2005 ME/CFS Research Forum

Adelaide Research Network 3 - 4 June 2005

Convenor: Alison Hunter Memorial Foundation

Emeritus Professor Barrie Marmion AO DSC MD FRACOP FRCPath
Q fever Research Group
IMVS & Hanson Institute
Adelaide Australia


Oral presentations

Acute & chronic Q Fever and its post-infection fatigue syndrome

Q fever is caused by a small obligate intracellular bacterium, Coxiella burnetii. In the body it grows in cells of the macrophage lineage - ie. monocytes, alveolar macrophages, tissue macrophages, and perhaps such macrophage analogues as the mesangium, synovial A cells, microglia etc. It is highly adapted to survive and replicate at the low pH of the macrophage phagolysosome, eventually forming cytoplasmic microcolonies. The interaction of the coxiella with the macrophage, a key cell for the generation and modulation of cellular immune responses has significant implications for acute primary Q fever and its resolution, and for development of chronic complications.

Clinically, acute primary Q fever mostly starts abruptly after an incubation period 15-25 days. There is high fever, rigors, profuse sweats, incapacitating fatigue and prostration, a savage headache, severe myalgia and arthralgia, photophobia, nausea, loss of appetite and disordered cerebral function--the last sometimes mimicking meningoencephalitis. The majority of patients have abnormal liver function tests and a small proportion develop 'atypical pneumonia'.

The acute illness lasts from 1-6 weeks or so; the more severe cases often experience substantial loss of weight. During the first 7-10 days of the acute illness the coxiella can be detected in the blood by animal inoculation or PCR assay for genomic DNA. However, the bacteraemia diminishes once antibody appears at 8-12 days after onset of illness. The acute phase of the disease is accompanied by raised inflammatory markers (eg. CRP, ESR) and sometimes by autoantibody formation including anticardiolipin. The general symptoms of the acute attack are considered to reflect the acute phase cytokine cascade from the developing cell-mediated immune response, rather than a direct toxic effect of the coxiella. (Cytokines are very potent immunological mediators that communicate between cells in the immune system). It is possible that organ-based changes such as those in hepatocyte function are also cytokine-mediated although small granulomas of activated macrophages are observed in the liver and other tissues (eg. bone marrow). Note that the range of general acute phase symptoms is essentially the same, for example, in

Q fever, Legionella spp and Mycoplasma pneumoniae infections, although there may be differences in the involvement of different organ systems.

This is not surprising as the acute phase cytokine response is involved in each case and is responsible for a major part of the symptoms.

Convalescence may be interrupted by febrile relapses. Coxiellas do not appear to be present in the blood in these episodes which may respond to corticosteroid therapy.

With more severe cases of acute Q fever it is not unusual for convalescent patients to complain of continuing inappropriate fatigability, Myalla and joint pain, nausea and ethanol intolerance, night sweats, interrupted sleep patterns, cognitive dysfunction and loss of libido. However, such patients are usually afebrile. This convalescent malaise and debility may last 6-9 months before complete recovery. It is the patients who fail to recover and in whom the debility lasts beyond a year that constitute the group regarded as having the post Q fever fatigue syndrome (QFS). This affects ~10% of patients exposed once but with a higher rate ~15-20% in abattoir populations in which there is frequent re-exposure to C.burnetii.

QFS generally follows on from the acute illness without a break but sometimes a patient will recover and some months later develop severe QFS. Epidemiological evidence is sometimes adequate to indicate that they have been re-exposed to the coxiella but often this is difficult to document with certainty.

Other chronic sequels to Q fever are (i) subacute Q fever endocarditis (~2%) which may be manifest around 2-5 years after the acute illness, but may have a longer incubation period; (ii) granulomatous lesions resembling osteomyelitis near joints or in vertebrae although less inflamed and more indolent than those due to the usual pyogenic bacteria; (iii) granulomas developing in testes and genital tract or other organs.

As with cattle, sheep and goats and smaller mammals such as dogs and cats, women infected in early pregnancy or even before conception may, late in pregnancy develop placentitis with numerous coxiellas, and deliver an infected foetus or neonate.

All of these sequelae strongly indicate persistence of the coxiella after an initial infection although the site of persistence and its control mechanisms are only now being studied in depth.

We first observed patients with the post Q fever fatigue (QFS) syndrome in the 1980s during our clinical trials of Q fever vaccine in South Australian abattoirs. The workers described second or repeated attacks of 'Q fever'. The conventional wisdom was (and is) that an acute attack of Q fever confers solid immunity to subsequent Q fever and by and large this is correct. However, it became apparent that the abattoir workers were describing a condition with a number of symptoms of an acute Q fever but without the high fever and the very severe incapacity of the original illness that confined them to bed. About the same time Dr Jon Ayres, a respiratory physician in Birmingham UK noted in a letter to the Lancet that some patients in an outbreak of Q fever in the general population of Solibull, South Birmingham remained ill with similar symptoms after their initial illness. We then made some systematic observations in 3 groups of abattoir workers - comparing those who had acute Q fever, those who had been subclinically infected and those who had been vaccinated - with a control group of seronegative Telecom workers. This showed a statistically significant association between a clinical, overt attack of Q fever and the symptom complex of QFS (ie., inappropriate fatigue on exertion with slow recovery, muscle aches and pains, muscle fasciculation, night sweats, ethanol and food intolerance, constant headaches, photophobia, disturbed sleep patterns, loss of libido, problems with short term memory etc). The findings were published in back to back letters to the Lancet from our group and that of Dr Jon Ayres (Lancet 1996 347 977-8, 978-9). Patients also described feeling as though they had a fever but on taking their temperature it was rarely raised more than half a degree. The symptoms the abattoir workers experienced were in essence a 'down sized' version of the original acute phase reaction in the primary Q fever illness but without the fever.

