In his article on blood pressure measurement by

Transcription

1 510 Clinical Conference Ambulatory Blood Pressure Monitoring Use in Hypertension Research and Clinical Practice Principal Discussants Giuseppe Mancia, Marco Di Rienzo, and Gianfranco Parati Cattedra di Medicina Interna, University di Milano and Ospedale S. Gerardo di Monza; Centro Auxologico Italiano, Milano; and LaRC, Centro di Bioingegneria (M.Di R.), Fondazione Pro-Juventute Don Gnocchi, Milano, Italy In his article on blood pressure measurement by sphygmomanometry published in 1897 in the Gazzetta Medica di Torino,' Riva Rocci wrote the following: "Blood pressure is acted upon, in a temporary but pronounced fashion, by the state of 'psychic' excitation of the patient. It is enough that the patient is spoken to, that he is invited to read, that he is even looked at suddenly, or that a sudden and even far noise strikes him (e.g., a carriage passing by in the outside street) that his blood pressure raises, and not at all to the same extent in all patients. This emotional reaction may be useful in psychiatry, but in other disciplines these blood pressure rises represent an inconvenience, and it is therefore necessary for the patient to be put in an environment as quiet as possible.... Furthermore, because even the application of the instrument causes a temporary blood pressure rise, it is necessary to take not only one but several consecutive blood pressures spaced by 3- or 5-minute intervals until a constant value is measured. This value, however, is not always the minimal value... It should not be necessary to add that it will not be proper to compare data unless obtained in identical conditions or environment, position, time of day, distance from meals, wakefulness, etc. This remark is certainly superfluous for the investigators, but it may be useful to practitioners whose hurry may in this case result in a waste rather than a gain of time." Thus, knowledge that emotional behaviors can increase blood pressure and that this affects the values taken by the physician goes back to almost the time blood pressure was first measured in an indirect, simple fashion. This is also the case for the current concepts that blood pressure is subjected to pronounced variations, that these variations are different in different subjects, and that for these reasons the prevailing blood pressure of a given individual is not easily reflected by clinical blood pressure values. However, the increasing interest in blood pressure variability that has occurred in the past 20 years depends on technical progress made in the 1960s, i.e., on Dr. Mancia was an invited speaker at the Cardiovascular Center, College of Medicine, University of Iowa, Iowa City, March 30, Address for correspondence: Prof. Giuseppe Mancia, Centro Fisiologia Qinica e Ipertensione, Via F. Sforza 35, Milano, Italy. Fax: the description of methods that allow ambulatory blood pressure to be measured noninvasivery or invasively for 24 hours and thus permit day and night blood pressure variations to be observed and quantified. 2-4 This article will review the information that has been derived from this progress using data from our group in Milan and will focus on four issues: 1) the effects of emotional behavior on blood pressure and their implications for the diagnosis of hypertension, 2) the mechanisms responsible for the magnitude of 24-hour blood pressure variability, 3) the alterations in blood pressure variability induced by hypertension and aging, and 4) the relevance of this phenomenon for the prognosis of the hypertensive patient and the evaluation of antihypertensive treatment. Discussion of these issues will be followed by a few considerations on some ongoing studies on ambulatory blood pressure monitoring and by an indication of which of the important issues raised by the increasing knowledge in blood pressure variability may receive an answer in the near future. Behavior and Blood Pressure Twenty-four-hour intra-arterial and noninvasive blood pressure monitoring in ambulatory conditions has offered a thorough description of the effects of behavior on blood pressure. 5 The major contribution has been made with intra-arterial monitoring, because these effects usually have such a dynamic profile as to be only approximately reflected by the limited number of blood pressure values (e.g., one every 15 minutes) obtained by the noninvasive approach. It is thus known that some behaviors (such as eating, drinking, or mental work when performed in the absence of stress) do not affect blood pressure substantially; that nighttime sleep, daytime sleep, and postprandial digestion cause hypotension; and that a variety of other daytime behaviors may raise blood pressure. 6 An example is emotion, which can cause a slight blood pressure rise when mild and a pronounced and prolonged pressor response when more marked and long-lasting. This has clinical implications, because sphygmomanometric blood pressure measurements by a physician are emotional for the patient, leading to a blood pressure elevation. 7-8 This was shown in 88 hospitalized patients whose blood pressure was checked with sphygmomanometry by an unfamiliar physician while 24-hour or 48-hour ambulatory blood pressure was measured intra-arterialry. 5-9 In

2 Mancia et al Ambulatory Blood Pressure 511 mmhg +60 -i Maximal change in MAP and HR with doctor's vitit (n=88) t X Y? bpm A 1 A 2. FIGURE 1. Plots show peak increases in mean arterial pressure (MAP) and heart rate (HR) in 88 subjects during a visit by a physician in charge of assessing blood pressure with a sphygmomanometer. In each subject, blood pressure was measured intra-arterialfy for 24 or 48 hours by the Oxford method. 3 Intra-arterial blood pressure and heart rate values observed during the physician's visit were compared with those observed 4 minutes before the visit. Data are shown for individual subjects (points) and as average changes for the group as a whole (horizontal bars) (redrawn from Mancia and Parati 10 with permission). 86 of the 88 patients, the appearance of the physician caused an increase in the patient's intra-arterial blood pressure and heart rate, which peaked within 4 minutes. The magnitude of the pressor response was large in some patients and small in others, with a striking between-subject difference (Figure 1). No relation was found between the response and the patient's age (16-67 years), baseline mean blood pressure ( mm Hg), sex, sex difference from the physician, and pressor or tachycardic responses to laboratory stressors. 5 ' 9-11 Thus, behavioral dependence of blood pressure is responsible for its overestimation by the traditional blood pressure-assessing procedure. Although reaching hypertensive values in only a subset of individuals (the "white coat" hypertensive patients, see Reference 12), the overestimation is a common phenomenon. It is also a phenomenon that cannot be corrected because of its wide between-subject range and the fact that it cannot be predicted by clinical information and laboratory tests. The clinical relevance of the white coat phenomenon will be further discussed below. It should be mentioned here, however, that the emotional component of the sphygmomanometric blood pressure measurements can be minimized in various ways. In the patients mentioned above, the blood pressure increase observed at the 10th or 15th minute of the physician's visit was approximately 50% of the peak response. 9 In addition, when blood pressure was measured in some patients by a nurse, the peak response was less, and at the 10th minute, the blood pressure value was similar to the value before or after the nurse's visit (Figure 2). 13 Finally, it is known that the pressor response is usually less when blood pressure is self-measured at home 14 and absent when automatic or semiautomatic devices are used (Figure 3). 15 The pressor response can be attenuated even when blood pressure is measured by a physician, provided the subject is familiar with the procedure because of repeated visits. 16 -' 8 Familiarization may not occur in all subjects, however. 19 Furthermore, attenuation of the hemodynamicalry dependent pressor response may take weeks rather than days, because in 16 subjects under intra-arterial ambulatory blood pressure monitoring, four visits by the same physician over a short period of time (2 days) were associated with similar pressor and tachycardic responses. 13 Determinants of Blood Pressure Variability Over the 24 hours blood pressure can vary by more than 50 mm Hg. A considerable portion of this variability is due to behavioral influences, presumably through central modulation of autonomic drive to the heart and systemic blood vessels, as further demonstrated by the reduced blood pressure variability that has been shown in subjects with high spinal cord transection. 20 But nonbehavioral factors are involved as well, because blood pressure varies even when subjects lie immobile with no apparent changes in activity. By and large, 24-hour blood pressure variations depend largely on central cardiovascular control. This is suggested by data obtained in normotensive and untreated essential hypertensive subjects in whom 24-hour blood pressure was recorded intra-arterially and half-hour average values were calculated from beat-to-beat analysis of the blood pressure tracing. Half-hour average blood pressure values were positively correlated with half-hour average heart rate values (Figure 4). 21 This was not due to the fact that daily life changes in blood pressure are caused by changes in heart rate and cardiac output, because a marked reduction in heart rate variability by atropine did not affect blood pressure variability. 22 This probably indicates that factors modulating the heart and blood vessels in a consensual fashion, i.e., exciting or inhibiting both targets together, play a major role in the tendency of blood pressure and heart rate to vary between half hours, the central nervous system being their most likely site of origin. Other possible factors involved in the determination of blood pressure variability continue to be investigated. At present, the evidence indicates that the circadian blood pressure profile that can be seen in most individuals, i.e., the daytime increase and nighttime fall in blood pressure, is quickly reversed in shift workers at the time they start the night working schedule, excluding an important role of time per se in 24-hour blood pressure variability. 23 It also indicates that blood pressure variability depends to some degree on circulatory changes mechanically induced by ventilation, and there are speculations that humoral vasoactive substances, changes in vascular smooth muscle reactivity, and other as yet unknown influences are involved. Finally, the search for factors responsible for blood pressure variability has allowed the clarification of two additional phenomena. First, 24-hour blood pressure is characterized not only by marked and irregular variations but also by rhythmic blood pressure oscillations within the range of Hz (i.e., the so-called low-, mid-, and high-frequency osculations), 24 " 26 which can be identified

