Comparison of Pulse Wave Analysis Between Persons With White Coat Hypertension and Normotensive Persons

Abstract

White coat hypertension (WCH) is considered by some but not all investigators to be a benign condition without increased cardiovascular risk. Pulse wave analysis is a noninvasive method to measure how the reflected pressure wave interacts with central aortic blood pressure (BP) and to assess how it is related to vascular stiffness. The purpose of the study was to compare central aortic BP in normotensive and WCH participants. WCH participants were identified after ambulatory BP monitoring. Normotensive participants served as controls. Using radial artery applanation tonometry, aortic pulse wave analysis was performed. Augmentation index (AI), AI₇₅, and differences in systolic BP between central aortic and peripheral vasculatures were calculated. Results show a difference in AI, AI₇₅, (AI standardized to a heart rate of 75 beats per minute), and central aortic systolic pressures between WCH and normotensive participants. The WCH group had significantly higher systolic BP and pulse pressure; however, these were still within the normal range. In summary, WCH participants had increased central aortic pressures compared with normotensives, supporting the potential for increased cardiovascular risk in WCH.

White coat hypertension (WCH) is defined as elevated office blood pressure (BP) measurements when BP is repeatedly normal at home, at work, or by ambulatory BP monitoring (ABPM). Although repeated home BP measurements have been used, 24‐hour ABPM is the traditional method to identify individuals with WCH. Studies suggest that the risk of cardiovascular (CV) events correlates more closely with ABPM than with office BP measurements. ¹ Several of these studies suggest no increase in CV morbidity or mortality. ² , ³ , ⁴ , ⁵ Other research reports, however, that individuals with WCH may be at increased risk for eventual sustained hypertension or CV disease. ¹ , ⁶ Pulse wave analysis (PWA) is a noninvasive method that measures pulse wave reflections and their effects on central aortic pressure; it is considered an indirect measure of vascular stiffness. ⁷ PWA utilizes the radial artery pulse contour and estimates the central aortic pressure profile via a software‐based algorithm. PWA estimates the extra systolic pressure load on the left ventricle (the augmentation index [AI]), AI₇₅ (AI standardized to a heart rate of [HR] 75 beats per minute), and differences in systolic BP (SBP) between the central and the peripheral vasculature. Several studies have indicated that PWA predicts CV outcomes. ⁸ , ⁹ We undertook this study to determine whether individuals with WCH have differences in central aortic pressures as determined by PWA when compared with normotensive individuals.

METHODS

Study Participants

Individuals with WCH were recruited after 24‐hour ABPM was performed at the Hypertension Clinic at the University of Pennsylvania, and a diagnosis of WCH was made usually within a month of the ABPM. We invited all participants on no medications with an average BP of <135/85 mm Hg on ABPM and a diagnosis of WCH to participate in the study over the period between April 2005 and December 2005. Normotensive participants, aged 18 to 70 years, were recruited as controls from among the office staff and by colleagues. No subjects were taking antihypertensive medications or had antihypertensive drug exposure within the preceding year. A limited health history was taken on each participant to exclude a personal history of diabetes, hypertension, or prevalent disease. A nominal sum was reimbursed to each participant to cover transportation costs and the inconvenience of the study visit. Institutional review board approval was obtained at the Hospital of the University of Pennsylvania in Philadelphia, and written informed consent was obtained from all participants.

Procedures

All study procedures took place in the Hypertension Clinic. Height and weight were measured and body mass index (BMI) was calculated. After a 5‐minute rest period, a mean of 2 BP readings was recorded over the brachial artery with the subject sitting, and the average SBP, diastolic BP, HR, and pulse pressure were recorded. PWA measurements were performed supine using the right radial artery after at least 5 minutes of rest. A Millar tonometer was attached to an electronic module interface (SphygmoCor PVx device; AtCor Medical, West Ryde, Australia) and was then placed perpendicular to the palpated radial pulse.

Radial artery pressure waveforms were sampled for 10 seconds and an ensemble average of the AI and AI₇₅ were obtained and recorded. To be specific, the radial artery waveforms were processed using proprietary software (SphygmoCor 2000 version 7.2, AtCor Medical, West Ryde, Australia), which has an internal algorithm that performs the calculations. The AI is the pulse pressure (systolic ‐ diastolic) in the aorta divided into the amount of systolic pressure increase resulting from the reflected peripheral pulse wave expressed as a percent. The AI₇₅ was determined at the time of study by the formula: AI₇₅ = ([HR − 75] × 0.39) + AI.

