Search

Funding Acknowledgements Type of funding sources: None. Introduction Left bundle branch area pacing (LBBAP) is a rising technique that aims to preserve physiologic activation of the left ventricle. However, left bundle branch capture (LBBC) is not always achieved despite good progression of the lead into the interventricular septum, leading to what has been called deep septal pacing (DSP). In the latter, the conduction system is not captured, whose consequences on systolic function are not well known. The goal of the present work is to evaluate the repercussion on transthoracic echocardiogram (TTE) of capturing or not the left bundle branch. Methods Unicentric prospective study where patients in whom LBBAP was performed were consecutively included if: Previous TTE was available with a left ventricular ejection fraction (LVEF) > 40% A follow-up TTE performed >30 days after the implant was available Ventricular pacing was >40% of the time at the device follow-up LBBC was accepted if any of the following were present: Sudden change of QRS morphology during ventricular threshold test Left bundle potential was registered with an interval to left ventricular activation time (LVAT) equal to stimulus-LVAT time. T-test was used to compare means between continuous variables and chi-squared to test differences between groups in the categorical ones. Two-tailed p value Results 20 patients were included, 12 (60%) males, with a mean age of 74.4±10.6 years. Half of the patients (n=10) fulfilled LBBC criteria whereas the rest received DSP. Image 1 summarizes baseline characteristics and differences observed in the control TTE. After a median follow-up of 322 (interquartile range 114-325) days, LVEF increased in LBBC group whereas it decreased in DSP patients. Image 2 shows a slopegraph of the evolution of LVEF by type of stimulation. Indexed LVEDV increased in DSP group, but it was diminished in LBBC patients (p=0.03). Control TTE showed no significant differences in other parameters. Conclusions In non-selected patients with comparable pre-implant baseline characteristics, deep septal pacing was associated with a decrease in LVEF and an increase in LVEDV, while LBB capture showed better outcomes in these two parameters.

Funding Acknowledgements Type of funding sources: None. Introduction Left bundle branch pacing (LBBP) has traditionally been performed with fixed helix lumenless leads. The aim of this work is to compare feasibility and results with an extendable helix stylet-driven lead. Methods Unicenter and prospective registry of consecutive cases where LBBP was attempted using 2 different systems: -B: stylet-driven extendable helix lead guided with preshaped delivery sheath. -M: lumenless and fixed helix lead guided with preshaped delivery sheath. Anthropometric, echocardiographic, procedural variables and long-term performance were compared using T-test for continuous variables, and chi-squared for categorical ones. A two-tailed p value Results 134 patients were included. 44% (n=59) underwent the B technique and 56% (n=75) the M technique. Table 1 summarizes baseline, procedural and follow-up results. Implant failed in 4 (6.8%) patients in the B group: 2 because of repetitive entanglement of the helix with the endocardial surface, and 2 because of dislodgment of the lead after slitting the sheath. In the M group only 1 (1.3%) implant failed due to endocardial entanglement precluding correct penetration into the septum. Of the successfully implanted leads, 78.2% (n=43) in the B group vs. 70.3% (n=52) of the M group fulfilled strict LBB capture criteria (p=0.31). Baseline characteristics, fluoroscopy and procedure time showed no differences. After a median device follow-up time of 66 (interquartile range 36-184) days, slightly inferior capture threshold was seen in M group. 2 patients (3.8%) of the B group showed a threshold increase >2.0V, one of them needing intervention due to lack of capture. All M leads performed well during follow-up. Conclusions In consecutive patients with comparable baseline characteristics, both B and M technique are equally feasible, with similar LBBP success, procedural outcomes and long-term performance. Larger registries are needed to assess whether implant and long-term failure are also comparable.

