Initial CMR Research Articles

2424Long-term outcome of patients with biopsy-proven viral myocarditis: 12-year results from a late gadolinium enhancement cardiac magnetic resonance study

Abstract Purpose Myocarditis is a common cardiac disease that is associated with significant mortality as demonstrated by several studies. Late gadolinium enhancement (LGE) by cardiac magnetic resonance imaging (CMR) is a valuable tool for risk stratification of patients with suspected myocarditis. Previous studies using CMR-LGE have reported a good negative predictive value over follow-up periods of 4–6 years, while its positive predictive value was only modest. However, there is a lack of data regarding the long-term prognosis (>10 years) of these patients. This study reports an extended long-term follow-up of a large cohort of patients with biopsy-proven viral myocarditis. Methods At initial presentation, all patients underwent endomyocardial biopsy and CMR for the work-up of suspected myocarditis or unclear cardiomyopathy and had evidence of virus in PCR analyses. The primary endpoints were: all-cause death, cardiac death, and the occurrence of sudden cardiac death (SCD). Results 183 patients with biopsy-proven viral myocarditis were followed for a median of 11.5 years. At baseline, patients were 52 years old, 31% were females, and the median ejection fraction was moderately reduced (44%). Initial CMR assessment revealed LGE in 101 (55%) patients, while 82 (45%) patients had LGE-negative CMR. During the follow-up period, the following endpoints occurred in the overall cohort: all-cause death (n=71, 39%), cardiac death (n=50, 27%) and SCD (n=20, 11%). Most importantly, only a single LGE-negative patient experienced a SCD during this 12-year follow-up, while all other SCDs occurred in patients with LGE-positive CMR (1% vs. 19%, p<0.001). Consequently, the negative predictive value (NPV) of normal CMR-LGE regarding SCD was 98%. In addition, cardiac mortality (12% vs. 40%, p<0.001, NPV=88%) and all-cause mortality (20% vs. 54%, p<0.001, NPV=79%) were significantly lower in patients without LGE. Conclusion This cohort of biopsy-proven viral myocarditis demonstrates substantial mortality (39% in 11.5 years). However, absence of LGE on CMR was associated with favorable prognosis. This was applicable regarding all-cause and cardiac mortality, but most importantly with regard to SCD with a NPV of 98% over almost 12 years median follow-up. Acknowledgement/Funding This work was funded by the Robert-Bosch-Stiftung, Stuttgart, Germany and the Berthold-Leibinger-Stiftung, Ditzingen, Germany.

Relation of Quantity of Subepicardial Adipose Tissue to Infarct Size in Patients With ST-Elevation Myocardial Infarction

According to the so-called obesity paradox, obesity might present a protective role in patients with myocardial infarction. We aimed to assess the influence of the epicardial adipose tissue (EAT) volume on cardiac healing and remodeling in patients with acute ST-elevation myocardial infarction. We prospectively included 193 consecutive patients presenting a first STEMI without known coronary artery disease. Cardiac magnetic resonance imaging was performed at baseline and after a 3-month follow-up. EAT volume was computed, and the population was divided into quartiles: the highest quartile of EAT defining the high EAT group (h-EAT). h-EAT was associated with increased body mass index, higher rate of history of hypertension, and smaller infarct size at initial CMR assessment (18.3 ± 11.9% vs 23 ± 13.7% of total left ventricular [LV] mass, p= 0.041). Moreover, microvascular obstruction was less frequent in the h-EAT group (36.2% vs 59.3%, p= 0.006). There were no differences in LV ejection fraction (LVEF), LV volumes, systolic wall stress, coronary artery burden, and clinical events during the index hospitalization between the EAT groups at baseline and at follow-up. Linear regression analysis showed h-EAT to be associated with smaller infarct size at baseline (β coefficient=-3.25 [95% CI-5.89 to-0.61], p= 0.016). h-EAT also modified positively the effect of infarct size on LV remodeling, as assessed by the change in LVEF (p= 0.046). In conclusion, h-EAT was paradoxically related to smaller infarct size and acted as an effect modifier in the relation between the extent of infarct size and LVEF changes. Patients with higher extent of EAT presented better cardiac healing.

