The Impact of Longitudinal Strain on Haematological and Cardiac Response and Survival in Patients with Systemic AL Amyloidosis

Background: Cardiac involvement is the major determinant of prognosis in systemic AL amyloidosis. The extent is assessed by cardiac biomarker-based staging system using N-terminal pro-brain natriuretic peptide (NT-proBNP) and Troponin T. Longitudinal strain evaluates the global and regional function of the left ventricle (LV) and may be preferable to both LV ejection fraction and NT-proBNP, which is limited by its sensitivity to changes in fluid balance, in determining prognosis. This is the first report of a large cohort of uniformly treated prospectively followed patients assessing the utility of changes in longitudinal function by 2-D strain (LS%), impairment of which is a hallmark of amyloidosis.

Methods: 915 newly diagnosed patients seen at the UK National Amyloidosis Centre (February 2010 - August 2017) were studied. All patients underwent comprehensive assessments including echo-cardiogram at baseline and each follow up visit.

Results: 628/915 (68.6%) patients had cardiac involvement. Mayo stage I, II, IIIa and IIIb in 144 (15.7%), 302 (33.0%) 344 (37.6%) and 125 (13.7%) respectively. Impairment of LS% correlated significantly with increasing Mayo stage (p<0.0001 between LS% for each Mayo stage). At 12 months, only patients with complete haematological responses (CR) had significant improvement in LS% (overall p=0.04; regional baso-lateral p=0.007, and baso-septal p=0.007). The median overall survival (OS) of the whole cohort was 61 months; survival of Mayo stage I and II patients was not reached whilst OS in Mayo stage IIIa and IIIb patients was 30 and 4 months respectively. Patients with cardiac involvement were stratified into 3 baseline LS% groups (≤17%; 10.3-16.9%; and ≥10.2%) with poor baseline LS% being associated with shorter OS (p<0.0001). These groups predicted survival independently of Mayo stage. OS was superior in patients who achieved a minimum absolute improvement in LS% of 1.5% when analysed at either 12 (not reached vs. 72 months, p=0.008) and 24 (not reached vs. 80 months, p<0.0001) months from diagnosis. Patients achieving a LS% response (1.5%) improvement survived longer than those achieving a traditional cardiac response alone or no cardiac response at both 12 and 24 months (p<0.0001).

Conclusion: Longitudinal strain is an informative functional marker that is independent of Mayo staging in predicting outcomes in patients with cardiac AL amyloidosis which can be incorporated in prognostic staging for these patients. Improvement in LS% was observed in patients who achieved a CR, and a value of 1.5% was associated with superior outcomes over and above achieving a cardiac response by international consensus criteria. An absolute improvement in LS% should be considered a criterion for cardiac response in AL amyloidosis.