14 / 66
CARDIOVASCULAR JOURNAL OF AFRICA • Volume 30, No 6, November/December 2019
322
AFRICA
In addition to differences in body fat distribution, the
association with cardiometabolic risk also differs according
to gender, age and ethnicity. For example, African American
women were shown to have a weaker association between
VAT and blood pressure, triglyceride, high-density lipoprotein
cholesterol (HDL-C) and total cholesterol (TC) concentrations
than Caucasian women, while African American men displayed
a stronger association between VAT, triglyceride and low HDL-C
concentrations and the metabolic syndrome (MetS) than their
Caucasian counterparts.
28
While differences in body fat distribution, and associations
with cardiometabolic risk between black, Caucasian and Asian
women have been described in SA,
24,29,30
no studies have examined
the mixed-ancestry population of SA, who present with a high
prevalence of the MetS (62%) and type 2 diabetes (28.2%),
placing this population at high risk for CVD.
31
The composition of the mixed-ancestry (collectively referred
to as ‘Coloured’) population of SA is Khoisan (32–43%), Bantu-
speaking Africans (20–36%), Europeans (21–28%) and a smaller
Asian contribution (9–11%).
32
This population accounts for 8.9%
of the South African population and 48.8% of the population
of the Western Cape Province.
33
The aims of the study were
therefore, for the first time, to investigate the relationship
between whole-body fat distribution and cardiometabolic risk
factors in mixed-ancestry SA men and women, and to explore
the effect of menopausal status on these relationships in women.
Methods
The study sample included all self-described mixed-ancestry
volunteers who completed a whole-body dual X-ray
absorptiometry (DXA) scan as part of the Cape Town Vascular
and Metabolic Health (VMH) study described previously.
34
Inclusion criteria were adults aged 20 years and older. Subjects
were excluded if they were pregnant or acutely ill. A total of 46
men and 207 women volunteered for the study.
Ethical approval was obtained from the Ethics Committees of
the Cape Peninsula University of Technology and Stellenbosch
University (respectively, NHREC: REC-230 408-014, CPUT/
HWS-REC 2015/H03 and N14/01/003). All participants signed
written informed consent and the study was conducted according
to the Code of Ethics of the World Medical Association
(Declaration of Helsinki).
Anthropometric measurements were taken and included body
weight, height and body mass index (BMI), as described in detail
previously.
34
Body composition (fat mass and fat-free mass) was
acquired by a suitably trained and experienced radiographer
using a Hologic Discovery W DXA whole-body scanner
configured with software version 13.4.1 (Hologic, Bedford,
MA). Participants were positioned as per the NHANES body
composition manual, as advocated by Hangartner.
35
DXA-derived measures of body composition regions included
six standard regions of interest (ROI), namely the whole body;
the trunk defined by the lower border of the mandible and
including the chest, abdomen and pelvic triangle; the arm
ROIs (right and left) defined by a line bisecting the shoulder
joint of the right and left arm; and the leg ROIs (right and left)
defined by a line bisecting the hip joint aligned with the iliac
crest and pubis.
27
For the android fat measurement, the ROI is
automatically defined with a caudal limit placed on top of the
iliac crests and its height is set to 20% of the distance from the
top of the iliac crest to the base of the skull as the cephalic limit.
36
The height of the gynoid ROI is double that of the android
ROI with the separation between the two regions equating to
1.5 times the height of the android ROI. VAT and SAT were
estimated within this android region.
DXA has proved to be as accurate as a clinical computed
tomography scan in the quantification of VAT and SAT in
adults.
36
Sub-total body fat % and kg, which excluded the head,
was used in the analysis. The head was excluded to reduce
the possibility of any artefacts in the head region, and total
body adipose tissue classification excludes the head. Regional
fat distribution (arms, legs, trunk, android and gynoid) are
expressed as a percentage relative to sub-total fat mass (%FM).
Blood pressure was measured according to the World Health
Organisation (WHO)
37
guidelines using a semi-automatic digital
blood pressure monitor (Omron M6 comfort-preformed cuff BP
monitor) on the right arm, in a seated position and at rest for 10
minutes. The lowest of three consecutive readings was taken in
the analyses.
34
After an overnight fast (eight to 14 hours), blood samples
were taken to measure levels of glycated haemoglobin (HbA
1c
),
glucose, insulin, lipid profile, and biochemical marker for
inflammation, high-sensitivity C-reactive protein (hsCRP). After
collection of the fasting blood sample, the subjects without
previously diagnosed diabetes underwent an oral glucose
tolerance test as per the WHO criteria.
38
Participants drank 75 g
of anhydrous glucose in 250–300 ml of water over the course of
five minutes,
39
following which blood samples were collected after
the two-hour test load. Blood samples were transported daily in
an icebox for processing using standard pathology practices.
Biochemical parameters were analysed at an ISO 15189
accredited pathology practice (Pathcare, Reference Laboratory,
Cape Town, South Africa) as described elsewhere.
34
Plasma
glucose level was measured by the enzymatic hexokinase method
(Beckman AU, Beckman Coulter, South Africa). HbA
1c
level
was assessed by high-performance liquid chromatography
(Biorad Variant Turbo, BioRad, South Africa). Insulin
concentration was measured with the paramagnetic particle
chemiluminescence assay (Beckman DXI, Beckman Coulter,
South Africa). Levels of HDL-C were measured by enzymatic
immuno-inhibition, triglycerides by glycerol phosphate oxidase-
peroxidase assays, and low-density lipoprotein cholesterol (LDL-
C) by enzymatic selective protection (Beckman AU, Beckman
Coulter, South Africa). Analysis of hsCRP was performed on
the BNA nephelometer (Dade Behring) by particle-enhanced
immunonephelometry with a detection limit of 0.18 mg/l and a
measuring range of 0.18–1 150 mg/l.
Homeostatic model assessment of insulin resistance (HOMA-
IR) was calculated from fasting glucose and insulin levels.
40
The
MetS was quantified using the Joint Interim Statement (JIS)
criteria
39
and the WHO glucose tolerance categories were used.
Statistical analysis
Data were analysed using SPSS
®
version 24 (Armonk, NY:
IBM Corp.) and STATA
®
version 14.2 (STATA corporation,
Texas, USA). The participants in the study were a convenient
sub-sample of the larger study. The 253 available volunteers
who participated in the study provided an 80% power at a 5%