Prevention of Q fever endocarditis
Gijs JM Limonard a, Marrigje H Nabuurs-Franssen a, PN Richard Dekhuijzen a, Cornelis AR Groot a
Confronted with more than 4000 human Q fever cases in the Netherlands since 2007, we appreciated the milestone article of Matthieu Million and colleagues' paper about Q fever endocarditis.1
Since cardiac valvular disease is regarded as a major risk factor for this rare but potentially fatal complication, the investigators emphasise their follow-up strategy: systematic screening for cardiac valvular abnormalities of every acute Q fever patient and administration of long-term prophylactic antibiotic treatment in case of any, be it trivial or clinically significant, valvular defect.2
However, this follow-up strategy has never been prospectively validated in an unselected population of patients with Q fever.
We present here the outcome of 134 Dutch patients with Q fever from the 2007 and 2008 outbreaks, who had screening echocardiography and prospective serological follow-up for 1 year (Table 1, Table 2). None of the patients received antibiotic prophylaxis and none developed clinical or serological signs of Q fever endocarditis. Additionally, none had developed clinical signs suggestive of chronic disease during the past 2 years.
Characteristics of patients and echocardiographic findings
Valvulopathy at screening echocardiography*
By contrast with existing publications, our data indicate a very low risk of progression to Q fever endocarditis. This contrasting result may be explained by selection bias and lack of specification of valvular defect severity in published retrospective research.3, 4
In these studies, the risk of endocarditis in patients with acute Q fever and pre-existing cardiac valve defects was estimated to be 39%, but many of the included patients had prosthetic valves.3
Million and colleagues1 describe a highly selective cohort with valve defects mostly caused by rheumatic fever (30%, about half of all specified valvular disease), a disorder that is rare in most developed countries. Furthermore, 30% of their patients had infected prosthetic valves.1 Degenerative valvular disease was present in only 5%,1 whereas this type of valvular defect is highly prevalent in the general population (22% in our study).
Minor, clinically insignificant valvular disease (trace and mild defects) has a high prevalence in any unselected population.5 In our study, these defects were not associated with development of chronic disease. In our setting, the harm of low-threshold administration of long-term antibiotic prophylaxis might therefore outweigh the benefit. Additionally, the associated health-care costs of this follow-up strategy on a large scale are substantial.
Screening echocardiography is therefore no longer undertaken in patients with acute Q fever in the Netherlands.
We declare that we have no conflicts of interest. We thank R Besselink and A Olde-Loohuis for clinical follow-up, G Weers and A Horrevorts for serological testing, C Wijkmans for collecting data, and J A Visser for providing the echocardiographic tests.