Heartbeat-to-heartbeat cardiac tissue characterization
van den Boomen, Maaike
DOI:
10.33612/diss.128413796
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below.
Document Version
Publisher's PDF, also known as Version of record
Publication date: 2020
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):
van den Boomen, M. (2020). Heartbeat-to-heartbeat cardiac tissue characterization. University of Groningen. https://doi.org/10.33612/diss.128413796
Copyright
Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
Take-down policy
If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.
Velthuis, R.H.J.A. Slart, R.J.H. Borra, N.H.J. Prakken – “Native T2and T2*reference values for
cardiomyopathies and heart transplantations: A systematic review and meta-analysis,”, Journal of Cardiovascular Magnetic Resonance, 2020;22(1):34
Chapter 3
Native T
2
and T
2
*
reference values for
cardiomyopathies and heart transplantations
A systematic review and meta-analysis
Abstract
This chapter introduces myocardial native T2- and T2*-mapping, which are
known to be sensitive to edema, necrosis, hemorrhage and iron deposition. How-ever, their clinical application is currently limited as the ranges for healthy and cardiac diseases are poorly defined. Therefore, here the clinically interesting ranges of native myocardial T2- and T2*-values in patients myocardial
infarc-tion (MI), heart transplantation, restrictive cardiomyopathy (RCM), hypertrophic cardiomyopathy (HCM), dilated cardiomyopathy (DCM), myocarditis (MC) and populations with hypertension (HT), type 2 diabetes mel-litus (DM) and obesity are determined. The standard mean difference (SMD) of the T2-values of patients with MI, heart transplantation, sarcoidosis, systemic
lupus erythematosus, amyloidosis, HCM, DCM, and MC were significantly increased compared to healthy controls which confirms the diagnostic applica-bility of native T2-mapping in those diseases. T2-values were not different in
iron overload patients, but the T2*-values for this group and for the MI patients
were significantly decreased, which demonstrates the diagnostic appliciability of native T2*-mapping in these patient populations. However, this meta-analysis
also showed that T2- and T2*-values vary strongly between studies which is not
solely related to the differences in population demographics, but also related to MR vendor, acquisition methods and analysis. Therefore, globally applicable reference values are currently difficult to define, but center specific references could serve as a solution.
3.1
Introduction
V
entricular dysfunction in ischemic cardiomyopathies is triggered by impairedcoronary blood supply to the myocardium (Felker et al. 2002). In non-ischemic cardiomyopathies (NICM) many factors contribute to heart failure (HF) including hypertrophic cardiomyopathy (HCM), dilated cardiomyopathy (DCM) and restrictive cardiomyopathy (RCM) (Seferovic et al. 2019, Wu 2007). The prevalence of HF has been rising since the year 2000 and is shown to be related to the current lifestyle in Western Society (Benjamin et al. 2017, Lam et al. 2011), with increasing populations with high cardiovascular risk (obesity, hypertension (HT) and type 2 diabetes mellitus (DM)) (Benjamin et al. 2019).Early diagnosis of cardiomyopathy is important to initiate appropriate treatment (Collins et al. 2015, Ponikowski et al. 2016). However, physical examination and the medical history are often non-specific in early diagnosis and therefore require additional assessments (Ponikowski et al. 2016, Kelder et al. 2011). Cardiovascu-lar magnetic resonance imaging (MRI) can be used to detect cardiac remodeling by measuring the left ventricular volume, mass, ejection fraction (EF) and myo-cardial fibrosis with late gadolinium enhancement (LGE) (Kuruvilla et al. 2014, El-liott et al. 2014). Nevertheless, early cardiac modifications of cardiomyopathies are often indistinguishable and difficult to differentiate from overlapping findings in patients with high cardiovascular risk such as obesity, HT and DM (Aurigemma et al. 2013, Cokkinos and Belogianneas 2016, Gonzalez et al. 2018, Yap et al. 2019). Consequently, misinterpretation of cardiac remodeling in these high cardiovascu-lar risk groups may result in incorrect cardiomyopathy suspicion, which could in-troduce overtreatment. However, the changes occurring in cardiomyopathies may affect myocardial tissue characteristics, which can be measured quantitatively by T1-, T2- and T2*-mapping as part of the cardiac MRI exam (Hamlin et al. 2014).
In line with this, the European Society of Cardiology recently described a shifting standards from the assessment of LGE towards the use of T1 and T2 mapping in
their latest statement (Celutkiene et al. 2018). Furthermore, the clinical utility of T1
-mapping has already been acknowledged and included in some diagnostic guide-lines (Ponikowski et al. 2016, Elliott et al. 2014, van den Boomen et al. 2018, Mess-roghli et al. 2017). In addition, other guidelines also advocate to include T2- and
T2*-mapping instead of T2-weighted imaging (Celutkiene et al. 2018, Messroghli
et al. 2017, Badano et al. 2015, Habib et al. 2017).
The Society for Cardiovascular Magnetic Resonance released clinical recommen-dations on parametric imaging in cardiac MRI (Messroghli et al. 2017). Since T2
-mapping values vary due to different water concentrations in the myocardium, it could be useful to assess infiltration related diseases such as iron overload and Fabry disease, and myocardial injury diseases featuring edema, necrosis, and hemorrhage formation (Messroghli et al. 2017, Thavendiranathan et al. 2012, Verhaert et al. 2011). Furthermore, T2 could contribute in the diagnosis of heart transplant rejections as
edema correlates with acute heart transplant rejection (Messroghli et al. 2017, Us-man et al. 2012). In addition to T2-mapping, T2*-mapping values mainly depend on
magnetic field inhomogeneities and are therefore clinically useful in iron related dis-eases, but also enable assessment of necrosis and hemorrhage formation (Messroghli et al. 2017, Carpenter et al. 2011, Lota et al. 2017).
Reference values of T2- and T2*-mapping in healthy subjects have been
investi-gated in multiple studies (Roy et al. 2017, Maceira et al. 2015, Granitz et al. 2019, von Knobelsdorff-Brenkenhoff et al. 2013). However, the heterogeneity of the data caused by different field strengths, imaging techniques and settings underlines the need for local reference values (Messroghli et al. 2017, von Knobelsdorff-Brenkenhoff et al. 2013). The objective of this study was to perform a meta-analysis to determine the reference values of myocardial T2and T2*in HF-related cardiomyopathies and
heart transplantations. Knowledge of these ranges can help determine the clinically applicability of quantitative techniques.
3.2
Methods
Search Strategy
The study was performed following the Preferred Reporting Items for Systemic Reviews and Meta-Analysis (PRISMA) statement (Shamseer et al. 2015) and the Cochrane Handbook for Systemic Review (Higgins and Green 2011). Three inves-tigators (G.J.H. Snel, M. van den Boomen and L.M. Hernandez) independently and systematically searched for eligible studies published between January 2011 and September 2019 in PubMed/MEDLINE and Embase that reported cardiac T2- and
T2*-mapping in humans. The search contained terms related to T2- and T2*-mapping
and specific cardiac diseases (See Appendix 3.A).
For this meta-analysis published results from randomized control trials, cohort studies and observational studies in peer-reviewed journals were accepted if they included adults with an age of 18 years and older with non-ischemic or ischemic
cardiomyopathy, heart transplant patients or people with an increased cardiovascu-lar risk and reported cardiac MRI derived T2- and/or T2*-mapping values. Studies
were excluded if the article was not available in English or full text.
Study Selection
Titles and abstracts proposed by the databases were assessed for eligibility by one author and checked by a second author (G.J.H. Snel, M. van den Boomen and L.M. Hernandez). After consensus between these investigators, the full-text reports of these eligible studies were independently assessed by two of the three investiga-tors for final inclusion. The study quality was systemically evaluated with the Newcastle-Ottawa quality assessment scale (NOS) (Margulis et al. 2014), which eval-uates quality on three domains:
• Selection and definition of study populations (0-4 stars); • Comparability of populations (0-2 stars);
• Ascertainment of exposure/outcome (0-3 stars).
Data Collection
Data was extracted from the included studies by one author and checked by a sec-ond author (either G.J.H. Snel, M. van den Boomen and L.M. Hernandez). Rele-vant data regarding patient characteristics, such as; study population, age, gender, body mass index, T2- and T2*-values, as well as MRI acquisition related information,
such as; field strength, vendor, sequence and sequence parameters were extracted. Data was reported as mean ˘ standard deviation (SD) and data reported as me-dian with interquartile or full range was converted using the methodology of (Hozo et al. 2005). Healthy control data was extracted if available.
Data Analysis
The included data were divided into two groups of reported T2- and T2*-values per
difference (SMD) and the 95% confidence intervals (CI). The heterogeneity of the in-cluded studies was defined as significant if I2ě50% (Pă0.05) by using a χ2test. This
heterogeneity was further tested by a meta-regression, sensitivity and bias analysis. Available covariates were tested for their association with the myocardial T2- and
T2*-values using a backwards elimination model and remaining significant
covari-ates (Pă0.05) were included into a mixed effect model of the data. Furthermore, publication bias was assessed by inspection of the funnel plots with the Egger re-gression asymmetry test and a sensitivity analysis was performed by omitting each study sequentially and recalculating the model. A meta-analysis was performed in each population with at least 10 published studies, as stated by PRISMA guidelines (Shamseer et al. 2015). Review Manager (RevMan, version 5.3, Cochrane Collabora-tion, Copenhagen, Denmark) was used to determine the random effect models and the package “metaphor” in R (version 3.4.1, R Foundation for Statistical Computing, Vienna, Austria) was used for the mixed effect models, bias and sensitivity analysis.
