HbA1c-A Marker of Subclinical Atherosclerosis in Asymptomatic Individuals

1 Jul, 21

Introduction

Glycosylated hemoglobin (HbA1c) is a valuable biomarker that can be used along with risk estimators such as Systematic Coronary Risk Estimation (SCORE) and atherosclerotic cardiovascular Disease (ASCVD) risk score.

Aim

To determine the association between HbA1c and the extent of subclinical atherosclerosis (SA)
To ascertain the additive value of HbA1c when used with key cardiovascular risk factors (CVRFs) to predict the presence and extent of SA

Profile of the Study Participants

A cohort of middle-aged participants from the PESA (Progression of Early Subclinical Atherosclerosis) study (age; 40-54 years, n=3,973)
The study participants had no history of CVD and had HbA1c levels in the nondiabetic range at baseline

Methods

Study Design

Observational study

Categorization and Assessments

HbA1c was categorized into 8 categories: [≤4.8% (reference), 4.9% to 5.0%, 5.1% to 5.2%, 5.3% to 5.4%, 5.5% to 5.6%, 5.7% to 5.8%, 5.9% to 6.0%, and 6.1% to 6.4%) to explore associations; and further in 3 categories (≤5.2%, 5.3% to 5.6%, 5.7% to 6.4%) for additional analysis.
Predicted probabilities of CV events were estimated individually with the SCORE risk algorithm, which predicts 10-year risk of CV death
The main analysis was repeated by replacing HbA1c by fasting plasma glucose (FPG) or by replacing the SCORE risk equation with ASCVD risk score
FPG was also categorized into 8 categories, including 3 categories in the pre-diabetes range as per the ADA criteria (≤79, 80 to 84, 85 to 89, 90 to 94, 95 to 99, 100 to 104, 105 to 109, and 110 to 125)
As per the predicted probabilities of CV events based on the 10-year ASCVD risk equations for asymptomatic adults, patients were categorized to be at low, borderline, or intermediate risk (<5%, 5% to 7.4%, and 7.5% to 19.9%).
Presence and extent of SA was ascertained using 2-dimensional vascular ultrasound and non-contrast cardiac computed tomography (CCT)
Coronary artery calcium scoring (CACS) was estimated from CCT

Results

The mean age of the study population was 45.7 years and 37.7% of them were females. Mean HbA1c was 5.4% and the median 10-year risk of CV death as per the SCORE index was 0.35%.
After adjusting for established CVRFs, HbA1c (≤4.8% as reference) was positively associated with the prevalence and multi-territorial extent of SA (Table 1)

Table 1: Odds ratio for SA as per the HbA1c category

HbA1c Category	Odds Ratio (95% CI)*
4.9% to 5.0%	1.05 (0.77-1.45)
5.1% to 5.2%	1.27 (0.95-1.70)
5.3% to 5.4%	1.27 (0.96-1.68)
5.5% to 5.6%	1.36 (1.03-1.80)
5.7% to 5.8%	1.80 (1.33-2.43)
5.9% to 6.0%	1.87 (1.33-2.63)
6.1% to 6.4%	2.47 (1.62-3.76)

*p<0.001

The association remained significant in all pre-diabetes groups as well as amongst those below the pre-diabetes cut-off (HbA1c 5.5% to 5.6% odds ratio: 1.36; p =0.033).
High HbA1c was associated with an increased risk of SA in low-risk individuals (p < 0.001), but not in moderate-risk individuals (p =0.335).
FPG had no association with SA.
Relative risk estimations using SCORE or ASCVD predictors reaffirmed that inclusion of HbA1c altered the risk of multi-territorial SA in most risk categories.

Conclusions

In a large cohort of asymptomatic individuals without diabetes, HbA1c levels were associated with SA. Thus, routine use of HbA1c along with traditional CVRFs can better identify asymptomatic individuals at higher risk of SA.
Lifestyle interventions and novel antidiabetic medications may have a key role in reducing both HbA1c levels and SA in individuals without diabetes.

J Am Coll Cardiol 2021;77:2777–91.