Novel echocardiographic pixel intensity quantification method for differentiating transthyretin cardiac amyloidosis from light chain cardiac amyloidosis and other phenocopies.

AIMS: Differentiating cardiac amyloidosis (CA) subtypes is important considering the significantly different therapies for transthyretin (ATTR)-CA and light chain (AL)-CA. Therefore, an echocardiographic method to distinguish ATTR-CA from AL-CA would provide significant value. We assessed a novel echocardiographic pixel intensity method to quantify myocardial calcification to differentiate ATTR-CA from phenocopies of CA and from AL-CA, specifically. METHODS AND RESULTS: 167 patients with ATTR-CA (n=53), AL-CA (n=32), hypertrophic cardiomyopathy (n=37), and advanced chronic kidney disease (n=45) were retrospectively evaluated. The septal reflectivity ratio (SRR) was measured as the average pixel intensity of the visible anterior septal wall divided by the average pixel intensity of the visible posterior lateral wall. SRR and other myocardial strain-based echocardiographic measures were evaluated with receiver operator characteristic analysis to evaluate accuracy in distinguishing ATTR-CA from AL-CA and other forms of left ventricular hypertrophy. Mean septal reflectivity ratio (SRR) was significantly higher in the ATTR-CA cohort compared to the other cohorts (p <0.001). SRR demonstrated the largest AUC (0.91, p<0.0001) for distinguishing ATTR from all other cohorts and specifically for distinguishing ATTR-CA from AL-CA (AUC=0.90, p<0.0001, specificity 96%, sensitivity 63%). There was excellent inter- and intra-operator reproducibility with an ICC of 0.91 (p <0.001) and 0.89 (p <0.001), respectively. CONCLUSION: The SRR is a reproducible and robust parameter for differentiating ATTR-CA from other phenocopies of CA and specifically ATTR-CA from AL-CA.

Echocardiographic parameters associated with less reverse left ventricular remodeling after transcatheter aortic valve implant in subjects with prosthesis patient mismatch.

BACKGROUND: Transaortic valve implant (TAVI) is the treatment of choice for severe aortic stenosis (AS). Some patients develop prosthesis patient mismatch (PPM) after TAVI. It is challenging to determine which patients are at risk for clinical deterioration. METHODS: We retrospectively measured echocardiographic parameters of left ventricular (LV) morphology and function, prosthetic aortic valve effective orifice area (iEOA) and hemodynamics in 313 patients before and 1 year after TAVI. Our objective was to compare the change in echocardiographic parameters associated with left ventricular reverse modeling in subjects with and without PPM. Our secondary objective was to evaluate echo parameters associated with PPM and the relationship to patient functional status and survival post-TAVI. RESULTS: We found that 82 (26.2%) of subjects had moderate and 37 (11.8%) had severe PPM post-TAVI. There was less relative improvement in LVEF with PPM (1.9 ± 21.3% vs. 8.2 + 30.1%, p = .045). LV GLS also exhibited less relative improvement in those with PPM (13.4 + 34.1% vs. 30.9 + 73.3%, p = .012). NYHA functional class improved in 84.3% of subjects by one grade or more. Echocardiographic markers of PPM were worse in those without improvement in NYHA class (mean AT/ET was .29 vs. .27, p = .05; DVI was .46 vs. .51, p = .021; and iEOA was .8 cm/m2 vs. .9 cm/m2 , p = .025). There was no association with PPM and survival. CONCLUSIONS: There was no improvement in LVEF and less improvement in LV GLS in those with PPM post-TAVI. Echocardiographic markers of PPM were present in those with lack of improvement in NYHA functional class.

Predictive Value of Relative Apical Sparing of Longitudinal Strain on Echocardiography for Cardiac Amyloidosis.

Relative apical longitudinal sparing (RALS) on echocardiography has become an increasingly used tool to evaluate for cardiac amyloidosis (CA), but the predictive value of this finding remains unclear. This is a retrospective analysis at a single tertiary care center across 3 years. Patients were included if they had RALS, defined by strain ratio ≥2.0 on echocardiography, and sufficient laboratory, imaging, or histopathologic workup to indicate their likelihood of CA. Patients were stratified by their likelihood of CA, and contributions of other co-morbidities previously shown to be associated with RALS. Of the 220 patients who had adequate workup to determine their likelihood of having CA, 50 (22.7%) had confirmed CA, 35 (15.9%) had suspicious CA, 83 (37.7%) had unlikely CA, and 52 (23.7%) had ruled-out CA. The positive predictive value of RALS for CA was 38.6% for confirmed or suspicious CA. The remaining 61.4% of patients who were unlikely or ruled out for CA had other co-morbidities such as hypertension, chronic kidney disease, malignancy, or aortic stenosis, 17.0% of this group had none of these co-morbidities. In our tertiary care cohort of patients with RALS pattern on echocardiography, we found that fewer than half of patients with RALS were likely to have CA. Given the increasing use of strain technology, further studies are warranted to determine the optimal strategy for assessing CA in a patient with RALS.

