Skip to main content


To the Editor:

Fluorodeoxyglucose (FDG)–positron emission tomography (PET) imaging for assessment of cardiac sarcoidosis (CS) is limited by physiological cardiomyocyte uptake of FDG (1). Techniques for suppressing physiological uptake include fasting, fatty meals, and heparin infusion; however, nonspecific uptake is present in up to 20% of patients (2). As a glucose analogue, FDG is inherently nonspecific, as numerous pathologies show increased uptake. 3′-Deoxy-3′-[18F]-fluorothymidine (FLT) is a PET radiotracer with uptake determined by cellular proliferation, including sarcoidosis granulomas; it does not require extensive patient preparation due to a lack of physiological myocardial uptake (1,3). Absence of normal cardiac activity and specificity of the uptake mechanism may translate into increased patient convenience and improved diagnostic accuracy of the test. This prospective study examined the use of FLT-PET imaging for the evaluation of CS.

Fourteen subjects were referred for FDG/perfusion PET imaging for assessment of active CS. Five of the 14 subjects had known CS, and the indication for the PET scan was for reassessment of disease activity; for 9 subjects, the indication was for initial assessment. All patients were enrolled and imaged with FLT-PET within 2 weeks of the FDG-PET scan, with no interval initiation or change in treatment. In all, 3 of 14 subjects were receiving treatment at the time of PET imaging (2 methotrexate, 1 prednisone). Twelve of 14 subjects had CS in keeping with the Heart Rhythm Society criteria (4). Before FDG-PET imaging, subjects were prepared with low-carbohydrate/high-fat/protein-permitted diets followed by fasting (12 h) and unfractionated heparin before administration of FDG (2). Rest perfusion PET imaging (13N ammonia or rubidium-82) was performed on the same day. Before FLT-PET imaging, participants were instructed to fast for 6 h. No other preparation was specified.

Two experienced, blinded readers independently reviewed the anonymized and randomized FLT, FDG, and perfusion studies. A third reader resolved differences in interpretation. Images were reviewed in conjunction with the perfusion images in accordance with the recent Society of Nuclear Medicine and Molecular Imaging–American Society of Nuclear Cardiology guidelines (2). Abnormal FDG or FLT uptake was interpreted as positive for active CS.

No studies were considered nondiagnostic. Six subjects had uptake consistent with CS on both FDG and FLT (FDG+/FLT+), and 1 subject had a positive FDG and a negative FLT study. In FDG+/FLT+ subjects, the mean number of affected segments was not significantly different (FDG vs. FLT: 8.3 ± 2.1 vs. 6.3 ± 2.0; p = 0.12), whereas the distribution of involved segments was significantly different (paired Student’s t-test, p = 0.048). In this subgroup, the segmental FLT and FDG mean standardized uptake value [SUVmean] was only weakly correlated (r = 0.23; 95% confidence interval [CI]: 0.03 to 0.40; p = 0.021). No adjustments for correlated observations within individuals were made.

The sum rest score (SRS) strongly correlated with FLT SUVtotal (r = 0.90; 95% CI: 0.33 to 0.99; p = 0.014) but not with FDG SUVtotal (p = 0.75) (Figure 1). In FDG+/FLT+ subjects, the distribution of perfusion defects was significantly different from the distribution of affected segments for both FDG and FLT (p < 0.001 and p = 0.034, respectively). Mean segmental SUV was weakly (FDG: r = 0.22; 95% CI: 0.05 to 0.39; p = 0.014) and moderately (FLT: r = 0.49; 95% CI: 0.32 to 0.62; p < 0.0001) correlated with perfusion defect severity.

Figure 1.
Figure 1.

Relationship Between FLT, FDG, and Perfusion PET

(A)(Top) Mean segmental 3′-deoxy-3′-[18F]-fluorothymidine (FLT) versus fluorodeoxyglucose (FDG) standardized uptake values (SUVs) (normalized to blood pool activity) in subjects with positive FDG studies. Some segments exhibited predominantly inflammatory activity (high FDG SUVs/low FLT SUVs), whereas some were more strongly associated with scarring (high FLT SUVs/low FDG SUVs). (Bottom) Relationship between SUVtotal and sum rest score (SRS) in FDG+/FLT+ subjects. (B) Cardiac uptake with FLT (top) and FDG (bottom) in the same subject. PET = positron emission tomography.

To the best of our knowledge, this study is the first to examine the relation between FLT-PET, FDG-PET, and perfusion PET imaging in CS. Our results suggest that FLT provides information distinct from FDG and perfusion imaging. Overall, FLT uptake was strongly correlated with myocardial scarring, yet differed significantly in distribution from both perfusion defects and FDG-positive segments reflecting differences in uptake mechanisms between the tracers. Given the strong correlation between FLT uptake and the SRS, we hypothesize that FLT may be identifying areas likely to develop myocardial scar. It is unclear whether the use of FLT-PET imaging will enhance prognosis or if treatment based on these findings would lead to better outcomes.

  • 1. Martineau P., Pelletier-Galarneau M., Juneau D., Leung E., Birnie D. and Beanlands R.S.B. : "Molecular imaging of cardiac sarcoidosis". Curr Cardiovasc Imaging Rep 2018; 11: 6.

    CrossrefGoogle Scholar
  • 2. Chareonthaitawee P., Beanlands R.S., Chen al. : "Joint SNMMI-ASNC expert consensus document on the role of 18F-FDG PET/CT in cardiac sarcoid detection and therapy monitoring". J Nucl Med 2017; 58: 1341.

    CrossrefMedlineGoogle Scholar
  • 3. Norikane T., Yamamoto Y., Maeda Y., Noma T., Dobashi H. and Nishiyama Y. : "Comparative evaluation of 18F-FLT and 18F-FDG for detecting cardiac and extra-cardiac thoracic involvement in patients with newly diagnosed sarcoidosis". EJNMMI Res 2017; 7: 69.

    CrossrefMedlineGoogle Scholar
  • 4. Birnie D.H., Sauer W.H., Bogun al. : "HRS expert consensus statement on the diagnosis and management of arrhythmias associated with cardiac sarcoidosis". Heart Rhythm 2014; 11: 1304.

    CrossrefGoogle Scholar


Please note: Drs. Beanlands and Birnie are co-senior authors of this work. The work has been supported in part by a UOHIAMO grant from the University of Ottawa Heart Institute; a research trial grant from the Ministry of Health and Long Term Care Research (grant no. 06374) for the Ontario PET Cardiac Sarcoidosis Trial in collaboration with the PET Steering Committee of Ontario; and by the Canadian Institutes of Health Research (Dr. Birnie, Principal Investigator) for CHASM-CS (Cardiac Sarcoidosis Cohort Study; grant no. 342139, NCT01477359). Dr. Beanslands is a uOttawa Tier 1 Research Chair and the UOHI Vered Chair in Cardiology. Dr. Birnie is a uOttawa Tier 1 Research Chair and the UOHI Chair in Arrhythmia Research. Dr. Beanslands reports grants and honoraria from Lantheus Medical Imaging, Jubilant DRAXImage, and GE Healthcare. Dr. Juneau has received consultant fees from AbbVie Canada (unrelated to this work). Dr. Leung has received speaker honoraria from Bayer. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.