Quantifying Inflammation in Infarcted Myocardial Tissue with Severely Reduced Flow: A Hybrid PET/MRI Approach Using a Prolonged Constant Infusion of 18F-FDG and Gd-DTPA

Introduction: Cardiovascular disease is the leading cause of death worldwide. Heart failure, specifically, is predominantly caused by disregulation of inflammation often seen after a heart attack ^[1]. MRI has shown promise in detecting characteristics that can increase the risk of heart failure, including: infarct size, presence of hemorrhage and presence and size of an area of extreme low blood flow within the infarct called the region of microvascular obstruction (MO) ^[2]. What is needed is an imaging method that can distinguish between pro-inflammatory (neutrophils and M1 macrophages) and anti-inflammatory (M2 macrophages) cells as disregulation occurs when the pro-inflammatory phase is prolonged. It has been shown that ¹⁸FDG/PET can, in principle, distinguish between the pro- and anti-inflammatory cell types ^[3] however post heart attack there are three problems: a) ¹⁸FDG cannot penetrate the MO after intravenous bolus injection, b) resting healthy myocardium also has uptake of ¹⁸FDG and c) partial volume effects interfere with identification of MO tissue. A constant infusion may be able to penetrate the MO while simplifying kinetic modelling ^[4,5]. Here we present initial results addressing these three issues using a canine model of heart failure post heart attack and PET/MRI imaging during a simultaneous prolonged constant infusion of ¹⁸FDG and an extracellular MRI contrast agent (a gadolinium chelate e.g. Magnevist).

Methods: Two animals were imaged at 5 days after a heart attack. During the 150-minute constant infusion of Magnevist and ¹⁸FDG, MRI T1-maps and 3D T1 weighted were acquired every 10 minutes; PET images were binned in 3-minute frames. Suppression of ¹⁸FDG uptake by cardiomyocytes was initiated at 40 minutes using a heparin injection and the start of a 50-minute lipid infusion.

Results: a) The MRI T1-maps allow the identification of the infarcted tissue and the zone of MO with relatively little partial volume (figure 1B). b) The MRI T1-maps can be used to infer the degree of penetration into the MO zone. In one animal, the MO was penetrated after 150 minutes (figure 1C) while the other did not have a visible infarct at any time (not shown). c) The ¹⁸FDG uptake in normal heart cells was eliminated after suppression. d) The suppression of myocardial glucose uptake lasted at least 60 minutes after the lipid infusion was stopped (figure 1A and 1G). e) The MRI contrast agent deposition in the infarcted tissue including the MO zone was not affected by suppression. f) The ¹⁸FDG uptake in infarcted tissue and the zone of MO was not affected by suppression. Discussion: This PET/MRI protocol can be used in the canine model of heart failure to investigate therapies to regulate the inflammatory response. This PET/MRI protocol can be used in patients to detect disregulation of the inflammatory responses and to evaluate therapy. Future Work: Additional animals need to be studied. [1] Frangogiannis, NG. (2014). Nat Rev Cardiol. 11(5): 255-265. [2] Kali A. (2016). Circulation Cardiovasc Imaging. 9(11): e004996. [3] Thackeray, J. (2017). SNMMI 2017. J Nucl Med. 58:302. [4] Prato, FS. (2015). J Nucl Med. 56(2): 299-304. [5] Wilk, B. (2016). SNMMI 2016. J Nucl Med. 57:120.

• Download figure
• Open in new tab
• Download powerpoint