Source: Clinical Infectious Diseases
        Vol. 45, #6, pp 732-735
Date:   September 15, 2007
URL:    http://www.journals.uchicago.edu/CID/journal/available.html


Postinfective Fatigue Syndrome Is Not Associated with Altered Cytokine
Production
----------------------------------------------------------------------
Ute Vollmer-Conna(1), Barbara Cameron(2) Dusan Hadzi-Pavlovic(1), Kristi
Singletary(4), Tracey Davenport(3), Suzanne Vernon(4), William C. Reeves(4),
Ian Hickie(3), Denis Wakefield(2), and Andrew R. Lloyd(2), for the Dubbo
Infective Outcomes Study Groupa
1 School of Psychiatry and
2 Centre for Infection and Inflammation Research, School of Medical Sciences,
  University of New South Wales, and
3 Brain & Mind Research Institute, Sydney University, Sydney, Australia; and
4 Division of Viral and Rickettsial Diseases, Centers for Disease Control
  and Prevention, Atlanta, Georgia
* Reprints or correspondence: Dr. Ute Vollmer-Conna, Dept. of Human Behaviour/
  Psychiatry, University of New South Wales, UNSW Sydney 2052, New South 
  Wales, Australia (ute@unsw.edu.au).
Study group members are listed at the end of the text.

Received 3 April 2007;
accepted 15 May 2007;
electronically published 6 August 2007.


Peripheral blood specimens and clinical data were obtained over a 12-month
period from subjects in the Dubbo Infection Outcomes Study to examine
cytokine production in postinfective fatigue syndrome. Ex vivo production of
8 cytokines was examined in 22 case patients and in 42 control subjects who
recovered promptly. No significant differences were found. Ongoing production
of the cytokines examined does not play a role in postinfective fatigue
syndrome.

Despite intensive research efforts, the pathophysiology of the enigmatic
clinical disorder chronic fatigue syndrome (CFS) remains obscure, and
effective therapies are not available [1]. Prospective cohort studies have
empirically verified the existence of a postinfective fatigue syndrome (PIFS)
consistent with CFS that is triggered by infection with Epstein-Barr virus
(EBV), Ross River virus (RRV), or Coxiella burnetii (the causative agent of Q
fever) [2-5].

Hypotheses that abnormalities in the immune response underpin the
pathogenesis of CFS have been extensively explored. However, a systematic
review suggested that reported alterations in T cell responses, cytokine
levels, and natural killer cell activity were inconsistent and rarely
correlated with the clinical condition [6]. Substantial heterogeneity among
patients fulfilling the diagnostic criteria for CFS [7, 8] is the likely
reason for such inconsistent findings. Accordingly, we established a disease
model for the onset and evolution of CFS, by means of a prospective cohort
study of subjects observed from the onset of documented acute EBV infection,
RRV infection, or Q fever: the Dubbo Infection Outcomes Study (DIOS) [5]. In
this study, PIFS was documented in 27% of subjects at 3 months, with 11%
fulfilling diagnostic criteria for CFS at 6 months after detailed medical and
psychiatric evaluation, to exclude alternative causes for illness. The PIFS
illness was stereotyped across infections, suggesting that the host response,
rather than microbiological factors, determined ongoing symptoms.

In subjects from DIOS studied early in the febrile phase of illness, the
severity of symptoms, including fatigue, correlated with levels of the
proinflammatory cytokines IL-1beta and IL-6, in unstimulated PBMC cultures,
indicative of cytokines "spontaneously" released ex vivo [9]. Thus, the
present study aimed to analyze cytokine production in patients with PIFS and
matched control subjects who had recovered uneventfully from the same
infection, to resolve whether ongoing, aberrant cytokine production mediates
prolonged fatigue states.


Subjects and methods.

Participants were enrolled after presentation with acute febrile illness and
detection of IgM antibodies against EBV, RRV, or C. burnetii. These
provisional serologic diagnoses were confirmed by testing acute- and
convalescent-phase serum samples [5].