3 512 Hypertension Vol 21, No 4 April 1993 A DBP 10 /mlr i P«ak * * T k \ S A HR lomln FIGURE 2. Line graphs show rises in intra-arterial systolic blood pressure (SBP), diastolic blood pressure (DBP), mean arterial pressure (MAP), and heart rate (HR) during the visit of a physician and nurse in charge of measuring blood pressure with a sphygmomanometer. Data are shown as mean ±SEM changes from control value taken 4 minutes before the visit and refer to 30 subjects under 24-hour intra-arterial blood pressure monitoring The nurse's visit preceded the physician's visit in 15 subjects and followed it in the remaining 15 (from Mancia et al 13 with permission). 4- * II A~ Paak S 80 m cc r so x 40 AUTOMATIC CUFF INFLATION n = 15 I I I I I I I I I - H-4-H H-H-r-r-r-Ho 9-5 T "V 10m In SEMI-AUTOMATIC CUFF INFLATION n.20 H+H Control -1 Control 1 -t-imin and quantified by power spectral analysis, taking 256- beat consecutive segments of the ambulatory intraarterial blood pressure tracing. This suggests that blood pressure variability can be divided into an "irregular" component, originating from the cardiovascular response to environmental stimuli, and several rhythmic fluctuations that are intrinsic to the cardiovascular system. These rhythmic fluctuations are in an early investigational phase, but several interesting features have already emerged. For example, low-, mid-, and FIGURE 3. Plots show intra-arterial blood pressure and heart rate during the minute immediately preceding beginning of cuff inflation ( 1 to 0), the minute immediately after cuff inflation (0 to 1), and the 1-minute control period 5 minutes before cuff inflation. Values measured every 15 seconds are shown. Data refer to hypertensive subjects in whom intrabrachial or radial blood pressure monitoring was performed for several hours in combination with automatic (n=15) or semiautomatic (n=20) noninvasive blood pressure monitoring in the contralateral arm. Automatic and semiautomatic blood pressure readings were obtained every 10 minutes. Data were averaged in each subject and mean ±SEM calculated for the group as a whole (from Parati et al' 5 with permission). M, mean arterial pressure; S, systolic; D, diastolic.

4 Mancia et al mmhg Ambulatory Blood Pressure 513 Mean Arterial Pressure n= FIGURE 4. Plots show mean arterial pressure and heart rate in 89 ambulant subjects with untreated essential hypertension. Data are shown as mean±sem for each half hour. Blood pressure was measured intra-arterialfy by the Oxford method3 (from Mancia et al2' with permission). 100 b/min 90 Heart Rate n = TIME OF THE DAY (hours) which, the power of blood pressure fluctuations is inversely proportional to their frequency).28 Interestingly, different components of blood pressure variations may be differently influenced by classic blood pressure control mechanisms. This was shown in unrestrained animals subjected to intra-arterial blood pressure monitoring in which sinoaortic denervation increased overall blood pressure variability29-31 but had a complex effect on specific blood pressure spectral components. That is to say, sinoaortic denervation markedly increased the amplitude of the very lowfrequency blood pressure oscillations (Figure 6). The low-frequency oscillations were modestly increased by sinoaortic denervation, whereas the mid-frequency oscillations were reduced and the high-frequency oscillations were not homogeneously altered high-frequency blood pressure fluctuations tend to account for 20-25% of overall blood pressure variability as quantified by the variance (or standard deviation) of all 24-hour blood pressure values, thereby being smaller than the "irregular" variability component.6-25 Furthermore, the rhythmic blood pressure oscillations do not have the same amplitude over the 24 hours, and in the mid-frequency range they show a marked systematic reduction during the night, in line with the nighttime reduction of several cardiovascular parameters (Figure S).25 Finally, we have recently shown27 that over 24 hours blood pressure fluctuates also at frequencies markedly lower than Hz and that these "very low"-frequency components contribute to a substantial degree to overall blood pressure variability, as predicted by the 1/f model (according to AVERAGE USE) POWERS NORMALIZED OVER TOTAL VARIANCE FIGURE 5. Plots show low-frequency (LF), mid-frequency (MF), and high-frequency (HF) powers of systolic (SBP) and diastolic (DBP) blood pressures throughout 24 hours. Data refer to mean half-hour powers computed from consecutive segments of 256 beats. Mean±SEM from 10 patients with untreated mild essential hypertension is shown. Powers are represented as percent of total variance. Note that nighttime (b) midfrequency powers were significantly less than daytime (a) mid-frequency powers (from Parati et al25 with permission). HF DBP 24 8 hours

5 514 Hypertension Vol 21, No 4 April 1993 db i SBP oooo db 40 0-, DBP Hz FIGURE 6. Tracings show effects of sinoaortic denervation on spectral components of systolic (SBP) and diastolic (DBP) blood pressures having a frequency from to 0.1 Hz. Data are shown as ratio between spectral densities obtained in eight conscious cats after and before sinoaortic denervation. Thin lines refer to single cats, thick lines to average group values. Note that sinoaortic denervation increased the power of the very low-frequency oscillations (modified from Di Rienzo et al 31 with permission) I oooo Hz The second phenomenon concerns the arterial baroreceptor reflex. In 62 subjects under 24-hour ambulatory intra-arterial blood pressure monitoring, various indexes of blood pressure variability almost always showed an inverse relation with the baroreceptor reflex sensitivity estimated by the bradycardic response to baroreceptor stimulation via an intravenous bolus of phenylephrine or the tachycardic response to baroreceptor deactivation via an intravenous bolus of trinitroglycerin, both responses showing a direct relation with heart rate variability (Figure 7). 33 This was the case also when the baroreceptor reflex sensitivity was estimated by the blood pressure response to the alteration in carotid baroreceptor activity induced by a neck chamber. 33 Most important, the inverse relation between baroreceptor reflex sensitivity and blood pressure variability, a finding confirmed by other studies, 34 " 36 was also found within individual subjects in whom hourly blood pressure variability was plotted versus hourly baroreceptor reflex sensitivity estimated by computer analysis of the beat-to-beat relation of the systolic blood pressure and pulse interval values (see below) (Figure 8). It is thus clear that, in line with classic animal data, the arterial baroreceptor reflex limits the blood pressure variations occurring in humans partly via opposite changes in heart rate and cardiac output " 38 The magnitude of these variations in any given individual thus depends on the balance of oscillatory and antioscillatory influences. Changes in Blood Pressure Variability With Hypertension and Aging Several studies have reported that in secondary hypertension and in essential hypertension with severe end-organ complications the nocturnal blood pressure fall is reduced or absent. 39 " 42 However, in most patients with essential hypertension, the alternation of high daytime and low nighttime blood pressures is preserved. 21 This can be seen in Figure 9, which refers to age-matched normotensive, moderate, and more marked essential hypertensive subjects under 24-hour ambulatory intra-arterial blood pressure monitoring. The circadian blood pressure profile, as well as the circadian heart rate profile, was superimposable in the three groups. The circadian heart rate and blood pressure profiles were also only slightly flattened in old normotensive subjects («=59; age, 74.5±0.9 years; mean±sem) compared with young normotensive subjects (n=57; age, 38.2±1.4 years). 43 In the young group, average mean arterial pressure (noninvasive ambulatory monitoring) was 97.1 ±1.1 mm Hg during the daytime and 85.3±1.1 mm Hg during the nighttime; respective figures for the old group were 93.3 ±1.0 and mmhg, the night-related hypotension thus being only slightly reduced. This was also the case for the night-related bradycardia, which amounted to 9.3 and 6.2 beats per minute in the young and old group, respectively (Zito, Parati, Abate, and Mancia, unpublished observations). Preservation of the circadian blood pressure profile does not mean that hypertension and aging do not affect blood pressure variability In subjects in whom 24-hour blood pressure was monitored intraarterially and the blood pressure signal was analyzed beat-to-beat, the within-half-hour standard deviation of mean arterial pressure (i.e., the average of all half-hour standard deviations within the recording period) increased progressively from normotensive to borderline, moderate, and more severe essential hypertensive subjects (Figure 10) Furthermore, in old normotensive subjects (see above), the number of blood pressure values differing more than ±20 mm Hg

