Effect of systolic blood pressure status on coronary inflammation and high-risk plaque characteristics

Abstract

BACKGROUND

Inadequately controlled hypertension often leads to an increased cardiovascular death rate in type 2 diabetes mellitus (T2DM). It remains unclear whether systolic blood pressure (SBP) status of hypertension is related to coronary inflammation and plaques in T2DM.

AIM

To evaluate whether SBP variability (SBPV) and levels of hypertension are related to coronary inflammation and plaques in T2DM patients using coronary computed tomography angiography (CCTA).

METHODS

This retrospective study involved 881 T2DM patients with CCTA images, including 668 hypertension and 213 normotension patients. Hypertension patients were subgroup based on SBP status: (1) SBPV: Low (< 8.96 mmHg) and high (≥ 8.96 mmHg) groups; and (2) SBP levels: Controlled (< 140 mmHg) and uncontrolled (≥ 140 mmHg) groups. Pericoronary adipose tissue (PCAT) attenuation, high-risk plaques (HRPs) and obstructive stenosis (OS) were evaluated by CCTA. Propensity score matching was utilized to compare these CCTA findings for these groups. The impact of SBPV and SBP levels of hypertension on these CCTA findings in T2DM patients were evaluated by multivariate logistic regression and multivariable linear regression.

RESULTS

PCAT attenuation of the left anterior descending artery (LAD), any low attenuation plaque (LAP), any spotty calcification (SC), any positive remodeling (PR), and OS had significant differences between the hypertension group and the normotension group, as well as between the high SBPV or uncontrolled SBP group and the low SBPV or controlled SBP group (all P < 0.05). Hypertension was independently positively correlated with LAD-PCAT attenuation (β = 1.815, P = 0.010), LAP (OR = 1.612, P = 0.019), SC (OR = 1.665, P = 0.013), PR (OR = 1.549, P = 0.033), and OS (OR = 1.928, P = 0.036) in all T2DM patients. Additionally, high SBPV and uncontrolled SBP were independently positively correlated with LAD-PCAT attenuation (high SBPV: β = 1.673, P = 0.048; uncontrolled SBP: β = 2.370, P = 0.004) and PR (high SBPV: OR = 1.903, P = 0.048; uncontrolled SBP: OR = 2.230, P = 0.013) in T2DM patients with hypertension.

CONCLUSION

Inadequately controlled hypertension, including high SBPV and/or uncontrolled SBP levels, may be related to increased coronary artery inflammation, HRPs, and OS in T2DM, leading to increased cardiovascular risk. Achieving both low SBPV and controlled SBP levels simultaneously, especially in individuals with T2DM and hypertension, warrants clinical attention.

Key Words: Type 2 diabetes mellitus; Hypertension; Coronary computed tomography angiography; Pericoronary adipose tissue attenuation; High-risk plaques

Core Tip: Pericoronary adipose tissue attenuation, high-risk plaques (HRPs), and obstructive stenosis (OS) are major risk factors for cardiovascular disease evaluated using coronary computed tomography angiography. Our findings demonstrated that inadequately controlled hypertension, including high systolic blood pressure variability (SBPV) and/or uncontrolled systolic blood pressure (SBP) levels, may be related to increased coronary inflammation, HRPs, and OS in type 2 diabetes mellitus (T2DM), leading to increased cardiovascular risk. Achieving both low SBPV and controlled SBP levels simultaneously, especially in individuals with T2DM and hypertension, warrants clinical attention.

INTRODUCTION

Concomitant hypertension is present in approximately two-thirds of type 2 diabetes mellitus (T2DM) patients, which has been linked to an increase of twofold for the likelihood of cardiovascular disease (CVD) risk in T2DM[1,2]. In individuals with T2DM, adequately controlled hypertension is pivotal in mitigating the comorbidity rates and mortality. Lowering 10 mmHg of the average systolic blood pressure (SBP) results in a 12% drop in any complications connected to diabetes and a 15% reduction in mortality associated with diabetes[3]. Additionally, SBP variability (SBPV), regardless of the average SBP level, may serve as a possible indicator for the onset of CVD in T2DM patients[4].

Both T2DM and hypertension are associated with endothelial dysfunction, vascular inflammation, and atherosclerosis[5]. Pericoronary adipose tissue (PCAT) attenuation reflecting coronary inflammation assessed by coronary computed tomography angiography (CCTA) could significantly predict CVD in patients with T2DM[6]. In addition, CCTA assessments revealed a significant occurrence of obstructive atherosclerotic CVD and high-risk plaques (HRPs) in T2DM patients[7,8]. PCAT attenuation reflects changes caused by early and persistent inflammation. In contrast, HRPs and obstructive stenosis (OS) signify a sequence of alterations that occur in the later stages of the inflammatory process[9]. Therefore, PCAT attenuation along with plaque characteristics detected by CCTA could offer valuable complementary insights into the condition of coronary arteries.

Evidence from a previous study suggests that the likelihood of developing more extensive coronary artery plaques and partially calcified plaques further increases when both hypertension and T2DM are present[10]. Furthermore, previous analysis found that variability and levels of SBP were important factors linked to atherosclerosis progression[11,12]. However, the effects of SBP status of hypertension in T2DM on major risk factors for CVD assessed by CCTA still lack further convincing research. Given this connection, we intended to evaluate whether SBP status of hypertension is related to coronary inflammation and plaques in T2DM patients using CCTA.

