A Ghosh Dastidar

Out of hospital cardiac arrest survivors with inconclusive coronary angiogram: Impact of cardiovascular magnetic resonance on clinical management and decision-making☆

BACKGROUND Non-traumatic out of hospital cardiac arrest (OHCA) is the leading cause of death worldwide, mainly due to acute coronary syndromes. Urgent coronary angiography with view to revascularisation is recommended in patients with suspected acute coronary syndrome. Diagnosis and management of patients with inconclusive coronary angiogram (unobstructed coronaries or unidentified culprit lesion) is challenging. We sought to assess the role of Cardiovascular Magnetic Resonance (CMR) in the diagnosis and management of OHCA survivors with an inconclusive coronary angiogram. METHODS AND RESULTS This is a retrospective multicentre CMR registry analysis of OHCA survivors with an inconclusive angiogram. Clinical, ECG and multi-modality imaging data were analysed. Clinical impact of CMR was defined as a change in diagnosis or management. Out of 174 OHCA survivors referred for CMR, 110 patients (63%, 84 male, median age 58) had an inconclusive angiogram. CMR identified a pathologic substrate in 76/110 patients (69%): ischemic heart disease was found in 45 (41%) and non-ischemic heart disease in 31 (28%). A structurally normal heart was found in 25 patients (23%) and non-specific findings in 9 (8%). As compared to trans-thoracic echocardiogram, CMR proved to be superior in identifying a pathologic substrate (69% vs 54%, p=0.018). The CMR study carried a clinical impact in 70% of patients, determining a change in diagnosis in 25%, in management in 29% and a change in both in 16%. CONCLUSIONS CMR showed a promising role in the diagnostic work-up of OHCA survivors with inconclusive angiogram and its wider use should be considered.

Diagnostic yield of cardiovascular magnetic resonance in young-middle aged patients with high-grade atrio-ventricular block

BACKGROUND Atrio-ventricular block (AVB) is a rare finding in young or middle-aged adults, often leading to pacemaker implantation (PM) without further investigation. We sought to assess the diagnostic role of cardiovascular magnetic resonance (CMR) in young and middle-aged adults with high-grade AVB. METHODS We consecutively enrolled young-middle aged (18-65years) patients with high grade AVB referred to CMR after standard clinical assessment (history, electrocardiogram and cardiac rhythm monitoring) prior to PM implantation. Cine and post-contrast imaging were performed in a 1.5T scanner. RESULTS 34 patients (59% male, mean age 42±12years) with high grade AVB were referred to CMR for suspected ischemic heart disease (IHD)(n=4) and non-ischemic heart disease (NIHD)(n=20); no clear cause was found in 9 patients prior to CMR and 1 patient had suspected lung disease. A pathologic substrate was found on CMR in 15 patients (44%), while a structurally normal heart was reported in 18 (53%). Non-specific findings were reported in 1 patient (3%). There was a fair agreement between CMR and echocardiographic findings (Cohens kappa 0.243), and CMR provided an entirely new diagnosis in 34% of patients. As compared to the standard clinical assessment, CMR had an additional role in 65% of patients and guided further testing (genetic testing, extra-cardiac imaging) in 9%. CONCLUSIONS CMR found a pathologic substrate in 44% of patients, mainly NIHD (32%). Half of the patients (53%) had a structurally normal heart. When added to the standard clinical assessment, CMR had an incremental diagnostic role in two thirds of patients.

