Improvement of diagnostic accuracy of 18fluorine-fluorodeoxyglucose PET/computed tomography in detection of infective endocarditis using a 72-h low carbs protocol

New criteria for the diagnosis of infective endocarditis using 18F-FDG PET/CT imaging

OBJECTIVES

The purpose of this study was to examine a qualitative scoring system and a semi-quantitative method of FDG-PET/CT imaging in the diagnosis of infective endocarditis (IE).

METHODS

We retrospectively included 108 patients who underwent 18F-FDG-PET/CT for suspected IE. PET/CT scans were interpreted using a 4-point score (0 = no uptake; 1 = cardiac uptake < blood-pool activity; 2 = blood-pool < uptake < liver activity; 3 = uptake > liver) and semi-quantitatively using SUVmax and SUVmean of the suspected valve lesion, liver, spleen, and of the bone marrow (BM). BM and spleen SUVmean were normalized to the liver, and hypermetabolism (HSBM) was defined as a BM or spleen-to-liver ratio > 1.

RESULTS

Comparing the score criteria results with the clinical final diagnosis, the sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of qualitative criteria were 93%, 81%, 84%, 91%, and 87%, respectively. A definite IE was documented in 83% of patients showing HSBM and abnormal cardiac uptake, 44% with abnormal cardiac uptake, 28% with HSBM, and 10% with neither one.

CONCLUSION

The qualitative scoring system is helpful in the diagnosis of IE. The HSBM is an additional, reliable indirect sign of IE.

18F-FDG-Uptake in Mediastinal Lymph Nodes in Suspected Prosthetic Valve Endocarditis: Predictor or Confounder?

Introduction: Prosthetic valve endocarditis (PVE) is a serious disease affecting ~0.4% of prosthetic valve recipients per year. ¹⁸F-FDG-PET/CT has high sensitivity and specificity for PVE and is included as major criterion for the diagnosis in recent guidelines of the European Society of Cardiology. We addressed the question whether increased FDG-uptake in mediastinal lymph nodes could help to support the visual diagnostic assessment of PVE. Methods: In this sub-analysis of a previously published retrospective multicentre study, 160 unique patients were identified who underwent ¹⁸F-FDG-PET/CT for evaluation of suspected PVE. ¹⁸F-FDG-PET/CT was performed in adherence to the European Association of Nuclear Medicine guidelines of 2015 and scans were assessed for signs of mediastinal lymph node activity by 2 experienced nuclear medicine physicians who were blinded to clinical context. Clinical diagnosis of PVE had been established based on surgical findings or multidisciplinary consensus after a 1-year follow-up in 80 of 160 patients (50%). Results: In total, 52 patients showed increased mediastinal lymph node activity. Mediastinal lymph node activity on ¹⁸F-FDG-PET/CT did not increase diagnostic accuracy when added to the visual analysis of scans for signs of PVE: X ²: 0.118, p = 0.731). After excluding patients with known confounders for ¹⁸F-FDG-PET/CT, namely use of Bioglue® during prosthetic valve implantation and C-reactive protein levels below 40 mg/L, mediastinal lymph node activity was still not of additional diagnostic value compared to visual analysis alone (X²:0.129, p = 0.723). Discussion: Assessment of mediastinal lymph node activity did not improve ¹⁸F-FDG-PET/CT diagnostic accuracy for suspected PVE compared to visual assessment of the valve alone, as it seems to be a rather a specific finding, that might be caused by sternal wound or mediastinal infections or even by subclinical respiratory infections. Future studies might elucidate whether increased FDG active lymph nodes indicate a high-risk patient group and whether more detailed assessment of mediastinal lymph nodes could improve their additional diagnostic benefit.