Q J Med 2002; 95: 539-546
© 2002 Association of Physicians
Long-term follow-up of patients from the 1989 Q fever outbreak: no evidence of excess cardiac disease in those with fatigue
From the Departments of 1 Respiratory Medicine, 2 Cardiology and 3 Microbiology, Birmingham Heartlands Hospital, Birmingham, UK
Received 3 August 2001 and in revised form 19 March 2002
 |
Summary |
|---|
 Top Summary IntroductionMethodsResultsComparison of fatigue-free cases...DiscussionReferences |
|---|
Background: In 1989, an outbreak of Q fever (C. burnetii infection) with 147 confirmed cases occurred in Solihull, West Midlands. Three patients developed cardiomyopathy in the subsequent 10 years. The cohort has been followed up with respect to the development of fatigue and, in this instance, cardiac effects after the original infection.
Aim: To determine whether persisting fatigue after Q fever represented sub-clinical cardiomyopathy.
Design: Prospective follow-up study.
Methods: All traceable subjects from the original outbreak, and community age-, sex- and smoking-matched controls, were studied. Questionnaires for idiopathic fatigue, 12-lead ECG, echocardiography, spirometry and shuttle walk distance were undertaken, and a subset with CDC-defined chronic fatigue syndrome had gated cardiac scans.
Results: Of the original cohort, 19 had died, three had emigrated and 10 were untraceable. Of the remaining 115, 108 responded to a mailed questionnaire and 87 were investigated further, of whom 85 provided complete data. Two developed aortic valve vegetations, one of whom died. Chronic fatigue syndrome was found in 20% of cases and 5.3% of controls (including those with co-morbidities), falling to 8.2% and 0 when excluding those with co-morbidities. There were no significant differences in ECG and echocardiographic investigations or shuttle-walk distance between those with fatigue and those without. Six of the seven patients with CFS had gated cardiac scans: all were within normal limits.
Conclusions: These findings do not support the existence of a sub-clinical cardiomyopathy in the patients in this cohort who suffer from fatigue after acute Q fever, although endocarditis can occur after acute infection.
|
Introduction |
|---|
|
TopSummaryIntroductionMethodsResultsComparison of fatigue-free cases...DiscussionReferences |
|---|
Q fever (Coxiella burnetii infection) is rarely reported in urban areas in the UK, although studies from the 1940s onwards have shown that cattle and sheep are infected, and that rural outbreaks and sporadic cases are not uncommon.1 In humans, Q fever can either present as an acute illness with high fever and dramatic weight loss, or as a chronic infection, presenting as endocarditis, granulomatous hepatitis or osteomyelitis.2 It is a particularly infectious organism (the minimal infective dose being 1–5 organisms) and is usually transmitted following contact with parturient ewes or cows. In April 1989, the largest UK outbreak of Q fever, with 147 serologically confirmed cases, occurred in an area to the south of Birmingham.3,4 It was probably a windborne outbreak, possibly from a flock of lambing ewes. Eighty three percent of the affected population were males of working age, children being spared. The acute illness was severe,5,6 and many patients were hospitalized. This was mostly because of pneumonia, although there was an unusually high incidence of neurological symptoms in this group.6 However, it is likely that others, less severely affected, were not identified in our initial study. Follow-up at 6 months showed resolution of the chest X-ray changes in the majority of cases,5 but many patients complained of feeling persistently tired and breathless. Three people involved in the Q fever outbreak had reportedly developed cardiomyopathy, which was originally attributed by their own physician to the acute Q fever infection.
In Australia, a prolonged post-infection fatigue syndrome, termed Q fever fatigue syndrome (QFS), has been observed following acute Q fever in abattoir workers, farmers and other individuals in contact with livestock.7 The existence of this syndrome has been contested in the past, the picture being coloured by litigation and the fact that abattoir workers are likely to be exposed recurrently to Coxiella. Chronic fatigue syndrome (CFS) is a condition for which there is now an accepted definition, determined by an International Working Group on the basis of a symptom complex, termed the CDC definition.8 The working group identified as a separate entity the symptom complex of CFS following infection as ‘post-infective CFS‘, but there is still some resistance in some quarters to identifying cases of CFS as post-infective.9 As the symptoms of QFS are difficult to distinguish from those of CFS, study of QFS characterized by persistent fatigue following Coxiella infection may help in identifying mechanisms for CFS with no recognized infectious cause.