6 Mancia et al Ambulatory Blood Pressure _ 12 a \ r = 0.35 r=0.40 r ; 0.47 r=0.32 ; C 5 4 c r :0.45 r = 0.M r = 0.50 r = 0.44 \ V v * r = 0.35 r =0.41 r = 0.40 FIGURE 7. Plots show regression lines of mean arterial pressure (MAP) and heart rate (HR) standard deviation (SD) and baroreceptor reflex sensitivities measured by the slope of changes in RR interval induced by increasing or reducing systolic blood pressure (SBP) through intravenous boluses of phenylephrine and trinitroglycerin. Only regression lines achieving statistical significance (p<0.05) have been drawn (from Mancia et al 33 with permission). ^ 4! S MAP Standard Deviation (mmhg) HR Standard Davlatlon within half hour SD 2 hour day SD among half hour SD 2 hour night SD 12 (b/i lin) from the mean daytime or nighttime value was two or three times more frequent than in younger normotensive subjects S - 50 Thus, hypertension and aging are associated with an increased degree of shortlasting blood pressure variations. It may be speculated that this is accounted for by greater pressor responses to emotional and other behavioral stimuli due to an increased central emotional reactivity, a greater response of resistance vessels to vasoconstrictor stimuli, or both. It is also possible, however, that hypertensive and elderly subjects have greater rhythmic oscillations or that the antioscillatory influence of the arterial baroreceptor reflex is impaired. This last hypothesis has received experimental support after development of a technique that allows arterial baroreceptor reflex sensitivity to be assessed throughout the 24 hours, avoiding use of vasoactive drugs and other artificial methods to stimulate and unload the arterial baroreceptors. The technique consists of scanning the 24-hour intra-arterial blood pressure tracing every 3 msec to identify the large number of episodes characterized by a progressive increase or decrease in systolic blood pressure and a linearly related increase or reduction in pulse interval, respectively. The baroreceptor reflex nature of these episodes was demonstrated in conscious cats by their striking reduction after sinoaortic denervation (Figure II). 54 When used in humans, the following data were obtained 55 : First, in any given individual, the sensitivity of the baroreceptor reflex was strikingly different throughout the day and night (Figure 12), indicating that the performance of BAROREFLEX GAIN AND SBP VARIABILITY SO S8P rnmhfl r L hour F.M yi FIGURE 8. Line graph shows hourly standard deviation of systolic blood pressure (SD SBP) (filled circles) and baroreceptor reflex sensitivity (open circles) in one subject (F.M.). Twenty-four-hour blood pressure was measured intra-arteriatiy by the Oxford method 3 in ambulatory conditions. Baroreceptor reflex sensitivity was assessed by the slope of the linear relation between 1) spontaneously occurring progressive increases in SBP and pulse interval and 2) spontaneously occurring progressive reductions in SBP and pulse interval. The two types of sequences pooled amounted to approximately two thousand over the 24 hours. There was a tendency for SD and baroreceptor reflex sensitivity to change in opposite directions (\ 2 test). 10

7 516 Hypertension Vol 21, No 4 April 1993 MEAN VALUEStSE mmhg 160 -i MAP bpm 130 HR SO J r > r normot«nilv«i (n-22) * mild hrp*rt«nilv«t (n~26)» > mor«fitrt hyp»rumtvm (n«41) FIGURE 9. Plots show circadian mean arterial pressure (MAP) and heart rate (HR) profiles in normotensive subjects (n=22) and untreated mild (n=26) and more severe (n=41) essential hypertensive patients. Data are shown as average half-hour values. Twenty-four-hour monitoring was obtained intra-arterialfy by the Oxford method. 3 blood pressure-controlling mechanisms is characterized by marked 24-hour variations. Second, throughout the 24 hours, the baroreceptor reflex sensitivity of essential hypertensive subjects was systematically less than that of normotensive controls. And third, the difference between the two groups was more marked during the night than during the day, because, whereas in normotensive subjects the night was associated with an increased baroreceptor reflex sensitivity, no increase occurred in hypertensive individuals (Figure 13). Similar differences characterized young and elderly normotensive subjects c o 5 6 I o 4- (D TJ S 2 W 24h MAP: 81*9 103* mmHg Cuff MAP: ii2 mmhg Mean ag«: years *p<0.05 FIGURE 10. Bar graph shows blood pressure variability in normotensive subjects (n=10), borderline untreated hypertensive subjects (n=31), and mild (n=69) and more severe (n=66) untreated essential hypertensive subjects. Blood pressure variability was calculated by averaging the standard deviation of the 48 half hours included in a 24-hour intraarterial monitoring period. Twenty-four-hour mean arterial pressure (MAP) (±SEM) and mean cuff values (±SEM) are shown at bottom. Clinical Use Twenty-four-hour intra-arterial blood pressure monitoring can help identification of pheochromocytoma, particularly when the hypertensive crises are too short to allow diagnosis by traditional methods such as sphygmomanometric measurements during the crisis and crisis-related increases in plasma catecholamines. Twenty-four-hour intra-arterial blood pressure monitoring may also exclude a pheochromocytoma by demonstrating little or no increase in blood pressure in the presence of seemingly typical signs and symptoms of this condition The main clinical interest of this approach, however, consists in its potential for providing a more precise diagnosis of the blood pressure elevation occurring in a given patient and thus a sharper definition of his or her blood pressure-related risk. This is supported by several observations: First, clinic blood pressure correlates with the incidence of cardiovascular morbidity and roo- 400 Q Intact B SAD * p*0.0» + PI/+SBP -PI/-9BP FIGURE 11. Bar graph shows number of sequences characterized by spontaneously occurring progressive increases in systolic blood pressure (SBP) and pulse interval (PI) (+PI/+SBP) and spontaneously occurring progressive reductions in SBP and PI (-PI/ SBP) in conscious cats before and after sinoaortic denotation (SAD). Data are shown as mean ±SEM from 10 animals. Note that after SAD sequences were strikingly reduced (from Bertinieri et al 54 with permission).