MATERIALS AND METHODS

Study population

We enrolled retrospectively 1072 T2DM subjects with CCTA from January 2022 to January 2023 at Liyuan Hospital, Tongji Medical College, Huazhong University of Science and Technology. The following criteria were used for exclusion: Previously diagnosed CVD, such as coronary revascularization, stroke, or myocardial infarction; severe kidney failure (glomerular filtration rate calculated to be below 30 milliliters per minute per 1.73 m²); suspected infectious diseases; concomitant neuroendocrine tumors; and poor image quality or missing image. Consequently, this study included 881 T2DM patients (Figure 1). Hypertension patients were sub-group based on SBP status: (1) Median split of SBPV: SBPV < 8.96 mmHg (low SBPV group) and SBPV ≥ 8.96 mmHg (high SBPV group); and (2) Controlled or uncontrolled SBP: SBP < 140 mmHg (controlled SBP group) and SBP ≥ 140 mmHg (uncontrolled SBP group)[13].

Figure 1 Flow diagram. CCTA: Coronary computed tomography angiography; T2DM: Type 2 diabetes mellitus; SBP: Systolic blood pressure; SBPV: Systolic blood pressure variability.

The Ethics Review Committee of Liyuan Hospital, Tongji Medical College, Huazhong University of Science and Technology approved the study protocol, which was carried out in keeping with the Declaration of Helsinki (No. [2024] IEC RYJ002). The committee waived the informed consent requirement because the study was retrospective in nature.

Date collection and definition

Two radiologists together gathered age, sex, body mass index (BMI), CAD risk factors, laboratory findings, and prior drug usage from inpatient clinical records. T2DM was defined as any of the following: Receiving hypoglycemic therapy; glycated blood glucose protein ≥ 6.5%; or fasting blood glucose ≥ 7.0 mmol/L[14]. Hypertension was characterized by the use of antihypertensive medication or diastolic blood pressure ≥ 90 mmHg and/or SBP ≥ 140 mmHg at rest[15]. Dyslipidemia was identified by any of the following: Diagnosis/treatment of dyslipidemia, serum triglycerides > 2.3 mmol/L, low-density lipoprotein (LDL) cholesterol > 4.1 mmol/L, high-density lipoprotein cholesterol < 1.0 mmol/L, or total cholesterol > 6.2 mmol/L[16]. Smoking status categorizes subjects as current smokers and non-smokers.

CCTA acquisition

All subjects had CCTA examinations performed with a 128-slice wide detector computed tomography (CT) scanner (Revolution HD, GE Healthcare, United States). The heart rates of patients were kept close to 70 beats per minute, with regular recommendations for oral metoprolol. CT angiography (CTA) scans were obtained with prospective electrocardiogram-triggered tube voltage, and the scanner's automatic exposure control system modified the tube current depending on the patient's body size.

SBPV and SBP levels

Brachial blood pressure was recorded at the same levels as the right atrium with an OMRON electronic blood pressure monitor on the upper arm[17]. Blood pressure was measured prior to taking the medication, ensuring no caffeine or tobacco had been consumed in the previous 30 minutes. Measurements were taken following a 5-minute period of comfortable rest while seated. Each recorded SBP value was the average of three readings taken at 5-minute intervals. Measurements were performed at 8:00 in the morning, midday, and 4:00 in the afternoon. Recordings were made for at least five consecutive days immediately before the CCTA examination, with three measurements each day, resulting in a total of at least 15 measurements. SBP levels are defined as the mean of all recorded SBP values. The standard deviation (SD) of these recorded SBP readings is known as SBPV. Among various metrics of SBPV, we chose to use the SD of SBP because the average actual variability and the coefficient of variation were both greater than the Akaike information criteria, suggesting that the SD of SBP measurements was the most suitable for the analysis, as previously reported[18].

CCTA analysis

Specialized software for coronary artery analysis (Shukun Technology, 1.0.4 version) was used for analyzing PCAT attenuation[19]. Measurements were taken from the first 40 mm of the left circumflex artery and left anterior descending artery (LAD) and 10-50 mm from the origin of the right coronary artery. The mean weighted CT attenuation of fat tissue (-190 to -30 HU) within the radial range equally distant from the mean width of the blood vessel was automatically calculated as PCAT attenuation[20]. Figure 2 illustrates an instance of PCAT attenuation.

Figure 2 Coronary computed tomography angiography evaluation of left anterior descending artery-pericoronary adipose tissue attenuation example. A: Heart reconstruction in three dimensions; B: The straightened view of the proximal coronary artery segment; C: The cross-sectional view of pericoronary adipose tissue attenuation ranging from -190 and -30 HU; D: Surrounding the first 40 mm of the left anterior descending artery. LAD: Left anterior descending artery.

We examined coronary segments > 2 mm in width and described plaque features[21]. The HRPs were characterized by positive remodeling (PR), napkin-ring sign (NRS), spotty calcification (SC), and low attenuation plaque (LAP). LAP was accessed by CT values of less than 30 HU in areas larger than 1 mm² within the plaque. SC was identified by separate calcifications in a non-calcified plaque measuring less than 3 mm in any plane, where the short diameter is under two-thirds of the vessel width and the length is less than 1.5 times the vessel width. The proportion of the largest coronary diameter (encompassing lumen and plaque) at the lesion site to the proximal normal vessel diameter is 1.1 or higher, indicating PR. A higher CT attenuation rim-like region surrounding a low CT attenuation area is the characteristic of the NRS[22-24]. The degree of stenosis was measured and classified according to a visual estimation scale derived from the Coronary Artery Disease-Reporting and Data System[21]: Stenosis grades: 0 (no plaques), 1 (< 25%), 2 (25%-49%), 3 (50%-69%), 4 (70%-99%), 5 (complete blockage). OS refers to stenosis exceeding 50%. Non-OS is characterized by the lack of OS. We characterized adverse CTA findings by identifying the existence of HRPs and/or OS. Figure 3 illustrates examples of representative HRPs. Two cardiovascular radiologists independently examined the images without knowledge of the clinical details of patients. They achieved consensus through discussion when disagreements arose.

Figure 3 Image of high-risk coronary plaques. A: Low attenuation plaque; B: Spotty calcification; C: Napkin-ring sign; D: Positive remodeling with a remodeling index = 1.57 [lesion plaque area (0.47 cm)/reference area (0.30 cm)].