Introduction and objective Cryoballoon pulmonary vein isolation (CB-VPI) is non-inferior to radiofrequency ablation in patients with atrial fibrillation. Although protocols aiming to reduce radiation exposure have been developed, most of them use intracardiac echocardiography (ICE) or electroanatomic mapping (EAM) systems, which significantly increase procedure costs or may not be widely available. Previous studies aiming at fluoroscopy reduction have been published, achieving median fluoroscopy times around 10 minutes. We developed a novel strategy to achieve Very Low rADIoscopic exposure to Minimize Ionizing Radiation during cryoablation of atrial fibrillation (VLADIMIR) (Table 1). The aim of this study was to assess its impact in acute procedural and clinical outcomes. Methods Consecutive patients undergoing CB-VPI were prospectively enrolled, treated with the VLADIMIR strategy and compared to a retrospective cohort of patients having undergone CB-VPI performed by the same operators with conventional fluoroscopy strategy. Primary end-points were fluoroscopy time and total dose area product (DAP). Secondary end-points were intraprocedural or early post-procedural complications. Results A total of 84 patients were enrolled. 50 patients underwent CB-VPI with conventional fluoroscopy strategy and in 34 patients the Vladimir strategy was applied. Vladimir group had significant lower median fluoroscopy time (2.98 min vs 20.65 min, p Conclusion The Vladimir protocol significantly reduces fluoroscopy time and radiation exposure during CB-VPI without affecting the rate of periprocedural complications. Funding Acknowledgement Type of funding sources: None. Table 1. The VLADIMIR strategyTable 2

Funding Acknowledgements Type of funding sources: None. Background Patients with left atrial appendage (LAA) thrombus are excluded for LAA occlusion (LAAO) in clinical trials. However, some of them may require this therapy due to repeated thromboembolisms despite accurate anticoagulation. Aim The aim of the study is to describe the prevalence of LAA thrombosis in patients submitted for LAAO in a high-volume center of LAAO procedures, explaining the employed technique and results in this scenario. Methods Single-center retrospective analysis including all consecutive patients referred for LAAO. Results LAA thrombus was detected by TEE in 8/76 patients referred for LAAO (10.5%). Five of them underwent LAAO despite the presence of thrombus due to clinical conditions. All the procedures were successfully performed with Amulet devices (Table). The following preventive measures were putting in place during the implants: avoidance of contrast injection into the LAA during the procedure (non-touch technique), TEE measure for device size election; and TEE device deployment guidance with special care to the the sheath orientation and depth enabling a slow and single implant of the device partially released in the LAA before complete implantation at the entrance of the LAA (Figure. A&B: Large LAA thrombus partially occupying the landing zone. C: Thrombus sealed by LAAO device. D&E: Non-touch technique: echo-guided progressive deployment of the LAAO device without thrombus shifting). No cerebral protection devices were used. TEE performed one month after the procedure ruled out the presence of residual thrombi or leaks in all cases. There were no events after a median follow-up of 17.1 months (range: 32.3-1.1) Conclusion LAAO in patient with LAA thrombus can be effectively and safely performed, in high-volume centers and with an accurate technique, under accurate TEE guidance. Abstract Figure. LAA thrombus and occlusion technique

