Incidence of chronic Q fever and chronic fatigue syndrome: a six year follow-up of a large Q fever outbreak, 2021, Ankert et al

[Andy]

Abstract

Objectives
Acute Q fever is a generally self-limiting infection caused by the intracellular gram-negative bacterium Coxiella (C.) burnetii. For yet unknown reasons, a subset of patients develops chronic a infection. Furthermore, a Chronic Fatigue Syndrome (CFS) as post-acute Q fever sequelae has been described. We here investigated the rates of chronic Q fever and incidences of CFS six years after one of the largest European Q fever outbreaks that occurred in Jena, Germany in 2005 with 331 reported cases, who lived in proximity of a grazing sheep herd.

Methods
A total of 80 patients and 52 non-diseased household members from the former outbreak, were enrolled six years after the outbreak, blood samples collected and tested for a chronic Q fever were determined by seroprevalence using referenced immunofluorescence tests. Also, the presence of a CFS was assessed using the Short Form Symptom Inventory developed by the Centers (United States) for Disease Control and Prevention (SF CDC- SI).

Results
In 80 out of 132 (60.6%) study participants, previous Q fever infection was confirmed serologically, while no previous infection was detected in the 52 household members. None of the participants fulfilled the serological criteria of chronic Q fever. The evaluation of the CDC-SI did not show any differences between the two groups. Also, there was no difference between both groups regarding fulfillment of CFS-defining criteria (n = 3 (3.8 %; sero-positive) vs. n = 2 (3.8 %; sero-negative), p = 0.655).

Conclusion
Our six-year follow-up study of a large Q fever outbreak did not find evidence for chronic Q fever or post Q fever CFS. There was no asymptomatic sero-positivity in household members of Q fever patients.

Open access, https://onlinelibrary.wiley.com/doi/10.1111/tbed.14224

 

[Hoopoe]

Strange finding. The author think that whether there is a chance of developing CFS or not depends on the pathogen strain.

 

[Hutan]

The approach to determine if someone had CFS is unusual, both in terms of the symptoms picked for the criteria, and the calculation of the score.

We asked the individuals to answer the Centers for Disease Control and Prevention Symptoms Inventory (SF CDC- SI) to detect chronic fatigue syndrome after C. burnetii infection (QFS) (Wagner et al., 2005). The SF CDC- SI is designed to detect concomitant symptoms associated with CFS. It collects information about the presence, frequency, and intensity of six CFS-defining symptoms (unusual fatigue after exertion, non-restorative sleep, muscle pain, sleep problems, problems with memory, and concentration problems). If at least four of the required concomitant symptoms occurred for at least six months, the SF CDC- SI specific Cut-off was judged to be met.

Perceived frequency of each symptom was rated on a four-point scale and severity or intensity of symptoms was measured on a three-point scale. The case-definition symptom scores were calculated by multiplying the frequency score by the intensity score. We transformed the intensity scores into equidistant scores before resulting in range 0–16 for each symptom. Afterwards the values for each symptom were summed up (SF CDC- SI-Score).

The SF CDC- SI allows to make a statement about the presence of a CFS if the Cut Off ≥4 and the SF CDC- SI-Score ≥25.

 

[cassava7]

The authors relied on the short form of the CDC CFS symptom inventory, which assesses 6 symptoms instead of the full form's 19. It was developed in 2005 by the CDC's Wagner, Unger, Reeves et al. from the answers to the full inventory of "164 subjects (with CFS, other fatiguing illnesses and non fatigued controls) identified from the general population of Wichita, Kansas" by only keeping "symptoms whose scores had a corrected item-total score correlation < 0.60" [1].

The list of symptoms in the short form seems to be at odds with the definition of the Fukuda criteria and the scoring of the full inventory. The Fukuda criteria list 8, not 6, defining symptoms for CFS*, and:

- while the full inventory distinguishes between memory and concentration problems, it states that only one of the two can be counted ("Note that only L or M can be counted, not both"), as the criteria gather them into one single item: "substantial impairment in short-term memory or concentration"

- while unrefreshing sleep is a defining symptom, sleeping problems are considered an accompanying symptom in the full inventory.

