Echocardiographic assessment of left ventricular radial strain to differentiate cardiac amyloidosis from Fabry disease

Background

Left ventricular (LV) wall thickening is a typical echocardiographic finding in infiltrative cardiomyopathies like cardiac amyloidosis (CA) and Fabry disease (FD). The discrimination of both infiltrative diseases remains challenging by standard as well as 2D speckle tracking echocardiography (2DSTE)-based analysis of longitudinal LV strain patterns. Over the recent years, a constant development in image quality and data processing provided better possibilities to analyse layer specific myocardial deformation indices. With regards to FD, specific layer LV strain patterns may be useful to rule-in FD in patients with suspected infiltrative cardiomyopathy.

Purpose

The aim of the present study was to investigate differences and the diagnostic value of layer specific 2D STE-based radial LV strain indices in CA and FD.

Methods

Next to standard parameters of a comprehensive echocardiographic assessment (Vivid E9 or E95, GE Vingmed, Horton, Norway, with an M5S 1.5–4.5MHz transducer), we retrospectively analysed the transmural radial LV strain (GSradial), the subendocardial radial LV strain (GSendo), the subepicardial radial LV strain (GSepi), and the strain gradient (GSendo − GSepi) (EchoPAC software, GE) in FD patients and CA patient from the amyloidosis registry at our study site. A Receiver operating curve (ROC) analysis was used to assess the diagnostic value of the respective LV strain values and the layer-specific strain gradient to discriminate FD and CA.

Results

A total of 38 FD and 40 CA patients were included in our analyses. In patients with FD, GSepi showed a marked impairment. LV radial and layer strain were significantly reduced in CA compared to FD patients [GSradial −12.0 (−16.2 to −9.9) in CA vs. −17.7 (−20.5 to −14.6) in FD; p<0.001); GSendo (−20.6 (−27.0 to −15.7) in CA vs. −30.0 (−32.0 to −25.6) in FD; p<0.001); and GSepi (−7.4 (−8.9 to −4.8) in CA vs. −9.1 (−11.86 to −7.6) in FD; p<0.001)]. The gradient of GSendo and GSepi was significantly lower in patients with CA compared to FD (−14.0±5.6 in CA vs. −19.4±4.3 in FD respectively; p<0.001). GSendo held the highest diagnostic accuracy to discriminate CA and FD (area under the curve [AUC] 0.83, 95% confidence interval [CI] 0.73–0.92). The layer-specific strain gradient GSendo − Gsepi showed an AUC of 0.79 (CI 0.69 to 0.89).

Conclusion

Layer-specific strain analysis demonstrated significantly reduced strain values in CA patients compared to FD. The analysis of GSendo held a high diagnostic accuracy to discriminate FD and CA in patients. The integration of layer-specific LV strain indices into the diagnostic work-up may improve the management of patients with an unclear thick heart pathology in future.

Funding Acknowledgement

Type of funding sources: Private company. Main funding source(s): Alnylam Pharmaceuticals (Cambridge, MA, USA)

New echocardiographic approaches to differentiate cardiac amyloidosis and Fabry disease: the right heart and mitral valve thickness

Background

Infiltrative cardiomyopathies such as cardiac amyloidosis (CA) and Fabry disease (FD) are associated with high cardiovascular morbidity and mortality. Diagnosis is often challenging as CA and FD may present similar cardiac phenotypes on standard echocardiographic imaging, especially in early stages of disease.

Purpose

Our study focuses on right ventricular (RV) two-dimensional speckle tracking imaging and mitral valve thickness as diagnostic strategies to discriminate CA from FD. Here, we report preliminary data of our study registry of CA and FD patients.

Methods

We retrospectively screened patients with diagnosed CA or with genetically confirmed FD from our study registry. Global and free wall RV strain and mitral leaflet thickening in both groups were analysed using 2D speckle tracking echocardiography and linear measurements from a parastenal long axis view, respectively (Vivid E9 or E95, GE Vingmed, Horton, Norway, M5S 1.5–4.5MHz transducer). The diagnostic accuracy for discriminating CA and FD in both groups was evaluated using receiver operating characteristic (ROC) curve analysis.

