Case Presentation: 81 year-old man with a history of atrial fibrillation on Warfarin and bioprosthetic aortic and mitral valves presented with dark urine and shortness of breath. He described progressive shortness of breath and reduced exercise tolerance, unable to walk upstairs to his apartment. His urine had become “the color of eggplant”. On exam, he had an elevated JVP, new lower extremity edema, and a holosystolic murmur.Initial laboratory evaluation revealed large hemoglobin on urinalysis without RBCs, hemoglobin 9.5 (baseline 11.2), platelets 230, creatinine 1.28 (baseline 0.93), total bilirubin 2.2. Additional testing revealed reticulocytosis, schistocytes 2-5/HPF on smear, haptoglobin < 6 mg/dL, and LDH 1722. Direct Coombs test was weakly positive. The patient underwent a transesophageal echocardiogram (TEE) showing several mobile echo-densities on the aortic cusps, moderate paravalvular mitral leak, pulsatile aneurysmal segment near the aortic valve, and severe tricuspid regurgitation. He was started on antibiotics for possible endocarditis and transferred to Cardiology for management of the paravalvular leak (PVL). He underwent aortic and mitral valve replacement, occlusion of the aortic cavitary lesion, and tricuspid valve repair. Intraoperative findings included inflammatory changes along prior suture lines without vegetations. Postoperatively, blood cultures remained negative and antibiotics were stopped. His hemoglobin stabilized at 8.6 and bilirubin, LDH, and recticulocyte count returned to normal. The patient was discharged and was asymptomatic a month later in clinic.
Discussion: Anemia is commonly encountered in the hospital. This patient presented with dark urine and heart failure, prompting consideration of entities where these findings overlap. His hemoglobinuria rather than true hematuria and schistocytes on peripheral smear warranted urgent evaluation for hemolysis. The undetectable haptoglobin and elevated LDH and plasma free hemoglobin were diagnostic: a haptoglobin of < 25 mg/dL has a sensitivity of 83% and specificity of 96% for hemolytic anemia.1 Clinically significant hemolysis after cardiac valve replacement is a known but rare occurrence, estimated to be < 1% of cases.2 It is more commonly seen with mechanical valves or due to either PVLs or valve degradation, both of which likely contributed in this case.2 In order to evaluate prosthetic valves, a TEE is often needed to fully evaluate the mitral position. In this case, the TEE characterized the severity of the PVL and better visualized the damaged aortic valve. Ultimately, management of PVLs include leak closure or valve replacement. 3In addition to following hemoglobin, LDH, and haptoglobin, monitoring the bone marrow response via reticulocyte count is critical in patients with hemolysis. Patients with an appropriate peripheral reticulocytosis, especially with acute on chronic hemolysis, often require folate and iron supplementation.
Conclusions: This case of a patient with bioprosthetic valve replacements and symptomatic hemolytic anemia illustrates the importance of evaluating mechanical causes for hemolysis. After repair of the mitral PVL and aortic valve replacement, the patient’s laboratory markers of hemolysis and symptoms of heart failure resolved. While clinically significant anemia from valve replacement is rare, the advent of new cardiac devices and techniques makes it likely that more inpatient and outpatient providers will encounter this in their practice.