journal of Musculoskeletal Pain 3 (supplement 1): 79 (abstract). Cleare AJ, Beam J, McGregor Aet al (1995) Contrasting neuroendocrine responses in depression and chronic fatigue syndrome. Journal of Affective Disorders 35: 283-289. Crofford LJ, Pillemer SR, Kalogeras KT et al (1994) Hypothalamic-pituitary-adrenal axis perturbations in patients with fibromyalgia. Arthritis and Rheumatism 37: 1583-1592. Dailey PA, Bishop GD, Russell IJ & Fletcher EM (1990) Psychological stress and the fibrositis/ fibromyalgia syndrome. Journal of RheumatoIogy 17: 1380-1385. Demitrack MA & Crofford LJ (1995) Hypothalmic-pituitary-adrenal axis dysregulation in fibromyalgia and chronic fatigue syndrome: an overview and hypothesis. Journal of Musculoskeletal Pain 3: 67-73. Demitrack MA, Dale JK, Straus S E e t al (1991) Evidence for impaired activation of the hypothalamic-pituitary-adrenal axis in patients with chronic fatigue syndrome. Journal of Clinical Endocrinology and Metabolism 73: 1224-1234. Demitrack MA, Gold PW, Dale JK et al (1992) Plasma and cerebrospinal fluid monoamine metabolism in patients with chronic fatigue syndrome: Preliminary findings. Biological Psychiatry 32: 1065-1077. Disdier P, Harle J-R, Brue T et al (1991) Severe fibromyalgia after hypophysectomy for Cushing's disease. Arthritis and Rheumatism 34: 493-495. Elam M, Johansson G & WaUin BG (1992) Do patients with primary fibromyalgia have an altered muscle sympathetic nerve activity? Pain 48: 371-375. Ferraecioli G, Cavalieri F, Salaffi F et al (1990) Neuroendocrinologic findings in primary fibromyalgia (soft tissue chronic pain syndrome) and in other chronic rheumatic conditions (rheumatoid arthritis, low back pain). Journal of Rheumatology 19: 869-873. Goldenberg DL, Simms RW, Geiger A & Komaroff AL (1990) High frequency of fibromyalgia in patients with chronic fatigue seen in a primary care practice. Arthritis and Rheumatism 33: 381-387. Griep EN, Boersma JW & de Kloet EP (1993) Altered reactivity of the hypothalamic-pituitatyadrenal axis in the primary fibromyalgia syndrome. Journal of Rheumatology 20: 469-474. Griep EN, Boersma JW & de Kloet ER (1994) Pituitary release of growth hormone and prolactin in the primary fibromyalgia syndrome. Journal of Rheumatology 21: 2125-2130. Haas DA & George SR (1987) Neuropeptide Y administration acutely increases hypothalamic corticotropin-releasing factor immunoreactivity: lack of effect in other rat brain regions. Life Sciences 41: 2725-2731. Holmes MC, DiRenzy G, Gillham B & Jones MT (1982) Role of serotonin in the control of secretion of corticotrophin releasing factor. Journal of Endocrinology 93" 151-160. Hrycaj P, Stratz T & Muller W (1993) Platelet 3H-imipramine uptake receptor density and serum serotonin in patients with fibromyalgia/fibrositis syndrome. Journal of Rheumatology 20: 1986-1987. Hudson JI, Goldenberg DL, Pope HG et al (1992) Comorbidity of fibromyalgia with medical and psychiatric disorders. American Journal of Medicine 92: 363-367. Jones MT, Hillhouse EW & Burden J (1976) Effect of various putative neurotransmitters on the secretion of cortic0trophin-releasing hormone from the rat hypothalamus in vitro: a model of the neurotransmitters involved. Journal of Endocrinology 69: 1-10. Kamilaris TC, Calogero AE & Johnson EO (1989) Effect of neuropeptide Y on hypothalamicpituitary-adrenal .tunction in the rat. 19th Annual Meeting of the Society for Neuroscience, Phoenix, AZ.
NEUROHORMONAL PERTURBATIONS IN FM
377
Koenig JI (1990) Regulation of the hypothalamo-pituitary axis by neuropeptide Y. Annals of the New York Academy of Sciences 611:317-328. Krieger DT & Rizzo F (1969) Serotonin mediation of circadian periodicity of plasma 17-hydroxycorticosteroids. American Journal of Physiology 217: 1703-1707. Lundberg JM, Franco-Cereceda A, Lacroix J-S & Pernow J (1990) Neuropeptide Y and sympathetic neurotransmission. Annals of the New York Academy of Sciences 611: 166-174. McCain GA (1989) Nonmedicinal treatments in primary fibromyalgia. Rheumatic Disease Clinics of North America 15: 73-89. McCain GA & Tilbe KS (1989) Diurnal hormone variation in fibromyalgia syndrome: A comparison with rheumatoid arthritis. Journal of Rheumatology 16 (supplement 19): 154-157. Moldofsky H & Warsh JJ (1978) Plasma tryptophan and musculoskeletal pain in non-articular rheumatism ('fibrositis syndrome'). Pain 5: 65-71. Neeck G & Riedel W (1992) Thyroid function in patients with fibromyalgia syndrome. Journal of Rheumatology 19:1120-1122. Nikolarakis KE, Almeida OFX & Herz A (1986) Stimulation of hypothalamic I]-endorphin and dynorphin release by corticotropin-releasing factor. Brain Research 399: 152-155. Qiao Z-G, Vaerc~yH & M0rkrid L (1991) Electrodermal and microcircalatory activity in patients with fibromyalgia during baseline, acoustic stimulation and cold pressor tests. Journal of Rheumatology 18: 1383-1389. Rowe PC, Bou-Holaigah I, Kan JS & Calkins H (1995) Is neurally mediated hypotension an unrecognised cause of chronic fatigue? Lancet 345: 623-624. Russell IJ (1989) Neurohormonal aspects of fibromyalgia syndrome. Rheumatic Disease Clinics of North America 15: 149-168. Russell IJ (1995) Neurohormonal: Abnormal laboratory findings related to pain and fatigue in fibromyalgia. Journal of Musculoskeletal Pain 3: 59-65. Russell IJ, Michalek JE & Vaprio GA (1989) Serum amino acids in fibrositis/fibromyalgia syndrome. Journal of Rheumatology 16 (supplement 19): 158-163. Russell IJ, Bowden CL & Michalek J (1992) Platelet 3H-imipramine uptake receptor density and serum serotonin levels in patients with fibromyalgia/fibrositis syndrome. Journal of Rheumatology 19: 104-109. Scapagnini U, Moberg GP, Van Loon GR et al (1971) Relation of