Title: Prognostic Significance of Strain Doppler Imaging in Light-Chain Amyloidosis
Topic: Noninvasive Cardiology
Date Posted: 6/11/2010
Author(s): Koyama J, Falk RH.
Citation: JACC Cardiovasc Img 2010;3:333-342.
Clinical Trial: No
Study Question: What is the prognostic relevance of strain and strain rate imaging in patients with light-chain (AL) amyloidosis both with and without clinical evidence of congestive heart failure (CHF)?
Methods: Detailed echocardiography and Doppler, tissue velocity imaging (TVI) for strain, and strain rate were performed in 119 patients with biopsy-proven AL amyloidosis. Patients were excluded for a history of significant hypertension or valvular heart disease. TVI was performed in basal, mid, and apical segments in the anterior, lateral, septal, and inferior walls. Cardiac amyloid (CA) was considered present if left ventricular (LV) wall thickness was >12 mm.
Results: There was no echo evidence of CA in 49 patients (Group 1). Thirty-three patients had echo evidence of CA but no clinical evidence of CHF (Group 2), and the remaining 37 had both clinical CHF and CA (Group 3). Diastolic LV wall thickness was 10.2 ± 1.2, 14.5 ± 1.8, and 15.8 ± 2.7 mm in Groups 1, 2, and 3, respectively (p < 0.0001 for Groups 2/3 vs. 1). Fractional shortening was lower in Group 3 (28.2 ± 9.6%) compared to Groups 1 and 2 (42.7 ± 7.9 and 41.3 ± 8.6%, p < 0.0001) and left atrial dimension was larger in Group 3 than Group 1 (46.6 ± 6.9 vs. 41.5 ± 7.5 mm, p < 0.0001). Mitral E/A ratio was substantially higher in Group 3 (2.71 ± 1.15) than in either Group 1 (1.08 ± 0.51) or Group 2 (1.15 ± 0.72) (both comparisons p < 0.0001). With TVI, peak systolic velocity was similar in Groups 1 and 2 across all sites, but was lower in Group 3 at the base and mid LV. Average strain in basal segments was -18.3 ± 4.9 in Group 1, -13.9 ± 5.0 in Group 2, and -6.5 ± 4.3 in Group 3 (p < 0.001 compared to Groups 1 and 2), with similar trends seen for mid and apical strain. There were 39 deaths during a mean follow-up of 285 ± 136 days, 22 of which were either sudden or related to CHF. Stepwise regression revealed that mean basal strain, TVI-E-apex, and mitral valve deceleration time were independent predictors of death, but that the only independent predictor on multivariable analysis was basal systolic strain. Receiver operating characteristic analysis suggested basal strain of -13% was the optimal cut-off value for the whole group as well as for the two groups without clinical CHF (Groups 1 + 2). Within the CHF group (Group 3), basal strain value of <-4.6 predicted all-cause mortality.
Conclusions: Among patients with AL amyloidosis, mean basal strain is an independent and powerful predictor of all-cause mortality and is superior to standard two-dimensional echocardiographic and Doppler flow measurements.
Perspective: Measures of myocardial deformation including strain, strain rate, and torsion have been applied to patients with known or suspected CA. Previous studies have suggested that subclinical CA can be identified using these sophisticated measures. Previous studies have also suggested the prognostic relevance of diastolic dysfunction and other echocardiographic parameters in patients with documented cardiac amyloidosis. This study evaluated a consecutive series of 119 patients, none of whom had significant underlying hypertensive cardiovascular disease, atrial fibrillation, or valvular heart disease and identified basal strain using TDI as an independent predictor of both all-cause and cardiac mortality. Both tissue Doppler imaging and speckle tracking techniques can be used to determine strain and strain rate at multiple levels of the left ventricle. It should be emphasized that in this study, it was only the basal strain values that retained prognostic relevance. Absolute strain and strain rate values are location dependent within the left ventricle and will be platform and technique dependent as well. As such, the threshold identified as optimal in this study (-13%) may not translate to clinical practice using other techniques such as speckle tracking or even other ultrasound platforms. As sophisticated measures of myocardial deformation have clearly shown promise for detection of preclinical amyloid not otherwise identified by echocardiography and for prognostic purposes, the next logical extension of this type of study will be to identify changes in myocardial mechanics related to therapy designed to combat amyloidosis. William F. Armstrong, M.D., F.A.C.C.
|