ABSTRACTS/PROCEEDINGS Multimodality imaging: bird’s eye view from The European Society of Cardiology Congress 2014 Barcelona, August 30-September 3, 2014 Jeroen J. Bax, MD, PhD,a Victoria Delgado, MD, PhD,a Stephan Achenbach, MD,b Udo Sechtem, MD, PhD,c and Juhani Knuuti, MD, PhDd a

Department of Cardiology, Heart Lung Centrum, Leiden University Medical Center, Leiden, The Netherlands b Department of Cardiology, University Hospital Erlangen, Erlangen, Germany c Department of Cardiology, Robert-Bosch-Krankenhaus, Stuttgart, Germany d Turku PET Centre, Turku University Hospital, University of Turku, Turku, Finland

doi:10.1007/s12350-014-0013-3

Continuing the tradition of the Journal publishing abstracts presented at the American Society of Nuclear Cardiology and other major international meetings, this issue covers the 2014 European Society of Cardiology (ESC) congress hosted in Barcelona from August 30 to September 3. At this ESC meeting, a total of 413 imaging abstracts (9% of the total) were presented. Four authors were invited with expertise in echocardiography (VD), cardiac computed tomography (CT) (SA), cardiac magnetic resonance (CMR) (US), and nuclear imaging (JK) and were asked to summarize the findings of 4-7 abstracts in these areas that were of most interest to them. These abstracts were integrated by one of the Editors of the Journal (JB). ECHOCARDIOGRAPHY Innovation was the main topic of the ESC congress of this year, and numerous abstracts have shown the increasing use of speckle tracking echocardiography (STE) in clinical practice and further advances in 3dimensional (3D) echocardiography. The role of 2-dimensional STE to detect subclinical left ventricular (LV) dysfunction was demonstrated in 1,065 type 1 diabetic patients without known heart disease.1 Jensen et al evaluated the presence of diabetic

Reprint requests: Jeroen J. Bax, MD, PhD, Department of Cardiology, Heart Lung Centrum, Leiden University Medical Center, Albinusdreef 2, 2300 RC Leiden, The Netherlands; [email protected] J Nucl Cardiol 2014;21:1245–51. 1071-3581/$34.00 Copyright Ó 2014 American Society of Nuclear Cardiology.

cardiomyopathy associated with kidney disease, defined by the presence of albuminuria. Despite similar LV ejection fraction (LVEF), LV global longitudinal strain impaired progressively across the groups from -18.5% in patients with normoalbuminuria (n = 739), to -17.9% in patients with microalbuminuria (n = 223) and -17.4% in patients with macroalbuminuria (n = 103) (P \ .001). At multivariate analysis, LV global longitudinal strain remained independently associated with micro- (P = .036) and macroalbuminuria (P \ .001). These findings indicate that LV global longitudinal strain is a more sensitive marker of LV systolic dysfunction than LVEF and permits an earlier identification of patients with diabetic cardiomyopathy. Recent advances in 2-dimensional STE allow for analysis of multi-layer myocardial mechanics (from endocardium to epicardium). Using multi-layer 2-dimensional STE Chan et al2 demonstrated that endocardial LV longitudinal strain was the most sensitive marker of ischemia during dobutamine stress echocardiography. LV longitudinal and circumferential strains were measured at the endo-, mid-, and epicardium at rest and peak dose in 30 patients. Of 480 segments, 46 showed ischemia at peak dose. Endocardial and mid-myocardial LV longitudinal strain were significantly impaired in ischemic segments compared with normal segments at peak dose (-16% ± 6.6% vs -24.3% ± 13.7%, P \ .001 and -15.3% ± 5.7% vs -20.9% ± 9.7%, P \ .001; respectively). In contrast, multi-layer LV circumferential strain showed a transmural gradient from endocardium to epicardium without significant differences between ischemic and normal segments. This novel algorithm may increase the sensitivity of dobutamine stress echocardiography by 1245

