705 - Factors Associated with Local Invasion in Infective Endocarditis: a Nested Case-control Study

Background: A substantial proportion of infective endocarditis (IE) cases are complicated by local invasion. The purpose of this study was to identify patient and disease characteristics that increase the odds of local invasion in surgically treated IE patients.

Methods: This was a nested case-control study. All episodes of IE for patients admitted to Cleveland Clinic from January 1^st, 2013 – June 30^th, 2016 were identified from the Cleveland Clinic IE Registry. Patients >18 years of age who underwent surgery for IE were included. Data was compiled by manual review of the Electronic Medical Record. Local invasion, defined as peri-annular extension, peri-valvular abscess, intra-cardiac fistula or pseudoaneurysm, was ascertained from the surgical operative note. Associations of selected factors with local invasion were examined in a multivariable logistic regression model.

Results: Among 510 patients who met inclusion criteria, 206 had local invasion. Mean age was 56 years and 369 (72 %) were male. Overall 344 (67 %) had aortic valve, 228 (45%) mitral valve, and 66 (13%) tricuspid or pulmonic valve involvement. Aortic valve involvement (OR 5.76, 95% CI 3.44 – 9.98), mechanical valve (OR 7.63, 95% CI 3.63 – 17.07), bioprosthetic valve (OR 3.20, 95% CI 1.99 – 5.19), significant paravalvular leak (OR 2.27, 95% CI 1.09 – 4.97), new atrioventricular nodal block (OR 3.05, 95% CI 1.57 – 6.09), Staphylococcus aureus infection (OR 2.11, 95% CI 1.20 – 3.76), coagulase negative staphylococcal infection (OR 2.38, 95% CI1.27 – 4.54), and non- viridans group streptococcal infection (OR 4.21, 95% CI 1.81 – 10.06) were significantly associated with local invasion.

Conclusion: Intra-cardiac and microorganism factors, but not comorbid conditions, are associated with local invasion in IE.