Thus, the short form appears to only assess 5 out of 8 of the defining CFS symptoms per the Fukuda criteria (with one of the latter separated into two in the short form, i.e. incorrectly counted twice). It is possible that with the 3 remaining symptoms, the participants in this study would have scored higher and met the cut-off for CFS.

However, Wagner et al. found that the short form score converged well with the full inventory (r = .94, p < 0.001) and the Fukuda case definition (r = .95, p < 0.001) but found with regards to construct validity that it doesn't adequately separate CFS from idiopathic chronic fatigue [1]:

The Total [inventory] score and the [Fukuda] Case Definition score distinguished between subjects classified CFS or ISF (Bonferroni post-hoc test; p < .05), while the Symptom Inventory Short Form revealed only a trend towards higher symptom impact for the CFS.

The authors do note that the IOM committee identified criticisms with the CDC CFS symptom inventory (in 2015, so post-data collection), but they justify their choice of its short form rather than "a battery of questionnaires" by "[wanting] to increase the motivation of patients to participate in this follow up". Strangely, they cite a letter to the editor by @Michiel Tack to back this justification, but the letter criticizes the (full) symptom inventory for being outdated and uncharacteristic of ME/CFS. It does not mention the short form or that filling fewer and shorter questionnaires is easier for CFS patients.

The discrepancy with other studies on the incidence of QFS is also noted in the discussion, but only explained with weak suppositions about psychosocial aspects of CFS and the strain of Coxiella burnetii:

Various factors influencing the prevalence of CFS after Q fever infection can be found in the literature. Sociological factors are also described as influential. Different geographic regions appear to be more exposed to CFS than others. Furthermore, media importance and financial interests in secondary illnesses should be taked into consideration when assessing chronic fatigue in general (Eldin et al., 2017; Pheby et al., 2020). Exemplarily, Morroy et al. (2016) described in their meta-analysis that neither psychological nor microbial factors appear to predict postinfectious fatigue. Yet a biopsychiological etiology could not be ruled out (Strauss, B. et al., 2012). Keijmel et al. (2015) showed in their work that cognitive-behavioural variables may also play a role in that there is a negative association between physical activity and CFS after Q fever infection. The bacterial strain additionally appears to play a decisive role (Sukocheva et al., 2010; Raijmakers et al., 2020).

Finally, it is intriguing that the authors are publishing their data 10 years after collection. In 2011, they contacted by mail 311 patients who had tested positive to Coxiella burnetii in 2005; 94 responded and, of those, 80 agreed to take part in the study. Presumably, most sequelae-free patients wouldn't have taken the time to respond, but whether there was any other form of selection bias is unknown.

Is there a German QFS patient association? They would likely know more about the patients who remain ill and about this study.

* post-exertional fatigue, unrefreshing sleep, problems remembering or concentrating, muscle aches and pains, joint pain, sore throat, tender lymph nodes and swollen glands, and headaches

[1] Wagner, D., Nisenbaum, R., Heim, C. et al. Psychometric properties of the CDC Symptom Inventory for assessment of Chronic Fatigue Syndrome. Popul Health Metrics 3, 8 (2005). https://doi.org/10.1186/1478-7954-3-8

ETA: the researchers are part of the consortium of the Q fever German Interdisciplinary Program for Research (Q-GAPS), which is supported by the German Federal Ministry of Education and Research. Dr Katharina Boden is the principal investigator of a subproject aiming to assess the efficacy of a disinfectant against C. burnetii.

The Q-GAPS FAQ entry on "What is Q fever fatigue syndrome" reads (bolding mine):

Therapeutically this symptomatic complex is a challenge, as the disease can't be influenced by administering antibiotics. In addition, a diagnostic laboratory test for detecting the disease is not possible. Therefore, psychosomatic and behavioural therapeutic rational approaches are recommended.

 

[Hutan]

Nice analysis cassava7, thank you.

