Left atrial structural and functional remodelling in Fabry disease and cardiac amyloidosis: A comparative analysis.

BACKGROUND: Fabry disease (FD) and transthyretin cardiac amyloidosis (TTR CA) are cardiomyopathies with hypertrophic phenotype that share several features, including left atrial (LA) enlargement and dysfunction, but direct comparative data are lacking. Aim of the present study was to perform a comparative analysis of LA remodelling between the two diseases. METHODS AND RESULTS: In this prospective study, a total of 114 patients (31 FD and 83 TTR CA) were included; all of them had left ventricular hypertrophy (LVH), defined as left ventricular (LV) wall thickness ≥ 12 mm. Despite similar degree of LVH, patients with TTR CA showed worse LV systolic and diastolic function. LA maximal volume index was not significantly different between the two groups (p = 0.084), while patients with TTR CA showed larger LA minimal volume index (p = 0.001). Moreover, all phases of LA mechanics were more impaired in the TTR CA group vs FD (reservoir: 6.9[4.2-15.5] vs 19.0[15.5-29.5], p < 0.001). After excluding patients with atrial fibrillation (AF), these differences remained clearly significant. In multivariable regression analyses, LA reservoir strain showed an independent correlation with TTR CA, controlling for demographic characteristics, AF and LV systolic and diastolic performance (p ≤ 0.001), whereas LV global longitudinal strain did not. Finally, among echocardiographic parameters, LA function demonstrated the highest accuracy in discriminating the two diseases. CONCLUSIONS: TTR CA is characterized by a more advanced LA structural and functional remodelling in comparison to patients with FD and similar degree of LVH. The association between TTR CA and LA dysfunction remains consistent after adjustment for potential confounders.

Performance and explainability of feature selection-boosted tree-based classifiers for COVID-19 detection.

In this paper, we evaluate the performance and analyze the explainability of machine learning models boosted by feature selection in predicting COVID-19-positive cases from self-reported information. In essence, this work describes a methodology to identify COVID-19 infections that considers the large amount of information collected by the University of Maryland Global COVID-19 Trends and Impact Survey (UMD-CTIS). More precisely, this methodology performs a feature selection stage based on the recursive feature elimination (RFE) method to reduce the number of input variables without compromising detection accuracy. A tree-based supervised machine learning model is then optimized with the selected features to detect COVID-19-active cases. In contrast to previous approaches that use a limited set of selected symptoms, the proposed approach builds the detection engine considering a broad range of features including self-reported symptoms, local community information, vaccination acceptance, and isolation measures, among others. To implement the methodology, three different supervised classifiers were used: random forests (RF), light gradient boosting (LGB), and extreme gradient boosting (XGB). Based on data collected from the UMD-CTIS, we evaluated the detection performance of the methodology for four countries (Brazil, Canada, Japan, and South Africa) and two periods (2020 and 2021). The proposed approach was assessed in terms of various quality metrics: F1-score, sensitivity, specificity, precision, receiver operating characteristic (ROC), and area under the ROC curve (AUC). This work also shows the normalized daily incidence curves obtained by the proposed approach for the four countries. Finally, we perform an explainability analysis using Shapley values and feature importance to determine the relevance of each feature and the corresponding contribution for each country and each country/year.

Valvular heart disease and cardiomyopathy: reappraisal of their interplay.

Cardiomyopathies and valvular heart diseases are typically considered distinct diagnostic categories with dedicated guidelines for their management. However, the interplay between these conditions is increasingly being recognized and they frequently coexist, as in the paradigmatic examples of dilated cardiomyopathy and hypertrophic cardiomyopathy, which are often complicated by the occurrence of mitral regurgitation. Moreover, cardiomyopathies and valvular heart diseases can have a shared aetiology because several genetic or acquired diseases can affect both the cardiac valves and the myocardium. In addition, the association between cardiomyopathies and valvular heart diseases has important prognostic and therapeutic implications. Therefore, a better understanding of their shared pathophysiological mechanisms, as well as of the prevalence and predisposing factors to their association, might lead to a different approach in the risk stratification and management of these diseases. In this Review, we discuss the different scenarios in which valvular heart diseases and cardiomyopathies coexist, highlighting the need for an improved classification and clustering of these diseases with potential repercussions in the clinical management and, particularly, personalized therapeutic approaches.

