Relationship Between Glycated Haemoglobin and Carotid Atherosclerotic Disease Among Patients with Acute Ischaemic Stroke

Abstract

Objectives

This study aimed to determine the relationship between glycaemic control and carotid atherosclerotic disease among patients with acute ischaemic stroke (AIS).

Methods

This retrospective cross-sectional study took place in the Neurology Department of King Fahad Hospital of University, Khobar, Saudi Arabia, from April to October 2017. Data were collected from the medical records of 244 patients with a diagnosis of AIS confirmed by computed tomography. Doppler ultrasounds of the carotid artery were performed to determine the presence of increased carotid intima media thickness (CIMT) and plaques.

Results

Significantly higher mean glycated haemoglobin (HbA1c) levels were noted in cases with high CIMT values (P = 0.002), but not in cases with carotid plaques (P = 0.360). In addition, there was a significant association between diabetes mellitus (DM) and high CIMT (P = 0.045), but not with carotid plaques (P = 0.075). Finally, while dyslipidaemia and age were independently correlated with high CIMT values (P = 0.034 and <0.001, respectively) and carotid plaques (P <0.001 each), no independent relationships were noted in terms of gender and other risk factors like DM, hypertension and smoking (P >0.050 each).

Conclusion

High HbA1c levels were associated with high CIMT values, but not with carotid plaques. Therefore, HbA1c levels may be useful as an indirect marker of the initial stages of carotid artery atherosclerosis.

Advances in Knowledge

• - The association between glycated haemoglobin (HbA1c) and carotid intima-media thickness (CIMT) has been studied in different populations. This study demonstrated a relationship between HbA1c and CIMT in patients with ischaemic stroke.

Application to Patient Care

• - The findings of this study may help in the early recognition of carotid artery atherosclerotic disease as well as in the formulation of strategies and guidelines for stroke prevention. However, further clinical research regarding different stroke risk factors would be beneficial.

Extracranial carotid artery atherosclerotic disease is an important thromboembolic cause of stroke and a key target in stroke prevention. 1 In addition, growing evidence suggests that carotid atherosclerotic disease is a risk factor for neurocognitive dysfunction.2 Approximately 30% of ischaemic strokes are caused by carotid occlusive disease.3 The sensitivity and specificity of a carotid Doppler ultrasound—which measures the reduction in carotid diameter as well as the carotid intima media thickness (CIMT)—is 90–95%, making it highly sensitive for the detection of atherosclerotic plaques.4 According to previous research, CIMT increases in patients with diabetes mellitus (DM).5 Moreover, raised glycated haemoglobin (HbA1c) levels are related to the risk of carotid atherosclerotic plaques as well as their subsequent thickening and hardening. Such plaque changes are an independent risk factor for stroke, with the risk abruptly increasing at an HbA1c level of >7% regardless of a diagnosis of DM.6–8

In addition to predicting overall stroke risk, poor glycaemic control also influences functional outcomes following an ischaemic stroke. Recovery largely depends on the intrinsic rewiring and remapping of damaged structures, along with surviving regions. The damage induced by chronically altered glycaemic levels is not limited to the extracranial large vessels but also involves the vascular and neural pathways in the brain.9 The effect of chronic hyperglycaemia-induced metabolic dysfunction on the structure and function of the cerebrovascular system leads to impaired blood supply to the salvageable penumbra zone through the collateral vessels and increases susceptibility to reperfusion injuries. Thus, chronic exposure to abnormally high glycaemic levels may significantly impair reparative neovascularisation and reduce neural circuit functional plasticity.9