Subjects were assessed at 1, 2, 3, 6, and 12 months after the onset of
infection. At each visit, physical and psychological health was assessed, and
a blood sample was collected. The 34-item Somatic and Psychological Health
Report [10] was used to monitor symptoms. A score of >=3 (maximum score, 12)
on the empirically derived subscale SOMA was used to record PIFS, because
this reliably predicts disability and reflects patients' and doctors' reports
of reasons for presentation to primary care [10]. Subjects were classified as
having provisional PIFS if their SOMA scores at all time points up to and
including 3 months exceeded this threshold. When symptoms persisted beyond 6
months, structured medical and psychiatric assessments and laboratory
investigations were undertaken to exclude alternative explanations for
ongoing illness and to designate CFS diagnosis (termed "confirmed PIFS") [1].
The items "days in bed" and the number of "days out of role" (Brief
Disability Questionnaire) served to index functional impairment [11]. The
factor score coefficients (symptom weights) from the original study [5] were
applied to each subject's responses on the

Somatic and Psychological Health Report to quantify illness severity in 6
domains: acute sickness, irritability, fatigue, musculoskeletal pain,
neurocognitive disturbance, and mood disturbance.

The participants were 22 subjects with confirmed PIFS (11 patients with EBV
infection, 6 with RRV infection, and 5 with Q fever) and 42 control subjects
(17 patients with EBV infection, 14 with RRV infection, and 11 with Q fever)
matched for age from the same cohort who had recovered within 6 weeks of
symptom onset. Written informed consent was obtained from all subjects. The
relevant institutional review boards approved the study.

Blood samples were collected and processed under strict endotoxin-minimized
conditions. PBMCs were separated (Lymphoprep; AXIS-SHIELD) and
cryopreserved in RPMI (Trace Biosciences) with 10% dimethylsulfoxide (Sigma)
and 50% autologous plasma, and aliquots were stored in liquid nitrogen. 
Serum was separated and stored in aliquots at -80 C.

After being thawed, PBMCs were resuspended in RPMI (GIBCO) supplemented with
10% heat-inactivated fetal bovine serum (Trace Biosciences) at 2x10^6
PBMCs/mL. The cell suspension was dispensed into 96-well plates (Nunc) in
medium only (unstimulated cultures), lipopolysaccharide (10 ng/mL; 
Salmonella typhimurium; Sigma), or mouse anti-human, anti-CD3, and anti-CD28
Dynabeads added at a ratio of 4 beads to 1 cell ratio (Dynal). Plates were
held for 24 h at 37 C. Supernatants were stored at -80 C until being
subjected to the multiplex immunoassay. Concentrations of the cytokines
IL-1beta, IL-2, IL-4, IL-6, IL-10, IL-12, TNF-alpha, and IFN-gamma in serum
and culture supernatants were analyzed (BioPlex; BioRad).

The linear mixed-effects models procedure (in SPSS for Windows, version 14;
SPSS Software) was chosen to analyze the influence of PIFS status on
longitudinal patterns of symptom severity and cytokine production. Because
the PIFS illness was stereotyped across different infective triggers,
acute-phase cytokine production was comparable across the 3 infections, and
neither psychological nor microbial factors were predictive of PIFS [5, 9],
the data set was collapsed across infections for analysis. Student's t test
was used for group comparisons of age, duration, and disability, and the
chi^2 test was used to assess differences in the sex and case/control
distributions.


Results.

Subjects with PIFS did not differ from the control subjects with regard to
age (mean p/m SD, 32 p/m 14 vs. 32 15 years; P=.99) or sex (P=.34 for the
ratio of male to female subjects). Of the 64 participants, 28 were infected
with EBV, 20 were infected with RRV, and 16 had Q fever. The proportion of
PIFS cases to controls (0.66 to 0.34) did not differ between infections
(P=.76). The first assessment in this study (denoted as 1 month) occurred a
mean (p/m SD) of 29 p/m 8 days after onset of symptoms. This interval did not
differ between case patients and control subjects (P=.46). The mean duration
of the entire illness (p/m SD) was 40 p/m 19 weeks for case patients and 5
p/m 3 weeks for control subjects (P<.001). The mean number of days in bed
(p/m SD) for patients with PIFS was 18 p/m 15, compared with 7 p/m 8 days
among control subjects (P=.03). Subjects with PIFS also reported
significantly more days out of role (PIFS group, 44 p/m 31 days; control
subjects, 16 p/m 13 days; P=.01).

