Dimitris A. Papanicolaou

The Pathophysiologic Roles of Interleukin-6 in Human Disease

Dr. Dimitris A. Papanicolaou (Developmental Endocrinology Branch, National Institute of Child Health and Human Development, National Institutes of Health [NIH], Bethesda, Maryland): During inflammation, the inflammatory cytokines tumor necrosis factor-, interleukin-1, and interleukin-6 are secreted, in that order [1, 2]. Interleukin-6 then inhibits the secretion of tumor necrosis factor- and interleukin-1 [3], activates the production of acute-phase reactants from the liver [4], and stimulates the hypothalamic-pituitary-adrenal axis [5] to help control the inflammation. In this sense, interleukin-6 is both a proinflammatory and an anti-inflammatory cytokine. It is produced not only by immune and immune accessory cells (such as monocytes, macrophages, lymphocytes, endothelial cells, fibroblasts, mast cells, astrocytes, and microglia) but also by many nonimmune cells and organs (such as osteoblasts, bone marrow stromal cells, keratinocytes, synoviocytes, chondrocytes, intestinal epithelial cells, Leydig cells of the testis, folliculostellate cells of the pituitary, endometrial stromal cells, trophoblasts, and vascular smooth-muscle cells) [4, 6-12]. What makes interleukin-6 particularly interesting to physicians is its marked pleiotropy and its involvement not only in inflammation but in the regulation of endocrine and metabolic functions. Its diverse actions are summarized in the Table 1 [13]. Table 1. Actions of Interleukin-6 Molecular Biology of Interleukin-6 Located on the short arm of chromosome 7, the interleukin-6 gene consists of 5 exons and 4 introns and has a fairly complex transcriptional regulation [14]. The interleukin-6 promoter has recognition sites for transcription factors NF-IL6 (C/EBP ), which belongs to the C/EBP family, and NF-B, which is a major mediator of inflammatory stimuli [15, 16] (Figure 1). Figure 1. Transcriptional regulation of the interleukin-6 (IL-6) promoter. Interleukin-6 exerts its broad range of action through the interleukin-6 receptor, a single-pass transmembrane receptor not directly involved in signal transduction. Instead, activation of the receptor by interleukin-6 induces homodimerization of another transmembrane receptor, gp130, which initiates the transduction cascade [13]. The interleukin-6 receptor has a second soluble form that consists of the extracellular domain of the membrane receptor. Interleukin-6 also activates gp130 through this soluble form, even on cells that lack the interleukin-6 receptor on their membranes [17, 18]. For example, interleukin-6 can cause cardiac hypertrophy through gp130, even though cardiac myocytes lack the interleukin-6 receptor. The gp130 receptor is shared by many cytokines and growth factors for signal transduction, including interleukin-11, oncostatin-M, leukemia inhibitory factor, ciliary neurotrophic factor, cardiotropin-1, and leptin [13] (Figure 2). Figure 2. Pleiotropy of interleukin-6 (IL-6) action. Endocrine and Metabolic Actions of Interleukin-6 As Figure 3 shows, interleukin-6 has a broad array of actions on the endocrine and metabolic systems. Figure 3. Regulation of the secretion and endocrine actions of interleukin-6 (IL-6). Hypothalamic-Pituitary-Adrenal Axis Animal studies have shown that interleukin-6 acutely activates the hypothalamic-pituitary-adrenal axis by acting primarily on the corticotropin-releasing hormone neuron. Specifically, a blockade of corticotropin-releasing hormone inhibits the effects of exogenous interleukin-6 on the hypothalamic-pituitary-adrenal axis in rats [19]. Subcutaneous administration of interleukin-6 to normal human volunteers resulted in elevated plasma levels of adrenocorticotropin hormone (ACTH) and then an increase in plasma levels of cortisol [20]. The plasma level of cortisol peaked after the plasma level of ACTH peaked; this indicates that, at least in this acute setting, cortisols response to interleukin-6 administration is mediated by release of ACTH [21]. Interleukin-6 seems to be one of the most potent stimuli of the hypothalamic-pituitary-adrenal axis in humans. Subcutaneous administration of interleukin-6 once a day for 7 days resulted in remarkable enlargement of the adrenal glands similar to that seen after prolonged activation of the adrenal glands by ACTH (as in Cushing disease or ectopic ACTH production) [22]. In animals and humans, glucocorticoids inhibit production of interleukin-6 in vitro and in vivo [23, 24]. In a recent study [25], administration of hydrocortisone or dexamethasone attenuated exercise-induced elevation of plasma levels of interleukin-6. Conversely, correction of hypercortisolism by surgical removal of a corticotroph adenoma when plasma levels of cortisol were undetectable increased plasma levels of interleukin-6 more than fourfold in patients with Cushing disease [26]. Therefore, interleukin-6 stimulates the hypothalamic-pituitary-adrenal axis and cortisol exerts negative feedback on secretion of