Because of sporadic reports of some patients from the 1989 outbreak having chronic symptoms, a questionnaire survey was conducted 5 years after the initial event, which showed that undue fatigue was reported by 69% of cases, significantly more than controls, although a surprising 51.5% of controls exhibited undue fatigue.10,11 A variety of other symptoms were significantly more common in cases, particularly sweating (53%; 32% in controls), breathlessness (51%; 31% in controls) and blurring of vision (34%; 18% in controls),11 thus confirming the Australian reports of QFS, but highlighting the low specificity of these symptoms given their high prevalence in the non-exposed population. In addition, we had been given information that three patients from the original outbreak had subsequently developed cardiomyopathy and one, endocarditis. We therefore hypothesized that a possible reason for the symptom complex of breathlessness and fatigue in the remaining patients with chronic symptoms after acute Q fever was a sub-clinical cardiomyopathy.
|
Methods |
|---|
|
TopSummaryIntroductionMethodsResultsComparison of fatigue-free cases...DiscussionReferences |
|---|
This study was part of a long-term follow-up of this cohort; details of the other components will be published elsewhere. As part of the 10-year follow-up, we compared the cardiac health of the Coxiella-exposed subjects to a matched control group. The study design included a nested case comparative study to determine whether cardiac indices, in particular radio-nuclide ejection fractions, were impaired in patients with chronic fatigue compared to those without chronic fatigue or in the Coxiella-exposed compared to the non-Coxiella-exposed controls. Coxiella-exposed subjects (defined as those showing either a single titre of >1:256 to the Phase 2 antigen or a >4-fold rise in titre at the time of the outbreak) were compared to age- and sex-matched controls identified from GP lists who had no serological evidence of previous Q fever. Controls were identified using randomized selection from all available individuals on two general practitioner lists from the broad geographical area of the outbreak. In the case-comparative study, subjects involved in the Q fever outbreak in 1989 who had fatigue were compared to subjects without fatigue who had had Q fever in 1989 using the same serological definition. The study had been passed by the East Birmingham Local Research and Ethics Committee (LREC).
Patients
We identified 115 patients from the original outbreak cohort of 147, of whom 110 replied positively after three mailings. Three patients had emigrated and 19 patients had died from a variety of causes (Table 1
) leaving just 10 patients untraced, a much higher identification rate than was achieved in the 1995 follow-up study,11 a positive response of 88% of survivors living in the UK. These subjects underwent a hierarchical questionnaire-based approach to defining chronic fatigue following that used by Wessely et al.12
|
Assessment
Fatigue was assessed by a hierarchy of questionnaires as described above. The questionnaire also addressed demographic details, including self-reported smoking status, which was not validated further. Chest X-rays were not performed for both ethical and logistical reasons. Each patient had a resting 12-lead ECG and M mode and 2D echocardiography using standard protocols. Following advice from the East Birmingham LREC, we undertook gated cardiac radio-nuclide scans only in those Coxiella-exposed subjects who fulfilled the CDC definition for chronic fatigue syndrome and who had no co-morbidities that might explain their fatigue. The radiologist involved was blind to the fact that all patients were fatigued rather than a mix of fatigued and non-fatigued patients. Shuttle walks were undertaken using standard methodology.13 Where necessary, hospital and general practitioner records were assessed to confirm clinical details and cause of death.
Analysis
The cardiac indices of Q-exposed fatigued cases were compared with Q-exposed non-fatigued cases. Matched cases and controls were compared studying only complete matched pairs using McNemar's test.
|
Results |
|---|
|
TopSummaryIntroductionMethodsResultsComparison of fatigue-free cases...DiscussionReferences |
|---|
Assessment of fatigue
This analysis is published in full elsewhere, but in summary, in those patients with complete data who underwent hospital assessment, 58 (68.2%) of the Coxiella-exposed cases reported fatigue of any duration. This total fell to 36 (42.4%) reporting fatigue after excluding subjects with those co-morbidities defined as exclusions in the 1994 CFS definition. Using the stricter CDC definition of CFS, and including in the denominator only the 85 patients seen in hospital who were fully assessed for CFS, 17 (20%) had CDC-defined CFS (including those with co-morbidities), which fell to seven (8.2%) when excluding those with co-morbidities. Demographic characteristics of the 85 patients who attended for further investigations are shown in Table 2
, categorized by degree of fatigue. Of 81 controls assessed, six were excluded on the basis of serological evidence of prior C. burnetii exposure; the characteristics of the remaining 75 are shown in Table 2.
|
Cardiac clinical cases in the cohort
We were aware that three patients had developed cardiomyopathy since contracting Q fever. Of these, two were alcohol abusers and one had had a pre-existing ischaemic cardiomyopathy. One male patient developed aortic valve vegetations in 1990 at the age of 43. He fulfilled the criteria for chronic fatigue but not the CDC definition, because of confounding co-morbidities. Ten years later, a serological IFA assay showed IgG antibody titres to Phase 1 and 2 Coxiella antigens at 10 and 160 respectively with negative values at <10 in the IgM and IgA sub-classes. The pattern was not that of Q fever endocarditis (see Discussion).
