18F-FDG PET/CT in the Detection of Prosthetic Valve Endocarditis: a Pilot Study

  • #CI/INT 01-EP-5
  • Cardiology and Imaging in Cardiac Surgery/Intraoperative and Early Assessment of the Results of Surgery. E-POSTER (ORAL) SESSION
  • E-Poster (oral)

18F-FDG PET/CT in the Detection of Prosthetic Valve Endocarditis: a Pilot Study

Diana M. Pursanova, Leo A. Bockeria, Iraklii P. Aslanidis, Renat M. Muratov, Olga V. Mukhortova, Irina V. Shurupova, Tatyana A. Trifonova, Irina V. Ekaeva

A.N.Bakoulev Scientific Center of Cardiovascular Surgery, Moscow, Russia

Date, time and location: 2018.05.26 17:00, Exhibition area, 1st Floor. Zone – A


Objective: The diagnosis of prosthetic valve endocarditis (PVE) remains challenging. The aim of this study was to investigate the diagnostic value of 18F-fluorodeoxyglucose positron emission tomography/computed tomography (18F-FDG PET/CT) in these patients and to compare 18F-FDG-uptake around prosthetic valves in patients with and without PVE.

Methods: This retrospective analysis included results of 18F-FDG PET/CT examinations performed in 26 patients with prosthetic valve – with possible infective endocarditis (n=13) and malignance used as negative controls (n=11).Examinations were done on PET/CT scanner (Biograph-64, Siemens) 90 min after 18F-FDG injection (175–200Mbq). Visual and semi-quantitative analysis was performed. Maximal standardized uptake value (SUVmax) in the valve area was measured and the SUVratio of SUVmax around the valve to SUVmean in the descending aorta was calculated. All PET/CT results were confirmed according to clinical, laboratory and morphological data. Ten patients had final diagnosis of definite PVE.

Results: The sensitivity, specificity and accuracy of 18F-FDG PET/CT were 90%, 86% and 88%, respectively; positive and negative predictive values – of 82% and 92%; positive and negative likelihood ratios – 6.3 and 0.1.

In patients with possible PVE PET/CT correctly confirmed the diagnosis in 9/13 of cases and ruled it out – in 3/13 of patients.

In semiquantitative analysis both SUVmax and SUVratio were higher in confirmed PVE patients compared to non-PVE patients: median 4.1 and 2.8 vs 2.7 and 1.5 (p<0.002).

SUVratio 2.3 was the optimal threshold for the PVE diagnosis using ROC-curve analysis (AUC=0.875) with sensitivity 80% and specificity 86%.

Conclusions: 18F-FDG PET/CT appears to be useful in the diagnosis of PVE – the level of 18F-FDG-uptake in the prosthetic valve area showed a good diagnostic performance. However, further studies on larger groups are needed to assess the role of 18F-FDG PET/CT in the detection of PVE.

To top