Real world experience of therapeutic monitoring of medically treated prosthetic valve infective endocarditis by 18F-FDG-PET/CT

INTRODUCTION

18F-FDG-PET/CT is recommended to improve the diagnosis of prosthetic valve infective endocarditis (PVIE) and is a major criterion in the ESC-2015 classification. However, there is little evidence for its usefulness in the follow-up of medically treated PVIE patients.

METHODS

A monocentric retrospective analysis of patients hospitalized for PVIE between January 2013 and December 2019 who were not treated with surgery and who had at least two 18F-FDG-PET/CT examinations during their medical management.

RESULTS

Among 170 patients with PVIE, 117 were treated with antibiotic therapy but no surgery. Of these, 36 (31%) had at least two 18F-FDG-PET/CT examinations. At initial imaging, 28 patients had heterogeneous FDG uptake on their prosthetic valve and eight on their associated aortic graft. Hypermetabolism of spleen and bone marrow (HSBM) was observed in 18 and 19 patients, respectively. At the first follow-up 18F-FDG-PET/CT, 21 (58%) patients still had heterogeneous uptake, indicating persistent active endocarditis. HSBM was still present at the last follow-up imaging in four of the six patients with recurrent PVIE.

CONCLUSION

18F-FDG-PET/CT monitoring of medically treated patients with PVIE provides valuable additional information and prospective multicentric study should be conducted to assess its usefulness.

Redo isolated tricuspid valve surgery: prediction of in-hospital mortality using the TRI-SCORE

Background

Redo isolated tricuspid valve surgery (ITVS) is rarely performed. The TRI-SCORE reliably predicts in-hospital mortality after ITVS on native valve but has not been tested in the setting of redo interventions.

Purpose

We aimed to compare the predictive value of the TRI-SCORE to other surgical risk scores for redo ITVS.

Methods

Using a mandatory administrative database, we identified all consecutive adult patients who underwent a redo ITVS at 12 French tertiary centers between 2007 and 2017. Baseline characteristics and outcomes were collected from chart review and the TRI-SCORE, Logistic EuroSCORE, EuroSCORE II and STS were calculated.

Results

We identified 70 patients who underwent a redo ITVS (mean age 54±15 years, 63% female). Prior intervention was a repair in 51% and a replacement in 49%. A tricuspid valve replacement was performed in all patients. In-hospital mortality was 10%. The TRI-SCORE was the only risk score associated with in-hospital mortality (p=0.01). Area under the receiver operating characteristic curve for the TRI-SCORE was 0.83, much higher than with logistic EuroSCORE (0.58), EuroSCORE II (0.61) or STS (0.59). The table presents the observed and predicted values of in-hospital mortality according to TRI-SCORE categories.

Conclusion

The TRI-SCORE accurately predicted in-hospital mortality after redo isolated tricuspid valve surgery and may guide the clinical decision-making process especially as transcatheter therapies are emerging.

Funding Acknowledgement

Type of funding sources: None.

Adherence to guidelines of diagnostic and therapeutic practices for suspected infections of cardiac implantable electronic devices

Background

Despite guidelines describing the optimal diagnostic and therapeutic procedures of patients with suspected infection of cardiac implantable electronic device (CIED), the management of such patients is often challenging.

Purpose

The aim of this study was to describe our diagnostic and therapeutic practices for suspected CIED infection and to compare them to European Heart Rhythm Association (EHRA) guidelines.

Methods

Patients hospitalized in our tertiary care hospital for suspected CIED infection from 2014 to 2019 were retrospectively included. We applied the EHRA classification and compared diagnostic and therapeutic management to EHRA guidelines.

Results

Among 184 patients (mean age 72.3±12.4 years), 137 had a proven infection of the lead (either by TTE/TEE, by [18F]FDG PET/CT, or by positive culture of the lead) or an isolated pocket infection without proof of lead infection and 47 had no proof of infection of CIED.

According to EHRA classification, CIED infection was considered as definite in 145 patients, possible in 31 and excluded in 8.

Regarding recommended diagnostic procedures, blood cultures were performed in 90.8%, TTE in 97.8%, TEE in 85.9%, [18F]FDG PET/CT in 50.5% and imaging for embolisms in 78.3% of the patients.

Compared to therapeutic recommendations for the 145 cases of definite CIED infection, device removal was performed in 96 patients (66.2%), antibiotic therapy was prescribed in 130 (89.7%), with a duration equal or superior to the one recommended in 105 (72.4%) of the patients.

One-year survival rate was 66.3% and was better in patients with device removal than in patients without, both in the whole population (respectively 77.7% and 48.6%) and in the patients with definite CIED infection (80.0% vs 49.0%). After adjustement on age and Charlson's score, patients without device removal had a significantly higher risk of 1-year mortality (HR=3.40 [1.84–6.26], p=0.0001).

Conclusion

Suspicion of CIED infection are difficult situations. The use of theoretical classification helps to define the different types of cases and the resultant therapeutic approach.