Finally, it is intriguing that the authors are publishing their data 10 years after collection.

Yes.

Patients with QFS are otherwise healthy people without underlying medical or psychological co- morbidities, who develop a complex set of symptoms dominated by paralyzing fatigue after acute Q fever infection.

I do wonder if it is related to the current ME/CFS guideline process.

It is an intriguing paper.

This chart (specifically the 'unusual fatigue after exertion' bars) does suggest that, even if the diagnostic criteria had been better, they might have not found any QFS in 2011.



There is a file of data giving participant symptoms in the supplementary material. Table 1 gives the sero-positive status for the participant IDs.

I guess it's worth noting that there were only 82 respondents who had had Q fever. We know that incidence of post-infectious fatigue decreases a lot over time. If we look at the Dubbo results, only a very small percentage of people still had symptoms after a couple of years. Given that frequency, it's probably not remarkable if not a single incidence of QFS was identified.

The largest reported Q fever outbreak occurred between 2007 and 2010 in the Netherlands with more than 4,000 registered human cases (Van der Hoek et al., 2010). A significant proportion of these Q fever patients showed an impairment of their state of health up to one year after infection. After four years, half of the patients complained of excessive fatigue and a severely impaired overall quality of life. (Limonard et al., 2016). Other follow-up studies reported prevalence rates of QFS of 19.2 % ten years after exposure (in the United Kingdom) (Wildman et al., 2002) and 28 % after five years (in Australia) (Hickie et al., 2006).

I found the reference to the Dubbo study (the Hickie et al study) odd. I'm sure that the Dubbo study itself did not find a 28% prevalence of QFS after 5 years (especially not in 2006 when the study had just been done.). So, if the reference is correct, the Dubbo study paper must have referred to another study. I wonder why the authors of this German paper did not note that the Dubbo study found that incidence of post-infectious fatigue dropped off to something more like 1% after a couple of years for each of Q fever, EBV and Ross River Fever. Is it because it would highlight that their study was not a big enough sample to be sure to identify a QFS case?

 

[cassava7]

I guess it's worth noting that there were only 82 respondents who had had Q fever. We know that incidence of post-infectious fatigue decreases a lot over time. If we look at the Dubbo results, only a very small percentage of people still had symptoms after a couple of years. Given that frequency, it's probably not remarkable if not a single incidence of QFS was identified.

That the study was statistically underpowered to detect years-long cases of QFS seems to be the most plausible argument behind the findings, and a much more compelling one than potential issues around the short form of the CDC CFS symptom inventory.

I found the reference to the Dubbo study (the Hickie et al study) odd. I'm sure that the Dubbo study itself did not find a 28% prevalence of QFS after 5 years (especially not in 2006 when the study had just been done.). So, if the reference is correct, the Dubbo study paper must have referred to another study. I wonder why the authors of this German paper did not note that the Dubbo study found that incidence of post-infectious fatigue dropped off to something more like 1% after a couple of years for each of Q fever, EBV and Ross River Fever. Is it because it would highlight that their study was not a big enough sample to be sure to identify a QFS case?

The authors note that one of the "strength of our study is the large number of sero-positive subjects (...)". Indeed, they likely did not want the low number of patients (with regard to CFS incidence) to be seen as a weakness.

In 2006, the Dubbo study group (Hickie et al.) reported an incidence of CFS, respectively at 6 and 12-month follow-up, of 11% (29/253 but with 28 meeting CFS criteria) with 3 from Q fever and 9% (22/253):

Incidence of post-infective fatigue syndrome

The case rate for provisional post-infective fatigue syndrome was 35% (87/250) at six weeks, 27% (67/250) at three months, 12% (29/250) at six months, and 9% (22/250) at 12 months. No difference in these case rates existed between the initial infective agents (fig 1).