Statistical analysis

Median with interquartile range or mean ± SD was used to present quantitative variables. Both counts and percentages (%) were used to display categorical variables. Quantitative variables were compared between the two groups using the Mann-Whitney U test or the paired Student's t-test, as appropriate, and between multiple groups using the Kruskal-Wallis rank test or a one-way ANOVA with Bonferroni post-hoc testing. Depending on the data cell sizes, either the Fisher's exact test or the χ² test was used to compare dichotomous variables. The one-sample Kolmogorov-Smirnov test was used to assess the assumption of normalcy. The influence of baseline feature discrepancies between the two groups was reduced by performing 1:1 propensity score matching using a 0.02 caliper. The dependent variable in the logistic regression model was hypertension, high SBPV, or uncontrolled SBP, and the independent variables included age, gender, BMI, smoking habits, dyslipidemia, tube voltage, and the use of statins. Due to findings from earlier studies that suggest statins affect coronary inflammation, these medications were included as matching factors[25]. To access the relationships between hypertension and PCAT attenuation, HRPs, and OS, we conducted univariate and multivariate analyses using linear regression and logistic regression. The independent variables included age, gender, BMI, smoking, dyslipidemia, hypertension, SBPV, SBP levels, the use of statins, antihypertensive medications, and glucose-lowering drugs. The dependent variables examined were LAD-PCAT attenuation, LAP, SC, PR, and OS. For the multivariate analysis, any independent variable that had a P value less than 0.05 in the univariate analysis was chosen. Statistical significance was found using P < 0.05. GraphPad Prism 7.0 and SPSS Statistical Software version 25.0 were used to perform the analyses.

RESULTS

Subjects and clinical baseline characteristics

The study involved 881 T2DM patients (mean age: 62 years; 447:434 M:F), comprised of 668 hypertension subjects and 213 normotension subjects.

The hypertension subjects were categorized into low and high SBPV groups, with clinical baseline features displayed in Table 1. The normotension group was generally younger and had lower levels of SBP, diastolic pressure, and glycated hemoglobin (HbA1c; P < 0.05). The high SBPV group showed a greater proportion of males, an elevated BMI value, an increased level of all laboratory findings, and an increased use of statins and oral medications for T2DM treatment than the normotension group (P < 0.05). Furthermore, the high SBPV patients were characterized by older age, a greater percentage of male patients, a higher level of all laboratory findings except for total cholesterol, and higher use of statins, angiotensin receptor blockers/angiotensin-converting enzyme inhibitors, calcium channel blockers, and oral medications for T2DM treatment than low SBPV patients (P < 0.05).

Table 1 Clinical baseline characteristics of the normotension group, low systolic blood pressure variability group, and high systolic blood pressure variability group.

Variables	Normotension group, n = 213	Low SBPV group, n = 334	High SBPV group, n = 334	P value
Baseline characteristic
Age (years)	58.24 ± 9.32	61.83 ± 10.19a	65.03 ± 10.88a,b	< 0.001
Male sex	90 (42.3)	164 (49.1)	193 (57.8)a,b	0.001
Body mass index (kg/m2)	23.71 ± 3.15	24.12 ± 3.13	24.73 ± 3.99a	0.009
Smoking	33 (15.5)	57 (17.1)	71 (21.2)	0.187
Dyslipidemia	94 (44.1)	135 (40.4)	163(48.8)	0.115
Systolic pressure (mmHg)	118 ± 9	136 ± 10a	149 ± 16a,b	< 0.001
Diastolic pressure (mmHg)	73 ± 8	78 ± 11a	85 ± 14a,b	< 0.001
Laboratory findings
Fast glucose (mmol/L)	6.68 ± 2.87	7.66 ± 2.82	7.72 ± 3.93a,b	0.005
HbA1c (%)	7.37 ± 2.36	7.42 ± 1.86a	7.79 ± 1.97a	0.027
Triglyceride (mmol/L)	1.36 (0.91, 1.82)	1.46 (0.81, 2.11)	1.55 (0.68, 2.41)a,b	0.001
Total cholesterol (mmol/L)	4.41 ± 1.20	4.44 ± 1.19	4.64 ± 1.26a	0.004
HDL-cholesterol (mmol/L)	1.25 ± 0.36	1.26 ± 0.35	1.28 ± 0.38a,b	0.007
LDL-cholesterol (mmol/L)	2.59 ± 0.85	2.69 ± 0.95	2.79 ± 0.96a,b	< 0.001
Lipid - lowering medication
Statin	128 (60.1)	214 (64.1)	261 (78.1)a,b	0.001
Hypertension treatment
ARB/ACEI	-	128 (38.3)	82 (24.6)b	< 0.001
Beta - blocker	-	87 (26.1)	65 (19.5)	0.064
Diuretics	-	27 (8.1)	22 (6.6)	0.830
Calcium channel blocker	-	119 (35.6)	163 (48.8)b	0.001
Diabetes treatment
Oral	77 (36.1)	142 (42.5)	183 (54.8)a,b	< 0.001
Insulin	36 (16.9)	57 (17.1)	70 (20.1)	0.192
Tube voltage of CT acquisition (kVp)				0.002
100	139 (65.3)	183 (54.8)	152 (45.5)
120	71 (33.4)	144 (43.1)	158 (47.3)
140	3 (1.4)	7 (2.1)	24 (7.2)

^aP < 0.005 vs the normotension group.

^bP < 0.005 vs the low systolic blood pressure variability group.

Data are showed as means ± SD, median (25^th and 75^th percentile), or n (%). SBPV: Systolic blood pressure variability; HbA1c: Glycated hemoglobin; LDL: Low-density lipoprotein; HDL: High-density lipoprotein; ARB/ACEI: Angiotensin receptor blockers/angiotensin-converting enzyme inhibitors; CT: Computed tomography.

