circadian blood pressure patterns in ambulatory hypertensive patients: effects of age

Circadian Blood Pressure Patterns in Ambulatory

Hypertensive Patients

Effects of Age

JAN I. M. DRAYER, M.D.

MICHAEL A. WEBER, M.D.

JANICE L. DeYOUNG

FREDERIC A. WYLE, M.D.

Long Beach and Irvine, California

From the Section of Clinical Pharmacology and Hypertension, Department of Medicine, Veterans Administration Medical Center, Long Beach, California, and the University of California, Irvine, California. Requests for reprints should be ad- dressed to Dr. Jan I. M. Drayer. Hypertension Center, V.A. Medical Center, 5901 East Seventh Street, Long Beach, California 90822. Manuscript accepted February 23, 1982.

Circadian blood pressure monitoring was performed in 50 untreated ambulatory hypertensive patients to study the effects of age on the pattern and variabtf!ty of Mood pressure and heart rate. Casual blood pressure, measured in the morning, was greater than the average of the blood pressures measuked at 7.5 minute intervals for 24 hours (148 f 2/g5 f 2 and 137 f 2/88 f 2 mm Hg, p <O.OOl). The correlation between casual systolic pressure and the 24 hour average was stronger (p <O.OS) in younger (less than 55 years of age) patients (r = 0.69, n = 24, p KO.001) than In older patients (r = 0.42, n = 26, p <O.l ). Similarly, diastolic pressures correlated more strongly (p <0.05) in younger patients (r = 0.71, p <O.OOl) than in older patients (r = 0.43, p <0.05). Variability of systolic pressure, defined as the standard deviation of all readfngs obtained during 24 hours, was greater than that of diastolic pressure (16.7 and 13.1 mm Hg, respectively, p <O.OOl). Moreover, the variability of systolic pressure was greater in older than in younger patients (18.1 and 15.2 mm Hg, respectively, p <O.Ol). The variability of diastolic pressure was slightly but not sfgnificantiy greater in older patients (13.7 and 12.5 mm Hg, not signfficant). The circadian pattern of bfood pressure, expressed as averages of readings obtained during consecutive 2 hour intervals, was similar in the two age groups. However, the level of systolic pressure was consistently higher (p <O.Ol ) and that of both diastolic pressure and heart rate consistently lower (p <O.Ol) in older patients. Thus, ambulatory circadian blood pressure monitoring reveals significant changes in blood pressure levels and its variability with age; the casual blood pressure does not accurately reflect these changes. Longer periods of blood pressure monitoring are required for accurate assessment of the characteristics of hypertension in the aged.

Noninvasive ambulatory circadian blood pressure monitoring was recently introduced as a technique to better define blood pressure in patients with borderline hypertension. This technique has also been used to quantify variability of blood pressure and to assess the effects of antihypertensive therapy [l-3]. Early experiences using blood pressure monitoring during the day [4] or at night [5] show that the variability of blood pressure increases with the height of the blood pressure, but not with patient’s age. Other studies using circadian ambulatory blood pressure monitoring equipment fail to agree whether variability of blood pressure is related to age [6-81. Important infor- mation about circadian patterns of blood pressure and heart rate also can be obtained using this technique. It is well known that blood pressure is lower during the night and higher during the day.

October 1982 The American Journal of Medicine Volume 73 493

CIRCADIAN BLOOD PRESSURE MONITORING AND AGE-DRAYER ET AL.

It is likely that the circadian rhythm of hormones such as renin and catecholamines is at least partly respon- sible for the circadian pattern of blood pressure. Indeed, a significant association was found between plasma norepinephrine levels and blood pressure levels measured at 30 minute intervals throughout the day [9]. The relation between the circadian patterns of blood pressure and the renin system remains to be elucidated.

In this study we report data obtained during noninvasive automated ambulatory blood pressure monitoring during 24 hours in 50 patients with hypertension. The results suggest that some of the characteristics of 24 hour blood pressure patterns change with age. Older patients have higher systolic blood pressures and lower diastolic blood pressures and heart rates throughout the 24 hour period. The variability of systolic blood pressure increases with age. Therefore, longer periods of blood pressure recording might be needed to accurately assess blood pressure levels in the aged.

METHODS

Patients with mild to moderate hypertension were selected for this study. They had supine diastolic blood pressures between 90 and 115 mm Hg as measured by a conventional mercury sphygmomanometer. The diagnosis of essential hypertension was established using standard clinical and biochemical procedures. Patients with diabetes mellitus, congestive heart failure, renal insufficiency, or clinical significant liver disease were excluded from the study. The patients were free of all antihypertensive medication for at least 2 weeks before the study. All patients signed an in- formed consent before ambulatory blood pressure monitoring.