1 Prevalence of extra-cardiac findings detected by cardiac MRI in inherited vs acquired cardiovascular diseases

Introduction With its large field of view, Cardiovascular Magnetic Resonance (CMR) allows the detection of extra-cardiac pathologies (ECP). Both cardiologists and radiologists should be able to recognise ECP and identify those requiring further investigation. The aim of our study is to assess the difference in prevalence of ECP in patients with suspected inherited cardiac conditions vs acquired heart disease. Materials and methods We reviewed 1.817 consecutive clinical CMR studies performed in the biggest CMR department in Southwest England to look for ECP. Demographic characteristics and scans indications were also recorded. For each scan the presence of ECP and its relevance (need for further investigation, i.e. suspected lung malignancy) was assessed. The internal record system (Picture Achievement and Communication System, PACS) was used to check whether the ECP were previously known, or whether it represents a new finding. Abstract 1 Figure 1 (A) Axial Haste showed a nodule (arrow) in the superior lobe of the right lung. (B) On the High Resolution Computed Tomography (HRCT), performed to further assess the ECP, the presence of the nodule in the superior lobe of the right lung was confirmed (head-arrow) Results We analysed 1,817 scans, referred for the assessment of inherited cardiac condition (Group A, n = 906) and acquired heart disease (Group B, n = 911). There was no significant difference in prevalence of ECP between the two groups (p = 0.63). ECP were found in 26% of patient in Group A, 4% of which requiring further assessment; 69% previously unknown (Figure 1). ECP were reported in 27% of patients in Group B, 5% requiring further assessment; 68% were previously unknown. Conclusion One in four patient has an extra-cardiac finding and the prevalence of ECP did not differ in patients presenting with inherited conditions vs acquired heart disease.

6 Prevalence and CMR characteristics of apical HCM

Hypertrophic cardiomyopathy (HCM) is the most common cause of sudden cardiac death in young adults. The 3 main phenotypes are asymmetric (most common), concentric and apical. Literature suggests apical HCM is rare and more benign, but data is scarce. We sought to describe prevalence and characteristics of apical HCM in a large CMR service. Methods We reviewed 3,100 scans (Jan 2014–Mar 2015). Protocol included cines, early and late gadolinium enhancement imaging. 114 consecutive HCM patients were identified. Asymmetric HCM defined as septal/ free wall thickness ratio > 1.3; apical HCM as apical wall thickness > 15mm or apical/basal wall thicknesses ≥ 1.3–1.5. Concentric HCM defined as symmetrical hypertrophy of ventricular wall without regional preferences. Non-apical HCM (asymmetric and concentric phenotypes) were compared with apical HCM. Fisher’s exact t-test and unpaired t-test were performed for statistical significance (P-value < 0.05, statistically significant). Univariate and multivariate logistic regression analyses were performed to determine CMR predictors of apical HCM. Results 10 patients were excluded, leaving 104 patients, median age 60years; 70% male. 70% had non-apical HCM (5 patients concentric HCM, the rest asymmetric HCM) and 30% apical HCM. Mean maximum LV wall thickness, indexed LV mass, stroke volume, prevalence of LVOTO and SAM were greater in non-apical group. Presence of LGE was high in both groups (>85%) and wasn’t statistically different. Univariate predictors of apical HCM included maximum LV wall thickness, indexed stroke volume, LVOT obstruction (Table 1). In the multivariate model, maximum wall thickness remained the only significant predictor. Abstract 6 Table 1 CMR characteristics of Apical vs non-Apical HCM CMR findings: Total Cohort (n = 104) Non-apical (n = 73) Apical (n = 31) P-value Mean LVEF (%) 69.7 68.4 72.6 0.0552 Mean LVEDVI (mL/m²) 73.7 76.7 66.8 0.0718 Mean LVESVI (mL/m²) 23.8 25.5 20.1 0.1177 Mean indexed stroke volume 53.1 55.9 46.4 0.0333 Mean max. LV wall thickness (mm) 18.2 19.3 15.6 0.0001 Mean indexed LV mass 93.5 98.4 82.4 0.0102 LVOTO 35.2 41.1 12.5 0.0403 SAM 31.4 38.9 6.25 0.0143 LGE (%) 86.9 85.7 89.7 0.8063 LVEF, left ventricular ejection fraction; LVEDVI, left ventricular end diastolic volume index; LVESVI, left ventricular end systolic volume index; LVOTO, Left ventricular outflow tract obstruction; SAM, systolic anterior valve motion; LGE, late gadolinium enhancement. Conclusions Our study suggests that prevalence of apical HCM is almost 1/3rd of all observed cases. It also demonstrates that prevalence of LGE was high in apical HCM, suggesting that better prognosis of apical HCM based on the absence of myocardial fibrosis, should be reconsidered. Further large trials are needed to better understand the pathophysiology.

