Prognostic factors in patients with infective endocarditis and persistently positive blood cultures

Background

The persistence of positive blood cultures (PPBC) after 48-72 hours of antibiotic therapy in infective endocarditis (IE) is considered a factor of poor prognosis. The aim of this study was to identify predictors of in-hospital mortality in these patients and to evaluate the prognostic role of cardiac surgery.

Methods

A total of 1,708 patients with left-sided IE were prospectively recruited from three tertiary hospitals between 1998 and 2024. Of these, 1,265 (74.1%) had positive blood cultures at admission, and 283 (22.4%) maintained PPBC at 48-72 hours. A univariate logistic regression analysis was performed to identify predictors of in-hospital mortality, followed by a multivariable analysis with stepwise variable selection. Multiple models were evaluated to determine the best fit and predictive capacity.

Results

The mean age of the analysed patients was 67 (14) years, and 64% were male. The most frequent comorbidities were diabetes mellitus (31.8%), anaemia (20.9%), and chronic kidney disease (15.2%). Regarding the microbiological profile, the most frequently isolated microorganisms were S. aureus (33.2%), coagulase-negative staphylococci (19.1%), and enterococci (19.1%).

Prosthetic IE accounted for 46.6% of patients, 87.6% had vegetations on transoesophageal echocardiography, and 30% had periannular complications. The main in-hospital complications were heart failure (55.4%), acute kidney injury (38.2%), systemic embolisms (31.8%), and septic shock (23.3%). Cardiac surgery was performed in 59.7% of patients, and 102 patients (36%) died during hospitalisation.

In the univariate analysis (Table), predictors of in-hospital mortality included Staphylococcus aureus infections, vegetation size, in-hospital complications such as acute heart failure, acute kidney injury, and septic shock, and surgical risk scales such as the RiskE Score and logistic EuroSCORE. Conversely, Streptococcus viridans infections and cardiac surgery were significantly associated with lower in-hospital mortality.

In the multivariable analysis, the development of septic shock and acute kidney injury were independently associated with increased in-hospital mortality. Additionally, compared to early cardiac surgery (<7 days from diagnosis), both a time from diagnosis to surgery greater than 7 days and failure to perform cardiac surgery were independently associated with higher in-hospital mortality. The final model showed good discriminatory capacity (area under the ROC curve= 0.801; 95% CI: 0.75-0.85) (Figure).

Conclusions

In patients with IE and PPBC, the development of septic shock and acute kidney injury, delayed cardiac surgery (more than 7 days from diagnosis), and failure to perform cardiac surgery were independent prognostic factors for in-hospital mortality.

Therefore, early intervention, before the time established by clinical practice guidelines to define persistent infection, appears to be crucial for improving the prognosis of these patients.Figure AUC, ROC curveUnivariate analyses in hospital mortality