Funding Acknowledgements Type of funding sources: None. Introduction Catheter ablation is recommended as first-line treatment in patients with atrioventricular nodal reentry tachycardias (AVNRT). However, the best therapeutic modality in patients with dual AV nodal physiology but non-inducible tachycardias in electrophysiological study (EPS) remains controversial, especially when no tachycardias have been documented. Our objective was to evaluate the results of empirical slow pathway ablation in patients showing dual AV nodal physiology but non-inducible AVNRT. Methods Multicenter, retrospective, observational registry of consecutive patients undergoing EPS due to clinical suspicion of paroxysmal supraventricular tachycardias (PSVT), but with no prior ECG documentation. Clinical, EPS and ablation (when performed) data were collected and andalyzed. Results 427 patients of 12 centers were included. Mean age was 46.3 ±16.1 and 297 (69.6%) were females. AVNRT was induced in 188 patients (typical in 181 cases, atypical in 7). Dual AV nodal physiology with or without single nodal echo beats, but with no sustained tachycardia and without evidence of accessory pathway was present in 68 patients. Ablation of the slow pathway was performed in 187/188 patients with AVNRT and in 30/68 patients with dual physiology. Among subjects with non-inducible tachycardia, ablation reduced significantly recurrences (39.5% in non-ablated vs. 16.7%; p = 0.04), with a level equivalent to those with ablated AVNRT(14.4% vs. 16.7%; p = 0.75). Procedure-related complications were similar in both groups: empirical ablation n = 1; 3.3% vs. induced tachycardia n = 6; 3.2% (p = 0.98). Conclusions In patients with high clinical suspicion of PSVT but non-documented and non-inducible arrhythmias, the presence of dual AV nodal physiology makes AVNRT a likely mechanism of the clinical tachycardia. Catheter ablation of slow pathway reduces the risk of recurrence to a level equivalent to those with inducible and ablated AVNRT. AVNRT (n = 188) Dual nodal physiology (n = 68) p value Age (years) 48.6 ± 16.3 41.9 ± 14.0

Funding Acknowledgements Type of funding sources: None. Introduction ECG documentation of paroxysmal supraventricular tachycardias (PSVT) may be challenging due to their short duration and usually abrupt end. As ablation represents first-line treatment of this pathologies, it is not uncommon that physicians refer patients to electrophysiological study (EPS) when symptoms are suggestive of PSVT even though no ECG documentation has been obtained. Our objective was to evaluate the results of ablation performed in patients with induced arrhythmias in EPS but no previous ECG documentation. Methods Retrospective and observational multicenter registry of consecutive patients undergoing EPS, followed or not by ablation, referred because of undocumented clinical suspicion of PSVT. We collected clinical, EPS and clinical follow-up data, along with data regarding ablation, when performed. Descriptive statistical analysis was carried out using Stata. Results 427 patients of 12 centers were included. Mean age was 46.3 ± 16.1 and 297 (69.6%) were females. Sustained arrhythmias were induced in most patients (n = 255; 59.7%) and ablation was performed in 274 (64.2%). 10 complications (2.3% of procedures) were reported: 3 transient AV block, 2 PR interval prolongation, 2 puncture-related hematoma, 2 painful sites of puncture and 1 catheter entrapment in mitral chordae. During a mean follow-up of 3.1 ± 2.0 years, 111 patients experienced recurrence of palpitations, mostly (n = 77; 69.4%) without ECG documentation. Clinical recurrence was more frequent in subjects who had not been ablated (41.8% vs 17.2% in ablated patients; risk ratio 2.44 (CI 95% 1.77-3.36); p Conclusions EPS in patients with high clinical suspicion of PSVT, followed by ablation when tachycardia is induced is effective in preventing future recurrences of palpitations. Early referral for EPS can, therefore, reduce visits to the emergency room, admissions and healthcare burden due to other low-yield, unnecessary tests. Type of recurrence ECG documentation 34(8.0%) ·Sinus rhythm 12(2.8%) ·Atrial tachycardia 6(1.4%) ·Typical atrioventricular nodal reentrant tachycardia 5(1.2%) ·Atrial fibrillation 5(1.2%) ·Premature ventricular beats 3(0.7%) ·Orthodromic atriventricular tachycardia 2(0.5%) Non documented recurrence 77(18.0%)