8 Diagnostic performance of ECG detection of left atrial enlargement in patients with arterial hypertension relative to the cardiac magnetic resonance gold-standard: Impact of obesity

ECG may demonstrate evidence of left atrial enlargement (LAE), which has adverse prognostic implications. We sought to determine the accuracy of 5 ECG criteria of LAE in a hypertensive cohort relative to CMR and to investigate the confounding effect of obesity. Methods Consecutive referrals for CMR from a tertiary hypertension clinic were reviewed. Patients with any concomitant cardiac pathology were excluded. ECGs were assessed, blinded to CMR data, for: 1) P wave >110ms, 2) P-mitrale (notched P wave with inter-peak duration >40ms), 3) P wave axis <30°, 4) Area of negative P terminal force in lead V1 (NPTF-V1) >40ms•mm and 5) Positive P terminal force in aVL (PPTF-aVL) >0.5mm. Maximal LA volume index (LAVI) was measured by the biplane area-length method. Abstract 8 Table 1 A) Diagnostic performance of the various ECG parameters at detecting left atrial enlargement; B) Obesity subgroup analysis of diagnostic performance of the various ECG parameters at detecting left atrial enlargement A) Diagnostic performance of the various ECG parameters Prevalence ECG LAE (%) ROC-AUC (95th CI) Sensitivity (%) Specificity (%) PPV (%) NPV (%) ACC (%) P > 110 ms 9 0.497 (0.384 – 0.610) 9 91 25 74 69 P mitrale 1 0.495 (0.382 – 0.608) 0 99 0 74 73 P axis < 30o 27 0.437 (0.328 – 0.546) 18 70 17 71 56 NPTF-V1 > 40 ms.mm 17 0.465 (0.355 – 0.576) 12 81 18 72 63 PPTF-aVL > 0.5 mm 8 0.502 (0.389 – 0.616) 9 92 27 74 70 Any ECG criteria for LAE 46 0.387 (0.279 – 0.495) 29 48 17 65 43 B) Subgroup analysis by obesity P > 110 ms Non-obese 10 0.515 (0.352 – 0.679 12 91 33 74 70 Obese 9 0.519 (0.357 – 0.681) 12 92 33 75 72 P mitrale Non-obese 2 0.529 (0.364 – 0.695) 0 98 0 74 73 Obese 0 0.500 (0.340 – 0.660) 0 100 0 75 75 P axis < 30o Non-obese 21 0.560 (0.395 – 0.725) 29 83 38 76 68 Obese 34 0.467 (0.309 – 0.625) 29 64 22 73 55 NPTF-V1 > 40 ms.mm Non-obese 11 0.504 (0.342 – 0.667) 12 89 29 73 68 Obese 21 0.478 (0.320 – 0.636) 18 78 21 74 63 PPTF-aVL > 0.5 mm Non-obese 6 0.537 (0.371 – 0.703) 12 96 50 75 73 Obese 10 0.469 (0.314 – 0.625) 6 88 14 73 67 Any ECG criteria for LAE Non-obese 41 0.620 (0.462 – 0.778) 59 65 38 81 63 Obese 51 0.475 (0.315 – 0.635) 47 48 24 73 48 (LAE = left atrial enlargement, ROC-AUC = receiver operator curve-area under curve, CI = confidence interval, PPV = positive predictive value, NPV = negative predictive values, ACC = accuracy) Results 130 patients were included (age: 51.4 ± 15.1 years, 47% male, 51% obese, systolic blood pressure: 171 ± 29 mmHg, diastolic blood pressure: 97 ± 15 mmHg). The prevalence of LAE by CMR was 26% and by ECG varied from 1% (P-mitrale) to 27% (P axis <30o), and was 46% when ≥1 ECG LAE criteria were present. There was no significant difference in mean LAVI when ≥1 ECG LAE criterion was present compared to when no ECG LAE criteria were present (47 ± 15 vs 50 ± 15 ml/m2, p = 0.235). All the individual ECG LAE criteria were more specific than sensitive (Table 1/A), with specificities ranging from 70% (P axis <30o) to 99% (P-mitrale). Obesity attenuated the specificity of most of the individual ECG LAE criteria (Table 1/B). Obesity correlated with significant lower specificity (48% vs 65%, p < 0.05) and a trend towards lower sensitivity (59% vs 43%, p = 0.119) when ≥1 ECG criteria of LAE were present. Conclusion Individual ECG criteria of LAE in hypertension are specific, but not sensitive, for identifying anatomical LAE, relative to CMR. LAE in hypertension should not be excluded on the basis of the ECG, particularly in obese subjects.