brain 5-hydroxytryptamine content to the diurnal variation of plasma corticosterone in the rat. Neuroendocrinology 7: 90-96. Sharpe M (1995) Cognitive behavior therapy and the treatment of chronic fatigue syndrome. Journal of Musculoskeletal Pain 3: 141-147. Sternberg EM (1993) Hypoimmune fatigue syndromes: Diseases of the stress response? Journal of Rheumatology 20: 418-421. Streeten DHP & Anderson GH (1992) Delayed orthostatic intolerance. Archives oflnternal Medicine 152: 1066-1072. Sutton RE, Koob GF, LeMoal M e t al (1982) Corticotropin-releasing factor produces behavioral activation in rats. Nature 297: 331-333. Swerdlow NR, Geyer MA, Vale WW & Koob GF (1982) Corticotropin-releasing factor potentiates acoustic startle in rats: Blockade by chlordiazepoxide. Psychopharmacology (Berlin) 88: 147-152. VaerCy H, Helle R, FCrre O et al (1988) Cerebrospinal fluid levels of 13-endorphin in patients with fibromyalgia (fibrositis syndrome). Journal of Rheumatology 15:1804-1806. VaerCy H, Qiao Z-G, M~rkrid L & Fc~rre 0 (1989) Altered sympathetic nervous system response in patients with fibromyalgia (fibrositis syndrome). Journal of Rheumatology 16: 1460-1465. Vamvakopoulos NC & Chrousos GP (1993) Evidence of direct estrogenic regulation of human corticotropin-releasing hormone gene expression. Journal of Clinical Investigation 92: 18961902. van Denderen JC, Boersma JW, Zeinstra Pet al (1992) Physiological effects of exhaustive physical exercise in primary fibromyalgia syndrome (PFS): Is PFS a disorder of neuroendoorine reactivity? Scandinavian Journal of Rheumatology 21: 35-37. Wasman J & Zatzkis SM (1986) Fibromyalgia and menopause: Examination of the relationship. Postgraduate Medicine 80: 165-171. Wolfe F, Smythe HA, Yunus MB et al (1990) The American College of Rheumatology 1990 criteria for the classification of fibromyalgia. Arthritis and Rheumatism 33: 160-172. Yehuda R & Meyer 1S (1984) A role for serotonin in the hypothalamic-pituitary-adrenal response to insulin stress. Neuroendocrinology 38: 25-32.
378
L . J . CROFFORD ET AL
Young EA (1996) The role of gonadal steroids in hypothalamic-pituitary-adrenal axis regulation. Critical Reviews in Neurobiology 9: 371-381. Young EA, Kotun J, Haskett RF et al (1993) Dissociation between pituitary and adrenal suppression to dexamethasone in depression. Archives of General Psychiatry 50: 395-403. Yunus MB, Masi AT, Calabro JJ et al (1981) Primary fibromyalgia (fibrositis): clinical study of 50 patients with matched normal controls. Seminars in Arthritis and Rheumatism 11: 1 5 1 - 1 7 1 . Yunus MB, Dailey JW, Aldag JC et al (1992) Plasma and urinary catecholamines in primary fibromyalgia: A controlled study. Journal of Rheumatology 19" 95-97.
Fibromyalgia (FM) is a poorly understood syndrome characterized by chronic, widespread musculoskeletal pain and the presence of tender points on physical examination (Wolfe et al, 1990). Other clinical manifestations include fatigue, sleep disturbance, headache, and neuropsychological complaints (Yunus et al, 1981). The concept that disorders such as FM might by associated with 'subtle and undetectable' disturbances in the central nervous system was introduced as early as 1869 by Beard in his description of neurasthenia, a syndrome characterized by marked somatic and constitutional complaints, but few physical findings (Beard, 1869). Although the aetiology and pathogenesis of FM remain elusive, recent years have seen great progress toward defining neurochemical abnormalities in FM. A clue to the pathophysiology in FM may be found in the clinical observation that patients with FM frequently report the onset of their illness following a significant physical or psychological stress (Crofford et al, 1994; Demitrack and Crofford, 1995). In addition, patients with FM report that the course of their syndrome is significantly influenced by stress (Yunus et al, 1981). Finally, patients with FM report a high level of perceived daily stress (Dailey et al, 1990). For these reasons FM falls into the spectrum of what might be termed 'stress-related illness'. Other clinical syndromes, including chronic fatigue syndrome (CFS), irritable bowel syndrome, headache syndromes, irritable bladder (female urethral syndrome, interstitial cystitis), temporomandibular joint syndrome, and dysmenorrhoeic syndromes, share substantial symptomatic overlap and cooccur with FM (Hudson et al, 1992). These syndromes are also associated with high levels of stress (Buchwald et al, 1987; Goldenberg et al, 1990; Hudson et al, 1992). It is well known that there is a stereotypic response to endogenous or exogenous stress, mediated primarily through activation of the hypothalamic-pituitary-adrenal (HPA) axis and the sympathetic nervous system (Chrousos and Gold, 1992). These physiological responses are designed to respond to any internal or external disturbance to homeostatic equilibrium, Bailli~re's Clinical Rheumatology-365 Vol. 10, No. 2, May 1996 1SBN 0-7020--2182-2 0950-3579/96/020365 + 14 $12,00/00
Copyright 9 1996, by Baillirre Tindall All rights of reproduction in any form reserved
366
L . J . CROFFORD ET AL