interleukin-6. Interleukin-6 thus functions as a hormone in the traditional sense: It participates in a feedback loop of the hypothalamic-pituitary-adrenal axis. Thermogenesis and Basal Metabolic Rate Several cytokines, especially interleukin-1, are pyrogenic in humans and animals [27]. Administration of interleukin-6 causes elevations in body temperature and resting metabolic rate in humans [20]. In animals, the fact that nonsteroidal anti-inflammatory agents can inhibit the thermogenic effect of interleukin-6 suggests that this effect may be mediated by prostanoids [28]. The Steroid Withdrawal Syndrome The concept of the steroid withdrawal syndrome was introduced in 1965 by Amatruda and colleagues [29]. The syndrome is characterized by fever; headache; nausea; fatigue; malaise; somnolence; anorexia; and, less commonly, flu-like symptoms, such as arthralgias and myalgias. These symptoms occur during an abrupt reduction in levels of circulating cortisol and have been seen in patients who became severely hypocortisolemic when they underwent curative transsphenoidal surgery for Cushing disease. At that time, plasma levels of interleukin-6 were greatly elevated [26]. Normal volunteers and patients who received interleukin-6 had similar symptoms; this suggests that interleukin-6 participates in the pathogenesis of the steroid withdrawal syndrome [20, 21, 30]. Vasopressin and the Syndrome of Inappropriate Secretion of Antidiuretic Hormone The release of arginine vasopressin by the posterior pituitary is controlled by changes in intravascular volume and by osmotic stimuli. The syndrome of the inappropriate secretion of antidiuretic hormone occurs in the absence of serum hyperosmolarity or hypovolemia and can be caused by several conditions, including certain types of trauma, infections (meningitis and pneumonia), and inflammation [31]. During the syndrome, production of inflammatory cytokines (including interleukin-6) increases. Because high doses of interleukin-6 increase plasma levels of vasopressin in humans [32], endogenous interleukin-6 may also participate in the pathogenesis of this syndrome. Interleukin-6 as a Stress Hormone Because it innervates many immune organs, such as the spleen and the thymus, the autonomic nervous system interacts directly with the immune system [33, 34]. Stress or administration of adrenaline to animals elevates levels of endogenous interleukin-6, but pretreatment with a -adrenergic antagonist abolishes this effect. These effects suggest that interleukin-6 secretion is stimulated through -adrenergic receptors [35, 36]. In a recent study [37], administering adrenaline to humans increased plasma levels of interleukin-6. In normal volunteers, treadmill exercise also increased levels of plasma interleukin-6. In addition, peak plasma levels of catecholamines were positively correlated with plasma levels of interleukin-6 [25]. These data indicate that interleukin-6 is secreted during stress, probably through a -adrenergic receptor mechanism, and that it participates in the stress response. Lipid Metabolism Normal volunteers had precipitous reductions in serum total cholesterol levels, apolipoprotein B levels (this reflects low-density lipoprotein cholesterol), and triglyceride levels within 24 hours of interleukin-6 administration [38]. During sustained elevation of plasma catecholamine levels (such as that which occurs immediately after myocardial infarction), serum lipid levels are temporarily reduced, rendering serum cholesterol measurements misleading [39]. Whether catecholamine-stimulated endogenous interleukin-6 contributes to the transient decrease in serum lipid concentrations observed in conditions with increased sympathoneural discharge requires further study. Thyroid Axis and the Euthyroid Sick Syndrome Exogenous interleukin-6 decreased the secretion of thyroid-stimulating hormone in animals in vivo [5], and interleukin-6 was recently shown to be associated with a decrease in serum levels of thyroid-stimulating hormone and triiodothyronine in humans within 4 hours of administration. Interleukin-6 seemed to have a more lasting effect on triiodothyronine levels; the decrease persisted for at least 24 hours after a single injection of interleukin-6 [20, 21]. Thus, interleukin-6 was associated with changes in thyroid function test results similar to those seen in the euthyroid sick syndrome, a condition of physiologic hypothyroidism that occurs during nonthyroidal illness, apparently in an attempt by the organism to conserve energy. Depending on the severity and duration of the illness, it ranges from an isolated decrease in serum triiodothyronine levels in mild cases to a decrease in serum levels of free thyroxine and, finally, to subnormal thyroid-stimulating hormone levels in more severe cases [40]. Interleukin-6 levels are frequently elevated in conditions that are associated with the euthyroid sick syndrome (such as infection or inflammation, major trauma or surgery, and prolonged stays in the intensive care unit) and