0094 : Influence of epicardial adipose tissue on infarct size after a STEMI with successful reperfusion: a CMR study

The Epicardial Adipose Tissue (EAT) is related to both the early development of atherosclerosis and the severity of coronary artery disease (CAD). We aimed to assess the influence of the EAT volume which was measured by cardiac magnetic resonance imaging (CMR) on infarct size and on clinical outcomes in acute ST-elevation myocardial infarction (STEMI) with successful reperfusion. This prospective study involved 193 patients with a first STEMI admitted to the University Hospital of Angers. CMR was performed at baseline and after a three-month follow-up in order to investigate common CMR findings, such as left ventricular (LV) volumes, ejection fraction (EF), infarct size, micro-vascular obstruction (MVO), systolic wall stress (SWS) and EAT volume. We divided the population into quartiles – the greater quartile of EAT defining the high EAT group (>33.5 ml). Clinical outcomes were recorded during the index hospitalization and a 1-year follow-up. Patients in high EAT group were significantly associated with increased body mass index and hypertension, presented a smaller infarct size (18.3±11.9 vs 23±13.7% of total LV, p=0.041) at initial CMR assessment and after a 3-month follow-up (p=0.01), MVO occurred less frequently (36.2 vs 59.3%, p=0.006). We found no difference between EAT groups in regard to LVEF, LV volumes, SWS as well as in regard to coronary burden and clinical events during the index hospitalization or the follow-up. EAT was an independent predictor of low infarct size in univariate and multivariate analysis; B=–3.25 [–5.894; –0.615], p=0.016.Fig. 1 EAT volume > 33.5 ml was paradoxically related to smaller infarct size in STEMI but didn’t seem to be related to changes in LVEF, LV remodelling or further clinical events. The author hereby declares no conflict of interest Figure Example of Epicardial Adipose tissue assessment

An unexpected cause of hypertrophic myocardium.

A 60-year-old man with known atrial fibrillation (AF) was referred for routine cardiovascular magnetic resonance imaging (CMR) prior to AF ablation. He was normotensive and had no other previous medical conditions. A transthoracic echocardiogram had shown good systolic function with moderate left ventricular hypertrophy and mildly dilated atria (Fig. 1). Figure 1: Four-chamber transthoracic echocardiography demonstrating moderate myocardial hypertrophy and mildly dilated atria. Initial CMR imaging for anatomy and function demonstrated normal ventricular size and mildly impaired systolic function (ejection fraction = 50%, normal >55%) with no significant valvular pathology. There was asymmetrical septal hypertrophy measuring 16 mm (normal <13 mm) in the basal septum versus 9 mm in the lateral wall at the same level with significantly elevated left ventricular mass at 241 g (Fig. 2a). There was no systolic anterior motion of the mitral valve or left ventricular outflow tract obstruction at rest. Both atria were slightly enlarged, the left atrium measuring 35 cm2 and the right atrium measuring 36 cm2. To identify the etiology of hypertrophy, conventional late gadolinium enhancement (Fig. 2b) was undertaken alongside novel native T1 mapping sequences (Fig. 3 and Supplementary Videos 1 and 2). Acquisition of late gadolinium images was technically challenging, giving the impression of diffuse enhancement associated with a relatively dark blood pool, both of which are suspicious of amyloid infiltration of the heart [1]. The native T1 maps showed an elevated septal T1 relaxation time (1148 ms, normal <1050 ms), supporting the initial impression of cardiac amyloid and further suggesting that this was most likely due to light chain deposition (AL amyloid) [2]. AL amyloid was subsequently confirmed by serum free light chain immunoassay and bone marrow trephine biopsy. Figure 2: (a) A hypertrophic myocardium with mildly dilated atria. (b) Diffuse enhancement associated with a relatively dark blood pool (difficult ‘nulling’) of the myocardium, typical of cardiac amyloid. Figure 3: T1 map following motion correction and co-registration to enable calculation of native T1 values. In this case, the septal value was elevated at 1148 ms suggestive of light chain amyloid deposition (AL amyloid) and demonstrating how new techniques can ... A combination of late gadolinium enhancement and novel T1 mapping sequences have the ability to correctly determine the underlying pathology in patients with hypertrophic myocardium and potentially help in the identification of the amyloid class.