which is the way we will define stress. A stressor may be metabolic, physiological, traumatic, inflammatory/infectious, psychological, or emotional in nature. Patients with FM were shown to have abnormalities of both HPA axis and sympathetic stress-response systems (Griep et al, 1993; Crofford et al, 1994). It is important to emphasize, however, that the primary stress-response systems share important connections with other neuroendocrine and neurochemical systems that may also be disturbed in patients with FM (Chrousos and Gold, 1992). In this chapter, we will review the HPA axis, sympathetic nervous system, and other neuroendocrine systems in patients with FM and related disorders. We will also present a coherent framework to link these biological findings to susceptibility and clinical manifestations of FM. THE HPA AXIS Consistent with the concept that FM is a stress-related illness, we and others have demonstrated perturbations of the HPA axis (Table 1). The first abnormality described was a low 24-hour urine-free cortisol compared with either normal subjects or patients with rheumatoid arthritis (McCain and Tilbe, 1989; Crofford et al, 1994). FM patients, however, exhibited normal morning (peak) and elevated evening (trough) cortisol levels resulting in a loss of the normal diurnal cortisol fluctuation (McCain and Tilbe, 1989; Crofford et al, 1994). Abnormal cortisol measurements were most prominent in patients with a longer duration (> 2 years) of disease (McCain and Tilbe, 1989). No differences in levels of cortisol binding globulin have been detected in FM patients (Crofford et al, 1994). Potential reasons for these appparently disparate findings become clearer when one considers the detailed circadian dynamics of the HPA axis. For example, there are normally eight to nine peaks of cortisol over a 24-hour period, presumably in response to basal surges of corticotrophin-releasing hormone (CRH) and adrenocorticotrophic hormone (ACTH). The majority of these cortisol surges occur during the night-time circadian peak. In FM, the amplitude of the cortisol peaks might be normal or even elevated, but the frequency of the peaks could be decreased resulting in a low cortisol output when examined over 24 hours. Further study of the basal circadian HPA axis dynamic in FM is clearly indicated. Table 1. HPA axis perturbations in FM. Basal Decreased 24-hour urine-free cortisol Normal peak, elevated trough total and free plasma cortisol levels Blunted circadian change in plasma cortisol levels Normal plasma cortisol binding globulin levels Stress Enhanced ACTH release after exogenous CRH and insulin-induced hypoglycaemia Blunted cortisol response after exogenous CRH and insulin-induced hypoglycaemia, even in the face of exaggerated ACTH release Blunted cortisol response to exercise
NEUROHORMONAL PERTURBATIONS IN FM
367
The HPA axis response to administration of exogenous CRH has been used as an index of the stress-induced activation of the axis, whereas insulin-induced hypoglycaemia tests the integrity of the stress-associated increase in endogenous hypothalamic CRH as a stimulus to pituitary-adrenal hormone secretion. CRH-stimulation tests performed in the evening, when FM patients exhibit elevated basal trough cortisol levels compared with control patients, revealed no statistically significant differences in CRH-stimulated ACTH levels between patients with FM and controls. However, a non-significant trend toward elevated mean basal and CRH-stimulated ACTH levels was noted (Crofford et al, 1994). Griep et al (1993) performed stimulation testing of HPA axis function using exogenous CRH and insulin-induced hypoglycaemia. They demonstrated significantly elevated ACTH responses to both stimuli in patients with FM. Their studies were performed in the morning at the circadian peak, when there are no differences in basal cortisol levels compared with controls. Elevated trough cortisol levels in patients with FM as compared with normal subjects may have blunted the effect of CRH on ACTH in the evening by negative feedback mechanisms. Both studies demonstrated a relatively blunted adrenal cortisol response to stimulated ACTH release (Griep et al, 1993; Crofford et al, 1994). Further evidence for perturbed HPA axis function is demonstrated by significantly lower cortisol levels after exercise in FM patients as compared to control subjects (van Denderen et al, 1992). The reasons for this response are unclear, but could result from a relatively hyporesponsive adrenal gland, could be due to intrinsic hypoactivity of the adrenal cortex itself, or could develop in the context of chronic understimulation due to deficiencies of CRH or ACTH (Sternberg, 1993). HPA axis function has also been examined in patients with the related chronic fatigue syndrome (CFS) (Demitrack et al, 1991). Comparison of HPA axis function in FM with patients with CFS reveals differences despite the significant clinical overlap between these patients. Low 24-hour urinefree cortisol excretion, similar to that seen in patients with FM, was demonstrated in patients with CFS (Demitrack et al, 1991). However, in contrast to FM, in CFS, evening plasma ACTH levels were elevated and cortisol levels were reduced compared to control subjects (Demitrack et al, 1991). CRH-induced stimulation of pituitary-adrenal secretion revealed an attenuated, rather than an exaggerated ACTH response in CFS patients. Additionally, stimulation of the adrenal gland revealed low maximal cortisol responses to ACTH, compatible with secondary adrenal atrophy. We postulated that these data suggested central CRH insufficiency in CFS patients. While preliminary studies performed by Beam et al (1995) failed to confirm changes in ACTH or cortisol in response to hypoglycaemia or serotonergic stimulation with d-fenfluramine in the morning in patients with CFS, further studies by the same group demonstrated decreased plasma cortisol responses to d-fenfluramine in CFS patients compared with normal and depressed controls (Cleare et al, 1995). Although there are differences in dynamic function of the HPA axis in FM and CFS patients, both patient groups share low 24-hour urine cortisol levels. The possibility exists that differences in provocative tests between
368
L.J.