NN2Poly: A Polynomial Representation for Deep Feed-Forward Artificial Neural Networks.

Interpretability of neural networks (NNs) and their underlying theoretical behavior remain an open field of study even after the great success of their practical applications, particularly with the emergence of deep learning. In this work, NN2Poly is proposed: a theoretical approach to obtain an explicit polynomial model that provides an accurate representation of an already trained fully connected feed-forward artificial NN a multilayer perceptron (MLP). This approach extends a previous idea proposed in the literature, which was limited to single hidden layer networks, to work with arbitrarily deep MLPs in both regression and classification tasks. NN2Poly uses a Taylor expansion on the activation function, at each layer, and then applies several combinatorial properties to calculate the coefficients of the desired polynomials. Discussion is presented on the main computational challenges of this method, and the way to overcome them by imposing certain constraints during the training phase. Finally, simulation experiments as well as applications to real tabular datasets are presented to demonstrate the effectiveness of the proposed method.

Clinical staging of Anderson-Fabry cardiomyopathy: An operative proposal.

As a slowly progressive form of hypertrophic cardiomyopathy (HCM), Anderson-Fabry disease (FD) resembles the phenotype of the most common sarcomeric forms, although significant differences in presentation and long-term progression may help determine the correct diagnosis. A variety of electrocardiographic and imaging features of FD cardiomyopathy have been described at different times in the course of the disease, and considerable discrepancies remain regarding the assessment of disease severity by individual physicians. Therefore, we here propose a practical staging of FD cardiomyopathy, in hopes it may represent the standard for cardiac evaluation and facilitate communication between specialized FD centres and primary care physicians. We identified 4 main stages of FD cardiomyopathy of increasing severity, based on available evidence from clinical and imaging studies: non-hypertrophic, hypertrophic - pre-fibrotic, hypertrophic - fibrotic, and overt dysfunction. Each stage is described and discussed in detail, following the principle that speaking a common language is critical when managing such complex patients in a multi-disciplinary and sometimes multi-centre setting.

Prognostic Implications of the Extent of Cardiac Damage in Patients With Fabry Disease.

BACKGROUND: There is limited evidence on the risk stratification of cardiovascular outcomes in patients with Fabry disease (FD). OBJECTIVES: This study sought to classify FD patients into disease stages, based on the extent of the cardiac damage evaluated by echocardiography, and to assess their prognostic impact in a multicenter cohort. METHODS: Patients with FD from 5 Italian referral centers were categorized into 4 stages: stage 0, no cardiac involvement; stage 1, left ventricular (LV) hypertrophy (LV maximal wall thickness >12 mm); stage 2, left atrium (LA) enlargement (LA volume index >34 mL/m2); stage 3, ventricular impairment (LV ejection fraction <50% or E/e' ≥15 or TAPSE <17 mm). The study endpoint was the composite of all-cause death, hospitalization for heart failure, new-onset atrial fibrillation, major bradyarrhythmias or tachyarrhythmias, and ischemic stroke. RESULTS: A total of 314 patients were included. Among them, 174 (56%) were classified as stage 0, 41 (13%) as stage 1, 57 (18%) as stage 2 and 42 (13%) as stage 3. A progressive increase in the composite event rate at 8 years was observed with worsening stages of cardiac damage (log-rank P < 0.001). On multivariable Cox regression analysis, the staging was independently associated with the risk of cardiovascular events (HR: 2.086 per 1-stage increase; 95% CI: 1.487-2.927; P < 0.001). Notably, cardiac staging demonstrated a stronger and additive prognostic value, as compared with the degree of LV hypertrophy. CONCLUSIONS: In FD patients, a novel staging classification of cardiac damage, evaluated by echocardiography, is strongly associated with cardiovascular outcomes and may be helpful to refine risk stratification.