Chronic Fatigue Syndrome – A clinically empirical approach to its definition and study

BackgroundThe lack of standardized criteria for defining chronic fatigue syndrome (CFS) has constrained research. The objective of this study was to apply the 1994 CFS criteria by standardized reproducible criteria.MethodsThis population-based case control study enrolled 227 adults identified from the population of Wichita with: (1) CFS (n = 58); (2) non-fatigued controls matched to CFS on sex, race, age and body mass index (n = 55); (3) persons with medically unexplained fatigue not CFS, which we term ISF (n = 59); (4) CFS accompanied by melancholic depression (n = 27); and (5) ISF plus melancholic depression (n = 28). Participants were admitted to a hospital for two days and underwent medical history and physical examination, the Diagnostic Interview Schedule, and laboratory testing to identify medical and psychiatric conditions exclusionary for CFS. Illness classification at the time of the clinical study utilized two algorithms: (1) the same criteria as in the surveillance study; (2) a standardized clinically empirical algorithm based on quantitative assessment of the major domains of CFS (impairment, fatigue, and accompanying symptoms).ResultsOne hundred and sixty-four participants had no exclusionary conditions at the time of this study. Clinically empirical classification identified 43 subjects as CFS, 57 as ISF, and 64 as not ill. There was minimal association between the empirical classification and classification by the surveillance criteria. Subjects empirically classified as CFS had significantly worse impairment (evaluated by the SF-36), more severe fatigue (documented by the multidimensional fatigue inventory), more frequent and severe accompanying symptoms than those with ISF, who in turn had significantly worse scores than the not ill; this was not true for classification by the surveillance algorithm.ConclusionThe empirical definition includes all aspects of CFS specified in the 1994 case definition and identifies persons with CFS in a precise manner that can be readily reproduced by both investigators and clinicians.

