Stress-Induced Laboratory Blood Pressure in Relation to Ambulatory Blood Pressure and Left Ventricular Mass Among Borderline Hypertensive and Normotensive Individuals
Our primary aim in the present study was to investigate the association between blood pressure measured in the laboratory and in the ambulatory state in a group of middle-aged borderline hypertensive men and age-matched normotensive control subjects. In addition, we examined the relation between stress-induced blood pressure measurements and left ventricular mass. Blood pressure and heart rate were measured noninvasively during a standardized laboratory stress protocol and four times per hour throughout 24 hours. Borderline hypertensive subjects had significantly higher systolic and diastolic pressures than normotensive subjects during both the daytime (systolic pressure, 141.1±9.7 versus 130.9±8.6 mm Hg; diastolic pressure, 88.8±7.0 versus 79.4±6.2 mm Hg, P<.001) and nighttime (systolic pressure, 114.0±9.9 versus 107.1±8.3 mm Hg; diastolic pressure, 71.5±7.5 versus 64.6±7.2 mm Hg, P<.001). The borderline hypertensive group also displayed increased systolic pressure reactivity in the laboratory compared with the normotensive group. The groups did not differ significantly in left ventricular mass (index). In both borderline hypertensive and normotensive individuals, blood pressure levels during stress testing were closely related to ambulatory blood pressure levels (r=.51 to .82). Furthermore, stress-induced blood pressure levels were significantly correlated to left ventricular mass in borderline hypertensive (r=.33 to .40) but not normotensive subjects. Since stress-induced blood pressure levels were significantly associated with both ambulatory blood pressure levels and left ventricular mass in borderline hypertensive subjects, the addition of standardized stress testing to casual blood pressure measurements may improve risk estimation.
Ambulatory blood pressure (ABP) readings are more closely related to hypertensive complications than are casual office BP readings.1 2 ABP levels also show a closer relationship to LVM and left ventricular wall thickness than do casual BP measurements in both hypertensive3 4 and BHT5 individuals. In addition, ABP levels have been found to predict the development of sustained hypertension from a borderline state better than casual BP measurements.6 Behaviorally evoked BP changes may also contribute to cardiovascular complications. Previous studies suggest that exaggerated cardiovascular responses to standardized stress tests in the laboratory can predict cardiovascular complications such as coronary heart disease7 and hypertension.8 9
Because stress reactivity assessment in the laboratory can be conducted with a higher degree of standardization and control than ABP determination, it is of interest to examine the relationship between cardiovascular measurements in the laboratory and in the ambulatory state. Several previous studies have found a strong positive relationship between ABP and laboratory BP levels.10 11 12 13 14 However, the relationship between laboratory-induced BP reactivity (ie, change from baseline in response to stress challenge) and ABP levels is not as consistent. Some studies have found cardiovascular reactivity to tasks requiring an active behavioral response to be related to ABP levels,15 16 17 whereas others have found no relationship between stress-induced BP reactivity and ABP levels.11 18 Most studies relating BP reactivity to ABP have been conducted in NT individuals, and only a few studies exist in which hypertensive subjects have been studied and compared with normotensive subjects.
BHT individuals have an increased risk of developing essential hypertension.19 20 Several studies indicate that BHT individuals show exaggerated BP and HR reactivity to laboratory stressors compared with NT control subjects, primarily to psychological tasks, such as mental arithmetic or stressful interviews.21 This hyperreactivity in BHT compared with NT individuals can be attributed to an enhanced sympathetic nervous system activity that may precede the hypertensive state, since sympathetic overactivity is less evident in individuals with established hypertension.22 Hence, BHT and NT control individuals might differ in the association between cardiovascular responses in the laboratory and ambulatory state.
Previous studies show rather weak correlations between casual clinic BP measurements and LVM. Since sympathetically mediated cardiovascular reactivity is implicated in the pathogenesis of left ventricular hypertrophy,23 stress-induced BP measurements may show a closer association with left ventricular dimensions than casual BP levels. Evidence exists of increased sympathetic tone among BHT compared with NT individuals.24 Thus, the relationship between stress-induced BP measurements and LVM might be more pronounced among BHT compared with NT control subjects.