On the day of the study, the patient’s height and body weight were obtained, and casual blood pressure and heart rate were measured. Casual blood pressure was defined as the average of two readings taken after 10 minutes of supine rest. Casual pressures were not obtained in six patients. A blood sample for measurements of serum potassium and creatinine and for plasma renin activity was obtained. The patients had been ambulatory for at least 1 hour before the

drawing of blood for plasma renin activity. The patients began a 24 hour urine collection for measurement of sodium excretion.

Automated ambulatory blood pressure monitoring was then performed during 24 hours using the Pressurometer Ill (Del Mar Avionics, Irvine, California [lo]). With this instrument, readings of systolic and diastolic blood pressure and heart rate were obtained at 7.5 minute intervals throughout the day. After completion of the monitoring, the data were printed using a 1979 Pressurometer Charter (Del Mar Avionics).

The following readings were deleted arbitrarily from each 24 hour tracing before further analysis: inconsistent increases or decreases in systolic blood pressure, diastolic blood pressure, or heart rate of greater than 20 mm Hg or 20 beats/min, and systolic and diastolic blood pressure readings with a calculated pulse pressure (systolic blood pressure - diastolic blood pressure) of less than 5 mm Hg. Further analysis of tracings was only performed if at least 75 percent of the maximal number of 192 readings per 24 hours passed the deletion criteria. Statistical analysis of the tracings was performed using an Apple II Plus microcomputer. The following analyses of systolic blood pressure, diastolic blood pressure, and heart rate were obtained from each tracing: average of all readings obtained during the 24 hour monitoring, standard deviation of all data (defined as variability of blood pressure or heart rate), average of the readings obtained during 12 consecutive 2 hour periods starting at mid- night, and the standard deviation of these 2 hour averages. Averages of daytime and nighttime blood pressures were calculated for each patient. Daytime was defined as the time interval between 0600 and 2200 hours (46 of the 50 patients had their highest pressures recorded during this period). Nighttime was defined as the time interval between 2200 and 0600 hours (42 of the 50 patients had their lowest blood pressures recorded during this period).

Chemical test results were obtained using standard labo- ratory techniques. Plasma renin activity was measured by radioimmunoassay [ 1 I].

Standard parametric statistical tests, paired and unpaired Student t tests, and Pearson correlation coefficients were used in the analysis of the data. As an aid to analysis, the patients were subdivided into younger (55 years of age or less) and older (greater than 55 years of age) subgroups. Data

TABLE I Comparison of Casual Blood Pressure and Heart Rate With Parameters Obtained During 24-Hour Monitoring in 50 Patients With Hypertension

24-hour Highest Lowest Casual’ p Average Variabilityt P-hour Average 2.hour Average

Systolic blood pressure 148 f 2 <O.OOl 137 f 2 16.7 f 0.6 155 f 2 116f 2

(mm H9) Diastolic blood pressure 95 f 2 <O.OOl 8842 13.1f0.4 104f 2 74 f 2

(mm H9) Pulse pressure 53 f 2 <O.OOl 48 f 2 14.3 f 0.5 51 f2 35 f 2

(mm H9) Heart rate 74 f 2 ns 75 f 2 11.8 f 0.5 89 f 2 62f 1

(beatsjmin)

f n = 44. 7 Variability is defined as the standard deviation of all readings obtained during 24 hours. ns = not significant.

494 October 1992 The American Journal of Medlclne Volume 73


obtained during automated ambulatory 24 hour blood pressure monitoring were compared between the two subgroups. All data are expressed as mean f standard error of the mean. Two-tailed p values are used to express significance of the results of the statistical tests.

RESULTS

Fifty male patients, aged 24 to 84 years (mean 50.9 f 2.3 years), were selected for the study. Three patients were black. Average height was 1.75 f 0.01 m and body weight 82.7 f 1.8 kg. All patients had normal concentrations of potassium (mean 4.2 f 0.1 mEq/liter) and creatinine (mean 1.20 f 0.03 mg percent) in serum. Plasma renin activity was measured in 39 of the 50 patients and averaged 2.6 f 0.2 ng/ml/hour. Urinary sodium excretion was 139 f 11 mEq/24 hours.