Responses of the sympathetic nervous system and the hypothalamic–pituitary–adrenal axis to interleukin‐6: A pilot study in fibromyalgia

OBJECTIVE To determine whether deficient activity of the hypothalamic corticotropin-releasing hormone (CRH) neuron, which stimulates the hypothalamic-pituitary-adrenal (HPA) axis and the central control nuclei of the sympathetic nervous system and inhibits ascending pain pathways, may be pathogenic in patients with fibromyalgia (FM). METHODS We administered interleukin-6 (IL-6; 3 microg/kg of body weight subcutaneously), a cytokine capable of stimulating hypothalamic CRH release, and measured plasma levels of adrenocorticotropic hormone (ACTH), cortisol, and catecholamines and their metabolites and precursors. Thirteen female FM patients and 8 age- and body mass index-matched female controls were studied. The diagnosis of FM was made according to American College of Rheumatology criteria. Tender points were quantitated by pressure algometry. All subjects had HPA axis studies. Seven FM patients and 7 controls also had catecholamine measurements. RESULTS After IL-6 injection, delayed ACTH release was evident in the FM patients, with peak levels at 96.9 +/- 6.0 minutes (mean +/- SEM; control peak 68.6 +/- 10.3 minutes; P = 0.02). Plasma cortisol responses to IL-6 did not differ significantly between patients and controls. Basal norepinephrine (NE) levels were higher in the FM patients than in the controls. While a small, although not significant, rise in NE levels occurred after IL-6 injection in the controls, NE levels dramatically increased over basal levels in the FM patients between 60 and 180 minutes after IL-6 injection. Both peak NE levels (mean +/- SEM 537.6 +/- 82.3 versus 254.3 +/- 41.6 pg/ml; P = 0.0001) and time-integrated NE responses (93.2 +/- 16.6 pg/ml x minutes(-3) versus 52.2 +/- 5.7 pg/ml x minutes(-3); P = 0.038) were greater in FM patients than in controls. Heart rate was increased by IL-6 injection in FM patients and controls, but rose to significantly higher levels in the FM patients from 30 minutes to 180 minutes after IL-6 injection (P < 0.03). CONCLUSION Exaggerated NE responses and heart rate increases, as well as delayed ACTH release, were observed among female FM patients compared with age-matched female controls. Delayed ACTH release after IL-6 administration in FM is consistent with a defect in hypothalamic CRH neuronal function. Exaggerated NE release may reflect abnormal regulation of the sympathetic nervous system, perhaps secondary to chronically deficient hypothalamic CRH. The excessive heart rate response after IL-6 injection in FM patients may be unrelated to the increase in NE, or it may reflect an alteration in the sensitivity of cardiac beta-adrenoceptors to NE. These responses to a physiologic stressor support the notion that FM may represent a primary disorder of the stress system.

Dose effects of recombinant human interleukin-6 on pituitary hormone secretion and energy expenditure.

Interleukin-6 (IL-6), the main circulating cytokine, is putatively a major mediator of the effects of the immune system on several endocrine axes and intermediate metabolism. We performed dose-response studies of recombinant human IL-6 on pituitary hormone secretion in 15 healthy male volunteers, using 5 single, escalating subcutaneous doses of IL-6 (0.1, 0.3, 1.0, 3.0 and 10.0 micrograms/kg body weight), each in 3 volunteers. We measured resting metabolic rate (RMR) with indirect calorimetry and plasma anterior pituitary hormones and vasopressin (AVP) at baseline and half-hourly over 4 h after the injection. All doses examined were tolerated well and produced no significant adverse effects. Dose-dependent RMR increases were observed in response to the 3.0- and 10.0-microgram/kg doses of IL-6, beginning at 60 min and slowly peaking between 180 and 240 min. Plasma adrenocorticotropic-hormone concentrations increased dramatically and dose-dependently in all the patients who received the 3.0- and 10.0-microgram/kg doses of IL-6, respectively, peaking to 150 and 255 pg/ml at 60 min, and slowly returning to normal by 4 h. Corresponding plasma cortisol levels peaked dose-dependently between 90 and 150 min, but remained elevated throughout the sampling period. In contrast, the growth hormone (GH) dose-response was bell-shaped, with maximum (approximately 100-fold) stimulation achieved by 3.0 micrograms/kg IL-6. Prolactin (PRL) showed a similar but less pronounced response pattern. Thyroid-stimulating hormone (TSH) dose-dependently and progressively decreased over the 240 min, while gonadotropins showed no clear-cut changes. In conclusion, subcutaneous IL-6 administration induced synchronized dose-dependent increases in the RMR and hypothalamic-pituitary-adrenal axis activity, suggesting that hypothalamic corticotropin-releasing hormone may mediate both of these functions in humans. IL-6 also acutely stimulated GH and PRL secretion and suppressed TSH secretion. The dose of 3.0 micrograms/kg could be used safely in the study of patients with disturbances of the hypothalamic-pituitary unit or of thermogenesis.