Consistent comparison of symptom-based methods for COVID-19 infection detection.

BACKGROUND: During the global pandemic crisis, various detection methods of COVID-19-positive cases based on self-reported information were introduced to provide quick diagnosis tools for effectively planning and managing healthcare resources. These methods typically identify positive cases based on a particular combination of symptoms, and they have been evaluated using different datasets. PURPOSE: This paper presents a comprehensive comparison of various COVID-19 detection methods based on self-reported information using the University of Maryland Global COVID-19 Trends and Impact Survey (UMD-CTIS), a large health surveillance platform, which was launched in partnership with Facebook. METHODS: Detection methods were implemented to identify COVID-19-positive cases among UMD-CTIS participants reporting at least one symptom and a recent antigen test result (positive or negative) for six countries and two periods. Multiple detection methods were implemented for three different categories: rule-based approaches, logistic regression techniques, and tree-based machine-learning models. These methods were evaluated using different metrics including F1-score, sensitivity, specificity, and precision. An explainability analysis has also been conducted to compare methods. RESULTS: Fifteen methods were evaluated for six countries and two periods. We identify the best method for each category: rule-based methods (F1-score: 51.48% - 71.11%), logistic regression techniques (F1-score: 39.91% - 71.13%), and tree-based machine learning models (F1-score: 45.07% - 73.72%). According to the explainability analysis, the relevance of the reported symptoms in COVID-19 detection varies between countries and years. However, there are two variables consistently relevant across approaches: stuffy or runny nose, and aches or muscle pain. CONCLUSIONS: Regarding the categories of detection methods, evaluating detection methods using homogeneous data across countries and years provides a solid and consistent comparison. An explainability analysis of a tree-based machine-learning model can assist in identifying infected individuals specifically based on their relevant symptoms. This study is limited by the self-report nature of data, which cannot replace clinical diagnosis.

Comparison of De-Kay repair versus De Vega suture for functional tricuspid regurgitation: a preliminary experience.

BACKGROUND: In patients undergoing mitral valve surgery, restrictive suture annuloplasty (De Vega) for less-than-severe functional tricuspid regurgitation has been proven to be safe and effective. The aim of this study is to determine whether the adjunct of the plication of the posterior tricuspid leaflet with the same running suture (bicuspidized De Vega or "De Kay") is equally safe and effective. METHODS: Single center, retrospective study on patients submitted to suture repair of the tricuspid valve during mitral valve surgery, with either conventional or De Kay, between January 2014 and December 2020. Comparison was based on degree of residual tricuspid valve regurgitation and right ventricular assessment at discharge. RESULTS: Over the course of the study period, 255 patients undergoing mitral valve surgery had a dilated (>40 mm or >20 mm/m2) tricuspid valve annulus, with less-than-severe tricuspid regurgitation. Conventional De Vega was employed in 166 patients (65.1%) and De Kay in the remaining 89 (34.9%). At discharge the adjunct of postero-septal commissure plication has similar outcomes to the classic De Vega repair. It seems to preserve right ventricular function. CONCLUSIONS: De Kay repair guarantees the same tricuspidal regurgitation reduction as compared with conventional De Vega early after surgery.

Inflammation across the spectrum of hypertrophic cardiac phenotypes.

The hypertrophic cardiomyopathy phenotype encompasses a heterogeneous spectrum of genetic and acquired diseases characterized by the presence of left ventricular hypertrophy in the absence of abnormal cardiac loading conditions. This "umbrella diagnosis" includes the "classic" hypertrophic cardiomyopathy (HCM), due to sarcomere protein gene mutations, and its phenocopies caused by intra- or extracellular deposits, such as Fabry disease (FD) and cardiac amyloidosis (CA). All these conditions share a wide phenotypic variability which results from the combination of genetic and environmental factors and whose pathogenic mediators are poorly understood so far. Accumulating evidence suggests that inflammation plays a critical role in a broad spectrum of cardiovascular conditions, including cardiomyopathies. Indeed, inflammation can trigger molecular pathways which contribute to cardiomyocyte hypertrophy and dysfunction, extracellular matrix accumulation, and microvascular dysfunction. Growing evidence suggests that systemic inflammation is a possible key pathophysiologic process potentially involved in the pathogenesis of cardiac disease progression, influencing the severity of the phenotype and clinical outcome, including heart failure. In this review, we summarize current knowledge regarding the prevalence, clinical significance, and potential therapeutic implications of inflammation in HCM and two of its most important phenocopies, FD and CA.

