The relationship of hypertension to coronary atherosclerosis and cardiac events in patients with coronary CT angiography

Abstract

Hypertension is an atherosclerosis factor and is associated with cardiovascular risk. We investigated the relationship between hypertension and the presence, extent and severity of coronary atherosclerosis in coronary computed tomography angiography as well as cardiac events risk. Of 17181 patients enrolled in the CONFIRM registry who underwent ≥64-detector row coronary computed tomography angiography, we identified 14803 patients without known coronary artery disease. Of these, 1434 hypertensive patients were matched to 1434 patients without hypertension. Major adverse cardiac events risk of hypertension and no-hypertension patients was evaluated with Cox proportional hazards models. The prognostic associations between hypertension and no-hypertension with increasing degree of coronary stenosis severity (non-obstructive or obstructive ≥50%) and extent of coronary artery disease (segment involvement score 1–5, >5) was also assessed.

HT patients less commonly had no coronary atherosclerosis and more commonly had non-obstructive as well as one, two, and three vessel disease than the no-hypertension group. During a mean follow-up of 5.2±1.2 years, 180 experienced cardiac events, with 104 (2.0%) occurring in the hypertension group and 76 (1.5%) occurring in the no-hypertension group (Hazard ratios 1.4, 95% confidence intervals 1.0–1.9). Compared to no-hypertension patients without coronary atherosclerosis, hypertension patients with no coronary atherosclerosis and obstructive coronary disease tended to have higher risk of cardiac events than the no-hypertension group. Similar trends were observed with respect to extent of coronary artery disease. Compared to no-hypertension patients, hypertensive patients have increased presence, extent, and severity of coronary atherosclerosis and tend to have an increase in major adverse cardiac events.

INTRODUCTION

Hypertension affects almost one-third of adults including >7 million patients in the US^1–3 and is strongly associated with cardiovascular morbidity and mortality^4–9. Although hypertension is a well-established risk factor for coronary artery disease (CAD)¹⁰, the relationships between hypertension and coronary atherosclerotic plaque stenosis, extent, characteristics and major adverse cardiac events (MACE) risk has not been examined. Coronary computed tomographic angiography (coronary CTA) has emerged as an accurate non-invasive modality to evaluate coronary atherosclerotic plaque and assess the risk of patients with suspected CAD^11–14. In this study, we utilized coronary CTA to investigate the relationship between hypertension and the presence, extent and severity of CAD and to explore whether hypertension adds to the assessment of atherosclerosis in prediction of MACE.

METHODS

Study population

From 17181 patients enrolled in the CONFIRM (Coronary CT Angiography Evaluation for Clinical Outcomes: An International Multicenter Registry) registry between 2002 and 2011 who underwent ≥64-detector row coronary CTA at 17 centers, we identified 14803 patients without known CAD who underwent coronary CTA. Of those, we sequentially excluded patients without information regarding MACE (n=6889), early revascularization <3 months after coronary CTA (n=1212), and any risk factors used for matching (n=2272), resulting in a population of 4430 patients. Hypertensive and non-hypertensive subjects were matched for age, gender, all other CAD risk factors including diabetes, dyslipidemia, current smoking or family history, chest pain symptoms (asymptomatic, atypical, non-cardiac, and typical chest pain) and dyspnea using propensity scores¹⁵. The resulting propensity score was then applied 1:1 to match every hypertensive subject (n=2791) to a corresponding non-hypertensive subject (n=1639) using a Mahalanobis nearest-neighbor matching algorithm with caliper <0.01¹⁵. After matching, 2868 patients (age 56.4±11.1 years, male 60.8%) comprised the final study population, with 1434 patients with hypertension (HT) and 1434 patients without hypertension (no-HT). The study was followed by Declaration of Helsinki Guidelines and each institution obtained Institutional Review Board approval. All patients had signed informed consent.

