Arterial stiffness is related to augmented seasonal variation of blood pressure in hypertensive patients

Abstract

Background. Seasonal variation in blood pressure (BP), a usual tendency of both systolic (SBP) and diastolic BP (DBP) to rise during winter in hypertensive patients, may be related to the higher cardiovascular mortality in winter. However, it is not yet clear what factors are relevant to the seasonal BP changes. We hypothesized that arterial stiffness is related to the BP changes between summer and winter. Methods and results. Eighty‐five elderly (>55 years) patients with essential hypertension (33 males, 64±6.0 years) were enrolled. Seasonal BP profiles over at least 2 years were studied along with arterial stiffness and clinical variables (age, gender, smoking, duration of hypertension, anti‐hypertensive medications and body mass index). Both SBP and DBP were significantly higher during winter compared with three other seasons (spring 128±10.0/79±7.3 mmHg, summer 127±9.8/78±7.1 mmHg, autumn 127±10.3/78±8.0 mmHg, winter 136±12.5/81±7.6 mmHg; SBP changes; p<0.001, DBP changes; p<0.001). There were no significant seasonal differences among spring, summer and autumn. Pulse wave velocity (PWV), a widely used clinical indicator of arterial stiffness was correlated with winter–summer differences in SBP (r = 0.272, p = 0.012), but not in DBP (r = 0.188, p = 0.085). Age, which was correlated with PWV strongly (p<0.001), was not significantly related to the seasonal changes in BP (SBP changes; p = 0.114, DBP changes; p = 0.298). No other clinical variables had significant correlation with seasonal BP changes. Multivariate regression analysis revealed that PWV is the only significant predictor for winter–summer SBP changes. Conclusions. Our results established a feasible link between arterial stiffness and seasonal BP variation. These findings may partly explain higher cardiovascular risk in patients with increased arterial stiffness.

• Arterial stiffness
• hypertension
• seasonal variation

Introduction

Seasonal changes in cardiovascular disease have been well documented, usually with an increase in winter [1–3]. Acute myocardial infarction exhibits a seasonal pattern with an increased incidence during winter. Data from the US National Registry of Myocardial Infarction report approximately 53% more cases in winter than during the summer, and in‐hospital case fatality rate shows a peak of 9% in winter [1]. Moreover, a winter peak in occurrence has been reported for ischemic and hemorrhagic stroke [2], and rupture or dissection of aortic aneurysms as well [3].

Several factors may play roles in the tendency of increased cardiovascular events in winter. Cold exposure, for example, results in an increase of sympathetic activity and blood pressure (BP) levels. Besides, a significant negative correlation between ambient temperature and BP has been reported [4]. Winter is also characterized by pro‐thrombotic state. Fibrinogen levels show wide seasonal variation, increasing up to 23% during the colder months [5]. Furthermore, a mild surface cooling can increase platelet and red cell count, and consequently blood viscosity [6]. Probably, such an unfavorable constellation of underlying factors may play a role in the increased cardiovascular disease in winter.

Among these, augmented seasonal variation of BP is one of the most important contributing factors to increased cardiovascular events. Seasonal variation of BP has been documented in normal population [7], in hypertensive patients [4], [8], in the elderly [9], [10], and in dialysis patients [11]. Ambient temperature [7], [10], age [9], body mass index (BMI) [9], [12], smoking [13], and interdialytic body weight gain [11] are reported to be related to the seasonal variation of BP. However, there is no single important affecting factor underlying the seasonal variation of BP, which is consistently observed in various studies [4], [7], [9–13].

Recently, arterial stiffness, which can be evaluated by several measurements including pulse wave velocity (PWV), has been shown to be a strong independent predictor of cardiovascular and all‐cause mortality in patients with essential hypertension[14], and in patients with older subjects over 70 years [15], as well as in patients with end‐stage renal disease on hemodialysis [16]. However, underlying mechanisms between arterial stiffness and increased cardiovascular mortality are not well known. Recent studies [17], [18] suggested possibilities that arterial stiffness is related to the variation of BP. Moreover, BP variation is more easily observed in elderly patients, in whom arterial is known to be increased. Therefore, we hypothesized that arterial stiffness is related to augmented seasonal variation of BP, and subsequently related to increased cardiovascular events during winter.

