Disturbed neuroendocrine immune interactions in CFS
Annemieke Kavelaars, Wietse Kuis, Lidewij Knook, Gerben Sinnema and Cobi J. Heijnen
DDepartments of Pediatric
Immunology (A.K., W.K., L.K., C.J.H.) and Psychology (G.S.), Wilhelmina
Children’s Hospital of the University Medical Center Utrecht, 3584 EA Utrecht,
Address correspondence and requests for reprints to: Dr. Annemieke Kavelaars, Wilhelmina Children’s Hospital of the University Medical Center Utrecht, Department of Pediatric Immunology, Room KC 03.068.0, Lundlaan 6, 3584 EA Utrecht, The Netherlands. E-mail: email@example.com.
Chronic fatigue syndrome (CFS) is a disease characterized by debilitating fatigue for at least 6 months that has resulted in a substantial reduction in the activity level of the individual and is not attributable to known clinical conditions. For study purposes, the Centers for Disease Control and Prevention (Atlanta, GA) have defined criteria for CFS (1).
In addition to the persistent or relapsing fatigue, at least four other symptoms from a list of eight should be present concurrently with the fatigue. These symptoms include unrefreshing sleep, postexertion malaise, multi-joint pain, new headaches, muscle pain, tender cervical or axillary lymph nodes, sore throat, and impaired memory or concentration (1).
The etiology of CFS is unknown. Viral infections have been suggested as precipitating events, and a number of studies suggest involvement of viruses in at least part of the patients (2, 3, 4). In a recent study on children and adolescents with CFS, 60% of the patients indicated an acute disease at onset (5). However, to date, there is no evidence for a specific virus associated with CFS (6).
Other studies have focused on immunological dysfunction in CFS patients and suggested changes in cytokine production, natural killer cell activity, and alterations in T-cell reactivity (7, 8, 9). Although immunological changes have been described, a consistent pattern of immunological abnormalities has not been found.
In adults with CFS, evidence has been presented for changes in the neuroendocrine system. Demitrack et al. (10) showed that the reactivity of the hypothalamo-pituitary-adrenal (HPA) axis is decreased in adult patients with CFS as compared with controls.
Unfortunately, treatment with hydrocortisone results in only limited improvement in CFS patients. Cleare et al. (11) presented data showing that oral administration of low doses of hydrocortisone could improve the clinical condition in about 30% of a selected group of patients. In another study it was shown that oral hydrocortisone administration did result in some improvement as measured by a change in Wellness score, which is a global health scale. However, fatigue and activity did not change significantly in this study (11).
In view of the generalized symptomatology in CFS, the search for a single factor cause does not seem to be the most adequate approach. We hypothesized that the symptomatology in CFS may result from abnormalities in interorgan communication rather than from abnormalities in a single organ system. One aspect of interorgan communication is based on the production and secretion of (neuro)endocrine mediators by a given organ system and the presence and reactivity of specific receptors in the target organ system(s). Thus, not only alterations in the actual level of neuroendocrine mediators, but also changes in the way target organs respond to these mediators, may result in inadequate communication. Inadequate communication could contribute to the pathophysiology in CFS and may explain the mixed results observed in various studies that focus on a single organ system.
It is now well established that the neuroendocrine system and the immune system closely interact. Psychological stress can modulate immune reactivity via complex interactions involving the HPA axis as well as the autonomic nervous system (12, 13, 14). Cells of the immune system, like cells in other organ systems, express receptors for hormones and neurotransmitters (14, 15). Triggering of these receptors results in modulation of immune reactivity. As a model system to investigate the integrity of neuroendocrine regulation we chose cells of the immune system that are easily accessible in the peripheral blood and can be studied ex vivo.
We determined the sensitivity of the immune system to regulation by the glucocorticoid agonist dexamethasone and the ß2-adrenergic receptor agonist terbutaline. It has been well established that glucocorticoid receptor agonists will inhibit the proliferative response of T cells (16, 17, 18). Therefore, we determined the effect of dexamethasone on T-cell proliferation in healthy individuals and in CFS patients. ß2-adrenergic receptor agonists are known to regulate cytokine production by monocytes (19, 20, 21). Thus, we examined changes in lipopolysaccharide (LPS)-induced production of the cytokines tumor necrosis factor (TNF)- and interleukin (IL)-10 in the presence of increasing concentrations of terbutaline. We also examined baseline levels of epinephrine and norepinephrine in the same blood samples. In addition, plasma cortisol and ACTH levels before and after infusion of CRF were determined.
