Ambulatory Systolic Blood Pressure and Carotid Atherosclerosis in Type 2 Diabetes: A Critical Role of Glycemic Control Stratification

Introduction

Diabetes mellitus is a chronic metabolic disorder that exerts widespread effects on vascular structures due to long-term metabolic dysregulation.^1,2 Studies^3,4 have shown that subclinical vascular changes, such as carotid plaques, are commonly observed in diabetic patients even in the absence of overt clinical events. These alterations reflect early stages of arterial wall damage and atherosclerotic progression, underscoring the importance of early vascular monitoring in diabetic populations. Since carotid plaques are direct precursors to stroke and other adverse vascular events, their early detection and timely intervention are critical for delaying disease progression.^5,6

Traditional office blood pressure (BP) measurements provide only a snapshot of an individual’s BP at a specific time point. In contrast, ambulatory blood pressure (ABPM) monitoring offers continuous BP readings across a 24-hour period and is considered a more accurate reflection of an individual’s true BP profile,^7,8 particularly useful for early risk assessment in high-risk populations.⁹ Previous studies have demonstrated strong associations between the average levels, variability, and circadian rhythm abnormalities of ABPM with carotid intima-media thickness (IMT) and plaque formation.^10–12 However, most prior ABPM studies in diabetic populations have focused on overall BP or long-term clinical endpoints such as cardiovascular events,^13,14 and few have examined whether the relationship between blood pressure and early atherosclerotic changes is modified by glycemic status.

Hemoglobin A1c (HbA1c), an integrated marker of average blood glucose levels over the past 2–3 months, is widely used to assess long-term glycemic control in diabetic patients.¹⁵ Increasing evidence indicates that elevated HbA1c levels are closely associated not only with atherosclerosis and cardiovascular events,^1,16,17 but also with enhanced endothelial dysfunction^9,20, inflammation, and oxidative stress responses,^18,19 which may in turn influence the vascular response to elevated blood pressure. A previous study in a Finnish cohort²⁰ revealed that as glycemic metabolism deteriorated from normal to diabetic states, individuals exhibited a significant increase in systolic blood pressure (SBP) along with a progressive rise in carotid IMT, suggesting that the relationship between BP and atherosclerotic changes may be more pronounced in individuals with abnormal glucose metabolism. Nevertheless, it remains unclear how dynamic 24-hour SBP patterns interact with HbA1c levels to affect subclinical carotid atherosclerosis in diabetic patients, highlighting a critical knowledge gap. The present study addresses these gaps by applying 24-hour SBP clustering combined with HbA1c stratification to investigate their joint associations with carotid plaque in patients with type 2 diabetes. This approach allows us to capture dynamic blood pressure patterns over the full day and to explore whether glycemic control modifies the relationship between SBP clusters and subclinical atherosclerosis, providing novel insights into early vascular risk stratification within diabetic populations.

We conducted a cross-sectional observational study to examine the association between ambulatory systolic blood pressure and carotid plaque and to determine whether this relationship varied across HbA1c strata. Using data from ABPM and carotid ultrasound, we evaluated the associations of systolic and diastolic blood pressure at different time intervals with carotid plaque. Stratified analyses by glycemic control were performed to inform early vascular risk identification and blood pressure management in patients with diabetes.

