The prediction of cardiovascular events by phenotype of central and peripheral blood pressure in subjects without hypertension

Introduction

High blood pressure (BP) has doubled worldwide since 1990, according to a report based on large data [1]. In addition, the importance of high BP is continuously emphasized because controlling BP with low cost is effective in preventing cerebrovascular accidents as well as vascular diseases such as ischemic heart disease and kidney disease [2]. The report also emphasized that the diagnosis rate or treatment rate is lower than the increasing prevalence rate. According to a fact sheet released in Korea in 2022, it is higher than the global average, but about 30% does not recognize hypertension (HBP), and the treatment rate tends to decrease [3]. There are several hurdles and limitations contributing to this. Depending on a number of factors, an individual’s BP status can change significantly [4]. However, office BP has limitations in evaluating such BP variability. It has recently been emphasized that BP variability is more important for targeting organ damage [5]. Even if a doctor tries to proceed with 24-hour ambulatory BP monitoring (ABPM), there are hurdles in examination cooperation, such as lack of patient insight.

One of the out-of-office BP measurements, central blood pressure (CBP), has long been considered a better predictor than brachial BP for CV event risk, and meta-analyses have been sufficiently conducted accordingly [6]. In addition, the non-invasive central hemodynamics measurement method was sufficiently validated [7, 8]. Given the anatomical proximity of the heart, brain, and kidney to the central arteries and the role of arterial stiffness in cardiovascular disease, CBP is thought to be more closely related than brachial BP. In addition, there have been reports that CBP may be different due to the characteristics of blood vessels even in people with the same brachial BP [9]. In recent meta-analysis, patients with increased CBP but normotensive peripheral BP had increased CV events [10]. However, CBP has not yet been included in the guidelines for BP management. Since the SPRINT trial emphasized the advantageous aspect of tight BP control [11], it may be necessary to focus on finding patients who need more aggressive management that can be overlooked only by office BP measurement. In addition, it is still controversial whether assessment of CBP is necessary even in the presence of normal brachial BP.

Therefore, in this study, we evaluate the role of CBP by measuring brachial systolic BP (SBP) and CBP in patients who have never been treated for HBP before. Also, we investigate the prediction of CV events by phenotype of central and peripheral BP in subjects without hypertension.

Method

Results

The median follow-up period for enrolled patients was 7.5 years, with an average age of 52.5 ± 13.7 years. Of the 720 patients in the study group, 47.5% were men and 52.5% were women. The average body mass index (BMI) was 23.75 ± 3.2 kg/m², and 21 patients (2.9%) had primary endpoint events during the follow-up period. One of them died of cardiac death, 9 patients were diagnosed with vascular disease including ACS and coronary reperfusion, and 6 experienced stroke, TIA, or brain hemorrhage. Heart failure occurred in 5 patients.

Table 1 shows the baseline characteristics of the participants by classifying them into patients with and without a primary endpoint event. There was a statistically significant difference between the two groups only in age. In addition, CBP, SBP, pulse pressure (PP), and central PP (CPP) were statistically significantly different between the two groups.

Table 2 presents the results of univariate Cox proportional hazards ratio models for each variable. Table 3 shows a multivariable Cox proportional hazards model adjusted for age and smoking. After adjustment, statistically significant increases in risk of 30% for CBP and 40% for SBP were confirmed, as was an increase in risk of about 40% for PP, but CPP did not acquire statistical significance.

Figure 1 shows a restricted cubic spline curve for the risk of CV event occurrence according to the increase in CBP in subjects without HBP. In this analysis, the point considered to be the optimal CBP was 110 mmHg. In addition, the risk of CV event occurrence according to the increase in BP showed a J-curve pattern.

• Download figure
• Open in new tab

Figure 1.

Restricted cubic spline curve of CBP; the lowest CBP of cumulative incidence of the primary outcome was 110 mmHg.

