Case Presentation: The patient is a 70-year-old woman with history of Streptococcus agalactiae native valve infectious endocarditis (IE) complicated by multifocal thalamic infarcts with subsequent porcine mitral valve repair 7 months prior who presented to an outside hospital for generalized weakness and fatigue. On day 4 of admission, she developed a fever and was found to have multiple blood cultures positive for MRSA. Transthoracic echocardiogram (TTE) showed echodensities concerning for mitral valve vegetations, so she was transferred for further evaluation and started on therapy for MRSA prosthetic valve endocarditis (PVE). A transesophageal echocardiogram (TEE) performed shortly after transfer showed no evidence of valvular vegetations. Cardiology reviewed the outside hospital TTE and felt that the echodensities were likely struts and not pathologic findings. Despite negative TEE, there was still high clinical suspicion and “possible endocarditis” based on Duke criteria, so PET/CT was pursued, ultimately showing intense circumferential FDG uptake in the mitral valve highly suspicious for mitral valve prosthesis infection and a cavitary lung lesion concerning for septic embolus. Cardiothoracic surgery evaluated the patient and recommended nonoperative management given her stability. She completed a course of gentamicin while inpatient and was discharged on vancomycin and rifampin to complete a 6-week course with follow-up with infectious disease.
Discussion: TEE remains the most sensitive and specific imaging modality for diagnosing IE. However, it has limitations, particularly in the presence of prosthetic material or when paravalvular complications are suspected, which can lead to delayed diagnosis. PVE can have inconclusive echocardiography and culture data in up to 20% of cases. In these settings, adjunct imaging modalities such as PET/CT have become increasingly valuable for improving diagnostic accuracy. PET/CT has moderate sensitivity and high specificity for detecting PVE and cardiac implantable electronic device endocarditis. It can also detect other foci of involvement, such as the cavitary lung lesion in this patient, or delineate a source. It is important to note that PET/CT demonstrates substantially lower sensitivity in native-valve IE and thus cannot be used to exclude the diagnosis. Cardiac CT (CCT) is another adjunct imaging modality that demonstrates higher sensitivity than TEE for detecting paravalvular abscesses and pseudoaneurysms but remains inferior in identifying vegetations, leaflet perforations, and paravalvular regurgitation. Both studies, when used in conjunction with TEE, provide a more comprehensive evaluation by compensating for each modality’s individual limitations. Reflecting their growing diagnostic importance, both CCT and PET/CT were incorporated as major imaging criterion in the 2023 revision of the Duke criteria for infective endocarditis.
Conclusions: PET/CT has emerged as an increasingly valuable adjunct imaging modality in the diagnosis of infective endocarditis. The revised 2023 Duke Criteria recommend its use in clinically challenging cases in which the suspicion for endocarditis remains high despite equivocal or negative TEE findings, which are more common with prosthetic valve involvement as illustrated in this case. The integration of PET/CT and CCT enhances diagnostic accuracy and supports more informed clinical decision-making, particularly regarding surgical intervention and optimization of antibiotic therapy.