The association between HbA1c and CIMT has been studied in various populations.10,11 In addition, previous studies have analysed the frequency of carotid artery stenosis among patients with stroke as well as those with severe triple vessel coronary artery disease and DM.12,13 Choi et al. studied the association between HbA1c levels and arterial stiffness and peripheral arterial disease among Korean patients with type 2 DM.14 However, there is a lack of local data from Saudi Arabia analysing the relationship between glycaemic control and carotid atherosclerotic disease. This study therefore aimed to investigate the relationship between elevated HbA1c levels and CIMT and carotid plaques in ischaemic stroke patients in Saudi Arabia. This population is susceptible to multiple vascular risk factors as well as a rising prevalence of neurocognitive disorders and stroke, with evidence of both small vessel disease and large vessel occlusion.15,16

Methods

This retrospective cross-sectional study was carried out in the Department of Neurology, King Fahad Hospital of University, Khobar, Saudi Arabia, between April and October 2017. Retrospective data were collected from the medical records of 400 adult patients with acute ischaemic stroke (AIS) seen at an affiliated hospital between February 2010 and September 2016. As per the guidelines of the World Health Organization, AIS was defined as “rapidly developed clinical signs of focal (or global) disturbance of cerebral function, lasting more than 24 hours or leading to death, with no apparent cause other than of vascular origin”.17 All AIS diagnoses were confirmed by cranial computed tomography. Patients with isolated transient ischaemic attacks, strokes of cardioembolic aetiology or due to vasculitis syndromes and those with coagulation disorders were excluded. Therefore, a final sample of 244 patients with AIS were included in the study.

High-resolution B-mode carotid Doppler ultrasonography of each patient was performed by a radiologist using a MyLab^™ Class C^® 7 MHz linear transducer (Esaote Group, Genoa, Italy), as per international guidelines. 18 The CIMT was calculated as the mean maximal intima media thickness of each carotid artery measured at both the bifurcation and the internal carotid arteries. A CIMT value of >0.8 mm was considered high, as this cut-off value has been found to correlate with vascular risk factors.19,20 A plaque was defined as a localised protrusion of the internal part of the vessel wall into the lumen with a thickness of >1.5 mm between the intimalumen and media-adventitia interfaces.21 HbA1c levels were measured using high-performance liquid chromatography with a G8 HPLC Analyser^® (Tosoh Bioscience Inc., San Francisco, California, USA) as per the National Glycohemoglobin Standardization Program guidelines.22

According to the American Diabetes Association criteria, DM was defined as either a fasting glucose level of ≥126 mg/dL or an HbA1c level of ≥6.5%.23 Other risk factors for stroke and carotid artery disease were also noted, such as hypertension (HTN), dyslipidaemia and smoking. HTN was defined as a systolic or diastolic blood pressure (BP) of >140 mmHg or 90 mmHg, respectively. Dyslipidaemia was defined as a fasting serum cholesterol level of >200 mg/dL, triglyceride level of >150 mg/dL, low-density lipoprotein level of >120 mg/dL and a high-density lipoprotein level of 35–60 mg/dL.24 However, a higher HDL cut-off point (<40mg/dL) was used when assessing cardiac risk in female patients, given the observed incidence of coronary disease in women with comparatively higher HDL-C levels than men.25

Data were analysed using the Statistical Package for the Social Sciences (SPSS), Version 20.0 (IBM Corp., Armonk, New York, USA). The results were presented as frequencies and percentages for age, gender, DM, HTN, smoking, dyslipidaemia, stroke subtype, high/normal CIMT and the presence/absence of carotid plaques and means and standard deviations for age and HbA1c level. An independent t-test was used to compare means and a Chi-squared test was used to compare categorical variables. A P value of <0.050 was deemed statistically significant. A logistic regression analysis was performed to ascertain the effects of age, gender, HTN, dyslipidaemia, smoking and HbA1c level on the likelihood of high CIMT values and carotid plaques.

Ethical approval for this study was received from the Institutional Review Board of Imam Abdulrahman Bin Faisal University (#IRB-2017-01026).

