Cardiac MRI for myocardial and vascular heart disease 1183 shown in table 1. There was a moderate-strong correlation between RVEF and both LS and eRVEF (r=-0.47, r=0.67, p=0.0012 and p<0.0001 respectively). TR severity by TRF was strongly correlated with LS and eRVEF (r=0.58, r=-0.90; p<0.0001 and p<0.0001 respectively). TRF was also correlated to RVEF, how- ever a less strong correlation was observed (r=-0.381, p=0.011). Table 1. Baseline characteristics Age (median ± IQR) 73±6 Women (n, %) 36, 80% Native tricuspid valve (n, %) 37, 82.2% Significant valve disease (different from TR) (n, %) 22, 48.9% RVEF, % (mean ± SD) 60.9±9% eRVEF (mean ± SD) 35±12.8% LS, % (mean ± SD) -15.2±3.7% TRF, % (mean ± SD) 42.1±15.1% Conclusions: LS and eRVEF are novel RV measures of systolic function. The stronger correlation with TR severity might suggest its role as an earlier param- eter of LV dysfunction. Its potential to predict a worse prognosis needs further evaluation. P5639 Myocardial extracellular volume assessed by cardiac magnetic resonance T1 mapping as a marker of diastolic function: comparative study of CMR and strain echocardiography S. Kato 1 , N. Saito 1 , N. Asahina 1 , N. Iinuma 1 , D. Kamimura 1 , T. Nakachi 1 , K. Fukui 1 , T. Iwasawa 1 , M. Kosuge 2 , K. Kimura 2 , K. Tamura 3 . 1 Kanagawa Cardiovascular and Respiratory Center, Cardiology, Yokohama, Japan; 2 Yokohama City University Medical Center, Cardiology, Yokohama, Japan; 3 Yokohama City University, Yokohama, Japan Background: Cardiac magnetic resonance (CMR) T1 mapping has been pro- posed to quantify myocardial extracellular volume fraction (ECV). Early diastolic strain rate (Esr) assessed by speckle tracking echocardiography reflects myocar- dial relaxation. Purpose: The aim of this study was to assess the relationship between ECV by CMR T1 mapping and Esr by speckle tracking echocardiography. Methods: One hundred and seventy one patients (mean age: 67±12 years, male: 135 (79%)) underwent echocardiography and CMR. T1 mapping was acquired by modified Look-Locker inversion recovery (MOLLI) sequence using 1.5T MR scanner equipped with 32 channel cardiac coils. Longitudinal Esr (LEsr, 1/sec) and circumferential Esr (CEsr, 1/sec) were quantified by speckle tracking echo- cardiography. Patients were divided into two groups based on ECV value (low ECV group: ECV<0.30, n=129; high ECV group: ECV≥0.30, n=42). Results: Both LEsr and CEsr were significantly reduced in high ECV group than low ECV group (LEsr: 0.32±0.14 vs 0.38±0.13, p=0.01; CEsr: 0.53±0.28 vs 0.67±0.25, p<0.01). Significant negative correlation was observed between LEsr and ECV (r=-0.244, p<0.01), CEsr and ECV (-0.249, p<0.01) (Figure). Conclusions: ECV assessed by CMR T1 mapping correlates with diastolic strain parameters. ECV evaluated by CMR T1 mapping might be useful as a non- invasive imaging marker of myocardial diastolic function. Funding Acknowledgements: Research Grant, MSD Life Science Foundation, Public Interest Incorporated Foundation P5640 Global longitudinal and circumferential strain in combination with ejection fraction: an approach to measure cardiac function in ischaemic, dilative and inflammatory disease L. Winau, F. Pathan, R. Heinke, J. Haslbauer, V. Puntmann, E. Nagel. JW Goethe University, kardiovaskuläre Bildgebung, Frankfurt am Main, Germany Background: Global longitudinal and global circumferential strain (GLS, GCS) are well-established techniques in FT-CMR, but are currently not widely utilised in everyday clinical practice. Our research focuses on the value of performing strain in combination with ejection fraction (EF) as a useful tool to assess cardiac function. Methods: We performed CMR on a total of 112 subjects with a 3T scanner (is- chaemic cardiomyopathy n=26, diffuse inflammatory disease n=20, dilated car- diomyopathy n=32 and healthy volunteers n=34). Short and long axis views were acquired by CINE imaging for feature tracking analysis to obtain global longitudi- nal and circumferential strain, as well as ejection fraction from a SAX stack. Results: Significant differences in longitudinal and circumferential strain were measured across patient subgroups. Compared to the controls, ischaemic and DCM patients showed