Paradoxical Response to Dexamethasone in the Diagnosis of Primary Pigmented Nodular Adrenocortical Disease

Primary pigmented nodular adrenocortical disease is a bilateral adrenal disorder that leads to adrenocorticotropin (ACTH)-independent cases of the Cushing syndrome. In most cases, this disease occurs as part of the Carney complex, an autosomal dominant, multiple neoplasia syndrome that consists of skin lentigines, myxomas, and other nonendocrine and endocrine tumors (1). Cardiac myxoma, a tumor that may cause stroke and death, is among the most frequent and is often the first manifestation of the Carney complex (1, 2). The diagnosis of primary pigmented nodular adrenocortical disease should be followed by screening for the Carney complex and, in particular, its potentially fatal cardiac component (1, 2). Establishing the diagnosis of primary pigmented nodular adrenocortical disease can be difficult because the associated hypercortisolism usually develops slowly over several years and the clinical manifestations may be subtle (3, 4). Results of radiologic imaging can be normal or indistinguishable from those that indicate adrenal nodularity, which is frequently present in other primary forms of the Cushing syndrome and in normal elderly persons (5). In addition, plasma ACTH levels may not be fully suppressed, especially in mild or periodic cases of the Cushing syndrome (3, 4). Previous reports have mentioned a paradoxical increase in glucocorticoid level after administration of various doses of dexamethasone in patients with primary pigmented nodular adrenocortical disease (6-11), but this observation has not been systematically investigated. The Liddle test (12), the administration of low-dose and high-dose dexamethasone, is used to differentiate between pituitary-dependent and non-pituitary-dependent forms of the Cushing syndrome (12, 13). Suppression of urinary free cortisol greater than 90% and suppression of 17-hydroxycorticosteroid greater than 64% during the Liddle test identified all patients with pituitary-dependent cases of the Cushing syndrome in a large study of 118 patients (13). This test, however, has not been evaluated for its ability to differentiate between primary adrenocortical causes of the Cushing syndrome. We analyzed data from 21 patients with primary pigmented nodular adrenocortical disease who had been evaluated at the National Institutes of Health (NIH) over the past 30 years. Sixteen patients underwent the Liddle test in addition to other testing. We compared the usefulness of the Liddle test in diagnosing primary pigmented nodular adrenocortical disease and in differentiating this disease from other primary adrenocortical disorders. For this purpose, we investigated two control groups of patients with primary adrenocortical diseases that lead to ACTH-independent cases of the Cushing syndrome: patients with macronodular adrenocortical disease, a condition that is almost always bilateral (14-16), and patients with unilateral, single adrenocortical adenomas. Methods Patients We reviewed records of patients with ACTH-independent cases of the Cushing syndrome who were seen at the NIH clinical center over the past 30 years. The Carney complex was diagnosed on the basis of published criteria (2). Primary pigmented nodular adrenocortical disease, macronodular adrenocortical disease, and single adrenocortical adenoma were confirmed by histologic analysis after adrenalectomy or at autopsy, according to published criteria (1, 14-17). Macronodular disease, in particular, was diagnosed in the presence of multiple, bilateral, nonpigmented adrenocortical adenomas and a substantial increase in the weight of the adrenal glands (14-16). For some patients, little information was obtained about levels of plasma ACTH and urinary free cortisol because radioimmunoassays, which are now used to determine such variables, were not available until the late 1970s. The two control groups consisted only of patients for whom all data were available. For each patient, we analyzed plasma ACTH levels at 8:00 a.m.followed by ovine corticotropin-releasing hormone stimulationand diurnal plasma cortisol variation, as described elsewhere (13, 18). A 6-day Liddle test was conducted for each patient, as described elsewhere (12, 13): After 2 days of baseline measurement of urinary steroid excretion, dexamethasone, 0.5 mg, was given orally every 6 hours for 2 days starting at 6:00 a.m.; the dosage of dexamethasone was then increased to 2 mg every 6 hours for the last 2 days of the test. For children, the lowest dose of dexamethasone was adjusted to 7.2 g/kg of body weight and the highest dose was adjusted to 28.5 g/kg (3). In all patients, the 24-hour urinary free cortisol level was expressed per square meter of body surface area (g/m2); 17-hydroxycorticosteroid excretion was expressed per grams of creatinine excreted in 24 hours (mg/g). We also performed computed tomography of the adrenal glands, as described elsewhere (19). Hormone Assays Plasma ACTH and cortisol levels