Measurements of casual blood pressure and heart rate, average and variability of the 24 hour tracing, and the highest and lowest 2 hour averages are given in Table I. Casual systolic blood pressure, diastolic blood pressure, and pulse pressure are significantly higher than the 24 hour average of these variables (p <O.OOl). The variability of systolic blood pressure was significantly greater than that of diastolic blood pressure (16.7 f 0.6 versus 13.1 f 0.4 mm Hg, p <O.OOl).

The prevalence of the highest and lowest averages for systolic blood pressure, diastolic blood pressure, pulse pressure, and heart rate for each of the 12 2-hour periods of the day are given in Figure 1. The lowest 2 hour averages for systolic blood pressure, diastolic blood pressure, and heart rate are most commonly obtained between 10 PM and 6 AM. In fact, 42 of the 50 patients recorded their lowest 2 hour average for systolic blood pressure during that time period, whereas the highest 2 hour averages were found between 6 AM

and 10 PM in 47 patients. Similarly, 42 patients had the lowest diastolic pressure between 10 PM and 6 AM and 45 had the highest averages between 6 AM and 10 PM.

The lowest averages for heart rate were obtained during the night in 38 patients and the highest during the day

SYSTOLIC BLOOD PRESSURE 0 HlGHE!ST Pk AVERAGE m LOWEST2hr AVERAGE

LIC BLOOD P

HEART RATE -

PULSE PRESSURE

IO

0

IO

0 4.m 8 12 4pm 8 12

TIME OF DAY (hn)

Figure 1. Prevalence (%) of highest and lowest 2 hour averages of blood pressure and heart rate related to time of day.

in 46 patients. Interestingly, the lowest 2 hour average for pulse pressure was obtained during the night in only 25 patients. The time relation of lowest and highest values for blood pressure and heart rate was similar for younger and older patients.

Correlations between casual blood pressure and heart rate and variables obtained during 24 hour monitoring are given in Table II. Casual systolic blood pressure correlated significantly with the average 24

TABLE If Correlations Between Cawal Supine Blood Pressure and Heart Rate Values and Values Obtained During 24-Hour Monitoring in 44 Patients Wtth Hypertension

24-hour Average

Casual supine Systolic blood

pressure Diastolic blood

pressure Pulse pressure

Heart rate

0.54

p <O.OOl 0.58

p <O.OOl 0.60

p <O.OOl 0.46

p <O.Ol

Variability’ Highest

2-hour average

0.08

0.19

0.24 p <O.l

0.19

0.64 p <o.ooi

0.45 p <O.Ol

0.53 p <O.OOl

0.43 p <O.Ol

Lowest P-hour aversae

0.25

0.55 p <O.OOl

0.47 p <O.Ol

0.38 p co.05

l Variability is defined as the standard deviation of all readings obtained during 24 hours.

October lM2 The American Journal of Medklne Volume 73 495

CIRCADIAN BLOOD PRESSURE MONITORING AND AGE-DRAYER ET AL

TABLE III Correlation Coefficients (r Values) Between Casual Supine Blood Pressure and Heart Rate Values and the Averages of Multiple Readings (Each 7.5 min) Obtained During Daytime (0600 to 2200 hours) or Nighttime (2200 to 0600 hours) in 44 Patients with Hypertension

Systolic blood pressure

Diastolic blood pressure

Heart rate

Daytime

0.54 p <O.OOl

0.62 p <O.OOl

0.39 p <0.05

Nighttime

0.17

0.38 p <0.05

0.34 p <0.05

hour systolic blood pressure and the highest 2 hour average, but not with the 24 hour variability or the lowest 2 hour average. Casual diastolic blood pressure did not correlate with the 24 hour variability, but correlated significantly with the 24 hour average diastolic blood pressure and with both the highest and the lowest 2 hour averages. Correlations between casual supine blood pressure and the average of the full 24 hour blood pressures were not different from those between the casual blood pressure and the average of the daytime (between 0600 and 2200 hours) blood pressures (Table III). However, correlations between casual supine blood pressure and the average of the nighttime (between 2200 and 0600 hours) blood pressures were significantly (p <O.Ol) weaker than those between casual pressure and daytime pressure. Correlations between standing casual pressure and daytime or nighttime pressure were not significantly different from those

TABLE IV Comparison of Clinical and Biochemical Measurements in Younger (55 Years of Age or Less, n = 24) and Older (Over 55 Years of Age, n = 26) Patients With Hypertension