3.3
Results
Results of Literature Search
The search in PubMed and Embase revealed respectively 555 and 545 articles and one paper was manually added (Luetkens et al. 2019). After removal of the du-plicates, 706 articles remained for evaluation of title and abstract which resulted in 154 articles included for the final meta-analysis. The PRISMA flow diagram with rationale for exclusion is provided in Figure 3.1. The number of studies per pop-ulation was described as total studies (number of T2 studies and number of T2*
studies): A total of 31 (22 T2 and 13 T2*) studies were included on MI (Verhaert
et al. 2011, Durighel et al. 2016, Bulluck et al. 2016a, Bulluck et al. 2017a, Carberry et al. 2017, Carrick et al. 2016a, Kali et al. 2013, Mohammadzadeh et al. 2018, Robbers et al. 2018, Roghi et al. 2015, Yilmaz et al. 2013, Zia et al. 2012, Chen et al. 2019, Za-man et al. 2015, Nakamori et al. 2019, Tahir et al. 2017, Carberry et al. 2018, Carrick et al. 2016, Haig et al. 2019, Hausenloy et al. 2019, Krumm et al. 2016, McAlindon et al. 2015, Masci et al. 2018, Park et al. 2013, Tessa et al. 2018, White et al. 2015, An et al. 2018, Bulluck et al. 2016b, Fischer et al. 2018, Layland et al. 2017, van Heeswijk et al. 2012), 11 T2studies on heart transplantation (Usman et al. 2012, Butler et al.
2015, Dolan et al. 2019, Dolan et al. 2019a, Markl et al. 2013, Miller et al. 2014, Miller et al. 2019, Vermes et al. 2018, Yuan et al. 2018, Bonnemains et al. 2014, Odille et al. 2016), 70 (5 T2and 70 T2*) studies on iron overload (Desai et al. 2014, Fragasso et al.
Figure 3.1: Overview of study review process according to the to the Preferred Reporting Items for Systemic Reviews and Meta-Analysis (PRISMA) flow diagram (Moher et al. 2009)
2011, Kritsaneepaiboon et al. 2018, Krittayaphong et al. 2017, Barrera Portillo et al. 2013, Saiviroonporn et al. 2011, Soltanpour and Davari 2018, Acar et al. 2012, Alam et al. 2016, Alp et al. 2014, Barzin et al. 2012, Bayraktaroglu et al. 2011, Camargo et al. 2016, Delaporta et al. 2013, Di Odoardo et al. 2017, Djer et al. 2013, Ebrahim-pour et al. 2012, Eghbali et al. 2017, Fahmy et al. 2015, Feng et al. 2013, Fernandes et al. 2011a, Fernandes et al. 2017, Garceau et al. 2011, Git et al. 2015, Hanneman
et al. 2013, Hanneman et al. 2016, Junqueira et al. 2013, Kayrak et al. 2012, Kirk et al. 2011, Kucukseymen et al. 2017, Li et al. 2016, Liguori et al. 2015, Mehrzad et al. 2016, Ozbek et al. 2011, Quatre et al. 2014, Roghi et al. 2015a, Sado et al. 2015, Sakuta et al. 2010, Torlasco et al. 2018, Chen et al. 2014a, De Assis et al. 2012, De Sanc-tis et al. 2016, Marsella et al. 2011, Mavrogeni et al. 2013, Meloni et al. 2012, Mel-oni et al. 2014, Pepe et al. 2018, Pistoia et al. 2019, Pizzino et al. 2018, Positano et al. 2015, Russo et al. 2011, Wijarnpreecha et al. 2015, Barbero et al. 2016, Bayar et al. 2015, Du et al. 2017, Ferro et al. 2017, Karakus et al. 2017, Karami et al. 2017, Monte et al. 2012, Parsaee et al. 2017, Pennell et al. 2015, Piga et al. 2013, Porter et al. 2013, Vlachaki et al. 2015, Yuksel et al. 2016, Gu et al. 2013), 2 T2 studies
on sarcoidosis (Greulich et al. 2016, Puntmann et al. 2017), 4 T2 studies on
sys-temic lupus erythematosus (lupus) (Mayr et al. 2017a, Zhang et al. 2015, Hinojar et al. 2016, Winau et al. 2018), 2 T2studies on amyloidosis (Kotecha et al. 2018,
Ri-Figure 3.2: Weighted mean T2*-values with weighted standard deviation (SD) of
all included 1.5T studies per myocardial infarction (MI), iron overload (IO), hyper-trophic cardiomyopathy (HCM), dilated cardiomyopathy (DCM), and hypertension (HT). patients (black) and healthy controls (grey)
Figure 3.3:Weighted mean T2*-values with weighted standard deviation (SD) of all
included 3T studies per myocardial infarction (MI), iron overload (IO), hypertrophic cardiomyopathy (HCM), dilated cardiomyopathy (DCM), and hypertension (HT). patients (black) and healthy controls (grey). With no reported values on hypertension at 3T.
douani et al. 2018), 2 T2 on Fabry disease (Messalli et al. 2012, Knott et al. 2019),
4 (2 T2 and 2 T2*) studies on HCM (Gastl et al. 2019, Kanzaki et al. 2016, Amano
et al. 2017, Park et al. 2018), 9 (7 T2 and 2 T2*) studies on DCM (Kanzaki et al.
2016, Nagao et al. 2015, Ito et al. 2015, Kono et al. 2014, Nishii et al. 2014, Spieker et al. 2018, Cui et al. 2018, Mordi et al. 2016, Child et al. 2018), 19 T2 studies on
myocarditis (Thavendiranathan et al. 2012, Luetkens et al. 2019, Baessler et al. 2017, Baessler et al. 2018, Baessler et al. 2019, Bohnen et al. 2015, Dabir et al. 2019, Gatti et al. 2019, Luetkens et al. 2017, Lurz et al. 2016, Radunski et al. 2014, Radunski et al. 2017, Spieker et al. 2017, Huber et al. 2018, Mayr et al. 2017, von Knobelsdorff-Brenkenhoff et al. 2017, Gang et al. 2019, Stirrat et al. 2018) and 1 T2*studie on HT
(Chen et al. 2018) (Appendix 3.B). The absolute T2- and T2*-values are known to be
mapping values were noted separately per field strength and this was also used as covariate in the meta-regression analysis. T2- and T2*-mapping obtained in control
subjects were recorded as values from healthy subjects, unless the control popula-tion was explicitly defined otherwise in the “populapopula-tion” column of Appendix 3.B.
Figure 3.4:Weighted mean T2-values with weighted standard deviation (SD) of all
included 1.5T studies per myocardial infarction (MI), transplantation (trans), iron overload (IO), sarcoidosis (SA), Lupus, amyloidosis (AM), Fabry disease (Fabry), hypertrophic cardiomyopathy (HCM), dilated cardiomyopathy (DCM), and myo-carditis (MC) patients (black) and healthy controls (grey)
Study Quality
None of the included studies received the maximum NOS quality score (Appendix 3.B). All studies without healthy controls automatically received limited scores in the matching and selection section. Only 57 of the 154 included studies reported control values of healthy subjects and were eligible for inclusion further analysis. However, the case definition of patients and the ascertainment of mapping values
Figure 3.5: Weighted mean T2-values with weighted standard deviation (SD) of
all included 3T studies per myocardial infarction (MI), transplantation (trans), iron overload (IO), sarcoidosis (SA), Lupus, amyloidosis (AM), Fabry disease (Fabry), hypertrophic cardiomyopathy (HCM), dilated cardiomyopathy (DCM), and myo-carditis (MC)). patients (black) and healthy controls (grey). No values were reported on cardiac transplantations, amyloidosis, Fabry disease and HCM at 3T.
were adequate in all studies.
Myocardial Infarction
The weighted mean T2*-values at 1.5T were 28.5˘6.8ms in MI patients and 34.7˘3.7
ms in healthy controls (Durighel et al. 2016, Bulluck et al. 2016a, Bulluck et al. 2017a, Carberry et al. 2017, Carrick et al. 2016a, Kali et al. 2013, Mohammadzadeh et al. 2018, Robbers et al. 2018, Roghi et al. 2015, Yilmaz et al. 2013, Zia et al. 2012) (Appendix 3.B and Figure 3.2). At 3T, these were 22.0˘3.7ms in MI patients and 29.6˘2.7ms in controls (Chen et al. 2019, Zaman et al. 2015) (Appendix 3.B and Fig-ure 3.3). The meta-analysis confirmed the significant decrease in T2*-values in MI
patients (SMD=-1.99, 95%CI [-2.70, -1.27], Pă0.01, I2=98%, Figure 3.6). All
stud-ies performed cardiac MRI in ST-elevation myocardial infarction (STEMI) patients post percutaneous coronary intervention (PCI) in the acute phase, except for one (Robbers et al. 2018). Some studies performed follow-up in these patient groups (Carberry et al. 2017, Carrick et al. 2016a, Kali et al. 2013, Roghi et al. 2015, Zia et al. 2012, Chen et al. 2019) and there was one study that including non-STEMI pa-tients (Mohammadzadeh et al. 2018). Most studies reported T2*-values of multiple
region of interest (ROI) in the myocardium (Appendix 3.B). Although none of the tested covariates were significant, the decrease in T2*-values seemed stronger in the
infarct cores compared to the infarct zone as a whole. Significant funnel asymme-try was found for the random effects model suggesting eight missing studies with negative instead of positive results (Pă0.01), while the mixed effects model did not show funnel asymmetry (P=0.60).
