Case Presentation: A 62 year old homeless male with no significant past medical history presents to a large tertiary referral center after being found unresponsive by police. Vital signs: 99.2 F, HR 86, RR 17, BP 160/81. Labs: Lactic acid 2.9, WBC 26.35, CRP 150.60, urine drug screen negative. CT imaging revealed evidence of ureteral stones, pyelonephritis and hydronephrosis, requiring percutaneous nephrostomy tube (PCN) placement. Empiric treatment for urosepsis was initiated with broad spectrum antibiotics. Imaging also identified an incidental 1.5cm spiculated left lung mass abutting the aortic arch with associated lymphadenopathy, raising concern for malignancy. After the PCN, patient was seen with a new facial droop and spastic hemiplegia of the left arm. MRI of the brain revealed ischemic strokes, with a possible embolic or watershed area etiology. An ischemic stroke workup identified an approximately 70% unilateral carotid artery stenosis. Vascular surgery was consulted but did not recommend intervention due to acuity, and stated stenosis was unlikely the stroke’s primary cause. A transthoracic echocardiogram was unremarkable, but transesophageal echocardiography revealed thickened, mildly mobile lesions on the aortic valve. Extensive workup included blood cultures and antibodies. Infectious disease consultation was obtained, and workup revealed positive Q fever antibodies, though the titer significance was unclear. Patient was started on empiric antibiotics for culture negative endocarditis and Coxiella endocarditis. Karius testing, which detects circulating bacterial DNA, was performed to confirm significance of titers, and determine duration of antimicrobial therapy which was added to the regimen. Neurology recommended therapeutic anticoagulation for Non Thrombotic bacterial endocarditis (NBTE).
Discussion: From the hospitalist’s perspective, the decision regarding anticoagulation poses a significant dilemma, given the complex factors at play. Our Neurology team recommended therapeutic anticoagulation with low-molecular-weight heparin (LMWH) for NBTE; however, there was concern about the risk of hemorrhagic transformation, particularly in the setting of septic emboli. On the other hand, the patient had a suspected lung malignancy, which on CT chest angiogram appeared to involve the aortic arch, potentially contributing to thrombus and embolic strokes in the future. While the American Heart Association guidelines recommend anticoagulation in the setting of infective endocarditis for patients with rheumatic mitral stenosis, mechanical heart valves, atrial fibrillation or atrial/ventricular thrombus, our patient did not have these complications. 2012 Chest guidelines on antithrombotic therapy for valvular disease has a grade 1C recommendation against the use of anticoagulation in endocarditis of a native valve. Furthermore, there is limited evidence of anticoagulation for malignancies invading the aortic arch, with surgery being preferred. There was discussion about starting anticoagulation if the Karius test was negative. However, the patient’s homelessness raised concerns about follow-up care and long-term anticoagulation therapy.
Conclusions: In conclusion, therapeutic anticoagulation was deemed more harmful than beneficial in this patient in the setting of factors such as ischemic stroke, infective endocarditis, malignancy, and compounded by his social situation. A conservative approach to anticoagulation and infective management was chosen.