Infective endocarditis: What are predisposing conditions in native valves?

Büchi, Annina Elisabeth; Hoffmann, Mario (2017). Infective endocarditis: What are predisposing conditions in native valves? (Dissertation, Universität Bern, Medizinische Fakultät, Universitätsklinik für Infektiologie, Inselspital Bern)

Preview

Text 17buechi_a-hoffmann_m.pdf - Published Version Available under License Creative Commons: Attribution-Noncommercial-No Derivative Works (CC-BY-NC-ND). Download (2MB) | Preview

The term ‘predisposing heart condition’ is used as an indication of antimicrobial prophylaxis to prevent infective endocarditis (IE) and as a criterion for diagnosing IE according to the modified Duke criteria. Whereas the use of the term for antimicrobial prophylaxis is well defined, the criterion for diagnosing IE is not.
The general objective of this thesis is to narrow the definition of a predisposing heart condition in ‘native’ valves for the diagnosis of IE. Therefore, we reviewed the literature and the evidence about specific heart conditions reported to be a risk factor for IE. In parallel, we reviewed the imaging technique available at the time these studies were published and compared the results with imaging from today’s perspectives and current definitions of a specific heart condition (i.e. valvular disease). Finally, we evaluated the knowledge and opinion of clinicians about the term predisposing heart condition.
Our literature review included 207 studies, the vast majority of which were descriptive. Only a few studies investigated valve pathology as a risk factor for IE via analytical statistics. In addition, three-quarters of all included studies involved patients who presented with IE prior to the publication of the modified Duke criteria.
Studies focussing on mitral valve prolapse (MVP, 116 publications), prior IE (96 publications), and bicuspid aortic valve (BAV, 78 publications) provided the most data. The odds ratio of developing IE for a patient who had previously experienced an episode of it was approximately 2.5. The mean proportion of patients with IE plus a history of previous IE was 8.3% (median 7.1%, interquartile range [IQR] 4.9%–10.2%). One study associated BAV with a higher risk of IE (hazard ratio 6.3). In 77 descriptive studies, a median of approximately 6% of patients with IE had BAV as an underlying condition. Our literature review on the evolution of imaging methods indicated, however, a considerable influence of medical progress on the diagnosis of MVP. Six analytical studies and 90 of the 110 descriptive studies included patients prior to the publication of the modified Duke criteria in 2000. For many years, MVP was diagnosed via auscultation only, and echocardiographic means for diagnosis were used in the late 90s. Therefore, both the risk of developing IE and the proportion of patients with IE and MVP as a predisposing factor could not be quantified.
The literature review on mitral valve stenosis (MS, 23 publications) and pathologies involving the pulmonary valve (18 publications) and the tricuspid valve (nine publications) provided little data. These publications had inconsistent results and low proportions of patients with IE had these valve pathologies.
The significance of aortic valve stenosis (AS, 46 publications), mitral valve insufficiency (MI, 41 publications), and aortic valve insufficiency (AI, 39 publications) as a predisposing heart condition
was difficult to assess from today’s perspective because of the progress made in imaging methods; of these studies, 75.6%, 78.6%, and 79.5%, respectively, included patients prior to the publication of the modified Duke criteria in the year 2000. In addition, except for AS (1989), the categorisation of mild, moderate, and severe valve pathology was established in 1998 or 2006. The publications had considerable heterogeneity with a wide distribution of results. An observational study indicated that with an increased incidence of AS, the risk of developing IE rises. Only one of these 126 publications for these three valve pathologies used analytical statistics. Congenital AS was associated with a higher risk of IE (hazard ratio of 4.9).
The results from the literature review parallel those from a survey that we performed to evaluate the knowledge and opinion of clinicians on the term predisposing heart condition. The survey indicated that there is significant uncertainty among clinicians regarding what is considered to be a Duke minor criterion for a predisposing heart condition in a native valve. The results from 318 questionnaires with responses from specialists in the fields of internal medicine, infectious diseases, and cardiology provided a wide range of answers. Their answers also showed that what the participants believed to be a current Duke minor criterion and what they thought should be a minor criterion had a median accordance of 33%.
Taken together, these findings demonstrate that there is uncertainty about what is considered a predisposing heart condition for the diagnosis of IE. This uncertainty is demonstrated in our extensive literature review and reflected in our survey among clinicians. The vast majority of studies used only descriptive statistics and included patients prior to the publication of the modified Duke criteria (2000). The tremendous progress in imaging methods and categorisation of valve pathologies since then makes it difficult to interpret the literature review analyses from today’s perspective. Nonetheless, studies on MVP, a prior episode of IE, and BAV had the highest representation in the literature. Among these three pathologies, MVP is most likely to be affected by the evolution of imaging methods, and therefore its risk cannot be quantified. Sensitivity analyses and mathematical models performed on the data obtained in this systematic review may help to further narrow the definition of a predisposing heart condition.

Interest & Impact

Downloads

191 since deposited on 18 Sep 2017
22 in the past 12 months

Citations

5 Citations in Web of Science ®
6 Citations in Scopus

Search

in Google Scholar™

Services

Actions (login required)

Edit item

Actions (login required)

Edit item