Ambulatory Blood Pressure Monitoring in Acute Stroke
The West Birmingham Stroke Project
Background and Purpose Ambulatory blood pressure monitoring (ABPM) devices are increasingly used in the assessment of hypertension, but their value in patients after a stroke is unknown, despite the fact that hypertension is an important cause of stroke and many patients have relatively high blood pressure (BP) levels at presentation. We therefore investigated the clinical use of a 24-hour oscillometric ABPM device in patients after acute stroke. We also investigated ABPM in different types of stroke (thrombosis, hemorrhage, and transient ischemic attack) and ethnic and sex differences.
Methods BP was measured manually with a standard mercury sphygmomanometer, and ABPM measurements were made with an oscillometric device. The first reading obtained with the ABPM device was compared with simultaneous manual BP measurements. Mean daytime and nighttime pressures were also analyzed to determine the frequency of nocturnal BP falls (“dipping”).
Results We studied 86 patients (48 men; mean±SD age, 64.2±9.2 years) admitted with acute-onset stroke (ictus within 12 hours) in a district general hospital. Thirty-one patients (36.0%) had a previous history of hypertension. The median percentage of successful BP readings by ABPM was 92% (interquartile range, 72 to 98). There was no significant difference in manual BP levels compared with the first simultaneous systolic or diastolic ABPM measurements. Systolic BPs recorded by ABPM were significantly higher in black patients with acute stroke and in patients with intracerebral hemorrhage, who also showed a trend toward higher nocturnal BPs. There was no difference in BPs between men and women and those who were alive or dead 6 months later (P=NS). There was also no difference between mean day and night systolic BP (mean difference, 1.9 mm Hg; P=.08), although mean daytime diastolic BP was higher than mean nighttime diastolic BP (mean difference, 2.4 mm Hg; P=.01). Patients with stroke therefore demonstrated a loss of diurnal BP rhythm and may be considered “nondippers”; there was also a trend toward “reverse dipping” in patients with intracranial hemorrhage.
Conclusions This study demonstrates higher systolic BPs as recorded by ABPM (but not manually) in patients with intracerebral hemorrhage than in those with cerebral infarcts; higher levels were also found in blacks. ABPM recordings are useful in the assessment of BP in patients with stroke, who may be considered nondippers.
Ambulatory blood pressure monitoring (ABPM) devices are increasingly used in the assessment of hypertension, but their value in patients after a stroke has not been studied systematically. This is an important issue because hypertension is a major cause of stroke, and many stroke patients demonstrate relatively high blood pressure (BP) levels when measured manually on admission to the hospital.1 2 The accurate measurement of BP after an acute stroke is important because antihypertensive therapy may be considered in some cases. However, BP may be falsely elevated or depressed immediately after a stroke, depending on level of consciousness, severity of neurological deficit, mobility, and physical activity. Conventional clinical or ward recordings may therefore be unreliable or misleading, leading to inappropriate prescribing or withholding of antihypertensive therapy. To overcome this problem, ABPM has been proposed as a method of obtaining an accurate clinical assessment. The large number of readings obtained by these devices reduces variability and abolishes the observer bias encountered with casually recorded BPs. An additional consideration is that fewer subjects are required for clinical research studies with ABPM.3 4
We therefore investigated the clinical use of the Spacelabs 90207 oscillometric ABPM device in patients admitted to the hospital with acute stroke to examine BP changes in three categories of stroke (cerebral infarct, hemorrhage, and transient ischemic attack), as well as differences between men and women and differences between ethnic groups. In addition, we examined 24-hour diurnal BP variations.
Subjects and Methods
All consecutive patients aged 75 years or younger who were admitted to the hospital after an acute stroke (ictus onset within 12 hours) were included in the study. Patients were diagnosed as having a stroke based on World Health Organization criteria: rapidly developing clinical signs of focal and at times global loss of cerebral function with symptoms lasting more than 24 hours or leading to death, with no apparent cause other than that of vascular origin. Patients with transient ischemic attacks were also included if objective evidence of motor or sensory loss, which recovered within 24 hours, was demonstrable on admission. All patients underwent CT scanning within 72 hours of the ictus to define the type of stroke.
Manual BPs were measured with a standard manual sphygmomanometer at the time that 24-hour ABPM was performed with a Spacelabs 90207 oscillometric device applied to the nondominant arm.5 The ABPM device was installed within 24 hours of admission as soon as informed consent had been obtained. Patients unable to give consent because of their neurological deficit, those with atrial fibrillation, and those already receiving antihypertensive drugs were excluded. The first reading obtained with the ABPM device was compared with manual BP measurements by the sequential same-arm technique. Mean daytime and nighttime ABPM readings were also analyzed to determine diurnal variation and any evidence of nocturnal BP falls (“dipping”).