Funding Acknowledgements Type of funding sources: None. Introduction Paroxysmal supraventricular tachycardias (PSVT) are common arrhythmias and catheter ablation is considered its first-line treatment. However, the duration of the episodes frequently precludes ECG documentation. Thus, patients may not be referred for ablation until the tachycardia is documented, leading to recurrences, emergency room visits and often unnecessary tests or treatments. Our objective was to evaluate the results of electrophysiological study (EPS) followed or not by ablation in patients with suspected but not documented PSVT. Methods Multicenter, retrospective, observational registry of consecutive patients undergoing EPS due to clinical suspicion of PSVT, but with no prior ECG documentation. Collection of clinical and EPS data, along with data regarding ablation, when performed. Results 427 patients of 12 centers were included. Mean age was 46.3 ±16.1 and 297 (69.6%) were females. Most frequent symptoms consisted on sudden onset (n = 360; 84.9%) and abrupt end (n = 304; 72.0%), with median episode duration of 10 minutes (interquartile range 5-20 min). Sustained arrhythmias were induced in most patients (n = 255; 59.7%). Specific types are summarized in Table 1. Ablation was performed in 274 (64.2%) patients. A total of 10 complications (2.3% of procedures) were reported: 3 transient AV block, 2 PR interval prolongation, 2 puncture-related hematoma, 2 painful site of puncture and 1 catheter entrapment in mitral chordae. Conclusions Electrophysiological study in patients with palpitations highly suggestive of PSVT is an effective and safe diagnostic and therapeutic tool that may be considered as a first-choice even in the absence of documented tachycardia. Results of EP study Results of electrophysiological study Typical AVNRT 183 (42.9%) Orthodromic AVRT 38 (8.9%) Dual AV nodal physiology 30 (7.0%) 1 nodal echo beat 21 (4.9%) Atrial tachycardia 19 (4.5%) >1 nodal echo beat 17 (4.0%) Atrial fibrillation 7 (1.6%) Atypical AVNRT 7 (1.6%) Atrial flutter 1 (0.2%) No abnormal findings 104 (24.4%) AVNRT atrioventricular nodal reentran tachycardia AVRT atrioventricular reentrant tachycardia

Introduction Prosthetic valve thrombosis (PVT) is a rare but feared complication of cardiac valve replacement which usually represents a difficult diagnostic challenge. Although PVT is not frequent in bioprosthetic valves, the advent of transcatheter heart valves and the advances in computed tomography, allowing the detection of subclinical thrombosis, have yielded growing interest in that field, specially taking into account that there is no consensus on optimal antithrombotic approach in these patients. Clinical Case A 74-year-old male with prior cardiac antecedents of aortic valve replacement (April 2010) with a bovine pericardial bioprosthesis (Mitroflow 25, Sorin group Inc.), ischemic dilated cardiomyopathy with moderately depressed left ventricle ejection fraction (33%) and carrier of VDD pacemaker due to third degree atrioventricular block was admitted to advanced heart failure unit. During routine follow-up echocardiograms, prosthetic valve presented no signs of dysfunction except slightly increased gradients (image C): max 38 mmHg (normal 95%) and dyspnea NYHA class III the patient was referred for upgrade to cardiac resynchronization therapy (CRT). Computed tomography to asses epicardial venous anatomy prior to CRT implant was performed. In addition to venous distribution, it was described a repletion defect in aortic bioprosthetic valve suggestive of leaflet thrombosis (image A). To complete the study the patient underwent a transesophageal echocardiography (TOE) revealing a swallow’s nest shaped hypoechoic occupation of non-coronary and left aortic leaflets (image B), and 3D effective orifice area of 0,9 cm2. Oral anticoagulation was started in association to previously taken acetylsalicylic acid (ASA). Control TOE was performed 3 months after diagnosis showing almost complete resolution of thrombi. During the follow-up a CRT-D was implanted, with significant response in systolic performance, reaching a LVEF of 45%. Interestingly, despite the increase in anterograde aortic flow, progressive decrease of aortic gradients (max 24 mmHg, med 15 mmHg) until normalization was found (image D). Clinical benefit was also patent, being the patient in NYHA class I at the moment. Discusion Valve thrombosis could be difficult to diagnose in the presence of left ventricle dysfunction as gradients shall remain low despite an important compromise in valve motion. We present a case of incidental diagnosis of non-obstructive leaflet thrombosis that was managed conservatively with oral anticoagulation and ASA. The descent in transaortic gradients, moreover taking into account the response to CRT increasing LVEF, indicates that gradients slightly increased or in the upper limit of normality should raise suspicion in valve dysfunction in the presence of decreased LVEF. Abstract 1645 Figure.