Figure 3.6:Standardized mean difference between native myocardial T2*of patients
with a myocardial infarction (MI) and healthy controls with associated random ef-fects weight factors, CI=confidence interval, IV=inverse variance, Std=standardized
Since the heterogeneity could not be corrected with the existing covariates, a sec-ond analysis was performed where the reported T2*-values were divided in infarct
zone or infarct core groups. The infarct zone is defined as the area with T2-values
2SD above the T2-value of remote myocardium caused by edema (Carrick et al. 2016)
and the infarct core is defined as the hypo-intense core is the center in the infarct zone with T2*-valuesă20ms due to presence of hemorrhage (Bulluck et al. 2016a).
Eight studies explicitly reported infarct zone values (Durighel et al. 2016, Bulluck et al. 2016a, Bulluck et al. 2017a, Carrick et al. 2016a, Kali et al. 2013, Moham-madzadeh et al. 2018, Yilmaz et al. 2013, Chen et al. 2019). The weighted mean T2*-value at 1.5T of the infarct zones was 32.3˘5.4ms and at 3T this was 22.4˘2.8ms
(Appendix 3.C). These T2*-values still resulted in significantly decrease compared
to controls (SMD=-1.21, 95%CI [-1.83, -0.59], Pă0.01, I2= 95%), and also contained
a significant heterogeneity. Furthermore, infarct core T2*-values were explicitly
re-ported in five studies (Bulluck et al. 2016a, Bulluck et al. 2017a, Carrick et al. 2016a, Robbers et al. 2018, Zaman et al. 2015). The weighted mean T2*-value at 1.5T of
in-farct cores was 16.1˘4.2ms and at 3T this was 16.1˘7.6ms (Appendix 3.C). These infarct core values showed a bigger SMD (SMD=-4.00, 95%CI [-5.67, -2.32], Pă0.01, I2=98%), while the heterogeneity remained significant. Multiple studies reported
the remote myocardium as control which had a mean T2*-value of 34.0˘4.9ms at
1.5T and 30.5˘1.0ms at 3T (Appendix 3.C).
The weighted mean T2-values at 1.5T were 58.5˘5.8ms in MI patients and 49.3
˘2.6ms in controls (Verhaert et al. 2011, Bulluck et al. 2016a, Bulluck et al. 2017a, Car-rick et al. 2016a, Zia et al. 2012, Nakamori et al. 2019, Tahir et al. 2017, Carberry et al. 2018, Carrick et al. 2016, Haig et al. 2019, Hausenloy et al. 2019, Krumm et al. 2016, McAlindon et al. 2015, Masci et al. 2018, Park et al. 2013, Tessa et al. 2018, White et al. 2015) (Appendix 3.B and Figure 3.4). At 3T, this was 60.3˘9.7ms in MI pa-tients and 44.0˘3.8ms in controls (Zaman et al. 2015, An et al. 2018, Bulluck et al. 2016b, Fischer et al. 2018, Layland et al. 2017, van Heeswijk et al. 2012) (Appendix 3.B and Figure 3.5). Most studies restricted their inclusion to STEMI patients, but some studies included specifically non-STEMI patients (Nakamori et al. 2019, Tessa et al. 2018, Layland et al. 2017) and others included both STEMI and non-STEMI patients (Verhaert et al. 2011, Tahir et al. 2017, Park et al. 2013, Fischer et al. 2018). Besides two studies (Nakamori et al. 2019, Tessa et al. 2018), patients in all studies underwent cardiac MRI post-PCI in the acute phase and a few studies also included follow-up data (Bulluck et al. 2016a, Carberry et al. 2017, Carrick et al. 2016a, Zia et al. 2012, Tahir et al. 2017, An et al. 2018). T2-values of different ROIs in the
myocar-dium were reported (Appendix 3.B) but all studies showed an increase of T2-values
et al. 2016a, Bulluck et al. 2017a). The meta-analysis confirmed this significant in-crease in T2-values in MI patients (SMD=2.17, 95%CI [1.79, 2.54], Pă0.01, I2=96%,
Figure 3.7). The age and percentage of men in the control group, the time between intervention and the MRI scan, the field strength, the type of control (remote myo-cardium versus healthy controls), the type of MRI sequence, the ROI location and the EF in patients were all significant covariates. No other significant covariates re-mained that could account for the remaining high heterogeneity (I2=78%), but this
significant I2indicates that there are probably other covariates which were not used
in this analysis. Publication bias was detected with five possibly missing studies, but no significant asymmetry was found for either the random effects model (P=0.10) or the mixed effects model (P=0.55).
Since the ROI location was one of the covariates, an additional analysis was per-formed where the reported T2-values were divided in infarct zone or infarct core
groups as described above. Infarct zone T2-values were reported in 18 studies
(Verhaert et al. 2011, Durighel et al. 2016, Carrick et al. 2016a, Zaman et al. 2015, Tahir et al. 2017, Carberry et al. 2017, Haig et al. 2019, Hausenloy et al. 2019, Krumm et al. 2016, Masci et al. 2018, Park et al. 2013, Tessa et al. 2018, White et al. 2015, An et al. 2018, Bulluck et al. 2016b, Fischer et al. 2018, Layland et al. 2017, van Heeswijk et al. 2012). The weighted mean T2-value at 1.5T of infarct zones was 63.7˘6.4ms and
at 3T this was 63.5˘10.5ms (Appendix 3.D). Restricting the analysis to infarct zone values, increased the difference between patients and controls (SMD=2.63, 95%CI [2.25, 3.01], Pă0.01, I2=93%). Furthermore the meta-analysis showed older patients,
a short period between the intervention and MRI scan, lower EF in patients and using a field strength of 1.5T, to increase the difference between patients and con-trols. The used MRI sequence was also found as significant covariate, however none of the specified sequences provided a clear advantage by providing bigger SMD. No other significant residual factors remained accounting for the hetero-geneity (I2=80%). Again, publication bias was found with two missing studies, but
no significant asymmetry was found for either the random effects model (P=0.76) nor the mixed effects model (P=0.58). Core T2-values were reported in five studies
(37,38,40,53,57). The weighted mean T2-value at 1.5T of infarct cores was 51.9˘4.6ms
and at 3T no values were reported (Appendix 3.D). Including only the T2-values
of the infarct cores resulted in a smaller difference between patients and controls (SMD=0.83, 95% Cl [0.37, 2.44], Pă0.01, I2=91%). Lastly, the weighted mean T
2
-value at 1.5T of remote myocardium was 49.2˘2.5ms and at 3T this was 45.0˘3.0ms (Appendix 3.D).
Figure 3.7:Standardized mean difference between native myocardial T2of patients
with a myocardial infarction (MI) and healthy controls with associated random ef-fects weight factors, CI=confidence interval, IV=inverse variance, Std=standardized
Heart Transplant
The weighted mean T2-values at 1.5T were 54.6˘5.2ms in heart transplant patients
and 49.2˘2.5ms in healthy controls (Usman et al. 2012, Butler et al. 2015, Dolan et al. 2019, Dolan et al. 2019a, Markl et al. 2013, Miller et al. 2014, Miller et al. 2019, Vermes et al. 2018, Yuan et al. 2018, Bonnemains et al. 2014, Odille et al. 2016) (Appendix 3.B and Figure 3.4). All studies showed an increase of T2-values in patients compared
to healthy controls which was significant for all subgroups including patients with positive rejection biopsy. The meta-analysis confirmed the increase in T2-values in
the myocardium of heart transplant patients (SMD=1.05, 95%CI [0.69, 1.41], Pă0.01, I2=65%, Figure 3.8). An exploratory meta-regression analysis indicated that the
re-jection status, the EF and patient age caused the heterogeneity without remaining significant residual factors (I2=1%). Transplant rejection, lower EF and older
pa-tients resulted in an increased SMD between papa-tients and controls.
Based on the fact that the cardiac transplant rejection was a significant covari-ate, the population was divided between positive and negative rejection biopsies. For patients with a positive biopsy the weighted mean T2-values were 56.4˘3.3ms
(Usman et al. 2012, Butler et al. 2015, Dolan et al. 2019a, Miller et al. 2014, Miller et al. 2019, Vermes et al. 2018) and in patients with a negative biopsy the weighted mean T2-values were 52.5˘3.9ms (Usman et al. 2012, Butler et al. 2015, Dolan et al.
2019a, Markl et al. 2013, Miller et al. 2014, Miller et al. 2019, Vermes et al. 2018, Yuan et al. 2018) (Appendix 3.D). None of these heart transplantation studies reported T2-values acquired at 3T or the application of T2*-values.