8 Mancia et al Ambulatory Blood Pressure beat sequences F.R.46yre Regression coefficients (+ m sec / + mm Hg) (-msec/- mmhg) + 10" r = 0.09 p=ns r = 0.05 p=ns mmhg SBP FIGURE 12. Scatterplots show regression coefficients of the sequences of spontaneously occurring progressive increases (top panel) and decreases (bottom panel) in systolic blood pressure (SBP) and pulse interval identified in a single subject (F.R.) over a 24-hour intra-arterial monitoring period. The coefficient is plotted vs. SBP measured at the beginning of the corresponding sequence. 60 SO h mcanttse + PI/+8BP SEQUENCES -*, SO X E 8 i c 2 0 left ventricular systolic and diastolic function, 63 ' 64 albuminuria, 65 and cerebral ischemia. 66 Overall or individual organ damage also bears a direct relation with blood pressures occurring at work, 62 with the number of daytime blood pressure peaks, 63 with nighttime blood pressure values, 4767 and with 24-hour blood pressure variability. 59 Thus, blood pressure averages and excursions in daily life appear to reflect the consequences of a high blood pressure condition on the cardiovascular system more satisfactorily than clinic blood pressure. Clinical use of ambulatory blood pressure monitoring is still unsettled, however, and two important caveats should be emphasized. First, with the exception of an uncontrolled follow-up study relating semiautomatic daytime blood pressure with subsequent cardiovascular morbidity and mortality, 68 the above-mentioned data on ambulatory blood pressure and end-organ damage refer to cross-sectional observations. Thus, no prospective controlled trial has ever documented that 24-hour average blood pressure, daytime and nighttime blood pressure, daytime blood pressure peaks, or blood pressure variability are prognostically useful or that ambulatory and clinic blood pressure together predict more accurately the chance of an individual developing cardiovascular complications. Second, because of limited sample size, inappropriate selection criteria, and other shortcomings, studies aimed at defining ambulatory blood pressure normalcy have not yet provided conclusive results. 69 Thus, lack of prognostic validation and reference values discourages routine diagnosis of hypertension by ambulatory blood pressure monitoring. In the authors' opinion, this diagnosis still should be based mortality, but, although significant, the correlation coefficient has a relatively low value. 57 Second, clinic blood pressure reflects to only a limited degree 24-hour average blood pressure, in both untreated and treated hypertension. 6 And third, compared with clinic blood pressure, 24-hour or daytime mean blood pressure correlates more closely with the overall end-organ damage of hypertension 59 ' 60 and with markers of individual organ damage such as left ventricular hypertrophy, 6162 FIGURE 13. Plots show sequences of spontaneously occurring progressive increases (+PI/+SBP) and decreases (-PI/ SBP) in pulse interval and systolic blood pressure during 2 hours of the morning, 2 hours of the evening, and 2 hours of nighttime sleep. Number of sequences is shown in top panels, slope of sequences in bottom panels. Mean±SEM from normotensive and untreated essential hypertensive subjects is shown (from Parati et al 55 with permission). m o»0 20 morning vtning o o Mormot*nil>*«night morning»h rp«rt«n»l»»«nlng night n = 10) *p«0.05

9 518 Hypertension Vol 21, No 4 April J SBP I * T I Control HR 1 T 1 Placebo mmhg 120-, b/m)n BO WO- 9O- 80- DBP I * - I Control o-o Office 24 h ABPM $ p<0.05 n =27 Placebo FIGURE 14. Plots show systolic blood pressure (SBP), diastolic blood pressure (DBP), and heart rate (HR) before and after 4 weeks ofplacebo treatment in 27 essential hypertensive patients. Dashed lines refer to clinic blood pressure, continuous lines to 24-hour ambulatory blood pressure (ABPM) means. Data are shown as mean±sem (from Mutti et al' 6 with permission). mainly on clinic blood pressure, whose prognostic importance has been documented by large-scale epidemiological studies. 70^74 Ambulatory blood pressure might help to avoid treatment in patients with a high clinic but a clearly low 24-hour or daytime blood pressure mean. However, caution is needed because there is evidence that subjects with white coat hypertension, i.e., with high clinic but low home blood pressure, have metabolic abnormalities and cardiovascular alterations similar to subjects with sustained hypertension, i.e., with high clinic and home blood pressures. 75 Thus, it is still debated whether a temporary blood pressure elevation in the clinic environment is devoid of an increased risk. 76 Similar caveats apply to the use of ambulatory blood pressure monitoring in the routine assessment of the efficacy of antihypertensive treatment. Ambulatory blood pressure monitoring, however, has become a widely used method for assessing new antihypertensive agents. This is justified by the ability of this approach to precisely determine whether the drug or drug regimen under study provides therapeutic coverage over 24 hours. Furthermore, at variance from clinic blood pressure, intra-arterial and noninvasive ambulatory blood pressures are largely devoid of a placebo effect (Figure 14) 15,77 Yjjjs means that 1) the actual blood pressure effect of the active drug is unmasked, increasing the study statistical power, and 2) a placebo group can be avoided, thereby avoiding the related ethical problems. Finally, the 24-hour ambulatory blood pressure mean is more reproducible than clinic blood pressure, 78 ' 79 the difference reaching two or three times when the mean is computed on 30 or more values (Figure 15). Given the dependence of the number of subjects to recruit on the square of the random blood pressure variations, this implies a striking reduction of the study size. 78 In this context, however, it should be emphasized that the mmhg SBP n = 14 DBP All All Values from 24 h Intra- Arterial ABPM Values from 24 h Intra-Arterial ABPM FIGURE 15. Bar graphs show mean difference and plots show standard deviation of mean difference between 24-hour systolic (SBP) and diastolic (DBP) blood pressure means obtained 4 weeks apart. Mean was obtained by including a progressively larger number of values. Twenty-four-hour ambulatory blood pressure (ABPM) was obtained intra-arteriauy in 14 subjects. The inverse of the standard deviation of the mean difference represents blood pressure reproducibility (from Trazd et al 79 with permission).

10 Mancia et al Ambulatory Blood Pressure mmhg 24 -i O SBP (n-15) power for the analysis of differences in the 24-hour blood pressure profiles. DBP (n-15) 0 CD CD CO SB eft CD O A Q CO o 6 - eft 0 J FIGURE 16. Plots show standard deviation (SD) of the mean difference between 24-hour (filled circles) and hourly (open circles) mean systolic (SBP) and diastolic (DBP) blood pressures. Data are from 15 subjects in whom 24-hour blood pressure was monitored twice 4 weeks apart noninvasively (From Mancia et al80 with permission). increased reproducibility of ambulatory blood pressure data is less evident for hourly blood pressure means, regardless of their daytime or nighttime location. In a recent study,80 our group found that the reproducibility of these means was consistently less than that of the 24-hour mean, the difference often being marked (Figure 16). Thus, when the efficacy of an antihypertensive drug is to be assessed throughout the 24 hours, a too-drastic reduction in the number of patients to be recruited should be avoided, lest there be inadequate D.G. ff 40 r««fs INTRA-ARTERIAL DBP Current Studies Several basic and clinical problems outlined in the previous sections are currently being investigated. Our group and the Department of Bioengineering at the University of Amsterdam are testing a device that allows finger blood pressure81-83 to be monitored beatto-beat in ambulatory conditions, the monitoring period lasting 24 hours.84 Comparison with 24-hour intraarterial blood pressure monitoring has shown that in most instances the device is accurate. It has also shown that, although there is some distortion (e.g., amplification of low-frequency components), the finger pressure signal is suited for power spectrum analysis and arterial baroreceptor reflex testing (Figure 17),85 indicating that investigation of mechanisms involved in dairy cardiovascular regulation may be pursued without resorting to the limiting invasive approach. We are also investigating the mechanisms responsible for the rhythmic blood pressure oscillations, and, in particular, the possibility that their so-called low-frequency ( Hz) and mid-frequency ( Hz) components reflect sympathetic vascular modulation In conscious normotensive rats subjected to prolonged intra-arterial blood pressure monitoring, chemosympathectomy by 6-hydroxydopamine increased systolic blood pressure variance by 54.2% compared with intact rats. The amplitude of the low-frequency blood pressure oscillations was increased by sympathectomy, and that of the mid-frequency blood pressure oscillations was reduced by approximately 60% (Figure 18).89 Additional a-blockade did not modify the changes observed with chemosympathectomy alone, leading to the conclusion that the low-frequency blood pressure RECORDING MAP PI FINGER BLOOD PRESSURE RECORDING SBP DBP 519 MAP Pi FIGURE 17. Plots show power spectral density over a half-hour period obtained by analysis of intra-arterial signal (radial artery) and noninvasive finger pressure device (Finapres) in one subject (D.G.) (from Omboni et al85 with permission). SBP, systolic blood pressure; DBP, diastolic blood pressure; MAP, mean arterial pressure; PI, pulse interval.

