Automated ambulatory blood pressure monitoring: A study in age-matched normotensive and hypertensive men

Noninvasive automated techniques have been used in this study in 29 ambulatory normotensive

and hypertensive men to monitor blood pressure during a full 24-hour period. The two groups of men were carefully matched for age. The data were analyzed with respect to possible

differences in the circadian pattern of blood pressure, in the variability of blood pressure, and differences in the prevalence of abnormal blood pressures between the two groups. The

circadian pattern of blood pressure in hypertensive patients was displayed parallel to that in

normotensive control subjects but at a significantly higher level. The difference between daytime and nighttime systolic blood pressure was greater in normotensive than in hypertensive men.

However, variability of blood pressure was not different between the two subgroups. Hypertensive men showed on average a greater incidence of abnormal blood pressures than

normotensive subjects. However, nine of the normotensive control subjects showed more than

25% of abnormal readings of systolic or diastolic blood pressures during a 24-hour recording period. In contrast, fewer than 25% of abnormal blood pressure readings were found in nine of

the hypertensive patients. The data reported here may help to better define the use of

ambulatory blood pressure monitoring techniques in the diagnosis of hypertension. (AM HEART J

109:1334, 1985.)

Jan I. M. Drayer, M.D., Michael A. Weber, M.D., and Dina K. Nakamura. Long Beach and Irvine, Calif.

Noninvasive automated blood pressure monitoring has now frequently been used in ambulatory hyper- tensive patients. It has served to better define characteristics of blood pressure in patients with borderline hypertension’-* and in elderly hyperten- sive patients.5 Only few studies have reported results of blood pressure monitoring in normotensive con- trol subjects.*~ “3 A, 7 The studies have shown that the whole-day average of blood pressures, the circadian pattern of blood pressure, and the incidence of abnormal blood pressure levels are important parameters of the results of blood pressure monitor- ing.

In this study, we have used these techniques in the analyses of blood pressure monitorings to allow comparison of results in normotensive and hyper- tensive men. The groups of patients and volunteers were carefully matched for their age and for sex so as

Frum the Section of Clinical Pharmacology and Hypertension. Veterans Administration Medical Center. and the Ilniversity of California.

Received for publication Aug. 6. 1984; revision received Dec. :i. 1984: accepted Jan. 7. 1985.

Reprint requests: Jan I. M. Drayer, M.D., Hypertension Center W130. VA Medical Center, 5901 E. Seventh St., Long Beach, CA 90822.

to exclude the effects of these parameters on the results of the study. The results of this study might help to better define normalcy of blood pressures using these noninvasive automated blood pressure monitoring techniques.

METHODS

Patients. In this study noninvasive ambulatory blood pressure monitoring was performed in 29 men with benign essential hypertension and in 29 normotensive male vol- unteers. The groups were carefully matched for age. Antihypertensive therapy had been discontinued in the hypertensive patients for at least 2 weeks prior to the day of the blood pressure monitoring. All normotensive volun- teers had a seated casual blood pressure less than 140/90 mm Hg on at least two occasions. In contrast, all hyper- tensive patients had repeated seated casual diastolic pressures greater than 90 mm Hg, and most had systolic blood pressures greater than 140 mm Hg.

On the day of the study, the patients’ height and body weight were obtained. Actual body weight was expressed as a percentage of the ideal body weight (Metropolitan Life Tables). The participants in the study were requested to perform their normal daily activities during the 24-hour monitoring period. The blood pressure monitorings were performed on working days.

1334

Volume 109

Number 6 Ambulatory BP monitoring in hypertension 1335

Table 1. Clinical characteristics of age-matched normoten- sive and hypertensive men

Normotensive Hypertensive

(n = 291 (n = 29)

Age (yr) 42 + 12 ns 42 zk 12 Height (cm) 179 It 9 ns 178 k 6 Body weight 81.6 t 13.9 ns 84.9 f 11.7 Degree of obesity (“c’ ) 113 + 15 ns 118 I 15

ns = not significant

BP monitoring and analyses. The blood pressure mon- itoring was performed using the Pressurometer III (Del Mar Avionics, Irvine, Calif.). With this instrument, syst.o- lit and diastolic blood pressure and heart rate were obtained at 7.5-minute intervals throughout the 24-hour day. After completion of the monitoring, the data were printed using a 1979 Pressurometer Charter. The tech- niques used for the analysis of the data have been described earlier!.” Readings which showed an inconsis- tent increase or decrease in systolic or diastolic blood pressure of greater than 20 mm Hg, and readings with a calculated pulse pressure of less than 5 mm Hg were deleted prior to further analysis of the data. More than 150 blood pressure readings were available for analysis in each tracing.