Our aim in the present study was to compare ABP levels with stress-induced BP and HR levels and reactivity in the laboratory among BHT and NT individuals. In addition, we examined the association between stress-induced BP measurements and LVM.
In 1985, a BP screening program started in the small town of Åkersberga, north of Stockholm. All men aged 35 to 55 years were asked by mail to visit the primary healthcare center to have their BP measured. Of a sample of 2694 people, 207 were found to have borderline hypertension, defined as a supine DBP of 85 to 94 mm Hg measured on at least two separate occasions. The subjects had yearly checkups for 3 years. A total of 81 subjects consistently had a DBP within the borderline range. These 81 BHT individuals were invited to participate in the present study together with 80 age-matched control subjects from the original population who had a DBP less than or equal to 80 mm Hg at the initial measurement. To obtain the 80 age-matched individuals in the present study, we asked 105 NT individuals to participate; 23 declined to participate and 2 had a DBP above 80 mm Hg. The BP of the control subjects was measured on two occasions a few weeks apart, both at the initial screening and before this study. To be able to participate in the study, the individual's DBP had to be less than or equal to 80 mm Hg on all occasions. The study was approved by the local Ethics Committee of Karolinska Hospital and was conducted in accordance with the Declaration of Helsinki. All subjects gave their written informed consent before entering the study. Of the 81 BHT and 80 NT individuals who agreed to participate in the study, 6 in each group discontinued. In the BHT group, 2 died (cancer and unknown cause), 1 moved to Canada, and 3 started in the study but did not wish to complete all the tests. In the NT group, 1 participant developed insulin-dependent diabetes mellitus and could not complete the study, and 5 subjects started but later withdrew.
All casual BP measurements during the recruitment procedure were made by a specially trained nurse with a mercury sphygmomanometer. The cuff size was adjusted to the circumference of the left arm, and the arm was placed with the cuff at heart level. Clinic BP was calculated as the mean of two measurements taken with subjects in the sitting position after 5 minutes of rest. SBP and DBP were defined according to Korotkoff phases I and V, respectively.
ABP and Laboratory BP
ABP and laboratory BP and HR were monitored with a Pressurometer-IV (model 1990, Del Mar Avionics).25 In the laboratory, an instructor was present in the room with the subject to manually activate the Pressurometer-IV during the various laboratory stress tests. For the ambulatory measurements, the Pressurometer-IV was set to automatically record SBP, DBP, and HR (in beats per minute [bpm]) every 15 minutes throughout 24 hours during an ordinary work day. At each measurement time, the Pressurometer-IV inflates and deflates an arm cuff and records the appearance and disappearance of Korotkoff sounds through a microphone placed over the brachial artery. The 24-hour data were stored in a memory unit and then transmitted to a computer that constructed an individual data file. Artifacts (defined as any of the following: SBP <50 mm Hg, SBP >250, DBP>SBP, DBP <30, DBP >150) were excluded. For every reading, subjects were asked to keep their arm as motionless as possible to prevent artifactual readings and were also asked to record their location, position, and activity in a diary every time the cuff recorded their BP.
Left ventricular wall thickness was assessed by M-mode echocardiography. The echocardiographic recordings were obtained with an ATL Ultramark 8 with a 3.0-MHz transducer and a 4-cm focusing point. The M-mode investigations were guided by a cross-sectional survey of left ventricular anatomy. Registration techniques and measurements of left ventricular diameters and wall thickness were performed in accordance with recommendations of the American Society of Echocardiography.26 All registrations were performed by the same investigator, who had no knowledge of the subjects' BP levels. LVM was measured according to Devereux et al.27 Since body size is the factor most closely correlated to LVM,28 LVMI was calculated as LVM divided by body surface area.29
Subjects arrived at the laboratory between 9 and 11 am. After the initial hookup procedure, all subjects were given a stress reactivity assessment that included a pretask resting period and two stress tests—mental arithmetic and isometric muscle contraction. The pretask measurements were taken with subjects in the sitting position during a 10-minute resting period. An average of two BP and HR measurements was obtained during rest. For the mental arithmetic, subjects were asked to sum one- and two-digit numbers in the presence of a 90-dB noise delivered through earphones during 4 minutes. Subjects were not harassed or asked to go faster. An average of four BP and HR measurements was obtained during mental arithmetic. Isometric muscle contraction was performed at one third of previously determined maximal voluntary contraction with the use of a pressure meter (Vigorimeter, Martin) during 2 minutes. Two BP and HR measurements were obtained during isometric exercise.