Results

Overall, 244 patients with AIS were included in the study. Of these, 65.6% were male and the mean age was 64.3 ± 14.5 years. Most patients were ≥61 years old (59%). A total of 193 patients were hypertensive (79.1%), 185 had dyslipidaemia (75.8%), 140 were diabetic (57.4%) and 49 were smokers (20.1%). Among the diabetics, the prevalence of HTN, dyslipidaemia and smoking was 84.3%, 84.3% and 22.9%, respectively. There were 93 patients (66.4%) with both HTN and dyslipidaemia, 26 smokers with HTN (18.6%) and 29 smokers with dyslipidaemia (20.7%). In 24 cases, all four risk factors were present (17.1%). More than half of the patients (53.7%) had small vessel disease, while 46.3% had large artery atherosclerosis. High CIMT values were observed in 140 cases (57.4%), while carotid plaques were present in 117 (48%). Mean HbA1c levels were higher among diabetic patients compared to non-diabetic patients (8.6 ± 2.2% versus 5.7 ± 0.8%) 7.4 ± 2.2% [Table 1]. This difference was statistically significant (P <0.001).

Table 1.

Demographic and clinical characteristics of acute ischaemic stroke patients according to diabetes status (N = 244)

Characteristic	n (%)
	Total	Diabetic patients (n = 140)	Non-diabetic patients (n = 104)
Gender
Male	160 (65.6)	93 (66.4)	67 (64.4)
Female	84 (34.4)	47 (33.6)	37 (35.6)
Age in years
≤50	74 (30.3)	41 (29.3)	33 (31.7)
51–60	26 (10.7)	19 (13.6)	7 (6.7)
61–70	54 (22.1)	32 (22.9)	22 (21.2)
>70	90 (36.9)	48 (34.3)	42 (40.4)
Mean ± SD	64.3 ± 14.5	64.4 ± 13.5	64.2 ± 15.9
Range	26–120	34–98	26–120
Risk factors*
HTN	193 (79.1)	118 (84.3)	75 (72.1)
Dyslipidaemia	185 (75.8)	118 (84.3)	67 (64.4)
Smoking	49 (20.1)	32 (22.9)	17 (16.3)
HTN and dyslipidaemia	144 (59)	93 (66.4)	51 (49)
HTN and smoking	35 (14.3)	26 (18.6)	9 (8.7)
Dyslipidaemia and smoking	40 (16.3)	29 (20.7)	11 (10.6)
HTN, dyslipidaemia and smoking	31 (12.7)	24 (17.1)	7 (6.7)
Type of stroke
LAA	113 (46.3)	66 (47.1)	47 (45.2)
SVD	131 (53.7)	74 (52.9)	57 (54.8)
CIMT in mm
>0.8	140 (57.4)	88 (62.9)	52 (50)
<0.8	104 (42.6)	52 (37.1)	52 (50)
Presence of carotid plaques
Yes	117 (48)	74 (52.9)	43 (41.3)
No	127 (52)	66 (47.1)	61 (58.7)
HbA1c level in %
Mean ± SD	7.4 ± 2.2	8.6 ± 2.2	5.7 ± 0.8
Range	3.8–14.0	5.0–14.0	3.8–10.0

SD = standard deviation; HTN = hypertension; LAA = large artery atherosclerosis; SVD = small vessel disease; CIMT = carotid intima media thickness; HbA1c = glycated haemoglobin.

^*Percentages do not add up to 100% as some patients may have had several risk factors.

In addition, mean HbA1c levels were significantly higher in patients with high CIMT values compared to those with normal CIMT values (7.8 ± 2.4% versus 6.9 ± 1.8%; P = 0.002). However, there was no significant difference in the mean HbA1c levels of patients with carotid plaques compared to those without (7.5 ± 2.3% versus 7.3 ± 2.2%; P = 0.360) [Table 2]. Finally, a significant association was observed between the frequency of high CIMT values and the presence of DM (62.9% versus 50%; P = 0.045). However, no significant association was found between the frequency of carotid plaques and the presence of DM (52.9% versus 41.3%; P = 0.075) [Table 3].