CROFFORD
ET A L
FM and CFS could be explained on the basis of differences in other ACTH secretagogues. For example, these syndromes could represent different forms of insufficient stimulation of the HPA axis, with both syndromes expressing low hypothalamic CRH, but FM characterized by increased exposure of the corticotrophs to arginine vasopressin (AVP) and CFS patients with decreased AVP levels. In fact, there is preliminary experimental data that support this view. Our group detected a trend towards increased release of AVP after postural change in patients with FM (Crofford et al, 1994). Five of 12 FM patients achieved AVP levels more than two standard deviations above the highest control value. In contrast, Bakheit et al (1993) evaluated a group of patients with postviral fatigue syndrome and demonstrated significantly lower basal AVP levels. CFS patients also demonstrated a lack of correlation between serum and urine osmolality with plasma AVR These data, however, should be interpreted with caution since plasma AVP reflects release of AVP from the magnocellular nucleus of the hypothalamus, not the parvocellular nucleus that releases AVP to the hypophyseal venous circulation. THE SYMPATHETIC NERVOUS SYSTEM The HPA axis and sympathetic nervous system are intimately linked as the predominant mediators of a co-ordinated stress-response system (Chrousos and Gold, 1992). Functional studies of sympathetic nervous system function in FM, as measured by changes in skin microcirculation, suggest dysfunction of the adrenergic component of peripheral sympathetic activity (Vaer0y et al, 1989; Qiao et al, 1991). In studies of muscle sympathetic activity, there was no difference between FM patients and controls at rest. However, after stimulation of muscle sympathetic activity by static hand grip, contraction of the jaw muscle, or mental stress, there was a lack of exaggerated sympathetic activity and, in fact, a tendency to lower muscle sympathetic activity (Elam et al, 1992). The observation that some patients with chronic fatigue syndromes may have identifiable neurally-mediated hypotension (Streeten and Anderson, 1992; Rowe et al, 1995) has prompted an evaluation of autonomic dysfunction as measured by tilt-table testing in patients with FM. Clauw et al (1995a) demonstrated that patients with FM, who were not chosen on the basis of symptomatology, displayed significantly smaller pulse pressures in the upright position and higher blood pressures on return to the supine position. Further studies will be needed to clarify the utility of functional tests of autonomic function in patients with both FM and CFS. Early studies of sympathetic nervous system hormones in FM patients either failed to show difference in plasma or urine catecholamine levels (Yunus et al, 1992), or demonstrated a subgroup of patients with elevated urinary norepinephrine levels (Russell, 1989). In CFS, patients were shown to have a significant reduction in basal plasma levels of the monoamine metabolite 3-methoxy-4-hydroxyphenylglycol (Demitrack et al, 1992). Neuropeptide Y (NPY) co-localizes with norepinephrine in the sympathetic
N E U R O H O R M O N A L PERTURBATIONS IN FM
369
nervous system, and plasma NPY can serve as a correlate of sympathoadrenal output (Lundberg et al, 1990). In human subjects, elevated plasma NPY levels are seen with heavy physical exercise or other situations of strong sympathetic activation (Lundberg et al, 1990). Our group demonstrated that patients with FM had significantly lower plasma NPY levels than matched control subjects (Crofford et al, 1994). These findings have been confirmed by Clauw et al (1995b) who demonstrated significantly lower basal NPY levels in FM patients, who also met criteria for CFS, compared with controls. While reduced NPY levels could result from a decrease in the level of physical activity in these patient groups, this is unlikely to be the sole explanation since the study by Clauw et al was performed on patients that were ambulatory and performing regular exercise. Centrally, NPY increases the concentration of CRH in the hypothalamus, and is present in high concentrations in the portal-hypophyseal venous circulation where it acts synergistically with CRH to stimulate release of ACTH (Haas and George, 1987; Koenig, 1990). NPY has also been reported to have ACTH-like activities at the level of the adrenal cortex (Kamilaris et al, 1989). These findings provide mechanisms by which NPY interacts with HPA axis function. The significance of low plasma NPY levels in patients with FM is unclear. Nevertheless, low NPY levels may represent another measure of hypofunction or depletion of the sympathetic stress axis. The relationship between autonomic dysfunction and neuroendocrine abnormalities in FM has not yet been fully defined. SEROTONIN
Serotonin influences the circadian fluctuation of HPA axis products (Krieger and Rizzo, 1969; Scapagnini et al, 1971). Numerous studies have shown that serotonin and serotonin agonists stimulate the pituitary-adrenal system, probably through stimulation of CRH release from the hypothalamus, and that density of serotonin receptors and serotonin levels parallel HPA axis activity (Jones et al, 1976; Buckingham and Hodges, 1979; Holmes et al, 1982; Burnett et al, 1992). There is continuing controversy about the role of serotonin during stress-induced stimulation of the HPA axis. However, serotonin enhances the HPA axis response during insulin-induced hypoglycaemia, perhaps due to a direct effect of insulin on the availability of tryptophan to the central nervous system (Yehuda and Meyer, 1984). It is not yet known whether abnormalities in serotonin pathways could result in the HPA axis perturbations as observed in FM patients. The metabolism of serotonin and its precursor, tryptophan, are abnormal in FM patients (Russell, 1989). The serum concentration of serotonin is lower in patients with FM than matched controls, and patients with FM have an increased number of serotonin re-uptake receptors on platelets (Russell et al, 1992). There is an inverse correlation between the level of free serum tryptophan, a precursor of serotonin, and severity of pain in FM patients, although tryptophan levels were in the normal range (Moldofsky and Warsh, 1978). In CFS, we have demonstrated significant increases in
370
L. J. C R O F F O R D
ET A L
plasma levels of the serotonin metabolite, 5-hydroxyindoleacetic acid (5HIAA) (Demitrack et al, 1992). In addition, serotonin neurotransmission, as assessed by prolactin responses to the serotonin receptor agonist, buspirone, is increased in CFS relative to healthy and depressed control groups (Bakheit et al, 1992). Taken together with the d-fenfluramine challenge abnormalities in CFS, these studies suggests that serotonergic systems may be disturbed in CFS patients as well (Cleare et al, 1995).
THE GROWTH HORMONE AXIS Patients with FM have been noted to have low basal levels of growth hormone (Griep et al, 1994) and insulin-like growth factor-1 (IGF-1 or somatomedin C), which reflects integrated growth hormone stimulation (Bennett et al, 1992). Analogous to the exaggerated release of ACTH after insulin-induced hypoglycaemic stress, exaggerated release of growth hormone is seen after the same stimulus in patients with FM (Griep et al, 1994). Bennett et al (1995a) reported blunted growth hormone secretion from the pituitary after administration of clonidine. This dichotomy suggests that growth hormone axis hypofunction most probably occurs at the level of the hypothalamus or other higher brain regions. The causes and consequences of these changes in growth hormone physiology are not completely understood. However, since secretion of growth hormone is maximum during sleep, it has been proposed that the sleep disturbance may be responsible f o r perturbations in the growth hormone axis in FM. It should also be noted that HPA axis hormones have profound effects on the growth hormone axis, through direct effects of CRH to stimulate somatostatin secretion, and suppressive effects of glucocorticoids on secretion of growth hormone itself from the pituitary (Chrousos and Gold, 1992). Mechanisms by which both cortisol and growth hormone/IGF-1 are low in patients with FM remain to be determined.