1246

Bax et al Multimodality imaging

assessing the longitudinal mechanics of the most vulnerable layer to ischemia. Fusion imaging combining anatomical data from multi-detector row CT coronary angiography and functional data from 3D STE represents a further step in the diagnosis of significant coronary artery stenosis. Casas Rojo et al3 demonstrated the feasibility of novel post-processing software that creates a 3D model overlaying the coronary artery anatomy onto a LV volume that displays the regional strain (Figure 1). Fifteen patients with coronary artery disease (CAD) underwent multi-detector row CT coronary angiography and 3D STE. Ten patients had abnormal regional strain at rest at the areas supplied by coronary arteries with significant stenosis. Of 5 patients who underwent stress echocardiography, 1 patient showed myocardial ischemia at peak stress with impaired LV strain in the regions supplied by a significantly diseased coronary artery. The clinical application of this hybrid imaging prototype (providing integrated information on coronary artery anatomy (i.e., stenosis) and the functional consequences, i.e., ischemia) needs additional studies demonstrating its diagnostic accuracy in larger populations. Risk stratification of patients with ischemic heart disease has been the focus of many abstracts. Evaluation of intra-ventricular flow patterns and vortices and its association with LV aneurysms and thrombus formation in ischemic heart disease patients was presented by Rodrı´guez Mun˜oz et al.4 Wall shear strain was assessed with a novel echocardiographic tool, the so-called Vector Flow Mapping, which integrates information of

Journal of Nuclear CardiologyÒ November/December 2014

intra-ventricular flow velocity variation next to the myocardial wall and shape and size of LV aneurysms. In 11 patients with LV aneurysms, the magnitude of wall shear stress along the cardiac cycle was semi-quantified using a 5-point scale. Patients with C4 wall shear stress had larger LV aneurysms in volume (15.2 ± 2.1 mL vs 12.1 ± 2.5 mL, P = .06) and length (13.2 ± 2.6 mm vs 7.9 ± 2.1 mm, P \ .001) compared with patients with \4 wall shear stress. Interestingly, 75% of patients with \4 wall shear stress had thrombus inside the aneurysm, whereas this was not observed in patients with C4 wall shear stress. Further validation of this hypothesis may permit better stratification and identification of patients that benefit from anticoagulation. Finally, comprehensive analysis of the proximal isovelocity surface area (PISA) of the mitral regurgitant jet throughout the systole using 3D echocardiography and customized software permits accurate quantification of mitral regurgitant volume taking into consideration the dynamic changes of the regurgitant jet. In 60 patients with mitral regurgitation (72% organic), Brugger et al5 compared 3D transthoracic (TTE) and transesophageal (TEE) echocardiographic measurements of mitral regurgitant volume with CMR. Compared with CMR, TTE underestimated the mitral regurgitant volume significantly more than TEE (27 ± 3 mL/beat vs 34 ± 3 mL/ beat; CMR: 39 ± 3 mL/beat; P = .006). The superior spatial resolution of TEE may explain these findings. Advances in TTE transducers may overcome this limitation.

Figure 1. Hybrid imaging with fusion of multi-detector row computed tomography coronary (CT) angiography and 3D speckle tracking echocardiography. A shows a total occlusion of the left anterior descending (LAD) coronary artery at distal level and a significant stenosis at mid level on the 3D rendering of the multi-detector row CT coronary angiography data overlaid onto a 3D cast of the LV showing the 3D strain pattern. The apex, supplied by the diseased LAD, is color-coded in blue which indicates abnormal strain (thinning and lengthening) suggesting ischemia. B shows the polar plot map of 3D LV strain data showing abnormal strain in the apical regions supplied by the diseased LAD. Reproduced with permission from Casas Rojo et al.3