2 Clinical application of cardiovascular magnetic resonance in patients with MR-conditional devices: safety, feasibility and clinical impact

Background Implanted cardiac devices were previously considered unsuitable for CMR. With the development of MR-conditional devices, access to CMR has increased, despite concerns regarding image quality and diagnostic accuracy. We aimed to assess the clinical application of CMR in patients wearing MR-conditional devices. Materials and methods We retrospectively enrolled patients wearing MR-conditional devices undergoing a comprehensive CMR protocol (cine, early and late gadolinium enhancement, LGE) in a 1.5T scanner (June 2012–November 2015). Every sequence was analysed by two independent observers and scored according to the effect of artefacts on image quality and interpretation (no, minor and major artefacts). Inter-observer agreement was assessed per sequence and as overall judgement on scan quality and interpretation. Clinical impact of CMR was defined as a change in diagnosis and in management. All devices were interrogated before and after CMR. Abstract 2 Table 1 Cohen’s kappa for inter-observer agreement on image quality and interpretation per sequence and as overall judgement Cohen’s kappa p HASTE 0.378 0.001 Long Axis Cine 0.356 0.001 Short Axis Cine 0.532 <0.001 EGE 0.398 0.003 Long Axis LGE 0.284 0.005 Short Axis LGE 0.516 <0.001 Overall Judgement 0.454 <0.001 EGE, early gadolinium enhancement; LGE, late gadolinium enhancement. Results We enrolled 46 consecutive patients (28 male, mean age 56 ± 16 years) wearing MR-conditional pacemaker (22, 48%) and implantable loop recorder (24, 52%). All CMR scans were successfully completed and diagnostic: minor artefacts were recorded in 17 scans (37%), major artefacts in 7 (15%), and no artefacts in 22 (48%). Additional FLASH sequences were performed in 9 patients (20%) to overcome artefacts. Inter-observer agreement on image quality and interpretation was moderate, both overall (kappa 0.454, p < 0.0001) and per sequence, with the exception of long-axis LGE sequences, for which it was fair (kappa 0.284, p = 0.005) (Table 1). Cine sequences were most affected by artefacts, mainly in the mid-apical left ventricular anterior wall and anteroseptum (Figure 1). No change in device parameters was reported after the scan. CMR had a clinical impact in 26 patients (57%), determining a change in diagnosis in 16 (35%), in management in 5 (11%) and a change in both in 5 patients (11%). Conclusion With dedicated protocols and under strict monitoring of cardiac devices, CMR is safe and feasible in patients wearing MR-conditional devices, and it also has major clinical impact. Abstract 2 Figure 1 Left ventricular segmental analysis to assess artefacts interference. Sixteen-segment model showing that artefacts mostly affect the mid-apical left ventricular anterior and anteroseptal walls, on cine and post-contrast sequences, respectively