Manual BPs were recorded by an observer who had been trained with the British Hypertension Society BP measurement video recording.6 The standard mercury sphygmomanometer was used to measure auscultatory manual BPs, with a standard cuff deflation rate of 2 to 3 mm Hg/s.6 7 8 The first appearance and final disappearance of Korotkoff sounds were recorded to determine systolic and diastolic BPs, as recommended by the British Hypertension Society7 and the Joint National Committee on Detection, Evaluation, and Treatment of High Blood Pressure.8 At the start and conclusion of the study, all the ABPM devices were calibrated against a standard mercury sphygmomanometer, and no calibration errors were found. Either a “large adult” (15×32 cm) or “normal adult” (12×23 cm) cuff size was used, appropriate to the subject's mid-upper-arm circumference, and the standard cuff deflation rate of the ABPM device (8 mm Hg/s) was used. ABPM readings were obtained each 30 minutes throughout the 24-hour monitoring period, which allowed approximately 48 readings throughout the period studied, and for each subject the data were edited and reduced to 24 consecutive 1-hour averages. Mean daytime ABPM readings were calculated as the mean of BPs from 7 am to 11 pm, and mean nighttime ABPM readings were calculated as the mean of BPs from 11 pm to 7 am. No other clinical procedures were performed during the day of monitoring.
The first reading obtained with the ABPM device was compared with the mean of the immediately preceding and following manual BP measurements.9 The manual heart rate measurement was taken for 60 seconds. Mean manual BPs were also compared with the mean 24-hour BPs measured with the ABPM device. Mean daytime and nighttime ABPM readings were also determined and analyzed to determine the presence of a nocturnal BP fall (dipping). Corresponding heart rates were recorded. The mean percentage of successful readings was calculated, as determined by each ABPM device and summarized on the standard computer analysis printout of the 24-hour recordings; the proportion of patients who achieved more than 70% of successful readings was also determined.
Statistical comparisons between BP and heart rate measurements were made with a paired t test, and the mean difference between the measurements compared was determined. Analysis of the type of stroke, influence of sex and ethnic origin, and clinical outcome was performed with one-way ANOVA, and their combined interaction was analyzed with ANOVA for unbalanced data (general linear model). Agreement for manual BP and ABPM measurements was expressed as mean±SD difference; a one-sample t test was performed to determine whether the mean difference was significantly different from zero. A value of P<.05 was considered statistically significant. All statistical calculations were performed on a microcomputer with a commercially available statistical package (Minitab v8, Minitab Inc).
We studied 86 patients (58 men; mean±SD age, 64.2±9.2 years) admitted with acute stroke (ictus within 12 hours) to a district general hospital; 12 patients had complete neurological recovery within 24 hours and were classified as having had transient ischemic attacks. Sixty-five patients were white, 8 were black/Afro-Caribbean, and 13 were Indo-Asian. Of the patients studied, 31 (36.0%) had a previous history of hypertension and 19 (22.1%) had a history of ischemic heart disease (defined as previous myocardial infarction by standard criteria, revascularization procedures [coronary artery bypass surgery or angioplasty], or a clear history of angina). CT scanning demonstrated an ischemic or thrombotic stroke in 59 patients, intracerebral hemorrhage in 15, and no gross abnormality in 12 patients.
BPs measured manually and with Spacelabs 09207 ABPM devices were studied. ABPM data were only obtained in 73 patients; 7 patients died before the ABPM device was applied or before a complete 24-hour recording was obtained, 3 patients were immediately transferred to a neurosurgical unit for further management before the ABPM was applied, and in 3 patients the ABPM did not make any recordings because of technical errors.
ABPM and Manual BP Measurements
The median percentage of successful BP readings by ABPM was 92% (interquartile range, 72% to 98%). The proportion of stroke patients in whom more than 70% of BP readings were successful was 77%. Compared with the mean of the two manual BP readings and manual heart rate measurement, the first ABPM systolic or diastolic BP or heart rate measurements were not significantly different (paired t test; all P=NS). The mean differences between manual and systolic and diastolic ABPM measurements were not significantly different from zero, suggesting satisfactory agreement between the measurements (systolic BP: mean±SD difference, 2.8±28.8 mm Hg; one-sample t test; T=0.80, P=.43; diastolic BP: mean±SD difference, 1.8±20.0 mm Hg, T=−0.73, P=.47).