Methods

Data Design and Participants

This study was a cross-sectional observational analysis based on data derived from the MMC (Metabolic Management Center) platform, a specialized database established by Yuhuan Second People’s Hospital in 2022 to support long-term monitoring and follow-up of patients with diabetes. Participants were consecutively recruited from a single center (Yuhuan Second People’s Hospital) between March 2021 and September 2023, and the platform enrolls new patients annually and typically follows them 3 to 4 times per year. Patients with type 2 diabetes mellitus (T2DM) who underwent 24-hour ABPM monitoring as part of routine clinical care were eligible for this study. The diagnosis of T2DM was determined according to the 2025 American Diabetes Association (ADA) criteria.¹⁵ Data collection for the present analysis was conducted between July 2022 and May 2023. Inclusion criteria included: (1) diagnosis of T2DM; (2) completion of valid 24-hour ambulatory blood pressure monitoring; (3) age ≥ 18 years. Exclusion criteria included: (1) presence of severe cardiovascular or cerebrovascular disease; (2) severe liver or kidney dysfunction; (3) incomplete or poor-quality monitoring data. A total of 498 eligible participants were ultimately included in the analysis. Written informed consent was obtained from all participants prior to enrollment. The study protocol was reviewed and approved by the Ethics Committee of Yuhuan Second People’s Hospital (ID:1708).This study was conducted in accordance with the principles of the Declaration of Helsinki.

Measurement of Important Variables

Glycated hemoglobin A1c (HbA1c) was measured from venous blood samples collected after an overnight fast. HbA1c levels were determined using high-performance liquid chromatography (HPLC), a standardized method certified by the National Glycohemoglobin Standardization Program (NGSP).²¹ Participants were stratified into three groups based on HbA1c levels: <5.7%, 5.7–6.4%, and >6.4%, in accordance with the American Diabetes Association (ADA) guidelines.¹⁵ In this study, since all participants had confirmed T2DM, HbA1c was used to further stratify glycemic control status.

ABPM monitoring was conducted using a validated oscillometric device worn on the non-dominant arm for a 24-hour period. Measurements were automatically recorded every 30 minutes during the daytime (6:00–22:00) and every 60 minutes during the nighttime (22:00–6:00). In this study, eight systolic blood pressure (SBP)-related indicators were measured. Abnormal SBP was defined as 24-hour SBP ≥130 mmHg, daytime SBP ≥135 mmHg, and nighttime SBP ≥120 mmHg. Abnormal DBP was defined as 24-hour DBP ≥80 mmHg, daytime DBP ≥85 mmHg, and nighttime DBP ≥70 mmHg.²² Blood pressure load was calculated as the proportion of readings exceeding these thresholds during the corresponding periods.

The primary outcome was the presence of carotid plaque, assessed using carotid ultrasonography performed by trained technicians following a standardized protocol. Carotid plaque was defined as a focal structure encroaching into the arterial lumen of at least 0.5 mm, or 50% of the surrounding IMT, or having a thickness >1.5 mm as measured from the intima-lumen interface to the media-adventitia interface.²³

Covariates

We collected socio-demographic characteristics and blood biochemical indices of the participants through questionnaires and health examinations. Covariates included gender (male or female), age (years), marital status (unmarried or married), education level (below undergraduate or undergraduate and above), smoking status (yes or no), drinking status (yes or no), BMI (kg/m²), fasting blood glucose (FBG, mmol/L),total cholesterol (TC, mmol/L), low-density lipoprotein cholesterol (LDL-C, mmol/L), and high -density lipoprotein cholesterol (HDL-C, mmol/L).

Statistical Analysis

Data were presented as means and standard deviation (SD) for normally distributed continuous variables, whereas medians and interquartile ranges were utilized for non-normal continuous variables. Categorical variables were described using frequency and percentage. Depending on the appropriateness of each method, baseline characteristics were compared between two groups using the Student’s t-test, Wilcoxon rank test, or chi-square test. The number of missing values is summarized in Table S1. To preserve the largest possible sample size, we employed a multiple imputation chain-equation method.

We summarized the general characteristics of the study population using descriptive statistics. To evaluate the association between dichotomized ambulatory blood pressure monitoring indicators and the presence of carotid plaque, logistic regression models were applied. Analyses were stratified by HbA1c categories (<5.7%, 5.7–6.4%, and ≥6.4%). Given that the selection of ambulatory SBP indicators was hypothesis-driven based on prior literature, formal adjustment for multiple comparisons was not applied.Two models were constructed: Model 1 was unadjusted, while Model 2 was adjusted for potential confounders including age, sex, BMI, history of hypertension, smoking, alcohol consumption, marital status, and educational attainment.