were measured as described elsewhere (13, 18). Urinary free cortisol excretion was measured by using direct radioimmunoassay (20, 21). The intra-assay coefficient of variation was 5%, and the interassay coefficient of variation was 10%. Urinary 17-hydroxycorticosteroid excretion was measured by using a modification of the colorimetric method described by Porter and Silber (20, 21). The intra-assay and interassay coefficients of variation were 6% and 11%, respectively. Statistical Analysis All data are expressed as the mean SE. For all statistical comparisons, a P value less than 0.05 was considered significant. Data were analyzed by using Statistica software (StatSoft, Inc., Tulsa, Oklahoma). Friedman analysis of variance was initially used within each group to determine differences in urinary free cortisol and 17-hydroxycorticosteroid levels in response to dexamethasone administration during the Liddle test. The Wilcoxon matched-pair test was used to determine which time points significantly differed from baseline. The Mann-Whitney U test was used to assess differences among groups. The specificity and sensitivity of the Liddle test were determined, and receiver-operating characteristic (ROC) curves were constructed, as described elsewhere (13, 18), to assess the usefulness of each method for differential diagnosis. Results Twenty-one patients (8 males and 13 females) with primary pigmented nodular adrenocortical disease were seen at the NIH clinical center over the past 30 years. Most patients were young: Age at diagnosis was 27.7 2.9 years. Four patients were children (2 boys and 2 girls; age at diagnosis, 10 1.8 years), and 17 patients were adults (6 men and 11 women; age at diagnosis, 31.9 2.6 years). No children were included in the two control groups. For 20 of 21 patients (95%), primary pigmented nodular adrenocortical disease occurred as a component of the Carney complex. Three of these patients had periodic cases of the Cushing syndrome (14%), which were characterized by periods of clinical symptoms and hypercortisolemia followed by periods of normalization of the body habitus and eucortisolemia, as described elsewhere (3, 4). The length of these periods varied considerably, from every 2 to 3 months (2 patients) to 1 to 2 weeks (1 patient). Four patients with primary pigmented nodular adrenocortical disease (19%) had subclinical cases of the Cushing syndrome, which were characterized by a relative paucity of clinical findings of the syndrome (with the exception of osteoporosis) and eucortisolemia with abnormal diurnal cortisol variation, as described elsewhere (4). Computed tomography revealed normal-sized adrenal glands in 19 patients with primary pigmented nodular adrenocortical disease (90%). Two patients had macronodules (nodules larger than 1 cm); one nodule had calcifications. An irregular contour of the adrenal glands, as described elsewhere (4, 5, 19), could be seen in 10 of the patients with primary pigmented nodular adrenocortical disease who had normal-sized adrenal glands (48%) (Figure 1). On computed tomography, all patients with macronodular adrenocortical disease had macronodules and all patients with single adenomas had a single mass. Figure 1. Macronodular appearance on computed tomography of the adrenal gland in a patient with primary pigmented adrenocortical disease ( left ) and the more typical bead-on-a-string appearance of an adrenal gland in a patient with primary pigmented nodular adrenocortical disease ( right ) Morning plasma ACTH levels were in the low end of the normal range but were still measurable in almost all of the patients in the three groups; however, no differences were seen in baseline ACTH values or responses of ACTH or cortisol to ACTH-releasing hormone stimulation (P>0.1 for all comparisons, Mann-Whitney U test) (data not shown). All patients demonstrated loss or reversal of the normal pattern of diurnal cortisol variation, but no differences were seen among groups (P>0.1 for all comparisons, Mann-Whitney U test) (data not shown). Data were available for 16 patients with primary pigmented nodular adrenocortical disease who had 17-hydroxycorticosteroid or urinary free cortisol responses to the Liddle test. Thirteen of these 16 patients had both glucocorticoids measured; for 3 patients only 17-hydroxycorticosteroid measurements were available. In these 16 patients, urinary excretion of both glucocorticoids increased gradually during the Liddle test (Figure 2). Friedman analysis of variance showed a P value less than 0.001 for 17-hydroxycorticosteroid and urinary free cortisol values in patients with primary pigmented nodular adrenocortical disease; the P value was greater than 0.2 for 17-hydroxycorticosteroid and urinary free cortisol values in patients with macronodular adrenocortical disease. For 17-hydroxycorticosteroid and urinary free cortisol values in patients with single adenomas, the P values were 0.05 and greater than 0.2, respectively. Figure 2. Mean percentage change in 24-hour excretion of 17-hyd