In contrast, there was a statistically significant difference between the mean manual systolic and diastolic BPs and the corresponding average 24-hour ABPM readings; for systolic BPs, the mean difference between mean manual and 24-hour ABPM values was 12.9 mm Hg (paired t test; T=3.8, P=.0003), and for diastolic BPs, the mean difference was 4.4 mm Hg (T=2.2, P=.035). The mean manual systolic BP was higher than the mean systolic ABPM daytime (mean difference, 11.0 mm Hg; T=3.2, P=.0019) and nighttime (mean difference, 13.6 mm Hg; T=4.0, P=.0002) readings. The mean manual diastolic BP was also higher than the mean ABPM nighttime diastolic reading (mean difference, 5.8 mm Hg; T=2.74, P=.0078) but not significantly different from the mean ABPM daytime diastolic reading (mean difference, 2.9 mm Hg; T=1.4, P=.15) (Table 1⇓).
Day and Night BPs by ABPM
There was also no difference between mean day and night systolic BPs (mean difference, 1.9 mm Hg; P=.08), although mean daytime diastolic BP was higher than mean nighttime diastolic BP (mean difference, 2.4 mm Hg; P=.01). Although the mean day-night difference for DBP was statistically significant, the mean day-night differences in systolic BP and diastolic BP were less than 10%.
Effects of Ethnic Origin and Type of Stroke
There were no statistically significant differences in manual BPs by ethnic origin or type of stroke (one-way ANOVA; P=NS) (Table 2⇓). Black patients had higher mean ABPM readings for 24-hour systolic BPs (P=.002), daytime systolic BP (P=.001), and nighttime systolic BP (P=.09).
Patients with intracerebral hemorrhage showed higher BPs than the other groups, with the highest daytime systolic BP (P=.025) and diastolic BP (P=.005) by ABPM. In addition. they demonstrated a tendency for “reverse dipping” with higher pressures at night, a trend that was statistically significant for systolic BP (Table 2⇑). Patients with cerebral infarcts or transient ischemic attacks showed a reduced nocturnal BP fall with loss of dipping but no reverse diurnal rhythm. There was no significant difference in ABPM measurements between men and women and between those who were alive or dead after 6 months of follow-up (unpaired t test; all P=NS).
With the use of ANOVA for unbalanced data (general linear model), analysis of the combined influences of sex, ethnic origin, type of stroke, and clinical outcome and their combined interactions on BP levels as determined by ABPM was performed. This showed that ethnic origin was significantly associated with higher mean 24-hour systolic BP (F=4.0, P=.024) and daytime systolic BP (F=4.6, P=.013); type of stroke was significantly associated with higher mean nighttime systolic BP (F=4.1, P=.022).
ABPM devices are increasingly used in the management of hypertensive patients, and reference values for normal ranges are accumulating.10 11 Almost all studies have shown that readings determined by ABPM correlate more closely than clinic pressure with the extent of target organ damage.12 13 14 15 The increasing demand for 24-hour ABPM has resulted in publication of a protocol for the evaluation of measuring devices by the British Hypertension Society9 and the Association for the Advancement of Medical Instrumentation (AAMI).16 The British Hypertension Society protocol has been used to evaluate the Spacelabs 90207 device, which achieved a B grading for both systolic and diastolic BPs and also satisfied the AAMI criteria.5 ABPM is also useful because it avoids the “white coat” and placebo effects in hypertension studies.3 4 17
This clinical study demonstrates that in patients with acute stroke, oscillometric BP measurement provides readings that are comparable to those obtained with a stethoscope and a mercury manometer, suggesting that ABPM can be used to assess BPs in patients with acute stroke. There was a high percentage of successful recordings (approximately 90%) in our patients, comparable to clinical studies in stable outpatients. In addition, there was also no significant difference in the mean manual systolic BP when the standard mercury sphygmomanometer was used compared with simultaneous ABPM measurement. Because the first manual readings are typical of the “one-off” casual BP recordings in the office or outpatient situation, their similarity to the first ABPM readings suggests that the two measurements may be comparable. This is particularly important because casual BP measurements after acute stroke have shown high levels after admission, which have tended to settle spontaneously within the first week of admission1 2 18
The mechanisms that account for the initial poststroke BP elevation are unclear, but psychosocial stress from the acute hospital admission, central mechanisms,19 20 catecholamine and cortisol release,21 22 and a white coat effect23 have been postulated. A recent study by Harper et al18 suggests that use of 24-hour ABPM measurements may abolish the white coat effect, although the latter was unlikely to be the sole factor in poststroke hypertension. In the present study, mean manually determined BPs on first admission were higher than mean 24-hour ABPM measurements and mean daytime or nighttime readings. This is despite the similarity between mean manual recordings and the first ABPM readings, suggesting a “high” initial reading, which spontaneously settles over a 24-hour period after application of the ABPM. This is consistent with previous observations by Harper et al18 that suggest a significant white coat effect, so that reliance on casual manually determined BPs may overestimate BPs in stroke patients, leading to inappropriate antihypertensive therapy.