Figure 3.8: Standardized mean difference between native myocardial T2 of
pa-tients with heart transplantation (Transplant) and healthy controls with asso-ciated random effects weight factors, CI=confidence interval, IV=inverse variance, Std=standardized
Iron Overload
The weighted mean T2*-value at 1.5T was 27.2˘13.7ms in iron overload patients
over-Figure 3.9: Standardized mean difference between native myocardial T2* of iron
overload (IO) patients and healthy controls with associated random effects weight factors, CI=confidence interval, IV=inverse variance, Std=standardized
load studies except for one (Gu et al. 2013) (Appendix 3.B and Figure 3.2). At 3T, these values were 21.8˘7.8ms in iron overload patients and 22.4˘3.8ms in controls (Kritsaneepaiboon et al. 2018, Alam et al. 2016, Meloni et al. 2012, Gu et al. 2013) (Appendix 3.B and Figure 3.3). The meta-analysis showed an overall significant decrease in T2*-values in iron overload patients (SMD=-2.39, 95%CI [-3.28, -1.49],
Pă0.01, I2=98%, Figure 3.9). All studies showed a significant decease
indepen-dent of each other, except for two studies that showed a non-significant decrease (Kritsaneepaiboon et al. 2018, Alam et al. 2016) and one study that showed an non-significant increase in patients compared to controls (Desai et al. 2014). The type of control was found as a covariate which showed that using non-cardiac involved iron overload subjects as controls caused an increase in SMD for cardiac-involved patients than using healthy controls. Moreover, the type of patients was found as covariate; using a population with proven cardiac involvement caused bigger dif-ferences with controls than using a mix of non-cardiac and cardiac involved iron overload patients. Furthermore, the number of echoes used in the T2* sequence
was determined as a covariate. However, these covariates accounted only partly for the heterogeneity in the mixed effects model (I2=80%), while other tested
re-gressors (age of patient and control population, percentage of men in patient and control population, MR vendor, field strength and the serum ferritin concentration in patients) had no significant influence. Significant funnel asymmetry (Pă0.01) was only found for the random effects model suggesting five missing studies with pop-ulations showing increased T2*-values instead of decreased compared to healthy
subjects.
Since the type of iron overload patient was one of the covariates, an additional analysis was performed on reported T2*-values from cardiac involved iron
et al. 2017, Garceau et al. 2011, Liguori et al. 2015, Mehrzad et al. 2016, De Assis et al. 2012, De Sanctis et al. 2016, Meloni et al. 2014, Positano et al. 2015, Bayar et al. 2015, Karakus et al. 2017, Pennell et al. 2015, Porter et al. 2013). The weighted mean T2*-value at 1.5T in cardiac involved iron overload patients was 11.8˘3.7ms
and at 3T no T2*-values were reported (Appendix 3.C). This analysis also showed a
significant decrease in T2*for cardiac involved iron overload patients compared to
controls (SMD=-3.59, 95%CI [-4.69, -2.48], Pă0.01, I2=97%) and this difference was
also bigger than controls compared to the iron overload overall as described above.
Figure 3.10: Standardized mean difference between native myocardial T2 of iron
overload (IO) patients and healthy controls with associated random effects weight factors, CI=confidence interval, IV=inverse variance, Std=standardized
The weighted mean T2-values at 1.5T were 56.0˘13.6ms in iron overload patients
and 58.3˘3.2ms in healthy controls (Kritsaneepaiboon et al. 2018, Krittayaphong et al. 2017, Feng et al. 2013) (Appendix 3.B and Figure 3.4). At 3T, this was 53.2˘6.2ms in iron overload patients and 52.0˘5.5ms in controls (Kritsaneepaiboon et al. 2018, Camargo et al. 2016) (Appendix 3.B and Figure 3.5). Kritsaneepaiboon et al. re-ported no significant changes in T2-values for iron overload patients at both 1.5T
and 3T, while Camargo et al. reported decreased T2-values in iron overload
pa-tients at 1.5T. The random effects models of all studies combined resulted in a no significant change in T2-values for iron overload patients compared to the controls
(SMD=-0.54, 95%CI [-1.56, 0.48], P=0.30, I2= 86%, Figure 3.10).
Figure 3.11: Standardized mean difference between native myocardial T2 of
sar-coidosis patients and healthy controls with associated random effects weight fac-tors, CI=confidence interval, IV=inverse variance,Std=standardized
Sarcoidosis
The weighted mean T2-values at 1.5T were 52.3˘3.8ms in sarcoidosis patients and
49.0˘1.6ms in controls (Greulich et al. 2016) (Appendix 3.B and Figure 3.4). At 3T, these were 54.0˘12.2ms in sarcoidosis patients and 45.0˘10.8ms in controls (Mayr et al. 2017a) (Appendix 3.B and Figure 3.5). This suggested an increase of T2-values
in sarcoidosis patients (SMD=0.87, 95%CI [0.55, 1.20], Pă0.01, I2= 0%, Figure 3.11).
However, insufficient studies were available for further analysis and there was no data that described T2*-values.
Lupus
The weighted mean T2-values at 1.5T were 55.7˘4.9ms in Lupus patients and 50.6˘
3.3ms in healthy controls (Mayr et al. 2017a, Zhang et al. 2015) (Appendix 3.B and Figure 3.4). At 3T, this was 57.3˘8.6ms in Lupus patients and 44.4˘4.0ms in con-trols (Hinojar et al. 2016, Winau et al. 2018) (Appendix 3.B and Figure 3.5). This suggested an increase of T2-values in Lupus patients (SMD=1.39, 95%CI [0.34, 2.44],
Pă0.01, I2=93%, Figure 3.12). However, insufficient studies were available for
fur-ther analysis and fur-there was no data that described T2*-values.
Figure 3.12: Standardized mean difference between native myocardial T2 of
lu-pus patients and healthy controls with associated random effects weight factors, CI=confidence interval, IV=inverse variance, Std=standardized
Amyloidosis
The weighted mean T2-values at 1.5T were 55.3˘4.2ms in amyloidosis patients and
50.2˘2.7ms in controls (Kotecha et al. 2018, Ridouani et al. 2018) (Appendix 3.B and Figure 3.4). All included studies reported an increase of T2-values in amyloidosis
in-sufficient studies were available for further analysis, both included studies reported higher T2-values in amyloid light-chain (AL) amyloidosis than in transthyretin
(ATTR) amyloidosis. Furthermore, there were no studies performed with T2-values
on 3T and there was no data that described T2*-values.
Figure 3.13: Standardized mean difference between native myocardial T2 of
amy-loidosis (AM) patients and healthy controls with associated random effects weight factors,CI=confidence interval, IV=inverse variance, Std=standardized
Figure 3.14: Standardized mean difference between native myocardial T2of Fabry
disease patients and healthy controls with associated random effects weight factors, CI=confidence interval, IV=inverse variance, Std=standardized
Fabry Disease
The weighted mean T2-values at 1.5T were 57.7˘3.0ms in Fabry disease patients
(Messalli et al. 2012, Knott et al. 2019) (Appendix 3.B and Figure 3.4). One study reported T2-values in controls of 47.5˘2.4ms (Knott et al. 2019), suggesting a trend
to higher T2-values in Fabry disease patients (SMD=0.52, 95%CI [-0.23, 1.28], P=0.17,
I2=71%, Figure 3.14). This increase was caused by the reported T
2-values in Fabry
disease patients with left-ventricle hypertrophy (LVH) (50.4˘3.8ms), while patients without LVH showed similar T2-values (47.8˘1.7ms) to controls. However,
insuf-ficient studies were available for further analysis, there were no studies performed with T2-values on 3T and there was no data that described T2*-values.
Hypertrophic Cardiomyopathy
The weighted mean T2*-values at 1.5T from one study were 26.4˘4.4ms in HCM
patients and 31.3˘4.3ms in controls (Gastl et al. 2019) (Appendix 3.B and Figure 3.2). At 3T, this was 22.3˘4.1ms in HCM patients and 21.0˘6.4ms in controls (Kanzaki et al. 2016) (Appendix 3.B and Figure 3.3). The study performed at 1.5T reported values in subgroups based on the presence of fibrosis (with or without LGE) and in both subgroups the T2*-value seemed decreased compared to controls, but this was
only significant in the group with LGE presence (Gastl et al. 2019). However, in the study performed at 3T there was no significant difference between the HCM with or without LGE presence (Kanzaki et al. 2016) and T2*-values also did not differ from
controls. As result, the analysis showed a no significant difference between HCM patients and controls (SMD=-0.61, 95%CI [-1.58, 0.36], P=0.22, I2=87%, Figure 3.15).
However, insufficient studies were available for further analysis.
Figure 3.15: Standardized mean difference between native myocardial T2* of
hypertrophic cardiomyopathy (HCM) patients and healthy controls with asso-ciated random effects weight factors, CI=confidence interval, IV=inverse variance, Std=standardized
The weighted mean T2-value at 1.5T was 56.3˘4.0ms in HCM patients (Amano
et al. 2017, Park et al. 2018) (Appendix 3.B and Figure 3.4). One study reported T2
-values in controls of 48.1˘3.2ms suggesting significantly higher T2-values in HCM
patients (Amano et al. 2017) (SMD=1.95, 95%CI [0.93, 2.97], I2=N/A, Pă0.01,
Fig-ure 3.16). However, in that same study the T2-values were measured in the patient
myocardium with visually high T2, which was present in 38% of the patients. For
the patients without LGE in that study the myocardial T2-value of 48.8˘2.4ms was
not significantly difference from controls. Furthermore, there were no studies per-formed with T2-values acquired at 3T and insufficient studies were available for
Figure 3.16: Standardized mean difference between native myocardial T2 of
hypertrophic cardiomyopathy (HCM) patients and healthy controls with asso-ciated random effects weight factors, CI=confidence interval, IV=inverse variance, Std=standardized
Dilated Cardiomyopathy
The weighted mean T2*-value at 3T was 22.7˘3.6ms in DCM patients (Kanzaki
et al. 2016, Nagao et al. 2015), and only one of those studies reported T2*-values in
controls of 21.0˘6.4ms (Kanzaki et al. 2016) (Appendix 3.B and Figure 3.3). Thefore, the random effects model was only based on that study and since they re-ported T2*-values of 18.7˘3.1ms in DCM patients there was no significant change in
T2*(SMD=-0.54, 95%CI [-1.09, 0.01], I2=N/A, P=0.06, Figure 3.17). In both studies,
patients had chronic established DCM and patients without myocarditis or other cardiomyopathies (Kanzaki et al. 2016, Nagao et al. 2015). Furthermore, there were no studies performed with T2*-values acquired at 1.5T and there were also
insuffi-cient studies available for further analysis.
