Case Presentation: A 54 year-old-woman presented with one week of progressive fatigue, shortness of breath, and productive cough.  Physical exam revealed fever of 101.6⁰ F, poor oral dentition, bilateral pulmonary crackles and left lower quadrant abdominal tenderness. Initial laboratory workup was remarkable for elevated procalcitonin which prompted initiation of broad spectrum antibiotics. Chest X-ray was concerning for a multifocal pneumonia and subsequent CT chest showed multifocal airspace opacities concerning for septic emboli.  CT abdomen/pelvis showed a probable splenic infarct. There was moderate tricuspid regurgitation with no vegetations or perforations, and preserved ejection fraction on echocardiogram. On second day of admission she became acutely confused without any focal neurological deficits. CT head showed recent left occipital and cerebellar infarcts and subsequent MRI brain revealed posterior and middle cerebral artery infarcts concerning for embolic phenomenon. Infectious workup was only significant for Methicillin-susceptible Staphylococcus aureus(MSSA) positive sputum culture; multiple blood cultures obtained throughout admission were negative for any microbial growth. Despite negative blood cultures, and absence of valvular vegetations on echocardiogram; the patient’s radiological findings were consistent with septic emboli to brain, lungs and spleen from presumed MSSA Infective Endocarditis.

Discussion: Infective endocarditis(IE) has significant morbidity and one-year mortality of 20-40% which makes timely diagnosis vital. Modified Duke criteria employs pathologic, clinical and imaging modalities to stratify patients as definite, possible or rejected IE. Given the poor prognosis without treatment, the sensitivity and specificity (each around 80%) of the modified Duke criteria are suboptimal. Echocardiography is the primary imaging modality used for diagnosis, however transesophageal echocardiography(TEE) can miss the diagnosis in up to 30% of cases. Clinicians should consider alternate imaging modalities such as CT if TEE is unrevealing in highly suspicious cases.

Conclusions:  Infective endocarditis although uncommon is associated with significant morbidity and mortality (in-hospital 14-22%). Early and accurate diagnosis is essential to prevent negative outcomes. The modified Duke criteria is a validated diagnostic tool, however it must be used along with clinical judgement and in context of pre-test probability. Our case signifies that hospitalists should have a low threshold for initiating therapy if the clinical presentation suggests IE even in the absence of Duke criteria.