Patients and Methods
Fifteen girls with CFS, according to the criteria defined by the Centers for Disease Control and Prevention, with a substantial decrease in activity level and no primary psychological morbidity were asked to enter our study. The CFS patients included in our study were not taking any medication at the time of the study or within 6 weeks before the study. Patients with a psychiatric history were excluded. The body mass index of patients was significantly higher in CFS patients than in controls. Control individuals were recruited from healthy schoolmates of the patients of similar age and the same sex. An iv line was inserted into the underarm between 0830 and 0900 h. After a 60-min rest, a blood sample was drawn for analysis of plasma catecholamines and for determination of receptor sensitivity.
The CRH infusion was done between 1300 and 1400 h. The experimental protocol was approved by the medical ethical committee of the Wilhelmina Children Hospital. Written informed consent was obtained from parents and from the children.
Ex vivo response of peripheral blood cells to dexamethasone Whole blood was diluted 1:10 in medium [RPMI 1640 (Gibco, Grand Island, NY) supplemented with 2 mM glutamine, 100 U/mL penicillin, and 100 µg/mL streptomycin]. Diluted blood (100 µL) was cultured for 96 h in round bottom 96-well plates (Nunc, Glostrup, Denmark) with 25 µL phytohemagglutinin (PHA) (HA 15; Murex Diagnostics, Dartford, UK), final concentration 25 µg/mL, and 25 µL DEX in the concentrations indicated. At 16–18 h before the end of the culture, 1 µCi (37 kBq) [3H]-thymidine was added. At the end of the culture period, cells were harvested by the use of an automated cell harvester, and incorporated radioactivity was determined in a liquid scintillation counter.
Ex vivo response of peripheral blood cells to terbutaline Whole blood was diluted 1:10 in medium, and 100 µL diluted blood was cultured with 50 µl LPS [Escherichia coli (DIFCO Laboratories, Detroit, MI); final concentration 2 ng/mL] and 50 µL medium or the ß2-adrenergic receptor agonist terbutaline (Sigma Chemical Co., St. Louis, MO). After 18 h of culture at 37 C, supernatants were harvested and stored at -80 C until analysis. TNF- and IL-10 levels were determined by enzyme-linked immuosorbent assay (Pelikine; CLB, Amsterdam, The Netherlands).
Determination of plasma
Two milliliters of blood were collected on ice in 0.25 mol/L EGTA and 0.2 mol/L glutathione. Plasma samples were stored at -80 C. Catecholamines were determined by high-performance liquid chromatography according to the method described by Willemsen et al. (22). The detection limits were: adrenaline, 2 pg/mL; and noradrenaline, 2 pg/ml; CV, <10%.
CRH induced changes in
ACTH and cortisol
CRH (100 µg) was infused via an iv line between 1300 and 1400 h. Before and at various time points after infusion of CRH, blood was collected in ethylenediaminetetraacetate-coated tubes on ice. Plasma cortisol was determined using a fluorescence polarization immunoassay (Abbott Laboratories, Abbott Park, IL) (detection limit, 0.64 µg.dL; CV, <5%).
Plasma ACTH was determined by RIA, using antiserum from IgG Corporation USA and 125I-ACTH from CIS Bioindustries (France) (detection limit, 20 ng/L; CV, <8%).
Dose-response curves were analyzed by nonlinear regression using GraphPad Software, Inc. Prism 3.0 software. Data are expressed as mean and SEM. Two-tailed Student’s t tests were used to compare group differences. P < 0.05 was considered statistically significant.
We examined 15 patients diagnosed with CFS and 14 healthy controls. The mean age in the patient group was 15.8 ± 0.4 yr (range, 11–17) and in the control group 14.5 ± 0.6 yr (range, 10–17). Mean duration of disease was 21.8 ± 3.9 months (range, 6–48).
Ex vivo response of peripheral blood cells to dexamethasone Whole blood cultures were stimulated with the T cell mitogen PHA to induce proliferation and increasing concentrations of the glucocorticoid agonist dexamethasone. In the absence of dexamethasone, proliferative responses in CFS patients were higher than in healthy controls (CFS, 45,170 ± 5,063 cpm, n = 15; controls, 31,700 ± 3,852, n = 14; P = 0.044).