As expected, the analysis of symptom factor scores (figure 1A) established
that PIFS status was associated with more-severe symptoms in all 6 domains
(P>.001 for each). A significant main effect for time was evident for acute
sickness (P=.03), fatigue (P=.001), and musculoskeletal pain (P=.006),
indicating a decrease in symptom severity over time in these domains. Older
age was associated with higher levels of musculoskeletal pain (P=.003) and
neurocognitive disturbance (P=.02).

Serum cytokine levels were almost exclusively less than the detection limit
of the assay system (8-15 pg/mL) and did not differ between the groups (data
not shown); therefore, the analysis focused on culture supernatants.
Mixed-effects modeling did not reveal any between-group differences in
cytokine levels over time. These results were consistent across all culture
conditions (P>.05 for all) (figure 1B). Older age was associated with higher
levels of IL-4 (P=.02) in anti-CD3/anti-CD28-stimulated cultures and with
higher levels of IL-1beta (P=.003), IL-6 (P=.02), and TNF-alpha (P=.02) in
lipopolysaccharide-stimulated cultures.


Discussion.

This comprehensive analysis of 8 different cytokine responses in PIFS failed
to reveal any association with disease status, thus providing strong evidence
against the hypothesis that prolonged fatigue is associated with altered
cytokine production. These results are concordant with our recent finding
that specific antiviral immune responses in the EBV subcohort of the DIOS did
not reveal substantive differences between subjects with PIFS and those who
recovered promptly [12]. It remains possible that testing with other
cytokine-inducing stimuli and/or assessing a broader panel may implicate
other cytokines in PIFS.

Irrespective of PIFS status, older subjects reported more pain and
neurocognitive symptoms and showed an increased pro-inflammatory response to
lipopolysaccharide challenge, as well as a bias toward Th2-type lymphokine
production (IL-4) in anti-CD3/anti-CD28-stimulated cultures. Similar
responses have previously been attributed to immune senescence [13].

Consistent with earlier reports from DIOS, the present data indicate that
symptoms of PIFS originate in the acute phase of the infections and persist
over weeks to months [5, 12]. Acute infections are typically accompanied by a
cluster of stereotyped symptoms, including fever, fatigue, hypersomnia,
hyperalgesia, mood disturbance, and impaired concentration [9]. With the
exception of fever, it is these symptoms that typify the PIFS illness.
Accumulated animal data indicate that, in acute infection, these symptoms are
mediated by the action of proinflammatory cytokines on the brain [14]. We
have confirmed this finding in subjects from DIOS studied early in the
febrile phase [9]. Nevertheless, no such association was found with
comparable symptoms in the prolonged course of PIFS.

Although the present study cannot exclude a role for other untested factors
or for cytokines at very low levels or produced in specific microenvironments
[15, the evidence suggests that immunologic stressors such as acute
infections constitute a "trigger" rather than the "driver" of symptoms in
PIFS. Consequently, alternative hypotheses of pathogenesis need to be
explored, including persistent production of inflammatory mediators by
activated glial cells in the absence of ongoing peripheral stimuli or
sensitization of neural response pathways to physiological signals from the
periphery [15].


Study group members.

The additional investigators of the Dubbo Infection Outcomes Study Group
include (in alphabetical order): Heather Dunckley (Australian Red Cross Blood
Service, Sydney), Andrew Geczy (Australian Red Cross Blood Service, Sydney),
Ray Harris (University of South Australia, Adelaide), Rajiv Khanna
(Queensland Institute for Medical Research, Brisbane), Barrie Marmion
(Institute for Medical and Veterinary Science, Adelaide), and Bill Rawlinson
(Prince of Wales Hospital, Sydney).