Table 2.

Relationship between glycated haemoglobin and carotid intima media thickness and carotid plaques in acute ischaemic stroke patients (N = 244)

Variable	n (%)	Mean HbA1c level in % ± SD	P value*
CIMT in mm
>0.8	140 (57.4)	7.8 ± 2.4	0.002
<0.8	104 (42.6)	6.9 ± 1.8
Presence of carotid plaques
Yes	117 (48)	7.5 ± 2.3	0.360
No	127 (52)	7.3 ± 2.2

HbA1c = glycated haemoglobin; SD = standard deviation; CIMT = carotid intima media thickness.

^*Using an independent sample t-test.

Table 3.

Association between carotid intima media thickness and carotid plaques presence in acute ischaemic stroke patients according to diabetes status (N = 244)

Variable	n (%)		P value*
	Diabetic patients (n = 140)	Non-diabetic patients (n = 104)
CIMT in mm
>0.8	88 (62.9)	52 (50)	0.045
<0.8	52 (37.1)	52 (50)
Presence of carotid plaques
Yes	74 (52.9)	43 (41.3)	0.075
No	66 (47.1)	61 (58.7)

CIMT = carotid intima media thickness.

^*Using a Chi-squared test.

According to the logistic regression analysis, dyslipidaemia (odds ratio [OR]: 2.06, confidence interval [CI]: 1.05–4.04; P = 0.034), HbA1c levels (OR: 0.81, CI: 0.68–0.96; P = 0.019) and age (OR: 1.95, CI: 0.87–2.17; P <0.001) were independent predictors of high CIMT values. Similarly, dyslipidaemia (OR: 2.88, CI: 1.30–6.37; P <0.001) and age (OR: 1.87, CI: 0.85–2.05; P <0.001) were independently correlated with carotid plaques. However, gender (OR: 0.82, CI: 0.44–1.15; P = 0.532), DM (OR: 0.78, CI: 0.38–1.60; P = 0.519), HTN (OR: 1.56, CI: 0.78–3.10; P = 0.205) and smoking (OR: 1.92, CI: 0.88–4.15; P = 0.097) were not independently correlated with CIMT. Similarly, gender (OR: 0.72, CI: 0.36–1.45; P = 0.367), HbA1c levels (OR: 0.97, CI: 0.82–1.16; P = 0.811), DM (OR: 1.41, CI: 0.63–3.15; P = 0.401), HTN (OR: 2.10, CI: 0.92–4.78; P = 0.070) and smoking (OR: 1.79, CI: 0.77–4.15; P = 0.170) were not independently correlated with carotid plaques [Table 4].

Table 4.

Predictors of carotid intima media thickness and carotid plaques in acute ischaemic stroke patients (N = 244)

Risk factor	High CIMT*		Carotid plaques
	Adjusted OR (95% CI)	P value	Adjusted OR (95% CI)	P value
Age	1.95 (0.87–2.17)	<0.001	1.87 (0.85–2.05)	<0.001
Gender	0.82 (0.44–1.15)	0.532	0.72 (0.36–1.45)	0.367
DM	0.78 (0.38–1.60)	0.519	1.41 (0.63–3.15)	0.401
HTN	1.56 (0.78–3.10)	0.205	2.10 (0.92–4.78)	0.070
Dyslipidaemia	2.06 (1.05–4.04)	0.034	2.88 (1.30–6.37)	<0.001
Smoking	1.92 (0.88–4.15)	0.097	1.79 (0.77–4.15)	0.170
HbA1c	0.81 (0.68–0.96)	0.019	0.97 (0.82–1.16)	0.811
Hosmer-Lemeshow goodness-of-fit	0.08		0.45

CIMT = carotid intima-media thickness; OR = odds ratio; CI = confidence interval; DM = diabetes mellitus; HTN = hypertension; HbA1c = glycated haemoglobin.

