Early Diagnostic Prediction of Infective Endocarditis: Development and Validation of EndoPredict-Dx

Area of Science:

• Cardiology
• Infectious Diseases
• Diagnostic Medicine

Background:

• Infective endocarditis presents diversely, complicating diagnosis and treatment.
• Clinical suspicion guides empirical treatment and diagnostic test selection.
• Early and accurate diagnosis is critical for managing this life-threatening condition.

Purpose of the Study:

• To develop and validate the EndoPredict-Dx score for early prediction of infective endocarditis.
• To identify key clinical and laboratory predictors for infective endocarditis.
• To stratify patients into risk groups for timely intervention.

Main Methods:

• A cohort of 805 patients with suspected infective endocarditis was analyzed.
• Logistic regression was employed to derive the EndoPredict-Dx scoring system.
• Internal validation was performed using bootstrapping, with left-sided infective endocarditis as the primary outcome per Duke criteria.

Main Results:

• The EndoPredict-Dx score incorporates factors such as male sex, prior endocarditis, petechiae, heart murmur, suspected embolism, symptom duration, low hemoglobin, elevated leukocytes, high C-reactive protein, and urine red blood cells.
• The scoring system demonstrated good predictive performance with an AUROC of 0.78 in the derivation cohort and 0.77 in internal validation.
• Patients were successfully categorized into three distinct risk groups.

Conclusions:

• The EndoPredict-Dx score provides an accurate method for predicting infective endocarditis likelihood.
• The score effectively utilizes readily available clinical and laboratory data at admission.
• This tool can aid clinicians in the early diagnosis and management of infective endocarditis.