Electrocardiograms
ECGs were normal in 76.5% of Coxiella-exposed patients and 69.3% of controls. The spread of abnormalities was as expected in a group of this age range, and is summarized in Table 3. There were no significant differences between the overall groups in the pairs in which matching was possible: 72 complete matched pairs were available for analysis and no statistically significant differences were found (Table 4).
|
|
Echocardiograms
The overall data are shown in Table 5. Matched pair analysis (n=72) showed significantly reduced fractional shortening in controls compared to cases (32.3% vs. 37.0%; p<0.001, paired t test) (Table 6). There were no significant differences in echocardiographic, ECG or shuttle-walk distance data between fatigued and non-fatigued Q-fever patients. In addition, the pack-year smoking burden in fatigued and non-fatigued Q-fever patients was not significantly different (Table 7).
|
|
|
MUGA scans
These were undertaken in the six Coxiella-exposed subjects (4 males, 2 females; mean age 51.0 years, range 39–61) with CDC-defined CFS. There were no abnormal scans, all indices being within the normal range (mean ejection fraction 67.7%; range 58.8–73.8%) and consequently we did not scan the paired, non-fatigued patients who were exposed to Q fever.
|
Comparison of fatigue-free cases with fatigued cases |
|---|
|
TopSummaryIntroductionMethodsResultsComparison of fatigue-free cases...DiscussionReferences |
|---|
The original hypothesis was that sub-clinical cardiomyopathy might explain the fatigue in cases. Analysing the data for these two groups, taking fatigue at any level, there were no differences between the two in any ECG or echo measurement (Table 7
).
Shuttle walks
There was no significant difference in shuttle walk distance between either of the Coxiella-exposed groups and control subjects (Table 6).
|
Discussion |
|---|
|
TopSummaryIntroductionMethodsResultsComparison of fatigue-free cases...DiscussionReferences |
|---|
Q fever is an infectious disease which can cause both an acute illness, usually characterized by pneumonia, and a chronic form that has recognized clinical effects, notably endocarditis, osteomyelitis and granulomatous hepatitis.1 The typical serological response to acute infection is a rise in IgM antibody to Phase II antigen, followed by IgG antibody to Phase II antigen.14 Presence of IgM antibody does not necessarily indicate a current infection, as it may persist at a low level for many months after an acute infection. In chronic infection, Phase I antibody becomes measurable14 and in Q fever endocarditis, there are usually high titres of IgA and IgM antibodies to Phase I and Phase II antigens. IgM antibody is not usually found unless the endocarditis follows soon after an acute attack of Q fever. The recognition of QFS as a chronic complication of Coxiella infection was first described in Australia in the early 1990s,7 and represented quite a different outcome from classical chronic Q fever. Its existence has in the past been contested when clarification of its existence as a true entity was hampered by legal claims and inadequate study design. Q fever is an occupational disease of abattoir workers in Australia who are repeatedly exposed to the agent. Consequently, the cohort we describe here is important as the exposure was single. We thus had the opportunity to determine whether there was any evidence of endocarditis or granulomatous disease in this cohort, and whether the fatigue that we had earlier documented five years after exposure11 could be related to a sub-clinical cardiomyopathy.