In addition, a joint analysis of systolic blood pressureclusters (24-hour, daytime, and nighttime SBP) was conducted to explore the cumulative effect of multiple abnormal SBP periods on the risk of carotid plaque, particularly among participants with HbA1c ≥ 6.4%. Odds ratios and 95% confidence intervals were reported. Multiplicative interaction terms between HbA1c stratification and SBP indicators were also tested by including interaction terms in the logistic regression models. The results of the stratified and categorized analyses are visualized, which illustrates the association between different SBP levels and the risk of carotid plaque.

Sensitivity analyses were conducted by redefining SBP and DBP load variables using their median values as binary cut-off points instead of clinical thresholds. The robustness of associations under different definitions was evaluated. Subgroup analyses were stratified by gender and age, and both two models were adjusted accordingly. Model diagnostics were performed to evaluate potential overfitting and multicollinearity. The number of outcome events per variable exceeded 10 in fully adjusted models. Variance inflation factors indicated no significant multicollinearity. Model calibration was assessed using the Hosmer–Lemeshow test. All statistical analyses were performed using R software (version 4.4.1). A two-tailed p-value < 0.05 was considered statistically significant.

Results

General Characteristics of the Study Population

Table 1 presents the baseline demographic and clinical characteristics of the study participants, stratified by the presence or absence of carotid plaque. The study included 498 participants with a mean age of 58.00 ± 10.97 years and a mean BMI of 25.88 ± 3.54 kg/m². The majority were male (59.2%), and 91.8% had an education level below high school. More than half (55.2%) had a history of hypertension, and the average HbA1c level was 7.54 ± 1.88%.

Table 1 Comparison of the Baseline Demographics and Clinical Characteristics of the Non-Carotid Plaque Group and Carotid Plaque Group

Compared to those without carotid plaque, participants with carotid plaque were significantly older (62.88 ± 9.47 vs. 53.95 ± 10.49 years, p < 0.001), had lower educational attainment (96.9% below high school vs. 87.5%, p < 0.001), and a higher prevalence of hypertension (61.9% vs. 49.6%, p = 0.008). They also had higher HbA1c levels (7.71 ± 1.87% vs. 7.40 ± 1.88%, p = 0.010).

We stratified the population into three groups based on HbA1c levels: <5.7%, 5.7%–6.4%, and >6.4%. Subsequently, we conducted an analysis of both SBP clusters (Figure S1) and DBP clusters (Figure S2) to investigate their associations with plaque presence. Notably, significant differences were observed exclusively in the SBP cluster within the HbA1c >6.4% group (Figure S3). The finding implies that elevated HbA1c levels (>6.4%) may amplify the association between SBP and plaque presence.

Associations of the ABPM and Carotid Plaque by HbA1c Stratification

Figure 1 presents the logistic regression results for the association between dichotomous ambulatory blood pressure monitoring variables and carotid plaque, stratified by HbA1c levels.

Figure 1 Associations of dichotomous ABPM variable with carotid plaque according to the HbA1c stratification. Model 1, without adjustment for any covariates. Model 2, adjust gender, age, BMI, hypertension, smoking status, drinking status, marital status, education,UACR, LDL and TG.

After adjusting for sex, age, BMI, history of hypertension, smoking status, alcohol consumption, marital status, Urine Albumin-to-Creatinine Ratio(UACR),and educational attainment, higher risks of carotid plaque were significantly associated with SBP cluster variables in individuals with HbA1c ≥ 6.4%. Specifically, elevated 24-hour SBP (≥130, OR: 2.14, 95% CI: 1.3, 3.56, P = 0.005), awake SBP (≥135 mmHg, OR: 1.87, 95% CI: 1.14, 3.11), asleep SBP (≥120 mmHg, OR: 1.73, 95% CI: 1.05–2.87, P = 0.030), and 24-hour SBP load (≥50, OR:1.12, 95% CI: 1.1, 3.03) were all independently associated with higher odds of carotid plaque. Furthermore, significant multiplicative interactions were detected between HbA1c stratification and several SBP-related indicators. These associations were not significant in all participants and participants with HbA1c < 5.7% or 5.7–6.4%.

