The optimal range of HbA1c between 6.8% and 7.0% at admission was estimated to have the minimum risk for composite vascular events and stroke recurrence, with notably different levels according to stroke subtype.
During a 1-year follow-up of patients with acute ischemic stroke (AIS) and diabetes, the risk of cardiovascular events, including stroke, myocardial infarction (MI), and vascular death, significantly increased in those with glycated hemoglobulin (HbA1c) at admission.
This retrospective cohort included 18,567 patients with either transient ischemic attack (TIA) or AIS from the stroke registry of the Clinical Research Center for Stroke in Korea. Senior author Moon-Ku Han, MD, PhD, professor of neurology, Seoul National University College of Medicine, and colleagues used a Fine-Gray model to calculate the association between HbA1c and composite vascular events. Fractional polynomial (FP) and linear-quadratic models were utilized to estimate the risk of composite vascular events according to the ischemic stroke subtype.
At 1-year follow-up, 1437 patients developed the composite vascular events and 954 patients developed stroke. Sensitivity analysis that included patients without HbA1c data showed that higher HbA1c levels (>6.8%-7.0%) were associated with an increased risk of composite vascular events and stroke recurrence.
READ MORE: Urgent Secondary Prevention of TIA or Minor Stroke Has Long-Term Positive Effect
Original multivariable-adjusted HRs for 1 unit increase of HbA1c on admission were 1.04 (95% CI, 1.01-1.07) for composite vascular events and 1.02 (95% CI, 0.99-1.06) for stroke recurrence. When comparing prespecified ranges of HbA1c (<6.5, ≥6.5-<7.0, and ≥7.0), HbA1c levels greater than 7.0% at admission were associated with a significantly higher HR for both the composite vascular event (1.27 [95% CI, 1.12-1.44]) and stroke recurrence (1.28 [95% CI, 1.10-1.49]) compared with HbA1c admission less than 6.5%.
A subgroup analysis of patients who had the same level of HbA1c, but a higher level of fasting serum glucose (>130 mg/dL) showed a significantly higher risk of composite vascular events (HR, 1.44 [95% CI, 1.16-1.78]) than the reference group (HbA1c <6.5, fasting glucose ≤130 mg/dL). When evaluating those with HbA1c between at least 6.5% and 7.0% at admission, only those with fasting serum glucose levels greater than 130 mg/dL demonstrated a significantly increased risk of composite vascular events compared with the reference group (HR, 1.44 [95% CI, 1.15-1.79]).
According to the FP model, the optimal range of HbA1c associated with a minimum cardiovascular risk was the lowest for small vessel occlusion (HR, 6.6 [95% CI, 6.8-7.9]), followed by large artery atherosclerosis (HR, 7.3 [95% CI, 6.8-7.9]) and cardioembolism (HR, 7.4 [95% CI, 6.3-8.5]).
The study authors wrote, "more stringent glucose control may be justified in patients with stroke with small vessel occlusion subtype, which is consistent with the current recommendation for preventing the microvascular complications of DM (diabetes myelitis)."
In patients with cerebral artery atherosclerosis greater than and less than 50%, the HbA1c with the lowest risk was 8.8% (95% CI, 8.3-9.2) and 7.8% (95% CI, 7.5-8.0), respectively, in the FP model. As for the linear-quadratic model, HbA1c with the lowest risk was still 8.8% (95% CI, 8.3-9.2) for the greater than 50% cerebral artery atherosclerosis group, and increased to 8.2% (95% CI, 7.9-8.6) in the less than 50% group.
Even after considering age, sex, stroke subtype, presence of cerebral artery atherosclerosis, history of endovascular recanalization therapy, and premorbid neurologic status, the effects of prestrike glucose control on composite vascular events and stroke recurrence remained consistent. Although, the investigators did note there was a significant interaction between composite vascular events and fasting glucose level at admission.
Marginally significant interactions were observed between stroke recurrence and history of antidiabetic medicating. HRs of 1.05 and 0.96 per 1 unit increase of HbA1c were recorded in patients with and without prior antidiabetic medication (P interaction = .022), respectively, while the HRs of composite vascular events in patients with and without prior antidiabetic medication were 1.06 and 0.99, respectively.
One of the notable limitations of the study was that investigators only recorded HbA1c measurements at admission with follow-up measurements not available. The also wrote, “The association between HbA1c at admission and subsequent cardiovascular disease may not reflect the effort on glycemic control during follow-up. Yet the deleterious effects of poor glycemic control can last despite adequate control afterward."