Journal of Nuclear CardiologyÒ Volume 21, Number 6;1245–51

Bax et al Multimodality imaging

1247

COMPUTED TOMOGRAPHY Abstracts presented during the 2014 ESC congress covered the wide spectrum of situations in which multidetector computed tomography (CT) may potentially be useful both for non-invasive and invasive and interventional cardiology. Coronary artery imaging remains the major application of cardiac CT. A major area of research is the functional assessment of coronary artery stenoses by CT. One potential approach is the noninvasive determination of fractional flow reserve based on computational fluid dynamics (FFRCT). Using a method that allows performing this analysis on-site, Coenen et al6 compared FFRCT with invasively measured FFR in 106 patients, and reported a sensitivity and specificity of 86% and 65%. FFRCT, once sufficiently validated, may constitute an excellent method to clarify cases in which coronary CT angiography (CTA) shows intermediate degree stenosis. However, the diagnostic accuracy needs further evaluation in large prospective trials. Another major research focus is the prognostic value of CT, including the important question whether obstructive lesions, non-obstructive plaque, specific plaque features, or coronary calcium constitute the most relevant prognostic parameter. In 254 patients with coronary CTA performed within one month after an acute coronary syndrome, Ito et al7 observed obstructive lesions in non-culprit vessels in 59% of patients; 24% of patients (n = 63) had obstructive stenoses with ‘‘high risk’’ features (positive remodeling and low CT attenuation). Cardiac events (7 acute coronary syndromes, 22 late revascularizations) occurred in 26 patients and incidence was 17.5% in patients with obstructive highrisk plaque, 8.6% in patients with obstructive non-highrisk plaque, and 6.4% in patients without obstructive plaque. Hazard ratios (HR) were significantly increased when comparing high-risk obstructive plaque to patients without obstructive plaque (3.62, 95% confidence interval [CI] 1.36-10.60, P = .01), but not when comparing patients with non-high-risk obstructive plaque to patients without obstructive lesions (HR 1.48, 95% CI 0.51-4.53, P = .46) (Figure 2). The results suggest that specific plaque features may be of decisive prognostic importance, although larger studies with hard end-points are needed. The relevance of non-calcified plaques on contrastenhanced CT remains of interest. Plank et al8 evaluated 711 consecutive asymptomatic individuals with highrisk for CAD and identified non-calcified plaque in 33% of individuals without coronary artery calcium. During a mean follow-up of 2.7 years, no patient without atherosclerosis on CT experienced an event (acute coronary syndrome or cardiac death), as compared to 1.2% of

Figure 2. Survival free of events (acute coronary syndrome and late revascularizations) in 254 patients in whom coronary computed tomography angiography was performed within 1 month after an index acute coronary syndrome. Event free survival is significantly decreased in patients with obstructive lesions that displayed high-risk features (HRP, defined as positive remodeling and low CT attenuation) were present in non-culprit vessels. Reproduced with permission from Ito et al.7

patients with any form of plaque in CT. The presence of any non-calcified plaque increased the event risk slightly (HR 2.36; 95% CI 1.20-4.63; P \ .01). Interestingly, Auscher et al9 postulate that calcium indicates ‘‘plaque stabilization’’ since they observed an increased dense calcium volume when patients with recent infarction were randomized to early very high dose as compared to normal dose statin therapy. On the other hand, Heo et al10 demonstrated that coronary artery calcium alone is a very strong risk marker. In a cohort of 9,715 asymptomatic individuals followed for 15 years, 4,864 had no calcium and the ‘‘warranty period’’ of a zero calcium score (with an event rate of less than 4 per 1,000 patient years) was 12 years both for men and women. The utility of cardiac CT for interventional cardiology was addressed in several interesting studies. Hamdan et al11 studied 52 patients receiving a balloon expandable aortic valve and found a mean length of the membranous septum of approximately 6 mm in 18 patients who experienced atrio-ventricular block as compared to approximately 9 mm in patients who did not. Especially, the relationship of implantation depth to length of the membranous septum was predictive of atrio-ventricular block. Of note, information regarding the dimensions of the membranous septum is available ‘‘for free,’’ since almost all patients undergo contrastenhanced CT imaging before the percutaneous aortic valve procedure. Finally, Major et al12 reported on the identification of left atrial thrombus by contrast-enhanced CT. In 201 patients with atrial fibrillation undergoing (pre-ablation)