Left atrial strain analysis improves left ventricular filling pressures non-invasive estimation in the acute phase of Takotsubo syndrome.

AIMS: The aim of our study is to assess the ability of left atrial (LA) strain values to improve left ventricular and diastolic pressure (LVEDP) non-invasive estimation as compared with traditional echocardiographic indexes in the acute phase of Takotsubo syndrome (TTS) and to predict adverse in-hospital outcomes in this population. METHODS AND RESULTS: Consecutive TTS patients were prospectively enrolled. Left ventricular and diastolic pressure was measured at the time of catheterization. Transthoracic echocardiography was performed within 48 h from hospital admission. In-hospital complications (acute heart failure, death from any cause, and life-threatening arrhythmias) were collected. A total of 62 patients were analysed (72.2 ± 10.1 years, female 80%) and in-hospital complications occurred in 25 (40.3%). Left ventricular and diastolic pressure mean value was 24.53 ± 7.92 mmHg. Left atrial reservoir and pump strain values presented higher correlation with LVEDP (r -0.859, P < 0.001 and r -0.848, P < 0.001, respectively) in comparison with E/e ' ratio, left atrial volume index (LAVi), and tricuspid regurgitation (TR) peak velocity. In addition, at receiver-operating characteristic curve analysis, LA reservoir and pump strain resulted to be better predictors of LVEDP above the mean of our population [0.909 (95% CI 0.818-0.999, P < 0.001) and 0.889 (95% CI 0.789-0.988, P < 0.001)], respectively] as compared with E/e' ratio, LAVi, and TR peak velocity.Finally, LA reservoir strain resulted to be an independent predictor of worse in-hospital outcomes, together with LVEDP and left ventricular ejection fraction (all P < 0.001). CONCLUSION: In our study, lower LA reservoir and pump strain values were better predictors of LVEDP as compared with traditional echocardiographic indexes in the acute phase of TTS syndrome. Moreover, LA reservoir strain was an independent predictor of adverse in-hospital outcomes.

Using survey data to estimate the impact of the omicron variant on vaccine efficacy against COVID-19 infection.

Symptoms-based detection of SARS-CoV-2 infection is not a substitute for precise diagnostic tests but can provide insight into the likely level of infection in a given population. This study uses symptoms data collected in the Global COVID-19 Trends and Impact Surveys (UMD Global CTIS), and data on variants sequencing from GISAID. This work, conducted in January of 2022 during the emergence of the Omicron variant (subvariant BA.1), aims to improve the quality of infection detection from the available symptoms and to use the resulting estimates of infection levels to assess the changes in vaccine efficacy during a change of dominant variant; from the Delta dominant to the Omicron dominant period. Our approach produced a new symptoms-based classifier, Random Forest, that was compared to a ground-truth subset of cases with known diagnostic test status. This classifier was compared with other competing classifiers and shown to exhibit an increased performance with respect to the ground-truth data. Using the Random Forest classifier, and knowing the vaccination status of the subjects, we then proceeded to analyse the evolution of vaccine efficacy towards infection during different periods, geographies and dominant variants. In South Africa, where the first significant wave of Omicron occurred, a significant reduction of vaccine efficacy is observed from August-September 2021 to December 2021. For instance, the efficacy drops from 0.81 to 0.30 for those vaccinated with 2 doses (of Pfizer/BioNTech), and from 0.51 to 0.09 for those vaccinated with one dose (of Pfizer/BioNTech or Johnson & Johnson). We also extended the study to other countries in which Omicron has been detected, comparing the situation in October 2021 (before Omicron) with that of December 2021. While the reduction measured is smaller than in South Africa, we still found, for instance, an average drop in vaccine efficacy from 0.53 to 0.45 among those vaccinated with two doses. Moreover, we found a significant negative (Pearson) correlation of around - 0.6 between the measured prevalence of Omicron in several countries and the vaccine efficacy in those same countries. This prediction, in January of 2022, of the decreased vaccine efficacy towards Omicron is in line with the subsequent increase of Omicron infections in the first half of 2022.