Pre-scan risk factor assessment

All CAD risk factors were prospectively ascertained prior to the coronary CTA examination by direct patient interview by a physician or nurse research coordinator, and by standardized site surveys. Hypertension was defined as a history of physician-diagnosed high blood pressure or treatment with blood pressure medications. Dyslipidemia was defined as physician-diagnosed dyslipidemia or current treatment with lipid lowering medications. Diabetes was defined by physician-diagnosed diabetes, or use of insulin or oral hypoglycemic agents. A smoking history was defined as current smoking or cessation of smoking within 3 months of testing. Family history of CAD was determined by self-report. Chest symptom characteristics (asymptomatic, atypical, noncardiac, and typical chest pain and dyspnea) were recorded¹⁶.

Imaging analysis

Coronary CTA was performed using multi-detector CT scanners with 64-slices or more detector rows as previously described^{14, 17}. CT data sets were evaluated for the presence of any plaque and plaque composition (stenosis and extent) on coronary CTA, employing a modified 16-segment American Heart Association coronary tree model in accordance with society of cardiovascular computed tomography guidelines¹⁸. Coronary plaque was identified as hyperdense structure adjacent to lumen of any size or hypodense structure distinct from lumen and per-arterial tissue >1mm² in largest area. Severity of luminal stenosis was classified into three groups; none (0% luminal stenosis), non-obstructive (1–49% luminal stenosis) and obstructive stenosis (≥50% luminal stenosis). For per-vessel analysis, we employed a five-group categorization: no plaque, non-obstructive CAD, and presence of obstructive CAD in 1-vessel, 2-vessels, or 3-vessels. Left main disease was categorized as a 3-vessel CAD equivalent. For measures of CAD extent and distribution, the segment involvement score (SIS) was defined as the total number of coronary artery segments involved with any plaque¹³. For per-location analysis, we employed a five-group categorization: left main, proximal, mid and distal coronary segments, and side branches including diagonal branches, obtuse marginal branches, posterior descending artery and posterior lateral branch. Detected plaques were visually classified as non-calcified plaque (NCP) [containing no calcification], partially calcified plaque (PCP) [containing calcification and non-calcified plaque], or calcified plaque (CP) [containing only calcification].

Statistical analysis

Continuous variables were expressed as the mean ± standard deviation. The Wilcoxon rank-sum test (for non-parametrically distributed variables) was used to conduct intergroup comparisons between no-hypertension and hypertension groups. Categorical variables were compared using Pearson Chi-squared tests.

MACE was defined as all-cause death or non-fatal myocardial infarction (MI). MI was defined by site physicians in accordance with ACC/AHA guidelines and the World Health Organization Universal Definition of Myocardial Infarction^{19, 20}. The log-rank test was used for comparing MACE event rates between the HT and no-HT groups, and MACE free survival was further assessed using Cox Proportional hazards models and Kaplan-Meier survival curves. We also assessed MACE risk by Cox Proportional hazards models in men and women.

In addition, degrees of stenosis severity (normal, non-obstructive and obstructive CAD≥50%) and extent of CAD (SIS 0, 1–5 and >5) were assessed among no-HT and HT groups in relation to time to MACE by Cox Proportional hazards models. Scaled Schoenfeld residuals were used to verify the assumption of proportional hazards of the Cox models²¹. Hazard ratios (HR) and 95% confidence intervals (CI) were calculated from the Cox models. Area under the curves (AUC) by receiver operator characteristics (ROC) for prediction of MACE were used to evaluate the added value of HT over assessment of coronary atherosclerosis alone (≥50% stenosis or SIS) or the combination of coronary atherosclerosis and clinical factors other than hypertension [“other clinical factors” (age, gender, diabetes, dyslipidemia, current smoking, family history and all chest symptoms)]. We also calculated continuous net reclassification index (cNRI)²² between the models to investigate if HT reclassified patients with respect to MACE risk over the combination of other clinical risk factors and atherosclerosis variables.

All statistical calculations were performed using STATA (Version 11.2, StataCorp LP, College Station, Texas, USA) for Windows.

RESULTS

Patient Characteristics

Table 1 demonstrates the baseline characteristics among patients with and without hypertension. Propensity matching resulted in no differences between the groups in age, male gender, other CAD risk factors and all chest symptoms (p>0.05 for all).

Table 1.