At the stage of the survey few if any of the abattoir workers with QFS had claimed compensation for the "second attacks" of Q fever. A history of psychiatric morbidity before the initial attack of Q fever was rare. Most psychiatrists who saw patients with fully developed QFS considered that their clinical state related in some unexplained way to the original attack of Q fever rather than to a superimposed, unrelated, attack of "depression" due to other causes.

Our hypothesis to explain QFS is that it is a failure of a minority of patients to 'switch off' or down regulate elements of the original cellular immune (cytokine) response from the primary Q fever and, further, that they continue to generate the cytokines in response to continued persistence of the coxiella or its undegraded antigens. These possibilities were investigated in two ways.

First, we examined the cytokine responses of peripheral blood mononuclear cells from QFS patients and controls stimulated in short term culture with Q fever antigens and other unrelated antigens such as measles.

The results showed (Penttila et al Cytokine dysregulation in the post-Q fever-fatigue syndrome. QJM ed. 1998; 91 549-560) that QFS patients had a hyperactive response to Q fever antigens and liberated larger amounts of the cytokine IL-6 than controls. A proportion also formed more IFN gamma and less IL-2. (Note that the therapeutic administration of IL-6 and IFN gamma produces symptoms of the fatigue state (CFS): see Penttila et al., 1998)

Second, we studied persistence of the coxiella in Australian patients and those in the Birmingham outbreak. This resulted in the somewhat unexpected finding that most patients who have had acute Q fever had evidence of persistence of the coxiella in their bone marrow; irrespective of their clinical state. However patients with QFS more often had the coxiella in peripheral blood cells when compared with those who had made an uncomplicated recovery and remained well. In other words the level of persistence was less well controlled in QFS (Harris et al. Long term persistence of Coxiella burnetii in the host afterprimary Q fever Epidemiol. Infect.2000; 124: 543-9. Marmion et al. Long term persistence of Coxiella burnetii after acute primary Q fever QJ Med 2005; 98 7-20.).

It was further hypothesised that the difference in controlling the level of persistence of the coxiella in the QFS compared with asymptomatic, recovered individuals might be related to allelic variations in their immune response genes. Preliminary investigations indicate that QFS patients have an unusual, statistically significant preponderance of the HLA DR B1*11 subtype and of a particular variant in a control gene for interferong. (Helbig et al. Immune response genes in the post-Q fever fatigue syndrome, Q fever endocarditis and uncomplicated acute primary Q fever. QJ Med 2005 in press July).

The effect of a single gene variation as a sole factor in influencing the course of infection is rare; most effects are multigenic. Further investigations are therefore required to confirm the work and to identify other gene variants.

The preliminary observations do, however, suggest that the processing and presentation of C. burnetii antigens to the immune system may differ in the QFS patients and those who remain asymptomatic after recovery. This may also provide some insight into the difficult problem of QFS/CFS.

In summary, in general we see Coxiella burnetii - host relations as bipolar and conditioned at least partly, by the immunogenetic background of the individual and possibly by variations in the coxiella antigens. At one pole is subacute Q fever endocarditis with anergy, numerous organisms in valve vegetations and high antibody levels. At the other pole there is a hypersensitivity state (QFS) with few organisms in bone marrow and PBMC, and lower levels of antibody. In addition there are recrudescences in pregnancy and episodic granulomatous involvement of testes, bone and other organs. There is some very recent evidence that fine antigenic differences exist between strains of Coxiella burnetii isolated from acute and chronic disease. It is conceivable that these add a further dimension of complexity to the variations in genetic background of the infected subjects.


Q fever is an acute febrile disease due to a small intracellular bacterium acquired mainly from infected cattle sheep or goats. About 10% of patients develop a post infection fatigue syndrome resembling the chronic fatigue syndrome (CFS). In the subgroup of CFS following Q fever the determinants appears to be long term persistence of the organism (Coxiella burnetii) and an unregulated cellular immune hypersensitivity stimulated by the persistent organism or its antigens. The failure to down regulate the aberrant immune response may be related to variation in host immune response genes but this requires further investigation.

Return to Top


Alison Hunter Memorial Foundation
PO Box 6132 North Sydney 2059 Australia
Phone/Fax +61 2 9958 6285

About Us
About ME/CFS
Medical Politics