Younger Older Patients P Patients

Age (yr) 36.4 f 1.9 <O.OOl 64.3 f 1.1 Height (m) 1.77 f 0.02 1.74 f 0.01 Body weight 85.7 f 2.8 80.1 f 2.2

(kg) Serum potassium 4.1 f 0.1 4.3 f 0.1

(mEq/liter) Serum creatinine 1.10 f 0.03 1.17 f 0.05

(mg/dl) Plasma renin activity 3.3 f 0.3 <0.005 2.0 f 0.3

(ng/ml/hour) Urinary sodium excretion 149.0 f 17.0 130.0 f 15.0

(mEq/24 hours)

between casual supine pressure and daytime or nighttime pressure. However, casual standing pressure correlated more closely with daytime pressure (r = 0.64 for systolic and diastolic pressure) than with the average of the pressures obtained during the full 24 hours (r = 0.46 for systolic and r = 0.51 for diastolic PreSSWe).

Casual pulse pressure correlated significantly with all four variables of the 24 hour monitoring (Table II). Heart rate did not correlate with variability, but correlated significantly with the other three variables. Correlations between 24 hour averages of blood pressure and heart rate and their variability were not significant for systolic pressure (r = 0.08), diastolic pressure (r = 0.19), and heart rate (0.19) and only weakly significant for pulse pressure (r = 0.24, p <O.l). Correlations between plasma renin activity and variability of systolic blood pressure (r = -0.24), variability of diastolic blood pressure (r = -0.02), variability of pulse pressure (r = -0.1 l), and that of heart rate (r = 0.16) were not significant.

To study the effect of age on characteristics of the 24 hour blood pressure pattern, the patients were subdivided into younger and older subgroups. Clinical and some biochemical data for these subgroups are given in Table IV. The mean age of the younger group was 36 years and that of the older group 64 years. Height, body weight, serum potassium, and serum creatinine were not different in the two groups. Plasma renin activity was significantly lower in the older group despite similar urinary sodium excretion levels in both groups. The pattern of the 2 hour averages for systolic blood pressure, diastolic blood pressure, and heart rate are given in Figure 2. Systolic blood pressure was consistently higher (p <O.Ol) and diastolic blood pressure (p <O.Ol) and heart rates (p <O.Ol) were consistently lower in the older subgroup. Moreover, the standard deviation of the 2 hour averages for systolic blood pressure was consistently greater in older patients than in the younger subgroup (p <O.Ol).

The data of the other variables of circadian blood pressure monitoring are given in Table V. Casual systolic blood pressure was slightly higher in the older patients. The variability of systolic blood pressure was significantly greater in older than in younger patients. No differences were observed between the two subgroups in diastolic blood pressure or its variability. However, casual pulse pressure as well as variability of pulse pressure was greater in older pati,ents. The variability of heart rate decreased with age.

Correlations between casual blood pressure and heart rate and the 24 hour averages of these data and between 2 hour averages of these data obtained between 8 and 10 AM and the 24 hour averages are given in Table VI. Two hour averages predicted the 24 hour

496 October 1962 The American Journal of Medicine Volume 73


Figure 2. Comparison of circadian blood pressure patterns in ambulatory younger (55 years of age or less, n = 24) and older (over 55 years of age, n = 26) patients with hypertension.

_ YOUNGER PATIENTS

1 1 I 1 I 1 I 1 I 1 1

2om 4 6 8 IO 12 2pm 4 6 6 IO 12

TIME OF DAY (hrs)

TABLE V Comparison of Different Variables of Blood Pressure and Heart Rate Between Younger (55 Years of Age or Less, n = 24) and Older (Over 55 years of age, n = 26) Patients With Hypertension

TABLE VI Correlations Between Casual Bupina Blood Pressure and Heart Rate and the Avemgas of Blood Rassums and Heart

YOUngef older Rates kcordad Batween 8’ and 10 AM

Patients p Patients and the Averagos of Blood pressures and

Casual systolic pressure 143f3 <O.l 153f3 Heart Rates Obtainad Dwimg the Whole

, . . . Day (Corrdations Given for tha Whole (mm w

Average 24 hour systolic 135 f 3 pressure

(mm Hg) Variability of systolic pressure 15.2 f 0.7 <O.Ol

(mm Hg)’ Casual diastolic pressure 96 f 3

(mm Hg) Average 24 hour diastolic 69 f 2

pressure

(mm Hg) Variability of diastolic pressure 12.5 f 0.5

(mm Hg)’ Casual pulse pressure+ 47 f 2 <O.Ol

(mm Hg) Average 24 hour pulse pressure 46 f 2

(mm Hg) Variability of pulse pressure 13.4 f0.6 <O.l

(mm Hg)’ Casual heart rate+ 74 f 2

(beatsimin) Average 24 hour heart rate 76 f 2

(beats/min) Variability of heart rate 12.6 f 0.6 <O.Ol

(beats/min)’