7 Clinical impact of cardiovascular magnetic resonance on the management of acutely hospitalised patients

Background Cardiac Magnetic Resonance (CMR) is invaluable for assessing ischaemic and non-ischaemic cardiomyopathies. However, evidence regarding the incremental impact of CMR in acutely hospitalised patients is scarce. We evaluated the impact of CMR on diagnosis and clinical decision-making in this cohort. Methods We evaluated 2481 consecutive scans (Jan 2014-Dec 2014) at a large tertiary cardiothoracic centre, identifying 283 patients referred for inpatient scans. Protocol included short axis-long axis cines, T2-weighted oedema sequences, early and late gadolinium enhancement (LGE) images. Definitions for “significant clinical impact” of CMR included change in pre-CMR diagnosis, influence on hospitalisation period, change in medication and on decision making for invasive medical procedures (CABG, angiography, ICD implantation). Results Of the 283 patients, 8 were excluded due to poor image quality, leaving 275 patients (66% male, mean age 59yrs), mean ejection fraction of 46% ± 19. Patients underwent CMR for further assessessment of ischaemic heart disease, cardiomyopathy or congenital heart disease. CMR demonstrated significant clinical impact on 68% of patients. This included a completely new diagnosis in 27% of patients, change in management in 31% and 10% of patients that had both a new diagnosis and change in management. CMR results promoted invasive procedures on 27%, avoided invasive procedures on 16%; and influenced on hospital discharge on 15% of the patients (Figure 1). 84% of the patients had echocardiography prior to CMR. CMR confirmed echo diagnosis in 11%, complemented echo findings with significant new information in 41% and changed the echo diagnosis in 30% of the cases. In a multivariable model that included clinical/imaging parameters, age and presence of LGE were the only independent predictors of “significant clinical impact” (LGE p-value .007, OR 2.782, CI 1.328–5.828) (Table 1). Conclusions CMR had significant impact in patient’s diagnosis and management in 68% of acutely hospitalised patients. Presence of LGE was the only independent predictor of significant clinical impact following CMR. Abstract 7 Table 1 Logistic Regression Variables in the Equation Sig. Odds ratio 95 Conf. Interval Lower Upper Sex .486 .766 .361 1.622 Age .028 1.026 1.003 1.050 Troponin .469 1.000 1.000 1.000 LVEF .945 .999 .972 1.027 iEDV .827 1.001 .989 1.014 RWMA .053 2.440 .987 6.033 LGE .007 2.782 1.328 5.828 Oedema .672 .904 .566 1.444 Variable(s): Sex, Age, Troponin, LVEF, iEDV, RWMA, LGE, Oedema. Abstract 7 Figure 1 Change in diagnosis after performing CMR in patients admitted with chest pain (A), shortness of breath (B) and arrhythmias-out of hospital cardiac arrest (C). (D) Overall significant clinical impact of CMR in change in management and new diagnosis