In Figure 2, using a categorical approach, CBP at 120mmHg and SBP at 140mmHg might be divided into normotensive and hypertensive levels based on our previous study and Korea guideline [3, 13-15]. Group 1 (isolated central hypertension group, consisting of high CBP and normal SBP) had a total of 162 patients, with average values for age, BMI, CBP, SBP, DBP, PP, and CPP of 57.2 ± 10.8 years, 24.2 ± 3.1 kg/m², 126.4 ± 4.5 mmHg, 132.1 ± 4.8 mmHg, 81.0 ± 7.9 mmHg, 51.0 ± 8.0 mmHg, and 45.5 ± 8.6 mmHg, respectively. There were 116 people in Group 2 (both systolic hypertension group, consisting of high SBP and high CBP). The average values for age, BMI, CBP, SBP, DBP, PP, and CPP were 58.0 ± 12.0 years, 24.1 ± 3.18 kg/m², 140.8 ± 11.5 mmHg, 150.3 ± 9.1 mmHg, 87.1 ± 11.3 mmHg, 63.2 ± 12.6 mmHg, and 53.6 ± 14.1 mmHg, respectively. However, there was no difference between the two groups in baseline characteristics. Isolated central systolic hypertension and both systolic hypertension showed 4.9% and 6% of the CV event rate, respectively. There was no difference between the two groups (p=0.897). Group 3 (consisting of normal SBP and CBP) had a total of 432 patients, and the average values for age, BMI, CBP, SBP, DBP, PP, and CPP were 49.6 ± 14.0 years, 23.3 ± 3.2 kg/m², 105.5 ± 9.6 mmHg, 117.0 ± 10.2 mmHg, 72.0 ± 8.4 mmHg, 45.0 ± 8.5 mmHg, and 33.5 ± 7.9 mmHg, respectively. Group 4 (normal CBP and high SBP) consisted of a small number of 10 people.

• Download figure
• Open in new tab

Figure 2.

CBP corresponding to each SBP is represented as a dot; SBP and CBP are in a direct proportional relationship

Discussion

This study found that CVD risk according to CBP in patients without HBP had a J-shaped curve, and the lowest level of CBP was 110 mmHg. Isolated central hypertension and both systolic hypertension showed 4.9% and 6% of the CV event rate, respectively.

Generally, those who have never been diagnosed with HBP and have never taken antihypertensive drugs may have fewer CV events than hypertensive patients. According to a fact sheet released in Korea in 2022 [3], about 30% does not recognize HBP. In our study, 17.5% of those without recognizing high BP diagnosed as hypertension when 140 mmHg was the cut-off value of peripheral BP. In addition, it was investigated that CV events occurred in 6.3%, so it seems that not a few events occurred.

In clinical practice, BP is routinely measured at the brachial artery. However, follow-up in hypertensive patients by also measuring CBP can be more advantageous as a predictor of CV disease, as seen in the Café study [16]. In addition, it is thought that CBP may be more advantageous than brachial BP, which is mainly measured at the office, in evaluating target organ damage in normotensive patients [17, 18]. Therefore, our previous study showed that CV risk increased rapidly at a CBP of 120 mmHg or higher through previously studied CBP data, irrespective of the brachial BP status [13-15]. In a recent meta-analysis, optimal CBP and central hypertension were defined as less than 110 mmHg and 120 mmHg, respectively, based on the primary CV endpoint with a 5-year probability of cohort data [10]. Similar to the previous study, the lowest CBP was 110 mmHg among the cases of CV events in our study. In addition, the threshold of central hypertension was categorized at 120 mmHg, and in this case, the incidence of CV events tended to increase regardless of brachial SBP.

In our study, the CV event rate was 6% in patients with both systolic hypertension who needed conventional hypertensive treatment, whereas the CV event rate in patients with isolated central hypertension, who were not subject to aggressive hypertension treatment, was 4.9%. In the previous study, the cumulative CV events increased gradually from concordant central and brachial normotensive over isolated brachial hypertension to isolated central hypertension and onward to concordant central and brachial hypertension with no difference between isolated central hypertension and concordant hypertension, although 130 mmHg was the cut-off value for peripheral BP. Based on these results, isolated central hypertension should not be overlooked even if brachial BP is normal [17, 19].

A study found that most patients with isolated systolic hypertension (ISH) and normal CBP had white coat hypertension (WCH) [20] [21], so these patients meet the diagnostic requirements for WCH and may require ABPM. However, if the ABPM test is not available or if a young patient with a high BMI has high BP, evaluation with CBP may be necessary rather than using many drugs. The tendency of CBP to be inconsistent with brachial SBP is probably because central hemodynamic measures show a closer correlation with vascular aging, and vascular aging-associated hypertension has been reported in Asia. Based on this, it is thought that this tendency was also reflected in this study [22, 23].

In our study, there is a limitation that ABPM was not performed. It is thought that more information would have been obtained by classifying patients with masked hypertension and WCH through ABPM [24-26] and analyzing their CBP patterns. This is because our study is a retrospective observational study, and the number of participants was insufficient because we excluded those who were diagnosed with hypertension or who were taking antihypertensive medications for other reasons. In addition, aggressive evaluation was not conducted on participants who met these criteria. In future studies, survival analysis should be considered with a longer study period and a larger number of participants.

Conclusion

In this study, irrespective of brachial BP status, isolated central hypertension increased CV events. In addition, CBP not only acts as a predictor in CV outcomes, but is also thought to play a role in risk stratification of patients, which may be insufficient with brachial BP measurement. Therefore, to prevent CV events, it is essential to control not only peripheral BP but also CBP.