THE THYROID HORMONE AXIS Symptoms of FM are remarkably similar to patients with hypothyroidism (Becker et al, 1963; Carette and Lefran~ois, 1988). Although basal levels of thyroid hormones are normal, two independent groups have reported blunted thyroid-stimulating hormone (TSH), but exaggerated prolactin response to administration of thyroid hormone releasing hormone (Ferraccioli et al, 1990; Neeck and Riedel, 1992). Similar to the growth hormone axis, CRH stimulates somatostatin, which exerts inhibitory effects on the release of TSH from the pituitary. In addition, glucocorticoids antagonize release of TSH and actions of thyroid hormone in the periphery (Chrousos and Gold, 1992). Again, further studies of the interactions between the HPA and thyroid axis hormones in FM are needed.
NEUROHORMONAL
P E R T U R B A T I O N S IN FM
371
THE H P G AXIS
The incidence of FM is dramatically higher in women, and the onset of disease often correlates with the onset of menopause (Russell, 1989). This has prompted the hypothesis that changes in levels of the gonadal steroids may modulate the development of FM (Wasman and Zatzkis, 1986). Carette et al (1992) compared levels of HPG axis hormones in 10 patients with FM compared with normal control subjects. They were unable to demonstrate significant differences in oestrogens, androgens, or gonadotrophins, with the exception that testosterone levels were decreased in FM patients. Nevertheless, gonadal steroids play a role in the regulation of the HPA axis. There is in vitro evidence that oestrogen directly increases CRH expression (Vamvakopoulos and Chrousos, 1993). Studies have also demonstrated an effect of gonadal steroids and glucocorticoid negative feedback, with oestrogen leading to impaired shut-off of ACTH and corticosterone secretion following stress (Young, 1996). In general, both oestrogen and progesterone appear to antagonize glucocorticoid negative feedback, but some effects of oestrogen may be reversed by progesterone (Young, 1996). Hormonal status may be important for development of HPA axis perturbation. For example, virtually all post-menopausal women develop HPA axis dysregulation during depression, while pre-menopausal women are relatively resistant to changes in HPA axis function in depressive states (Young et al, 1993). Finally, it should be noted that HPA axis hormones exert reciprocal effect on the reproductive hormone axis, leading to decreased overall activity of the axis (Chrousos and Gold, 1992). N E U R O H O R M O N A L SYSTEMS AND THE AETIOPATHOGENESIS OF FM
The observed neuroendocrine abnormalities in FM patients could be a contributing cause to the aetiology or pathogenesis of FM, a consequence of the illness, or an association with other factors causal for FM. In support of the hypothesis that neuroendocrine dysfunction may be important to susceptibility to FM or to clinical features of the syndrome (Figure 1), it is important to note that the symptoms of FM are similar to many neuroendocrine deficiency states. For example, it is of considerable interest to note the clinical consequences of inadequate HPA axis activation. Subtle grades of HPA axis insufficiency may occur due to abnormalities at the level of the hypothalamus or higher central nervous system centres, the pituitary, the adrenal, or even at the level of target tissue responsiveness to cortisol. By way of example, cortisol resistance due to a genetic abnormality of the glucocorticoid receptor was reported to present with fatigue as the only symptom (Br6nnegard et al, 1986). It has been suggested that a number of syndromes that share clinical features with FM, including CFS, the depressed phase of seasonal affective disorder or other atypical depressive syndromes, and hypothyroidism may share a functional deficit of hypothalamic CRH (Chrousos and Gold, 1992; Sternberg, 1993).
372
L. J, CROFFORD ET AL
Vulnerability to Stress Disorders 9 Genetic 9 Hormonal 9 Environmental
?Mechanism: 9SNS contribution 9 SerotonergicJ system ~F" input
Syndrome Triggers 9 Trauma 9 Infection 9 Emotional stress
Perpetuating factors: Physical deconditioning
Psychological response to illness
9 Stress
Adrenal FM:
~CRH
"I'AV___PP CFS: ~,AVP
FM: 'I~ACTH
Cortisol
CFS:~ACTH
" Central Sequelae 9 Sleep disorder 9 Cognitive dysfunction ? Fatigue .i
eripheral Sequela= 9 Pain syndromes
Figure 1. Neurohormonal hypothesis of the pathogenesis of flbromyalgia (FM). Potential relationships between the HPA axis and aetiopathogenic mechanisms of disease in FM. We hypothesize that low levels of CRH and cortisol may contribute to both central and peripheral disease manifestations, and that factors which perpetuate disease may be integrated through HPA axis activity.
CRH serves not only as a principal stimulus to the HPA axis, but also as a behaviourally active neurohormone whose central administration to animals and non-human primates induces signs of physiological and behavioural arousal including activation of the sympathetic nervous system, hyper-responsiveness to sensory stimuli, and increased locomotion (Sutton et al, 1982; Swerdlow et al, 1982). Hence, a relative or absolute deficiency of hypothalamic CRH could contribute to the profound lethargy and fatigue that are characteristic of these syndromes (Chrousos and Gold, 1992; Stemberg, 1993). Although this is an attractive model, it should be emphasized that, to date, the assessment of CRH activity in these patient
NEUROHORMONAL
P E R T U R B A T I O N S IN FM
373