In daily practice, our study shows that the strict adherence to these guidelines is not always possible. This study shows a relatively good use of diagnostic procedures. However, theoretical advice for medical therapy and removal procedures are not always properly followed. To improve patients' prognosis, infected device removal must be systematically discussed in multidisciplinary endocarditis team, weighing the benefit-risk ratio of the strict appliance of the guidelines.

Funding Acknowledgement

Type of funding sources: None. Diagnostic comparisonTherapeutic comparison

Antistaphylococcal penicillins vs. cefazolin in the treatment of methicillin-susceptible Staphylococcus aureus infective endocarditis: a quasi-experimental monocentre study

Whether cefazolin is as effective and safer than antistaphylococcal penicillins (ASPs) for the treatment of methicillin-susceptible Staphylococcus aureus (MSSA) infective endocarditis (IE) is still debated in the absence of a randomized controlled trial. In this quasi-experimental study, we aimed to assess the effectiveness and safety of these two treatments in MSSA-IE, using the ASPs nationwide shortage in April 2016 as a unique opportunity to overcome the indication bias associated with observational studies. In this single-centre study, we compared patients with Duke-Li definite MSSA-IE treated with ASPs from January 2015 to March 2016 versus those treated with cefazolin from April 2016 to December 2018, when ASPs were not available. Effectiveness outcome was 90-day all-cause mortality. Safety outcomes included significant decrease in GFR and significant increase in serum liver enzymes. Logrank test was used to compare survival rates. Of 73 patients with MSSA-IE, 35 and 38 were treated with ASPs and cefazolin, respectively. Baseline patients' characteristics (demography, native or prosthetic valve IE, clinical characteristics, cardiac and septic complications) were similar between groups. Ninety-day all-cause mortality was 28.6% and 21.1%, in patients treated with ASPs and cefazolin, respectively (logrank p = 0.5727). There was no difference between groups for incident renal or liver toxicity events: acute kidney injury 45.7% vs. 44.7% (p = 0.933), increased ALT 5.7% vs. 13.2% (p = 0.432), bilirubin increase 5.7% vs. 10.5% (p = 0.676), in ASPs vs. cefazolin groups, respectively. In this quasi-experimental, effectiveness and safety did not statistically differ between ASPs and cefazolin for MSSA-IE treatment.

Age-related characteristics of infective endocarditis: prospective data from the Euro-Endo registry

Introduction

The profile of infective endocarditis (IE) is continuously evolving over time, and this may probably be partly due to increasing age of the patients.

Methods

All data were collected from the EURO-ENDO registry, which is a one-year prospective international multicentre observational survey on patients with definite or possible IE included between 2016 and 2018. Subjects were stratified into 3 groups according to their age at index hospitalization.

Results

Among the 3113 patients included, 1670 patients (54%, Young group) were <65, 1068 (34%, Medium group) between 65 and 80 and 375 (12%, Old group) ≥80 years old. The most striking age-related differences were (old group vs others) (table 1) 1) the higher comorbidity burden and Charlson index; 2) the lower rate of embolic events on admission and under therapy; 3) the higher rate of Enterocci and digestive streptococci; 4) the lower rate of surgery during acute IE despite a theoretical indication; 5) the higher in-hospital and 1-year mortality. With regards to surgery, young and medium age were predictive of more frequent performance of surgery as compared to old age (Young: OR 4.33, 95% CI [3.09–6.06], Medium: 3.62, [2.57–5.10], p<0.001). In multivariable analysis, age per se was not predictive of in-hospital and 1yr FU mortality, but lack of surgical procedures when indicated (27% of the old group), was strongly predictive.

Conclusion

This is the largest contemporary registry showing the strong influence of age on the demographic, clinical, therapeutic, and prognostic profile of IE. Non-performance of surgical procedures when indicated is frequent in old patients and is a strong predictor of mortality while age per se is not. Endocarditis Teams should take these results into account when considering surgery in elderly patients.

Funding Acknowledgement

Type of funding source: Other. Main funding source(s): EORP grant

Determinants and prognostic significance of focal myocardial fibrosis by cardiac magnetic resonance imaging in patients with mitral valve prolapse

Background

Mitral valve prolapse (MVP) is a frequent disease that can be complicated by mitral regurgitation (MR), heart failure, arterial embolism, rhythm disorders and death.

Purpose

We sought to evaluate the correlates of late gadolinium enhancement (LGE), and the prognostic significance of ventricular size, function, and LGE by cardiac magnetic resonance (CMR) imaging on cardiovascular morbi-mortality in patients with MVP

Methods

We included 426 patients (54±15 years, 57% male) with MVP (trace to severe MR) between 2010 and 2019 who underwent a comprehensive echocardiography and CMR imaging. Gadolinium was injected in 411 (96%) patients. The main outcome endpoint was a composite endpoint of cardiovascular death, heart failure, new onset atrial fibrillation or arterial embolism, censored at the time of mitral valve intervention.

