Association Between Left Ventricular Noncompaction and Vigorous Physical Activity

Left ventricular (LV) hypertrabeculation fulfilling noncompaction cardiomyopathy criteria has been detected in athletes. However, the association between LV noncompaction (LVNC) phenotype and vigorous physical activity (VPA) in the general population is disputed. The aim of this study was to assess the relationship between LVNC phenotype on cardiac magnetic resonance (CMR) imaging and accelerometer-measured physical activity (PA) in a cohort of middle-aged nonathlete participants in the PESA (Progression of Early Subclinical Atherosclerosis) study. In PESA participants (n = 4,184 subjects free of cardiovascular disease), PA was measured by waist-secured accelerometers. CMR was performed in 705 subjects (mean age 48 ± 4 years, 16% women). VPA was recorded as total minutes per week. The study population was divided into 6 groups: no VPA and 5 sex-specific quintiles of VPA rate (Q1 to Q5). The Petersen criterion for LVNC was evaluated in all subjects undergoing CMR. For participants meeting this criterion (noncompacted-to-compacted ratio ≥2.3), 3 more restrictive LVNC criteria were also evaluated (Jacquier, Grothoff, and Stacey). LVNC phenotype prevalence according to the Petersen criterion was significantly higher among participants in the highest VPA quintile (Q5 = 30.5%) than in participants with no VPA (14.2%). The Jacquier and Grothoff criteria were also more frequently fulfilled in participants in the highest VPA quintile (Jacquier Q5 = 27.4% vs. no VPA = 12.8% and Grothoff Q5 = 15.8% vs. no VPA = 7.1%). The prevalence of the systolic Stacey LVNC criterion was low (3.6%) and did not differ significantly between no VPA and Q5. In a community-based study, VPA was associated with a higher prevalence of CMR-detected LVNC phenotype according to diverse established criteria. The association between VPA and LVNC phenotype was independent of LV volumes. According to these data, vigorous recreational PA should be considered as a possible but not uncommon determinant of LV hypertrabeculation in asymptomatic subjects.