136f3 Group of Patients and for subgroups of Younger and Older Patients)

18.1 f 0.8

94 f 2

87 f 2

13.7 f 0.6

59 f 3

49 f 3

15.2 f 0.7

74 f 3

74 f 2

11.0 f 0.7

l Variability is defined as the standard deviation of all readings obtained during 24 hours. + n = 22 in each subgroup.

24-hoer Averqes Youa9WPEumlts older Pattents

whets Group ((55 Yesrs f&f) 055 Years Old) In = 50) In = 24) In = 26)

Casual readings Systolic r = 0.56

blood p <O.OOl pressure l

Diastolic r = 0.58 blood p <O.OOl pressure l

Heart rate* r = 0.37 p <0.05

8- 10 AM averages Systolic r = 0.85

blood p <O.OOl pressure

Diastolic r = 0.79 blood p <O.OOl pressure

Heart rate r = 0.64 p <O.OOl

r = 0.69 p <O.OOl

r = 0.71 p <O.OOl

r = 0.43 p <O.l

r = 0.52 r = 0.13 p <0.05 ns

r = 0.88 p <O.OOl

r = 0.84 p <O.OOl

r = 0.88 p <O.OOl

r = 0.90 p <O.OOl

r = 0.42

p <O.l

r = 0.70

p <O.OOl

r = 0.36 p <O.l

l n = 44 for group as a whole and n = 22 within each subgroup. ns = not significant.

October 1982 The American Journal d Medhdne Volume 73 497


average of blood pressure and heart rate more closely than did casual readings, particularly in older patients. Moreover, 2 hour averages of blood pressure obtained in the morning correlated significantly with both daytime and nighttime averages of pressure (r = 0.89 and r = 0.71 for systolic pressure and r = 0.87 and r = 0.62 for diastolic pressure, respectively, in all cases p <O.OOl). Correlations between 2 hour and 24 hour averages of the data were significantly greater than those between casual readings and 24 hour averages. Correlations between the variability of blood pressure observed between 8 and 10 AM and the variability of all pressures recorded during a 24 hour day were 0.37 for systolic blood pressure and 0.08 for diastolic blood pressure for the group as a whole.

COMMENTS

Noninvasive automated ambulatory monitoring of blood pressure was performed in 50 male patients with mild to moderate hypertension. Using this technique infor- mation was obtained about circadian blood pressure patterns and variability of blood pressure. We used the averages of blood pressure and heart rate obtained during 12 2-hour periods to represent circadian patterns for the full 24 hours. This approach does not allow the evaluation of acute changes in these measurements. The variability of blood pressure and heart rate was expressed as the standard deviation of all data obtained during the day of monitoring as described in other studies [5,6,8]. In contrast to others [4,5,8], we found evidence of significant changes in variability of blood pressure and heart rate with age. Compared with younger patients (55 years old or less), older patients showed an increase in variability in systolic blood pressure and pulse pressure and a decrease in variability in heart rate. In agreement with Bevan et al. [ 121, we found that the variability of systolic blood pressure was greater than that of diastolic blood pressure. This difference was significant in both younger and older subgroups but was somewhat more pronounced in the older patients.

The circadian pattern of blood pressure and heart rate did not change with age, but systolic pressure was consistently higher and diastolic pressure and heart rate consistently lower in older patients. In both younger and older patients, lowest pressures were more frequently found between 10 PM and 6 AM and highest between 6 AM and 10 PM. Indeed, blood pressure increased most clearly early in the morning. In contrast to results from an earlier study [7], maximal blood pressure for the day was most commonly reached between 10 AM and 12 noon and not early in the evening. Heart rate followed the systolic and diastolic pressure patterns closely,

lowest values being most common during the night and highest during the day. Interestingly, pulse pressure did not show such a pattern. Highest values occurred as commonly during the day as during the night. Thus, pulse pressure does not change appreciably during the full day and in particular does not decrease during the night even when systolic and diastolic blood pressures reach their lowest values.

