Relationship of beta-hydroxybutyrate levels and ketosis duration with diagnostic FDG-PET studies performed for the evaluation of active cardiac sarcoidosis

Introduction: FDG-PET is the non-invasive standard for detecting and monitoring myocardial inflammation related to active cardiac sarcoidosis (CS). Preparation with the ketogenic diet (KD) is required to induce a metabolic switch that suppresses physiologic myocardial glucose uptake (MGU), thereby highlighting pathologic MGU. Despite this, incomplete physiologic myocardial glucose suppression (MGS) continues to limit the diagnostic yield of FDG-PET. We have recently shown that beta-hydroxybutyrate (BHB) levels correlate with MGS in healthy volunteers. Yet, its use in clinical practice and relationship with ketosis duration remain unexplored.

Methods: We prospectively included patients who underwent FDG-PET for the evaluation or management of CS. Serum BHB levels were measured immediately before FDG injection. PET images were classified as 1) diagnostic (complete physiologic MGS) when there was either no MGU (negative) or focal/multifocal uptake (positive), or 2) nondiagnostic (incomplete physiologic MGS) when there was either diffuse (left ventricular) or nonspecific uptake (e.g., lateral wall or basal ring patterns) (Figure A). Area under the receiver operating characteristic curve (AUC) was used to evaluate the ability of BHB to predict a diagnostic scan.

Results: This study included 290 unique patients (age 59.3 ± 10.6 years; 63.4% male; 45.5% non-White; LVEF 48.4 ± 15.2%; 51.4% with biopsy-proven sarcoidosis) who underwent FDG-PET scans that were interpreted as the following: 219 (75.5%) negative; 44 (15.2%) positive; 9 (3.1%) diffuse; 18 (6.2%) nonspecific uptake. The KD was followed for 1-2 days in 154 patients and for ≥3 days in 136 subjects. There was an association between MGS and KD duration as the rate of diagnostic scans increased (87.0% vs. 94.8%), whereas the proportion of nondiagnostic scans (13.0% vs. 5.2%; p=0.02) decreased significantly in subjects following the KD for ≥ 3 days (versus <3 days). BHB levels increased significantly with ketosis duration (p<0.0001). Patients with diagnostic scans had higher BHB levels than patients with nondiagnostic scans (negative: median [IQR] BHB level 0.5 mmol/L [0.2, 0.9]; positive: 0.3 mmol/L [0.1, 0.7]; diffuse: 0.1 mmol/L [0.1, 0.1]; nonspecific uptake: 0.1 mmol/L [0.1, 0.3]; p<0.001). BHB levels showed excellent discrimination for a diagnostic study (AUC 0.81; 95% confidence interval [0.74, 0.89]) (Figure B). A BHB ≥ 0.53 mmol/L showed a specificity of 96% to predict a diagnostic FDG-PET scan, whereas BHB ≤ 0.1 mmol/L only had modest sensitivity (65%) to predict incomplete MGS.

Conclusions: Longer ketosis and higher BHB levels are strongly associated with diagnostic FDG-PET studies in patients undergoing evaluation for CS. Measurement of BHB levels prior to FDG-PET may help clinicians differentiate between pathologic and nonspecific/physiologic MGU.

• Download figure
• Open in new tab
• Download powerpoint