Home / Resources / Articles / Relationship Between HbA1c and Stroke Risk

Relationship Between HbA1c and Stroke Risk

Jun 16, 2018
 

New meta-analysis striving to expand on the notion that glycemic control and stroke risk are closely associated.

In May 2018, the Journal of the American Heart Association published a systematic review and meta-analysis discussing the relationship between HbA1c and stroke risk. Two previous meta-analyses, performed in 2004 by S. Marinopoulos et al. and in 2012 by N. Zhang et al., also studied HbA1c in people with diabetes and cardiovascular disease, outcomes, and death. In this new meta-analysis, John Peter Mitsios et al. hoped to expand upon this link.

What do we know about stroke risk and glycemic targets? Diabetes is considered a risk factor for stroke development; simply having type 2 diabetes increases a patient’s risk by 1.5 to 4-fold. But with appropriate glycemic control of the disease, diabetes inherent mortality risk and costliness could be reduced. Current ADA guidelines recommend a target HbA1c < 7.0% in most adults. This recommendation comes from studies showing microvascular risk reduction when the goal is achieved, but whether or not macrovascular risk reductions can be attained is not yet established. However, two randomized controlled studies have found benefits, specifically macrovascular event reduction, when stringent goals are reached. Evidence has shown that atypical glucose tolerance and hyperglycemia is often found in those presenting with acute stroke. This finding really warrants the need for further investigation.

This systematic review performed between February 2017 to March 2017 included 5 databases (MEDLINE, Embase, PubMed, Web of Science, and Cochrane Library). Additional information regarding the search strategy, specifically the medical subject headings (MeSH) terms, inclusion criteria, and exclusion criteria, can be found in Table 1: Search Strategy and Inclusion Criteria, and Exclusion Criteria.

Table 1: Search Strategy, Inclusion Criteria, and Exclusion Criteria
MeSH Terms
  • Glycosylated hemoglobin, HbA1c, glycated hemoglobin,
  • Stroke, cerebral infarction, cerebral hemorrhage, and transient ischemic attack
Search Restrictions
  • Human, ≥18 years old, English
Inclusion Criteria
  • Adjusted hazard ratios or risk ratios looking at various HbA1c and stroke risk; stroke risk defined as first or recurrent and subtype
  • Minimum follow-up period of ≥12 months
Exclusion Criteria
  • Failed automatic exclusion criteria found in the Scottish Intercollegiate Guidelines Network tool
  • Specific populations: end-stage kidney disease, dialysis, post-thrombolysis, post-myocardial infarction, and post-operative cohorts
  • Insufficient or missing data
  • Comparing diabetes mellitus to non-diabetes mellitus cohorts

The objective of this study was to determine the relationship between categorical or 1% increases in HbA1c and stroke risk. Analysis was performed and stratified by diabetes status, stroke subtype, and stroke duration. Applying the search strategy listed above, 5,831 articles were obtained. Subsequent analysis and exclusions eliminated most studies. Ultimately, 29 studies, with 532,779 participants, were used for both the meta-analyses and sensitivity analyses. The following associations were found: 1) First-ever stroke risk in American Diabetes Association (ADA) defined HbA1c ranges, 2) First-ever stroke risk and study-quoted 1% HbA1c, and 3) First-ever stroke risk and linear regression estimated 1% HbA1c. Association results presented in Table 2: Association Studied & Results.

Table 2: Associations Studied & Results
Association Results
First-ever stroke risk in ADA defined HbA1c ranges1
  • Non-diabetes mellitus compared to pre-diabetes mellitus. No significant increase in risk of first-ever stroke [average HR 1.19 (95% CI 0.87-1.62)].
  • Diabetes mellitus compared to non-diabetes mellitus. Significant increase in first-ever stroke risk in those with diabetes mellitus [average HR 2.15 (95% CI 1.76-2.63)].
First-ever stroke risk and