To obtain measures of BP and HR reactivity, we calculated change scores by subtracting averaged pretask resting levels from mean values during the stress tests. The 24-hour period consisted of all measurements taken during 24 hours, with laboratory measurements excluded. The daytime period was defined as all waking hours, and the nighttime period was defined as all sleeping hours, as judged from participants' diaries. All readings for SBP, DBP, and HR within each period of time (full 24-hour period, daytime, and nighttime) were averaged to represent that period.
ANOVA was performed on BP and HR measurements recorded at the clinic, in the laboratory, and in the ambulatory state. Values are given as mean±SD. Product-moment correlations were computed for investigation of the associations between casual BP levels, stress-induced BP levels and reactivity, and ABP levels. Product-moment correlations were also computed for investigation of the associations between stress-induced BP levels and reactivity and echocardiographic measurements.
Mean age was 50.0±5.7 and 49.9±5.7 years in the BHT and NT groups, respectively. The groups did not differ in smoking habits, with 25 smokers in the BHT group and 27 in the NT group, with an average consumption of five to six cigarettes per day. All individuals were employed and working full-time on the day of the ABP recordings.
Casual BP and ABP Measurements
Casual office BP levels were 140.6±9.5/89.4±2.4 mm Hg for the BHT group and 124.6±10.8/75.4±3.9 mm Hg for the NT group.
A total average of 89.2±12.2 and 88.5±15.3 ABP measurements were recorded in the BHT and NT groups, respectively. Seventy-two percent of the measurements were recorded during the daytime and 28% during the nighttime, with no significant difference in the number of measurements between BHT and NT participants.
ABP and HR Levels
For the entire 24-hour period and during the daytime, measurements of SBP, DBP, and HR were significantly higher in the BHT than the NT group. During the nighttime, SBP and DBP levels were significantly higher in the BHT group, but the groups did not differ significantly in HR levels (Table 1⇓). BHT individuals had significantly greater day-night differences than NT subjects in SBP (27.0±9.8 versus 24.0±8.1 mm Hg, P<.05) and DBP (17.5±6.1 versus 14.7±6.4 mm Hg, P<.01), reflecting a greater rise of BP levels during the day compared with the night in BHT individuals. Day-night differences in HR levels were not significant (11.7±6.0 versus 10.5±5.4 bpm, P=.23).
Standing Versus Sitting ABP Measurements
An average of 45% of the daytime BP and HR measurements were taken with subjects sitting, with no significant difference between the BHT and NT groups. The influence of posture on ABP and ambulatory HR was calculated by subtracting sitting from standing measurements. The effect of posture on BP and HR did not differ significantly between the groups: SBP was 3.9±6.6 versus 5.9±8.9 mm Hg (P=.12); DBP was −0.8±5.4 versus −1.8±6.4 mm Hg (P=.28); and HR was 2.6±6.4 versus 2.1±7.4 bpm (P=.86) for the BHT and NT groups, respectively.
Laboratory BP and HR Levels and Reactivity
SBP and DBP levels during resting, mental arithmetic, and isometric exercise were significantly higher among BHT than NT subjects. HR levels were significantly higher during resting and mental arithmetic in the BHT than the NT group. The groups did not differ in HR levels during isometric exercise (Figure⇓).