1248

Bax et al Multimodality imaging

CT and TEE, CT had a sensitivity of 100% and negative predictive value of 100% compared to TEE, but while specificity was 90%, the positive predictive value was only 17% (19 false-positive findings and 4 true-positive findings). A useful reminder of two facts: CT can rule out, but not positively identify, left atrial thrombus and also, sensitivity and specificity are useless parameters when the prevalence of disease is not taken into account. CARDIAC MAGNETIC RESONANCE There were two interesting foci of CMR research presented at the ESC meeting in Barcelona. Several abstracts reported on the use of T1 and T2 mapping techniques in various diseases. The second focus was on the application of CMR in electrophysiology. Hinojar et al13 studied 165 patients with a clinical diagnosis of viral myocarditis. Patients underwent CMR at 1.5 and 3 Tesla for T2 imaging, late gadolinium enhancement (LGE) and T1 mapping of the LV prior to and [20 minutes after contrast administration. Compared to controls, T1-indices were increased in patients with acute symptoms and in convalescence. Native T2 values were helpful to distinguish between healthy controls and patients independent of the stage of the disease. A combination of LGE and native T1 was useful to define the convalescent stage. T1 values in patients with acute myocarditis were[5 standard deviation above the normal range, whereas T1 values were less elevated in the convalescent stage ([2 standard deviation). The same group also presented data on myocardial T2 mapping in a smaller group of myocarditis patients.14 This study was performed on a 3 Tesla scanner. Midventricular short-axis T2 values were found to be elevated both in patients with acute and chronic myocarditis. Although T2 values tended to be concordant with native T1 values, the authors did not comment on the respective value of each of the measurements. The use of 3 Tesla magnets, however, makes this new diagnostic method difficult to apply at many centers which still use 1.5 T magnets for their clinical work. Kammerlander et al15 used LV biopsy samples to validate T1 mapping both in the native state and postcontrast administration. Biopsies were performed in 22 heart failure patients, 16 of whom had heart failure with preserved LVEF. The amount of LV extracellular matrix correlated significantly with modified look-locker inversion recovery (MOLLI) measured extracellular volume and with multiple breath-hold post-contrast T1 times. In contrast, there was no correlation with native T1 times. Nakamori et al16 also studied heart failure patients with preserved LVEF using a MOLLI sequence. The LV extracellular volume was assessed using hematocritadjusted myocardial and blood T1 values measured

Journal of Nuclear CardiologyÒ November/December 2014

before and after administration of gadolinium contrast medium and compared findings to a control group. Patients with heart failure with preserved LVEF had significantly greater mean extracellular volume, (31.6% ± 7.4% vs 26.9% ± 3.0%, P \ .05) and significantly higher pre-contrast T1 values than control group persons although there was some overlap (1372 ± 86 vs 1305 ± 34 ms, P \ .05). The somewhat conflicting results of these two abstracts indicate that T1 mapping may still not be at a stage where standard measurements are reliable to identify diffuse myocardial fibrosis. Left ventricular scar is a potential substrate for ventricular arrhythmias. Nucifora et al17 investigated the relation between the site of origin of monomorphic ventricular arrhythmias and myocardial tissue characteristics among patients with non-ischemic LV structural heart disease. The site of origin of the arrhythmia was identified using QRS morphology on ECG during the arrhythmia. The arrhythmias usually originated from the scar zone depicted by CMR which was intramyocardial in 6 patients and subepicardial in 16 patients. This confirms that contrast-enhanced CMR may be of value for mapping and ablation procedures. Border zone channels have been identified as critical isthmuses of ventricular tachycardia. Penela et al18 used 3D high-resolution 3T contrast-enhanced CMR within the first 6 days after ST-segment elevation myocardial infraction and at 6 months follow-up. A 3D color-coded shell map was obtained to depict the core and the border zone distribution. Myocardial scar was classified as arrhythmogenic if border zone channels were present. Forty-four border zone channels were present at baseline and 41 of them remained in the same segment/orientation at 6 months. Two patients had spontaneous sustained ventricular arrhythmias during follow-up and both had border zone channels in the acute phase. These findings suggest that 3D contrast-enhanced CMR may be useful for sudden cardiac death risk stratification. The specificity of these findings, however, is rather low. Finally, two large studies on the value of LGE imaging deserve mentioning. Shanbhag et al19 studied the prevalence and mortality risk for atypical patterns of LGE in a community-based cohort of older adults. The ICELAND-MI (Imaging Cardiac Evaluation to Locate Areas of Necrosis and Detect Myocardial Infarction) study with a total sample size of 931 patients included 67 patients with larger areas of atypical LGE which predicted a higher risk of all-cause mortality compared to normal subjects (HR 1.78, P = .014) (Figure 3). Minor patterns of atypical LGE (LGE near the aortic root, mitral annulus, and right ventricular insertion points) were present in 237 patients and carried a HR for all-cause mortality of 1.24 (P = .19). These data show that atypical larger LGE is associated with increased