Additionally, the hypertension group was divided into controlled SBP and uncontrolled SBP groups, with main clinical characteristics detailed in Table 2. The normotension patients were generally younger, showed a lower proportion of males, and had lower levels of SBP, diastolic pressure, and LDL-cholesterol (P < 0.05). The uncontrolled SBP patients exhibited an increased BMI value, elevated levels of HbA1c, triglycerides, and total cholesterol than normotension patients (P < 0.05). The uncontrolled SBP patients showed fasting increased blood glucose levels than controlled SBP patients (P < 0.05).

Table 2 Clinical baseline characteristics of the normotension group, controlled systolic blood pressure group, and uncontrolled systolic blood pressure group.

Variables	Normotension group, n = 213	Controlled SBP group, n = 313	Uncontrolled SBP group, n = 355	P value
Baseline characteristic
Age (years)	58.15 ± 9.24	62.80 ± 10.07a	63.16 ± 11.16a	< 0.001
Male sex	90 (42.3)	156 (49.8)a	201 (56.6)a	0.00
Body mass index (kg/m2)	23.71 ± 3.16	24.39 ± 3.40	24.43 ± 3.78a	0.030
Smoking	33 (15.5)	51 (16.3)	72 (20.3)	0.208
Dyslipidemia	94 (44.1)	121 (38.7)	157(44.2)	0.379
Systolic pressure (mmHg)	118 ± 9	125 ± 8a	153 ± 15a,b	< 0.001
Diastolic pressure (mmHg)	73 ± 8	76 ± 10a	87 ± 13a,b	< 0.001
Laboratory findings
Fast glucose (mmol/L)	6.45 ± 2.48	7.31 ± 4.10	7.45 ± 3.69	0.065
HbA1c (%)	7.37 ± 2.36	6.76 ± 1.84	7.11 ± 1.96a,b	0.040
Triglyceride (mmol/L)	1.36 (0.91, 1.82)	1.43 (0.83, 2.03)	1.50 (0.62, 2.38)a	0.027
Total cholesterol (mmol/L)	4.41 ± 1.20	4.61 ± 1.24	4.80 ± 1.05a	0.006
HDL-cholesterol (mmol/L)	1.25 ± 0.36	1.21 ± 0.35	1.22 ± 0.38	0.530
LDL-cholesterol (mmol/L)	2.59 ± 0.85	2.73 ± 1.00a	2.99 ± 0.90a	< 0.001
Lipid - lowering medication
Statin	127 (59.6)	207 (66.1)	260 (73.2)a,b	0.001
Hypertension treatment
ACEI/ARB	-	105 (33.5)	106 (29.9)	0.387
Beta - blocker	-	51 (16.3)	50 (14.1)	0.261
Diuretics	-	26 (8.3)	21 (5.9)	0.394
Calcium channel blocker	-	167 (53.4)	162 (45.6)	0.099
Diabetes treatment
Oral	70 (32.9)	100 (31.9)	124 (34.9)	0.762
Insulin	15 (7.0)	27 (8.6)	31 (8.7)	0.647
Tube voltage of CT acquisition				0.001
100	139 (65.3)	174 (55.6)	161 (45.4)
120	71 (33.4)	131 (41.9)	171 (48.2)
140	3 (1.4)	8 (2.6)	23 (6.5)

^aP < 0.005 vs the normotension group.

^bP < 0.005 vs the controlled group.

Data are showed as means ± SD, median (25^th and 75^th percentile), or n (%). SBP: Systolic blood pressure; HbA1c: Glycated hemoglobin; LDL: Low-density lipoprotein; HDL: High-density lipoprotein; ARB/ACEI: Angiotensin receptor blockers/angiotensin-converting enzyme inhibitors; CT: Computed tomography.

Comparison of CCTA parameters between normotension and hypertension groups and within the hypertension group

Comparisons of clinical characteristics and CCTA parameters between normotension and hypertension groups and within the hypertension group before matching are shown in Supplementary Tables 1 and 2.

Once age, gender, BMI, smoking, dyslipidemia, SBP, statin use, and tube voltage have been taken into consideration, we observed significant differences in LAD-PCAT attenuation, LAP, SC, PR, and odds ratio (OR) between the normotension and hypertension groups (LAD-PCAT attenuation: -77.59 ± 7.44 vs -76.23 ± 7.45, P = 0.008; LAP: 22.2% vs 34.1%, P = 0.015; SC: 30.5% vs 31.7%, P = 0.013; PR: 52.1% vs 64.1%, P = 0.027; and OR: 14.4% vs 25.7%, P = 0.009) in all T2DM patients (Table 3). Moreover, after adjusting for these variables, as well as SBP levels or SBPV, significant differences in LAD-PCAT attenuation, LAP, SC, PR, and OR were observed between the low and high SBPV groups (LAD-PCAT attenuation: -77.41 ± 7.50 vs -74.70 ± 8.86, P = 0.002; LAP: 26.7% vs 44.0%, P < 0.001; SC: 30.9% vs 51.8%, P < 0.001; PR: 61.3% vs 81.2%, P < 0.001; and OR: 18.5% vs 28.6%, P = 0.026), as well as between the controlled and uncontrolled SBP groups (LAD-PCAT attenuation: -76.53 ± 7.67 vs -74.58 ± 8.57, P = 0.016; LAP: 26.0% vs 46.6%, P < 0.001; SC: 30.3% vs 52.9%, P < 0.001; PR: 59.1% vs 82.2%, P < 0.001; and OR: 18.9% vs 33.1%, P = 0.001) in T2DM patients with hypertension (Table 4). Figures 4 and 5 illustrate the comparisons of PCAT attenuation, HRPs, and OS among T2DM patients before matching.