Intuitively it might seem unlikely that Coxiella infection could lead to long-term symptoms, but earlier animal studies showed that C. burnetii can persist for long periods after inoculation. In humans, recrudescence of infection in late pregnancy as long as 3 years after initial infection has been documented.15 Recent Australian observations on QFS patients by PCR detected C. burnetii genomic DNA sequences in 65% of bone-marrow aspirates, in 17% of peripheral blood mono-nuclear cells and 14% of liver biopsy samples.16 Similar findings have been reported in QFS/CFS from Japan,17 suggesting that persistent organisms or antigen may be acting as a chronic immune stimulus resulting in specific cytokine release and consequently the symptoms characteristic of QFS. Patients with either acute Q fever or endocarditis have high serum levels of the cytokines IL6 and tumour necrosis factor,18 while patients with QFS demonstrate increased production of IL6 from peripheral blood monocytes in culture when stimulated by C. burnetii antigen, compared to unfatigued controls and to patients with a history of exposure to Coxiella but without fatigue.19
That infection by C. burnetii can result in heart disease is well known. Acute myocarditis has been reported,20 while chronic perimyocarditis, and consequent sudden cardiac death has been attributed to the related organism Rickettsia helvetica21 although there have been no reported cases of cardiomyopathy following acute Q fever. Chronic Q fever is characterized by endocarditis on both diseased and artificial valves,1 which can be very difficult to treat.22 Recently, a study of a large number of patients involved in an outbreak of Q fever in 1983 in Switzerland suggested that, 12 years after the event, subjects had an increased risk of vascular disease.23 However, the study did not adequately account for smoking history,24 shown to be a risk factor for infection in our original outbreak group. Our findings, 10 years after the outbreak, show no evidence of an excess of ECG abnormalities in the Coxiella-exposed group with fatigue compared to those without fatigue, while abnormalities were not statistically more common on the matched analysis (Table 6). This might seem surprising, but the mortality in the exposed group (19/147, 12.9% over 10 years) is greater than expected, which, along with the high incidence of co-morbidities in the surviving members of the group, would be compatible with the initial infection being more likely to affect patients with other underlying diseases at the time of infection. Consequently, we could be considering a survivor population where those more susceptible to the longer-term effects of Q fever have died.
The spread of causes of death is not particularly surprising given the ages of the patients at the time of death. Of the three subjects in the outbreak group with reported cardiomyopathy, two were found to be alcoholics and one was re-classified as ischaemic cardiomyopathy. One patient died from endocarditis on a stenosed aortic valve (the aortic stenosis reportedly preceding the outbreak) but no further details were available to us. A further Coxiella-exposed patient developed aortic valve endocarditis, but as his serological response was not that classically seen in chronic Q fever and is still under investigation, the significance of this finding in relation to his original infection is unclear. The original identification of florid Q fever endocarditis was followed by a survey of patients with valve disease which revealed a case of apparently ‘burnt out’ Q fever endocarditis without obvious vegetations.25 Complement-fixing antibodies to Phase 1 and 2 antigens were <300, but C. burnetii was isolated from the valve in guinea pigs.26 Recent observations from Australia have found a number of patients with fibrosed and calcified valves, similar to those described above, which histologically reveal no specific findings but which are positive to C. burnetii DNA sequences by PCR. Some of the patients had ‘classical’ serological profiles of Q fever endocarditis, while others had only medium level IgG antibody against Phase 1 and 2 antigens and no IgA antibody (Marmion BP, personal communication).
In summary, we have found no evidence of involvement of cardiac muscle in the patients involved in the 1989 Q fever outbreak in the West Midlands, although two subjects have subsequently developed aortic valve endocarditis. Thus chronic heart disease following acute Q fever is rare and is limited to endocarditis, without clear evidence of myocardial dysfunction of any sort in the long term.
|
Acknowledgments |
|---|
This study was supported by a Locally Organised Research (LORS) grant from the West Midlands NHS Executive. We are grateful to Dr Gordon Jones for performing and analysing the gated scans. We are also grateful to Professor B.P. Marmion (University of Adelaide) for his constructive and helpful comments during both the course of the study and the writing of the manuscript.
|
Notes |
|---|
Address correspondence to Professor J.G. Ayres, Department of Respiratory Medicine, Birmingham Heartlands Hospital, Birmingham. e-mail: ayresj{at}heartsol.wmids.nhs.uk
|
References |
|---|
|
TopSummaryIntroductionMethodsResultsComparison of fatigue-free cases...DiscussionReferences |
|---|
1. Marmion BP, Stoker MGP. The epidemiology of Q fever in Great Britain an analysis of the findings and some conclusions. Br Med J1958; 2:809–16.
2. Ferguson RJ, Shaw TRD, Kitchin AH, et al. Subclinical chronic Q fever. Q J Med1985; 222:669–76.
3. Smith G, Ayres JG, Blair I, et al. Q fever outbreak in Birmingham, UK. Lancet1989; ii:557.
4. Hawker JI, Ayres JG, Blair I, et al. (Q fever group). A large Q fever outbreak in the West Midlands: windborne spread into a metropolitan area? Commun Dis Pub Health1998; 1:180–7.
5. Smith DL, Wellings R, Walker C, Ayres JG, Burge PS. The chest x-ray in Q fever: a report on 69 cases from the West Midlands outbreak. Br J Radiol1991; 64:1101–8.