Comparative analysis of right ventricular strain in Fabry cardiomyopathy and sarcomeric hypertrophic cardiomyopathy.

AIMS: To perform a comparative analysis of right ventricle (RV) myocardial mechanics, assessed by 2D speckle-tracking echocardiography (2D-STE), between patients with Fabry disease and patients with sarcomeric disease. METHODS AND RESULTS: Patients with Fabry cardiomyopathy (FC) (n = 28) were compared with patients with sarcomeric hypertrophic cardiomyopathy (HCM), matched for degree of left ventricle hypertrophy (LVH) and demographic characteristics (n = 112). In addition, patients with Fabry disease and no LVH [phenotype-negative carriers of pathogenic α-galactosidase gene mutations (GLA LVH-)] (n = 28) were compared with age and sex-matched carriers of sarcomeric gene mutations without LVH [Phenotype-negative carriers of pathogenic sarcomeric gene mutations (Sarc LVH-)] (n = 56). Standard echocardiography and 2D-STE were performed in all participants. Despite a subtle impairment of RV global longitudinal strain (RV-GLS) was common in both groups, patients with FC showed a more prominent reduction of RV free wall longitudinal strain (RV-FWS) and lower values of difference between RV-FWS and RV-GLS (ΔRV strain), in comparison to individuals with HCM (P < 0.001 and P = 0.002, respectively). RV-FWS and ΔRV strain demonstrated an independent and additive value in discriminating FC from HCM, over the presence of symmetric LVH, systolic anterior motion of the mitral valve and RV hypertrophy. Similar results were found in GLA LVH- patients: they had worse RV-FWS and lower values of ΔRV strain as compared to Sarc LVH- patients (both P < 0.001). CONCLUSION: Patients with FC show a specific pattern of RV myocardial mechanics, characterized by a larger impairment of RV-FWS and lower ΔRV strain in comparison to patients with HCM, which may be helpful in the differential diagnosis between these two diseases.

Right ventricular strain in Fabry disease: Prognostic implications.

INTRODUCTION: Left ventricular (LV) hypertrophy is the main feature of cardiac involvement in Anderson-Fabry disease (FD), but the right ventricle (RV) is also frequently affected. Previous studies failed to demonstrate an independent association between conventional parameters of RV performance and outcomes in FD. Nevertheless, if RV free wall strain (RV-FWS), assessed by 2D speckle tracking analysis, may provide a better prognostication is currently unknown. METHODS: We retrospectively evaluated the association between RV-FWS and the occurrence of cardiovascular events in a cohort of 56 patients with FD. The study endpoint comprises cardiovascular mortality, severe heart failure symptoms, new-onset atrial fibrillation and major arrhythmias requiring device implantation. RESULTS: Reduced RV-FWS, defined by values lower than 23%, was found in 25 (45%) patients. During a median follow-up of 47 months, 16 (29%) patients met the study endpoint. A ROC-curve analysis confirmed the threshold of reduced RV-FWS (<23%) as the best cut-off for predicting cardiovascular events, but with a lower power compared to left-sided parameters. On univariable Cox regression analysis, RV-FWS, expressed as continuous variable, was significantly associated with the study endpoint (HR: 0.795, 95% CI: 0.710-0.889, p < 0.001). However, RV-FWS did not retain a significant association with outcomes, after adjustment for LV global longitudinal strain or indexed left atrial volume (p = 0.340 and p = 0.289 respectively). CONCLUSIONS: RV-FWS was not independently associated with the occurrence of cardiovascular events in FD, confirming previous observations that prognosis is mainly driven by the severity of LV cardiomyopathy.