Background Patients with mid-range ejection fraction heart failure (HFmrEF) are a new category defined in actual guidelines. The aim of this study was to establish the prognostic value of several parameters obtained by Cardiac Magnetic Resonance (CMR) and Transthoracic Echocardiography (TTE) in patients with HFmrEF. Methods Thirty patients, defined as HFmrEF by TTE and CMR, were included between 2012-2018. Patients with structural heart disease different from Left Ventricular (LV) dysfunction were excluded. Cine sequences in CMR (SSFP) were used to obtain atrial and ventricular volumes and mass. Myocardial fibrosis was quantified by late gadolinium enhancement. TTE was used to obtain anatomical and functional parameters as LV and Right Ventricular (RV) ejection-fraction, LV and RV global longitudinal strain, and RV free-wall longitudinal strain. The primary endpoint was the combination of all-cause death or heart failure admission. The median follow-up was 1.9 (0.5-3.3) years. Results Mean age was 59.3 ± 12.4 years, and 67.9% patients were male. The aetiology of LV dysfunction was mainly ischemic (n = 16, 53.3%). Results are shown in Table1. Patients who presented the primary endpoint had a lower RV ejection-fraction by CMR and a lower absolute value of RV free-wall longitudinal strain by TTE(Figure 1). Conclusions In HFmrEF patients, worse RV function (by CMR and TTE-Speckle Tracking) may be associated with a worse prognosis. Larger studies are needed to confirm this hypothesis. Table1 All-cause death or HF admission (n = 5; 16.7%) No all-cause death of HF admission (n = 25; 83.3%) p iRVEDV (ml/m2) by CMR 65.5 ± 13.5 66.2 ± 12.3 0.906 iRVESV (ml/m2) by CMR 30.4 ± 7.5 24.4 ± 6.2 0.065 iLVEDV (ml/m2) by CMR 85.8 ± 23.7 98 ± 19.5 0.225 iLVESV (ml/m2) by CMR 47.8 ± 15.3 54.9 ± 11.7 0.246 Indexed LA Volume (ml/m2) by CMR 42.8 (36.5 - 49) 48.4 (42 - 63.5) 0.386 LVEF (%) by CMR 44.9 ± 3.3 44 ± 2.6 0.506 RVEF (%) by CMR 52.2 ± 7.2 61.7 ± 7.2 0.012 RV-FAC (%) by TTE 43.4 ± 4.4 44.7 ±7.5 0.378 TAPSE by TTE 2.1 ± 0.3 2.5 ± 0.1 0.032 LV Longitudinal Global Strain by TTE -14.3 ± 3.3 -15.5 ± 4.9 0.663 RV Longitudinal Global Strain by TTE -11.3 (-13.2 - -7.0) -19.5 (-23.7 - -10) 0.089 RV Free-Wall Longitudinal Strain by TTE -11.5 (-14.2 - -8.2) -20 (-26 - -13.7) 0.043 HF: Heart Failure; CMR: Cardio Magnetic Resonance; LV: Left Ventricle; RV: Right Ventricle; iRVEDV: Indexed RV End-Diastolic Volume; iRVESV: Indexed RV End-Systolic Volume; iLVEDV: Indexed LV End-Diastolic Volume; iLVESV: Indexed LV End-Systolic Volume; LA: Left Atrium; LVEF: LV Ejection Fraction; RVEF: RV Ejection Fraction; RV-FAC: RV Fractional Area Change; TAPSE: Tricuspid Annular Plane Systolic Excursion Abstract P816 Figure 1