11 520 Hypertension Vol 21, No 4 April 1993 I MF * LF JL FIGURE 18. Bar graph shows effects of sympathectomy on systolic blood pressure spectral components in 10 vehicletreated rats (hatched bars) and 10 sympathectomized rats (open bars) (from Daffonchio et al 89 with permission). HF, high-frequency band ( Hz); MF, mid-frequency band ( Hz); LF, low-frequency band ( Hz). p<0.05. oscillations are not generated primarily by oscillations in sympathetic firing. On the other hand, the mid-frequency oscillations appear to depend on sympathetic firing. They may also depend to a substantial degree on nonsympathetic mechanisms, their specificity as an index of sympathetic drive therefore being limited. Finally, we are currently performing 24-hour ambulatory blood pressure monitoring in more than 2,000 individuals of different ages randomly taken from the population of Monza, simultaneously measuring clinic blood pressure, random-zero blood pressure, home blood pressure, and left ventricular mass (echocardiography). 90 These data, and those of a less complex population study under way in Belgium, 91 should define the range of ambulatory blood pressure normalcy, its difference from traditional blood pressure, and its possible relation with the cardiac structure. Obtaining information on the ambulatory blood pressure relation with the prognosis of hypertension vis-a-vis clinic blood pressure will be more difficult. Clearly, a prospective study based on cardiovascular morbidity and mortality ought to include so many hypertensive subjects as to be difficult. Thus, use of surrogate end points is useful, and some studies are currently examining the ability of ambulatory versus clinic blood pressure to predict the development of left ventricular hypertrophy or the regression of hypertrophy as a result of antihypertensive treatment Left ventricular hypertrophy may be an appropriate end point because, although there is still no indication that its regression improves prognosis, its presence is related to increased cardiovascular and total death rates Questions and Answers Dr. Gerald F. DiBona (University of Iowa, Iowa City): What accounts for residual power of some 40% after sympathectomy? Dr. Mancia: I don't know, but among others, one possibility is rhythmic fluctuations of vascular tone directly generated by periodical smooth muscle contraction. In a more general context, I wish to emphasize that overall blood pressure variance is increased after sympathectomy (possibly because the metabolic vasodilatation accompanying movement cannot be compensated by vasoconstriction in inactive districts, tachycardia, and an increase in cardiac output), suggesting that in the rat the sympathetic nervous system has a prevailing antioscillatory role. Dr. Allyn L. Mark (University of Iowa, Iowa City): You emphasized the variability in white coat hypertension between individuals. Does the magnitude of white coat hypertension relate to baroreceptor reflex sensitivity or the spontaneous rhythmic or nonrhythmic variability of blood pressure? Dr. Mancia: We have not studied whether the magnitude of the white coat response is related to baroreceptor reflex sensitivity and rhythmic variability of blood pressure. Concerning whether this response relates to overall blood pressure variability, the answer is no. We were not surprised, because blood pressure variability has a nonbehavioral component. Furthermore, pressor reactivity to an emotional stimulus does not reflect reactivity to emotional stimuli of a different nature. 11 Dr. Allyn L. Mark (University of Iowa, Iowa City): What is the prognostic significance of white coat hypertension? Dr. Mancia: The issue is not yet settled. On the one hand, epidemiological studies show that occasionally taken blood pressures, i.e., blood pressures inclusive of a white coat component, are predictive of future cardiovascular complications. 57 On the other hand, the Australian trial on treatment of mild hypertension (and other intervention studies) shows that the patients under placebo in whom the initially elevated blood pressure decreased to normal values after a few months (i.e., what we would call white coat hypertensive patients) were a low-risk group. 19 The issue recently has been taken up again by Julius and his associates, 75 who showed, as I briefly recalled above, that white coat hypertensive patients may share to some extent the metabolic and hemodynamic abnormalities of "true" hypertensive patients, rather than being similar to the normotensive subjects. To me, this suggests that it is not at all certain that the marked increase in blood pressure that occurs in a number of subjects during the doctor's visit is a totally innocent phenomenon. 76 Dr. Mark W. Chapleau (University of Iowa, Iowa City): You showed, as expected, that baroreceptor reflex sensitivity as determined by changes in heart rate is inversely correlated with blood pressure variability in young subjects. In old subjects, there was no correlation; despite low baroreceptor reflex sensitivity, they still demonstrate decreased pressure variability at night. Is it possible that the elderly rely more on baroreceptor reflex control of vascular resistance? If this is preserved, it may explain the decreased pressure variability at night. Do you have any data on the influence of age on baroreceptor reflex control of sympathetic nerve activity or vascular resistance? Dr. Mancia: Your question is well posed because we have indeed seen in rats that aging impairs baroreceptor control of heart rate to a much greater extent than baroreceptor control of blood pressure. 97 There is some evidence that this is the case also in humans. 98 We have to consider, however, that the marked reduction in blood pressure (and heart rate) variability induced by sleep is also a centrally originated phenomenon, its occurrence surviving sinoaortic denervation. 99 These central influences governing circulation during sleep may exist in young and old subjects.

12 Dr. John B. Stokes III (University of Iowa, Iowa City): You indicated that parallel increases and decreases in blood pressure and pulse rate reflect changes in sympathetic outflow. What is the meaning of changes in these parameters in the opposite direction? Dr. Mancia: I believe that parallel increases and de- CTeases in blood pressure and heart rate reflect central influences, increasing and reducing sympathetic outflow simultaneously to the heart and the peripheral circulation, although other factors capable of consensual cardiac and vascular modulation (such as reflexes from skeletal muscles, sympathetic afferents, epinephrine, and other humoral factors) may play a role. When these parameters are changed in opposite directions, the circulation is under baroreceptor reflex control, as shown by the virtual disappearance of these events after sinoaortic denervation. Note that the time scales of the two phenomena are different: parallel increases and decreases in blood pressure and heart rate can be shown over half-hour periods, whereas opposite blood pressure and heart rate changes last no more than a few beats. Dr. William J. Lawton (University of Iowa, Iowa City): Most of the data shown involved intra-arterial pressures. Are there any words of caution or differences in extrapolating data from intra-arterial pressures to noninvasive ambulatory blood pressure recording? Dr. Mancia: Several phenomena documented by intraarterial monitoring (such as nighttime fall in blood pressure and blood pressure variability, lack of placebo effect, and improved reproducibility) have been seen also with noninvasive automatic monitoring. Caution is mandatory, however, because automatic blood pressure-monitoring devices are somewhat inaccurate when used in ambulatory conditions. 6 ' 100 ' 101 Furthermore, with noninvasive monitoring, there is little hope that one will be able to precisely assess blood pressure variablity by taking one blood pressure value every minutes or so. Nevertheless, any clinical largescale future of ambulatory blood pressure monitoring is based on the noninvasive approach, which justifies technical research in this area. Dr. William J. Lawton (University of Iowa, Iowa City): We have seen patients with relatively normal or acceptable office and home blood pressures but elevated ambulatory blood pressure measurements (opposite to the example you showed). Could you comment on this pattern? Dr. Mancia: We have also come across the pattern you mention, but in our experience this is extremely rare. It is unknown whether having low office and home blood pressures but a high ambulatory blood pressure means an increased risk or represents just a normal variant within the ambulatory blood pressure distribution of the healthy population. This, I believe, might be answered by the Pamela study, 90 which collects office, home, and ambulatory blood pressures. Dr. Annette E. Fitz (University of Iowa, Iowa City): Since nocturnal blood pressure falls in most patients, does including these values in a 24-hour average reduce the differentiating effects of blood pressure averages, or are there periods during the day that would better differentiate groups of hypertensive patients, as, for example, 6 AM to noon or noon to evening? Mancia et al Ambulatory Blood Pressure 521 Dr. Mancia: Your question addresses the crucial issue of whether some blood pressure values within the 24 hours are clinically more important than others. The answer is, I am afraid, that it is not known, although various investigators have described the correlation of organ damage with blood pressure at work, daytime, or nighttime or emphasized (with no clinical evidence) the danger of morning blood pressure rise. 102 We need prospective controlled trials to answer these questions. Dr. Gerald F. DiBona (University of Iowa, Iowa City): Is it true that, although the statistical power is adequate for evaluation of efficacy over a 24-hour period, it is not adequate for evaluation of peak-to-trough efficacy over shorter (i.e., hourly) time periods? Dr. Mancia: Twenty-four-hour average blood pressure is much more reproducible when taken weeks apart than office blood pressure. Hourly average blood pressures are less reproducible than 24-hour average blood pressure, although on the whole not worse than office blood pressure. This means that studies on antihypertensive drug efficacy can combine an adequate power with a marked reduction in the number of patients only if one takes 24-hour average blood pressure. If one wants to look at the blood pressure-lowering effect hour by hour (and focus in particular on the trough effect), then the number of patients should be adequately increased. The increase should be such as to account for the hourly average value that has shown the worst reproducibility. Dr. Lawrence G. Hunsicker (University of Iowa, Iowa City): Couldn't you get around the problem of variability in hour-by-hour determinations of blood pressure by having the same number of patients, but have each patient wear the ambulatory blood pressure monitor for a greater number of days? It seems that what is needed is not more patients, but more blood pressure determinations at each point in each patient. Dr. Mancia: I couldn't agree more. The phenomenon is probably due, at least in part, to the fact that means are more reproducible than single values, and one could probably obtain the same advantage by increasing the number of clinical blood pressure values obtained in a given session or the days in which blood pressure is taken. Ambulatory blood pressure, however, has the additional advantage that it allows one to document the antihypertensive effect in conditions similar to those of daily life. Dr. Sheng Xing Ma (University of Iowa, Iowa City): How long do patients stay in the hospital when they are having 24-hour arterial pressure measured? Is each one the same? Dr. Mancia: Twenty-four-hour arterial blood pressure is performed in hospitalized patients who are free to move within the hospital area and live the social life of inpatients not confined to bed. Usually, the recording is done after 1 week of hospitalization. Noninvasive ambulatory blood pressure is more commonly done on an outpatient basis, the device being placed in the morning and removed when the subject comes back the following day. We do not make any attempt to rigidly standardize behavior during the recording, but all recordings are made on working days. Dr. Sheng Xing Ma (University of Iowa, Iowa City): Do you have data related to changes in sympathetic activi-