0341: Predictor of left ventricular remodeling after acute myocardial infarction: a study of end-systolic wall stress

investigate whether the end-systolic wall stress is a predictor of left ventricular remodeling (LVR) in the aftermath of an acute myocardial infarction after successful reperfusion. LVR is a poor prognosis outcome associated with a greater number of major adverse cardiovascular events. It remains difficult to predict which patients will remodel. 169 STEMI patients were prospectively included in a CMR study. We calculate, among other parameters, end-systolic wall stress (WS) by three-dimensional MRI method with an home-made software. CMR was performed at day 5±2 and repeated at 3 months follow-up. LVR was defined as a LV end-diastolic volume indexed (LVEDVi) >120 ml/m 2 at 3 months. 13 patients presented LVR, including 11 due to anterior MI. LVR patients presented worse initial CMR parameters: WS 25.9±6 vs 16.0±4103N•m-2 (p <0.001), a LVEDVi 117.2±20 vs 84.6±16 ml/m 2 (p<0.001), an infarct size 46.6±20 vs 22.8±15g (p <0.001), a microvascular obstruction size 8.7±1.9 vs 9±3g (p <0.001) and a LV ejection fraction 31.0±8 vs 49.0±9% (p <0.001). The initial global WS emerges as an independent predictor of LVR (OR 1.298 [1.046 to 1.612], p <0.018) as LVEDVi (OR 1.093 [1.013 to 1.180], p <0.022) and the occurrence of heart failure (OR 9.912 [1.094 to 89.842], p <0.041). LVEF as infarct and microvascular obstruction size were not independent predictors. Patients with an initial global WS below 20.88 103 N•m-2 will not develop LVR in 98.5% of cases (sensitivity and specificity of 84.6%). end-systolic wall stress is an independent predictor of LV remodeling in post-MI. Patients with an initial global WS below 20.88 103N•m-2 presented lower risk for LVR. Abstract 0341 – Figure: AUC: area under the curve

Effect of ivabradine on left ventricular remodelling after reperfused myocardial infarction: A pilot study

Heart rate is a major determinant of myocardial oxygen demand; in ST-segment elevation myocardial infarction (STEMI), patients treated with primary percutaneous intervention (PPCI), heart rate at discharge correlates with mortality. Ivabradine is a pure heart rate-reducing agent that has no effect on blood pressure and contractility, and can reverse left ventricular (LV) remodelling in patients with heart failure. To evaluate whether ivabradine, when added to current guideline-based therapy, improves LV remodelling in STEMI patients treated with PPCI. This paired-cohort study included 124 patients between June 2011 and July 2012. Ivabradine (5mg twice daily) was given promptly after PPCI, along with beta-blockers, to obtain a heart rate<60 beats per minute (ivabradine group). This group was matched with STEMI patients treated in line with current guidelines, including beta-blockers (bisoprolol), according to age, sex, infarct-related coronary artery, ischaemia time and infarct size determined by initial cardiac magnetic resonance imaging (CMR) (control group). Statistical analyses were performed according to an intention-to-continue treatment principle. CMR data at 3 months were available for 122 patients. Heart rate was lower in the ivabradine group than in the control group during the initial CMR (P=0.02) and the follow-up CMR (P=0.006). At the follow-up CMR, there was a smaller increase in LV end-diastolic volume index in the ivabradine group than in the control group (P=0.04). LV end-systolic volume index remained unchanged in the ivabradine group, but increased in the control group (P=0.01). There was a significant improvement in LV ejection fraction in the ivabradine group compared with in the control group (P=0.04). In successfully reperfused STEMI patients, ivabradine may improve LV remodelling when added to current guideline-based therapy.