Figure 3.17:Standardized mean difference between native myocardial T2*of dilated
cardiomyopathy (DCM) patients and healthy controls with associated random ef-fects weight factors, CI=confidence interval, IV=inverse variance, Std=standardized
The weighted mean T2-values at 1.5T were 62.9˘5.7ms in DCM patients and
55.4˘3.5ms in healthy controls (Ito et al. 2015, Kono et al. 2014, Nishii et al. 2014, Spieker et al. 2018, Cui et al. 2018, Mordi et al. 2016) (Appendix 3.B and Figure 3.4). At 3T, these values were 47.0˘5.0ms in DCM patients and 45.0˘3.0ms in controls (Child et al. 2018) (Appendix 3.B and Figure 3.5). All studies reported significantly
higher T2-values in DCM patients compared to controls, except for the single study
performed at 3T (Child et al. 2018). The similar T2-values of patients and controls
in this study might be related to their ROI placement rather than the field strength, since they explicitly excluding LGE regions from the ROI. The overall meta-analysis confirmed the significantly increased T2-values in DCM patients (SMD=1.90, 95%CI
[1.07, 2.72], Pă0.01, I2=89%, Figure 3.18) and an exploratory meta-regression
anal-ysis indicated the MRI vendor and the age difference between DCM patients and controls as possible covariates. This showed in the use of a Philips scanner and a bigger age difference between control and patient groups resulted in an increased SMD between DCM patients and controls.
Figure 3.18:Standardized mean difference between native myocardial T2of dilated
cardiomyopathy (DCM) patients and healthy controls with associated random ef-fects weight factors, CI=confidence interval, IV=inverse variance, Std=standardized
Myocarditis
The weighted mean T2-values at 1.5T were 61.9˘11.5ms in myocarditis patients and
54.4˘5.9ms in healthy controls (Thavendiranathan et al. 2012, Luetkens et al. 2019, Baessler et al. 2017, Baessler et al. 2018, Baessler et al. 2019, Bohnen et al. 2015, Bohnen et al. 2017, Dabir et al. 2019, Gatti et al. 2019, Luetkens et al. 2017, Lurz et al. 2016, Radunski et al. 2014, Radunski et al. 2017, Spieker et al. 2017, Huber et al. 2018, Mayr et al. 2017, von Knobelsdorff-Brenkenhoff et al. 2017) (Appendix 3.B and Figure 3.4). At 3T, this was 63.8˘8.0ms in myocarditis patients and 53.3˘3.3ms in controls (Gang et al. 2019, Stirrat et al. 2018) (Appendix 3.B and Figure 3.5). The meta-analysis confirmed the significantly increased T2-values in myocarditis
pa-tients (SMD=1.33, 95%CI [1.00, 1.67], Pă0.01, I2=84%, Figure 3.19). Multiple
sig-nificant covariates were identified including; the difference in EF between patients and controls, the difference in percentage men between patients and controls, the
time between symptoms and the MRI scan, the number of echoes used in the MR acquisition sequence, the MR vendor and the slice thickness. These covariates to-gether corrected for the total heterogeneity (I2 = 0%) and resulted in an increased
SMD between myocarditis patients and controls when the same percentages of men was used in both groups, a smaller EF was seen in patients, 6 echoes were acquired for mapping, a Siemens MRI was used, a bigger slice thickness was used, and when the patients were scanned in the acute phase of myocarditis. No significant asym-metry was found for either the random effects model (P=0.12) or the mixed effects model (P=0.10).
Figure 3.19: Standardized mean difference between native myocardial T2of
myo-carditis patients and healthy controls with associated random effects weight factors, CI=confidence interval, IV=inverse variance, Std=standardized
Since the time between symptom onset and the MRI scan was found as signifi-cant covariate, the population was divided between T2-values from patients in the
acute phase and non-acute phase (Kindermann et al. 2012). The weighted mean T2-value of myocarditis patients in the acute phase was 63.5˘15ms at 1.5T and this
was 63.8˘8ms at 3T (Thavendiranathan et al. 2012, Luetkens et al. 2019, Baessler et al. 2018, Baessler et al. 2019, Bohnen et al. 2015, Bohnen et al. 2017, Dabir et al. 2019, Gatti et al. 2019, Luetkens et al. 2017, Lurz et al. 2016, Radunski et al. 2014, Spieker et al. 2017, Huber et al. 2018, Mayr et al. 2017, von Knobelsdorff-Brenkenhoff
et al. 2017, Gang et al. 2019, Stirrat et al. 2018) (Appendix 3.D). The weighted mean T2-value of myocarditis patients in the non-acute phase was 58.3˘4.3ms at 1.5T
(Baessler et al. 2019, Bohnen et al. 2017, von Knobelsdorff-Brenkenhoff et al. 2017, Lurz et al. 2016) and no T2-values were reported at 3T (Appendix 3.D). Furthermore,
there were no studies that described T2*-values for myocarditis.
Hypertension
One study reported T2*-values at 1.5T of 26.3˘3.7ms in HT patients and 30.8˘2.7ms
in healthy controls (Chen et al. 2018) (Appendix 3.B and Figure 3.2). This suggested a decrease of T2*-values in HT patients, but this was not significant (SMD=-1.46,
95%CI [-3.21, 0.29], P=0.10, I2=92%, Figure 3.20). In addition, this study classified
the included HT population in the presence of LVH or not, and showed in both subgroups a significant decrease of T2*-values. However, in HT patients with LVH
this decrease was bigger. Furthermore, insufficient studies were available for further analysis and there were no studies that described T2*-values acquired at 3T or T2
-results. Also, no published data was found on T2or T2*for the cardiovascular risk
populations obesity and diabetes.
Figure 3.20:Standardized mean difference between native myocardial T2of
hyper-tension (HT) patients and healthy controls with associated random effects weight factors, CI=confidence interval, IV=inverse variance, Std=standardized
3.4
Discussion
Quantitative analysis of factors that modulate myocardial T2and T2*, such as edema,
lipids and paramagnetic iron-containing species can potentially provide additional diagnostic information to distinguish between myocardial diseases and healthy myo-cardium. This meta-analysis confirmed that T2-mapping can help differentiate
and rejection in heart transplant patients. Although T2-mapping has been expected
to be sensitive to iron overload as well (Messroghli et al. 2017), no significant de-crease was found between iron overload related diseases and healthy myocardium (P=0.30). On sarcoidosis, lupus, amyloidosis, sarcoidosis, Fabry disease and HCM insufficient studies were reported for further analysis, but the available data sug-gested T2to be increased in these diseases, with an exception for Fabry disease
pa-tients without LVH. Furthermore, this meta-analysis confirmed that T2*-mapping
can differentiate between healthy myocardium and myocardium affected by MI and iron overload, since T2*-values were decreased in both of these populations. For
HCM, DCM and HT patients, the limited available T2*-mapping studies also gave
some indication of a decrease compared to controls, but this was overall not sig-nificant. Eventually, the similar directions of the changes in T2- and T2*-values of
the cardiac diseases makes it difficult to differentiation between the in diseases, as opposed to differentiation from the healthy.
Reported T2- and T2*-values in healthy subjects showed large differences between
studies, which could partly be due to the lack of acquisition standardization. In the standardized cardiac MRI guidelines and protocols published in 2013 (Kramer et al. 2013), T2*-mapping was only described as a clinical applicable technique to
assess cardiac iron deposition and T2-mapping was defined as a research-domain
technique (Kramer et al. 2013, Schulz-Menger et al. 2013). T2-mapping sequences
were stated as optional since they were not yet standardized (Kramer et al. 2013), which led to different acquisition approaches and therefore potentially acquisition related variation in reported T2-values. In 2017, clinical recommendations were
re-leased regarding parametric imaging of both T2- and T2*-mapping and defined
stan-dardized data acquisition and analysis (Messroghli et al. 2017). Here was stated that local healthy T2- and T2*-values should be determined in order to clinically use these
quantitative techniques, which is now confirmed by this metaanalysis (Figure 3.2 -3.5). The use of normal scan results of clinically referred patients could be used to determine these healthy reference values, but this is not recommended due to po-tential referral bias. Also age- and gender-matching of the control group was recom-mended since both are known to influence T2- and T2*-values (Roy et al. 2017,
Mess-roghli et al. 2017).