groups is inferential, based primarily on peripheral pituitary-adrenal responses to hormonal challenge. Even direct measurement of CRH in the cerebrospinal fluid may merely reflect cortical or spinal sources of CRH, and not functional CRH activity in the paraventricular nucleus (PVN). Moreover, since the behavioural effects of CRH are produced at disparate sites within the central nervous system, there is no reason to presume that a functional deficit of CRH in the PVN is associated with similar reductions of CRH activity in limbic or cortical locations. The musculoskeletal pain that is the dominant clinical feature of FM could also be influenced by HPA axis abnormalities. Myalgia, arthralgia, and muscle weakness are features of steroid withdrawal syndromes, and FM had been documented after hypophysectomy for Cushing's disease (Disdier et al, 1991). Both CRH and AVP neurons, as well as sympathetic neurons, have connections to brain pro-opiomelanocortin containing neurons that produce opioids (Nikolarakis et al, 1986). Descending opioid peptidergic pathways may be important in influencing the biochemical determinants of pain at brain stem and spinal cord levels, such as substance R which is elevated in patients with FM (Vaer0y et al, 1988; Russell, 1995). Other neuroendocrine or neurochemical systems may influence HPA axis function in FM. The high prevalence of FM in women, as well as the observed increase in incidence at the time of menopause, could be the result of sexual dimorphism in the development of stress-response systems and/or the influence of gonadal steroids on stress-response systems. For example, oestrogen impairs glucocorticoid negative feedback at the level of the hippocampus leading to a functional increase of HPA axis reactivity. Decreased oestrogenic stimulation at the time of menopause could be permissive for relative hypofunction of the HPA axis, contributing to the development of FM or increased severity of FM symptoms. The role of sympathetic nervous abnormalities as a contributing factor to the pathogenesis and persistence of symptoms in FM and related syndromes, such as CFS, remains to be fully elucidated. Ongoing studies will assess autonomic function in larger groups of patients with FM and other fatigue states, and future controlled treatment trials should be performed in patients demonstrating abnormalities (Streeten and Anderson, 1992; van Denderen et al, 1992; Clauw et al, 1995a; Rowe et al, 1995). Abnormalities in the tryptophan/serotonin metabolic pathway were among the first neurochemical abnormalities to be demonstrated in patients with FM (Russell et al, 1989; Russell et al, 1992; Hrycaj et al, 1993; Russell, 1995). The HPA axis abnormalities observed in patients with FM could be related to alterations in serotonin receptor activation (Russell, 1995). Serotonin abnormalities could be related to the sleep disturbance seen in FM patients, and which may contribute to other symptoms of pain and fatigue. In addition, serotonin may contribute to aberrant pain perception in FM, via interactions with substance P at the level of the spinal cord (Russell, 1995). Finally, the growth hormone-IGF-1 axis is an important anabolic stimulus for musculoskeletal tissues (Bennett et al, 1992). Hence, reductions in
374
L.J. CROFFORD ET AL
circulating levels of these substances, as seen in FM, may be related to the diffuse myalgias which are characteristic of FM. IMPLICATIONS OF NEUROENDOCRINE PERTURBATIONS FOR FM TREATMENT STRATEGIES
One of the most compelling reasons for considering HPA axis and sympathetic system abnormalities as the key to understanding FM, is that many therapeutic strategies useful in FM either modulate patient perceptions of stressors (cognitive behavioural therapy) or directly influence function of stress-response systems (aerobic exercise, tricyclic antidepressants) (Bradley, 1989; McCain, 1989; Carette, 1995; Sharpe, 1995). These observations support the hypothesis that central determinants of HPA axis function, not primary adrenal insufficiency, are operative in FM. This is particularly true since cortisol replacement has never been shown to be useful as a treatment (Clark et al, 1985). Further investigation into the nature of the HPA axis abnormalities and the role of other central nervous system hormones in modulating HPA axis function in patients with FM may lead to improved treatment strategies. Direct testing of the hypothesis that low growth hormone and IGF-1 levels play a role in the symptoms of FM are currently underway. Bennett and co-workers have undertaken a double-blind placebo controlled trial of growth hormone replacement therapy in FM patients with low IGF-1 levels (Bennett et al, 1995). There was significant improvement within the group of patients treated with growth hormone over a 9-month period, and growth hormone resulted in significant improvement compared with placebo (Bennett et al, 1995). Further studies comparing growth hormone with other treatment modalities may clarify its role in the overall treatment of patients with FM. A therapeutic goal in patients with FM may be to restore physiological neuroendocrine and neurochemical function. Since it is likely that the FM syndrome represents a final common pathway for a number of aetiopathogenic mechanisms, therapy may need to be directed towards a number of different central nervous system pathways. SUMMARY
Fibromyalgia (FM) falls into the spectrum of what might be termed 'stressassociated syndromes' by virtue of frequent onset after acute or chronic stressors and apparent exacerbation of symptoms during periods of physical or emotional stress. Patients with FM exhibit disturbances of the major stress-response systems, the HPA axis and the sympathetic nervous system. Integrated basal cortisol levels measured by 24-hour urine-free cortisol are low. FM patients display a unique pattern of HPA axis perturbation characterized by exaggerated ACTH response to exogenous CRH or to endogenous activators of CRH such as insulin-induced hypo-
NEUROHORMONAL PERTURBATIONS IN FM
375
glycaemia. The cortisol response to increased ACTH in these stress paradigms is blunted, as is the cortisol response to exercise. Functional analysis suggests that FM patients may also exhibit disturbed autonomic system activity. For example, plasma NPY, a peptide co-localized with norepinephrine in the sympathetic nervous system, is low in patients with FM. Abnormalities of related neuronal systems, particularly decreased serotonergic activity, may contribute to the observed neuroendocrine perturbations in FM. Finally, other neuroendocrine systems, including the growth hormone axis, are also abnormal in FM patients. Many clinical features of FM and related disorders, such as widespread pain and fatigue, could be related to the observed neuroendocrine perturbations. This hypothesis is supported by the observation that many useful treatments for FM affect the function of these central nervous system centres. Further clarification of the role of neuroendocrine abnormalities in patients with FM, and the relationship of these disturbances with particular symptoms, may lead to improved therapeutic strategies.