The following analyses of systolic and diastolic blood pressure were performed using an Apple II Plus micro- computer: averages of all readings obtained during the 24-hour monitoring period, averages of readings obtained during 12 consecutive 2-hour periods starting at midnigyht, and the averages of daytime (0600 to 2200 hours) and nighttime (2200 to 600 hours) blood pressures. In addi- tion, a histogram was derived using blood pressures of each of the 24-hour monitoring periods, and the incidence of “abnormal” systolic (>140 mm Hg) and/or diastolic (>90 mm Hg) blood pressures was determined for each monitoring period. This study was approved by the Research and Development Committee and all subjects signed an approved informed consent.

Statistics. Standard statistical tests, paired and unpaired Student t tests, and Pearson correlation coeffi- cients were used in the analyses of the data. All data are expressed as mean t one standard deviation. Two-tailed p values are used to express the significance of the results of the statistical tests.

RESULTS

Noninvasive automated ambulatory blood pres- sure monitorings were performed in 29 normoten- sive and hypertensive men. The clinical characteris- tics of the two groups are given in Table I. The age in the groups of participants ranged from 23 to 60 years. The groups did not differ significantly with respect to their height, body weight, or degree of obesity. A special cuff was not required in any patient or volunteer. Thus, the differences in BP

. NORMOTENSIVE MEN

t ,, HYPERTENSVE MEN

50

I I I I I I I I SAM 12 4PM 8 12 4 8 AM

TlME OF DAY (hrs)

Fig. 1. Circadian pattern of blood pressure in 29 age- matched normotensive and hypertensive men.

patterns found in this study between the normoten- sive and hypertensive subjects are not likely to be a consequence of sex, age, or the degree of obesity of the participant.

BP patterns. The average blood pressure measured on the day of the monitoring with a standard mercury sphygmomanometer was 121 f 11176 -t 8 mm Hg in the normotensive subjects and 151 2 16/103 + 9 mm Hg in the hypertensive men.

The average of the blood pressures obtained during various time periods are depicted in Table II. In both normotensive and hypertensive men, night- time blood pressure averages are significantly lower and daytime averages and morning (800 to 1000 hours) averages significantly higher than the aver- age of the whole-day blood pressure. The difference between the daytime and nighttime average systolic blood pressure was greater in normotensive volun- teers than in hypertensive men (19 t 8 and 14 s 13 mm Hg, respectively, p < 0.05). The difference between daytime and nighttime diastolic blood pres- sures was similar for the two groups of men (12 + 7 and 10 f 8 mm Hg, not significant). The difference between daytime and nighttime blood pressure was significantly greater for systolic than for diastolic blood pressure in normotensive volunteers (p < 0.01) and in hypertensive patients (p < 0.05).

The variability of systolic and diastolic blood pressure, defined as the standard deviation of all blood pressures measured during the 24-hour moni- toring period, was similar for normotensive and hypertensive men (15 +- 4/11 + 3 mm Hg in normo- tensive men and 16 + 5/12 + 3 mm Hg in hyperten- sive patients). However, in both groups of men, the

June, 1985

1336 Drayer, Weber, and Nakamura American Heart Journal

Table II. Comparison of averages of blood pressures (mm Hg) obtained during various periods of blood pressure monitoring in normotensive and hypertensive men

Whole-day

aoerage

Daytime

average

Nighttime

average

8:0o-lO:00 A.M.

average

Normotensive

(n = 29) Systolic blood pressure 120 +- 11

Diastolic blood pressure 75 k 7 Hypertensive

(n = 29)

Systolic blood pressure 140 k 18

Diastolic blood pressure 92 k 12

*p < 0.01 versus whole-day average blood pressure.

126 + ll* 107 f 10* 125 * 11*

79 _t 7* 67 +- 8’ 80 i- 7%

145 t 18* 132 f 21’ 147 t 24*

96 + ll* 86 If 13* 97 * 1?l*

variability was significantly 0, < 0.05) greater for systolic blood pressure than for diastolic blood pressure.

The averages of systolic and diastolic blood pres- sures obtained during the 12 consecutive 2-hour periods are shown in Fig. 1. The circadian pattern of blood pressure is not different in normotensive and hypertensive men, but it is established at higher levels of blood pressure in the hypertensive patients.

Prevalence of abnormally high blood pressures. The incidence of arbitrarily defined abnormal blood pressures observed during a 24-hour day in depicted for each of the participants in the study in Fig. 2. The incidence of systolic blood pressures >140 mm Hg ranges from 0% to 67 % in normotensive (aver- age 12 f 15 % ) and between 2 % and 97 % in hyper- tensive individuals (average 46 + 28%). The inci- dence of diastolic blood pressures >90 mm Hg ranged between 0 % and 64 % (average 14 +- 15 % ) in normotensives and between 8% and 99% (average 56% + 28%) in hypertensive men.

