Improved prediction of electrical storm in patients with prior myocardial infarction and implantaImproved prediction of electrical storm in patients with prior myocardiable cardioverter defibrillator

Funding Acknowledgements

Type of funding sources: None.

Aims

To evaluate predictors of electrical storm (ES), including chronic total occlusion in an infarct-related coronary artery (infarct-related artery CTO, IRACTO), in a cohort of patients with prior myocardial infarction (MI) and implantable cardioverter-defibrillators (ICD).

Methods

Multicenter observational cohort study including 643 consecutive patients with prior MI and a first ICD implanted between 2005 and 2018 at three tertiary hospitals. All the patients included in the study had undergone a diagnostic coronary angiography before ICD implantation. The variable prior ventricular arrhythmias (VA+) was positive in patients with secondary prevention ICDs and in those with at least one appropriate ICD therapy after primary prevention implantation.

Results

During a median follow-up of 42 months 59 patients (9%) suffered ES. The presence of at least one IRACTO not revascularized (IRACTO-NR) was associated with a significantly higher cumulative incidence of ES (14.5% vs 4.8%, p<0.001). IRACTO-NR maintained a significant association with ES after adjustment for potential confounders (HR 2.3, p=0.005) and was an independent predictor of ES together with VA+ and LVEF. The best cut-off of LVEF to predict ES was ≤38%. A risk-prediction model based on IRACTO-NR, VA+ and LVEF≤38% identified three categories of ES risk (low, intermediate and high), with progressively increasing cumulative incidence of ES (2.2%, 9% and 20%).

Conclusion

In a cohort of patients with prior MI and ICD, IRACTO-NR is an independent predictor of ES. A new risk-prediction model allowed the identification of three categories of risk, with potentially important clinical implications.

P971 The echocardiographic phenotype in patients with cardiac amyloidosis and heart failure with preserved ejection fraction

BACKGROUND

New echocardiographic phenotypes of heart failure (HF) are focused on myocardial systolic involvement of the left ventricle (LV), either endocardial and/or transmural.

PURPOSE.

To study the pattern of myocardial involvement in patients (p) with HF with preserved left ventricular ejection fraction (pLVEF) and cardiac amyloidosis (CA).

METHODS.

Comparative study of 16 p with CA and HF with pLVEF, considering as cut point LVEF > 50%, in NYHA class ≥ II / IV, and a control group of 16 healthy people. Longitudinal Strain (LS) and Circumferential Strain (CS) were calculated using 2D speckle-tracking echocardiography, along with Mitral Annulus Plane Systolic Excursion (MAPSE) and Base-Apex distance (B-A). Also, the following indexes were calculated: Twist (apical rotation + basal rotation, º); Classic Torsion (TorC): (twist/B-A, º/cm); Torsion Index (Tor.I): (twist/MAPSE, º/cm) and Deformation Index (Def.I): (twist/LS, º).

We suggest the introduction of these dynamic torsion indexes as Tor.I and Def.I that include twist per unit of longitudinal systolic shortening of the LV instead of using TorC which is the normalisation of twist to the end-diastolic longitudinal diameter of the LV.

RESULTS

There were no differences of age between the groups (68.2 ± 11.5 vs 63.7 ± 2.8 years, p = 0.14). Global values of LS and CS were lower in p with CA indicating endocardial and transmural deterioration during systole, while TorC and Twist of the LV remained conserved in p with CA.

However, there is an increase of dynamic torsion parameters such as Tor.I and Def.I that show an increased Twist per unit of longitudinal shortening of the LV in the CA group (Table).

CONCLUSIONS

In p with CA and HF with pLVEF, the impairment of LS and CS indicates endocardial and transmural systolic dysfunction. In these conditions, LVEF would be preserved at the expense of a greater dynamic torsion of the LV.

Table LS (%) CS (%) Twist (º) TorC (º/cm) Tor.I (º/cm) Def.I (º/%) CA pLVEF (n = 16) -11.7 ± 4.2 17.2 ± 4.8 19.8 ± 8.3 2.5 ± 1.1 27.7 ± 13.5 -1.8 ± 0.9 Control Group (n = 15) -20.6 ± 2.5 22.7 ± 4.9 21.7 ± 6.1 2.7 ± 0.8 16.4 ± 4.7 -1.0 ± 0.3 p < 0.001 < 0.01 0.46 0.46 < 0.01 < 0.01 Dynamic Torsion Indexes and Classic Torion Parameters in pLVEF CA patients vs Control group.