13 522 Hypertension Vol 21, No 4 April 1993 ties or norepinephrine concentrations and changes in arterial pressure during the 24-hour period? Dr. Mancia: We do not have data on 24-hour distribution of plasma catecholamines, but this has been shown by other groups. Plasma catecholamines fall during the night and increase during the morning concomitant with the blood pressure and heart rate increase. 103 It appears that the catecholamine increase does not precede but accompanies and follows arousal and resumption of physical activity. References 1. Riva Rocci S: The technique of sphygmomanometry [in Italian]. Gazz Med Torino 1897;10: Sokolow M, Werdegar S, Kain H, Hinman AT: Relationship between level of blood pressure measured casually and by portable recorders and severity of complications. Circulation 1966;34: Bevan AT, Honour AJ, Stott FH: Direct arterial pressure recording in unrestricted man. Clin Sci 1969^6: Richardson DW, Honour AJ, Fenton GW, Stott FH, Pickering GW: Variation in arterial pressure throughout the day and night Clm Sci 1964;26: Mancia G, Zanchetti A: Blood pressure variability, in Zanchetti A, Tarazi RC (eds): Handbook of Hypertension. Amsterdam, Elsevier, 1986, vol 7, pp Mancia G: Ambulatory blood pressure monitoring: Research and clinical applications. J Hypertens 1990;8(suppl 7):Sl-S13 7. Pickering GW: High Blood Pressure. London, Churchill, Mancia G, Parati G, Casadei R, Groppelli A, Zanchetti A: Effect of stress on diagnosis of hypertension. Hypertension 1991;17(suppl III):III-56-III Mancia G, Bertinieri G, Grassi G, Parati G, Pomidossi G, Ferrari A, Gregorini L, Zanchetti A: Effects of blood pressure measurements by the doctor on patient's blood pressure and heart rate. Lancet 1983;2: Mancia G, Parati G: Reactivity to physical and behavioural stress and blood pressure variability in hypertension, in Julius S, Bassett DR (eds): Handbook of Hypertension: Volume 9, Behavioural Factors in Hypertension. Amsterdam, Elsevier, 1987, pp Parati G, Pomidossi G, Casadei R, Ravogli A, Groppelli A, Cessna B, Mancia G: Comparison of the cardiovascular effects of different laboratory stressors and their relationship with blood pressure variability. J Hypertens 1988;6: Pickering TG, James GD, Harshfield GA, Blank S, Laragh JH: How common is white coat hypertension? JAMA 1988;259: Mancia G, Parati G, Pomidossi G, Grassi G, Casadei R, Zanchetti A; Alerting reaction and rise in blood pressure during measurement by physician and nurse. Hypertension 1987;9: Julius S, Ellis CN, Pasqual AV, Matice M, Hansjon L, Hunyor SN, Sandier L: Home blood pressure determination: Value in borderline (labile) hypertension. JAMA 1974;229: Parati G, Pomidossi G, Casadei R, Mancia G: Lack of alerting reactions and pressor responses to intermittent cuff inflations during non-invasive blood pressure monitoring. Hypertension 1985;7: Mutti E, Trazzi S, Omboni S, Parati G, Mancia G: Effect of placebo on 24h non-invasive ambulatory blood pressure. J Hypertens 1991;9: Beevers DG: Blood pressure that falls on rechecking. Br Med J 1982;284: Mathieu G, Biron P, Roberge F, Picard JM, Boulet C, Allard C: Blood pressure determinations during medical examinations: How many? Can J Public Health 1974;65: Management Committee: The Australian therapeutic trial in mild hypertension. Lancet 1980;l: Krum H, Louis WS, Brown DS, Jackman GP, Howes LG: Diurnal blood pressure variation in quadriplegic chronic spinal cord injury patients. Clin Sci 1991;80: Mancia G, Ferrari A, Gregorini L, Parati G, Pomidossi G, Bertinieri G, Grassi G, Di Rienzo M, Pedotti A, Zanchetti A: Blood pressure and heart rate variabilities in normotensive and hypertensive human beings. Circ Res 1983^3: Parati G, Pomidossi G, Casadei R, Groppelli A, Trazzi S, Di Rienzo M, Mancia G: Role of heart rate variability in the production of blood pressure variability in man. / Hypertens 1987;5: Baumgart P, Walger P, Fuchs G, Dorst KG, Wetter H, Rahn KH: Twenty four hour blood pressure is not dependent on endogenous circadian rhythm. Hypertension 1989;7: Furlan R, Guzzetti S, Crivellaro W, Dassi S, Tinelli M, Baselli G, Cerutti S, Lombardi F, Pagani M, Malliani A: Continuous 24-hour assessment of the neural regulation of systemic arterial pressure and RR variabilities in ambulant subjects. Circulation 1990;81: Parati G, Castiglioni P, Di Rienzo M, Omboni S, Pedotti A, Mancia G: Sequential spectral analysis of 24 hour blood pressure and pulse interval in humans. Hypertension 1990;16: Di Rienzo M, Castiglioni P, Mancia G, Parati G, Pedotti A: 24h sequential spectral analysis of arterial blood pressure, and pulse interval in free-moving subjects. IEEE Trans Biomed Eng 1989; 36: Di Rienzo M, Castiglioni P, Frattola A, Omboni S, Pedotti A, Mancia G, Parati G: Evaluation of the 1/f model to describe 24h blood pressure (BP) and heart rate (HR) profiles in man. (abstract) / Hypertens 1992;10(suppl 4):S Kobayashi M, Musha T: 1/f fluctuation of heart beat period. IEEE Trans Biomed Eng 1982;29: Cowley AW, Liard LF, Guyton C: Role of the baroreceptor reflex in dairy control of arterial blood pressure and other variables in dogs. Circ Res 1973^2: Ramirez AJ, Bertinieri G, Belli L, Cavallazzi A, Di Rienzo M, Pedotti A, Mancia G: Reflex control of blood pressure and heart rate by arterial baroreceptors and by cardiopulmonary receptors in the anaesthetized cat. J Hypertens 1985^3: Di Rienzo M, Parati G, Castiglioni P, Omboni S, Ferrari AU, Ramirez AJ, Pedotti A, Mancia G: Role of sinoaortic afferents in modulating BP and pulse interval spectral characteristics in unanesthetized cats. Am J Physiol 1991;261:H1811-H Persson PB, Ehmke H, Kohler WW, Kichheim HR: Identification of major slow blood pressure oscillations in conscious dogs. Am J Physiol 1990;259:H1050-H Mancia G, Parati G, Pomidossi G, Casadei R, Di Rienzo M, Zanchetti A: Arterial baroreceptor reflexes and blood pressure and heart rate variabilities in humans. Hypertension 1986;8: Watson RDS, Stallard TJ, Flinn RM, Littler WA: Factors determining pressure and its variability in hypertensive man. Hypertension 1980;2: Mann S, Altaian DG, Raftery EB, Bannister R: Circadian variation of blood pressure in autonomic failure. Circulation 1983;68: Floras JS, Hassan MO, VannJones J, Osikowska BA, Sever PS, Sleight P: Consequences of impaired arterial baroreceptor reflexes in essential hypertension: Effects on pressor responses, plasma noradrenaline and blood pressure variability. / Hypertens 1988;6: Conway J, Boon N, Davies C, Jerves VJ, Sleight P: Neural and humoral mechanisms involved in blood pressure variability. J Hypertens 1984;2: Ferrari AU, Daffonchio A, Albergati F, Mancia G: Inverse relationship between heart rate and blood pressure variability in rats. Hypertension 1987;10: Imai Y, Abe K, Munakata M, Sakuma H, Hashimoto J, Imai K, Sekino H, Yashinaga K: Does ambulatory blood pressure monitoring improve the diagnosis of secondary hypertension? J Hypertens 1990;8(suppl 6):S71-S Imai Y, Munakata M, Abe K, Sekino M: The circadian blood pressure variation in patients with disturbance of reninangiotensin-aldosterone system and with renal disease [in Japanese], in Koshikawa S, Nagasawa T, Koiso K, Ito H (eds): Annual Review Neurology. Tokyo, Chugaiga Kusha, 1990, pp Mancia G, Ferrari A, Gregorini L, Parati G, Pomidossi G, Zanchetti A: Prolonged intra-arterial blood pressure recording in the diagnosis of phoeochromocytoma. Lancet 1979;2: Tanaka T, Natsume T, Shibata H, Nozawa K, Kojima S, Tsuchiya M, Ashida T, Ikeda M: Circadian rhythm of blood pressure in primary aldosteronism and renovascular hypertension. Jpn Circ J 1983;47: Zito M, Parati G, Omboni S, Cervone C, Ulian L, D'Aviero M, Abate G, Mancia G: Effect of ageing on blood pressure variability. J Hypertens 1991;9(suppl 6):S328-S Millar-Craig MW, Mann S, Balasubramanian V, Raftery EB: Blood pressure variability in hypertension, in Stott FD, Raftery EB, Goulding L (eds): Proceedings, 3rd International Symposium

