Infective endocarditis: Absence of microbiological diagnosis is an independent predictor of inhospital mortality

Abstract Background Infective endocarditis (IE) is associated with high inhospital mortality. New microbiological diagnostic techniques have reduced the proportion of patients without etiological diagnosis, but in a significant number of patients the cause is still unknown. Our aim was to study the association of the absence of microbiological diagnosis with in-hospital. prognosis. Methods Prospective cohort of 2000 consecutive patients with IE. Data were collected in 26 Spanish hospitals. Modified Duke criteria were used to diagnose patients with suspected IE. Results A total of 290 patients (14.8%) had negative blood cultures. Etiological diagnosis was achieved with other methods (polymerase chain reaction, serology and other cultures) in 121 (6.1%). Finally, there were 175 patients (8.8%) without microbiological diagnosis (group A) and 1825 with diagnosis (group B). In-hospital mortality occurred in 58 patients in Group A (33.1%) vs . 487 (26.7%) in Group B, p = 0.07. Patients in Group A had a lower risk profile than those in Group B, with less comorbidity (Charlson index 1.9 ± 2.0 vs. 2.3 ± 2.1, p = 0.03) and lower surgical risk (EuroSCORE 23.6 ± 21.8 vs. 29.6 ± 25.2, p = 0.02). However they presented heart failure more frequently (53% vs. 40%, p = 0.005). Multivariate analysis showed that the absence of microbiological diagnosis was an independent predictor of inhospital mortality (Odds Ratio 1.8, 95% Confidence Interval 1.1–2.9, p = 0.016). Conclusion Approximately 9% of patients with IE had no microbiological diagnosis. Absence of microbiological diagnosis was an independent predictor of inhospital mortality.