Acknowledgments

We gratefully acknowledge the support of the general practitioners, the
public health unit, and the diagnostic pathology services in the Dubbo region
and the enduring cooperation of the participants in the research. We also
acknowledge the assistance provided by Dr. Rosane Nisenbaum in the
statistical analysis.

Financial support.
National Health and Medical Research Council of Australia Project Grants
(157092 and 157062); Meat & Livestock, Australia; and Centers for Disease
Control and Prevention (Cooperative Research Agreement U50/CCU019851-01).

Potential conflicts of interests.
All authors: no conflicts.


Figure caption

Figure 1.
A, Severity of illness in different symptom domains over a 12-month period
for subjects with postinfective fatigue syndrome (PIFS; shaded columns) and
control subjects who recovered uneventfully from the same infections (open
columns). B, Cytokine levels produced in unstimulated peripheral blood
mononuclear cell cultures collected over a 12-month period from subjects with
PIFS (shaded columns) and control subjects (open columns). Bars, median; box
extremities, 25th and 75th percentiles.


References

 1. Fukuda K, Straus SE, Hickie I, et al. The chronic fatigue syndrome: a
    comprehensive approach to its definition and study. International
    Chronic Fatigue Syndrome Study Group. Ann Intern Med 1994; 121:
    953-9.
 2. White PD, Thomas JM, Amess J, et al. Incidence, risk and prognosis
    of acute and chronic fatigue syndromes and psychiatric disorders after
    glandular fever. Br J Psychiatry 1998; 173:475-81.
 3. Buchwald DS, Rea T, Katon W, et al. Acute infectious mononucleosis:
    characteristics of patients who report failure to recover. Am J Med
    2000; 109:531-7.
 4. Candy B, Chalder T, Cleare AJ, et al. Predictors of fatigue following
    the onset of infectious mononucleosis. Psychol Med 2003; 33:847-55.
 5. Hickie I, Davenport TA, Wakefield D, et al. Viral and non-viral path-
    ogens precipitate post-infective and chronic fatigue syndromes: a pro-
    spective cohort study. BMJ 2006; 333:575-8.
 6. Lyall M, Peakman M, Wessely S. A systematic review and critical eval-
    uation of the immunology of chronic fatigue syndrome. J Psychosom
    Res 2003; 55:79-90.
 7. Wilson A, Hickie I, Hadzi-Pavlovic D, et al. What is chronic fatigue
    syndrome? Heterogeneity within an international multicentre study.
    Aust N Z J Psychiatry 2001; 35:520-7.
 8. Vollmer-Conna U, Aslakson E, White PD. An empirical delineation of
    the heterogeneity in chronic unexplained fatigue in women. Phar-
    macogenomics 2006; 7:355-64.
 9. Vollmer-Conna U, Fazou C, Cameron B, et al. Production of pro-
    inflammatory cytokines correlates with the symptoms of acute sickness
    behaviour in humans. Psychol Med 2004; 34:1289-97.
10. Hickie IB, Davenport TA, Hadzi-Pavlovic D, et al. Development of a
    simple screening tool for common mental disorders in general practice.
    Med J Aust 2001; 175:S10-7.
11. von Korff M, Ustun T, Ormel J, Kaplan I, Simon G. Self-report dis-
    ability in an international primary care study of psychological illness.
    J Clin Epidemiol 1996; 49:297-303.
12. Cameron B, Bharadwaj M, Burrows J, et al. Prolonged illness following
    infectious mononucleosis is associated with altered immunity but not
    elevated viral load. J Infect Dis 2006; 193:664-71.
13. Weng NP. Aging of the immune system: how much can the adaptive
    immune system adapt? Immunity 2006; 24:495-9.
14. Hart BL. Biological basis of the behavior of sick animals. Neurosci
    Biobehav Rev 1988; 12:123-37.
15. Pollmacher T, Haack M, Schuld A, et al. Low levels of circulating
    inflammatory cytokines - do they affect human brain functions? Brain
    Behav Immun 2002; 16:525-32.

--------
(c) 2007 Infectious Diseases Society of America