Predictability of 24-hour blood pressure averages. The relationships between the average of blood pressures measured between 800 and 1000 hours in the morning and the whole-day averages of systolic and diastolic blood pressures are given in Fig. 3. The calculated regression line as well as the correlation coefficient are similar for these age-matched normo- tensive and hypertensive men. In most subjects, the morning average is higher than the whole-day aver- age of systolic or diastolic blood pressure.

in age-matched groups of the same sex. In addition, both groups of participants were comparable with respect to the degree of obesity. Thus, the results observed cannot be related to these parameters. In both groups of men, nighttime blood pressure aver- ages were significantly lower than daytime averages of blood pressure. The amplitude of the difference between daytime and nighttime systolic blood pres- sures was greater in normotensive than in hyperten- sive subjects. Diastolic blood pressure decreased slightly more during the night in normotensive than in hypertensive individuals. Irving et a1.7 also men- tioned a greater fall in blood pressure during sleep in normotensive control subjects. In the study by Pick- ering et a1.,2 the difference between highest daytime and lowest nighttime blood pressure was similar for normotensive individuals and hypertensive subjects. It is possible that the increased daytime to night- time blood pressure amplitude observed in our study is one of the characteristics of the normotensive control subjects and this characteristic may disap- pear during the development of high blood pres- sure.

DISCUSSION

Results of 24-hour ambulatory blood pressure monitoring also can be expressed as a circadian pattern of blood pressure. In general, the circadian pattern of blood pressure is not different for normo- tensive individuals, patients with borderline hyper- tension,3 and younger or older patients with estab- lished hypertension. 3* 5 However, the blood pressure patterns of these subgroups are displayed at differ- ent levels of blood pressure. The whole-day average of blood pressures best defines this difference between the subgroups.1~3~4

BP patterns. The results of this study in normoten- BP variability. Variability of blood pressure com- sive and hypertensive men reveal a similarity in the monly is defined as the standard deviation of all pattern of blood pressure observed during the 24- measurements of blood pressure obtained during a hour day in both groups of participants. It is the first certain time period. We found that the variability of study published in which the results were obtained blood pressure is similar for normotensive control

Volume 109

Number 6 Ambulatory BP monitoring in hypertension 1337

.

NORMOTENSIVE HYPERTENSIVE

. . -

:

I NORMOTENSIVE HYPERTENSIVE

Fig. 2. Incidence of “abnormal” blood pressures in nor- motensive and hypertensive men.

subjects and hypertensive patients. The lack of a difference in blood pressure variability in normoten- sive and hypertensive men has frequently been reported.‘e3 However, variability is, to a certain extent, related to the height of the blood pressure. Thus, variability of systolic blood pressure is signif- icantly greater than that of diastolic blood pres- sure.2v3s5 Variability of blood pressure may increase with age.5z8

The difference between the daytime and night- time systolic blood pressure was greater in normo- tensive volunteers than in hypertensive men. This contrasts with the similarity of the variability of systolic blood pressure between the two groups. The daytime-nighttime amplitude of systolic blood pres- sure might define a different aspect of blood pres- sure control than variability of blood pressure. Obvi-

L

A

SYSTOLIC BLOOD PRESSURE

/

y=x

/

A

.

A.. l NORMOTENSIVE m

Y = .7x + 32 R = -75

AHYPERTEKSIVE

Y = .7x + 33 R = -93

I I 1 I 120 140 160 160 200

8:oo - io:oo AM AVERAGE BLOOD PRESSURE (MM HG)

DlASTOLIC BLOOD PRESSURE /

. NORMOTENSIVE

Y = .8x + 10 R = -84

b/6. . l r : . t AHYPERTENSIVE

Y = .8x + 14 R = .83

80 100 LO 140

8:oo - lo:00 AN AVERAGE BLOOD PRESSURE iMM HG)

Fig. 3. Relationship between 8:00 to 10~00 A.M. and whole-day blood pressure averages in normotensive and hypertensive men.

ously, various normal daily activities, e.g., physical activity, play a major role in the determination of blood pressure variability, whereas a decrease in sympathetic nervous system activity during sleep will determine blood pressure levels observed during sleep.