Methods

The subjects analyzed were 85 hypertensive elderly patients (>55 years) who were on regular follow‐up at Severance Hospital, Yonsei University College of Medicine between April 2002 and March 2005. Hypertension was defined as systolic BP (SBP) >140 mmHg or diastolic BP (DBP) >90 mmHg, according to the Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure: the JNC 7 report [19]. Definition of each four seasons is as follows; spring (March–May), summer (June–August), autumn (September–November), winter (December–February). The inclusion criteria were: (i) patients who were followed regularly every 3 months over 2 years who can show the seasonal profile of the BP; (ii) patients at age 55 years or more. Exclusion criteria were: (i) recent myocardial infarction, unstable angina, congestive heart failure, previous cerebrovascular accident, malignant debilitating disease, severe respiratory disease, renal failure (creatinine>1.4 mg/dl), anemia (hemoglobin<12 g%), secondary hypertension, hyper‐ or hypothyroidism and severe dementia; (ii) patients with any kind of changes in hypertensive medication during the follow‐up period, which can influence the seasonal profile of BP

After the patients had sat down and rested for 15 min, two consecutive readings were recorded by a mercury sphygmomanometer with a standard‐sized cuff (12×35 cm), and their mean values were calculated. The average indoor temperature was maintained by an air conditioner at 23–25°C, irrespective of outdoor environment.

Arterial stiffness was estimated by brachial‐ankle PWV (baPWV, cm/s), which was measured using a volume‐plethysmographic apparatus (from PWV/ABI, VP‐2000®; Colin, Co., Ltd., Komaki, Japan) as described previously [20]. Briefly, four cuffs matched with oscillometric sensors were wrapped around the upper arms and ankles, and then the pulse volume records of the bilateral brachial and tibial arteries were monitored during a continuous deflation of the cuffs. BP of each lesion could be obtained by this oscillometric method. Electrodes of the electrocardiograph were placed on both wrists, and a microphone was placed on the left edge of the sternum to detect heart sounds. Transit time between the brachial and ankle pulse waves was automatically measured from the time delay between the feet (sharp initial systolic upstroke) of the wave at the two sites. The distance between the two recording sites of baPWV was calculated automatically using the height of the patient.

All the baPWV measurements are made after the BP were stabilized at least 2 weeks after starting or changing the antihypertensive medication, because the PWV is highly influenced by the actual BP itself. Those patients could be enrolled only if they are followed up over 2 years after starting or changing the medication. Moreover, estimation of arterial stiffness using volume‐plethysmographic apparatus was done only during the winter except for four patients. To avoid interobserver variation, one experienced examiner measured all baPWV of the analyzed patients.

Age, gender, smoking, BMI, duration of hypertension, types of anti‐hypertensive medications, laboratory data and echocardiographic data were analyzed. The statistical analysis was performed using SPSS statistical analysis program (ver. 12.0, SPSS Inc., Chicago, IL, USA). Values are expressed as mean±SD. Data were analyzed by independent t‐test, paired t‐test, correlation analysis and multivariate linear regression. A p‐value of <0.05 was considered statistically significant. The study was approved by the Ethics Committee of Yonsei University, and informed consent was obtained from all participants.

Results

Demographic data such as age, sex, smoking, duration of hypertension, body weight, BMI, and types of hypertensive medications, laboratory findings and echocardiographic data are shown in Table I. Mean and SD of measured baPWV were 1589±286.2 cm/s.

Table I. The characteristics of the study group.

Characteristics	Values
Demographic data
Age	64±6.0
Sex	33:52
Smoking	2 (2.4%)
Dyslipidemia	32 (37.6%)
Bodyweight (kg)	63±9.2
BMI (kg/m^2)	24.8±2.70
Duration of HTN (years)	11±9.3
Types of hypertensive medications
Calcium channel blocker	35 (41.2%)
Diuretics	24 (28.2%)
Angiotensin receptor blocker	18 (21.2%)
Beta blocker	15 (17.6%)
ACE inhibitor	9 (10.6%)
Nitrate	2 (2.4%)
Laboratory findings
Hb (g/dl)	13.7±1.18
BUN (mg/dl)	15.8±3.90
Cr (mg/dl)	0.9±0.9
Total cholesterol (mg/dl)	192±34.5
Triglyceride (mg/dl)	160±89.1
HDL‐cholesterol (mg/dl)	50±11.6
LDL‐cholesterol (mg/dl)	117±27.2
hsCRP (mg/l)	0.9±0.77
Echocardiographic data
LVEF (%)	72±6.3
LVEDD (mm)	48.1±4.04
LVESD (mm)	29.5±3.98
LA (A‐P) (mm)	36.2±4.83
LA Vol. index (ml/m^2)	25.1±9.54

