Article Text

Download PDFPDF

1 Inflammatory cardiomyopathy in fabry disease
  1. JB Augusto1,
  2. S Nordin1,
  3. R Kozor2,
  4. R Vijapurapu3,
  5. K Knott1,
  6. R Hughes1,
  7. S Rosmini1,
  8. U Ramaswami4,
  9. T Geberhiwot3,
  10. RP Steeds3,
  11. S Baig3,
  12. D Hughes4,
  13. JC Moon1
  1. 1Barts Health NHS Trust, London, UK
  2. 2University of Sydney, Sydney Medical School, Sydney, Australia
  3. 3Queen Elizabeth Hospital Birmingham, Department of Cardiology, Birmingham, UK
  4. 4Royal Free Hospital, Lysosomal Storage Disorder Unit, London, UK


Background Fabry disease (FD) is an X-linked lysosomal storage disorder caused by mutations in α-galactosidase A. Cardiovascular magnetic resonance (CMR) has helped unveil the pathogenesis of Fabry cardiomyopathy: sphingolipid storage (low T1 mapping values), left ventricular hypertrophy (LVH) and myocardial fibrosis with late gadolinium enhancement (LGE) characteristically present in the basal inferolateral (BIFL) wall. Recent evidence has suggested that the LGE may be inflammation and oedema as part of this pathogenic process.

Purpose To assess the presence of inflammation in patients with FD using T2 mapping (for oedema/inflammation) supported by blood troponin levels (showing myocyte death and by inference inflammation).

Methods A multi-centre international study in gene positive FD patients using CMR and blood biomarkers. All participants underwent CMR at 1.5 T. Native T1 and T2 mapping were performed. The T1 mapping sequence was MOLLI with sampling scheme in seconds. LGE used a phase sensitive inversion recovery sequence. Global longitudinal 2D strain (GLS) values were obtained using feature tracking analysis. Blood high-sensitivity troponin T (hsTnT) was measured on the same day.

Results 100 FD patients (age 43.8±1.3 years, 42% male) were included. 45% had LVH, 35% LGE. Low T1 mapping (normal <943 ms) was found in 49% and 33% had high hsTnT values (normal <15 ng/L). Mean T2 mapping values were 52.6±0.6 ms in the BIFL wall and 49.5±0.3 ms in the remote myocardium/septum (p<0.001, normal <53 ms). T2 values in the BIFL wall were significantly higher among patients with LGE (58.2±6.1 ms vs 49.2±3.1 ms, p<0.001, figure 1). In a per-segment analysis of 1600 segments, higher T2 values correlated positively with percentage of LGE per segment (r=0,262, p<0.001), T1 values (r=0,205, p<0.001), maximum wall thickness (r=0,253, p<0.001) and GLS values (r=0,212, p<0.001). HsTnT values were higher among patients with LGE (median of 31 vs 3 ng/L in patients without LGE, p<0.001). There was a strong positive correlation between T2 values in the BIFL wall and ln(hsTnT) (r=0.776, p<0.001, figure 2). The strongest predictor of increased hsTnT in multivariate analysis (age, sex, LVH, septum T1, T2 in the BIFL, GLS, LGE) was T2 in the BIFL wall (β=0.4, p=0.001).

Abstract 1 Figure 1

Box-plot graph showing T2 values in remote and BIFL (LGE) areas

Abstract 1 Figure 2

Scatter-plot graph illustrating a positive correlation between T2 values in the BIFL wall and blood hsTnT

Abstract 1 Figure 3

Native T1, T2 and ECV mapping values measured in the septum and basal inferolateral wall. Typical BIFL wall scar is seen. Note the increased T2 mapping values matching the areas of LGE

Conclusions Cardiac involvement in FD goes beyond storage (low T1 values). When LGE is present, this is almost always associated with a high T2 and troponin elevation supporting FD as a chronic inflammatory cardiomyopathy. Initial reports of LGE being fibrosis are too simplistic – LGE in FD appears to have a significant chronic inflammation/oedema component.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.