Introduction Aortic stenosis (AS) is the most prevalent valve disease. It involves increased left ventricle (LV) postcharge leading to LV hypertrophy. The aim of the study was to assess the burden of LV hypertrophy and its different patterns among a large cohort of patients with severe AS. Methods Observational, multicentre and prospective study of consecutive cases. Patients with severe AS defined as aortic valve area (AVA) 95 g/m2 in females or >115 g/m2 in males. LV geometry was assessed by relative wall thickness ratio, considering a cut-off value of 0.42. A p value Results A total of 805 patients with severe AS were included, 49.7% (n = 400) of them females and 50.3% (n = 405) males. LV indexed mass was available in 665 subjects, most of them (74.9%, n = 498) presenting LV hypertrophy. Females showed higher prevalence of LV hypertrophy than males (82.7% vs. 60.5%; p Conclusions LV hypertrophy affects most of patients with severe AS. LV remodelling is different between gender, with women developing higher grade of hypertrophy despite lesser AS severity. LV hypertrophy is associated with poor prognosis echocardiographic signs (increased PSAP and indexed LAV). Differences related to LV mass Normal LV mass LV hypertrophy P value Indexed AVA (cm/m2) 0.42 ± 0.01 0.41 ± 0.01 0.52 Mean gradient (mmHg) 42 ± 0.7 36 ± 1.1 Abstract 619 Figure. Sex related LV geometry patterns

Introduction and aim Aortic stenosis (AS) represents the main valve disease, thus, addressing its epidemiology and natural history becomes crucial. The aim of the present study is to shed light in the differences appearing with age in severe AS. Methods Observational, multicentre and prospective study of consecutive cases. Patients with severe AS, defined as aortic valve area (AVA) volume (iSV) A p-value Results A total of 636 patients were included, with slight predominance of females (53.9%; n = 343). No differences were found in AVA, LVEF, end-diastolic indexed values, LV mass or LV geometry between both age groups. MG and SVi were significantly lower in people over 85, who also presented more frequently with atrial fibrillation (AF) and significant mitral (MR) and tricuspid regurgitation (TR). Flow and gradient pattern vary significantly as shown in the graphic. Conclusions In our cohort, flow and gradient pattern in severe AS varies with age, despite no differences in LV geometry. This contrasts with previously assumed, relating LF and LG to more severe concentric hypertrophy. LF and LG can be explained by AF, significant MR and TR affecting more frequently those over 85. Age-related differences in severe AS Less than 85 y.o. ≥85 y.o. p-value AVA (cm2) 0.73 (± 0.16) 0.72 (±0.18) 0.46 LVEF (%) 65 (±7.9) 64 (±7.8) 0.21 MG (mmHg) 41 (±14.2) 37 (±13) 0.04 iSV (mL/m2) 40 (±11) 36 (±11) 0.01 iLVEDV (mL/m2) 47 (±16) 45 (±16) 0.07 iLVMass (g/m2) 126 (±36) 127 (±36) 0.66 Normal geometry (%) 3.9 2.5 0.64 Concentric remodelling (%) 24.3 24.4 Concentric hypertrophy (%) 64.2 66.4 Eccentric hypertrophy (%) 7.6 6.7 AF (%) 12.3 19.3 0.04 Significant MR (%) 11.3 18.7 0.02 Significant TR (%) 50.4 65.7 Abstract 616 Figure. Age-related flow and gradient pattern