Figure 4 Pericoronary adipose tissue attenuation, high-risk plaques, and obstructive stenosis detected by coronary computed tomography angiography. A: The Pericoronary adipose tissue attenuation value of three main coronary arteries; B: The proportion of different high-risk plaques; C: The proportion of obstructive stenosis. LAD: Left anterior descending artery; LCX: Left circumflex artery; RCA: Right coronary artery; PCAT: Pericoronary adipose tissue; NRS: Napkin-ring sign; SC: Spotty calcification; LAP: Low attenuation plaque; PR: Positive remodeling; OS: Obstructive stenosis; SBPV: Systolic blood pressure variability.

Figure 5 Pericoronary adipose tissue attenuation, high-risk plaques, and obstructive stenosis identified by coronary computed tomography angiography. A: The Pericoronary adipose tissue attenuation value of three main coronary arteries; B: The proportion of different high-risk plaques; C: The proportion of obstructive stenosis. LAD: Left anterior descending artery; LCX: Left circumflex artery; RCA: Right coronary artery; PCAT: Pericoronary adipose tissue; NRS: Napkin-ring sign; SC: Spotty calcification; LAP: Low attenuation plaque; PR: Positive remodeling; OS: Obstructive stenosis; SBP: Systolic blood pressure.

Table 3 Comparison of the coronary computed tomography angiography parameters between normotension group and hypertension group after matching.

	Normotension group, n = 167	Hypertension group, n = 167	P value
Age (years)	58.65 ± 8.75	59.16 ± 8.90	0.598
Male sex, n (%)	69 (41.3)	62 (37.1)	0.433
Body mass index (kg/m2)	23.74 ± 3.14	24.03 ± 3.25	0.657
Smoking, n (%)	29 (17.4)	32 (19.2)	0.671
Dyslipidemia, n (%)	77 (46.1)	76 (45.5)	0.913
Statin, n (%)	107 (64.1)	109 (65.3)	0.819
Tube voltage (kVp), n (%)			0.325
100	107 (64.1)	104 (62.3)
120	58 (34.7)	63 (37.7)
140	2 (1.2)	0 (0.0)
LAD-PCAT (HU)	-77.59 ± 7.44	-76.23 ± 7.45	0.008
LCX-PCAT (HU)	-70.99 ± 7.10	-71.24 ± 7.23	0.759
RCA-PCAT (HU)	-76.80 ± 6.90	-77.05 ± 8.13	0.759
High-risk plaques, n (%)
Any LAP	37 (22.2)	57 (34.1)	0.015
Any SC	51 (30.5)	53 (31.7)	0.013
Any NRS	54 (32.3)	53 (31.7)	0.907
Any PR	87 (52.1)	107 (64.1)	0.027
OS, n (%)	24 (14.4)	43 (25.7)	0.009

Data are showed as means ± SD, median (25^th and 75^th percentile), or n (%). LAD: Left anterior descending artery; LCX: Left circumflex artery; RCA: Right coronary artery; PCAT: Pericoronary adipose tissue; NRS: Napkin-ring sign; SC: Spotty calcification; LAP: Low attenuation plaque; PR: Positive remodeling; OS: Obstructive stenosis.

Table 4 Comparison of the coronary computed tomography angiography parameters between two groups of hypertension patients after matching, n (%).

	Low SBPV group vs high SBPV group			Controlled SBP group vs uncontrolled SBP group
	Low SBPV group, n = 191	High SBPV group, n = 191	P value	Controlled SBP group, n = 191	uncontrolled SBP group, n = 191	P value
Age (years)	64.21 ± 10.90	62.74 ± 9.88	0.168	63.72 ± 10.79	63.26 ± 9.86	0.653
Male sex	112 (58.6)	102 (53.4)	0.303	116 (55.8)	111 (53.4)	0.622
Body mass index (kg/m2)	24.40 ± 3.43	24.47 ± 3.88	0.862	24.39 ± 3.48	24.38 ± 3.79	0.953
Smoking	42 (22.0)	44 (23.0)	0.806	41 (19.7)	47 (22.6)	0.806
Dyslipidemia	83 (43.5)	82 (42.9)	0.918	93 (44.7)	93 (44.7)	1.000
Systolic pressure (mmHg)	136 ± 11	136 ± 13	0.876	-	-	-
SBPV (mmHg)	-	-	-	7.56 ± 1.64	9.07 ± 1.25	0.085
Statin	136 (71.2)	134 (70.2)	0.882	152 (73.1)	151 (72.6)	0.912
Tube voltage (kVp)			0.436			0.951
100	102 (53.4)	101 (52.9)		105 (50.5)	107 (51.4)
120	86 (45.0)	83 (43.5)		96 (46.2)	95 (45.7)
140	3 (1.6)	7 (3.7)		7 (3.4)	6 (2.9)
LAD-PCAT (HU)	-77.41 ± 7.50	-74.70 ± 8.86	0.002	-76.53 ± 7.67	-74.58 ± 8.57	0.016
LCX-PCAT (HU)	-70.95 ± 7.45	-70.18 ± 8.37	0.345	-70.39 ± 7.41	-69.67 ± 8.07	0.354
RCA-PCAT (HU)	-76.38 ± 7.81	-75.80 ± 8.59	0.496	-75.90 ± 7.81	-75.28 ± 8.62	0.381
High-risk plaques
Any LAP	51 (26.7)	84 (44.0)	< 0.001	54 (26.0)	97 (46.6)	< 0.001
Any SC	59 (30.9)	99 (51.8)	< 0.001	63 (30.3)	110 (52.9)	< 0.001
Any NRS	66 (34.6)	72 (37.7)	0.523	70 (33.7)	84 (40.4)	0.155
Any PR	117 (61.3)	155 (81.2)	< 0.001	123 (59.1)	171 (82.2)	< 0.001
OS	35 (18.5)	46 (28.6)	0.026	39 (18.9)	59 (33.1)	0.001

Data are showed as means ± SD, median (25^th and 75^th percentile), or n (%). SBP: Systolic blood pressure; SBPV: Systolic blood pressure variability; LAD: Left anterior descending artery; LCX: Left circumflex artery; RCA: Right coronary artery; PCAT: Pericoronary adipose tissue; NRS: Napkin-ring sign; SC: Spotty calcification; LAP: Low attenuation plaque; PR: Positive remodeling; OS: Obstructive stenosis.