Relationship between left atrial reservoir strain, volumes, and geometry: Insights from simple theoretical model.

BACKGROUND: Left atrial (LA) volume is related to LA reservoir strain (LASR ), but the relationship is not fully resolved. We sought to model the relationship between LA end-diastolic and end-systolic volumes (LAEDV and LAESV) and LASR based on a geometrical approach to exploit the relationship between LASR and volume. METHODS: Modeling the LA as a hemisphere with radius r, LASR was recognized to vary linearly with r and LA volume with r3 . Expanding this cubic relation as a Taylor series resulted in a simple linear equation: LAESV/LAEDV = 1 + 3 × LASR . To validate this, 52 transthoracic echocardiograms were analyzed from 18 patients who underwent transcatheter edge-to-edge repair (TEER) with MitraClip with serial assessment pre-procedure, 1 month post-clip, and 12 months post-TEER. Linear regression was performed to compare the geometric equation to a statistical model created by a line of best fit to relate LAESV/LAEDV to LASR . RESULTS: The statistical and geometric model both resulted in a strong correlation (r = .8, p < .001, respectively). The slope of the line in the statistical model was 3.3, which was statistically indistinguishable from the expected slope of 3 based on the geometric model (Figure 2A). Using the geometric model to compare the measured versus calculated LAESV/LAEDV also resulted in a strong correlation (r = .8, p < .001)(Figure 2B). CONCLUSION: We describe the relationship between LA volume and strain mathematically by considering the geometry of the LA. This model enhances our understanding of the interaction between atrial strain and volume. Further research is necessary to validate this using 3D atrial volumes in a broader cohort of subjects.

A Practical Approach to Echocardiographic Imaging in Patients With Hypertrophic Cardiomyopathy.

Hypertrophic cardiomyopathy (HCM) is frequently unrecognized or misdiagnosed. The recently published consensus recommendations from the American Society of Echocardiography provided recommendations for the utilization of multimodality imaging in the care of patients with HCM. This document provides an additional practical framework for optimal image and measurement acquisition and guidance on how to tailor the echocardiography examination for individuals with HCM. It also provides resources for physicians and sonographers to use to develop HCM imaging protocols.

Artificial Intelligence and Echocardiography: A Primer for Cardiac Sonographers.

Artificial intelligence (AI) is emerging as a key component in diagnostic medical imaging, including echocardiography. AI with deep learning has already been used with automated view labeling, measurements, and interpretation. As the development and use of AI in echocardiography increase, potential concerns may be raised by cardiac sonographers and the profession. This report, from a sonographer's perspective, focuses on defining AI, the basics of the technology, identifying some current applications of AI, and how the use of AI may improve patient care in the future.

Automated Echocardiographic Quantification of Left Ventricular Ejection Fraction Without Volume Measurements Using a Machine Learning Algorithm Mimicking a Human Expert.

BACKGROUND: Echocardiographic quantification of left ventricular (LV) ejection fraction (EF) relies on either manual or automated identification of endocardial boundaries followed by model-based calculation of end-systolic and end-diastolic LV volumes. Recent developments in artificial intelligence resulted in computer algorithms that allow near automated detection of endocardial boundaries and measurement of LV volumes and function. However, boundary identification is still prone to errors limiting accuracy in certain patients. We hypothesized that a fully automated machine learning algorithm could circumvent border detection and instead would estimate the degree of ventricular contraction, similar to a human expert trained on tens of thousands of images. METHODS: Machine learning algorithm was developed and trained to automatically estimate LVEF on a database of >50 000 echocardiographic studies, including multiple apical 2- and 4-chamber views (AutoEF, BayLabs). Testing was performed on an independent group of 99 patients, whose automated EF values were compared with reference values obtained by averaging measurements by 3 experts using conventional volume-based technique. Inter-technique agreement was assessed using linear regression and Bland-Altman analysis. Consistency was assessed by mean absolute deviation among automated estimates from different combinations of apical views. Finally, sensitivity and specificity of detecting of EF ≤35% were calculated. These metrics were compared side-by-side against the same reference standard to those obtained from conventional EF measurements by clinical readers. RESULTS: Automated estimation of LVEF was feasible in all 99 patients. AutoEF values showed high consistency (mean absolute deviation =2.9%) and excellent agreement with the reference values: r=0.95, bias=1.0%, limits of agreement =±11.8%, with sensitivity 0.90 and specificity 0.92 for detection of EF ≤35%. This was similar to clinicians' measurements: r=0.94, bias=1.4%, limits of agreement =±13.4%, sensitivity 0.93, specificity 0.87. CONCLUSIONS: Machine learning algorithm for volume-independent LVEF estimation is highly feasible and similar in accuracy to conventional volume-based measurements, when compared with reference values provided by an expert panel.