RESUMO
AIMS: Adult congenital heart disease (ACHD) predisposes to infective endocarditis (IE). Surgical advancements have changed the ACHD population, whereas associated prosthetic material may constitute additional IE targets. We aimed to prospectively determine contemporary incidence, risk factors, and predictors of IE in a nationwide ACHD cohort, focusing on the presence of prosthetics. METHODS AND RESULTS: We identified 14 224 patients prospectively followed in the CONCOR ACHD registry (50.5% female, median age 33.6years). IE incidence was determined using Poisson regression, risk factors and predictors using Cox regression. Overall incidence was 1.33 cases/1000 person-years (124 cases in 93 562 person-years). For risk-factor analysis, presence of prosthetics was forced-as separate time-updated variables for specific prosthetics-into a model with baseline characteristics univariably associated with IE. Valve-containing prosthetics were independently associated with greater risk both short- and long term after implantation [0-6 months: hazard ratio (HR) = 17.29; 7.34-40.70, 6-12 months: HR = 15.91; 6.76-37.45, beyond 12 months: HR = 5.26; 3.52-7.86], non-valve-containing prosthetics, including valve repair, only in the first 6 months after implantation (HR = 3.34; 1.33-8.41), not thereafter. A prediction model was derived and validated using bootstrapping techniques. Independent predictors of IE were baseline valve-containing prosthetics, main congenital heart defect, multiple defects, previous IE, and sex. The model had fair discriminative ability and provided accurate predictions up to 10 years. CONCLUSIONS: This study provides IE incidence estimates, and determinants of IE risk in a nationwide ACHD cohort. Our findings, essentially informing IE prevention guidelines, indicate valve-containing prosthetics as a main determinant of IE risk whereas other prosthetics, including valve-repair, are not associated with increased risk long term after implantation.
