International experts' practice in the antibiotic therapy of infective endocarditis is not following the guidelines.

OBJECTIVE: The management of infective endocarditis (IE) may differ from international guidelines, even in reference centres. This is probably because most recommendations are not based on hard evidence, so the consensus obtained for the guidelines does not represent actual practices. For this reason, we aimed to evaluate this question in the particular field of antibiotic therapy. METHODS: Thirteen international centres specialized in the management of IE were selected, according to their reputation, clinical results, original research publications and quotations. They were asked to detail their actual practice in terms of IE antibiotic treatment in various bacteriological and clinical situations. They were also asked to declare their IE-related in-hospital mortality for the year 2015. RESULTS: The global compliance with guidelines concerning antibiotic therapy was 58%, revealing the differences between theoretical 'consensus', local recommendations and actual practice. Some conflicts of interest were also probably expressed. The adherence to guidelines was 100% when the protocol was simple, and decreased with the seriousness of the situation (Staphylococus spp. 54%-62%) or in blood-culture-negative endocarditis (0%-15%) that requires adaptation to clinical and epidemiological data. CONCLUSION: Worldwide experts in IE management, although the majority of them were involved and co-signed the guidelines, do not follow international consensus guidelines on the particular point of the use of antibiotics.

Warfarin therapy and incidence of cerebrovascular complications in left-sided native valve endocarditis.

Anticoagulant therapy has been anticipated to increase the risk of cerebrovascular complications (CVC) in native valve endocarditis (NVE). This study investigates the relationship between ongoing oral anticoagulant therapy and the incidence of symptomatic CVC in left-sided NVE. In a prospective cohort study, the CVC incidence was compared between NVE patients with and without ongoing warfarin. Among 587 NVE episodes, 48 (8%) occurred in patients on warfarin. A symptomatic CVC was seen in 144 (25%) patients, with only three on warfarin. CVC were significantly less frequent in patients on warfarin (6% vs. 26%, odds ratio [OR] 0.20, 95% confidence interval [CI] 0.06-0.6, p = 0.006). No increase in haemorrhagic lesions was detected in patients on warfarin. Staphylococcus aureus aetiology (adjusted OR [aOR] 6.3, 95% CI 3.8-10.4) and vegetation length (aOR 1.04, 96% CI 1.01-1.07) were risk factors for CVC, while warfarin on admission (aOR 0.26, 95% CI 0.07-0.94), history of congestive heart failure (adjusted OR 0.22, 95% CI 0.1-0.52) and previous endocarditis (aOR 0.1, 95% CI 0.01-0.79) correlated with lower CVC frequency.