The mean day-night difference in both systolic and diastolic BPs was much less than the 10% day-night difference normally seen in dippers. Patients with acute stroke can therefore be classified as “nondippers.” The phenomenon of nondipping has been seen in medical conditions as diverse as malignant hypertension, renal disease, Cushing's disease, pheochromocytoma, polycystic kidneys, stroke, diabetes with autonomic neuropathy, heart failure, and preeclamptic toxemia.12 24 The exact significance of dippers and nondippers is unclear. The absence of a nocturnal fall in BP may simply be due to the patient not sleeping or being disturbed by the inflation of the ABPM cuff.24 This would be a limitation intrinsic to all studies with ABPM devices. However, it is also argued that the absence of the normal dipping results in a higher 24-hour BP “load” and may have prognostic implications. In particular, hypertensive patients who are nondippers may have more target-organ damage than those with the normal diurnal variation in BP.25 The finding of nondipping in patients with stroke requires further study and may be related to the known diurnal variation in onset of acute stroke and heart attacks.26
In the present study we also demonstrated higher manual BP and ABPM readings among patients with intracerebral hemorrhage. This is consistent with observations from previous studies in which casual BP measurements were determined manually.1 27 These patients tended to have the highest daytime BPs by ABPM and a trend toward higher nocturnal BPs. This phenomenon of reverse dipping has not been reported in other patient groups and may perhaps reflect the high BPs and lack of diurnal variation in patients with intracerebral hemorrhage. Further studies of ABPM in larger numbers of such patients are needed to explore this hypothesis. We have also shown that black patients had higher mean ABPM readings for 24-hour systolic BP, daytime systolic BP, and nighttime systolic BP. This is in keeping with clinical observations of higher hypertension-related morbidity and mortality, such as stroke, in black patients than in whites and Asians.28 However, we did not find any significant difference in ABPM readings between men and women and between those who were alive or dead after 6 months of follow-up.
We suggest that the clinical use of an oscillometric device such as the Spacelabs ABPM device may possibly minimize errors from casual BP measurement, may minimize the white coat effect, and may reflect the usual BP in patients with acute stroke. This hypothesis can only be fully tested by a comparison of concurrent ABPM with simultaneous intra-arterial BP monitoring. However, intra-arterial BP monitoring is not practical for everyday clinical practice because it is highly invasive. Invasive monitoring may also increase the stress response with resultant excess catecholamine release, which may further increase the heart rate and result in recordings significantly different from usual BPs in these patients. In contrast, ABPM application should not confer any further discomfort than an ordinary BP measurement cuff.
This study is limited by our use of only one 24-hour period for ABPM measurement; our objective was to investigate ABPM in acute stroke, and we only studied patients who presented with ictus onset within 12 hours. However, ABPM readings were obtained each 30 minutes throughout the monitoring period, which allowed many BP readings throughout the period studied. The present study of BP changes (dippers and nondippers) in the acute stage of stroke may also not necessarily be the same as changes in BP in the same patients at 6 or even 12 months after stroke. Nevertheless, the main concerns in management usually occur when the patient is acutely admitted and involve whether ABPM devices can be used to assess BPs in acute stroke patients in this period (which was the primary aim of our study). An additional long-term study is therefore needed to ascertain whether this information can be translated to management decisions for poststroke hypertension. Finally, we did not compare ABPM with the usual ward nursing shift–measured BPs because BP measurements for individual patients would have been measured by a great many nurses of different seniority and training during shifts; hence, such comparisons would have been meaningless in view of the wide variation between observers. In our study, comparisons between mercury sphygmomanometer and ABPM were undertaken by a single highly trained observer.