Clinical characteristics (n=2868)

Clinical characteristics	No hypertension (n=1434)	Hypertension (n=1434)	P value
Age	56.3±11.0	56.6±11.1	0.47
Male Gender (%)	60.8	60.8	1.00
Diabetes (%)	7.5	8.5	0.30
Dyslipidemia (%)	48.0	49.0	0.60
Smoking (%)	16.5	18.3	0.20
Family history (%)	33.3	34.8	0.41
Chest pain status (%)
Asymptomatic	39.8	37.7	0.17
Non-cardiac	14.2	15.8
Atypical	37.3	35.9
Typical	8.8	10.7
Dyspnea	12.8	14.6	0.44

CAD Characteristics on coronary CTA

Extent and severity of CAD as observed on coronary CTA in patients with and without hypertension are shown in Table 2. Regarding plaque extent and severity stenosis, compared to the no-HT patients, HT patients manifested a greater SIS, and a lower prevalence of absent plaque. HT patients possessed greater prevalence of obstructive lesions in 1-, 2-, or 3-vessels (p<0.001). HT patients had more ≥50% stenosis in the proximal and mid coronary arteries and side branches. Regarding plaque characteristics, any NCP or CP was more observed in HT patients compared to no-HT patients [Table 2].

Table 2.

CAD Characteristics on CCTA (n=2868)

CAD Characteristics on CCTA	No hypertension (n=1434)	Hypertension (n=1434)	P value
SIS (Median; IQR)	0 (0, 2)	1 (0, 3)	<0.0001
Number of vessels with plaque and ≥50% stenosis (%)
No plaque	53.0	43.7	<0.001
Non-obstructive plaque (1–49%)	33.3	37.7
1-vessel disease (≥50%)	9.5	12.1
2-vessel disease (≥50%)	2.6	4.4
3-vessel disease (≥50%)/Left main	1.6	2.2
Distribution for any coronary artery disease (%)
Left main	12.6	14.1	0.24
Proximal	38.2	46.6	<0.001
Mid	29.3	36.0	<0.001
Distal	14.1	18.6	0.002
Side branches	14.7	18.4	0.01
Distribution for coronary artery disease with ≥50% stenosis (%)
Left main	0.3	0.6	0.18
Proximal	7.5	10.0	0.02
Mid	8.2	10.3	0.06
Distal	3.4	4.6	0.12
Side branches	4.7	7.0	0.01
Plaque characteristics (%)
Non-calcified plaque	13.2	17.8	0.001
Partially calcified plaque	19.5	22.2	0.08
Calcified plaque	24.3	28.2	0.02

Abbreviations: SIS- segment involvement score,

MACE risk

One hundred eighty patients (6.3% of study population) experienced MACE at a mean follow-up of 5.2±1.2 years, occurring in 104 patients of the HT group and 76 patients of the no-HT group (42 deaths, 34 non-fatal MI) (7.3% vs. 5.3%, p=0.03) (Table 3). Kaplan-Meier curve demonstrated that MACE were more common in the HT vs. the no-HT subjects (p<0.01) (Figure 1). By Cox proportional analysis, HT subjects experienced higher MACE risk than no-HT subjects (HR 1.4, 95% CI 1.0–1.9, p=0.03). On a sub-analysis by gender, in both of men and women, MACE tended to be more common in the HT vs. the no-HT subjects (HR 1.4, 95% CI 0.9–2.0, p=0.12 for men, HR 1.4, 95% CI 0.9–2.3, p=0.14 for women).

Table 3.

MACE risk and Predictors of MACE

MACE risk
MACE	No hypertension (n=1434)		Hypertension (n=1434)			P value
MACE (%, n)	5.3 (76)		7.3 (104)			0.03
Deaths (%, n)	2.9 (42)		3.9 (56)
Non-fatal MI (%, n)	2.4 (34)		3.4 (48)
Predictors of MACE
Models	cNRI (95% CI)	P value	%Events reclassified	%Non-Events reclassified	AUC	AUC p value
Stenosis severity of CAD
Model 1	–	–	–	–	0.648
Model 2 (vs. Model 1)	0.39 (0.24–0.54)	<0.0001	14%, p=0.053	24%, p<0.0001	0.729	<0.0001
Model 3 (vs. Model 2)	0.17 (0.02–0.32)	0.03	16%, p=0.04	1%, p=0.59	0.734	0.055
Extent of CAD
Model 4	–	–	–	–	0.678
Model 5 (vs. Model 4)	0.41 (0.26–0.56)	<0.0001	17%, p=0.03	24%, p<0.0001	0.721	0.001
Model 6 (vs. Model 5)	0.17 (0.02–0.32)	0.03	16%, p=0.04	1%, p=0.59	0.726	0.052