The medical, psychiatric, and laboratory assessments of the 29 provisional cases of post-infective fatigue syndrome at six months led to exclusion of one participant on medical grounds and none on psychiatric grounds. The 28 cases of chronic fatigue syndrome, termed here confirmed post-infective fatigue syndrome, included 14 men and 14 women with a mean age of 37 (range 17-63) years, including five participants with confirmed Epstein-Barr virus infection, three with Q fever, 13 with Ross River virus, and eight with unconfirmed infection. The 28 cases did not differ in age or sex when compared with either all participants with serological confirmation—age 36.0 versus 32.2 years, difference = 3.8 (−3.0 to 10.6) years; sex (per cent male) 55% versus 58%, difference = −3.4% (−19% to 28%)—or all enrolled participants—age 36.0 versus 34.1 years, mean difference = 1.9 (−8.7 to 4.9) years; sex (per cent male) 55% versus 58%, difference = −3% (−20% to 26%).

There was no follow-up data after 1 year in the Dubbo study. The German authors may have mistaken the number of confirmed cases of CFS at 6 months (28) with the percentage (11%).

Unfortunately the large 4-year follow-up study by Limonard et al. (2016, NL outbreak in 2010) did not evaluate CFS caseness but assessed patients with the Nijmegen Clinical Screening Instrument [1], which was originally developed for COPD and later applied to Q fever. This is problematic because, while it found that 50% of patients had excessive fatigue and reduced QoL, the 10-year follow-up study by Waldman et al. (2002, UK outbreak in 1989) that the authors also cite shows that fatigue, ICF and CFS must be clearly delineated: respectively, they affected 66.7%, 34.7%, 19.4% of the cohort [2]. Excluding comorbidities, this dropped to 8% CFS cases.

[1] Limonard GJ, Peters JB, Besselink R, Groot CA, Dekhuijzen PN, Vercoulen JH, Nabuurs-Franssen MH. Persistence of impaired health status of Q fever patients 4 years after the first Dutch outbreak. Epidemiol Infect. 2016 Apr;144(6):1142-7. doi: 10.1017/S0950268815002216. Epub 2015 Oct 28. PMID: 26508155.

[2] Wildman MJ, Smith EG, Groves J, Beattie JM, Caul EO, Ayres JG. Chronic fatigue following infection by Coxiella burnetii (Q fever): ten-year follow-up of the 1989 UK outbreak cohort. QJM. 2002 Aug;95(8):527-38. doi: 10.1093/qjmed/95.8.527. PMID: 12145392.

 

[cassava7]

There was no follow-up data after 1 year in the Dubbo study. The German authors may have mistaken the number of confirmed cases of CFS at 6 months (28) with the percentage (11%).

I am mistaken. The 28% figure was taken from the systematic review by Morroy et al. (2016), specifically the paragraph on CFS (bolding mine):

In Australia, QFS is the most common sequel of acute Q-fever reported to affect 10–15% of patients [70]. Higher percentages were described, with up to 28% of patients meeting the Centres for Disease Control and Prevention criteria for CFS 5 to 14 years after acute Q-fever, compared to none in the control group [8, 15]. The highest percentage of reported fatigue was 69% five years after acute Q-fever [9]. CFS criteria were met by 42% of C. burnetii-infected patients and 26% of controls [9, 15]. Ten years after acute Q-fever, 68% of patients reported fatigue of any duration [54], of whom 20% met the CFS criteria [15]. Excluding co-morbidity, 8% of patients met the CFS criteria compared to none of the controls [54]. C. burnetii-exposed compared to non-exposed subjects reported ten years later a fatigue prevalence of 65% vs. 35%, respectively, and 19% vs. 4% met the CFS criteria [7, 15].

This 28% figure comes from a 2+ year follow-up retrospective survey of the 1989 UK outbreak of 147 cases. The survey included 39 patients with confirmed acute Q fever and 3 control groups, each with 39 participants.

The 42% CFS figure comes from the 10-year follow-up of the same outbreak, with 71 responses (less than half the initial number of cases). However, doubt can be cast on this prevalence given that 26% controls were also diagnosed with CFS.

Unlike the Dubbo study, neither of these were prospective, so it is understandable that they would report higher rates of CFS.