study-quoted 1% HbA1c

Study-quoted 1% increase (or equivalent) in HbA1c and first-ever stroke
  • Non-diabetes mellitus cohort. [1.12 average HR (95% CI 0.91-1.39)].
  • Diabetes mellitus cohort. [average HR 1.17 (95% CI 1.09-1.25)].
Study-quoted 1% increase (or equivalent) in HbA1c and first-ever ischemic stroke
  • Non-diabetes mellitus cohort. [average HR 1.49 (95% CI 1.32-1.69)].
  • Diabetes mellitus cohort. [average HR 1.24 (95% CI 1.11-1.39)].
First-ever stroke risk and linear regression estimated 1% HbA1c 1% increase (or equivalent) in HbA1c and first-ever stroke
  • Non-diabetes mellitus cohort. [1.12 average HR (95% CI 1.02-1.34)].
  • Diabetes mellitus cohort. [average HR 1.17 (95% CI 1.01-1.36)].
1% increase (or equivalent) in HbA1c and first-ever ischemic stroke
  • Non-diabetes mellitus cohort. [average HR 1.35 (95% CI 0.91-2.02)].
  • Diabetes mellitus cohort. [average HR 1.32 (95% CI 1.23-1.42)].
American Diabetes Association HbA1c Definitions1

  • Pre-diabetes mellitus: Hba1c 5.7-6.5
  • Non-diabetes mellitus: HbA1c < 5.7%
  • Diabetes mellitus: HbA1c ≥ 6.5%

Overall, the study discovered several significant associations. Firstly, a significant increase was noted in the comparison between diabetes mellitus and non-diabetes mellitus based on ADA definitions; a 2.15-fold increase in first-ever stroke risk was found. Secondly, significance in the non-diabetes cohort was noted but only for the association between first-ever stroke and estimated 1% increase in HbA1c as well as first-ever ischemic stroke and study-quoted 1% increase. Furthermore, in the diabetes mellitus cohort, significance was noted in both study-quoted 1% increase and estimated 1% increase in HbA1c in both stroke subtypes. The results of this study are consistent with existing research, stating that diabetes mellitus is a risk factor more in ischemic stroke than other stroke subtypes. On the other hand, some non-significance was noted regarding first-ever stroke risk and pre-diabetes HbA1c, which agrees with former studies.

Investigators stated some study limitations. They are as follows: 1) Inclusion criteria used studies containing only hazard ratio and relative risk data, 2) lack of consistent definitions for diabetes mellitus across studies, 3) data obtained from studies with differing stroke outcome classification systems, 4) estimation of 1% increase in HbA1c using linear regression model is an approximation of the relationship, 5) exclusion of studies not in published in English, 6) too few articles limited analysis for each subgroup, 7) relevant articles may have been missed.

Ultimately, the investigators found that increased HbA1c is associated with increased first-ever ischemic stroke in both non-diabetes and diabetes mellitus. Investigators wonder if macrovascular complication prevention should begin sooner when HbA1c levels are lower, but future studies are needed to look at more stringent glycemic management.

Practice Pearls:

  • Diabetes mellitus is considered a major risk factor for stroke development.
  • It has been found that an increased HbA1c is associated with an increase in first-ever ischemic stroke regardless of diabetes subtype.
  • Elevations in HbA1c, approximately 1%, was associated with first-ever stroke risk in those with diabetes mellitus.

References:

Practice Update: Stroke Risk Increases as HbA1c Rises. Journal of the American Heart Association. 23 May 2018. Available from: https://www.practiceupdate.com/c/68323/1/8/?elsca1=emc_enews_expert-insight&elsca2=email&elsca3=practiceupdate_diab&elsca4=diabetes&elsca5=newsletter&rid=OTE0MTIxOTE4NTkS1&lid=10332481.

Mitsios, J., Ekinci, E., Mitsios, G., Churilov, L., and Thijs, V. Relationship Between Glycated Hemoglobin and Stroke Risk: A Systematic Review and Meta‐Analysis. Journal of the American Heart Association. 7 (2018): e007858. https://doi.org/10.1161/JAHA.117.007858.

Kaytie A. Weierstahl, Pharm.D. Candidate, LECOM School of Pharmacy