Results

A total of 33 FD and 27 CA patients, including 17 patients with wild-type transthyretin-related (ATTR) CA, 8 patients with hereditary ATTR CA, one patient with light chain CA, and one patient with AA CA, were included. RV function was significantly reduced in CA compared to FD patients as measured by global longitudinal RV strain (−13.6±5.1 in CA vs. −18.4±3.9 in FD, p<0.001) and free wall longitudinal RV strain (−17.5±5.7 in CA vs. −21.3±4.4 in FD, p=0.005). Significantly lower longitudinal RV strain was observed in the basal to mid ventricular region in CA compared to FD patients, while RV apical strain was not significantly different between the two groups. RV global longitudinal strain had the highest diagnostic accuracy of strain analyses (area under the curve [AUC] 0.79, 95% confidence interval [CI] 0.66–0.93) to discriminate CA from FD. Additional measurement of the mitral valve in mid diastole in the parasternal long axis view revealed a significantly thickened leaflet in CA patients (4.6±1.2 mm in CA vs. 2.9±1.0 mm in FD, p<0.001) corresponding to an AUC of 0.84 (95% CI 0.72–0.96) in ROC analysis.

Conclusion

Global longitudinal RV strain was significantly reduced in CA patients while the mitral valve leaflets were significantly thickened compared to FD patients. Therefore, both parameters may be useful to discriminate CA from FD in the echocardiographic workup of patients with unclear left ventricular wall thickening.

Funding Acknowledgement

Type of funding sources: Other. Main funding source(s): Alnylam Pharmaceuticals (Cambridge, MA, USA)

Comparative analysis of phasic left atrial strain and left ventricular posterolateral strain to discriminate Fabry cardiomyopathy from other forms of left ventricular hypertrophy

Background

“Classic” echocardiographic signs of Fabry cardiomyopathy (FC), such as left ventricular hypertrophy (LVH) and posterolateral strain deficiency (PLSD) have a low diagnostic accuracy in clinical practice.

Purpose

Our aim was to evaluate the diagnostic accuracy of phasic left atrial strain impairment compared to PLSD to discriminate FC from other forms of LVH.

Methods

40 patients with LVH due to bioptically and genetically confirmed FC or with LVH due to other causes, defined by exclusion of storage diseases, such as Amyloidosis or FC, by myocardial biopsy, were retrospectively analysed. Standard echocardiographic views (Vivid E9, GE, Vingmed, Horton) were used to analyse left atrial (LA) reservoir, conduit, and contraction strain using 2D speckle tracking echocardiography (2DSTE; EchoPAC software, GE) as well as the PLSD, obtained by the mean of deformation values in basal posterior and lateral segments in a 17-segment model. Receiver operating characteristic (ROC) curve analysis and a logistic regression model were performed to assess the diagnostic accuracy of LA and LV strain impairment.

Results

FC was confirmed in 20 patients by genetic testing and myocardial biopsy. In the LVH group, 12 patients were classified to have hypertrophic cardiomyopathy, two had hypertensive heart disease, and six expressed the pattern of LV hypertrophy combined with borderline myocarditis. LV septum thickness (15.8mm±3.4 in FC; 17.9mm±4.3 in LVH) and left atrial volume index (LAVI) (36.7ml/m2±11.3 in FC; 45.7ml/m2±16.3 in LVH) as well as LVEF (54.2%± 9.8 in FC; 52.5%±7.7 in LVH,) were not statistically different between groups. LV filling parameters such as E/A (1.2±0.5 in FC; 1.2±0.7 in LVH) and E/e' (11.0±4.9 in FC; 13.2±5.3 in LVH) showed a slightly more advanced impairment in the LVH group.

Global and regional LV function was not different between groups (LVGLS −13.8±3.7% in FC and −12.8±3.7% in LVH; PLSD −10.7±5.2% in FC and −8.85±3.9% in LVH; p-value?). LA reservoir strain (LASr) and LA conduit strain (LAScd) were significantly impaired in FC compared to the LVH group (LASr 14,6±2.5% in FC and 26.3±8.5% in LVH, p<0.01; LAScd −5.9±2.6% in FC and −15.8±4.7% in LVH, p<0.01). In ROC analysis, LASr, with an area under the curve (AUC) of 0.81 (95% CI 0.64–0.97) and LAScd with an AUC of 0.85 (95% CI 0.71–0.99), respectively, showed the highest diagnostic accuracy to discriminate FC. PLSD, in contrast, held a low diagnostic accuracy with an AUC of only 0.47 (95% CI 0.27–0.68).

Conclusion

A substantially higher diagnostic accuracy could be shown for LASr and LAScd impairment in discriminating FD and other forms of LVH compared to PLSD. The echocardiographic assessment of phasic LA strain may help to identify FC in patients with unclear LVH.

Funding Acknowledgement

Type of funding sources: None. ROC analysisRepresentative examples