Linear and logistic regression analyses of CCTA findings between normotension and hypertension groups and within the hypertension group

Univariate linear regression showed that hypertension, sex, BMI, dyslipidemia, and triglyceride were risk factors for LAD-PCAT attenuation in all T2DM patients, while high SBPV, uncontrolled SBP, sex, BMI, and triglyceride were risk factors for LAD-PCAT attenuation in T2DM patients with hypertension. In multivariable linear regression, hypertension was confirmed to be independently positively correlated with LAD-PCAT attenuation in all patients with T2DM (β = 1.815, P = 0.010), and both high SBPV and uncontrolled SBP were confirmed to exist independently positively connected with LAD-PCAT attenuation in T2DM patients with hypertension (high SBPV: β= 1.673, P = 0.048; uncontrolled SBP: β = 2.370, P = 0.004). Tables 5 and 6 show the detailed analysis results.

Table 5 Univariate and multivariate linear analysis on pericoronary adipose tissue attenuation of left anterior descending artery in all type 2 diabetes mellitus patients.

	Univariate		Multivariate
	β (95%CI)	P value	β (95%CI)	P value
Subgroup
Normotension group	Reference
Hypertension group	1.564 (0.217-2.911)	0.023	1.815 (0.429-3.201)	0.010
Age	0.009 (-0.046-0.064)	0.747
Male sex	2.397 (1.246-3.549)	< 0.001	2.245 (1.037-3.454)	< 0.001
Body mass index	-0.238 (-0.412 to -0.064)	0.008	-0.307(-0.480 to -0.133)	0.001
Smoking	0.729 (-0.780-2.237)	0.343
Dyslipidemia	-1.213 (-2.391 to -0.035)	0.044	-0.682 (-2.081-0.718)	0.339
Fast glucose	-0.054 (-0.226-0.119)	0.543
HbA1c	0.160 (-0.178-0.498)	0.352
Triglyceride	-0.480 (-0.793 to -0.167)	0.003	-0.371 (-0.774-0.033)	0.072
Total cholesterol	-0.239 (-0.739-0.262)	0.350
HDL-cholesterol	1.244 (-0.356-2.844)	0.127
LDL-cholesterol	-0.230 (-0.850-0.389)	0.466
Statin	-0.885 (-2.129-0.359)	0.163
Diabetes treatment	-0.805 (-2.028-0.418)	0.197

HbA1c: Glycated hemoglobin; LDL: Low-density lipoprotein; HDL: High-density lipoprotein; 95%CI: 95% confidence interval.

Table 6 Univariate and multivariate linear analysis on pericoronary adipose tissue attenuation of left anterior descending artery in type 2 diabetes mellitus patients with hypertension.

	Univariate		Multivariate
	β (95%CI)	P value	β (95%CI)	P value
Subgroup
Low SBPV group	Reference
High SBPV group	1.055 (0.510-2.527)	0.029	1.673 (0.016-3.330)	0.048
Subgroup
Controlled SBP group	Reference
Uncontrolled SBP group	2.064 (0.577-3.552)	0.017	2.370 (0.755-3.984)	0.004
Age	0.007 (-0.056-0.071)	0.824
Male sex	2.288 (0.933-3.643)	0.001	2.237 (0.824-3.651)	0.962
Body mass index	-0.306 (-0.505 to -0.106)	0.003	-0.317 (-0.513 to -0.120)	0.002
Smoking	0.319 (-1.422-2.061)	0.719
Dyslipidemia	-0.996 (-2.382-0.391)	0.159
Fast glucose	-0.047 (-0.261-0.166)	0.663
HbA1c	0.283 (-0.140-0.705)	0.189
Triglyceride	-0.414 (-0.748 to -0.080)	0.015	-0.445 (-0.817 to -0.072)	0.019
Total cholesterol	-0.135 (-0.704-0.435)	0.643
HDL-cholesterol	1.110 (-0.759-2.979)	0.244
LDL-cholesterol	-0.108 (-0.829-0.613)	0.770
Statin	-0.798 (-2.297-0.700)	0.296
Hypertension treatment	0.829 (-0.534-2.192)	0.233
Diabetes treatment	-0.578 (-2.010-0.854)	0.428

SBPV: Systolic blood pressure variability; SBP: Systolic blood pressure; LDL: Low-density lipoprotein; HDL: High-density lipoprotein; HbA1c: Glycated hemoglobin; 95%CI: 95% confidence interval.

Univariate logistic regression showed that positive connections between hypertension and LAP (OR = 2.337), SC (OR = 1.906), PR (OR = 2.328), and OS (OR = 2.757) in all T2DM patients, with all P < 0.05. High SBPV and uncontrolled SBP were confirmed to be positively correlated with LAP (OR = 1.831 or 2.190), SC (OR = 1.820 or 2.003), PR (OR = 2.445 or 3.016), and OS (OR = 2.537 or 3.097) in T2DM patients with hypertension, with all P < 0.05 (Table 7). In multivariable logistic regression, hypertension was confirmed to be positively correlated with LAP [OR (95%CI): 1.612 (1.081-2.405), P = 0.019], SC [OR (95%CI): 1.665 (1.112-2.462), P = 0.013], PR [OR (95%CI): 1.549 (1.036-2.317), P = 0.033], and OS [OR (95%CI): 1.928 (1.045-3.555), P = 0.036] independently in all T2DM patients. Additionally, high SBPV [OR (95%CI): 1.903 (1.005-3.603), P = 0.048] and uncontrolled SBP [OR (95%CI): 2.230 (1.185-4.197), P = 0.013] were confirmed to be independently positively correlated with PR in T2DM patients with hypertension (Table 8).