3 Clinical utility of cardiac MRI in young-middle aged patients with high-grade atrio-ventricular block

Background Atrio-ventricular (AV) block is a rare event in young-middle aged adults, often leading to pacemaker implantation without further investigation. We sought to assess the clinical utility of CMR in young-middle aged adults with high-grade AV block. Methods We retrospectively analysed the CMR registry to collect data on consecutive high-grade AV block patients (18–60yrs) referred for CMR (September 2012–November 2015). High-grade AVB was defined as Mobitz II 2nd degree or complete AVB. All patients underwent a transthoracic echocardiogram (TTE) and a comprehensive CMR protocol (cine and late gadolinium enhancement, LGE). A change in diagnosis was defined as a new diagnosis compared to a multi-parametric pre-CMR diagnosis (based on clinical, ECG and TTE data). Results We identified 34 patients (20 male, mean age 44 ± 12 years); 12 patients (34%) had II degree AVB and 22 (66%) complete AVB. Patients were referred to CMR for suspected ischaemic heart disease (IHD) in 4 patients (11%) and non ischaemic heart disease (NIHD) in 24 (71%); in 6 patients (18%) pre-CMR diagnosis was unclear. CMR showed IHD in 3 patients (9%) and NIHD in 11 patients (32%); a structurally normal heart was found in 18 patients (53%) and non-specific findings in 2 (6%) (Table 1) (Figure 1). LGE was found in 12 patients (34%), with predominant mid-wall pattern (58%). There was moderate agreement between CMR and TTE final diagnosis (Cohen’s kappa 0.435, p 0.001). CMR determined a change in diagnosis in 14 patients (40%). Abstract 3 Figure 1 CMR findings. Post-contrast four chamber long-axis (1A) and short-axis (1B) view showing structurally normal heart. Post-contrast four chamber long-axis (2A) and short-axis (2B) view showing epicardial LGE in the basal to mid-cavity lateral wall (white arrow) in a patient with myocarditis. Post-contrast four chamber long-axis (3A) and short-axis (3B) view showing transmural myocardial LGE in the basal to apical lateral wall in a patient with left ventricular non compaction Conclusions CMR was diagnostic in 94% of young-middle aged patients presenting with high grade AVB. As compared to a multi-parametric pre-CMR diagnosis, CMR led to a change in diagnosis in 40% of patients. Abstract 3 Table 1 CMR diagnosis CMR diagnosis n = 3 4 Ischaemic Heart Disease, n (%) 3 (9) Non-ischaemic Heart Disease, n (%) 11 (32) Structurally Normal Heart, n (%) 18 (53) Non-specific Findings, n (%) 2 (6)

5 Structural predictors of atrial fibrillation in hypertrophic cardiomyopathy using cardiac magnetic resonance imaging: Abstract 5 Table 1

Background Atrial fibrillation (AF) is the most common sustained arrhythmia in hypertrophic cardiomyopathy (HCM) and is associated with major adverse cardiovascular events. Cardiac magnetic resonance (CMR) with its superior tissue characterisation property is currently the imaging modality of choice for HCM. Aims To identify the structural predictors of AF in HCM using CMR. Methods 114 consecutive HCM patients were identified after reviewing approximately 3,100 CMR scans from our registry (Jan 2014 to Mar 2015). Comprehensive CMR protocol was used including cines, early and late gadolinium enhancement imaging. The diagnosis of HCM was based on left ventricular (LV) maximum wall thickness ≥15 mm (or 13–14 mm in the presence of familial history and/or ECG changes), in the absence of other cardiac/systemic disorders producing a similar degree of hypertrophy. Clinical notes were evaluated to identify a documented episode of AF. Univariate and multivariate logistic regression analyses were performed to determine the CMR imaging predictors of AF in HCM. Results The final study sample consisted of 104 patients with HCM with median age 60years (IQR = 54–70) and 70% male, (10 patients excluded due to uncertain/overlapping diagnosis). 70% had non-apical HCM; the remainder 30% apical HCM. 16% (n = 17) had a documented episode of atrial fibrillation. The univariate predictors of AF included left atrial volume and the ratio of left atrial volume to LV end systolic volume whereas in the multivariate model the ratio of left atrial volume to LV end systolic volume remained the only significant predictor (p = 0.034, OR = 2.236, CI = 1.06–4.70) (Table 1). Conclusion Our study suggests that the ratio of left atrial volume to LV end systolic volume is the best predictor of AF in HCM. The simple CMR derived ratio may have potential role for AF risk stratification in HCM. Abstract 5 Table 1 Predictors of AF in HCM Sig. OR 95% C. I. Sig. OR 95% C. I. Lower Upper Lower Upper Age 0.061 1.048 0.998 1.1 0.4 1.025 0.968 1.085 LA volume 0.001 1.043 1.022 1.064 0.056 0.905 0.817 1.003 LVEF 0.318 0.976 0.93 1.024 0.068 1.025 0.998 1.052 LA/ESV 0.001 1.81 1.286 2.547 0.034 2.236 1.063 4.702 LGE 0.349 2.743 0.332 22.648 0.855 1.298 0.079 21.262 Apical HCM 0.955 1.033 0.33 3.237 0.571 0.589 0.094 3.677 Max. Thickness 0.881 1.009 0.893 1.141 0.902 1.015 0.797 1.293 LV mass 0.229 0.988 0.968 1.008 0.417 0.986 0.952 1.02 Variable(s) entered on step 1: Age, Left atrium volume, LVEF- left ventricular ejection fraction, LA/ESV- left atrial volume/left ventricular end systolic volume, LGE- late gadolinium enhancement, Apical HCM, Max. Thickness.