SBP reactivity to mental arithmetic and isometric exercise was significantly higher in the BHT than the NT group. DBP and HR reactivities to any laboratory stressor did not differ significantly between the two groups (Table 2⇓).
Relation Between Casual, Ambulatory, and Laboratory BPs
Casual office SBP levels were correlated to daytime SBP levels in both the BHT and NT groups. Casual DBP levels were significantly associated with daytime DBP levels in the BHT but not the NT group (Table 3⇓). BP levels attained in the laboratory were consistently and significantly associated with daytime ABP levels in both BHT and NT participants. BP reactivity in the laboratory was not significantly associated with daytime BP levels in either the BHT or the NT group (Table 3⇓).
Correlations Between Laboratory BP and LVMI
LVMI did not differ significantly between the BHT and NT groups (114±22 and 109±22 g·m−2, respectively). Details of the echocardiographic findings as related to ABP among BHT and NT subjects have been reported previously.5
Correlations between casual and laboratory BP measurements and LVMI are presented in Table 4⇓. Casual clinic BP measurements were not significantly associated with LVMI for either BHT or NT subjects. Resting and stress-induced SBP levels correlated significantly with LVMI in the BHT group but not in the NT group. SBP reactivity did not correlate significantly with LVMI in either the BHT or NT group. DBP levels or reactivity and LVMI did not correlate significantly in either the BHT or NT group.
In the present study, BHT subjects had significantly higher ABP levels than NT subjects during both the daytime and nighttime. There was a greater awake-asleep difference in BP levels among BHT than NT participants, reflecting greater BP elevations during the daytime in the BHT group. These daytime BP elevations among BHT individuals are not an effect of postural differences between the groups because the number of measurements taken with subjects standing and sitting was the same in both groups. However, given that BHT individuals show an increased reactivity to orthostatic stress, it might be argued that BP could increase more in the BHT than the NT group as a result of posture. Among BHT individuals, BP responses to upright posture have been reported to be both normal30 and exaggerated.31 In the present study, there were no differences between the groups when the effect of body posture (sitting versus standing) on ABP levels was estimated. Hence, daytime BP elevations seem psychologically induced, possibly through an increased sympathetic drive, which is consistent with a higher HR among BHT individuals.
The higher BP during sleep in BHT than NT participants despite similar HR could suggest the presence of structural abnormalities as a basis for increased BP. This may apply for nighttime but not daytime BP because the higher HR levels during the daytime and the greater awake-asleep difference in BP levels in BHT than NT subjects indicate an increased sympathetic nervous system drive during the day. Several other lines of evidence support the suggestion that increased sympathetic nervous system activation is associated with hypertension. For example, Anderson et al32 assessed sympathetic nervous system activity through direct nerve recordings (microneurography) and reported higher sympathetic nervous system activation in BHT subjects than in matched control subjects. In addition, Fredrikson et al33 observed an increased rate of nonspecific electrodermal fluctuations during a passive habituation task, reflecting increased sympathetic nervous system activity in a group of young BHT subjects. Results from the present study indicate that sympathetic nervous system differences are not evident during sleep (HR is similar) but are manifest while an individual is awake during ordinary everyday activities.
In most previous studies, BHT versus NT control subjects have exhibited exaggerated BP and HR reactivity to laboratory tasks, mainly to stressors requiring an active behavioral response. In the present study, stress-induced BP was increased in the BHT compared with the NT group, whereas HR reactivity in the laboratory was similar in both groups. Increased BP but normal HR reactivity is a pattern mainly observed in individuals with essential hypertension.21
The early stages of hypertension are characterized by increased sympathetic drive and decreased parasympathetic cardiac inhibition. As hypertension advances, β-adrenergic responsiveness and cardiac compliance tend to decrease and hypertrophy of the resistance vessels increases. It should be noted that the present participants were middle-aged and had borderline hypertension for at least 4 to 5 years. Thus, the higher BP but similar HR during the nighttime and in the laboratory may indicate that the BHT group is changing hemodynamic pattern from high cardiac output/low vascular resistance to normal cardiac output/high vascular resistance. Since the BHT subjects had relatively modestly increased BP levels, results from the present study indicate that a hemodynamic shift may occur at low BP levels, before hypertension is established.