Associations of Ambulatory SBP Cluster and Carotid Plaque in the Highest HbA1c Category

Figure 2 and Table 2 display the joint effects of ambulatory SBP clusters on carotid plaque in individuals with HbA1c > 6.4%. Compared to those with normal 24h, awake, and asleep SBP, participants with all three SBP measures elevated had a significantly higher risk of carotid plaque (adjusted OR: 2.21, 95% CI: 1.22–4.08). Figure 1 shows that in the HbA1c ≥ 6.4% group, elevated 24h SBP (≥130 mmHg), awake SBP (≥135 mmHg), and asleep SBP (≥140 mmHg) were all significantly associated with increased odds of carotid plaque. Adjusted hazard ratios were 2.04 (95% CI: 1.01–4.18), 2.16 (95% CI: 1.11–4.27), and 2.46 (95% CI: 1.12–5.59), respectively.

Table 2 Joint Analysis of Ambulatory SBP Cluster and Carotid Plaque in the Highest HbA1c Category (HbA1c>6.4%)

Figure 2 Associations of Ambulatory SBP cluster and carotid plaque in the highest HbA1c category (HbA1c>6.4%). Model 1, without adjustment for any covariates. Model 2, adjust gender, age, BMI, hypertension, smoking status, drinking status, marital status, education, UACR, LDL and TG.

Sensitivity Analyses and Subgroup Analyses

The results of the sensitivity and subgroup analyses are detailed in Tables S2–S7. Most results are consistent with the main analysis. We reclassified the blood pressure load variables using the median cutoff instead of the original thresholds (Table S2) and additionally analyzed ABPM variables as continuous measures (Table S3). The associations between blood pressure indicators and carotid plaque remained consistent across both analytical approaches. Notably, only the female population and those aged less than 60 years had results consistent with the main analysis, this indicates that there is heterogeneity in terms of sex and age.

Discussion

This study revealed that in diabetic patients with poor glycemic control (HbA1c ≥ 6.4%), elevated SBP clusters were independently associated with an increased risk of carotid plaque. Notably, the highest risk was observed in individuals with elevated SBP across all periods. In contrast, no similar associations were found in patients with HbA1c < 6.4%. Building on our previous work²⁴ focusing on microvascular outcomes in type 2 diabetes, the present study extends the scope to macrovascular disease by examining the association between ambulatory systolic blood pressure and carotid atherosclerosis across different levels of glycemic control.

Numerous studies have demonstrated the superior predictive value of ABPM over traditional office BP in assessing atherosclerosis risk.^25,26 For example, Cardoso et al reported a strong association between elevated nighttime pulse pressure and the formation of arterial plaques, as well as the incidence of cardiovascular and cerebrovascular events.²⁷ ABPM comprehensively reflects BP load across different physiological states. Notably, nocturnal hypertension has been closely associated with autonomic dysfunction^28,29 and serves as a sensitive indicator of early vascular damage. Additional studies have shown that the standard deviation of daytime SBP, as well as the difference between average daytime and nighttime SBP, are positively correlated with carotid IMT and plaque formation.³⁰ These findings suggest that beyond mean levels, BP variability may also contribute to the pathogenesis of atherosclerosis.

In the present study, we observed a significant association between dynamic SBP clusters and carotid plaque presence only in participants with HbA1c levels ≥ 6.4%. This association was not observed among those with HbA1c < 6.4%, indicating that glycemic status may modulate the effect of BP on atherosclerosis. These findings are consistent with a prior study conducted in an elderly Finnish cohort,²⁰ which found that as glycemic metabolism deteriorated from normal to diabetic levels, both SBP and carotid IMT increased, suggesting a stronger BP–plaque relationship in individuals with dysglycemia. However, unlike that study, all participants in our analysis had diabetes. Our findings indicate that even among diabetic individuals, good glycemic control may be associated with a weaker relationship between elevated SBP clusters and plaque development, whereas poor glycemic control is associated with a stronger relationship.