Table 7 Univariate logic regression analysis on adverse coronary computed tomography angiography findings, odds ratio.

	All T2DM patients				T2DM patients with hypertension
	LAP	SC	PR	OS	LAP	SC	PR	OS
Subgroup
Normotension group	Reference
Hypertension group	2.337a	1.906a	2.328a	2.757a	-	-	-	-
Subgroup
Low SBPV group					Reference
High SBPV group	-	-	-	-	1.831a	1.820a	2.445a	2.537a
Subgroup
Controlled SBP group					Reference
Uncontrolled SBP group	-	-	-	-	2.190a	2.003a	3.016a	3.097a
Age	1.039a	1.054a	1.049a	1.030a	1.031a	1.050a	1.040a	0.019
Male sex	1.831a	1.440a	2.092a	1.464a	1.741a	1.490a	1.949a	1.514a
Body mass index	1.006	1.034	1.087a	0.938	1.000	1.026	1.089a	0.972
Smoking	1.766a	0.975	2.392	0.985	1.488a	0.857	2.107a	0.959
Dyslipidemia	1.234	1.146	1.259	1.646a	1.183	1.044	1.191	1.791a
Fast glucose	1.085a	1.045a	1.090a	1.094a	1.088a	1.069a	1.136a	1.106a
HbA1c	1.185a	1.112a	1.116a	1.272a	1.165a	1.131a	1.157a	1.302a
Triglyceride	1.101a	1.040	1.166a	1.070	1.089a	1.020	1.153a	1.052
Total cholesterol	0.842a	0.839a	0.832a	0.720a	0.856a	0.846a	0.857a	0.721a
HDL-cholesterol	0.513a	0.682	0.376a	0.325a	0.512a	0.751	0.371a	0.250a
LDL-cholesterol	0.770a	0.758a	0.776a	0.613a	0.778a	0.763a	0.796a	0.613a
Statin	1.414a	1.949a	1.791a	1.764a	1.147	1.744a	1.625a	1.430
Hypertension treatment	-	-	-	-	1.457a	1.315	1.529a	1.841a
Diabetes treatment	2.648a	2.001a	2.241a	3.064a	2.402a	1.847a	2.219a	3.319a

^aP < 0.05.

T2DM: Type 2 diabetes mellitus; SC: Spotty calcification; LAP: Low attenuation plaque; PR: Positive remodeling; OS: Obstructive stenosis; OR: Odds ratio; SBPV: Systolic blood pressure variability; SBP: Systolic blood pressure; LDL: Low-density lipoprotein; HDL: High-density lipoprotein; HbA1c: Glycated hemoglobin.

Table 8 Multivariate logistic analysis on adverse coronary computed tomography angiography findings, odds ratio (95% confidence interval).

	Model 1			Model 2			Model 3
	Normotension group (n = 213)	Hypertension group (n = 668)	P value	Low SBPV Group (n = 334)	High SBPV group (n = 334)	P value	Controlled group (n = 313)	Uncontrolled group (n = 355)	P value
	Reference			Reference			Reference
LAP		1.612 (1.081-2.405)	0.019		-	0.684	-		0.230
SC		1.665 (1.112-2.462)	0.013		-	0.261		-	0.117
PR		1.549 (1.036-2.317)	0.033		1.903 (1.005-3.603)	0.048		2.230 (1.185-4.197)	0.013
OS		1.928 (1.045-3.555)	0.036		-	0.989	-		0.205

OS: Obstructive stenosis; SC: Spotty calcification; LAP: Low attenuation plaque; PR: Positive remodeling.

DISCUSSION

We explored the impacts of inadequately controlled hypertension on the PCAT attenuation and adverse CTA findings in T2DM. The following were the main findings of our study. Initially, the hypertension group showed higher LAD-PCAT attenuation than the normotension group in all T2DM patients. Additionally, the high SBPV group and the uncontrolled SBP group in T2DM patients with hypertension showed increased LAD-PCAT attenuation than the low SBPV group and the controlled group, respectively. Second, the hypertension group had more LAP, SC, PR, and OS than the normotension group. Additionally, the high SBPV group and uncontrolled SBP group had more PR than the low SBPV group and controlled SBP group, respectively. Ultimately, the multivariate analysis additionally demonstrated that in patients with T2DM with hypertension, including high SBPV and/or uncontrolled SBP, further increased the risk of higher LAD-PCAT attenuation and adverse CCTA findings.

The effect of high SBPV and uncontrolled SBP of hypertension on PCAT attenuation in T2DM

This study revealed the potential relationship between inadequately controlled hypertension and coronary inflammation in T2DM. Hypertension and diabetes have significant overlap in their disease mechanisms, with inflammation considered as one of the common pathways[26]. Hypertension may induce T cell activation via the generation of new antigens, which in turn triggers an inflammatory response[27]. However, to date, the effect of SBP status of hypertension in T2DM patients on PCAT attenuation as an innovative non-invasive method utilizing CCTA scans to directly identify coronary inflammation has not been extensively explored. Hence, our study distinguished itself from earlier studies by focusing on PCAT attenuation not only in T2DM patients but also in those with inadequately controlled hypertension. In this study, our results showed that the hypertension group showed higher LAD-PCAT attenuation than the normotension group. Furthermore, it was observed that LAD-PCAT attenuation was notably increased in the high SBPV group compared with the low SBPV group, as well as in uncontrolled SBP patients compared with controlled SBP patients. This further demonstrates that inadequately controlled hypertension, including high SBPV and uncontrolled SBP, may be related to higher coronary inflammation in T2DM patients. Consistently, a prior study found that by altering the biomechanical cues from pulsatile blood flow, raised blood pressure could increase vascular inflammation[28]. The production of soluble intercellular adhesion molecule-1, which is essential for inflammation[29], and monocyte adherence to the endothelium-an early stage of the inflammatory response-are both increased by cyclic strains of vessels[30]. A significant relationship between SBPV and systemic immune inflammation index was demonstrated in hypertensive patients in a recent study[31]. Systemic and coronary inflammation are strongly correlated, which further supported our results[32]. Therefore, SBP status of hypertension is very important and deserves clinical attention in T2DM patients.

