Clonal hematopoiesis and left atrial remodeling in the young: a UK biobank cardiac MRI study
EP Europace Journal

Abstract
Clonal hematopoiesis of indeterminate potential (CHIP) has been associated with cardiovascular diseases through cardiac remodeling and impaired function. However, a large-scale evaluation of the association between CHIP and cardiac dysfunction, especially in the left atrium (LA), based on magnetic resonance imaging (MRI) is unknown.
From UK biobank, we included 41,665 participants with cardiac MRI and no history of cancer, cardiomyopathy, and valvular heart disease. Using the whole-exome sequencing data of the participants, we called somatic variants and identified CHIP mutations of variants allele fraction (VAF) ≥ 2.0% or 10.0%. The cardiac MRI parameters of LA – including volume index (LAVI) and ejection fraction (LAEF) – were compared based on the presence of CHIP mutations across different age groups (≤60 years and >60 years). The dilated LA and decreased LAEF was defined as LAVImax >40 mL/m2 and LAEF<50%, respectively. We defined LA remodeling either dilated LA or decreased LAEF. The occurrence of atrial fibrillation (AF), heart failure (HF), or stroke was compared according to the combination of CHIP status and LA remodeling.
Of the total (mean age 64.1±7.8years), CHIP mutations were present in 1,272 (3.1%) participants. In young individuals of ≤60 years (n=13,233; mean age 54.9±3.3 years), those with CHIP mutations presented significantly higher proportion of LA remodeling than those without CHIP mutations: adjusted OR (95% CI) as 1.362 (1.008-1.841), p=0.045. The association between LA remodeling and CHIP mutation became stronger when we applied VAF ≥10.0% to define CHIP mutation: adjusted OR (95% CI) as 1.765 (1.185-2.630), p=0.005. There was no significant difference in the LA remodeling between individuals with and without CHIP mutations in individuals of >60 years.
Across the age, the occurrence of the composite of AF, HF, or stroke was the highest among those with both CHIP mutation and LA remodeling than those without CHIP mutation and LA remodeling: adjusted HR (95% CI) as 1.980 (1.274-3.078), p=0.002 for VAF ≥2.0% and 2.471 (1.473-4.146), p=0.001 for VAF ≥10.0%.
CHIP is associated with LA remodeling, especially with a 1.4 to 2.6-fold increased risk in young individuals. The presence of CHIP mutation stratified the risk of the composite of incident AF, HF, or stroke, suggesting potential link between CHIP and atrial cardiomyopathy which is a key pathophysiology of AF, HF, or stroke.