Journal of Nuclear CardiologyÒ Volume 21, Number 6;1245–51

Bax et al Multimodality imaging

1249

Figure 3. Asymptomatic subject from the ICELAND study. Left panel systolic frame from cine CMR shows normal LV wall thickening. Right panel midwall LGE consistent with scar from prior myocarditis or non-ischemic cardiomyopathy. Reproduced with permission from Shanbhag et al.19

mortality in older community-dwelling subjects. This may be important in the interpretation of CMR studies of the heart performed for various indications in older patients. The prognostic information of the presence of myocardial scar was reported by Giusca et al20 in patients with diabetes undergoing LGE CMR. There were 1,969 consecutive patients grouped according to the presence or absence of diabetes and LGE. Hard events were cardiac death and non-fatal myocardial infarction. Revascularization procedures were also noted after a blanking period of the first 3 months after the CMR study. The median follow-up time was almost 3 years and 90 patients experienced a hard cardiac event. Both diabetes and LGE were predictors of hard events. Patients with both LGE and diabetes had a markedly higher likelihood for future hard cardiac events as compared to all other groups (HR of 5.6 for both features present vs. both features absent, HR of 4.3 for both features present vs diabetes without LGE and HR of 2.5 for both features present vs LGE only). These data show that in diabetics the presence of LGE is a strong predictor of an adverse clinical course. NUCLEAR IMAGING In the field of single-photon emission computed tomography (SPECT) imaging, various trends can currently be identified. Several studies addressed radiation dose reduction. This can be achieved by using new, more sensitive imaging devices or using stress-only imaging protocols in patients without previous myocardial infarction. In 4,852 patients without known CAD, Sharir et al21 reported on the feasibility and rates of very

low isotope-dose stress-only myocardial perfusion SPECT with cadmium-zinc-telluride (CZT) technology as compared to conventional scans. The stress radiation dose with CZT technology was B2 mSv in 84.7% of patients and B3 mSv in 99.9% of patients. The stressonly imaging protocol (without a rest scan) was more often successfully performed with the CZT technique as compared to the standard scan approach (in women 58.8% vs 45.6% and in men 32.1% vs 22.9%, respectively, P \ .001), which resulted in very low radiation doses. It has recently been concluded that the standard predictive models (such as the Diamond and Forrester model) overestimate the incidence of significant CAD and new updated models have been suggested. This could lead to the situation that in patients believed to have intermediate probability of CAD, the ischemic findings become less common. Interestingly, signs of this phenomenon could be identified in a large registry study by Pizzi et al.22 During a period of 11 years, 10,529 gated-SPECT myocardial perfusion scans were performed. The presence of ischemia gradually reduced from the years 1997-2000 to 2006-2008 from 24.3% to 12.4% despite the age of the population has increased. This suggests that imaging is commonly used in lower range of intermediate pretest probability of CAD if the predictive models are not revised; alternatively, the use of more aggressive medical therapy may also have resulted in reduced the presence of ischemia. As quantification of myocardial perfusion in absolute terms has become clinically feasible, microvascular disease can also be accurately detected. Coronary microvascular dysfunction is defined as impaired myocardial perfusion in the absence of epicardial artery

1250

Bax et al Multimodality imaging

Journal of Nuclear CardiologyÒ November/December 2014

Figure 4. Inflammation may have significant contribution to the pathogenesis of abdominal aortic aneurysm (AAA). The arrow points the inflammatory focus in the aorta of the patient with AAA. Adapted with permission from Joshi et al.24

obstruction. However, the prevalence of this phenomenon remains unknown among symptomatic patients with an intermediate probability of CAD. Soukka et al23 investigated 189 patients using CT coronary angiography and quantitative positron emission tomography (PET) perfusion imaging. In this population with an intermediate probability of CAD, some features of coronary microvascular dysfunction could be identified in 9%. However, coronary microvascular dysfunction without any coronary atherosclerosis was rare (1%). Coexistence of coronary microvascular dysfunction with non-obstructive CAD (3%) and obstructive CAD (6%) was more common. Molecular imaging is another evolving area of research; PET imaging for inflammatory cardiac diseases and vessel wall inflammation is one of the focuses. Joshi et al24 hypothesized that inflammation and calcification may have a differential contribution to the pathogenesis of abdominal aortic aneurysm. Using a hybrid scanner, aortic inflammation was assessed with 18F-fluorodeoxyglucose (FDG) PET imaging and calcification with CT in 63 patients with asymptomatic abdominal aortic aneurysm and in 19 age- and sexmatched patients with atherosclerosis. Aortic inflammation was more pronounced in abdominal aortic aneurysm patients than in patients with atherosclerosis suggesting an exaggerated systemic vascular inflammatory response in abdominal aortic aneurysm patients (Figure 4). Moreover, the same group questioned whether acute myocardial infarction causes generalized atherosclerotic inflammation and progression.25 Using PET/CT imaging with 18F-FDG, 40 patients with recent myocardial