14 Mancia et al Ambulatory Blood Pressure 523 on Ambulatory Monitoring. London, Academic Press, 1979, pp Goldberg AD, Raftery EB, Cashman PMM, Stott FD: Study of untreated hypertensive subjects by means of continuous intraarterial blood pressure recordings. Br Heart J 1978;40: Floras JS, Hassan MO, VannJones J, Osikowska BA, Sever PS, Sleight P: Factors influencing blood pressure and heart rate variability in hypertensive humans. Hypertension 1988;11: Mancia G, Parati G, Albini F, Villani A; Circadian blood pressure variations and their impact on disease. / Cardiovasc Pharmacol 1988;12(suppl 7):S11-S Lakatta EG: Alterations in the cardiovascular system that occur in the advanced age. Fed Proc 1979;38: Ogawa J, Takata J, Nomura G, Arai S, Iketa T, Takabatake T, Mattori N: Effects of aging on the variability of arterial pressure in essential hypertensive patients. Clin Sci 1982;63:411s-414s 50. Ferrari AU, Grassi G, Mancia G: Alterations in reflex control of circulation with ageing, in Amery A, Staessen S (eds): Handbook of Hypertension, Hypertension in the Elderly. Amsterdam, Elsevier, 1989, pp Folkow B, Halback MIL: Physiopathology of spontaneous hypertension in rats, in Genest J, Koiw E, Kuchel O (eds): Hypertension. New York, McGraw-Hill Book Co, 1977, pp Folkow B: Structural myogenic humoral and nervous factors controlling peripheral resistance, in Harrington M (ed): Hypotensive Drugs. London, Pergamon Press, 1956, pp Eliasson K, Heimdahl P, Kahan T: Circulatory and sympathoadrenal responses to stress in borderline and established hypertension. J Hypertens 1983;1: Bertinieri G, Di Rienzo M, Cavallazzi A, Ferrari AU, Pedotti A, Mancia G: Evaluation of baroreceptor reflex by blood pressure monitoring in unanesthetized cats. Am J Physiol 1988;254: H377-H Parati G, Di Rienzo M, Bertinieri G, Pomidossi G, Casadei R, Groppelli A, Pedotti A, Zanchetti A, Mancia G: Evaluation of the baioreceptor-heart rate reflex by 24-hour intra-arterial blood pressure monitoring in humans. Hypertension 1988;12: Imai Y, Abe K, Miura Y, Nihei M, Sasaki S, Minami N, Munakata M, Taira N, Sekino H, Yamakoshi K: Hypertensive episodes and circadian fluctuations of blood pressure in patients with phoeochromocytoma: Studies by long-term monitoring based on a volume oscillometric method. / Hypertens 1988;6: Kannel WB: Role of blood pressure in cardiovascular morbidity and mortality. Prog Cardiovasc Dis 1974;17: Floras JS, Jones JV, Hassan MD, Osikowska B, Sever PS, Sleight P: Cuff and ambulatory blood pressure in subjects with essential hypertension. Lancet 1981^2: Parati G, Pomidossi G, Albini F, Malaspina D, Mancia G: Relationship of 24-hour blood pressure mean and variability to severity of target organ damage in hypertension. J Hypertens 1987;5: Perloff D, Sokokjw M, Cowan R: The prognostic value of ambulatory blood pressure. JAMA 1983;249: Rowlands DB, Ireland MA, Glover DR, McLeay RAB, Stallard TJ, Littler WA: The relationship between ambulatory blood pressure and echocardiographically assessed left ventricular hypertrophy. Clin Sci 1981;61:101s-103s 62. Devereux R, Pickering TG, Harshfield GA: Left ventricular hypertrophy in patients with hypertension: Importance of blood pressure response to regularly recurring stress. Circulation 1983; 68: White WB, Dey HM, Shulman P: Assessment of dairy blood pressure load as a determinant of cardiac function in patients with mild to moderate hypertension. Am Heart J 1989; 118: White WB, Schulman P, McCabe UJ, Dey HM: Average daily blood pressure, not office blood pressure determines cardiac function in patients with hypertension. JAMA 1989;261: Giaconi S, Levanti C, Fommei E, Innocenti F, Seghieri G, Palla L, Palombo C, Ghione S: Microalbuminuria and casual and ambulatory blood pressure monitoring in normotensives and in patients with borderline and mild essential hypertension. Am J Hypertens 1989^: Sbimada K, Kawanoto A, Matsubayishi K, Ozawa T: Silent cerebrovascular disease in the elderly: Correlation with ambulatory pressure. Hypertension 1990;16: 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: Perloff D, Sokolow M, Cowan RM, Juster RP: Prognostic value of ambulatory blood pressure measurements: Further analyses. J Hypertens 1989;7(suppl 3):S3-S Staessen J, Fagard R, Lijnen P, Thijs L, VanHoof R, Amery A: Reference values for ambulatory blood pressure: A metaanarysis. J Hypertens 1990;68(suppl 6):S57-S Dawber TR: The Framingham Study: The Epidemiology of Atherosclerotic Disease. Cambridge, Mass, Harvard University Press, Dyer A: An analysis of the relationship of systolic blood pressure, serum cholesterol, and smoking to 14-year mortality in the Chicago People Gas Company Study II. Coronary and cardiovascular renal mortality in two competing risk models, (abstract) J Chronic Dis 1975;28:57s 72. Dyer AR, Stamler J, Paul O, Berkson DM, Lepper MH, McKean H, Shekelle RB, Iindberg HA, Garside D: Alcohol consumption, cardiovascular risk factors, and mortality in two Chicago epidemiologic studies. Circulation 1977^6: Stamler J, Berkson D, Iindberg H: Risk factors: Their role in the etiology and pathogenesis of the atherosclerotic diseases, in Wissler R, Geer J (eds): The Pathogenesis of Atherosclerosis. Baltimore, Md, Williams & Wilkins Co, 1971, p MacMahon S, Peto R, Cutler J, Callius R, Sorlie P, Neaton J, Abbott R, Godwin J, Dyer A, Stamler J: Blood pressure, stroke and coronary heart disease: Part I. Prolonged differences in blood pressure: Prospective observational studies corrected for the regression dilution bias. Lancet 1990^335: Julius S, Mejia A, Jones K, Krause L, Shork N, VandeVen C, Johnson E, Petrin J, Sekkarie MA, Kjeldsen SE, Schmouder R, Gupta R, Ferraro J, Nazzaro P, Weissfeld J: "White coat" versus "sustained" borderline hypertension in Tecunseh, Michigan. Hypertension 1990;16: Mancia G, Parati G: Clinical significance of "white coat" hypertension, (editorial comment) Hypertension 1990;16: Gould BA, Mann S, Davies AB, Altaian DG, Raftery EB: Does placebo lower blood pressure? Lancet 1981;2: Conway J, Coats A: Value of ambulatory blood pressure monitoring in clinical pharmacology. J Hypertens 1989;7(suppl 3):S29-S Trazzi S, Mutti E, Frattola A, Imholz B, Parati G, Mancia G: Reproducibility of non-invasive and intra-arterial blood pressure monitoring: Implications for studies on antihypertensive treatment. / Hypertens 1991;9: Mancia G, Omboni S, Parati G, Trazzi S, Mutti E: Limited reproducibility of hourly blood pressure values obtained by ambulatory blood pressure monitoring: Implications for studies on antihypertensive drugs. / Hypertens 1992;10: Penaz J: Photoelectric measurement of blood pressure, volume and flow in the finger, in Digest of the International Conference on Medicine and Biological Engineering. Dresden, Conference Committee of the 10th International Conference on Medicine and Biological Engineering, 1973, p Wesseling KH, Settels JJ, De Wit B: The measurement of continuous finger arterial pressure non-invasively in stationary subjects, in Schmidt TH, Dembrosky TM (eds): Biological and Physiological Factors in Cardiovascular Diasease. Berlin, Springer- Veriag, 1986, pp Parati G, Casadei R, Groppelli A, Di Rienzo M, Mancia G: Comparison of finger and intra-arterial blood pressure monitoring at rest and during laboratory testing. Hypertension 1989;13: Imholz BPM, Langewouters GJ, van Montfrans GA, Parati G, van Goudoever J, Wesseling KH, Wieling W, Mancia G: Feasibility of ambulatory, continuous 24-hour finger arterial pressure recording. Hypertension 1993;21: Omboni S, Parati G, Frattola A, Castiglioni P, Di Rienzo M, Pedotti A, Mancia G: Spectral analysis of blood pressure and pulse interval variability from finger and intra-arterial recordings, in Di Rienzo M, Mancia G, Pedotti A, Parati G, Zanchetti A (eds): Blood Pressure and Heart Rate Variability. Amsterdam, IOS Press, 1992, pp Pagani M, Lombardi F, Guzzetti S, Rimoldi O, Furlan R, Pizzinelli P, Sandrone G, Malfatto G, DeH'Otto S, Piccaluga E, Turiel M, Baselli G, Malliani A: Power spectral analysis of heart rate and arterial pressure variabilities as a marker of sympathovagal interaction in man. Ore Res 1986^9: Rimoldi O, Pierini S, Ferrari A, Cerutti S, Pagani M, Malliani A; Analysis of short term oscillations of RR and arterial pressure in conscious dogs. Am J Physiol 1990;258:H967-H976

