Statistics from Altmetric.com
Today, cardiovascular MRI (CMR) is widely used for diagnosis and therapeutic decision making in the setting of different cardiac diseases. CMR allows to analyse anatomical and functional parameters, and enables a non-invasive, accurate and repeatable assessment of changes in myocardial tissue. For example, contrast-enhanced CMR techniques such as late gadolinium enhancement (LGE) imaging enable an accurate, however, unspecific detection of myocardial damage, caused by myocardial infarction (MI). In addition, T2 and T2*-weighted CMR techniques allow to detect myocardial oedema as well as myocardial haemorrhage (in case of acute MI) and thereby provide additional information on infarct pathology and prediction of adverse outcome—at least in some cases.
Regarding the initiation of timely and adequate therapy, diagnosis of myocardial inflammation in the early phase of heart disease—before the occurrence of structural changes in the myocardium—is crucial. Unfortunately, T2-weighted CMR techniques that promise to depict myocardial oedema as the first morphological change in the sequence of events have a limited sensitivity, and are questioned with regard to their diagnostic value by some groups.1 Also novel T2-mapping techniques promise an improved diagnostic yield, neither LGE-based imaging nor T2- and T2*-weighted imaging do allow a direct visualisation of infiltrating macrophages in the infarcted myocardium, but rather depict myocardial tissue changes that are ‘associated’ with or the consequence of ongoing myocardial inflammation and healing. Hence, new and more specific CMR tools are needed to improve both early diagnosis and specificity of diagnosis.
From a clinical point of view, the opportunity to directly image infiltrating macrophages in the infarcted myocardium will be of broad diagnostic as well as therapeutic value since the extent and degree of myocardial disease might be monitored more accurately and the therapeutic success be assessed more appropriately.2 In this context, two distinct phases in the process of infarct healing were described recently3: a first inflammatory phase (until day ~4 following acute MI) that is characterised by the accumulation of proinflammatory monocytes (CD16−) that primarily remove necrotic cells and debris and a second resolution phase (following day ~4 after acute MI) in which reparative monocytes/macrophages (CD16+) predominate and orchestrate myocardial tissue repair. Noteworthy, recent preclinical studies suggest that a misbalance between these two phases (eg, an excessive accumulation of inflammatory monocytes during the first phase) may have adverse effects on infarct healing.3 4 Therefore, an improvement of imaging opportunities aiming at detecting even specific subsets of monocytes/macrophages in the human myocardium in order to allow an improved understanding of monocyte/macrophage trafficking and a better treatment of the infarcted heart muscle is needed.
Theoretically, a CMR approach based on targeting of macrophages infiltrating the infarct zone and promoting infarct healing by using ultrasmall superparamagnetic iron oxide nanoparticles (USPIOs) may lead to a more detailed characterisation of infarct pathology in patients with acute MI. USPIOs are composed of a magnetic core and a non-magnetic coating. The core consists of nanometre-sized, monodispersed iron oxide particles of magnetite (Fe3O4) and/or maghemite (γ-Fe2O3) while this core is coated by a polymer shell that stabilises those nanoparticles and determines their pharmacokinetic properties (depending on the surface chemistry such as composition, density and charge). Since USPIOs are ‘superparamagnetic’ compounds, their magnetic moments align in the presence of an external magnetic field and thereby cause a much larger magnetic susceptibility compared with most paramagnetic (eg, gadolinium-based) agents. Compared with gadolinium-based compounds, USPIOs are referred to as negative contrast agents, since the magnetic field inhomogeneities caused by USPIOs alter the relaxation properties in neighbouring tissue and thereby result in a signal void in T2/T2*-weighted images.
In the past years, USPIO-based molecular imaging agents have been developed and successfully applied in animal models of MI.5 6 Recently, ferumoxytol—an USPIO-based compound—was investigated as a potential CMR contrast agent to detect cellular inflammation in patients with acute MI.7 8 The results of these first clinical studies suggested that ferumoxytol enables a detailed characterisation of acute MI pathology by detecting infiltrating macrophages and altered perfusion kinetics. In this study of Stirrat et al, serial CMR scans were performed in 31 patients with acute MI—comprising both patients with ST-elevation myocardial infarction (STEMI) and non-ST-elevation myocardial infarction (NSTEMI)—and the authors compared their results of T2*-mapping-based detection of USPIO accumulation to T2-mapping-based detection of myocardial oedema.9 Moreover, the authors obtained myocardial biopsy specimens from the region of healing myocardium (mentioned to be taken from around the infarct zone) in three out of four patients undergoing bypass surgery and demonstrated co-localisation of iron and macrophage distribution in the area of healing myocardium whereas neither iron nor macrophages were seen within the adjacent regions of healthy myocardium. The authors’ major results can be summarised as follows: while myocardial oedema (detected by T2 mapping) persisted for at least 3 months, presence of macrophages/monocytes in the infarcted myocardium (detected by T2* mapping following USPIO administration) was observed for 2 weeks only—confirming the aforementioned monocyte/macrophage-driven resolution phase of infarct healing. Therefore, the authors argue that the observed pattern of cellular inflammation (using USPIO) is distinct, and provides complementary information to the more prolonged myocardial oedema in the area of infarcted myocardium.