infarction and 40 patients with stable angina pectoris were evaluated. Compared with patients with stable angina, patients with myocardial infarction had increased aortic 18F-FDG uptake. The study supports the hypothesis that in patients with recent myocardial infarction metabolic activity in remote atherosclerotic plaques is increased and this may destabilize remote atheromatous plaques and potentially cause an increase in recurrent athero-thrombotic events. Another hybrid imaging approach is PET/CMR. Langwieser et al26 evaluated PET/CMR imaging for predicting LV function after myocardial infarction. Both PET imaging with 18F-FDG and MRI using LGE have been used separately to detect myocardial viability. In this study, regional myocardial FDG uptake and LGE were studied 5-7 days after acute myocardial infarction. The standard viable vs. non-viable threshold using FDG PET ([50% uptake compared with remote myocardium) performed well in discriminating patients with and without functional recovery. For LGE-CMR, \25% transmurality was associated with functional recovery. Eventually, the integration between FDG PET and LGECMR may provide the optimal prediction of functional recovery.

References 1. Jensen MT, Sogaard P, Andersen HU, Bech J, Hansen TF, Biering-Sorensen T, et al. Global longitudinal strain by speckletracking echocardiography is impaired in type 1 diabetes patients with albuminuria—the Thousand&1 Study. Eur Heart J. 2014;35:1119.

Journal of Nuclear CardiologyÒ Volume 21, Number 6;1245–51

2. Chan J, Burstow D, Ischenko M, Hamilton-Craig CR, Platts DG, Haseler L, et al. The impact of myocardial ischemia on transmural mechanics using multi-layer speckle tracking echocardiography. Eur Heart J. 2014;35:959. 3. Casas Rojo E, Fernandez-Golfin C, Gorissen W, Mogelvang R, Hassager C, Fuchs A, et al. Feasibility of a new hybrid imaging system featuring fusion of multislice coronary tomography and 3D speckle-tracking echocardiography. Eur Heart J 2014;35:958 4. Rodrı´guez Mun˜oz D, Carbonell San Roman A, Moya Mur JL, Reyes S, Pastor Pueyo P, Del Val Martin D, et al. Echocardiographic assessment of wall shear stress: association with the development of adverse remodeling after myocardial infarction. Eur Heart J 2014;35:958 5. Brugger N, Wustmann K, Huerzeler M, Steck H, Seiler C. Impact of imaging modality, TTE vs. TEE, on the measurement of mitral regurgitant volume by three-dimensional integrated PISA. Eur Heart J. 2014;35:40. 6. Coenen A, Lubbers MM, Kurata A, Kono A, Dedic A, Saru RD, et al. Coronary CT angiography-derived fractional flow reserve, performed on-site using a novel reduced-order model, validated by invasive fractional flow reserve. Eur Heart J. 2014;35:715. 7. Ito H, Motoyama S, Sarai M, Kawai H, Nagahara Y, Takada K, et al. Prediction of cardiac events at non-culprit lesions in patients with acute coronary syndromes by plaque characteristics on coronary computed tomography angiography. Eur Heart J. 2014;35:459. 8. Plank F, Friedrich GJ, Dichtl W, Franz W, Feuchtner GM. Coronary computed tomography angiography in asymptomatic highrisk patients: non-calcifying and total coronary plaque burden predict outcome. Eur Heart J. 2014;35:453. 9. Auscher S, Heinsen L, Vinther K, Lambrechtsen J, Egstrup K. Effects of intensive lipid lowering therapy on coronary plaques composition in patients with acute myocardial infarction. Assessment with serial coronary CT-angiography. Eur Heart J. 2014;35:452. 10. Heo R, O’Hartaigh B, Gransar H, Valenti V, Gomez M, Shaw L, et al. Gender-based warranty period of a coronary artery calcium score of zero: a 15-year follow-up study. Eur Heart J. 2014;35:683. 11. Hamdam A, Guetta V, Konen E, Raanani E, Klempfner R, Segev A, et al. Computed-tomographic assessment of membranous septal anatomy for risk prediction of atrioventricular block associated with transcatheter aortic balloon expandable valve implantation. Eur Heart J. 2014;35:200. 12. Major GP, Szilveszter B, Horvath T, Kovacs A, Pataki S, Vago H, et al. Diagnostic performance of cardiac CT in detecting left atrial thrombus. Eur Heart J. 2014;35:201. 13. Hinojar R, Arroyo E, Foote L, Dabir D, Mahmoud I, Jackson T, et al. Native T1 values discriminate between health and disease in acute and convalescent stage of disease in clinically suspected myocarditis. Eur Heart J. 2014;35:119–20. 14. Hinojar R, Foote L, Arroyo Ucar E, Dabir D, Schnackenburg B, et al. Myocardial T2 mapping for improved detection of