Although plasma renin activity is lower in older patients [ 131, correlations between plasma renin activity and the variability of blood pressure or heart rate were not significant. Watson et al. [ 141 reported a significant correlation between the variability of systolic blood pressure and age-corrected values of upright plasma renin activity. We could not confirm these data, and correlations between variability in blood pressure and renin activity were not significant within the subgroups of younger or older patients.

Noninvasive automated monitoring of blood pressure would not be a major advance if casual pressures or short-term monitoring could provide reliable measures of whole-day blood pressure and its variability. Casual blood pressure values and heart rate did not correlate with variability of these measurements during the day. Moreover, only weak correlations were found between the variability of blood pressure observed during 2 hours in the morning and the variability of blood pressure recorded during the day. Therefore, the assessment of variability of blood pressure might necessitate 24 hour monitoring. Casual blood pressure predicted the 24 hour blood pressure fairly accurately (r = 0.57) in the group as a whole. Both supine and standing casual pressures correlated significantly with daytime pressure averages but not significantly or only weakly with the nighttime pressure averages. Interestingly, the average of blood pressures obtained during 2 hours in the morning correlated significantly with the 24 hour, daytime, and nighttime averages.

Casual pressures and heart rates were poor predic- tors of 24 hour averages of these data particularly in older patients. In this subgroup, 2 hour averages of systolic and diastolic blood pressure obtained between 8 and 10 AM were significantly more accurate in pre- dicting 24 hour averages of these pressures (r = 0.88 and 0.84, p X0.001). The data suggest that longer periods of monitoring might be needed in older patients to adequately assess their whole-day blood pressure. Monitoring for 2 hours in the morning could be an ade- quate alternative to the full 24 hour study and certainly would be less cumbersome and more convenient. It remains to be shown whether or not longer periods of blood pressure monitoring are mandatory before the institution of antihypertensive therapy in the elderly patient.

499 October 1982 The American Journal of Medlclne Volume 73


REFERENCES

1. Clement DL: Blood pressure variability. Baltimore: University Park Press, 1979; 13-107.

2. Horan MI, Kennedy HL, Padgett NE: Do borderline hyper- tensives have labile blood pressure? Ann Intern Med 1981; 94: 466-466.

3. Mann S, Millar Craig MW, Balasubramanian V, Raftery EB: Propranolol LA and ambulatory blood pressure. Br J Clin Pharmacol 1960; 10: 443-447.

4. Kain HK, Hinman AT, Sokolow M: Arterial blood pressure measurements with a portable recorder In hypertensive patients. Circulation 1964; 30: 682-892.

5. Drayer JIM, Hcefnagels WHL, Kloppenbrog PWC: Automated blood pressure recording versus conventional manometry: comments on blood pressure variability. Neth J Med 1976; 19: 8-15.

6. Mancia G, Ferrari A, Gregorini L, et at.: Blood pressure variability in man: its relation to high blood pressure, age and baroreflex sensitivity. Clin Sci 1980; 59: 4OlS- 404s.

7. Richardson DW, Honour AJ, Fenton GW, Stott FH, Pickering GW: Variation in arterial pressure throughout the day and night. Clin Sci 1964; 26: 445-460.

8. Watson RDS, Stallard TJ, Luttler WA: Factors determining the variability of arterial pressure in hypertension. Clin Sci 1979; 57: 2838-2858.

9. Sowers JR: Dopam&@ control of cfrcadii norepinephrine levels in patients with essential hypertension. J Clin En- docrinol Metab 1981: 53: 1133-1137.

10. Harshfield GA, Pidtering TG, Laragh JH: A validation study of the Del Mar Avionics ambulatory blood pressure system. Ambulatory Electrocardiography 1979; 4: 7-12.

11. Chen TC, Weber MA: Effects of di-idopropylfluorophosphate on the measurement of plasma renin activity. Clin Chim Acta 1980; 107: 217-222.

12. Bevan AT, Honour AJ, Stott FH: Direct arterial pressure re- cordings in unrestricted man. Clin Sci 1969; 36: 329- 344.

13. Drayer JIM, Weber MA, Sealey JE, Laragh JH: Low and high renin essential hypertension: a comparison of clinical and biochemical characteristics. Am J Med Sci 1981; 281: 135-142.

14. Watson RDS, Stallard TJ, Flurn RM, Littler WA: Factors determining direct arterial pressure and its variability in hypertensive man. Hypertension 1980; 2: 333-341.

October 1992 TheAmerfcanJoumrlef~ vdurn73 499

circadian blood pressure patterns in ambulatory hypertensive patients: effects of age

Documents