Results

Among the 411 patients, LGE+ was found in 118 (29%; 100 myocardial wall including 78 basal inferolateral midwall, 30 papillary muscles) and was more frequent with worse MR volume and LV remodeling. Correlates of LGE in multivariable analysis were LV mass index (OR 1.02, P=0.008), moderate-severe MR (OR 2.15, P=0.017) and coronary artery disease (OR 5.8, P=0.013). At 4 years, survival without cardiovascular events was decreased in patients with LGE+ (50.0±11.9 vs 73.5±6.3%, p<0.0001). In a stepwise multivariable analysis including classical predictors of outcome, grade of MR, LA volume index and the presence of LGE on CMR imaging (HR: 2.15 [1.15–4.02], P=0.017) were associated with impaired event-free survival.

Conclusion

CMR imaging provides additional information to echocardiography in the clinical work-up of MVP. Localized LV myocardial fibrosis is a predictor of cardiovascular event in MVP patients.

Funding Acknowledgement

Type of funding source: Public Institution(s). Main funding source(s): DGOS (PHRCI), Fondation Cœur et Recherche

P668 Echocardiographic description of mitral annular disjunction in mitral valve prolapse and implication in arrhythmic risk stratification

Introduction

Mitral annular disjunction (MAD) is an anatomical variation of the mitral annulus, characterized by an atrial displacement of the leaflet’s hinge points. It is associated with severe ventricular arrhythmias (VA) in mitral valve prolapse (MVP).

Purpose

The aim of this study was to assess MAD in MVP by echocardiography, analyze the reproducibility of measurements and evaluate its importance for arrhythmic risk stratification along with strain analysis of myocardial deformation.

Methods

Two hundred and sixty patients with MVP were included. MAD was evaluated and measured by two observers in the parasternal long axis and in the apical views. Myocardial longitudinal strain was analyzed by speckle-tracking.

Results

Ninety four patients (36.2%) of MVP patients presented MAD. These patients were younger (53.7 ± 15.1 vs 58.4 ± 17.6, p = 0.033) with higher rate of atypical chest pain (21.3% vs 11.5%, p = 0,041) and bileaflet prolapse (50.5% vs 32.3%, p = 0.004). Para-sternal long-axis view was the incidence of choice to detect MAD with a moderate inter-observer concordance (Kappa of 0.55), good correlation (r = 0.69, p < 0.01) and inter-class correlation coefficient (0.82; 0.67 – 0.90). Twenty patients (7.7%) had a history of severe VA. Among them, no difference was noted in terms of presence (35% vs 36.3%, p = 0.911) or length of MAD (11.1 ± 2.5 vs 11.2 ± 3.1, p = 0.937). However, deformation analysis showed reduced global longitudinal strain (18.6 ± 3.1 vs 21.3 ± 3.3%, p = 0.001) and higher mechanical dispersion values (46 ± 13 vs 37.4 ± 12.9 ms, p = 0.002)in comparison to the rest of the MVP population.

Conclusion

No significant association was found between severe VA and the presence or severity of MAD in MVP patients. Increased mechanical dispersion and reduced global longitudinal strain may be helpful for arrhythmic risk stratification.

Abstract P668 Figure. Comparison of MD and GLS

Comprehensive geriatric assessment in older patients suffering from infective endocarditis. A prospective multicentric cohort study

OBJECTIVES

The aim was to describe the impact of infective endocarditis (IE) on functional, cognitive and nutritional statuses, and to estimate the influence of these parameters on surgical management and mortality.

METHOD

This was a prospective study over 13 months in 14 French hospitals, including patients ≥75 years of age with definite or possible IE. A comprehensive geriatric assessment (CGA) was performed during the first week of hospitalization, including a retrospective estimation of functional status 2 months before hospitalization, and 3 months after.

RESULTS

A total of 120 patients were included (mean age 83.1 ± 5.0 (75-101) years). IE was associated with a dramatic impairment of functional status between 2 months prior hospitalization and the first geriatric evaluation (90.8% able to walk vs. 35.5% (p < 0.0001), ADL (Activities in Daily Living) 5.0 ± 1.7 vs. 3.1 ± 2.1 (p < 0.0001)). The 19 operated patients (15.8%) had less comorbidities (cumulative illness rating scale geriatric 10.8 ± 8.2 vs. 15.3 ± 7.1 (p 0.0176)), better functional (ADL 5.9 ± 0.4 vs. 4.9 ± 1.8 (p 0.0171) and nutritional (mini nutritional assessment 20.4 ± 5.0 vs. 17.3 ± 6.2 (p 0.0501)) statuses than non-operated patients. Among all infectious, cardiac and geriatric parameters, body mass index (HR 0.9, range 0.8-1, p 0.05) and ADL at the time of the first evaluation (HR 0.7, range 0.6-0.9, p 0.002) were the sole independent predictors of the 3-month (32.5%) and 1-year mortality (42.5%). Three months later, the 57 assessed patients only partially recovered their ADL (3.7 ± 1.9 vs. 5.3 ± 1.4 2 months prior hospitalization and 4.6 ± 1.9 at the first CGA; p < 0.0001).

CONCLUSION

Functional and nutritional abilities are crucial components that can be accurately explored through a CGA when managing IE in oldest patients.