DISCUSSION:
In this prospective cohort study, high allostatic load in early
pregnancy was associated with subsequent composite outcome 2 to 7 years
after delivery. After adjustment for covariates, high allostatic load
was significantly associated with composite outcome, HTN, and MD.
Results were similar in sensitivity analysis, where we adjusted for
prior APOs. The components of allostatic load most strongly associated
with CVD outcomes in exploratory analyses were BMI, SBP, DBP, and CRP.
These data support growing evidence that cumulative stress is associated
with subsequent CVD-related outcomes.
Others also have noted allostatic load as a risk factor for increased
risk of contemporaneous coronary artery disease, ischemic heart disease,
and peripheral arterial disease. 24-27 Similarly,
individuals with type 2 diabetes, elevated blood pressure and worse
glycemic control have higher allostatic load. 24-27Again, these studies assessed contemporaneous allostatic load.33
Generally, subclinical higher levels of blood pressure and glucose
metabolism are associated with subsequent HTN and MD.28 As such blood pressure and metabolic parameters
were individually associated with subsequent CVD and high allostatic
load remained significantly associated with subsequent CVD, even when we
excluded blood pressure and insulin from the allostatic load index for
the composite, HTN and MD.
Self-reported race has been included as a proxy for social experience,
systematic and interpersonal racism, and other unmeasured social
determinants of health. As such allostatic load has been strongly
associated with the non-Hispanic Black race in several studies.29, 30 Black individuals reporting greater perceived
racial discrimination had a higher allostatic load. 31However, one study noted that this was somewhat mitigated by additional
educational attainment. 32 We found that non-Hispanic
Black race was associated with high allostatic load and CVD outcomes.
The addition of allostatic load only partially mediated the relationship
between race and CVD outcomes in our study and not between race and each
component of the composite outcome as such these associations were not
significantly different by race.
Our study has several limitations. Our cohort lacks some
generalizability since it was limited to nulliparous individuals who
could access tertiary medical care centers and had the means to
participate in a complex longitudinal study. This analysis was
restricted to individuals in a follow-up study, introducing potential
bias in the cohort. Also, we only evaluated allostatic load during the
first trimester of the index pregnancy. Thus, we could not assess the
trajectory relationship between allostatic load later in pregnancy or
allostatic load and CVD outcomes longitudinally. Also, we could not
assess the impact of subsequent pregnancies between the index pregnancy
and HHS in person visit 2-7 years postpartum. Although we assessed a
12-factor index and individual components of allostatic load, we did not
assess other combination, including additional metabolic and
cardiovascular indicators that make up a broader allostatic load index,
which may be more robust for evaluating allostatic load and CVD.33
This study had notable strengths. We utilized a large,
well-characterized prospective cohort with standardized data collection
by trained research personnel. Outcomes used rigorous definitions, and
physicians adjudicated uncertain cases. 11Importantly, our population was geographically, racially, and ethnically
diverse and is somewhat representative of the US population. Another
strength is that each allostatic load biomarker component was weighted
equally, a scientifically sound approach. Evidence suggests no
significant difference between empirical and clinical cut-off
assessments. 34
In summary, early pregnancy high allostatic load is associated with
composite maternal CVD outcomes 2-7 years postpartum, particularly HTN
and MD. High allostatic load early in pregnancy could indicate increased
risk for subsequent HTN and MD. High allostatic load modestly mediated
the association between self-reported race and composite outcome, but
not individual components of the composite outcome. Discovery of early
pregnancy biomarkers that are associated with increase long-term CVD
risk might have impact on public health. Thus, pathways contributing to
allostatic load such as stress and inflammation should be investigated
as therapeutic targets intended to decrease CVD.