P308 Changes in the echocardiographic phenotype during the evolution of cardiac amyloidosis from preserved to reduced left ventricle ejection fraction

Introduction

Cardiomyopaties like Cardiac Amyloidosis (CA), are an important cause of Heart Failure (HF). They can cause endocardial or transmural involvement. It is possible to characterize the kind of affectation thanks to different phenotypes identified by 2D speckle tracking echocardiography. Purpose: To study the pattern of myocardial involvement in patients (p) affected by CA and HF. Methods: Comparative study of 30 p with CA and HF, in NYHA class ≥II/IV, of which 16 had preseved left ventricle ejection fraction (pLVEF) and 14 had reduced LVEF (rLVEF), considering as cut point a LVEF > 50%. There was a control group (CG) of 16 healthy subjects. Twist, radial strain (RS), circunferential strain (CS) and longitudinal strain (LS) were determined using 2D speckle-tracking echocardiography, along with mitral annulus plane systolic excursion (MAPSE) and basal-apex distance (B-A). The following indexes were calculated: Twist (apical rotation + basal rotation, °); Torsion (twist/B-A, °/cm); Torsion Index (TorI: twist/MAPSE, °/cm), and Deformation Index (DefI:twist/LS,°). The last indexes are dynamic parameters that allow for a more realistic assessment of LV torsion, since they include longitudinal shortening measures such as MAPSE and LS, describing in a more complete and physiological way the global LV systolic movement.

Results

There were differences of age between the three gropus, being older the p with rLVEF and younger the ones in the CG (63,7 ± 2,8; 68,2 ± 11,5; y 73,9 ± 12,9 years respectively). LS and CS were lower in rLVEF group when compared with pLVEF group, as well as in pLVEF group compared with the CG. The p with pLVEF showed increased values of the dynamic torsion parameters (DefI and TorI), indicating a compensatory increase of LV twist that disappears in p with rLVEF. Twist and Torsion are significantly lower only in the rLVEF group (see table).

Conclusions

In both CA groups, LS and CS deterioration indicates endocardial and transmural involvement. The loss of compensation given by the increased LV twist, reflected by DefI and TorI, marks the transition to the deterioration of LVEF.

Results Table LVEF (%) LS (%) CS (%) TWIST (°) Torsion (°/cm) TorI (°/cm) DefI (°/%) Control Group (n = 15) 68.2 ± 6.3 -20.6 ± 2.5 -22.7 ± 4.9 21.7 ± 6.1 2.7± 0.8 16.4 ± 4.7 -1.0 ± 0.3 CA pLVEF (n = 16) 60,6 ± 5.4* -11.7 ± 4.2* -17.2 ± 4.8* 19.8 ± 8.3 2.5± 1.1 27.7 ±13.5* -1.8 ± 0.9* CA rLVEF (n = 14) 37.2 ± 8.8** -8.7 ± 3.2** -13.0 ± 3.4** 8.3 ± 5.6** 1.0 ± 1.7** 13.4 ± 9.6** -1.0 ± 0.7** *:p value <0,01 between CG and pLVEF group; **:p value <0,01 between pLVEF and rLVEF

P882 Heart murmur in a 74-year-old patient with a Dacron aortic graft admitted for sepsis after a laparoscopic left colectomy

Introduction

Aortic graft infections (AGI) can have catastrophic consequences with an operative mortality of nearly 50%. The majority of AGI are a result of bacterial exposure at the time of operation (surgical-site related), nearly three quarters a caused by the Staphylococcus organisms. Late onset infections are less common. The mechanism can be through hematogenous spread and bacterial invasion of the graft. Diagnosis is challenging, done by a combination of clinical, radiological and laboratory findings in which echocardiography plays an important role. Fundamental tenets of AGI management are removal of the infected device and adjunctive antimicrobial therapy.

Case

A 74-year-old man, who had undergone supracoronary ascending aortic replacement in 2015 for an aneurism, visited our hospital with fever of 39°, general malaise and abdominal pain for the last 3 days. He had been discharged one week ago after laparoscopic left colectomy due to descending colon neoplasia. Physical examination showed a systolic heart murmur loudest over the left-upper sternal border. Hematological findings included C-reactive protein (CRP) of 82 mg/l and white blood cell count of 16 700/μl. Blood culture was positive por Pseudomonas aeruginosa. Transthoracic echocardiography revealed a supravalvular pulmonary stenosis (figure 1, A) Transesophageal examination showed an extensive peritubular collection (figure 1, B) that extended into the main pulmonary artery, conditioning extrinsic compression and severe stenosis. No blood flow was observed inside the collection. A 6 mm long and filiform image was detected on the right coronary leaflet, causing a moderate aortic regurgitation. Chest CT revealed a low density area around the vascular graft (figure 1, C) and the positron emission tomography (PET)-CT (figure 1, D) showed increased glucidic metabolism in the aneurysmal sac, periaortic fat and the proximal and distal portion of the prosthesis. With the diagnosis of prosthetic vascular graft infection, the patient was referred to cardiac surgery. The surgical sample cultures (graft and mediastinal pus) were all positive for P. aeruginosa. The patient completed antibiotic therapy with ceftazidime and gentamicine.

Discussion

AGI is an extremely complex clinical challenge. Mortality is high, and diagnostic and treatment approaches are controversial. Cardiovascular imaging is one of the most important diagnostic tools in the diagnosis. An echocardiogram should be done in every patient to look for findings of endocarditis. CT is the most informative radiologic study and can be also very helpful in identifying characteristics and extension of AGI. A concurrent PET-CT study has significant potential in improving diagnosis of AGI and monitoring response to treatment.Nevertheless there is an unavoidable degree of subjective judgment in the interpretation of imaging findings making clinical suspicion and laboratory findings crucial in determining whether an AGI exists.

Abstract P882 Figure.