Article Text

Download PDFPDF

149 A novel internally validated risk prediction model for adverse cardiac outcome in fabry disease
  1. Christopher Orsborne1,
  2. Joshua Bradley2,
  3. Laura J Bonnett3,
  4. Luke A Pleva2,
  5. Josephine H Naish2,
  6. David G Clark3,
  7. Nik Abidin4,
  8. Peter Woolfson4,
  9. Gaetano Nucifora3,
  10. Matthias Schmitt2,
  11. Ana Jovanovic4,
  12. Christopher A Miller2,
  13. Anna Reid2
  1. 1The Univeristy of Manchester, MCMR, Manchester University Foundation Trust, Manchester, LAN M23 9LT, United Kingdom
  2. 2Division of Cardiovascular Sciences, School of Medical Sciences, Faculty of Biology, Medicine and
  3. 3Department of Health Data Science, University of Liverpool, a Member of Liverpool Health Partners
  4. 3Manchester University NHS Foundation Trust, Southmoor Road, Wythenshawe, Manchester, M23 9LT
  5. 4Salford Royal NHS Foundation Trust, Stott Lane, Salford, UK


Introduction The cardiac manifestations of Fabry disease are the leading cause of death, but risk stratification remains inadequate. Identifying patients who are at risk of adverse cardiac outcome may facilitate more evidence-based treatment guidance. Contemporary cardiovascular magnetic resonance imaging (CMR) biomarkers have become widely adopted but their prognostic value remains unclear. Our objective was to develop, internally validate, and evaluate the performance of, a prognostic model, including contemporary deep phenotyping, which can be used to generate individual risk estimates for adverse cardiac outcome in patients with Fabry disease.

Methods Longitudinal prospective cohort study of 200 consecutive patients with Fabry disease undergoing clinical CMR. Median follow-up 1,640 (987 – 2,293) days. Prognostic models were developed using Cox proportional hazards modelling. Outcome was a composite of adverse cardiac events. Model performance was evaluated.

Results The highest performing, internally validated, parsimonious multivariable model included age, native myocardial T1 dispersion (standard deviation of per voxel myocardial T1 relaxation times – Figure 1), and indexed left ventricular mass. Median optimism-adjusted c-statistic across 5 imputed model development datasets was 0.77. Model calibration was excellent across the full risk profile. A risk calculator, which provides 5-year estimated risk of adverse cardiac outcome for individual patients, including males and females, was generated (Figure 2 - survival free of the composite outcome according to predicted probability divided into three quantiles according to predicted probability).

Conclusion This study developed and internally validated a risk prediction model that accurately predicts 5-year risk of adverse cardiac outcome for individual patients with Fabry disease, including males and females, which could easily be integrated into clinical care. External validation is warranted.

Conflict of Interest CO has received travel support from Biomarin and Amicus therapeutics, as well as a research grant from Amicus therapeutics

  • Fabry disease
  • Cardiac Magnetic Resonance Imaging
  • Cardiomyopathy

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.