Chronic Systemic Inflammation in Overweight and Obese Adults

To the Editor: Nitric oxide–mediated damage has been implicated in a number of neurological diseases including stroke1, 2 and multiple sclerosis (MS).3 For instance, monocytes expressing high levels of nitric oxide synthetase have been found in plaques from the brains of patients with MS.4 The proximal agent of neuronal cell damage may be peroxynitrite, which is formed in vivo from the synthesis of nitric oxide and superoxide.

Cognitive Dysfunction Relates to Subjective Report of Mental Fatigue in Patients with Chronic Fatigue Syndrome

Patients with chronic fatigue syndrome (CFS) frequently complain of cognitive dysfunction. However, evidence of cognitive impairment in CFS patients has been found in some, but not other, studies. This heterogeneity in findings may stem from the relative presence of mental fatigue in the patient populations examined. The present study assessed this possibility in a population-based sample of CFS patients. In all, 43 patients with CFS defined by the criteria of the 1994 research case definition using measurements recommended by the 2003 International CFS Study Group, and 53 age-, sex-, and race/ethnicity-matched nonfatigued subjects were included in the study. Mental fatigue was assessed using the mental fatigue subscale of the multidimensional fatigue inventory. Cognitive function was evaluated using an automated battery of computerized tests (Cambridge neuropsychological test automated battery (CANTAB)) that assessed psychomotor function, planning and problem-solving abilities, and memory and attentional performance. CFS patients with significant complaints of mental fatigue (score of mental fatigue 2 standard deviations above the mean of nonfatigued subjects) exhibited significant impairment in the spatial working memory and sustained attention (rapid visual information processing) tasks when compared to CFS patients with low complaints of mental fatigue and nonfatigued subjects. In CFS patients with significant mental fatigue, sustained attention performance was impaired only in the final stages of the test, indicating greater cognitive fatigability in these patients. CFS patients with low mental fatigue displayed performance comparable to nonfatigued subjects on all tests of the CANTAB battery. These findings show strong concordance between subjective complaints of mental fatigue and objective measurement of cognitive impairment in CFS patients and suggest that mental fatigue is an important component of CFS-related cognitive dysfunction.