Data are presented as mean ± SE. Data were analyzed using SPSS version 12 (SPSS, Inc, Chicago, IL). Standard descriptive statistics were used. Comparisons were made between the WCH and normotensive groups using independent Student t test, and a P value of <.05 was considered significant.

RESULTS

Thirty‐six participants were studied, 18 normotensive participants and 18 participants with WCH. The demographics are shown in the Table. The majority of participants were women, with 71% of women in the normotensive group and 62% in the WCH group. Participants were younger in the normotensive group, with a mean age of 41.9±2.1 years compared with 51.4±3.9 years in the WCH group. BMI was similar in both groups, with a mean BMI of 27.3±1.6 kg/m². Ethnicity of participants is shown in the Table.

Table.

Demographic and Hemodynamic Results

	Normotensive	White Coat Hypertension	P Value
Sex (female/male)	14/4	13/5	NS
Age, y	41.9±2.1	51.4±3.9	.04
Body mass index, kg/m2	26.3±1.3	28.7±2.0	NS
Ethnicity (white/African American)	11/7	17/1
Peripheral systolic BP, mm Hg	110.7±2.6	129.1±2.5	<001
Aortic systolic BP, mm Hg	97.9±2.5	115.2±2.9	<001
Change in systolic pressure centrally to peripherally, mm Hg	12.7±0.9	12.3±1.8	.82
Diastolic BP, mm Hg	70.7±2.0	76.2±2.3	.08
Heart rate, beats per minute	69.0±2.3	68.4±2.7	.88
Pulse pressure, mm Hg	40.0±1.4	52.9±2.1	<001
AI, %	11.7±3.3	22.3±3.5	<05
AI75, %	8.8±2.9	18.9±3.5	<05
Data are expressed as mean ± SEM. Abbreviations: AI, augmentation index; AI75, AI standardized to a heart rate of 75 beats per minute; BP, blood pressure; NS, not significant.

Hemodynamic data, AI, and AI₇₅ are shown in the Table. AI was significantly higher (P<.05) in WCH participants than normotensive participants, and AI₇₅ was significantly higher (P<.05) in WCH participants than normotensive participants (Figure 1). The central aortic systolic pressures (ASP) were also significantly higher in the WCH group (P<.001) (Figure 2). SBP was significantly higher in the WCH group; these SBP measurements, however, would still be considered nonhypertensive BP readings with a mean SBP of 129±2.5 mm Hg compared with mean SBP readings of 110.7±2.6 in the normotensive group. Diastolic BP and HR were not significantly different in the 2 groups. Pulse pressure was also significantly increased in the WCH group vs the normotensive group (P<.001). We also assessed the difference between centrally measured systolic pressure and the peripheral SBP. There was no difference in these measures between the 2 groups.

Figure 1.

AI indicates augmentation index; AI₇₅, AI standardized to a heart rate of 75 beats per minute; WCH, white coat hypertensive.

Figure 2.

Differences in blood pressure (BP) measured in white coat hypertensive (WCH) and normotensive groups. ^aP<.001. SBP indicates systolic BP; ASP, aortic systolic pressures; Δsp difference between the peripheral and control BP.

DISCUSSION

Our results show that participants with a clinical diagnosis of WCH have higher AI, AI₇₅, and ASP values when compared with normotensive participants. WCH is typically defined as an office BP >140/90 mm Hg on at least 3 separate office or clinic visits with 2 separate BP measurements performed at each visit, ¹⁰ usually in the absence of target organ damage. ABPM confirms the diagnosis of WCH with an ambulatory BP <135/85 mm Hg with no evidence of target organ damage. All participants in our study were diagnosed with WCH after having office BP measurements of >140/90 mm Hg at their primary care physicians who then referred them for ABPM. If participants had mean BP readings of <135/85 mm Hg during ABPM, the diagnosis of WCH was then made. These participants were then asked to enroll in the study.