In conclusion, oscillometric ABPM devices can be used in clinical practice to assess BPs in patients with acute stroke and may minimize errors in overestimation or underestimation of blood pressures. This study also demonstrates higher systolic BPs as recorded by ABPM (but not manually) in black patients with acute stroke and in patients with intracerebral hemorrhage. ABPM recordings may be useful in the assessment of BPs in patients with stroke. These patients demonstrate a loss of diurnal BP rhythm and may be considered nondippers, with some evidence of reverse dipping in those with intracerebral hemorrhage.
Dr Lip is a recipient of the Edith Walsh and Ivy Powell research awards for cardiovascular disease from the British Medical Association. We thank Michèle Beevers, SRN, and Dr David Churchill for helpful comments during the preparation of this manuscript and Dr Paul Davies for statistical support.
- Received July 9, 1996.
- Revision received August 26, 1996.
- Accepted August 26, 1996.
- Copyright © 1997 by American Heart Association
Britton M, Carlsson A, de Faire U. Blood pressure course in patients with acute stroke and matched controls. Stroke.. 1986;17:861-864.
Coats AJS, Conway J, Sleight P. Ambulatory monitoring: clinical trial size and precision. J Hypertens. 1989;7(suppl 6):s357. Abstract.
Coats AJS. Reproducibility or variability of casual and ambulatory blood pressure data: implications for clinical trials. J Hypertens. 1990;6(suppl 6):s17-s20.
O'Brien E, Mee F, Atkins N, O'Malley K. Accuracy of the Spacelabs 90207 determined by the British Hypertension Society protocol. J Hypertens. 1991;9(suppl 5):s25-s31.
Petrie J, Jamieson M, O'Brien E, Littler W, Padfield P, de Swiet M, for the Working Party on Blood Pressure Measurement. Blood Pressure Measurement [videotape]. London, UK: British Medical Journal Publications; 1990.
Petrie JC, O'Brien ET, Littler WA, de Swiet M. British Hypertension Society recommendations on blood pressure measurement. Br Med J.. 1986;293:611-615.
O'Brien E, Petrie J, Littler W, de Swiet M, Padfield PL, O'Malley K, Jamieson M, Altman D, Bland M, Atkins N. The British Hypertension Society Protocol for the evaluation of automated and semi-automated blood pressure measuring devices special references to ambulatory systems. J Hypertens.. 1990;8:607-619.
Cox J, O'Malley K, Atkins N, O'Brien E. A comparison of the twenty-four hour blood pressure profile in normotensive and hypertensive subjects. J Hypertens. 1991;9(suppl 1):S3-S6.
Staessen J, Celis H, De Cort P, Thijs L, Amery A. An inventory of studies on ambulatory blood pressure in large groups of subjects. J Hypertens. 1991;9(suppl 8):S48-S50.
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 recurring stress. Circulation.. 1983;68:470-476.
Association for the Advancement of Medical Instrumentation. American National Standard for Electronic or Automated Sphygmomanometers. Arlington, Va: Washington, DC: AAMI, USA; 1987.
Olsson T, Marklund N, Gustafson Y, Nasman B. Abnormalities at different levels of the hypothalamic-pituitary-adrenocortical axis early after stroke. Stroke.. 1992;23:1573-1576.
Meyer JS, Shimazu K, Fukuuchi Y, Ohuchi T, Okamoto S, Koto A. Impaired neurogenic cerebrovascular control and dysautoregulation after stroke. Stroke.. 1973;4:169-186.
Carlberg B, Asplund K, Hagg E. High blood pressure in acute stroke: is it ‘white coat’ hypertension? J Intern Med.. 1990;228:291-292.
Pickering TG. Ambulatory Blood Pressure Monitoring and Blood Pressure Variability. London, England: Science Press Ltd; 1990.
Verdecchia P, Schillaci G, Guerrieri M, Gatteschi C, Benemio G, Boldrini F, Porcellati C. Circadian blood pressure changes and left ventricular hypertrophy in essential hypertension. Circulation.. 1990;81:528-536.
Muller JE, Tofler GH, Tone PH. Circadian variation and triggers of onset of cardiovascular disease. Circulation.. 1989;79:733-743.
Carlberg B, Asplund K, Hagg E. Factors influencing admission blood pressure levels in patients with acute stroke. Stroke.. 1991;22:527-530.
Cruickshank JK, Beevers DG, Osbourne VL, Haynes RA, Corlett JCR, Selby S. Heart attack, stroke, diabetes and hypertension in West Indians, Asians and whites in Birmingham, England. Br Med J.. 1980;281:1108.