both significantly reduced GLS, GCS and EF (p<0.001). Patients with diffuse inflammatory disease did not present differences in GLS, but a significant decrease in GCS (p=0.038) and EF. However, EF was still in normal range (57.79±9.43). Correlation analysis revealed the association between both types of strains and EF. While GLS and GCS appear to be equally correlating with cardiac function in patients with ischemia, GCS and EF have a higher correlation than GLS and EF in DCM patients (r=-0.9 vs. r=-0.84). In patients with inflamma- tory disease, the correlation between GLS and EF is stronger than GCS and EF (r=-0.81 vs. r= -0.69). Table 1 n GLS p GCS p EF P mean std mean std mean std HV 34 -22.09 3.36 -27.36 6.15 63.46 6.09 ISC 26 -15.28 7.25 <0.001 -17.30 7.52 <0.001 46.38 17.00 <0.001 DIF-INF 20 -20.64 4.36 0.193 -23.62 5.73 0.038 57.79 9.43 0.012 DCM 32 -16.02 6.11 <0.001 -18.48 7.90 <0.001 43.16 13.43 <0.001 GLS = Global Longitudinal Strain; GCS = Global Circumferential Strain; EF = Ejection Fraction; HV = Healthy Volunteers; ISC = Ischeamic Heart Disease; DIF-INF = Diffuse Inflammatory Dis- ease; DCM = Dilatative Cardiomyopathy. Conclusion: GLS, GCS and EF provide additional invaluable information on car- diac function in ischaemic, dilated and inflammatory disease. Strongly reduced GCS in DCM indicates that contraction impairment is more severe than in is- chaemic patients. In patients with diffuse inflammatory disease, preserved EF in conjunction with decreased GCS and normal GLS imply that longitudinal shorten- ing compensates circumferential inwards movement. Therefore, we predict that a rapid decrease in GLS would result in acute loss of cardiac function and poorer prognosis. P5641 CMR and 4dflow-based analysis of alterations in post ischemic dilated cardiomiopathy before and after surgical ventricular restoration F. Piatti 1 , A. Camporeale 2 , S. Pozzi 1 , G. Di Giovine 2 , S. Pica 2 , S. Castelvecchio 3 , L. Menicanti 3 , A. Greiser 4 , E. Votta 1 , A. Redaelli 1 , M. Lombardi 2 . 1 Milan Polytechnic, Electronic, Information and Bioengineering, Milan, Italy; 2 IRCCS Polyclinic San Donato, Multimodality Cardiac Imaging, Milan, Italy; 3 IRCCS, Policlinico San Donato, Department of Cardiac Surgery, San Donato Milanese, Italy; 4 Siemens Healthcare GmbH, Erlangen, Germany Background: The contribution of Cardiac Magnetic Resonance (CMR) in the study of post-ischemic dilated cardiomyopathy (iDCM) consist of the analysis of LV volume, function and location/extension of ischemic scar. Recently, the use of 4D Flow in the characterization of intraventricular flow patterns allowed to quantify novel in vivo makers, such as wall shear stresses base-to-apex gradi- ents (DeltaWSS). These aspects related to intraventricular flow have never been analysed in iDCM, most likely due to 4D Flow extensive and challenging post- processing. Purpose: By exploiting a novel, fast and semi-automated software tool we aimed to assess through 4D Flow the fluid-dynamic adaptation of iDCM and to charac- terize changes after surgical ventricular restoration against controls. Methods: Study population included 24 patients (all males, age=65.9±9.4years) with iDCM and LV dysfunction and 18 healthy controls (9 males, age=37.1±13.5years). CMR protocol on a Magnetom Aera 1.5 T (Siemens Healthcare, Erlangen, Germany) included CINE images, LGE images and 4D Flow with acquisition volume targeted on LV cavity. 5 patients repeated CMR 6 months after surgical intervention (coronary-artery bypass + surgical ventricular restoration). Nt-proBNP was dosed in patients with iDCM at baseline and at 6 months follow up. A custom in-house software tool was exploited to quantify: global LV function (ESV, EDV, EF) from CINE; location, extent and transmurality of scar tissue from LGE; diastolic wall shear stress (WSS) distributions and DeltaWSS from 4D Flow (Fig. 1). Results: The analyses were successful on all datasets (n=47) requiring 20±5 minutes and few minor manual interventions. In patients with iDCM, WSS was lower on scar tissue than remote myocardium, both in diastole (p=0.082) and Downloaded from https://academic.oup.com/eurheartj/article-abstract/39/suppl_1/ehy566.P5641/5082031 by guest on 05 June 2020