Prevalence of abnormal blood pressures. Alterna- tively, one can define the various levels of blood pressure in the subgroups as the prevalence of arbitrarily chosen “abnormal” blood pressures (e.g., systolic blood pressure >140 mm Hg and/or diastolic blood pressure >90 mm Hg) during the day. It has been shown that the incidence of abnormal blood pressures is greater in older patients with estab- lished hypertension than in younger hypertensive patients, in patients with borderline hypertensives, or in normotensive control subjects.“6 Arbitrarily

1338 Drayer, Weber, and Nakamura

chosen abnormal levels of systolic blood pressure were found in 46% of all systolic blood pressures observed during 24 hours in hypertensive subjects, but also in 12% of systolic blood pressures recorded on tracings of normotensive men. Abnormal diastol- ic blood pressures were found in 56 % of each tracing in hypertensive individuals and in 14% of data in tracings of normotensive control subjects. The data in our study reveal a marked range in the incidence of abnormal pressures in normotensive and hyper- tensive men. Three normotensive men (10%) had more than 25% of abnormal systolic blood pressures and five hypertensive subjects had fewer than 25? abnormal readings of systolic blood pressure. Simi- larly, five normotensive subjects (14%) had more than 25% of abnormal diastolic blood pressures and three hypertensive patients had fewer than 25% of abnormal readings. Kennedy et al.fi also have shown a high incidence of abnormal systolic (0 % to 45 % ) or diastolic (0% to 19%) blood pressures in normo- tensive subjects, and they reported that abnormal pressures were more often found in older than in younger control subjects. In contrast, 13 of the 42 patients with borderline or established hypertension studied by Horan et al.’ had fewer than 25% of abnormal systolic or diastolic blood pressures during the 24 hours of blood pressure monitoring.

Predictability of 24-hour blood pressure averages.

Casual blood pressures do not necessarily predict the level of blood pressure recorded during a 24- hour period. Correlation coefficients reported thus far all show a fairly weak relationship between casual blood pressure and the whole-day average blood pressure.2, *, 7, g We have shown that averages of multiple blood pressures obtained during the morn- ing hours correlate much better with 24-hour aver- ages of blood pressures4, 5. 9 than, casual blood pres- sures. In this study, we were able to show that the relationship between these two parameters of blood pressure is similar for normotensive and hyperten- sive subjects. Casual blood pressures do not reliably predict whole-day averages of blood pressure in normotensive or hypertensive men. This also is

June. 1985

American Heart Journal

reflected in the marked overlap found in the preva- lence of abnormal blood pressures between the two groups. The correlation coefficients, the slopes of the regression line, as well as the intercept were not different for the relationship between 8:00 to 10:00 A.M. average and whole-day average of systolic or diastolic blood pressure for normotensive or hyper- tensive men. Thus, in both groups the average of the blood pressures obtained during 2 hours in the morning may be used to reliably predict the 24-hour average of blood pressure.

The data in this study reveal that 24-hour blood pressure monitoring may serve to better define different characteristics of blood pressure in normo- tensive and hypertensive men. Long-term follow-up studies need to be performed so as to show that a better definition of hypertension also will lead to a better prediction of the prognosis in hypertensive patients. REFERENCES

Horan Md. Kennedy HL, Padgett NE: Do borderline hyper- tensive patients have labile blood pressure? Ann Intern Med 94:466. 1981. Pickering TG. Harshfield GA, Kleinert HD, Blank S, Laragh .JH: Blood pressure during normal daily activities. sleep, and exercise. JAMA 247:992, 1982. Messerli FH, Glade LB, Ventura HO, Dreslinski CR, Suarez DH. et al: Diurnal variat.ions of cardiac rhythm, arterial pressure, and urinary catecholamines in borderline and established essential hypertension. AM HEART .J 104:109. 198’. Drayer JIM, W-eber MA, DeYoung JL: The use of short,-term amhulatory BP monitoring in differing forms of hyperten- sion. Clin Exp Hypertens A5:1597, 1983. Drayer *JIM, Weber MA, DeYoung JL, Wyle FA: Circadian Mood pressure patterns in ambulatory hypertensive patients. Am J Med 73:493, 1982. Kennedy HL, Horan MH. Sprague MK, Padgett NE, Shriver KK: Ambulatory blood pressure in healthy normotensive males. AM HEART J 106:717, 1983. Irving .JB, Brash HM, Kerr F, Kirby BJ: The value of ambulatory monitoring in borderline and established hyper- tenwn. PC&grad Med J 52(suppl 7i:137, 1976. Mancia G. Ferrari A, Gregorini L, Parati G, Pomidossi G, et al: Blood pressure variability in man: Its relation to high blood pressure, age and baroreflex sensitivity. Clin Sci 59:4Ols, 1980. Weher MA, Drayer JIM, Wyle FA, Brewer DD: A representa- tive value for whole-day BP monitoring. JAMA 248:1626, 1982.