Effect of ivabradine on left ventricular remodelling after reperfused myocardial infarction: a pilot study

Purpose: To evaluate whether ivabradine, when added to current guideline-based therapy, improves Left Ventricular (LV) remodelling in STEMI patients treated with successful PPCI. Methods: This case-control study included sixty-two patients, between June 2011 and July 2012 (the ivabradine group). Ivabradine, at the dose of 5 mg twice daily, was given promptly after PPCI, along with beta-blockers, to obtain HR <60 b.p.m. This group was matched with STEMI patients according to age, sex, infarct-related coronary artery, ischemia time and infarct size, determined by initial CMR (the control group). Six patients in the ivabradine group were excluded from the final analysis. Follow-up CMR data at 3 months were available for 112 patients. Results: The HR showed a significant reduction between 1 hour after angioplasty and 3 months of treatment in both groups (76.8±7.3 vs 54.7±5.8 and 71.5±11.9 vs 57.7±6.8 b.p.m, respectively, p<0.001). Furthermore HR in the ivabradine group was lower than the control group during initial CMR (56.3±5.0 vs 58.9±6.2 b.p.m, p=0.04). This difference increased during follow-up CMR (54.7±5.8 vs 57.7±6.8 b.p.m, p=0.02). Infarct sizes were similar between both groups. At follow-up CMR, treatment with ivabradine was associated with a smaller increase in LVEDV index (75.0±19.1 vs 72.6±15.1 ml/m2) vs control group (78.7±17.3 vs 72.8±15.5 ml/m2) (p=0.04). In the ivabradine group LVESV index remained unchanged (31.0±14.7 vs 30.9±11.2 ml/m2), whereas in the control group LVESV index increased (35.5±14.3 vs 32.5±10.4 ml/m2) (p=0.04). There was a nonsignificant trend toward greater improvement of LVEF in the ivabradine group (60.1±10.1 vs 58.1±9.8%) versus the control group (56.3±11.0 vs 56.4±9.2%) (p=0.09). Conclusions: In successful reperfused STEMI patients, ivabradine improves LV remodelling when added to current guideline-based therapy including beta-blockers. Given these promising results, larger studies are necessary to confirm our findings

A Decade Follow-up of a Thalassemia Major (TM) Cohort Monitored by Cardiac Magnetic Resonance Imaging (CMR): Significant Reduction In Patients with Cardiac Iron and In Total Mortality