Stress-induced BP levels, but not reactivity measures, predicted ABP levels in both the BHT and NT groups. In a study by de Faire et al,6 task-induced BP levels in the laboratory and reactivity measures could not predict BP development, whereas Light et al34 found that laboratory reactivity added to the prediction of BP development. However, the studies differ in time span. De Faire et al used ABP and laboratory measures to predict the development of sustained hypertension from a borderline state at 1 year of follow-up; Light et al used laboratory measurements as predictors of clinic and ABP measurements 15 years later in a group of normotensive individuals. Thus, stress-induced laboratory reactivity may possibly be predictive of BP development over more extended periods of time (several years rather than months).
Pickering and Gerin2 pointed out that discontinuous BP measurements during laboratory stressors may produce results of limited accuracy and low reliability. In a recent meta-analysis, Swain and Suls35 reviewed studies that assessed responses to the same laboratory stressors across occasions using various discontinuous BP measurement techniques. The highest reactivity reproducibility was obtained when three or more BP measurements were recorded during each stressor and when adults rather than young subjects, nonspeech rather than speech stressors, and ambulatory auscultatory equipment rather than manual or oscillometric methods were used. Overall, SBP reactivity showed higher reproducibility than DBP reactivity. In the present study, the participants were adults, the stressors were of a nonspeech character, ambulatory auscultatory equipment was used, and two to four measurements were recorded for each stressor. Thus, the BP reactivity measurements obtained in the present study are most likely reliable enough to support the conclusions.
Left ventricular hypertrophy is a major risk factor for arrhythmias and sudden death,36 and enhanced sympathetic activity may be involved in its development. Sympathetic stimulation increases the release of norepinephrine and angiotensin, both of which act as trophic factors and aggravate the pressure-induced tendency for cardiac hypertrophy.37 38 In the present study, LVMI did not differ significantly between the BHT and NT groups. However, there were significant associations between stress-induced BP levels and LVMI in the BHT but not the NT group. This is in line with previous results in which we found ABP levels to be associated with echocardiographic measures in the BHT but not the NT group.5 Thus, both ambulatory and stress-induced BP levels are more closely related to LVM than are casual BP measurements among BHT individuals.
ABP measurements are more strongly associated with morbidity and mortality than casual clinic BP measurements,1 2 and hypertensive individuals with left ventricular hypertrophy have a worse prognosis than those without.39 40 In the present study, stress-induced BP levels were strongly associated with ambulatory BP levels, and casual clinic BP measurements were not. Stress-induced BP levels were also significantly associated with LVM in the BHT group, and casual BP measurements were not. Since stress-induced BP levels were associated with both ambulatory BP levels and LVM among BHT individuals, the addition of standardized stress testing to casual BP measurements may improve risk estimation for hypertension-prone individuals.
Selected Abbreviations and Acronyms
|ABP||=||ambulatory blood pressure|
|DBP||=||diastolic blood pressure|
|LVM||=||left ventricular mass|
|LVMI||=||left ventricular mass index|
|SBP||=||systolic blood pressure|
This study was funded by grants from the Bank of Sweden Tercentenary Fund.
- Received February 21, 1996.
- Revision received March 26, 1996.
- Accepted May 3, 1996.
Perloff D, Sokolov M, Cowan R. The prognostic value of ambulatory blood pressure. JAMA. 1983;259:2793-2798.
Palatini P, Marmino P, Di Marco A. Ambulatory blood pressure versus casual blood pressure for the evaluation of target organ damage in hypertension: complications of hypertension. J Hypertens. 1985;3(suppl 3):S425-S427.
Verdecchia P, Schillaci G, Guerrieri M, Gatteschi C, Benemio G, Boldrini F. Circadian blood pressure changes and left ventricular hypertrophy in essential hypertension. Circulation. 1990;81:528-536.