Furthermore, the clinical recommendations also stated specific imaging protocols, technical requirements of sequences and image planning for T2- and T2*-mapping,
which should reduce variability in image acquisition from then onward (Messroghli et al. 2017). Since this meta-analysis includes multiple studies that were published prior to these guideline the heterogeneity was shown to be significantly influences
by the sequence based covariates, which has previously already been concluded from a direct comparison between sequences (Schwitter et al. 2018). In addition to explaining the study heterogeneity, this analysis also showed the variation in quantitative mapping in healthy controls between MRI vendors. In healthy vol-unteers scanned on a 1.5T from Philips the mean T2-value was 54.9˘3.3ms (n=13)
and from Siemens this was 50.0˘2.5ms (n=22). Furthermore, the mean T2*-values
on 1.5T from Philips was 34.1˘6.5 (n=5), from Siemens was 30.8˘4.5ms (n=3), and from GE was 55.0˘13.0ms (n=1) scanners. On 3T these T2-values from Philips were
44.7˘5.8ms (n=6) and from Siemens were 48.0˘3.0ms (n=5). Furthermore the mean T2*-values from Philips were 23.9˘4.7ms (n=2), for Siemens were 21.0˘4.8ms (n=1),
and for GE were 21.0˘6.4ms (n=1). These differences in vendor and field strenght should be kept in mind when T2- and T2*-values are used for quantitative clinical
diagnosis.
In addition to the clinical guidelines on T2and T2*acquisitions (Messroghli et al.
2017), following the recommendations in image analysis could reduce the non phys-iological variation of T2- and T2*-values. Since the clinical recommendations on
acquisition and ROI placement are described specifically per disease (Messroghli et al. 2017) this meta-analysis shows different approaches in analysis as well. Anal-ysis of T2 in diffuse diseases, such as HCM and DCM, were mostly performed
based on one or three short axis (SAX) slices and using global assessment (Park et al. 2018, Ito et al. 2015, Kono et al. 2014, Nishii et al. 2014, Spieker et al. 2018, Cui et al. 2018, Mordi et al. 2016), as recommended (Messroghli et al. 2017). In patchy diseases, such as amyloidosis and Fabry disease, the recommendation state that the T2 analysis should not only be based on basal and mid-ventricular SAX slices
but included an additional single 4 chamber (4CH) or 3 chamber (3CH) acquisi-tion (Messroghli et al. 2017). Only one study actually followed these recommenda-tions (Knott et al. 2019), while for the other cardiac patchy disease studies one or more of the recommended slices were not included (Kotecha et al. 2018, Ridouani et al. 2018, Messalli et al. 2012). Furthermore, in focal diseases such as MI and myo-carditis the ROI differs between patients because the location of the abnormality is different and therefore the guideline recommend multiple SAX acquisition to cover the whole myocardium (Messroghli et al. 2017). This resulted in that most included studies in this meta-analysis acquired multiple SAX slices (Zaman et al. 2015, Car-berry et al. 2017, Carrick et al. 2016, Hausenloy et al. 2019, Park et al. 2013, White et al. 2015, Bulluck et al. 2016b), but some studies placed only one (Masci et al. 2018) or three (Zia et al. 2012) SAX slices at the level of the infarcted area, which is more prone to missing the infarct core. In the studies with myocarditis patients mapping acquisitions were generally also performed over multiple SAX covering the whole
myocardium (Luetkens et al. 2019, Baessler et al. 2017, Baessler et al. 2018, Baessler et al. 2019, Bohnen et al. 2015, Dabir et al. 2019, Gatti et al. 2019, Luetkens et al. 2017, Lurz et al. 2016, Radunski et al. 2014, Spieker et al. 2017, Huber et al. 2018, von Knobelsdorff-Brenkenhoff et al. 2017), but some studies only acquired the T2-values
from a LGE hyperintense based ROI (Thavendiranathan et al. 2012, Bohnen et al. 2017, Radunski et al. 2017, Mayr et al. 2017, Stirrat et al. 2018). Also, studies includ-ing MI often distinclud-inguish between the infarct region and the infarct core, and use remote myocardium as the healthy control tissue. In these studies the ROI place-ment was generally based on LGE hyperintense regions (Verhaert et al. 2011, Bul-luck et al. 2017a, Zia et al. 2012, Nakamori et al. 2019, Hausenloy et al. 2019, Krumm et al. 2016, Masci et al. 2018, Park et al. 2013, Tessa et al. 2018, White et al. 2015, Bul-luck et al. 2016b, Layland et al. 2017, van Heeswijk et al. 2012), 2SD change of T2
signal intensity (Bulluck et al. 2016a, Carrick et al. 2016a, Carberry et al. 2017, Haig et al. 2019, McAlindon et al. 2015, Masci et al. 2018) or 2SD change of T2*-values
(Bulluck et al. 2017a, Carrick et al. 2016a, Haig et al. 2019). However, the clinical recommendation advises ROI placement in visually abnormal myocardium, inde-pendent of LGE hyperintensity, and this meta-analysis also shows that ROI place-ment significantly influences the T2 and T2*outcome. In the separate analysis the
infarct zone showed a bigger T2difference with controls than the infarct core, while
the infarct core showed a bigger T2* difference with controls then the infarct zone.
Lastly, for studies including iron overload patients most T2* measurements were
performed only in the intraventricular septum because the lateral wall often con-tains dephasing artefacts. Nevertheless, some studies reported an average of the mid-ventricular SAX slice (Acar et al. 2012, Ozbek et al. 2011, Sakuta et al. 2010, Wi-jarnpreecha et al. 2015) or the entire myocardium (Hanneman et al. 2013, Marsella et al. 2011, Meloni et al. 2012, Meloni et al. 2014, Pepe et al. 2018, Pistoia et al. 2019, Pizzino et al. 2018, Positano et al. 2015), which especially on 3T (Meloni et al. 2012) could lead to some unrealistic T2*-values due to aforementioned artefacts.
Myocardial Infarction
In the meta-analysis for MI other covariates aside from the ROI placement had a significant effect on T2- and T2*-mapping outcomes. These covariates included; the
use of remote myocardium as control values instead of truly healthy controls, the timing of the MRI acquisition after reperfusion, and the sequence that was used. First the covariate that includes the use of the remote myocardium as control, which is shown to be physiologically different from healthy tissue and therefore is not an appropriate control tissue (Biesbroek et al. 2017, Manrique et al. 2009). Followed
by the second covariate for timing of the MRI examination after PCI, for which his-tologically is verified in pigs that edema and haemorrhage formation peaks in the acute phase at two hours and seven days post-PCI (Fernandez-Jimenez et al. 2015a). These peaks were also detected in the acquired T2-values in humans at the same
day and at ten days post-PCI, compared to low T2-values at three days post-PCI
(Carrick et al. 2016a). However, these results were contradicted by another study that reported higher T2-values at three days post-PCI compared to the same day or
at day seven post-PCI (An et al. 2018). This makes the best timing of a MRI exam-ination that includes T2-mapping still undefined. Than the third covariate showed
that the use of a spin-echo (SE) based sequence provides bigger differences between MI patients and controls than the gradient-echo spin-echo (GESE) or T2
-prepared-balanced steady-state free procession (T2prep-SSFP) sequences, while the latter two are currently recommended in the general guideline (Messroghli et al. 2017). Lastly, due to the remaining high heterogeneity of the MI meta-analysis other covariates are expected to influence the T2- and T2*-mapping outcomes in addition to the ones
identified here.
Heart Transplantation
In the meta-analysis for heart transplantation patients the main distinct covariate was the rejection status of the transplanted heart. Acute cellular rejection is char-acterized by infiltration of inflammatory cells accompanied with edema resulting in an increase in T2-values (Messroghli et al. 2017, Butler et al. 2009), which was
also reported in most included studies (Usman et al. 2012, Dolan et al. 2019a, Markl et al. 2013, Miller et al. 2014, Vermes et al. 2018, Yuan et al. 2018). Nevertheless, T2-values in patients with negative biopsies were also significantly higher than
con-trols (Butler et al. 2015, Dolan et al. 2019a, Vermes et al. 2018). While it is now known that higher biopsy grades are related to higher T2-values (Usman et al. 2012, Markl
et al. 2013, Miller et al. 2014, Bonnemains et al. 2014), further research in T2-mapping
is needed to provide clinical applicability for early detection of rejection.
Iron Overload
In this iron overload meta-analysis all transfusion-dependent diseases leading to iron overload were evaluated as one group including thalassemia, sickle cell dis-ease and anaemias (de Montalembert et al. 2017). The overall average T2*-value
the established iron overload cut-off of T2*ă20ms (Schulz-Menger et al. 2013). This could be due to the fact that most studies reported T2*-values without
distinguish-ing between cardiac or non-cardiac involved iron overload. However, some studies did provided T2*-values of cardiac involved iron overload patients using T2*ă20ms
as a clinical cut-off. Consequently, the mean T2*-value of these cardiac involved
pa-tients was only 11.8˘3.7ms, which was significantly lower than the controls. How-ever, the type of controls should technically only include healthy volunteers, but in some cases also non-cardiac involved iron overload patients were used as controls by using a cut-off of T2*ě20ms. Since the T2*-values from real healthy volunteers of
32.4˘5.6ms (Desai et al. 2014, Kritsaneepaiboon et al. 2018, Seldrum et al. 2011, Alam et al. 2016, Camargo et al. 2016, Hanneman et al. 2013, Sado et al. 2015, Russo et al. 2011) was lower than the 35.7˘6.4ms from non-cardiac iron overload patients (Delaporta et al. 2013, Di Odoardo et al. 2017, Garceau et al. 2011, Liguori et al. 2015, Mehrzad et al. 2016, De Sanctis et al. 2016, Meloni et al. 2012, Positano et al. 2015) the accuracy of the T2*ă20ms cut off to establish cardiac involvement could be
chal-lenged.