REFERENCES Bakheit AMO, Behan PO, Dinan TG et al (1992) Possible upregulation of hypothalamic 5-hydroxytryptamine receptors in patients with postviral fatigue syndrome. British Medical Journal 304: 1010-1012. Bakheit AMO, Behan PO, Watson WA & Morton JJ (1993) Abnormal arginine-vasopressin secretion and water metabolism in patients with postviral fatigue syndrome. Acta Neurologica Scandinaviea 87: 234-238. Beard G (1869) Neurasthenia, or nervous exhaustion. Boston Medical and Surgical Journal III: 217-221. Beam J, Allain T, Coskeran Pet al (1995) Nearoendocrine responses to d-fenfluramine and insulininduced hypoglycemia in chronic fatigue syndrome. Biological Psychiatry 37: 245-252. Becker KL, Ferguson RH & McConahey WM (1963) The connective tissue disease and symptoms associated with Hashimoto's thyroiditis. New England Journal of Medicine 268: 275-280. Bennett RM, Clark SR, Burckhardt CS & Cook D (1995a) 1GF-1 assays and other GH tests in 500 fibromyalgia patients. Journal of Musculoskeletal Pain 3 (supplement 1): 109(abstract). Bennett RM, Clark SR, Burckhardt CS & Walczyk J (1995b) A double blind placebo controlled study of growth hormone therapy in fibromyalgia. Journal of Musculoskeletal Pain 3 (supplement 1): 110(abstrac0. Bennett RM, Clark SR, Campbell SM & Burckhardt CS (1992) Low levels of somatomedin C in patients with the fibromyalgia syndrome. Arthritis and Rheumatism 35:1113-1116. Bradley LA (1989) Cognitive-behavioral therapy for primary fibromyalgia. Journal of Rheumatology 16 (supplement 19): 131-136. Br6nnegard M, Wemer S & Gustafsson J-A (1986) Primary cortisol resistance associated with a thermolabile glucoeorticoid receptor in a patient with fatigue as the only symptom. Journal of Clinical Investigation 78: 1270-1278. Buchwald D, Goldenberg DL, Sullivan JL & KomaroffAL (1987) The 'chronic active Epstein-Barr virus infection' syndrome and primary fibromyalgia. Arthritis and Rheumatism 30:1132-1136. Buckingham C & Hodges JR (1979) Hypothalamic receptors influencing the secretion of corticotrophin releasing hormone in the rat. Journal of Physiology 290:421-431. Bumett PWJ, Mefford IN, Smith CC et al (1992) Hippocampal 8-[3H]hydroxy-2-(di-n-propylamino)tetralin binding site densities, serotonin receptor (5HT1A) messenger ribonucleic acid abundance, and serotonin levels parallel the activity of the hypothalamopituitary-adrenal axis in rat. Journal of Neurochemistry 59: 1062-1070. Carette S (1995) What have clinical trials taught us about the treatment of fibromyalgia? Journal of Masculoskeleml Pain 3: 133-140.
376
L . J . CROFFORD ET AL
Carette S & Lefrangois L (1988) Fibrositis and primary hypothyroidism. Journal of Rheumatology 15: 1418-1421. Carette S, Dessureault M & Btlanger A (1992) Fibromyalgia and sex hormones. Journal of Rheumatology 19:831 (letter). Chrousos GP & Gold PW (1992) The concepts of stress and stress system disorders: Overview of physical and behavioral homeostasis. Journal of the American Medical Association 267: 1244-1252. Clark S, Tindall E & Bennett RM (1985) A double blind crossover trial of prednisone versus placebo in the treatment of fibrositis. Journal of Rheumatology 12- 980-983. Clauw DJ, Radulovic D, Katz P e t al (1995a) Tilt table testing as a measure of dysantonomia in fibromyalgia. Journal of Musculoskeletal Pain 3 (supplement 1): 10 (abstract). Clauw DJ, Sabol M, Radulovic D et al (1995b) Serum neuropeptides in patients with both fibromyalgia (FM) and chronic fatigue syndrome (CFS). Journal of Musculoskeletal Pain 3 (supplement 1): 79 (abstract). Cleare AJ, Beam J, McGregor Aet al (1995) Contrasting neuroendocrine responses in depression and chronic fatigue syndrome. Journal of Affective Disorders 35: 283-289. Crofford LJ, Pillemer SR, Kalogeras KT et al (1994) Hypothalamic-pituitary-adrenal axis perturbations in patients with fibromyalgia. Arthritis and Rheumatism 37: 1583-1592. Dailey PA, Bishop GD, Russell IJ & Fletcher EM (1990) Psychological stress and the fibrositis/ fibromyalgia syndrome. Journal of RheumatoIogy 17: 1380-1385. Demitrack MA & Crofford LJ (1995) Hypothalmic-pituitary-adrenal axis dysregulation in fibromyalgia and chronic fatigue syndrome: an overview and hypothesis. Journal of Musculoskeletal Pain 3: 67-73. Demitrack MA, Dale JK, Straus S E e t al (1991) Evidence for impaired activation of the hypothalamic-pituitary-adrenal axis in patients with chronic fatigue syndrome. Journal of Clinical Endocrinology and Metabolism 73: 1224-1234. Demitrack MA, Gold PW, Dale JK et al (1992) Plasma and cerebrospinal fluid monoamine metabolism in patients with chronic fatigue syndrome: Preliminary findings. Biological Psychiatry 32: 1065-1077. Disdier P, Harle J-R, Brue T et al (1991) Severe fibromyalgia after hypophysectomy for Cushing's disease. Arthritis and Rheumatism 34: 493-495. Elam M, Johansson G & WaUin BG (1992) Do patients with primary fibromyalgia have an altered muscle sympathetic nerve activity? Pain 48: 371-375. Ferraecioli G, Cavalieri F, Salaffi F et al (1990) Neuroendocrinologic findings in primary fibromyalgia (soft tissue chronic pain syndrome) and in other chronic rheumatic conditions (rheumatoid arthritis, low back pain). Journal of Rheumatology 19: 869-873. Goldenberg DL, Simms RW, Geiger A & Komaroff AL (1990) High frequency of fibromyalgia in patients with chronic fatigue seen in a primary care practice. Arthritis and Rheumatism 33: 381-387. Griep EN, Boersma JW & de Kloet EP (1993) Altered reactivity of the hypothalamic-pituitatyadrenal axis in the primary fibromyalgia syndrome. Journal of Rheumatology 20: 469-474. Griep EN, Boersma JW & de Kloet ER (1994) Pituitary release of growth hormone and prolactin in the primary fibromyalgia syndrome. Journal of Rheumatology 21: 2125-2130. Haas DA & George SR (1987) Neuropeptide Y administration acutely increases hypothalamic corticotropin-releasing factor immunoreactivity: lack of effect in other rat brain regions. Life Sciences 41: 2725-2731. Holmes MC, DiRenzy G, Gillham B & Jones MT (1982) Role of serotonin in the control of secretion of corticotrophin releasing factor. Journal of Endocrinology 93" 151-160. Hrycaj P, Stratz T & Muller W (1993) Platelet 3H-imipramine uptake receptor density and serum serotonin in patients with fibromyalgia/fibrositis syndrome. Journal of Rheumatology 20: 1986-1987. Hudson JI, Goldenberg DL, Pope HG et al (1992) Comorbidity of fibromyalgia with medical and psychiatric disorders. American Journal of Medicine 92: 363-367. Jones MT, Hillhouse EW & Burden J (1976) Effect of various putative neurotransmitters on the secretion of cortic0trophin-releasing hormone from the rat hypothalamus in vitro: a model of the neurotransmitters involved. Journal of Endocrinology 69: 1-10. Kamilaris TC, Calogero AE & Johnson EO (1989) Effect of neuropeptide Y on hypothalamicpituitary-adrenal .tunction in the rat. 19th Annual Meeting of the Society for Neuroscience, Phoenix, AZ.
