Positron Emission Tomography/Computed Tomography for Diagnosis of Prosthetic Valve Endocarditis: Increased Valvular 18F-Fluorodeoxyglucose Uptake as a Novel Major Criterion

Study Questions:

Does 18F-fluorodeoxyglucose positron emission tomography/computed tomography (18F-FDG PET/CT) imaging improve the ability to identify and exclude prosthetic valve endocarditis (PVE)?


This study examined 72 consecutive patients with suspected PVE and with echocardiography (both transthoracic and transesophageal) and microbiology testing, who underwent cardiac PET/CT imaging on admission. The diagnostic accuracy of PET/CT alone and in combination with existing criteria was assessed, with the final diagnosis defined by modified Duke criteria during 3-month follow-up used as the reference standard.


Mean age was 67 years, and 61% were male. Prostheses were bioprosthetic and mechanical in 44 and 28 cases, respectively. Mean radiation dose was 15 mSv. Final diagnoses were definite, possible, and rejected PVE in 30, 22, and 20 individuals, respectively. When considering only patients with a final diagnosis of definite or rejected PVE, the sensitivity and specificity of PET/CT to identify PVE were 73% and 80%. In patients with a final diagnosis of “definite PVE,” the addition of abnormal PET/CT uptake as a major criterion for PVE using the modified Duke criteria increased the proportion of individuals with an initial diagnosis of “definite PVE” from 70% to 97% (p = 0.008). Further, in patients with a final diagnosis of “rejected PVE,” the incorporation of PET/CT findings into the initial modified Duke criteria decreased the proportion with “rejected PVE” as an initial diagnosis from 50% to 40%.


The authors concluded that use of 18F-FDG PET/CT imaging increases early identification of PVE.


While early diagnosis of PVE is important, it is often challenging, particularly as acoustic shadowing from the prosthesis may limit the ability of echocardiography to visualize valve cusps. By identifying evidence of an active inflammatory/infectious process at the site of a prosthetic valve, addition of 18F-FDG PET/CT findings to the modified Duke criteria significantly enhanced the diagnostic sensitivity of these criteria to identify PVE. The increase in sensitivity from 70% to 97% is striking, although future studies are needed to verify these findings in other populations and to determine which patients may benefit from PET/CT imaging.

Clinical Topics: Noninvasive Imaging, Computed Tomography, Echocardiography/Ultrasound, Nuclear Imaging

Keywords: Prostheses and Implants, Fluorodeoxyglucose F18, Follow-Up Studies, Cytoskeletal Proteins, Tomography, X-Ray Computed, Radiopharmaceuticals, Thorax, Endocarditis, Bacterial, Echocardiography, Positron-Emission Tomography

< Back to Listings