Data are mean±SD. Dyslipidemia – LDL>130 mg/dl or TC>200 mg/dl or HDL‐C<40 mg/dl or TG>150 mg/dl; BMI, body mass index; HTN, hypertension; ACE, angiotensin‐converting enzyme; Hb, hemoglobin; BUN, blood urea nitrogen; Cr, creatinine; HDL, high‐density lipoprotein; LDL, low‐density lipoprotein; hsCRP, high‐sensitivity C‐reactive protein; LVEF, left ventricular ejection fraction; LVEDD, left ventricular end diastolic dimension; LVESD, left ventricular end systolic dimension; LA(A‐P), antero‐posterior diameter of left atrium; LA Vol. index, left atrial volume index.

We took the mean of four seasonal BP profiles over at least 2 years. The results of BP measurements during the seasons are shown in Table II.

Table II. Seasonal systolic blood pressure (SBP) and diastolic blood pressure (DBP): Absolute values and their differences (Δ).

		BP, mean±SD	BP range (mmHg)
SBP	Spring	128±10.0	93–145
	Summer	127±9.8	99–150
	Autumn	127±10.3	98–155
	Winter	136±12.5	101–160
DBP	Spring	79±7.3	57–90
	Summer	78±7.2	55–98
	Autumn	78±8.0	45–95
	Winter	81±7.6	55–96
		BP, mean±SD	p‐value
ΔSBP	Winter–summer	9±10.7	<0.001
	Winter–spring	8±10.9	<0.001
	Winter–autumn	9±11.4	<0.001
	Summer–spring	1±10.4	0.303
	Summer–autumn	0±10.6	0.789
	Spring–autumn	0±9.0	0.382
ΔDBP	Winter–summer	3±7.3	<0.001
	Winter–spring	2±6.8	0.002
	Winter–autumn	3±8.0	<0.001
	Summer–spring	0±7.3	0.222
	Summer–autumn	0±8.3	0.941
	Spring–autumn	1±7.8	0.225

The winter–summer DBP difference was 3±7.3 mmHg (p<0.001). The winter–spring DBP was 2±6.8 mmHg (p = 0.002) and the winter–autumn DBP was 3±8.0 mmHg (p<0.001). There were no significant seasonal differences between spring–autumn, summer–spring and summer–autumn DBP differences.

We analyzed the correlation of the seasonal BP changes with age, BMI and PWV. PWV was correlated with winter–summer differences in SBP changes (r = 0.272, p = 0.012) (Figure 1), but not in DBP changes (r = 0.188, p = 0.085). Age, which was correlated with PWV strongly (p<0.001), was not significantly related to the seasonal changes in BP (SBP changes; p = 0.114, DBP changes; p = 0.298). While PWV showed high correlation with actual BP (SBP; r = 0.384, p<0.001, DBP; r = 0.318, p = 0.003), seasonal BP changes were not influenced by actual BP itself. No other clinical variables, including laboratory findings, duration of hypertension, types of antihypertensive medication and echocardiographic data, had a significant correlation with seasonal BP changes.

Figure 1 Correlation analysis of pulse wave velocity(PWV) with the winter‐summer systolic blood pressure (SBP) changes.

Multivariate regression analysis taking seasonal BP changes as a dependent variable revealed that PWV was the only predicting factor for seasonal SBP changes (Table III).

Table III. Multivariate regression analysis.

Dependent variable; winter–summer SBP changes, adjusted R square; 0.049
	β	Standard error	p‐value
Age	0.129	0.214	0.549
BMI	0.301	0.424	0.480
PWV	0.009	0.004	0.045
Dependent variable; winter–summer DBP changes, adjusted R square; 0.024
	β	Standard error	p‐value
Age	0.066	0.147	0.655
BMI	0.403	0.291	0.170
PWV	0.004	0.003	0.167

SBP, systolic blood pressure; DBP, diastolic blood pressure; BMI, body mass index; PWV, pulse wave velocity.

Discussion

We presented the relationship between arterial stiffness and seasonal BP variation in hypertensive elderly patients. Multivariate regression analysis taking SBP and DBP changes as a dependent variable with age, and BMI and PWV as an independent factor revealed that PWV was the only predicting factor for seasonal SBP changes. Our results established a feasible link between arterial stiffness and seasonal BP variation.

