Diagnostic accuracy and prognostic value of simultaneous hybrid 18F-fluorodeoxyglucose positron emission tomography/magnetic resonance imaging in cardiac sarcoidosis

Abstract

Aims: Cardiac death is the leading cause of mortality in patients with sarcoidosis, yet cardiac involvement often remains undetected. Cardiovascular magnetic resonance imaging (CMR) and 18F-fluorodeoxyglucose (FDG)-positron emission tomography (PET) have been used to diagnose cardiac sarcoidosis (CS) yet never simultaneously in a cohort. This study sought to assess the diagnostic and prognostic utility of simultaneous hybrid cardiac PET/MR. Methods and results: Fifty-one consecutive patients with suspected CS (age 50 ± 13 years, 31 males) underwent simultaneous PET/MR following a high-fat/low-carbohydrate diet and 12-h fast. Blinded image analysis of FDG uptake and late gadolinium enhancement (LGE) was performed using the American Heart Association (AHA) 16-segment model. The sensitivity and specificity of PET/MR for diagnosing CS was estimated using the Japanese Ministry of Health and Welfare guidelines. The primary endpoint was a composite of death, aborted sudden cardiac death, sustained ventricular arrhythmia, complete heart block, and hospital admission with decompensated heart failure. The secondary endpoints were a fall in left ventricular ejection fraction (LVEF) >10%, non-sustained ventricular tachycardia and other cardiac-related hospital admission. The prevalence of CS was 65% (n = 33). The sensitivity of PET and CMR alone for detecting CS was 0.85 and 0.82, respectively. Hybrid PET/MR was superior for detecting CS with sensitivity, specificity, positive, and negative predictive values of 0.94, 0.44, 0.76, and 0.80, respectively. There was poor inter-modality agreement for the location of cardiac abnormalities (k = 0.02). Over the median follow-up of 2.2 years, there were 18 (35%) adverse events. Cardiac RV PET abnormalities and presence of LGE were independent predictors of adverse events. Abnormalities found on both PET and magnetic resonance imaging was the strongest predictor of major adverse cardiac events. Conclusion: Simultaneous PET/MR is an accurate method for diagnosing CS. FDG-PET and CMR combined offers complementary information on disease pathophysiology. The presence of LGE and FDG uptake on PET/MR identifies patients at higher risk of adverse events. PET and CMR should therefore be considered in the assessment of disease presence, stage, and prognosis in CS.