6. Smith DL, Ayres JG, Blair I, et al. A large Q fever outbreak in the West Midlands: clinical aspects. Respir Med1993; 87:509–16.[Web of Science][Medline]
7. Marmion BP, Shannon M, Maddocks I, Storm P, Penttila I. Protracted fatigue and debility after acute Q fever. Lancet1996; 347:977–8.[Medline]
8. Fukuda K, Straus SE, Hickie I, et al. The International Chronic Fatigue Syndrome Study Group: a comprehensive approach to it's definition and study. Ann Intern Med1994; 121:953–9.
9. Ramsay M. Post-Viral Fatigue Syndrome: The Saga of Royal Free Disease. London, Gower Medical, 1986.
10. Ayres JG, Smith EG, Flint N. Protracted fatigue and debility after acute Q fever. Lancet1996; 347:978–9.[Web of Science][Medline]
11. Ayres JG, Flint N, Smith EG, et al. Post-infection fatigue syndrome following acute Q fever: follow up study of patients involved in the 1989 outbreak in the West Midlands. Q J Med1998; 91:105–23.[Web of Science]
12. Wessely S, Chalder T, Hirsch S, Wallace P, Wright D. The prevalence and morbidity of chronic fatigue and the chronic fatigue syndrome: a prospective primary care study. Am J Pub Health1997; 87:1449–55.
13. Singh S, Morgan MDL, Scott S, Walters D, Hardman AE. Development of a shuttle walking test of disability in patients with chronic airways obstruction. Thorax1992; 47:1019–24.
14. Guigno D, Coupland B, Smith EG, Farrell ID, Desselberger U, Caul EO. Primary humoural antibody response to Coxiella burnetii, the causative agent of Q fever. J Clin Microbiol1992; 30:1958–67.
15. Syrucek L, Sobeslavsky O, Gutwirth I. Isolation of Coxiella burnetii from human placentas. J Hyg Epidemiol Microbiol Immunol1958; 2:29–35.[Medline]
16. Harris RJ, Storm PA, Lloyd A, Arens M, Marmion BP. Long-term persistence of Coxiella burnetii in the host after primary Q fever. Epidemiol Infect2000; 124:543–9.[Medline]
17. Kato K, Arashima Y, Asai S, Furuya Y, Yoshida Y, Murakami M, et al. Detection of Coxiella burnetii specific DNA in blood samples from Japanese patients with chronic non-specific symptoms by nested polymerase chain reaction. FEMS Immunol Med Microbiol1998; 21:139–44.[Web of Science][Medline]
18. Capo C, Amirayan N, Ghigo E, Raoult D, Mege J-L. Circulating cytokine balance and activation markers of leucocytes in Q fever. Clin Exp Immunol1999; 115:120–3.[Web of Science][Medline]
19. Penttila IA, Harris RJ, Storm P, Haynes D, Worswick DA, Marmion BP. Cytokine dysregulation in the post-Q-fever fatigue syndrome. Q J Med1998; 91:549–60.[Web of Science]
20. Chevalier P, Vandenesch F, Brouqui P, Kirkorian G, Tabib A, Etienne J, Raoult D, Loire R, Touboul P. Fulminant myocardial failure in a previously healthy young man [clinical conference]. Circulation1997; 95:1654–7.
21. Nilsson K, Lindquist O, Pahlson C. Association of Rickettsia helvetica with chronic perimyocarditis in sudden cardiac death. Lancet1999; 354:1169–73.[Web of Science][Medline]
22. Raoult D, Houpikian P, Tissot Dupont H, Riss JM, Arditi-Djiane J, Brouqui P. Treatment of Q fever endocarditis: comparison of two regimes containing doxycycline and ofloxacin or hydrochloroquine. Arch Intern Med1999; 159:167–73.
23. Lovey P-Y, Morabia A, Bleed D, Peter O, Dupuis G, Petite J. Long term vascular complications of Coxiella burnetii infection in Switzerland:cohort study. Br Med J1999; 319:284–6.
24. Wildman M, Ayres JG. Long term vascular complications of Coxiella burnetii infection. Br Med J2000; 320:58.
25. Andrews PS, Marmion BP. Chronic Q fever. 2: Morbid anatomical and bacteriological findings in a patient with endocarditis. Br Med J1959; 2:983–8.
26. Marmion BP, Higgins FE, Bridges JB, Edwards AT. A case of sub-acute rickettsial endocarditis; with a survey of cardiac patients for this infection. Br Med J1960; 2:1264–7.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||