Dose-Dependent Effects of Recombinant Human Interleukin-6 on the Pituitary-Testicular Axis

Inflammatory cytokines are soluble mediators of immune function that also regulate intermediate metabolism and several endocrine axes. To examine the effects of interleukin-6 (IL-6), the main circulating cytokine, on the hypothalamic-pituitary-testicular axis in men, we performed dose-response studies of recombinant human IL-6 (rHuIL-6) in normal volunteers. Increasing single doses of IL-6 (0.1, 0.3, 1.0, 3.0, and 10.0 microg/kg body weight) were injected subcutaneously into 15 healthy male volunteers (3 at each dose) in the morning. We measured the circulating levels of testosterone, luteinizing hormone (LH), follicle-stimulating hormone (FSH), and sex hormone binding globulin (SHBG) at baseline and then at 24 h, 48 h, and 7 days after the IL-6 injection. LH and FSH levels were also measured half-hourly for the first 4 h after the IL-6 injection. All IL-6 doses were tolerated well and produced no significant adverse effects. Mean peak plasma IL-6 levels achieved after IL-6 administration were 8 +/- 1, 22 +/- 5, 65 +/- 22, 290 +/- 38, and 4050 +/- 149 pg/ml, respectively for the five doses. We observed no significant changes in plasma testosterone levels after the two smaller IL-6 doses. The three higher IL-6 doses, however, caused significant decreases in testosterone levels by 24 h, which persisted at 48 h and returned to baseline by 7 days. The higher testosterone suppression was after the 3.0 microg/kg dose, making the dose-response curve bell-shaped. There also appeared to be small but not significant increases in LH levels after the three higher IL-6 doses, which were not acute and seemed to follow temporally the testosterone decreases. The concurrent plasma levels of FSH and SHBG were not appreciably affected by any IL-6 dose. In conclusion, subcutaneous IL-6 administration, which caused acute elevations in circulating IL-6 levels of a similar magnitude to those observed in severe inflammatory and noninflammatory stress, induced prolonged suppression in testosterone levels in healthy men without apparent changes in gonadotropin levels. This suggests that IL-6 might induce persistent testicular resistance to LH action or suppression of Leydig cell steroidogenesis or both, with potential adverse effects on male reproductive function.

Neuroendocrine Aspects of Chronic Fatigue Syndrome

Chronic fatigue syndrome (CFS) is a serious health concern affecting over 800,000 Americans of all ages, races, socioeconomic groups and genders. The etiology and pathophysiology of CFS are unknown, yet studies have suggested an involvement of the neuroendocrine system. A symposium was organized in March 2001 to explore the possibility of an association between neuroendocrine dysfunction and CFS, with special emphasis on the interactions between neuroendocrine dysfunction and other abnormalities noted in the immune and autonomic nervous systems of individuals with CFS. This paper represents the consensus of the panel of experts who participated in this meeting.

Immunologic Aspects of Chronic Fatigue Syndrome

Chronic fatigue syndrome (CFS) is a serious health concern affecting over 800,000 Americans of all ages, races, socioeconomic groups and genders. The etiology and pathophysiology of CFS are unknown, yet studies have suggested an involvement of the immune system. A symposium was organized in October 2001 to explore the possibility of an association between immune dysfunction and CFS, with special emphasis on the interactions between immune dysfunction and other abnormalities noted in the neuroendocrine and autonomic nervous systems of individuals with CFS. This paper represents the consensus of the panel of experts who participated in this meeting. Data suggest that persons with CFS manifest changes in immune responses that fall outside normative ranges, but current research does not provide definitive evidence on whether these immune abnormalities are a cause or result of the illness. It has become clear that CFS cannot be understood based on single measurements of immune, endocrine, cardiovascular, or autonomic nervous system dysfunction. This panel encourages a new emphasis on multidisciplinary research into CFS.