In addition, current recommendation advises to perform T2*-mapping on 1.5T,
since higher field strengths showed more susceptibility artefacts (Messroghli et al. 2017). Nonetheless, two studies were performed at 3T as well as at 1.5T, in which the ROI was placement at the mid-ventricular septum to avoid inclusion of suscep-tibility artefacts (Kritsaneepaiboon et al. 2018, Alam et al. 2016). As expected, these studies showed an increased SMD between healthy controls and iron overload pa-tients at 3T compared to 1.5T, since the transversal relaxivity of iron increases with field strength (Chavhan et al. 2009). These last findings show that using 3T would be a trade-off between an increased risk on artefacts and a higher iron sensitivity.
Furthermore, T2-mapping was also expected to be sensitive for iron overload
(Messroghli et al. 2017), but this was not unequivocally confirmed by this meta-analysis. Where one study performed on 1.5T and 3T showed no statistically sig-nificant T2-changes in iron overload patients (Kritsaneepaiboon et al. 2018), others
did show a clear changes in T2-values (Krittayaphong et al. 2017, Camargo et al.
2016, Feng et al. 2013). However, in the non-significant study only 6% of the iron overload patients had cardiac involvement, which might explain the lack of change in T2. The other studies showed a high correlation between T2 and T2* changes
(Kritsaneepaiboon et al. 2018, Feng et al. 2013) and a significant decrease of T2-values
in patients with cardiac involved iron overload compared to healthy controls, sug-gesting that T2could indeed be sensitive to iron overload. However, more research
Dilated Cardiomyopathy
The increase of T2-values in DCM patients found in this meta-analysis confirmed
the immune-histological evidence of chronic myocardial inflammation for this dis-ease (Hinojar et al. 2015a). However, studies reporting T2-values of DCM subgroups
seemed contradicting, since one study showed significantly higher T2-values in
se-vere DCM compared to mild DCM (Nishii et al. 2014), while another suggested lower T2-values in severe DCM compared to mild DCM(Spieker et al. 2018).
Never-theless, overall an increase in T2was confirmed by this meta-analysis.
Myocarditis
The meta-analysis for myocarditis patients confirmed the expected increase of T2
-values in the acute phase. While all studies reported a significant increase in T2
-values, one study showed a non-significant increase of T2-values in the acute phase
compared to healthy controls, with 65.3˘45.4ms and 53.7˘31.0ms, respectively, which was mainly due to the broad SD of both groups (Mayr et al. 2017). Aside from the increased T2-values in the acute phase, a follow up study showed that three
and twelve months after the onset of the symptoms the T2-values were back to
nor-mal again (Bohnen et al. 2017). Another follow up study confirmed these nornor-mal T2-values at 189 days after the onset of the symptoms, but also showed that after
40 days the T2-values were still significantly increased compared to healthy
con-trols, with 52.4˘1.0ms and 50.4˘2.3ms, respectively (von Knobelsdorff-Brenkenhoff et al. 2017). These follow-up studies suggest that T2-mapping in myocarditis is most
valuable in the acute phase in addition to the Lake Louise criteria that include his-tology and cardiac MRI with T1- and T2-weighted imaging.
Hypertrophic Cardiomyopathy and Hypertension
The single study that reported T2-values from HCM patients and controls showed a
significant increase in patients (Amano et al. 2017), but two other studies repored contradicting results. Where one study at 1.5T reported significantly lower T2*
-values in HCM patients than in controls with 26.2˘4.6ms and 31.3˘4.3ms, respec-tively (Gastl et al. 2019), the other study at 3T reported no significant difference with 22.3˘4.1ms and 21.0˘6.4ms, respectively (Kanzaki et al. 2016). Since early treatment is key for HCM patients it is important to distinguish HCM from HT. However, only
one study reported on HT patients and showed decreased T2*-values at 3T for both
HT patients with and without LVH, with 23.8˘3.1ms and 28.6˘4.2ms, respectively (Chen et al. 2019). Based on these limited available studies no conclusion can be drawn on the clinical relevance of T2- and T2*-mapping. However, T1-mapping has
already shown be promising in distinguishing HT with and without LVH and HCM (van den Boomen et al. 2018, Hinojar et al. 2015a).
Obesity and type 2 Diabetes Mellitus
As the incidence of cardiomyopathies is related to obesity and DM (Ponikowski et al. 2016) it is important to determine whether these high cardiovascular risk fac-tors cause myocardial tissue adaptation and if these are distinguishable from either healthy or cardiomyopathies with quantitative techniques. Unfortunately, no T2- or
T2*-mapping data of these risk populations is available yet, which causes us to rely
on the reference values of cardiac diseases without considering these risk factors. In conclusion, this meta-analysis showed that T2- and T2*-values of both patients
and healthy controls demonstrate variation between studies related to differences in population demographics, MRI vendor, acquisition methods and analysis approach. Therefore, local reference values of matched healthy controls, imaging, and analysis approaches are required to interpret clinical results. When those local reference val-ues are established, T2- and T2*-mapping can help distinguish affected myocardium
in cardiomyopathies from healthy myocardium. However, it should be taken into account that the similar changes in T2 and T2*between cardiac diseases limits the
differentiation.
Conclusion
This chapter showed the clinical utility of T2- and T2*-mapping to distinguish
affected myocardium in patients with cardiomyopathies and rejected heart trans-plantation from healthy myocardium. However, variation of mapping values between studies caused by differences in acquisition methods and subject de-mographics complicate comparison between studies and therefore local healthy reference values are recommended to correctly interpret these quantitative val-ues clinically. Furthermore, since changes in T2- and T2*-mapping values of
most cardiomyopathies are similar, differentiation between the cardiac diseases remains difficult.
3.A
Search term Native T
2and T
2*Meta-Analaysis
The search term used in PubMed /MEDLIN for Literature Search:
(((T2 [tiab] OR “T 2”[tiab] OR t2star[tiab]) AND (mapping [tiab]OR maps [tiab] OR relaxation [tiab] OR value [tiab] OR values[tiab] OR time[tiab] OR times[tiab])) AND (“Magnetic Resonance Imaging”[Mesh] OR “Magnetic Resonance” [tiab] OR MRI[tiab] OR “cardiac MR” [tiab] OR CMR[tiab] OR “MR imaging”[tiab] OR MR[tiab])) AND ((“heart diseases” [Mesh] OR “heart disease” [tiab] OR myocard* [tiab] OR “heart failure”[Mesh] OR “heart failure”[tiab] OR “cardiac injury”[tiab]) OR ((“Heart” [Mesh] OR Heart [tiab] OR cardiac[tiab]) AND (“Sarcoidosis” [Mesh] OR Sarcoidosis [tiab]OR Sarcoidoses [tiab] OR “Besnier Boeck Schaumann” [tiab] OR “Amyloidosis” [Mesh] OR Amyloidosis [tiab] OR amyloidoses [tiab] OR “Churg-Strauss Syndrome”[Mesh] OR ”Churg-“Churg-Strauss*” [tiab] OR “Fabry Disease”[Mesh] OR “Anderson Fabry”[tiab] OR “Fabry Disease”[tiab] OR “Anderson-Fabry”[tiab] OR “Cardiomyopathies”[Mesh] OR cardiomyopathy [tiab] OR cardiomyopathies [tiab] OR “Myocarditis” [Mesh] OR myocarditis[tiab] OR “Iron Overload” [Mesh] OR “Iron overload”[tiab] OR ”Hemochromatosis”[Mesh] OR hemochromatosis[tiab] OR haemochromatosis[tiab] OR Hemosiderosis [tiab] OR Haemosiderosis [tiab] OR “Coronary artery disease”[ Mesh] OR “Coronary artery disease”[tiab]OR “Coronary Atherosclerosis”[tiab] OR “Transplantation”[Mesh] OR Transplantation [tiab])) OR (“Edema, Cardiac”[Mesh] OR “cardiac edema”[tiab] OR “cardiac oedema”[tiab]) OR ((Myocardial[tiab] OR Cardiac[tiab]) AND (inflammation [tiab] OR infiltration [tiab] OR edema[tiab] OR necrosis[tiab] OR “hemorrhage”[Mesh] OR hemorrhage[tiab] OR haemorrhage [tiab] OR “Connective Tissue Diseases”[Mesh] OR “Connective Tissue”[tiab])) OR ((Myocardial[tiab] OR Cardiac[tiab] OR “Heart” [Mesh] OR Heart [tiab]) AND (”Obesity”[Mesh] OR ”Overweight”[Mesh] OR “body weight”[tiab] OR “body mass index”[tiab] OR obese [tiab] OR obesity[tiab] OR overweight [tiab] OR ”Diabetes Mellitus”[Mesh] OR diabetes [tiab] OR diabetic[tiab] OR ”Hypertension”[Mesh] OR “blood pressure”[Mesh] OR hypertension [tiab] OR hypertensive[tiab] OR ”high blood pressure”[tiab]))) AND (”2011/01/01”[PDat] : ”2019/08/01”[PDat]) NOT ((“Animals”[Mesh] NOT “Humans”[Mesh]) OR pig[tiab] OR swine[tiab] OR mice[tiab] OR rats[tiab] OR rabbit[tiab] OR rabbits[tiab] OR “child”[mesh] OR children[tiab] OR Case Reports[ptyp] OR cancer[tiab] OR chemo-therapy[tiab] OR “Neoplasms”[Mesh] OR “cartilage” [Mesh] OR cartilage[tiab] OR “stroke”[Mesh] OR “Brain Infarction”[Mesh]).