A 85 years old woman with shortness of breath and leg edema with proggresive intensity last three weeks is admited in the cardiology department. As significant comorbidities, the patient has arterial hypertension (although last months all the medication that she needed was discontinued due to tendency toward low blood pressure) and paroxysmal atrial fibrillation. As previous surgical background, the patient only refers a carpal tunnel syndrome surgery (in both sides). In the physical exam she has a blood pressure of 110/60 Hg mm, lung crackles in the lower third of both lungs and a systolic murmur in the aortic area with a very low second sound. After being admitted in the cardiology department an echocardiography was done, showing severe left ventricle hypertrophy. As seen in the image (picture A), the myocardium, showed a characteristic granular ang bright pattern. It also showed a moderate to severe aortic stenosis. Although she was in atrial fibrillation during the echocardiography, the transmitral Doppler showed signs of significantly elevated filling pressures as for example, the presence of L waves. Finally, the exam was completed with an analysis of longitudinal deformation of the left ventricle, showing a clearly depressed global longitudinal strain, with a base-apex gradient (preserved at the apex, picture D). With these findings a Tc-PYP scintigraphy was performed, showing moderate and difuse uptake of the radioactive tracer in both ventricles (as shown in the image C). To complete the diagnosis, a cardiac magnetic resonance was performed showing a subendocardial ring pattern of delayed enhancement (image B). The diagnosis of transthyretin cardiac amyloidosis was done. Cardiac amyloidosis is clearly underdiagnosed entity, and sometimes differential diagnosis is really hard. Patients with cardiac amyloidosis are not always so "typical" as in this case. In fact, even a 5% of patients with history of aortic stenosis surgery may have also a cardiac amyloidosis. Maybe the use of myocardial deformation study and nuclear imaging can be useful solving this problem. Several studies have shown that nuclear imaging may be useful in the early diagnosis of this disease, even before that with echocardiography or cardiac magnetic resonance. Besides, new sequences of cardiac magnetic resonance may be very useful solving the problem of the differential diagnosis in left ventricle hypertrophy: The extracellular volume mapping. In cardiac amyloidosis hypertrophy is due to the deposit of amiloid in extracellular space. Because of that, the extracellular space is significantly increased in comparison with almost every other cause of left ventricle hypertrophy, making this sequence very useful in differential diagnosis. CONCLUSION Transthyretin cardiac amyloidosis is a rare and very under-diagnosed disease. However these problems with the advance of cardiac imaging and the development of new treatments could change very soon. Abstract P874 Figure. Image 1

Introduction A non-invasive diagnosis of cardiac amyloidosis is a challenge, especially in cases of atypical phenotypic presentation. Differential diagnosis includes hypertrophic cardiomyopathy (HCM), hypertensive cardiomyopathy, and other infiltrative disorders. Multimodality imaging is essential to make a final diagnosis. Case We present the case of a 65 years old woman, with a personal history of resistant arterial hypertension and mild hypertensive cardiomyopathy. She was diagnosed 6 years earlier with multiple myeloma, treated with chemotherapy and allogeneic hematopoietic stem cell transplant, presenting with several relapses and in a stable situation at the moment of our first consult. The patient was referred for heart failure in context of acquired community pneumonia one month earlier. Transthoracic echocardiography showed severe asymmetric left ventricle (LV) hypertrophy (Image A), systolic anterior motion of the mitral valve and diastolic dysfunction suggestive of HCM, not present in the previous examination. Strain imaging of the LV showed a typical amyloid infiltration pattern, with lower longitudinal strain values in the base compared to the apical segments (Image D). Further characterization of the myocardial tissue established the diagnosis along with performing a cardiac magnetic nuclear imaging (MRI). It showed the presence of inferior septum severe LV hypertrophy along with extensive patchy late gadolinium enhancement (LGE) of the lateral wall involving the endocardium (Image C), with normal LV contractility. There was no pleural effusion, but a small pericardial effusion was seen (Image B). With the suspicion of infiltrative heart disease, probably amyloid with an atypical LGE pattern, an oral mucosal biopsy was performed confirming amyloid diagnosis (Images E1,E2). Heart failure treatment was continued, but clinical evolution was poor with the deceasement of the patient 3 years after diagnosis. Discussion This case represents an example of the variety of imaging patterns we can see in cardiac amyloidosis. Despite the classical pattern of cardiac amyloidosis, with concentric LV hypertrophy, up to 8% of cases may present with asymmetrical LV hypertrophy, mimicking HCM. LGE extension and pattern can shows this variability as well: global transmural or subendocardial LGE is the most common, but focal patterns (up to 6% of cases) are described. The complexity of the diagnosis in these cases require a clinical and multimodality image approach. Abstract P726 Figure. Images of the case