As expected, The addition of dexamethasone to the cultures resulted in a dose-dependent inhibition of the proliferative response (Fig. 1 ). Interestingly, however, the effect of dexamethasone is much less pronounced in cultures with cells from CFS patients (Fig. 1 ). The maximal effect of dexamethasone was 88.1 ± 3.1% inhibition of T-cell proliferation in healthy controls. In contrast, in CFS patients the maximal effect was 66 ± 4.9% inhibition, which is significantly lower (P = 0.001). The IC50 was similar in CFS patients and controls (CFS, 31 nM; controls, 47 nM).
Figure 1. Dexamethasone inhibition of T-cell proliferation. Whole blood cultures were stimulated with PHA (25 µg/mL) in the presence of increasing concentrations of dexamethasone. After 72 h, cultures were pulsed with 1 µCi 3H-thymidine. Cells were harvested after 96 h of culture, and incorporation of 3H-thymidine was determined as a measure of T-cell proliferation. Data are expressed as percentage of proliferation in the absence of dexamethasone and represent the mean and SEM. , controls, n = 14; •, CFS, n = 15.
Ex vivo response of
peripheral blood cells to terbutaline
To examine the sensitivity of the immune system to ß2-adrenergic regulation, we investigated the effect of the ß2-adrenergic receptor agonist terbutaline on cytokine production by peripheral blood cells. Whole blood cultures were stimulated with LPA for 18 h to induce monocyte cytokine production, and increasing concentrations of terbutaline were added to the cultures.
In the absence of terbutaline, TNF- production did not differ significantly between patients and controls (CFS, 473.9 ± 104.6 pg/mL, n = 15; controls, 799.1 ± 205.1 pg/mL, n = 14; P = 0.17). The data depicted in Fig. 2 clearly demonstrate that the addition of the ß2-adrenergic agonist results in inhibition of TNF- production. More importantly, our data demonstrate that the inhibitory effect of the ß2-adrenergic agonist on TNF- production is significantly lower in CFS patients than in controls. In control subjects, the maximal effect of terbutaline on TNF- production was 67 ± 1.3% inhibition. In CFS patients, maximal inhibition of TNF- production was only 37.1 ± 3.3% (P < 0.0001). There was no difference in IC50 between CFS and controls (CFS, 3.9 nM; controls, 8.2 nM).
Figure 2. ß2-adrenergic regulation of monocyte cytokine production. Whole blood cultures were stimulated with LPS (1 ng/mL) in the presence of the ß2-adrenergic agonist terbutaline for 18 h. Culture supernatants were harvested, and the amount of TNF- (A) and IL-10 (B) were determined. Data are expressed as percentage of cytokine levels in the absence of terbutaline and represent the mean and SEM. , controls, n = 14; •, CFS, n = 15.
ß2-adrenergic receptor agonists inhibit TNF- production, but enhance IL-10 production. If the decreased inhibition of TNF- production in CFS patients is the result of alterations in ß2-adrenergic receptor function, then we expect a smaller ß2-adrenergic agonist-induced increase in IL-10 production in CFS patients, as well. The data in Fig. 2 demonstrate that the ß2-adrenergic agonist is less capable of increasing IL-10 production in CFS patients than in controls (maximal increase: CFS, 45 ± 6.3; controls, 70 ± 5.8%; P = 0.007). In the absence of terbutaline, there was no difference in IL-10 production (CFS, 64.1 ± 11.4 pg/mL, n = 14; controls, 44.3 ± 8.7 pg/mL, n = 13; P = 0.18).
Plasma adrenaline and
Plasma noradrenaline levels in CFS patients did not differ from levels in healthy subjects (CFS, 1.47 ± 0.1 nmol/L, n = 14; control, 1.46 ± 0.2 nmol/L, n = 14; P = 0.98). There was a statistically significant increase in plasma adrenaline levels in CFS patients as compared with controls (CFS, 0.14 ± 0.03 nmol/L, n = 14; control, 0.07 ± 0.01 nmol/L, n = 14; P = 0.04).
Reactivity of the
At baseline, plasma ACTH and cortisol levels were similar in patients and controls. Plasma cortisol levels were similar in patients and controls (cortisol: CFS, 0.28 ± 0.03 µmol/L, n = 15; control, 0.29 ± 0.03 µmol/L, n = 14; P = 0.85. ACTH: CFS, 42.7 ± 4.8 ng/L, n = 15; control, 35.4 ± 3 ng/L, n = 14; P = 0.21). The data in Fig. 3 show that the CRH-induced increase in plasma ACTH and plasma cortisol is also similar in CFS patients and controls.