First, this is the first clinical study being performed so far that focused on cardiac imaging using serial (up to three times) intravenous administration of ferumoxytol. Since this was obviously an investigator-initiated trial with an off-label use of ferumoxytol, which was approved at that time for iron replacement therapy in patients with anaemia due to chronic renal failure, the authors have to be congratulated for successfully performing such a challenging clinical study and shouldering the exhausting medicolegal challenges and barriers that are associated with such a clinical trial. Keeping this in mind and considering the fact that in contrast to gadolinium-based contrast agents, there is no relevant renal elimination of ferumoxytol and also no reports of brain deposits following the administration of ferumoxytol, the data presented by Stirrat et al deserve great attention.
Obviously, there are some minor shortcomings and limitations in this study. The study design is somewhat challenging, since the study patients received up to seven CMR scans over a 3-month period following acute MI using ‘a variable regime’ (with a quite large range for each time point) and therefore, some patients received up to three infusions of ferumoxytol while others received just one single infusion. Although the authors’ respective explanation that their study design allowed them to address a couple of study questions at the same time is conceivable, a more standardised study protocol (still answering all the questions raised by the authors) might have been possible. Moreover, the first CMR study following ferumoxytol administration was performed not before 24 hours after ferumoxytol infusion. In this context, previous studies suggested that ferumoxytol effects are detectable already 6 hour after ferumoxytol administration while differentiated human macrophages demonstrated ferumoxytol uptake not before 24 hours of incubation.8 Therefore, an (additional) direct ferumoxytol effect caused by altered tissue partitioning of USPIOs in different layers of infarcted myocardium with varying severity of structural damage was supposed. Due to the lack of CMR studies within the first hours of ferumoxytol administration, this study does not allow to address this important issue.
However, these aforementioned limitations do not really weaken the major results of this study: (1) T2* mapping following USPIO administration allows to depict a different pathology (supposably the presence of macrophage-based cellular inflammation) compared with T2-based sequences that are routinely used for oedema imaging; (2) while USPIO-associated signal void in T2* mapping suggestive of macrophage-based cellular inflammation is present for 2 weeks after acute MI only, myocardial oedema (detected by T2 mapping) persists for at least 3 months. The authors’ explanation of this observation by a prolonged restoration of capillary integrity taking several months after acute MI and thereby allowing myocardial oedema to be present in T2-weighted images up to 3 months is at least questionable, since leaky capillaries will theoretically also allow USPIOs to leave the capillary bed and enter the interstitial myocardial tissue.
Nevertheless, USPIO-based T2* mapping clearly allows to non-invasively detect and monitor myocardial changes (supposably the presence of macrophage-based cellular inflammation) occurring within the first 2 weeks following acute MI, which was not possible so far. In contrast to established techniques such as LGE imaging and/or T2-weighted oedema imaging, this novel USPIO-based approach allows to safely assess whether a recent MI occurred within the last 2 weeks or rather before. Future studies need to show whether this (timely more accurate) information will have clinical implications such as whether or not to perform a more aggressive therapy or a better assessment of a potential future recovery of left ventricular systolic function following MI (that can be expected in case of a recent MI within the last 2 weeks, but will be rather absent in case of an already 3 months old MI). Moreover, future applications of USPIO may also comprise other cardiac diseases that are characterised by myocardial inflammation such as myocarditis, cardiac transplant rejection or cardiac sarcoidosis. In this context, the authors’ efforts in performing a multicentre study and addressing those aforementioned cardiac diseases are promising and highly appropriate. In view of the unique features of USPIO such as ferumoxytol and the ongoing discussions regarding gadolinium-based compounds, iron oxide nanoparticles will continue to play an important role in future non-invasive imaging.
Competing interests None declared.
Provenance and peer review Commissioned; internally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.