NEUROHORMONAL PERTURBATIONS IN FM
377
Koenig JI (1990) Regulation of the hypothalamo-pituitary axis by neuropeptide Y. Annals of the New York Academy of Sciences 611:317-328. Krieger DT & Rizzo F (1969) Serotonin mediation of circadian periodicity of plasma 17-hydroxycorticosteroids. American Journal of Physiology 217: 1703-1707. Lundberg JM, Franco-Cereceda A, Lacroix J-S & Pernow J (1990) Neuropeptide Y and sympathetic neurotransmission. Annals of the New York Academy of Sciences 611: 166-174. McCain GA (1989) Nonmedicinal treatments in primary fibromyalgia. Rheumatic Disease Clinics of North America 15: 73-89. McCain GA & Tilbe KS (1989) Diurnal hormone variation in fibromyalgia syndrome: A comparison with rheumatoid arthritis. Journal of Rheumatology 16 (supplement 19): 154-157. Moldofsky H & Warsh JJ (1978) Plasma tryptophan and musculoskeletal pain in non-articular rheumatism ('fibrositis syndrome'). Pain 5: 65-71. Neeck G & Riedel W (1992) Thyroid function in patients with fibromyalgia syndrome. Journal of Rheumatology 19:1120-1122. Nikolarakis KE, Almeida OFX & Herz A (1986) Stimulation of hypothalamic I]-endorphin and dynorphin release by corticotropin-releasing factor. Brain Research 399: 152-155. Qiao Z-G, Vaerc~yH & M0rkrid L (1991) Electrodermal and microcircalatory activity in patients with fibromyalgia during baseline, acoustic stimulation and cold pressor tests. Journal of Rheumatology 18: 1383-1389. Rowe PC, Bou-Holaigah I, Kan JS & Calkins H (1995) Is neurally mediated hypotension an unrecognised cause of chronic fatigue? Lancet 345: 623-624. Russell IJ (1989) Neurohormonal aspects of fibromyalgia syndrome. Rheumatic Disease Clinics of North America 15: 149-168. Russell IJ (1995) Neurohormonal: Abnormal laboratory findings related to pain and fatigue in fibromyalgia. Journal of Musculoskeletal Pain 3: 59-65. Russell IJ, Michalek JE & Vaprio GA (1989) Serum amino acids in fibrositis/fibromyalgia syndrome. Journal of Rheumatology 16 (supplement 19): 158-163. Russell IJ, Bowden CL & Michalek J (1992) Platelet 3H-imipramine uptake receptor density and serum serotonin levels in patients with fibromyalgia/fibrositis syndrome. Journal of Rheumatology 19: 104-109. Scapagnini U, Moberg GP, Van Loon GR et al (1971) Relation of brain 5-hydroxytryptamine content to the diurnal variation of plasma corticosterone in the rat. Neuroendocrinology 7: 90-96. Sharpe M (1995) Cognitive behavior therapy and the treatment of chronic fatigue syndrome. Journal of Musculoskeletal Pain 3: 141-147. Sternberg EM (1993) Hypoimmune fatigue syndromes: Diseases of the stress response? Journal of Rheumatology 20: 418-421. Streeten DHP & Anderson GH (1992) Delayed orthostatic intolerance. Archives oflnternal Medicine 152: 1066-1072. Sutton RE, Koob GF, LeMoal M e t al (1982) Corticotropin-releasing factor produces behavioral activation in rats. Nature 297: 331-333. Swerdlow NR, Geyer MA, Vale WW & Koob GF (1982) Corticotropin-releasing factor potentiates acoustic startle in rats: Blockade by chlordiazepoxide. Psychopharmacology (Berlin) 88: 147-152. VaerCy H, Helle R, FCrre O et al (1988) Cerebrospinal fluid levels of 13-endorphin in patients with fibromyalgia (fibrositis syndrome). Journal of Rheumatology 15:1804-1806. VaerCy H, Qiao Z-G, M~rkrid L & Fc~rre 0 (1989) Altered sympathetic nervous system response in patients with fibromyalgia (fibrositis syndrome). Journal of Rheumatology 16: 1460-1465. Vamvakopoulos NC & Chrousos GP (1993) Evidence of direct estrogenic regulation of human corticotropin-releasing hormone gene expression. Journal of Clinical Investigation 92: 18961902. van Denderen JC, Boersma JW, Zeinstra Pet al (1992) Physiological effects of exhaustive physical exercise in primary fibromyalgia syndrome (PFS): Is PFS a disorder of neuroendoorine reactivity? Scandinavian Journal of Rheumatology 21: 35-37. Wasman J & Zatzkis SM (1986) Fibromyalgia and menopause: Examination of the relationship. Postgraduate Medicine 80: 165-171. Wolfe F, Smythe HA, Yunus MB et al (1990) The American College of Rheumatology 1990 criteria for the classification of fibromyalgia. Arthritis and Rheumatism 33: 160-172. Yehuda R & Meyer 1S (1984) A role for serotonin in the hypothalamic-pituitary-adrenal response to insulin stress. Neuroendocrinology 38: 25-32.
378
L . J . CROFFORD ET AL
Young EA (1996) The role of gonadal steroids in hypothalamic-pituitary-adrenal axis regulation. Critical Reviews in Neurobiology 9: 371-381. Young EA, Kotun J, Haskett RF et al (1993) Dissociation between pituitary and adrenal suppression to dexamethasone in depression. Archives of General Psychiatry 50: 395-403. Yunus MB, Masi AT, Calabro JJ et al (1981) Primary fibromyalgia (fibrositis): clinical study of 50 patients with matched normal controls. Seminars in Arthritis and Rheumatism 11: 1 5 1 - 1 7 1 . Yunus MB, Dailey JW, Aldag JC et al (1992) Plasma and urinary catecholamines in primary fibromyalgia: A controlled study. Journal of Rheumatology 19" 95-97.