15 524 Hypertension Vol 21, No 4 April Parati G, Frattola A, Omboni S, Castiglioni P, Di Rienzo M, Pedotti A, Mancia G: Blood pressure variability over 24 hours: A complex phenomenon, in Cerutti S, Minuco G (eds): Spectral Analysis of Heart Rate Variability Signal Methodological and Clinical Aspects. Pavia, La Goliardica Pavese, Quaderni di Medicina del Lavoro e Medicina Riabilitativa, 1991, pp Daflonchio A, Franzelli C, Di Rienzo M, Castiglioni P, Ramirez A, Parati G, Mancia G, Ferrari AU: Effect of sympathectomy on blood pressure variability in the conscious ral J Hypertens 1991; 9(suppl 6):S70-S Cesana G, De Vito G, Ferrario M, Libretti A, Mancia G, Mocarelli P, Sega R, Valagussa F, Zanchetti A; Ambulatory blood pressure normalcy: The Pamela study. / Hypertens 1991: 9(suppl 3):S17-S Staessen J, Bulpitt CJ, Fagard R, Mancia G, O'Brien ET, Thijs L, Vyncke G, Amery A: Reference values for the ambulatory blood pressure and the blood pressure measured at home: A population study. / Hum Hypertens 1991^: Mancia G, Zanchetti A, on behalf of the investigators: Prognostic value of ambulatory blood pressure: The SAMPLE study. High Blood Press 1992^: Clement DL, on behalf of the Steering Committee: Office versus ambulatory recordings of blood pressure (OvA): A European multicenter study. / Hypertens 1990;8{suppl 6):S39-S Kannel WB: Left ventricular hypertrophy as a risk factor: The Framingham experience. / Hypertens 1991;9(suppl 2):S3-S9 95. Levy D, Garrison RJ, Savage DD, Kannel WB, Castelli WP: Prognostic implications of echocardiographically determined left ventricular hypertrophy mass in the Framingham heart study. N Engl J Med 1990^22: Devereux RB, Pickering TG: Relationship between ambulatory or exercise blood pressure and left ventricular structure: Prognostic implications. / Hypertens 199O;8(suppl 6):S125-S Ferrari AU, Daffonchio A, Albergati F, Mancia G: Differential effects of aging on the heart rate and blood pressure influences arterial baroreceptors in awake rats. J Hypertens 1991;9: Mancia G, Grassi G, Bertinieri G, Ferrari A, Zanchetti A: Arterial baroreceptor control of blood pressure in man. JAuton Nerv Syst 1984;11: Mancia G, Zanchetti A: Cardiovascular regulation during sleep, in Orem J (ed): Physiology During Sleep. New York, Academic Press, Inc, 1980, pp Casadei R, Parati G, Pomidossi G, GroppeUi A, Trazzi S, Di Rienzo M, Mancia G: 24-hour blood pressure monitoring: Evaluation of Spacelabs 5300 monitor by comparison with intraarterial recording in ambulant subjects. / Hypertens 1988;6: GroppeUi A, Omboni S, Parati G, Mancia G: Evaluation of noninvasive blood pressure monitoring devices Spacelabs and versus resting and ambulatory 24-hour intra-arterial blood pressure. Hypertension 1992;20: Millar-Craig M, Bishop C, Raftery E: Orcadian variation of blood pressure. Lancet 1978;l: Turton MB, Deegan T: Orcadian variations of plasma catecholamine, cortisol, and immunoreactive insulin concentrations in supine subjects. Clin Chim Ada 1974^5: KEY WORDS blood pressure monitoring, ambulatory antihypertensive therapy behavior baroreceptor sympathetic nervous system aging reflex Presentation and publication of this Clinical Conference is supported by an educational grant from Health Sciences Service, Merck Sharp & Dohme.