P731 When cardiac imaging saves the day - a rare cause of embolic stroke

INTRODUCTION Hypereosinophilic syndrome (HES) is a rare disorder characterised by infiltration of tissues by eosinophils. Myocardial infiltration occurs in 50-60% of HES and leads to a condition called Loeffler’s endocarditis. Eosinophilic protein toxicity initiates endomyocardial necrosis. This is followed by a thrombotic stage and finally by endomyocardial fibrosis leading to a form of restrictive cardiomyopathy. Thrombosis is often located in the apical region of the ventricles and can result in stroke, which is the most devastating neurological consequence of hyperoesinophilia. We describe a case of a patient that presented with neurological symptoms and was found to have multiple embolic strokes secondary to Loeffler’s endocarditis. CASE A 57-year-old female presented to our institution with new onset confusion and reduced level of consciousness. Initial neurological assessment was consistent with encephalopathy. She had a 2-year history of eosinophilia that had been investigated by the haematology and rheumatology teams with no obvious aetiology identified. Initial haematological investigations showed a raised eosinophil count at 13mmol/L. Her cerebral MRI scan showed multiple embolic infarcts and therefore a transthoracic echo (TTE) was booked. This did not show any obvious intracardiac cause of emboli although the appearances of the LV apex were suspicious of thrombus. This was confirmed later, on contrast imaging (Fig 1). Staphylococcus aureus was grown in a single blood culture specimen raising the suspicion of infective endocarditis and a transoeosophageal echocardiogram (TOE) ruled out vegetations but again illustrated the apical filling defect despite absence of wall motion abnormalities (Fig 2). Finally, a cardiac MRI was arranged and this confirmed the diagnosis of Loeffler’s endocarditis with endomyocardial fibrosis and superimposed LV thrombus (Fig 3). She was treated with anticoagulation and steroids and her eosinophil count normalised before discharge. She remains well with no recurrence at two months post-event. DISCUSSION The diagnosis of Loeffler’s endocarditis depends on the presence high eosinophil count in combination with cardiac involvement on imaging. Transthoracic echocardiography can provide useful information such as apical thickening and thrombus in the left ventricle. As in our case, contrast TTE often provides further detail however contrast-enhanced cardiac MRI remains a key tool in the diagnosis and monitoring of this condition. It provides an assessment of systolic and diastolic function, tissue characterisation and typical features notably endomyocardial fibrosis and thrombosis on late enhancement imaging. CONCLUSIONS We presented a case where cardiac imaging has revealed the diagnosis in a patient presenting with systemic symptoms. We encourage clinicians to use multi-modality cardiac imaging as this has an invaluable role in the diagnostic process of complex patients. Abstract P731 Figure.

393Ischemic burden by vasodilator stress CMR predicts long-term all-cause death and the effect of revascularization in patients with known or suspected stable ischemic heart disease

Background In patients with stable ischemic heart disease (SIHD) the effect of revascularization on all-cause death (the most verifiable clinical event) is unknown. Objectives We explored the potential of the ischemic burden as derived from vasodilator stress cardiovascular magnetic resonance (CMR) to guide decision-making in this scenario. Methods In a large prospective multicenter registry, we recruited 6389 patients (mean age 65±11 years, 38% female) submitted to undergo vasodilator stress CMR for known or suspected SIHD. Baseline and CMR characteristics were prospectively recorded. The ischemic burden (at vasodilator stress first-pass perfusion imaging) and necrosis extent (at late enhancement imaging) were computed (17-segment model). The effect of CMR-related revascularization (within the following three months) on all-cause death (revised using the unified regional electronic health system registry) was explored. Results During a 5.75-year median follow-up, 717 (11.2%) all-cause deaths were documented. In multivariable analyses, more extensive ischemic burden (per 1-segment increase) independently related to all-cause death (1.05 [1.03–1.07], p<0.001). In 1034 patients (517 revascularized, 517 non-revascularized) strictly 1:1 matched for the independent predictors of outcome and of undergoing CMR-related revascularization (age, diabetes mellitus, male sex, LVEF, ischemic burden and necrosis extent), CMR-related revascularization did not significantly alter all-cause death rate (13.3% vs. 13.3%, p=0.54). Nevertheless, a potent interaction existed with the ischemic burden (p<0.001). CMR-related revascularization independently reduced the risk of all-cause death in 430 patients with ischemic burden >5 segments (9.3% vs. 16.3%, HR 0.56 [0.32–0.98], p=0.02) but it independently increased risk in 604 patients with ischemic burden ≤5 segments (16.2% vs. 11.3%, HR 1.59 [1.03–2.45], p=0.037). Figure 1. CMR-related revascularization Conclusions In patients with known or suspected stable ischemic heart disease the ischemic burden as derived from vasodilator stress CMR can be helpful to predict the effect of revascularization on long-term all-cause death. Acknowledgement/Funding Funded by “Instituto de Salud Carlos III”/FEDER (PIE15/00013, PI17/01836, and CIBERCV16/11/00486 grants) and Generalitat Valenciana (GV/2018/116).