WCH has been considered by some investigators to be a benign condition; however, this is controversial. Some but not all guidelines do not recommend treating WCH in the absence of target organ damage. Studies focusing on target organ damage have shown a mixed picture. Various studies have looked at echocardiographic changes and the presence of left ventricular hypertrophy in healthy patients; more recent studies have assessed other markers of risk including carotid atherosclerosis, endothelial dysfunction, and homocysteine. ¹¹ , ¹² , ¹³ There is disagreement in these studies as to whether the risk of WCH is closer to that of true normotension or true hypertension. Prognostic studies have shown that the CV risks of WCH are probably lower than those of true hypertension but higher than those of true normotension. ⁵ , ⁸ , ¹⁴ In light of the recent Conduit Artery Function Evaluation (CAFE) study ¹⁵ findings, our observation of a significantly higher central aortic systolic pressure in the WCH group leads us to agree with those investigators who suggest that WCH is not an innocuous condition.

PWA is an indirect measure of arterial stiffness. Increases in arterial stiffness hasten the return timing of the reflected wave. When the returning wave reflects back to the heart more quickly, the cardiac cycle is more likely to be in the latter stages of systole rather than early diastole. In this way, the returning wave adds an additional late systolic pressure burden to left ventricular ejection increasing the left ventricle load. Increases in AI or AI₇₅ have been shown to be stand‐alone markers of increased CV risk, although the number of studies demonstrating this are few because of the relative newness of PWA compared with the more standard measure of vascular stiffness (pulse wave velocity).

The purpose of our study was to determine whether there were differences in central hemodynamics by assessing PWA in participants with WCH, using normotensive participants as controls. An assessment of the Table reveals a number of limitations we encountered when performing and analyzing this study. Because we intentionally enrolled participants with WCH, their office BP (which was the reason they underwent ABPM evaluation) was higher, thus we could not match participants based on BP alone. Ideally, this study should have been performed in a larger group of patients, with controls who were matched for baseline BP and age instead of a normotensive group of medical personnel.

Another limitation of our study was the role that differences in BP between the 2 groups at the time of observation may have played in the central AI. Since the reflected wave travels more quickly at a higher BP, part of the difference we noted in central AIs between the 2 groups may be related to this. On the other hand, the magnitude of the reflected wave, which depends on the systemic vascular resistance and reflectance site locations also contributes to the central AI. We cannot disentangle these contributions (stiffness and magnitude) given the current technology; thus, we acknowledge the inability to match for BP level at the time of AI measurement as a limitation in our data.

Ethnicity also varied between the groups, with most participants in the WCH group being white (17 of 18); a lesser percentage of participants in the control group were white (11 of 18). Ethnicity in the WCH group is representative of our referral base of patients with the possible diagnosis of WCH. Similarly, although the sex distribution and BMI were comparable between the groups, the subjects with WCH were on average a decade older. This older age could also account for some of the differences in AI that we observed, as noted by McEniery and colleagues ¹⁶ who showed a 5% increase in men and women from the fifth to sixth decade. Our data had about twice this amount of change in AI and probably was not related alone to differences in age.

CONCLUSIONS

Our study found a significant difference in AI, AI₇₅, and ASP between WCH and normotensive individuals. Individuals in the WCH group had significantly higher SBP and pulse pressure; these were, however, still within the normal range. In view of recent Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure guidelines, ¹⁷ these participants also meet the criteria for prehypertension. Other hemodynamic parameters were similar between the groups. The difference between central and aortic SBP measurements was similar between the groups, but the WCH group had higher central aortic pressures. Our data support the notion that WCH probably represents something of an intermediate risk between normotension and hypertension. This implies increased arterial stiffness in these participants despite normal BP readings; this could be considered a surrogate measure of target organ damage. Although many hypertension specialists would not advocate antihypertensive treatment for patients with WCH, our data raise an important question as to whether this is the best approach in patients with WCH and whether treatment might be initiated based on findings such as those obtained noninvasively. We agree with the recent review ¹⁸ that supports the proposal that a study needs to be done where participants identified as having WCH would be randomized to either placebo or treatment with an antihypertensive agent. Outcomes such as changes in the central aortic SBP and the possibility that hypertension could be delayed or prevented by drug therapy, as was noted in the recent Trial for Preventing Hypertension (TROPHY) study, ¹⁹ would be monitored.