Of note, we further compared the predictive values of SBP and DBP in relation to plaque formation. Only the SBP cluster was significantly associated with plaque, while DBP, whether measured during the day or at night, did not show a statistically significant relationship. This finding underscores the clinical importance of identifying and managing elevated dynamic SBP, which aligns with previous reports linking SBP more strongly than DBP with IMT, plaque formation, and cardiovascular outcomes.^31,32 From a clinical perspective, our findings suggest that incorporating dynamic SBP pattern assessment into routine evaluation of patients with type 2 diabetes (particularly those with suboptimal glycemic control) may improve early identification of individuals at higher risk of subclinical atherosclerosis. This may support more individualized blood pressure management strategies, including closer monitoring and potentially earlier therapeutic adjustment in patients exhibiting persistently elevated SBP across multiple time periods.

Interaction between blood pressure variability, glycemic control, and carotid plaque in patients with type 2 diabetes and hypertension remains intricate and the effect may be largely rendered by renal dysfunction. In other studies, the effects of albuminuria and glomerular filtration rate on vascular. This highlights the importance of considering renal function when interpreting our results.Therefore, renal function indicators should be carefully considered when interpreting these associations. Future longitudinal studies are warranted to determine whether dynamic SBP clustering predicts plaque progression and cardiovascular events and to evaluate whether targeted intervention on SBP patterns improves long-term vascular outcomes.

Our results suggest that the association between SBP clusters and carotid plaques is modified by HbA1c levels. Several potential mechanisms may underlie this interaction. First, chronically elevated HbA1c reflects sustained hyperglycemic exposure, may activate oxidative stress and inflammatory pathways, induce endothelial dysfunction, and promote the progression of atherosclerotic plaques.^33,34 Second, diabetic patients often experience autonomic dysfunction, particularly sympathetic–parasympathetic imbalance, which may disrupt circadian BP patterns, lead to elevated nighttime BP, and increase BP variability, potentially contributing to plaque instability and arterial stiffness.³⁵ Additionally, elevated HbA1c is often accompanied by worsening insulin resistance and accumulation of advanced glycation end-products (AGEs), both of which may contribute to vascular remodeling and amplify the detrimental effects of BP on the arterial wall.³⁶ These mechanisms provide a plausible explanation for the stronger associations observed in individuals with higher HbA1c, although causality cannot be inferred from our study.

The strengths of this study include its novel stratified analysis of diabetic individuals by HbA1c level to explore the relationship between ABPM and carotid plaques. Our findings highlight the importance of paying close attention to elevated SBP cluster variables in individuals with higher HbA1c, as they are significantly associated with atherosclerotic risk and hold meaningful clinical warning value. However, this study has several limitations. First, as a cross-sectional observational analysis, causal relationships cannot be established. Ongoing follow-up of this cohort will allow future longitudinal assessment of these associations. Second, participants were recruited from a single center in one geographic region, which may limit generalizability. Third, detailed information on diabetes duration and medication use (including antihypertensive and glucose-lowering therapies) was not available, which may have resulted in residual confounding. Finally, although multiple ambulatory blood pressure indicators were examined, the analyses were hypothesis-driven; nevertheless, residual confounding and multiple comparisons cannot be entirely excluded. Future multicenter prospective studies are warranted to confirm these findings.

Conclusions

This cross-sectional study found that among individuals with HbA1c ≥ 6.4%, elevated SBP clusters were independently associated with a higher prevalence of carotid plaque, with the strongest association observed in those with elevated SBP across all periods (adjusted OR = 2.21). These findings suggest a relationship between ambulatory blood pressure patterns and subclinical atherosclerosis in patients with poorer glycemic control, without implying causality. Future multicenter, longitudinal studies are needed to further elucidate the temporal relationships and confirm these findings.