Aorto-cavitary fistulae in infective endocarditis: understanding a rare complication through collaboration.

Even in the modern era of antimicrobial chemotherapy and advanced diagnostic imaging, infective endocarditis continues to surprise, frustrate, and perplex, and remains an evolving disease with a persistently high mortality and morbidity. Almost all aspects of the disease, including its natural history, pre-disposing factors, sequelae, and causative organisms are virtually unrecognizable compared with Osler’s initial descriptions from the nineteenth century. In particular, chronic rheumatic heart disease is now an uncommon antecedent, whereas mitral valve prolapse, prior valve replacement, intravenous drug use, and preceding vascular instrumentation have become increasingly frequent, coinciding with an increase in staphylococcal infections and those due to fastidious or atypical organisms. Clinical studies have been slow to adapt to these shifting epidemiological patterns, partly on account of the relative scarcity of infective endocarditis, contemporary series indicating a current incidence of 1.7–6.2 cases per 100 000 patient years. Knowledge of the clinical features and natural history of the disease has therefore relied largely on small, uncontrolled, outdated studies; modern, welldesigned studies reflecting current disease patterns are long overdue. Anguera et al. report a large case series of 76 patients with surgically or autopsy proven aorto-cavitary fistulous tract formation, identified from a multi-centre clinical database of infective endocarditis over a 10 year period. Previously, data on aorto-cavitary fistulae were limited to isolated reports and small case series, and despite its retrospective nature, the study contributes substantially to knowledge of this rare but important complication of infective endocarditis. Within the study, aorto-cavitary fistulae occurred with a prevalence of 1.7%, rising to 5.8% in those with prosthetic valve endocarditis. Fistulae occurred uniquely in aortic valve endocarditis, usually in association with identifiable peri-valvular abscess formation, and with equal distribution between all three coronary sinuses and all four cardiac chambers. Almost all fistulae were successfully identified echocardiographically, though transthoracic imaging appeared unreliable in this regard, and the diagnosis was frequently only made after transoesophageal assessment. Fistulae were identified a median of 25 days after initial symptom onset, and 5 days after hospitalization, by which time moderate or severe aortic regurgitation and significant heart failure were present in over 60% of patients. Surgical correction was undertaken in 87% of the patients, largely in conjunction with aortic valve replacement, and in-hospital post-operative mortality was high at 42%. This adverse outcome was independently related to the preoperative presence of moderate or severe heart failure and the need for urgent or emergency surgery. In surviving patients, late post-operative prognosis was substantially worse in the presence of residual fistulae. Although the present study does not demonstrate a major survival advantage associated with surgery, the findings suggest that aorto-cavitary fistulae are a surrogate for severe and extensive tissue destruction and support the need for early intervention. In general, fistulae appear to be associated with peri-valvular abscess