AbstractAbstract 1011 Background.CMR was introduced in London to assess myocardial iron loading in 1999 and some of these patients now have 10 years of follow-up, most with contemporary CMR determinations. The impact of long-term monitoring of myocardial iron loading in thalassemia major (TM) on the proportion of patients with increased myocardial iron (cT2* <20ms) and on patterns of mortality has not been previously described in a longitudinal cohort over this duration. Patients and Methods.All patients regularly attending two London thalassemia centres, who received their first CMR Jan 1999 - Dec 2000 were analyzed as a cohort. Patients underwent initial CMR at the Royal Brompton Hospital and received CMR follow up (FU) either there or at the Heart Hospital (UCLH). 132 patients were identified as having received a CMR in 1999–2000. A minimum 7 years CMR FU was required for inclusion in the long-term CMR analysis. 109 patients had at least 7 years of CMR follow up (range 7.0–10.6 years, median 9.2). The median age at 1st CMR was 27.9 years (range 7.7 – 49.5 years). At baseline, patients were receiving chelation with deferoxamine (DFO) monotherapy (70%), deferiprone (DFP) monotherapy (21%), or a combination of these agents (9%). At latest FU, patients were receiving DFO (32%), deferasirox (DFX) (28%), DFP (22%), or combined DFP and DFO therapy (18%). Results: Improvement in cardiac iron:In 1999–2000, 60% of TM patients had cT2* values ≤20ms and 17% had cT2* values <10ms. By contrast, at long term FU, only 23% now have cT2* ≤20ms, 7% have cT2* values <10ms (p<0.001). Changes to chelation therapy:31% of patients stayed on the same chelator; 33% had 1 chelator switch, 26% 2 switches and 11% 3 or more switches. 18 switches in chelation therapy were due to side-effects (12 DFP, 5 DFX, 1 DFO). There were 9 breaks in chelation therapy during pregnancy in 8 different women. The proportions of patients with T2* < 20ms fell significantly for those who remained on DFO or DFP monotherapies throughout, or who changed chelation modalities on only one (p=0.002) or two (p=0.02) occasions. Patients who received had 3 or more switches did not show a improvement in this respect. The latter group was also the only subset that showed significant deterioration in myocardial iron (p<0.001). Mortality rates:the overall mortality rate for the initial cohort was 1.65 per 1000 patient years (95% CI 0.71 – 3.24); median age at death 35.6 years (range 27.3–48.4). This is a substantial improvement in the mortality index compared with the UK thalassemia registry data, of 4.3 per 1000 patient years during the period 2000–2003 (Modell et al, JCMR, 2008). The incidence rate ratio is 0.387 (95% CI 0.11–0.961), p<0.05, with patients in our cohort 61% less likely to die than those in the 2000–2003 cohort. Causes of death:there were 8 deaths during the FU period: 3 with complications of hepatitis C (all with cT2* > 20ms), 3 with sepsis (2 with cT2* <10ms and impaired ejection fraction, 1 with cT2* of 18ms), 1 with breast cancer, 1 with sudden unexplained death (cT2* > 20ms). Thus in only 2 patients could excessive cardiac iron loading be considered a causal/contributory factor. There was no significant difference in the baseline cT2* between those who died and those currently still alive (p= 0.2), meaning that death as a drop-out cause does not explain iron loading trends over FU. Chelators at death: DFO (4), DFP (2), DFX (1), combination (1). Conclusions:Over a decade we have seen an almost 3 fold fall in the proportion of patients with myocardial iron overload. Mortality has become substantially lower and cardiac iron overload is no longer the leading cause of mortality. In addition to CMR, this decade has seen the advent of two new oral iron chelators and many patients switched chelation regimen, sometimes several times, during the follow up period. Whilst the contribution of the individual components of this practice to the improved outcome cannot be concluded without randomized studies, it is clear that this modern management of TM is associated with reduced TM mortality. Disclosures:Off Label Use: Deferiprone is off label in the USA but licensed in Europe. Shah:Novartis: Honoraria, Speakers Bureau; Apotex/ Swedish Orphan: Honoraria. Pennell:Siemens: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding; Novartis: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding; Apotex: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees; Cardiovascular Imaging Solutions: Director of CVIS, Equity Ownership. Porter:Novartis: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding, Speakers Bureau.

Pulmonary valve replacement in tetralogy of Fallot improves the repolarization

Objective To assess the effect of pulmonary valve replacement (PVR) on the repolarization of patients with tetralogy of Fallot. Background Pulmonary valve regurgitation may cause right ventricular failure in adult patients with Fallot's tetralogy. In these patients, prolonged depolarization and disturbed repolarization are associated with ventricular arrhythmias and sudden cardiac death. Methods Thirty Fallot patients (age 32 ± 9 years, 19 male) eligible for PVR were studied with cardiac magnetic resonance imaging (CMR) before and 6 months after PVR. Electrocardiograms obtained during initial and follow-up CMR were analyzed and occurrence of ventricular arrhythmias was studied. Results Right ventricular end-diastolic volume (RV EDV) decreased from 322 ± 87 to 215 ± 57 ml after PVR ( P < 0.0001). The spatial QRS-T angle normalized from 117 ± 34 to 100 ± 35°, P = 0.0004 (normal angle < 105°). QT dispersion and T-wave complexity did not change significantly. T-wave amplitude decreased from 376 ± 121 to 329 ± 100 μV ( P = 0.01). T-wave area decreased from 43 ± 15 to 38 ± 13 μV s ( P = 0.02). Decreases in T-wave amplitude and area were most prominent in the right precordial leads overlying the RV. Three patients had sustained ventricular arrhythmias and one patient died suddenly. These patients had a QRS duration > 160 ms. No severe ventricular arrhythmias were found in patients with a RV EDV < 220 ml, QRS-T angle < 100°, QT dispersion < 60 ms or T-wave complexity < 0.30. Conclusion Normal repolarization indices may be associated with the absence of severe ventricular arrhythmias. PVR in Fallot patients with dilated right ventricles has a beneficial effect on electrocardiographic indices of repolarization heterogeneity.