Poster Presentation Australian & New Zealand Obesity Society 2015 Annual Scientific Meeting

Which body composition measures best predict adolescent cardiovascular structure and function: Community-based cross-sectional study (#236)

Alanna N Hanvey 1 2 , Susan Clifford 1 2 , Fiona Mensah 1 2 , Melissa Wake 1 2 3
  1. The University of Melbourne, Parkville, VIC, Australia
  2. Murdoch Childrens Research Institute, Parkville, VIC, Australia
  3. The Royal Children's Hospital, Parkville, VIC, Australia

Background: Adolescent cardiovascular functional and structural phenotypes can now be directly measured in community studies, but not yet in the primary or secondary care setting. Determining if brief, non-invasive body composition measures predict these phenotypes could aid clinicians in identifying adolescents at risk of poor cardiovascular outcomes.

Objectives: To assess 1) patterns of association between body composition and cardiovascular structure and function, and 2) whether multiple body composition measures explain more variation in cardiovascular outcomes than single body composition measures.

Methods: Design: Cross-sectional outcomes of a longitudinal community-based study in Melbourne, Australia. Participants: 203 adolescents aged 14-17 years. Measures: Body composition (‘exposures’) – BMI z-score; waist z-score; fat and lean mass indices. Cardiovascular (‘outcomes’) – vascular function (systolic and diastolic blood pressure (S/DBP), augmentation index (AIx), pulse wave velocity (PWV)); macrovascular structure (carotid intima-media thickness (cIMT)); microvascular structure (retinal arteriolar-to-venule ratio (AVR)). Analyses: adjusted linear regression models describing associations between body composition and cardiovascular outcomes.

Results: For each one unit increase in BMI z-score; waist z-score; fat mass index; and lean mass index SBP increased by 2.5mmHg (CI: 0.9-4.0); 1.9mmHg (CI: 0.5-3.4); 0.5mmHg (CI: 0.0-1.1); and 1.4mmHg (CI: 0.4-2.3), respectively, and AIx increased by 2.9% (CI: 1.2-4.6); 2.0% (CI: 0.4-3.6); 1.0% (CI: 0.4-1.6); and 1.9% (CI: 0.9-2.9), respectively. No associations between body composition and DBP, PWV, cIMT or AVR were observed. Models with multiple body composition measures included explained 1-6% more variation in cardiovascular outcomes compared to models with single body composition measures.

Conclusions: No single or combined body composition measures clearly predicted underlying CV structure and function at age 14-17 years, and small associations of body composition with blood pressure and augmentation index did not translate into measurable structural cardiovascular change. This implies that obesity-associated CV disease can still be averted, at least to the mid-teen years.