The effect of uncontrolled SBP and high SBPV of hypertension on high-risk coronary plaque in T2DM

Concerning plaque risk, we confirmed the relationship between inadequately controlled hypertension and atherosclerotic plaque in T2DM patients. Prior study by Im et al[33] noted that atherosclerosis and plaque characteristics are substantially correlated with different blood pressure levels. Moreover, Liu et al[34] concluded that blood pressure variability influences the structure and makeup of coronary plaques. High blood pressure variability and levels impair endothelial function through changes in elevated oxidative stress and wall shear stress forces. This initiates a sequence of significant pathophysiological changes, such as increasing cellular permeability and adhesion molecule levels, vascular remodeling, and vascular smooth muscle cell migration and proliferation, ultimately speeding up plaque development[35]. In this study, the hypertension group had increased odds of any LAP, SC, and PR compared with normotension patients. Furthermore, high SBPV or uncontrolled SBP patients showed higher likelihoods of any PR than low SBPV or controlled SBP groups. This result indicates that inadequately controlled hypertension may further exacerbate the risk of HRPs in T2DM patients. LAP was previously found to be a significant and independent indicator of myocardial infarction[36]. Additionally, the presence of SC is linked to diffuse and more extensive coronary atherosclerosis, which accelerates the progression of the illness even when medicinal treatments are used[37]. Earlier studies determined that all lesions linked to or stemming from plaque rupture exhibit PR, potentially serving as a key indicator for identifying lesion vulnerability[38,39]. While this study suggested that hypertension did not elevate the likelihood of NRS in the presence of T2DM, it is important not to overlook that NRS indicates plaques that are prone to rupture[40]. Regardless of other high-risk features identified in coronary CCTA, a strong association was found between the NRS shown on coronary CCTA and future acute coronary syndrome events[41]. HRPs accessed by CCTA are linked to increased odds of upcoming cardiovascular events[42]. Considering the above, this observation could suggest that CVDs develop more rapidly in the presence of high SBPV and uncontrolled SBP of hypertension in patients with T2DM, necessitating further clinical focus.

The effect of hypertension on OS in T2DM

In a prior study, Tay et al[43] found that both T2DM and hypertension are linked to increased odds of coronary OS. This was consistent with our research findings. Based on the current study, the hypertension group showed higher odds of the OS than the normotension group. Im et al[33] found that different blood pressure levels are significantly linked to the odds of OS, which further supported our results. Obstructive CAD patients have a greater annual adverse cardiac event incidence than those with nonobstructive CAD[44]. Consequently, evaluating coronary OS in individuals with both T2DM and hypertension may be essential for effective clinical decision-making.

The significance of our study lies in the finding that inadequately controlled hypertension, including high SBPV and uncontrolled SBP, may further increase coronary inflammation and the risk of adverse CTA findings in T2DM patients. PCAT attenuation could effectively identify underlying coronary inflammation prior to vascular morphological changes. LAD-PCAT attenuation served as a precise indicator for detecting initial perivascular inflammation before the progression to HRPs or OS in comorbid T2DM and inadequately controlled hypertension patients. Specifically, the study indicates that LAD-PCAT attenuation is closely related to SBP status. However, future research is essential to investigate if dynamic changes in SBP status are linked to variations in PCAT attenuation. This enables personalized anti-inflammatory treatments[45] and allows for dynamic non-invasive monitoring of treatment effects using this novel parameter. In addition, adverse coronary CTA findings, including HRPs and OS, are established risk factors for cardiovascular events[46]. Unlike PCAT attenuation, HRPs serve as an imaging hallmark indicative of a more advanced stage. Nevertheless, the causal relationship between SBP status and these major risk factors for CVD assessed by CCTA is still not well investigated, highlighting the necessity for more comprehensive longitudinal studies in this field.

Limitation

Our study is subject to some limitations. Initially, this study was conducted retrospectively as a cross-sectional analysis, which may introduce selection bias in patient selection and are unable to establish a causal link between the SBP status of hypertension and the CCTA findings. Therefore, our findings require validation through more extensive prospective longitudinal cohort studies that incorporate additional potential confounding variables, including dietary patterns, levels of physical activity, and details of medication use. Second, all measured SBP values in our study may not adequately reflect 24-hour dynamic SBP levels or long-term continuous SBP levels, which may potentially influence the results. Additionally, pulse pressure and diastolic blood pressure were not systematically assessed, which may omit some dynamic information about blood pressure. Third, our study did not consider or collect information on other comorbidities (e.g., chronic obstructive pulmonary disease). Future studies should clearly define the inclusion or exclusion criteria for all comorbidities to reduce potential bias and enhance the reliability and scientific rigor of the conclusions. Finally, our study focused solely on T2DM patients, excluding individuals with hypertension alone. Consequently, further research is needed on these aspects.

CONCLUSION

Our study indicated that high SBPV and uncontrolled SBP levels of hypertension in T2DM increased the risk of higher LAD-PCAT and adverse CTA findings in patients with T2DM. Our study reveals an initial indication that inadequately controlled hypertension in T2DM, including high SBPV and uncontrolled SBP levels, may increase the coronary artery inflammation, HRs, and OS. These findings could aid healthcare providers in pinpointing comorbid T2DM and hypertension patients at heightened cardiovascular risk, potentially preventing subsequent cardiovascular-related mortality.