^*A CIMT value of >0.8 mm.

Discussion

Assessing HbA1c levels is a useful method of determining glucose control. Such testing is now recommended not only for diagnosing DM and pre-DM, but also for predicting cardiovascular risk.26 It has been suggested that glucose variability strongly correlates with HbA1c levels among elderly male patients with type 2 DM.27 Fluctuations in glucose levels can harm the arterial wall due to oxidative stress and endothelial dysfunction.28 Furthermore, glycaemic status is strongly associated with all grades of carotid artery disease.29 The risk of major adverse cardiovascular events is significantly higher among type 2 DM patients who have had the disease for a long time and who have advanced carotid artery stenosis.30

B-mode ultrasonography is a noninvasive method of measuring intimal thickness, plaque formation and stenosis in the peripheral arteries. An increase in CIMT is an early manifestation of carotid artery atherosclerosis; this is subsequently followed by plaque formation and then carotid stenosis due to the narrowing of the lumen.29,31 Several studies have shown that Doppler ultrasonography of the carotid artery can help to predict subsequent major adverse cardiovascular events.31,32

The demographic characteristics of the sample included in the present study were similar to those of previous research concerning ischaemic stroke patients. In accordance with other studies conducted in the Gulf region, the majority of patients were male.16,33 Critically, 30.3% of cases were under 50 years of age, indicating a relatively young age of stroke onset. Dyslipidaemia was the second most frequent risk factor; however, this finding requires further verification with population-based prospective studies. The prevalence of DM in the current study was slightly lower than reported in a previous study of AIS patients in Khobar (57.4% versus 62.7%).16 The mean HbA1c level was comparable to that reported by Singh et al. among stroke patients in India (7.4 ± 2.2% versus 7.51 ± 1.75%).20

In the current study, high CIMT values and carotid artery plaques were present in 57.4% and 48% of cases, respectively. The relation between HbA1c levels and CIMT and carotid plaques has been addressed in several studies. Saha et al. found that the prevalence of high CIMT in stroke patients was ≈71%.34 Singh et al. noted high HbA1c concentrations in ischaemic stroke patients with high CIMT values and carotid plaque while Lee et al. noted an increase in CIMT and plaque formation among ischaemic stroke patients with DM.20,35 The present study similarly found that HbA1c level—as a reflection of poor glycaemic control—was significantly higher among AIS patients with high CIMT values. In contrast, HbA1c level was not significantly associated with carotid plaques. This might be due to the high frequency of HTN and dyslipidaemia in the sample which may have confounded the results.

According to a logistic regression analysis, dyslipidaemia was an independent risk factor for both high CIMT and plaque formation in the current study. Carotid plaque formation is thought to be multifactorial. Cardoso et al. reported that type 2 DM, advanced age, being male, smoking and ambulatory BP (particularly night-time pulse pressure) were the main independent predictors of atherosclerosis.36 In another study, Kowall et al. concluded that CIMT is highly dependent on traditional cardiovascular risk factors; however, the association between blood glucose levels and CIMT disappears after risk factor adjustment.37

The present study was subject to certain limitations. First, as the study was retrospective, the exact duration of DM and increased HbA1c levels could not be established. Second, some of the patients had multiple risk factors which can confound associations. Performing a risk factor adjustment would therefore have strengthened the results. Third, data regarding BP variability—which affects stroke outcomes in carotid artery occlusion and is an important predictor of carotid artery disease—were lacking.38,39 Finally, although Doppler ultrasonography has high sensitivity, there is nevertheless a risk of operator-dependent variability.

Conclusion

Among the studied sample of AIS patients, high HbA1c levels were significantly associated with high CIMT values, but not with carotid plaques. Accordingly, HbA1c may be beneficial as an indirect marker of the initial stages of carotid artery atherosclerosis. Further clinical studies adjusting for risk factors are recommended to determine the true relationship between glycaemic control and carotid artery disease.