Bax et al Multimodality imaging

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

1251

inflammatory myocardial involvement in acute and chronic myocarditis. Eur Heart J. 2014;35:120. Kammerlander A, Tufaro C, Pfaffenberger S, Marzluf B, Aschauer S, Babayev J, et al. Myocardial biopsy for the validation of cardiac magnetic resonance T1 mapping for quantification of extracellular matrix. Eur Heart J 2014;35:176 Nakamori S, Dohi K, Onishi K, Ishida M, Fujimoto N, Ichikawa Y, et al. Diffuse myocardial fibrosis assessed by cardiac magnetic resonance contrast-enhanced T1 mapping in heart failure with preserved ejection fraction. Eur Heart J. 2014;35:331. Nucifora G, Muser D, Masci PG, Barison A, Piccoli G, Rebellato L, et al. Relation between site of origin of monomorphic ventricular arrhythmias and myocardial tissue characteristics in nonischemic left ventricular heart disease. Eur Heart J. 2014;35:705– 6. Penela D, Acosta J, Fernandez-Armenta J, Andreu D, De Caralt TM, Mont L, et al. Identification of the arrhythmogenic substrate in acute phase of STEMI. Eur Heart J. 2014;35:269. Shanbhag SM, Aspelund T, Cao JJ, Schelbert EB, Kellman P, Eiriksdottir G, et al. Prognosis of atypical late gadolinium enhancement by cardiovascular MR in older adults in the ICELAND-MI study. Eur Heart J. 2014;35:37. Giusca S, Kelle S, Nagel E, Buss SJ, Puntmann V, Wellnhofer E, et al. The presence of myocardial scar carries additional prognostic information in patients with diabetes who undergo late gadolinium enhancement magnetic resonance imaging. Eur Heart J. 2014;35:353. Sharir T, Kovalski G, Pinskiy M, Rochman A, Prochorov V, Bojko A, et al. Increased rate of very low isotope-dose stress-only myocardial perfusion SPECT using CZT technology and quantitative analysis. Eur Heart J. 2014;35:1133. Pizzi MN, Romero-Farina G, Aguade-Bruix S, Limeres J, AndresVillareal MN, Garcia-Dorado D, et al. Progressive reduction in positive ischemic gated-SPECT studies in the last years. Eur Heart J. 2014;35:1135. Soukka I, Maaniitty T, Saraste A, Pikkarainen E, Uusitalo V, Ukkonen H, et al. Characterization of coronary microvascular dysfunction in patients with suspected coronary artery disease. Eur Heart J. 2014;35:1138. Joshi NV, Elkhawad M, Rajani NK, Tarkin JM, Melville AJ, Richards JMJ, et al. Differential aortic inflammation and calcification in abdominal aortic aneurysm and atherosclerosis. Eur Heart J. 2014;35:455. Joshi NV, Toor I, Shah ASV, van Beek EJR, Fletcher A, Mills NL, et al. Systemic atherosclerotic inflammation following acute myocardial infarction. Eur Heart J. 2014;35:521. Langwieser N, Rischpler C, Batrice A, Laugwitz KL, Schwaiger M, Nekolla S, et al. Simultaneous PET/MR imaging for predicting left ventricular function after myocardial infarction. Eur Heart J. 2014;35:458.