An Investigation of the Relationship between Indices of Body Composition and Cardiovascular Risk Factors in Adult Females

Abstract

Background: According to the World Health Organisation [1] obesity is now a “global epidemic”, ranking as the fifth most common cause of death worldwide. Obesity prevalence has more than doubled over the past two decades [2], with particularly high levels in Scotland [3]. Obesity shows strong associations with cardiovascular disease (CVD), which is the largest single cause of death in the UK [4]; accounting for one in three deaths. Currently NICE [5] recommend using body mass index (BMI) and waist circumference (WC) for obesity assessment. Recently Krakauer and Krakauer [6] proposed the novel “a body shape index” (ABSI) for better predicting mortality hazard. This study aimed to investigate the relationships between BMI, WC, percentage body fat (%BF), ABSI and various cardiovascular risk factors in adult females. Methods: The study was granted university ethical approval had an observational cross-sectional design and recruited through convenience sampling. International Society for the Advancement of Kinanthropometry methodologies were used to measure height, weight and WC. Single frequency bioelectrical impedance analysis enabled estimation of %BF. BMI (kg/m2) was calculated by dividing weight (kg) by height squared (m2). ABSI (m11/6 kg-2/3) was calculated by dividing WC (m) by BMI2/3 (kg/m2) height½(m) using an online calculator. Physical activity levels (PAL) and sitting time were estimated using the International Physical Activity Questionnaire, and self-reports of alcohol intake and alcoholic binges were also obtained. Vascular health was determined via: blood pressure (BP); carotid-femoral pulse wave velocity (PWV) and the augmentation index (AIx) using a Vicorder™ device. SPSS v.19 was used to determine Pearson product-moment correlation coefficients for normally distributed data (WC, ABSI, sitting time, systolic BP, diastolic BP, mean BP) and Spearman’s correlation coefficients for all other data. Results: 29 healthy females aged 27±8years were recruited from Queen Margaret University. The average BMI was 23.3±4.2kg/m2, %BF 23.4±7.4%, WC 73.8±8.7cm and ABSI 0.07±0.004m11/6kg-2/3. ABSI correlated weakly (r=0.1-0.3) and non-significantly (p>0.05) with all CVD risk factors. Established body composition indices revealed significant moderate-strength correlations (r=0.36-0.67), including: alcohol intake with BMI (p=0.04, r2=0.15), %BF (p=0.04, r2=0.14) and WC (p=0.02, r2=0.20); alcoholic binges with BMI (p=0.03, r2=0.16) and WC (p=0.01, r2=0.24); sitting time with BMI (p=0.02, r2=0.19); systolic BP with BMI (p=0.03, r2=0.17); diastolic BP with WC (p=0.02, r2=0.20); and mean BP with both BMI (p=0.04, r2=0.15) and WC (p=0.02, r2=0.19).  Discussion: In this small-scale study no evidence of a significant correlation was found between ABSI and the CVD risk factors investigated. Furthermore, several limitations were apparent: The sample was small and consisted of a particularly young healthy population, thus suggesting insufficient external validity and it is debatable whether all participants fully complied with the pre-test protocol, which may have influenced the results. Conclusion: Findings corroborate current recommendations for using established indices of body composition, particularly BMI and/or WC for identifying early risks of CVD. However, definitive conclusions regarding whether or not ABSI can be used as a predictor for CVD risk cannot be made. Further work in a larger more diverse population involving multiple regressions is warranted.