Abstract 16660: Precision Diagnosis of Hypertrophic Cardiomyopathy: A New Definition of Inappropriate Hypertrophy
Hunain Shiwani, Rhodri Davies, Constantin-Cristian Topriceanu, Anjali T Owens, Betty Raman, Raffaello Ditaranto, Joao Augusto, Camilla Torlasco, Matthew Webber, Benjamin Dowsing, Rebecca Hughes, Inês Miranda, Walter Witschey, Elizabeth Thompson, Silvia Castelletti, Lia Crotti, Luigi Lovato, Alberto Ponziani, Konstantinos Moschonas, George Joy, Iain Pierce, Peter Kellman, Alun Hughes, Elena Biagini, Saidi Mohiddin, Luis Lopes, Victor Ferrari, Harold Litt, Gabriella Captur, James Moon- Physiology (medical)
- Cardiology and Cardiovascular Medicine
Introduction: The definition of hypertrophic cardiomyopathy (HCM), unaltered for 50 years, requires unexplained left ventricular hypertrophy with a maximum wall thickness (MWT) ≥15mm in probands. However, this doesn’t consider age, sex and body-size which is inadequate for a precision therapy era.
Aim: To develop a personalised definition of inappropriate hypertrophy using cardiac MRI and evaluate potential care implications.
Methods: Healthy reference cardiac MRIs from the Framingham Heart Study, UK Biobank, and multiple healthy volunteer studies were analysed by a validated AI algorithm. Generalized additive mixed models accounting for age, sex, and body surface area (BSA) established a personalized hypertrophy threshold for MWT (>95% prediction interval) and conditional Z-scores. We assessed the discordance in HCM diagnosis between a “≥15mm” and “personalized hypertrophy” threshold applied to the UK Biobank and clinical HCM cohorts.
Results: In healthy subjects (n=5,255), 36% of MWT variation was explained by age, sex and BSA. In the UK Biobank (n=44,690), using ≥15mm, there is a substantial sex skew; 8% of males and 1% of females are classified as hypertrophic. With a personalized threshold, this reduces to 3% of males and 2% of females classified as hypertrophic. 17% of subjects have a predicted hypertrophy threshold of 15mm (with 46% predicted ≤ 14mm, 37% predicted ≥16mm). In clinical HCM cohorts (n=1,854) across 5 centres in 4 countries (UK, USA, Italy, Portugal), females had thinner hearts (17.7 vs 19.1mm; p<0.001) but more relative hypertrophy evidenced by greater deviation (Z-scores) from their predicted MWT (5.4 vs 5.1; p=0.05).
Conclusion: We propose a new, personalised definition for hypertrophy that mitigates for significant age, sex and BSA confounding inherent in a 15mm cut point. We have identified potential for under/over diagnosis in a population cohort and suboptimal risk stratification in smaller, younger and/or female HCM patients.