Model 1: Stenosis severity ≥50%

Model 2: Age, gender, other risk factors +stenosis severity ≥50%

Model 3: Age, gender, other risk factors +stenosis severity ≥50%+HT

Model 4: SIS

Model 5: Age, gender, other risk factors + SIS

Model 6: Age, gender, other risk factors + SIS +HT

Abbreviations: “other risk factors” =diabetes, dyslipidemia, smoking, family history, chest symptoms; cNRI= continuous net reclassification index; AUC=area under the curve; CAD- coronary artery disease, HT=hypertension; SIS- segment involvement score;

Figure 1.

Kaplan-Meier curve for MACE in patients with no known CAD among the absence and presence of hypertension

Abbreviations: MACE- major adverse cardiac events, CAD- coronary artery disease.

Considering the CTA findings, the risk of MACE was progressively higher in the subjects with non-obstructive CAD and those with obstructive CAD when compared to those with no-CAD (Figure 2a). A trend toward a higher odds ratio was observed among HT patients with normal coronary arteries; however, it did not reach statistical significance (HR 1.9, 95% CI 1.0–3.6, p=0.06). In patients with non-obstructive CAD, the hazard ratios of the HT and no-HT groups were similar. In the obstructive CAD group, the hazard ratio in the HT group was only slightly higher than that in the no-HT group (Figure 2a).

Figure 2.

a (left). Cox proportional hazard models by normal, non-obstructive and obstructive CAD among no-hypertension and hypertension groups

Abbreviations: CAD- coronary artery disease.

b (right). Cox proportional hazard models by SIS 0, SIS 1–5 and SIS >5 among no-hypertension and hypertension groups

Abbreviations: CAD- coronary artery disease. SIS- segment involvement score

Regarding the extent of CAD, similar findings were observed. The risk of MACE was also progressively higher in the subjects with SIS 1–5 and those with SIS >5 when compared to those with no-CAD. The hazard ratios of the HT group in the SIS 1–5 and SIS >5 groups were only slightly higher than the no-HT group (Figure 2b).

The incremental added value of hypertension in prediction of MACE over clinical and CTA variables is shown in Table 3. The presence of obstructive CAD was predictive of MACE (Model 1, AUC 0.648). The combination of “other clinical factors” increased this prediction (Model 2; p<0.0001 for cNRI and p<0.0001 for AUC). When HT was then added (Model 3), there was a significant increase in cNRI (p=0.03), and a trend toward increase in the AUC (p=0.055). Similar results were observed with respect to the extent of CAD. The SIS alone was predictive MACE (Model 4, AUC 0.678). The combination of “other clinical factors” increased this prediction (Model 5; p<0.0001 for cNRI and p=0.0001 for AUC). When HT was added (Model 6), a significant increase in cNRI (p=0.03) and a trend toward increase in the AUC were observed (p=0.052).

DISCUSSION

This current study demonstrated that patients with hypertension had more advanced coronary atherosclerosis by coronary CTA and future MACE risk compared to those without hypertension. When stratifying by gender, there was a trend toward increased MACE risk in HT in both men and women. Patients with hypertension more frequently had any CAD, with higher prevalence of any CAD and of 1, 2 and 3 vessel obstructive CAD in each category, as well as CAD ≥50% stenosis in the proximal, mid and side branches. HT patients also had greater prevalence of any NCP and CP. When stratifying by the extent and severity of CAD, MACE risk in the HT group was slightly higher than that in the non-HT group across the CAD categories. There was a trend toward incremental predictive value of HT over other risk factors and the extent or severity of CAD.

