The Duke minor criterion "predisposing heart condition" in native valve infective endocarditis - a systematic review

BACKGROUND The term “predisposition” is used as an indication of antimicrobial prophylaxis to prevent infective endocarditis and as a criterion for diagnosing infective endocarditis according to the modified Duke criteria. The criterion for diagnosing infective endocarditis in native valves is not well defined. OBJECTIVES To identify conditions that increase the risk for infective endocarditis in native valves, for the diagnosis of infective endocarditis according to the modified Duke criteria. In parallel, we compared the results with the year of patient inclusion for each study and echocardiographic techniques. RESULTS Our systematic review included 207 studies published from January 1970 to August 2015. Studies that focused on mitral valve prolapse (112 studies), prior infective endocarditis (96) and bicuspid aortic valve (78) provided the most data. However, only six (5.3%), three (3.1%) and one (1.3%) of these studies, respectively, used analytical statistical methods. Three (2.7%), two (2.1%) and one (1.3%), respectively, were graded as good quality studies. Odds ratios (ORs) for developing infective endocarditis were 3.5–8.2 for mitral valve prolapse, and 2.2 and 2.8 for prior infective endocarditis. The hazard ratio for developing infective endocarditis was 6.3 for bicuspid aortic valve. The mean prevalence proportion of infective endocarditis in patients with these three heart conditions were 8.5% (mitral valve prolapse), 8.3% (prior infective endocarditis) and 8.8% (bicuspid aortic valve). The proportions of publications prior to the publication of the modified Duke criteria were 81.8, 75.6 and 74%, respectively. Evolution of the imaging method and echocardiographic technique was estimated to be considerable for mitral valve prolapse. The literature review on aortic valve stenosis (46 studies), mitral valve insufficiency (41) and aortic valve insufficiency (39) provided two analytical studies for aortic stenosis. One study was graded as good quality and reported a hazard ratio 4.9. The mean prevalence of these heart conditions in patients with infective endocarditis were 7.3, 19.9 and 10.2%, respectively. The proportions of publications prior to the publication of the modified Duke criteria were 78, 75.6 and 79.5%, respectively. The evolution of both the echocardiographic technique and the categorisation of valve disease severity was considerable for all three entities. CONCLUSIONS The evidence for native valve heart conditions predisposing to infective endocarditis is mainly based on studies with only descriptive statistics published prior to the release of the modified Duke criteria. Mitral valve prolapse, prior infective endocarditis and bicuspid aortic valve are frequently cited as predisposing heart conditions for infective endocarditis. The evolution in echocardiographic techniques over the past decades and its influence on diagnosis was considerable for mitral valve prolapse, aortic stenosis, mitral insufficiency and aortic insufficiency.