40 / 61
CARDIOVASCULAR JOURNAL OF AFRICA • Volume 32, No 4, July/August 2021
212
AFRICA
age has been linked to deteriorating health.
29,30
Decreasing SBP
with age has been associated with dementia, depression, poly-
pharmacy (use of a large number of medications) and increased
number of co-morbidities.
31,32
A steeper decrease in both SBP
and DBP has been associated with a diagnosis of diabetes,
33
and an increase in all-cause and cardiovascular mortality.
34
However, longer-term decreases in BP have also been shown
to occur with or without the presence of hypertension, heart
failure, atrial fibrillation or stroke,
35
implying that decreasing BP
could be due to low cardiac output, a feature of ageing.
36
The
deceleration and decline in BP in old age is also associated with
use of antihypertensive medication,
22
which we did not account
for in this study.
The results of the three group-based trajectories show that
the averages for both BPs were either at the pre-hypertensive
or hypertensive levels for the medium and highly elevated BP
groups. The proportion of women in these two groups of
trajectories aged 45 years or above was more than those in
the lower BP trajectory group. Persistently elevated BP and
hypertension trajectories have been associated with increased
incidence of atrial fibrillation, with the associations being
stronger in women than men.
37
It has also been associated with
a higher risk of subclinical renal damage (SRD) since evidence
shows that the higher the levels of SBP in early life, the higher
the urinary albumin-to-creatinine ratio and risk of SRD later
in life.
38
Some studies have suggested that individuals with high
BP, especially SBP in midlife, are at a higher risk of arterial
stiffening.
39,40
Evidence has shown that continuously high BP
for years is closely correlated with subclinical atherosclerosis,
41
intima–media thickness and left ventricular mass index.
42
The unadjusted trajectories reflect the added effect of ageing
and the influence of other life-course risk factors such as BMI.
Generally, the effect of BMI on BP trajectories affected the
baseline value (intercept) more than the rate of change (slope).
This could be an indication that the effect of BMI on BP is as a
result of BMI increases that usually and rapidly take place early
in life. Accelerated weight gain and increased BMI in childhood
and early adult life increase the risk of elevated BP and the
development of hypertension in later life.
43,44
Surveys in South Africa have shown that the average BMI
for women by age 30 years is more than 28.0 kg/m
2
,
45-49
which is
in the upper range of the overweight level. The majority of the
women in this study were at least 30 years old, and the average
BMI at the first encounter was 29.9 kg/m
2
. An increase in BMI
could lead to arterial stiffness, which may cause the development
of higher BP levels.
50
Higher SBP levels reflect the stiffening of
the arterial walls in areas exposed to increased pressure,
51
while
coronary perfusion of the myocardium may be related to DBP.
52
Allowing the effect of BMI to vary by individual brought
about three more important findings. First, the standard
deviation of the random intercepts increased, an indication that
baseline BPs varied greatly by individual. This was evident from
the clear distinction in the group trajectories, which did not
interact at any time point. Second, the results showed that the
effect of BMI on BP changes was higher in women who initially
were in the normotensive status (low BP group), and the effect
progressively became less in those initially in the pre-hypertensive
(medium BP group) and hypertensive (highly elevated BP) states.
The third issue relates to the correlationbetweenmeasurements
from the same individual, which became more pre-eminent after
allowing the effect of BMI to be specific to each individual.
A stronger correlation can be helpful in tracking those likely
to have persistently high BP. Tracking of a characteristic is
the stability of a certain feature over time or its predictability
based on earlier measurements.
53,54
Tracking the stability of BPs
is of considerable public health interest because those at high
risk of developing hypertension could be identified at an early
stage,
55
through screening. The influence of change in BMI on
BP tracking emphasises the importance of weight control at
an early age. Maintenance of normal weight gain in childhood
may prevent clustering of hypertension and CVD risk factors in
adulthood.
56
The ideal trajectory for BP is one with minimal fluctuations
within the normotensive ranges across all ages. Favourable (less
steep trajectory) BP trends are attributed to socially patterned
and modifiable BP-related exposures such as lifestyle and diet.
57,58
Few studies from isolated communities such as forager-farmers,
have shown minimal age-related BP increases in comparison to
Western societies.
59
These communities however have adopted
a predominantly vegetarian diet with very low salt content, a
physically active lifestyle, and very low or non-existent obesity
levels.
59-61
Individuals who undergo an urban migration from one
of these isolated communities have been found to adapt quickly
to BP profiles of their adopted communities.
60
The main limitation to the study was that we were unable
to account for subjects on antihypertensive medications, which
could also have contributed to the decreasing BPs with increasing
BMI in higher age groups, and the majority of the women (>
70.0%) having SBP in the lowest of the three trajectory groups.
Nonetheless, controlling BMI for this and similar populations
should be prioritised as it could be beneficial in many ways and
possibly cheaper for BP control than medication alone. Another
limitation was that the sample size for women above 60 years at
any time point was small and this was the reason for the volatile
trajectory in this age range.
Notwithstanding, the major strength was in using repeated BP
and anthropometric measurements, which helped in analysing
long-term trends in BP changes as they were influenced by age
and BMI. This could be useful in guiding clinical practitioners
to focus on population segments with particular risk profiles.
Another strength was the study result showing that the effect of
BMI on elevation of BP was not similar for all individuals, and
this could help in clinical practice by designing individualised
interventions.
Conclusions
Three subgroups of increasing SBP and DBP trajectories were
identified, with the majority of the women in each BP type
falling in the lowest group, which on average was initially in
the normotensive state. The effect of BMI on the BP trajectory
for age was highest in women who initially had relatively lower
(mostly in the normotensive state) initial BPs. This BMI effect
gradually dropped in tandem with increasing initial BP. The
study also showed that steep increasing trajectories could be
avoided if preventative interventions are implemented between
30 and 40 years of age fromwhen the BP starts to increase steeply.
Follow-up study is required to find out if these trajectories would
be similar to findings from a larger and more diverse nationally
representative sample.