Seasonal variation of BP, which was previously found in various subsets of patients, is also well documented in our study group. We found both SBP and DBPs were significantly higher during winter compared with three other seasons. The degree of seasonal differences in BP that we found were similar to those reported in previous studies on hypertensives or elderly populations [4], [8], [10].

Increased arterial stiffness was related to the winter–summer SBP changes, but not to the DBP changes in our study. These results can be explained by the inclusion criteria of over 55‐year‐old patients. SBP increases steadily with age, whereas DBP increases until about age 55 and then it declines [21]. As the degree of SBP rise with aging usually outweighs the degree of DBP fall [21], the relationship between arterial stiffness and seasonal BP variation can be more clearly manifested in SBP changes rather than in DBP changes.

Age [9], BMI [9], [12] and smoking [13], which were previously known to be related to the seasonal variation of BP, were not shown to be related in our study. As for the smoking, the fact that smoking results in peripheral vasoconstriction may be related with the augmented BP response in winter. However, because the number of smoker in the study group was relatively small – only two – the effects of smoking on BP variations could be masked in our results. Kristal‐Boneh et al. [12] reported the inverse relationship between BMI and seasonal SBP changes, which means the augmented seasonal variation of BP in leaner patients. However, probably due to the differences of study population – lower BMI in our study group compared with previous study [12] – there was no significant relationship between BMI and seasonal variation of BP.

It should be noted that age, BMI and smoking, which were known to be related to the seasonal variation of BP, are also influencing factors on the arterial stiffness itself [22], [23]. Even though the mechanisms of how increased arterial stiffness is accompanied by the augmented seasonal BP changes are not yet clear, there are possibilities that many of previously known influencing factors are involved in the stiffening of the arterial conduit system first, and then in the seasonal variation of BP subsequently.

Epidemiological and clinical studies have shown that increased pulse pressure is an independent cardiovascular risk factor in general population [24], [25]. Pulse pressure is influenced by left ventricular ejection (stroke volume and ejection time) and arterial stiffness (principally that of the aorta and large central arteries). Because left ventricular ejection remains stable or even decreases with age, arterial stiffness is the principal factor responsible for increased pulse pressure in various subsets of patients with increased cardiovascular risk [24]. Recently, arterial stiffness itself, which can be evaluated by several measurements including PWV, has been shown to be a strong independent predictor of cardiovascular and all‐cause mortality in patients with essential hypertension [14] and with age over 70 years old [15], as well as in patients with end‐stage renal disease on hemodialysis [16]. However, underlying mechanisms between arterial stiffness and increased cardiovascular mortality are not well known yet. BP variation, which was known to be related to the increased cardiovascular event [26], could be the bridging point, which can explain the increased cardiovascular mortality in patients with increased arterial stiffness. Here we presented the feasible mechanism of increased cardiovascular mortality in patients with increased arterial stiffness by demonstrating the relationship between arterial stiffness and seasonal variation of BP.

These findings may give a meaningful contribution to the treatment of elderly hypertensive patients. Physicians should not overlook the BP surge during winter as an insignificant common finding. Based on our results, we can identify high‐risk patients who require more strict BP control during winter. To reduce the cardiovascular event, it is necessary to treat more carefully during winter, especially in high‐risk patients with increased arterial stiffness. Moreover, even for the normotensive patients, it is important to assess hypertension during winter, but not in other seasons, if they have increased arterial stiffness or other cardiovascular risk factors.

The present study has several potential limitations. First, our study population was relatively small. There were difficulties recruiting a large number of patients due to the relatively long follow‐up period of over 2 years and the strict exclusion criteria of our study. Second, PWV is influenced by BP for the times when PWV is measured. Even though all the PWV measurements were made after the BP was stabilized, the relationship between PWV and BP at the time of PWV measurements could not be overlooked. Third, PWV measurements using volume‐plethysmographic apparatus can pose a problem of reproducibility. However, it is well documented that intraobserver–intersession reproducibility of PWV measurements is very high, provided that the measurements are done by a single examiner [27]. In this study, one experienced examiner measured all the baPWV of analyzed patients. Moreover, estimation of arterial stiffness was done only during winter except four patients. However, prospective studies in a large population would be needed to reveal more clear relationship among arterial stiffness, augmented seasonal BP variation and major adverse cardiovascular event.

In conclusion, our study of 85 elderly hypertensive patients revealed a significant increase of BP in winter. In this population, arterial stiffness, which was known as an independent cardiovascular risk factor, was found to be related to augmented seasonal variation of BP. These findings may partly explain the higher cardiovascular risk in patients with increased arterial stiffness.