Figure 3. CRH-induced ACTH and cortisol in plasma. CFS patients (•, n = 15) and controls ( , n = 14) were infused with 100 µg CRH. Plasma ACTH (A) and plasma cortisol (B) were determined as described in Patients and Methods. Data represent the mean and SEM.
The pathophysiology of CFS is poorly understood. Research over the past 10 yr indicates that the syndrome cannot be explained by defects in one single organ system. The present study did not aim at defining alterations in one or the other system, but was designed to investigate the integrity of interorgan communication in CFS patients.
As a model system for interorgan communication, we chose the interaction between neuroendocrine factors and the immune system. The reactivity of the immune system can be tested ex vivo, and modulatory effects of glucocorticoids, as well as of ß2-adrenergic receptor agonists, have been clearly defined. Our results demonstrate that in adolescents with CFS, communication between neuroendocrine system and immune system is altered. In ex vivo studies, using peripheral blood of CFS patients, we demonstrated that the sensitivity of the immune system to regulation by neuroendocrine factors is decreased. T-cell proliferation is less sensitive to the inhibitory effects of dexamethasone, and monocyte cytokine production is relatively resistant to the modulatory effects of a ß2-adrenergic receptor agonist.
Interestingly, the decreased sensitivity to GC and to a ß2-adrenergic receptor agonist becomes apparent as a decreased maximal effect rather than as a change in the EC50. These results suggest that either the number of functional receptors is reduced or that the transduction of the signal from the receptor to the intracellular effector system is diminished in CFS. A reduced number of functional receptors is often associated with high plasma levels of the hormone.
In that case, the receptor is already occupied by hormone in vivo, and a lower number of receptors is available for exogenous ligand added ex vivo. However, in our study group, we have no evidence for disturbances in plasma cortisol that could explain the relative resistance to dexamethasone of T cells from CFS patients on this level. Baseline cortisol and CRH-induced increases in cortisol were similar in CFS patients and healthy subjects. Moreover, plasma ACTH levels before and after CRH infusion are similar in CFS and controls.
Therefore, we conclude that there are no major abnormalities in the reactivity of the HPA-axis in adolescents with CFS. In line with our data, baseline cortisol and ACTH in adults with CFS were not significantly different from controls in a number of studies (23). Demitrack et al. (10) also reported normal baseline cortisol levels, however, decreased 24-h excretion of cortisol in urine and a blunted response to infusion with CRH in a group of adults with CFS. It is possible that we do not observe changes in HPA-axis reactivity in our group of CFS patients because we are studying a much younger population. In our group of adolescents with CFS, mean age was 15.8 yr, whereas Demitrack et al. (10) studied adults with a mean age of 36.9 yr. Moreover, mean duration of disease in the adult study was 7.2 yr, whereas in our patient group mean duration of disease was less than 2 yr (10).
Our present data showing that cells of CFS patients are relatively resistant to a glucocorticoid receptor agonist may be of interest in view of the limited effect of hydrocortisone treatment in patients with CFS who do have a reduced activity of the HPA-axis (11, 24). If the resistance to GC agonist in CFS is a more generalized phenomenon, then normalization of plasma GC levels may not be sufficient to restore communication.
Alterations in neuroendocrine-immune communication are not specific for CFS. In a previous study, we demonstrated that the maximal effect of a ß2-adrenergic agonist on TNF- production is increased in patients with rheumatoid arthritis (25). Interestingly, peripheral blood cells of rheumatoid arthritis patients are more sensitive to the ß2-adrenergic receptor agonist, whereas in CFS we observe decreased reactivity. In addition, in both diseases we did not observe alterations in EC50, but only changes on the level of the maximal effect of the agonist (25).
The increased sensitivity of peripheral blood cells of rheumatoid arthritis patients to ß2-adrenergic modulation seemed to be associated with decreased expression of GRK-2, an intracellular kinase that plays a major role in receptor desensitization (25). Thus, altered sensitivity to a ß2-adrenergic agonist can be due to a change in the coupling efficiency of the receptor, a process in which GRK-2 plays a crucial role (26). In animal models it has been shown that chronic infusion of a ß2-adrenergic agonist will result in increased levels of GRK-2 and concomitantly in relative resistance to regulation by ß2-adrenergic agonists (27, 28).