Impact of severe valvular heart disease in adult congenital heart disease patients.

Background: The clinical impact of valvular heart disease (VHD) in adult congenital heart disease (ACHD) patients is unascertained. Aim of our study was to assess the prevalence and clinical impact of severe VHD (S-VHD) in a real-world contemporary cohort of ACHD patients. Materials and methods: Consecutive patients followed-up at our ACHD Outpatient Clinic from September 2014 to February 2021 were enrolled. Clinical characteristics and echocardiographic data were prospectively entered into a digitalized medical records database. VHD at the first evaluation was assessed and graded according to VHD guidelines. Clinical data at follow-up were collected. The study endpoint was the occurrence of cardiac mortality and/or unplanned cardiac hospitalization during follow-up. Results: A total of 390 patients (median age 34 years, 49% males) were included and S-VHD was present in 101 (25.9%) patients. Over a median follow-up time of 26 months (IQR: 12-48), the study composite endpoint occurred in 76 patients (19.5%). The cumulative endpoint-free survival was significantly lower in patients with S-VHD vs. patients with non-severe VHD (Log rank p < 0.001). At multivariable analysis, age and atrial fibrillation at first visit (p = 0.029 and p = 0.006 respectively), lower %Sat O2, higher NYHA class (p = 0.005 for both), lower LVEF (p = 0.008), and S-VHD (p = 0.015) were independently associated to the study endpoint. The likelihood ratio test demonstrated that S-VHD added significant prognostic value (p = 0.017) to a multivariate model including age, severe CHD, atrial fibrillation, %Sat O2, NYHA, LVEF, and right ventricle systolic pressure > 45 mmHg. Conclusion: In ACHD patients, the presence of S-VHD is independently associated with the occurrence of cardiovascular mortality and hospitalization. The prognostic value of S-VHD is incremental above other established prognostic markers.

Myocardial infarction with non-obstructive coronary arteries in hypertrophic cardiomyopathy vs Fabry disease.

BACKGROUND: Little is known about prevalence and predictors of myocardial infarction with non-obstructive coronary arteries (MINOCA) in Fabry disease (FD) and hypertrophic cardiomyopathy (HCM). We assessed and compared the prevalence and predictors of MINOCA in a large cohort of HCM and FD patients. METHODS: In this multicenter, retrospective study we enrolled 2870 adult patients with HCM and 267 with FD. The only exclusion criterion was documented obstructive coronary artery disease. MINOCA was defined according to guidelines. For each patient we collected clinical, ECG and echocardiographic data recorded at initial evaluation. RESULTS: Overall, 36 patients had MINOCA during a follow-up period of 4.5 ± 11.2 years. MINOCA occurred in 16 patients with HCM (0.5%) and 20 patients with FD (7.5%; p < 0.001). The difference between the 2 groups was highly significant, also after adjustment for the main clinical, ECG and echocardiographic variables (OR 6.12; 95%CI 2.80-13.3; p < 0.001). In the FD population MINOCA occurred in 17 out of 96 patients with left ventricle hypertrophy (LVH, 17.7%) and in 3 out of 171 patients without LVH (1.7%; OR 12.0; 95%CI 3.43-42.3; p < 0.001). At multivariable analysis, voltage criteria for LVH at ECG (OR 7.3; 95%CI 1.93-27.7; p = 0.003) and maximal LV wall thickness at echocardiography (OR 1.15; 95%CI 1.05-1.27; p = 0.002) maintained an independent association with MINOCA. No major significant differences were found in clinical, ECG and echocardiographic findings between HCM patients with or without MINOCA. CONCLUSIONS: MINOCA was rare in HCM patients, and 6-fold more frequent in FD patients. MINOCA may be considered a red flag for FD and aid in the differential diagnosis from HCM.