Borghi C, Costa FV, Boschi S, Mussi A, Ambrosioni E. Predictors of stable hypertension in young borderline subjects: a five-year follow-up study. J Cardiovasc Pharmacol. 1986;8(suppl 5):S138-S141.
Matthews KA, Woodall KL, Allen MT. Cardiovascular reactivity to stress predicts future blood pressure status. Hypertension. 1993;22:479-486.
Fredrikson M, Robson A, Ljungdell T. Ambulatory and laboratory blood pressure in individuals with negative and positive family history of hypertension. Health Psychol. 1991;53:1-12.
Harshfield GA, James GD, Schlussel Y, Yee LS, Blank SG, Pickering TG. Do laboratory tests of blood pressure reactivity predict blood pressure changes during everyday life? Am J Hypertens. 1988;1:168-174.
Langewitz W, Ruddel H, Schachinger H, Schmieder R. Standardized stress testing in the cardiovascular laboratory: has it any bearing on ambulatory blood pressure values? J Hypertens. 1989;7:S41-S48.
Van Egeren LF, Sparrow AW. Laboratory stress testing to assess real-life cardiovascular reactivity. Psychosom Med. 1989;51:1-9.
Fredrikson M, Blumenthal JA, Evans DD, Sherwood A, Light KC. Cardiovascular responses in the laboratory and in the natural environment: is blood pressure reactivity to laboratory-induced mental stress related to ambulatory blood pressure during everyday life? J Psychosom Res. 1989;33:753-762.
Lund-Johansen P. Hemodynamics in early essential hypertension. Acta Med Scand Suppl. 1967;482:1-105.
Lund-Johansen P. Central haemodynamics in essential hypertension at rest and during exercise: a 20-year follow-up study. J Hypertens. 1989;7(suppl 6):S52-S55.
Devereux RB, Pickering TG, Harshfield GA, Kleinert HD, Denby L, Clark L, Pregibon D, Jason M, Kleiner B, Borer JS, Laragh JH. Left ventricular hypertrophy in patients with hypertension: importance of blood pressure response to regularly occurring stress. Circulation. 1983;68:470-476.
Julius S. Changing role of the autonomic nervous system in human hypertension. J Hypertens. 1990;8(suppl 7):59-65.
Sahn DJ, De Maria A, Kisslo J, Weyman A. The committee on M-mode standardization of the American Society of Echocardiography: recommendations regarding M-mode echocardiography. Results of a survey of echocardiographic measurements. Circulation. 1978;58:1072-1083.
Sannerstedt R, Julius S, Conway J. Hemodynamic response to tilt with beta-adrenergic blockage in young patients with borderline hypertension. Circulation. 1970;42:1057-1064.
Frohlich ED, Tarazi RC, Ulrych M, Dustan HP, Page IH. Tilt test for investigating a neural component in hypertension: its correlation with clinical characteristics. Circulation. 1967;36:387-393.
Anderson EA, Sinkey CA, Lawton WJ, Mark AL. Elevated sympathetic nerve activity in borderline hypertensive humans. Hypertension. 1989;14:177-183.
Fredrikson M, Edman G, Levander S, Shalling G, Svensson J, Tuomisto M. Electrodermal responsivity in young hypotensive and hypertensive men. Psychophysiology. 1990;6:649-655.
Light KC, Dolan CA, Davis MR, Sherwood A. Cardiovascular responses to an active coping challenge as predictors of blood pressure patterns 10-15 years later. Psychosom Med. 1992;54:217-230.
Simpson P. Norepinephrine-stimulated hypertrophy of cultured rat myocardial cells is an alpha-1 adrenergic response. J Clin Invest. 1983;72:732-738.
Dzau VJ, Gibbons GH, Pratt RE. Molecular mechanisms of vascular renin-angiotensin system in myointimal hyperplasia. Hypertension. 1991;18(suppl II):II-100-II-105.
Kannel WB, Castelli WP, McNamara PM, McKee PA, Feinleib M. Role of blood pressure in the development of congestive heart failure: the Framingham study. N Engl J Med. 1972;287:781-787.