4 Troponin positive acute coronary syndromes and unobstructed coronary arteries: Improved diagnostic and clinical impact by performing cardiovascular magnetic resonance early after presentation

Background 7–15% of acute coronary syndrome (ACS) patients have unobstructed coronary arteries. In these patients cardiac magnetic resonance (CMR) can identify different underlying aetiologies. Aim Evaluate the diagnostic and decision making implications of CMR timing (early versus late) in patients with ACS and unobstructed coronary arteries. Methods 204 consecutive patients (mean age 55yrs, 51% males) with troponin positive ACS and unobstructed coronary arteries, referred for a CMR between September 2011 and July 2014 were evaluated. Comprehensive CMR was performed “early” (≤2weeks from presentation) in 98 patients and “late” (>2weeks from presentation) in 106. “Significant clinical impact” was predefined as change in diagnosis/management. Propensity matching was performed between early and late CMR groups to minimise selection bias. Results Overall, a cause was found in 70% of patients. CMR had significant clinical impact in 66%, including change in the final diagnosis in 54%. (Figure 1) In a multivariable model (included clinical and imaging parameters), presence of late gadolinium enhancement (LGE) and age were the only independent predictors of “significant clinical impact” (LGE OR 2.3, p = 0.02) (Table 1). In a propensity score analysis, 58 pair of patients was matched for early and late CMR. The diagnostic pick up rate in the “early” group was significantly higher than in the “late” group (88% vs 50% p < 0.0001). Myocarditis (33%) was the most common diagnosis in the “early” group, whereas myocardial infarction (22%) in the “late” group. The clinical impact also improved significantly in the early group compared to the propensity score matched late group (76% vs 51%, p = 0.01). Conclusion CMR was able to establish final diagnosis in overall 70%. CMR made significant additive clinical impact on management and diagnosis in 66%, with LGE being the best independent predictor of impact. Moreover, the diagnostic value as well as the clinical impact of CMR was highest when performed early. Abstract 4 Figure 1 Showing the change in diagnosis following CMR Abstract 4 Table 1 Predictors of clinical impact – univariate and multivariate logistic regression analysis Univariate analysis Multivariate analysis Variables Sig. OR 95% CI Sig. OR 95% CI Lower Upper Lower Upper Age 0.008 1.024 1.006 1.041 0.002 1.035 1.013 1.058 Sex 0.77 1.091 0.609 1.954 0.604 0.831 0.413 1.673 Troponin 0.209 1 1 1.001 0.474 1 1 1.001 STEMI 0.224 1.63 0.742 3.577 0.966 0.981 0.412 2.338 iEDV 0.291 1.006 0.995 1.017 0.316 1.006 0.994 1.019 LVEF 0.597 0.995 0.975 1.015 0.847 1.002 0.98 1.025 RWMA 0.121 1.616 0.881 2.966 0.959 1.02 0.475 2.192 Oedema 0.078 1.765 0.938 3.323 0.527 1.298 0.579 2.912 LGE 0.004 2.393 1.318 4.345 0.017 2.411 1.17 4.968