By Michael H. Crawford, MD, Editor

SYNOPSIS: 18F-FDG PET/CT imaging in a consecutive series of patients with proven infective endocarditis was of diagnostic value in those with prosthetic value endocarditis (sensitivity = 83%) and of prognostic value in all cases.

SOURCES: San S, Ravis E, Tessonier L, et al. Prognostic value of 18 F-Fluorodeoxyglucose positron emission tomography/computed tomography in infective endocarditis. J Am Coll Cardiol 2019;74:1031-1040.

Rouzet F, Iung B, Duval B. 18 F-FDG PET/CT in infective endocarditis: New perspectives for improving patient management. J Am Coll Cardiol 2019;74:1041-1043.

PET plus CT scanning has been shown to be of value in the diagnosis of prosthetic valve endocarditis (PVE), but less is known about its prognostic value in PVE and native valve endocarditis (NVE). Investigators from a hospital in Marseille, France, performed 18 F-fluorodeoxyglucose (FDG) PET/CT as soon as possible after admission in consecutive patients with a definite diagnosis of left-sided PVE or NVE from 2011 to 2017.

Exclusion criteria included need for urgent cardiac surgery, hemodynamic instability, and blood glucose > 180 mg/L after a 12-hour fast before the PET study. The primary endpoint was a composite of major adverse cardiac events: death, recurrent endocarditis, heart failure, embolic events, and cardiac hospitalization. The study population included 173 patients: 109 with PVE and 64 with NVE. Guideline indications for surgery were present in 127 patients, 93 of whom eventually underwent surgery. Median time from first antibiotics to PET/CT scan was seven days. PET/CT was positive in 100 patients, more often in PVE (83%) than NVE (16%). After a mean follow-up of 225 days, the primary endpoint occurred in 54% of patients (58% PVE, 48% NVE). One-year mortality was 20% (23% PVE, 16% NVE). Recurrent endocarditis occurred in 9% of patients, embolic events in 13%. In PVE patients, but not the NVE patients, a positive PET/CT was associated with more frequent occurrence of the primary endpoint (hazard ratio [HR], 2.7; 95% confidence interval [CI], 1.1-6.7; P = 0.04). Also, moderate or intense positivity was associated with embolic events (HR, 7.5; 95% CI, 1.24-45.2; P = 0.03) in PVE patients and NVE patients (HR, 8.8; 95% CI, 1.1-69.5; P = 0.02). The authors concluded that PET/CT is not only of diagnostic value in endocarditis, but is predictive of major adverse cardiac events during follow-up in PVE patients and subsequent systemic emboli in NVE patients.


FDG PET/CT has been used for a long time for diagnosis and prognosis in oncology to detect areas of inflammation associated with tumors. Also, since echocardiography of prosthetic values can be challenging, it has been recommended in guidelines as a diagnostic tool for PVE if the echo is not conclusive. This study confirms the diagnostic value of PET/CT in PVE with a demonstrated sensitivity of 83%. It was not as useful for NVE (sensitivity = 16%). What this study added was information on predicting subsequent major adverse cardiac events during a one-year follow-up. Surprisingly, about half the patients experienced major events on follow-up, and a multivariate analysis showed that PET/CT was predictive of these bad outcomes in PVE patients (HR, 2.7). The only other predictor that was significant was a CRP > 100 mg/L (HR, 1.9), which makes sense. Interestingly, patients with false-negative PET/CT usually scored low CRP values. Also, PET/CT strongly predicted subsequent systemic emboli in PVE and NVE patients. Thus, this study shows that PET/CT certainly is of prognostic value, too.

Knowing a patient’s prognosis is of value only if one can do something about it. In the case of endocarditis, surgery is an option. The traditional indications for surgery are hemodynamic failure, persistent infection, large mobile vegetations, and recurrent emboli. However, surgeons often are reluctant to operate on these sick patients with a poor prognosis. Perhaps a positive PET/CT would help convince them that early surgery is worth the risk. Even in NVE patients, a strongly positive PET/CT was highly predictive of subsequent emboli. In this study, as in others, about three-quarters of emboli go to the brain. Also, a persistently positive PET/CT on antibiotic therapy may bolster the perception that the patient carries a persistent infection.

There were limitations to this study. The patient cohort was relatively small, with a low number of events — especially with NVE patients. One could consider this a study of PVE patients. There was no information offered on the effect of antibiotics on PET/CT, which could be important clinically. At this point, PET/CT certainly is worth performing in left-sided PVE. There were no data provided on right-sided or device endocarditis, but it could be useful in these patients. Also, noncardiac foci were detected in 36% of study patients, which potentially is of value to know. Some of these extra cardiac foci could be the source of the infection, which would warrant attention.

Finally, since no contrast is used, PET/CT is easy on the kidneys. On the other hand, it is a somewhat cumbersome test, with a special meal required, followed by a 12-hour fast and a documented glucose level < 180mg/L before it can be performed. The scan does involve radiation; the mean dose in this study for the whole body was 15 mSv.