It is conceivable that in CFS the decreased reactivity of the immune system to a ß2-adrenergic agonist is associated with increased levels of this kinase since plasma adrenaline levels are increased in CFS patients.
The question remains how the altered sensitivity of the immune system for neuroendocrine regulation developed. Was it a preexisting condition or is it the result of the disease? It may well be possible that the altered neuroendocrine-immune communication is associated with a high level of psychological stress. We know that the psychological stress of bereavement results in a significant decrease in the sensitivity of the immune system to dexamethasone (manuscript in preparation). A high level of psychological distress has not only been reported in adults, but also in adolescents with CFS (29, 30, 31).
Interestingly, a large study on postinfection fatigue shows that there is an association between psychological distress and fatigue prior to viral infection and the likelihood to develop chronic fatigue later on (32). We hypothesize that the abnormalities in neuroendocrine-immune communication in chronic fatigue syndrome result from the level of preexisting psychological distress and a precipitating event (e.g. a viral infection).
In summary, the present study demonstrates that the interaction between neuroendocrine mediators and a target system, the immune system, is disturbed in CFS. Additional studies should be performed to get insight in the role of these abnormalities in the pathophsyiology of CFS.
We gratefully acknowledge Marijke Tersteeg and Jitske Zijlstra for excellent technical assistance. Footnotes
(1) Supported by a grant of the "ME-fonds" of The Netherlands. Received August 25, 1999. Revised October 21, 1999. Accepted October 25, 1999.
1. Fukuda K, Straus SE, Hickie I, Sharpe MC, Dobbins JG, Komaroff A. 1994 The chronic fatigue syndrome: a comprehensive approach to its definition and study. Ann Intern Med. 121:953–959.[Medline]
2. Strauss SE, Tosato G, Armstrong G, et al. 1985 Persisting illness and fatigue in adults with evidence of Epstein-Barr virus infection. Ann Intern Med. 102:7–16.[Medline]
3. DeFreitas E, Hilliard B, Cheney PR, et al. 1991 Retroviral sequences related to human lymphotropic virus type II in patients with chronic fatigue immune dysfunction syndrome. Proc Natl Acad Sci USA. 88:2922–2926.[Abstract]
4. Buchwald D, Cheney PR, Peterson DL, et al. 1992 A chronic illness characterized by fatigue, neurologic and immuologic disorders and active human herpesvirus type 6 infection. Ann Intern Med. 116:103–113.[Medline]
5. Krilov L, Fisher M, Friedman S, Reitman D, Mandel FS. 1998 Course and outcome of chronic fatigue in children and adolescents. Pediatrics. 102:360–366.[Abstract/Free Full Text]
6. Marshall G. 1999 Report of a workshop on the epidemiology, natural history, and pathogenesis of chronic fatigue syndrome in adolescents. J Pediatr. 134:395–405.[Medline]
7. Caligiuri M, Murray C, Buchwald D, et al. 1987 Phenotypic and functional deficiency of killer cells in patients with chronic fatigue syndrome. J Immunol. 139:3306–3313.[Abstract/Free Full Text]
8. Gold D, Bowden R, Sixbey J, et al. 1990 Chronic fatigue: a prospective clinical and virologic study. J Am Med Assoc. 264:48–53.[Abstract]
9. Whiteside TL, Friberg D. 1998 Natural killer cells and natural killer cell activity in chronic fatigue syndrome. Am J Med. 105:27s–34s.[Medline]
10. Demitrack M, Dale JK, Straus SE, Laue L, Listwak SJ, Kruesi MJP. 1991 Evidence for impaired activation of the hypothalamic-pituitary-adrenal axis in patients with chronic fatigue syndrome. J Clin Endocrinol Metab. 73:1234.
11. McKenzie R, O’Fallon A, Dale J, et al. 1998 Low-dose hydrocortisone for treatment of chronic fatigue syndrome. A randomnized controlled trial. J Am Med Assoc. 280:1061–1066.[Abstract/Free Full Text]
12. Kiecolt-Glaser JK, Garner W, Speicher CE, et al. 1984 Psychosocial modifiers of immunocompetence in medical students. Psychosom Med. 46:7–14.[Abstract]
13. Sheridan JF, Dobbs C, Brown D, Zwilling B. 1994 Psychoneuroimmunology: stress effects on pathogenesis, and immunity during infection. Clin Microbiol Rev. 7:200–212.[Abstract]