It is well-known that hypertension is a main cardiovascular risk factor and related to worsening prognosis^{4, 5, 7–9}. To our knowledge, however, no prior studies have shown the direct relation of hypertension to CAD characteristics including the presence, extent, and severity of CAD on coronary CTA and MACE risk that is described in this manuscript.

Using the current registry, our group previously reported the relationship between diabetes and current smoking and the presence, extent, and severity of coronary atherosclerosis on coronary CTA, as well as the relationship of the coronary CTA findings to future adverse outcomes^{23, 24}. The findings of the current study suggest that the presence of hypertension per-se may not add as much incremental prognostic value as these other risk factors after taking into account the presence, extent, and severity of coronary CTA.

Hypertension has been previously shown to be a predictor of the extent of coronary atherosclerosis as assessed by CAC²⁵. Both diabetes and smoking were both found to be stronger predictors of coronary atherosclerosis than hypertension. Extensive epidemiologic data has also demonstrated that hypertension is an independent risk factor for coronary atherosclerosis and for future cardiac events^{26, 27}, but that it is less strong a predictor than diabetes and smoking²⁵. Our findings are concordant with the prior data with respect to coronary atherosclerosis including a relationship to increasing amounts of obstructive CAD and with showing a trend toward association with MACE events.

How hypertension results in increase in coronary atherosclerosis has been extensively studied. The principal underlying pathophysiologic mechanism is considered to a mechanical one, related to pulse pressure²⁸. Wide pulse pressure has been reported to be associated with increased cardiac events^29–33. Both increased pulse pressure and systolic pressure contribute to endothelial dysfunction, which facilitates the entry of low density lipid cholesterol into the blood vessel wall, initiating the atherosclerotic process³⁴. Hypertension also is a cause of left ventricular hypertrophy (LVH)^{35, 36}, which has been implicated as a cause of coronary atherosclerosis, myocardial infarction, arrhythmia, cardiac failure or cardiac death^{34, 37–39}. LVH is associated with collagen deposition within the left ventricle³⁴. This process is considered to explain the frequent association of LVH with mid-myocardial scarring on cardiovascular magnetic resonance, which is associated with increase in cardiac events⁴⁰.

Of interest, in the current study, in patients with no evidence of CAD on coronary CTA, those with hypertension had a two-fold greater MACE risk compared to no-HT. An increased risk of events in these patients could have been related to LVH; however, information regarding LVH was not present in the CONFIRM database.

Limitations

There are several limitations in this study. Data regarding duration and severity of HT, as well as information regarding LVH did not exist in the current registry. Information was not uniformly available regarding specific antihypertensive medications at the time of testing. Further, no information was available regarding effectiveness of blood pressure control following testing which may have affected MACE risk⁴¹. We have included various CAD descriptors, including the extent, stenosis severity, basic characteristics and location of CAD. However, other variables such as vulnerable plaque features or bifurcation lesions that might be associated with MACE risk were not available in the current study. The number of events by gender was small and may have led to the finding a trend toward increase MACE in each gender that did not reach statistical significance.

Perspectives

Compared to patients without hypertension, hypertensive patients have increased presence, extent, and severity of coronary atherosclerosis and tend to have an increase in MACE events. The findings support the concept of lifestyle modification regardless of gender to optimize CAD risk factors, including hypertension, in order to reduce future cardiovascular events as suggested by current guidelines⁴².

CONCLUSION

Hypertensive patients had greater amount of coronary atherosclerosis and greater risk of MACE risk compared to non-hypertensive patients, independent of other clinical risk factors as well as of the presence of obstructive CAD or extent of CAD. Further, hypertensive individuals with an increasing degree of CAD stenosis severity and extent of CAD experienced modestly increase rates of MACE compared to non-hypertensive patients.

Novelty and Significance.

What is new

• This study is the first study showing the relation of hypertension to the presence, extent, and severity of CAD on coronary CTA and to risk of MACE among patients without known CAD.

What is relevant

• The presence of hypertension per-se may not add as much incremental prognostic value as other risk factors after taking into account the presence, extent, and severity of coronary CTA.

Summary

• Patients with hypertension had greater prevalence of extent and stenosis severity of CAD as well as modestly increased MACE risk compared to those without hypertension.