3.B
Table of Included T
2and T
2*Publications
Author (year) Disease/ Contr ol (n) T2 /T 2 * Disease (ms) T2 /T 2 * Contr ol (ms) P-value ROI place Seq. NOS Population MYOCARDIAL INF ARCTION (T 2 * ) 1.5T GE (Zia et al. 2012) F0:62 F1:62 F2:62/62 32.4 37.7 37.3 37.4 38.4 38.2 ă .01 NS NS 3SAX GRE 2,0,2 STEMI patients within 2d(F0), 3w(F1) and 6m(F2) after PCI. Remote as contr ol 1.5T Philips (Durighel et al. 2016) H+:30 H-:30 /30 33.8 ˘ 14.1 a 54.0 ˘ 17.9 b 45.0 ˘ 9.4 c 0.16 bc Single SAX mid infar ct GRE 1,0,2 STEMI patients referr ed for CMR in 7 days post-PCI. Haemorr hagic hypointense LGE infar ct(H+) or non-haemorr hagic infar ct(H-). Re-mote as contr ol.1.5T Siemens (Bulluck et al. 2016a) CF0:15 CF1:15 IF0:13 IF1:13 /28 11.3 ˘ 1.5 15.0 ˘ 1.5 29.7 ˘ 10.0 32.0 ˘ 5.8 32.3 ˘ 3.9 33.3 ˘ 3.1 Segments of 3SAX 1,0,2 STEMI 4d(F0) and 5m(F1) post-PCI. Hypo-cor e (T 2 *ă 20ms)(C), in-far ct(I) 2SD above remote myocar -dium. Remote as contr ol. (Bulluck et al. 2017a) 26/26 13 ˘ 3 33 ˘ 4 ă .01 Segments 2,0,2 STEMI PCl ă 2h, CMR at 4d post-PCI. Hypo-cor e (T 2 * ă 20ms) mea-sur ed. Remote as contr ol. (Carberry et al. 2018) CF0:203 CF1:203 ZF0:203 ZF1:203 /203 14.2 ˘ 3.6 16.6 ˘ 2.1 32.4 ˘ 7.6 25.7 ˘ 4.4 31.5 ˘ 2.4 3SAX 2,0,2 STEMI at 2d(F0) and 6m(F1) post-PCI. Hypo-cor e(C) (T 2 * ă 20ms) and infar ct zone(Z). Remote as contr ol.
1.5T Siemens (Carrick et al. 2016a) CF0: ˘ 0 CF1:30 CF2:30 CF3:30 ZF0:30 ZF1:30 ZF2:30 ZF3:30 /30 17.8 ˘ 6.0 14.1 ˘ 4.1 16.7 ˘ 5.9 18.9 ˘ 6.2 29.2 ˘ 5.8 26.6 ˘ 4.8 28.6 ˘ 3.3 29.2 ˘ 4.0 31.9 ˘ 2.0 32.9 ˘ 1.9 32.6 ˘ 1.6 32.4 ˘ 2.3 3SAX 1,0,3 STEMI 4-12h(F0), 3d(F1), 10d(F2) and 7m(F3) post-PCI. T2 * in infar ct zone(Z) (T 2 ą 2SD remote) and infar ct cor e(C) (center in the infar ct with mean T2 /T 2 * ă 2SD T2 / T2 * pe-riphery). Remote as contr ol. (Kali et al. 2013) H+:7 H-:7/14 15.9 ˘ 4.5 a 37.8 ˘ 2.5 b 35.2 ˘ 2.1 c ă .01 ac SAX whole LV GRE 1,0,2 STEMI within ˘ 3d post-PCI. LGE+ infar cts. Hypo-cor es on the T2 *-weighted ă 2SD refer ence ROI (H+), non-haemorr hagic (H-). Remote as contr ol.
(Moham- madzadeh et al. 2018) I:20 P:20 /20 35.5 ˘ 3.6 30.7 ˘ 4.9 29.4 ˘ 4.5 ă .01 NS 3SAX 2LAX 1,0,2 non-STEMI ě 6m af-ter MI. T2 *fr om in-far ct(I) (LGE+ myo-car dium) and peri-infar ct(P). Remote as contr ol (Robbers et al. 2018) C:43 B:43 /43 26.3 ˘ 10.7 30.7 ˘ 7.7 27.3 ˘ 6.9 1SAX at mid infar ct 2,0,2 STEMI 4-6d post-PCI. inf a rct cor e(C) (LGE+ based) and bor der zone(B). Remote as contr ol. (Roghi et al. 2015) H+F0:7 H+F1:6 H-F0:8 H-F1:8 17 18 31 31 3SAX necr otic ar ea GRE 1,0,1 STEMI ă 5d(F0) and 6m(F1) af-ter PCI. LGE+ as myocar dial haemor -rhagic(H+) or non-haemorr hagic(H-). (Y ilmaz et al. 2013) I:14 P:14 /14 24.0 ˘ 12.4 35.7 ˘ 10.7 32.0 ˘ 4.9 3SAX in-far ct GRE 1,0,2 STEMI within 2-7d post-PCI. Infar ct cor e(I)(LGE+ with hyper enhanced T2) and peri-infar ct(P) (LGE+ without hyper enhanced T2 ar ea). Remote as contr ol.
3T Philips (Chen et al. 2019) F0:22 F1:22 F2:22 F3:22/22 22.0 ˘ 3.1 23.9 ˘ 3.3 22.1 ˘ 4.0 21.5 ˘ 2.8 31.2 ˘ 1.6 30.0 ˘ 0.7 30.4 ˘ 0.8 30.3 ˘ 0.7 3SAX TFE 2,0,2 STEMI ă 2d(F0), ă 3d(F1), ă 7d(F2) and ă 30d(F3) after PCI. Infar ct val-ues (LGE+ based). Remote as contr ol. (Zaman et al. 2015) 6/15 16.1 ˘ 7.6 24.2 ˘ 6.7 Stack SAX GRE 2,0,2 STEMI 2d post-PCI. Intramyocar dial haemorr hage (hypo-cor e on LGE+). MYOCARDIAL INF ARCTION (T 2 ) 1.5T GE (Zia et al. 2012) F0:62 F1:62 F2:62/62 56.7 51.8 39.8 43.4 39.5 39.5 ă .01 ă .01 NS 5SAX in-far ct T2- prep 2,0,2 STEMI 2d(F0), 3w(F1) or 6m(F2) post-PCI. LGE+ segments. Remote as contr ol. 1.5T Philips (Nakamori et al. 2019) 14 45 Mean 16 AHA 1,0,1 Patients with cor o-nary artery disease.
(T ahir et al. 2017) F0:67 F1:50 F2:44 F3:45 /67 84 ˘ 10 68 ˘ 9 61 ˘ 7 58 ˘ 4 55 ˘ 3 Mid- SAX TSE 2,0,3 Acute MI with PCI after 8d(F0), 7w(F1), 3m(F2) and 6m(F3). Infar ct (LGE+ ar ea without hypo-intense ar ea). Remote as contr ol 1.5T Siemens (Bulluck et al. 2016a) F0:15 F1:15 /13 49.7 ˘ 5.7 47.3 ˘ 4.1 49.3 ˘ 2.5 46.7 ˘ 2.5 Segments 3SAX 1,0,2 STEMI 4d(F0) and 5m(F1) post-PCI. Hypo-cor e (T 2 *ă 20ms). Remote as contr ol. (Bulluck et al. 2017a) H+C:26 H+S:26 /26 H-C:13 H-S:13 /13 50 ˘ 4 66 ˘ 6 57 ˘ 4 66 ˘ 7 51 ˘ 3 50 ˘ 3 Segments 3SAX 2,0,2 STEMI PCl ă 2h, CMR at 4d post-PCI. Hypo-cor e (T 2 *ă 20ms)(H+) and without (H-) in infar ct cor e LGE+ (C) or salvage (S). Remote as contr ol.
(Carberry et al. 2017) F0:283 F1:283 /283 66.3 ˘ 6.1 56.8 ˘ 4.5 49.7 ˘ 2.3 ă .01 ă .01 SAX whole LV T2- prep True- FISP 1,0,2 STEMI 2d(F0) and 6m(F1) post-PCI. In-far ct (SI ą 5SD above remote region). Re-mote as contr ol. (Carrick et al. 2016a)
CF0:30 CF1:30 CF2:30 IF0:30 IF1:30 IF2:30 IF3:30 /50
55.5 ˘ 6.9 51.8 ˘ 4.6 59.2 ˘ 3.6 62.8 ˘ 6.7 61.4 ˘ 4.1 68.1 ˘ 3.7 54.0 ˘ 2.8 49.5 ˘ 2.5 SAX T2- prep True- FISP 1,1,3 STEMI 4-12h(F0), 3d(F1), 10d(F2) and 7m(F3) post-PCI. Infar ct(I) (T 2 ą 2SD above remote) and infar ct cor e(C) (center infar ct with a mean T2 /T 2 * ą 2SD below periphery). (Carrick et al. 2016) 171 54 ˘ 5 SAX whole LV T2- prep True- FISP 2,0,2 STEMI 2d post-PCI. Infar ct cor e (T1 ă 2SD of periph-ery). (Haig et al. 2019) C:245 I:245 /245 53.9 ˘ 4.8 62.9 ˘ 5.1 49.7 ˘ 2.1 SAX whole LV T2- prep True- FISP 1,0,3 STEMI 2d post-PCI. Infar ct zone (I) (T 2 ą 2SD above remote) and cor e (C) (center infar ct with a mean T2 /T 2 * ą 2SD below periphery). Remote as contr ol.