14. Ader R, Felten DL, Cohen N. 1991 Psychoneuroimmunology, 2nd ed. San Diego: Academic Press.
15. Blalock JE, Smith EM. 1985 A complete regulatory loop between the immune and neuroendocrine systems. Fed Proc. 44:108–112.[Medline]
16. Gillis S, Wong-Staal F, Crabtree GR, Smith KA. 1994 Glucocorticoid-induced inhibition of T cell growth factor production. I. The effects on mitogen-induced lymphocyte proliferation. J Immunol. 123:1624–1629.[Medline]
17. Northrop JP, Crabtree GR, Mattila PS. 1992 Negative regulation of interleukin-2 transcription by the glucocorticoid receptor. J Exp Med. 1755:1235–1245.
18. Kavelaars A, Zijlstra J, Bakker JM, et al. 1995 Increased dexamethasone sensitivity of neonatal leukocytes: different mechanisms of glucocorticoid inhibition of T cell proliferation in adult and neonatal cells. Eur J Immunol. 25:1346–1351.[Medline]
19. Kavelaars A, Van de Pol M, Zijlstra J, Heijnen CJ. 1997 ß2-adrenergic receptor activation stimulates the production of IL-8 by human monocytes. J Neuroimmunol. 77:211–216.[Medline]
20. Van der Poll T, Coyle SM, Barbosa K, Braxton CC, Lowry SF. 1996 Epinephrine inhibits tumor necrosis factor- and potentiates interleukin-10 production during human endotoxemia. J Clin Invest. 97:713–719.[Abstract/Free Full Text]
21. Severn A, Rapson NT, Hunter CA, Liew FY. 1992 Regulation of tumor necrosis factor production by adrenaline and ß-adrenergic agonists. J Immunol. 148:3441–3445.[Abstract/Free Full Text]
22. Willemsen JJ, Ross HA, Jacobs MC, et al. 1995 Highly sensitive and specific HPLC with fluorometric detection for determination of plasma epinephrine and norepinephrine applied to kinetic studies in human. Clin Chem. 41:1455–1460.[Abstract]
23. MacHale SM, Cavanagh JT, Bennie J, Carroll S, Goodwin GM, Lawrie SM. 1998 Diurnal variation of adrenocortical activity in chronic fatigue syndrome. Neuropsychobiology. 38:213–217.[Medline]
24. Cleare AJ, Heap E, Malhi GS, Wessely S, O’Keane V, Miell J. 1999 Low-dose hydrocortisone in chronic fatigue syndrome: a randomnised cross over trial. Lancet. 353:455–458.[Medline]
25. Lombardi MS, Kavelaars A, Schedlowski M, et al. 1999 Decreased expression and activity of G protein-coupled receptor kinases by peripheral blood mononuclear cells of patients with rheumatoid arthritis. FASEB J. 13:715–725.[Abstract/Free Full Text]
26. Chuang TT, Iacovelli L, Sallese M, De Blasi A. 1996 G-protein coupled receptors:heterologous regulation of homologous desensitization, and its implications. Trends Pharmacol Sci. 17:416–421.[Medline]
27. Iaccarino G, Tomhave ED, Lefkowitz RJ, Koch WD. 1998 Reciprocal in vivo regulation of myocardial G protein-coupled receptor kinase expression by ß-adrenergic receptor stimulation and blockade. Circulation. 98:1783–1789.[Abstract/Free Full Text]
28. Werner C, Poller U, Brodde OE. 1999 Chronic terbutaline treatment desensitizes ß-adrenergic inhibition of lymphocyte activation in human volunteers. J Auton Pharmacol. 17:237–242.
29. Carter B, Edwards J, Kronenberger W, Michalczyk L, Marshall G. 1995 Case control study of chronic fatigue in pediatric patients. Pediatrics. 95:179–186.[Abstract]
30. Walford G, McNelson W, McCluskey D. 1993 Fatigue, depression, and social adjustment in children with chronic fatigue syndrome. Arch Dis Child. 68:384–388.[Abstract]
31. Pelcovitz D, Septimus A, Friedman S, Krilov L, Mandel F, Kaplan S. 1995 Psychosocial correlates of chronic fatigue syndrome in adolescent girls. J Develop Behav Pediatr. 16:333–338.
32. Pawlikowska T, Chalder T, Hirsch S, Wallace P, Wright D, Wessely S. 1994 A population-based study of fatigue and psychological distress. Br Med J. 308:743–746.[Abstract/Free Full Text]