ACE-Inhibitor Therapy with Spirapril Increases Nocturnal Hypotensive Episodes in Elderly Hypertensive Patients

Home , Spirapril

Journal of Human Hypertension (2001) 15, 873–878  2001 Nature Publishing Group All rights reserved 0950-9240/01 $15.00 www.nature.com/jhh ORIGINAL ARTICLE ACE-inhibitor therapy with spirapril increases nocturnal hypotensive episodes in elderly hypertensive patients

I Kantola1, A Tere´nt2, M Kataja3 and E Breig-A˚ sberg4 1Department of Medicine, Turku University Central Hospital, Turku, Finland; 2Department of Medicine, Uppsala University Hospital, Uppsala, Sweden; 3National Public Health Institute, Helsinki, Finland; 4Department of Medicine, So¨derhamn Hospital, So¨derhamn, Sweden

The purpose of this double-blind, randomised trial with sodes were observed during the antihypertensive treat- a 4-week placebo run-in period followed by an active ment (one in the 3 mg group and 10 in the 6 mg group, treatment period using either spirapril 3 mg or 6 mg P Ͻ 0.01 between the two treatment groups). Fifty-four once a day was to clarify the existence of hypotensive episodes of MAP Ͻ70 mm Hg were observed during the episodes in elderly hypertensive patients treated by an placebo period and 117 during the treatment period ACE-inhibitor. Forty hypertensive patients aged 60–76 (P Ͻ 0.001). During the placebo period low MAPs were years underwent 24-h ABPM at the end of the run-in observed only during night time. During the treatment (week 4) and active treatment (week 9) periods. The period they were seen also from 11 am to 4 pm. In con- mean 24-h systolic blood pressure (SBP) decreased clusion, ACE-inhibitor therapy with spirapril signiﬁ- from 161.9 (26.7) mm Hg to 150.6 (29.9) mm Hg cantly increased hypotensive episodes in elderly hyper- (P Ͻ 0.001) and diastolic blood pressure (DBP) from tensive patients which may worsen their cerebral and 91.70 (14.7) mm Hg to 84.2 (17.3) mm Hg (P Ͻ 0.001). No myocardial circulation episodes of mean arterial pressure (MAP) Ͻ50 mm Hg Journal of Human Hypertension (2001) 15, 873–878 were seen during the placebo period. Instead 11 epi-

Keywords: hypotension; ambulatory blood pressure measurement; angiotensin-converting enzyme inhibition

Introduction lowest SBP 100–116 mm Hg) compared to dippers (nocturnal SBP fall 13–15%, nocturnal lowest SBP Hypertension, apart from being a risk factor for cor- 116–122 mm Hg). onary artery disease and stroke, causes adaptive cer- In hypertensive patients the occurrence of cardiac ebral vascular changes, leading to a shift of the lower ischaemic episodes could be influenced either by limit of autoregulation towards higher pressure with 1 high BP levels, through an increase of myocardial an impaired tolerance to pressure decrease. In oxygen demand, or by low BP values, through a young and middle-aged effectively treated hyperten- reduction of coronary perfusion due to an upward sive patients, cerebral blood flow autoregulation shift of the lower limit of coronary autoregulation in may or may not be readapted towards normal.1 In the subendocardium.3–5 Pierdomenico et al6 found elderly hypertensive patients non-reversible vascu- nocturnal ischaemia significantly more frequently in lar structural changes would be expected to prevail. non-dippers than in dippers and overdippers among Further, atherosclerotic narrowing of the larger cerebral arteries may contribute to impairment of auto- untreated hypertensive patients with coronary regulation by exhausting autoregulatory arteriolar artery disease. Drug therapy significantly reduced dilatation. In fact Kario et al2 have shown in their nocturnal ischaemia in non-dippers, had no signifi- study that the extent of silent cerebrovascular dam- cant effect in dippers and significantly increased age was more severe in extreme dippers (nocturnal night time ischaemia in overdippers. During treat- systolic blood pressure (SBP) fall 21–25%, nocturnal ment, night time ischaemia was significantly more frequent in overdippers than in dippers and non- dippers. Besides the risk of lowering BP too much in eld- Correspondence: Ilkka Kantola, Department of Medicine, Turku erly people, there is also the risk of over-diagnosing University Central Hospital, FIN-20520 Turku, Finland. E-mail 7 ilkka.kantolaȰtyks.fi hypertension. The white-coat effect has been Received 4 June 2001; revised and accepted 12 July 2001 shown to correlate positively with the age of the Nocturnal hypotensive episodes I Kantola et al 874 patient.8 In the study by Kario et al2 the frequency of Procedures silent cerebrovascular damage in the patients with sustained hypertension was greater than that in the Heart rate and BP were registered on three occasions patients with white coat hypertension. during the 4-week placebo phase. Patients entered This study is a part of a Finnish-Swedish efficacy the active treatment phase if office DBP was 100–114 and safety study of spirapril, an ACE-inhibitor, in mm Hg and SBP р200 mm Hg. BP was measured in elderly hypertensive patients. In this part of the the sitting position on the right arm in the morning study 24-h ambulatory BP measurement was perfor- after a 10-min rest, before the intake of placebo or med on the Swedish participants of the study to active drug. The mean value of two measurements clarify the 24-h BP profile of spirapril, especially its was used. nocturnal antihypertensive effects, in elderly hyper- Ambulatory BP measurement was performed at tensive patients. the end of the placebo period and at 5 weeks of spirapril treatment. For ambulatory BP measurement, the Accutracker II (version 30/23, Suntech Medical Subjects and methods Instruments, Raleigh, NC, USA)9 was used. The BP cuff was placed on the left upper arm. The quality Study design of the ambulatory BP measurement was checked by simultaneous auscultatory BP by one of the authors The patients were randomised in a double-blind ˚ fashion to spirapril either 3 or 6 mg for 6 weeks after (EBA). The BP was recorded approximately every a 4-week single-blind placebo wash-out period. The 20 min (randomly distributed by the device within dose remained stable during the period of active the 15–25 min range). The device automatically repeated the measurement if any of the following medication and was taken once daily before break- Ͼ Ͻ fast. were recorded: SBP 220 mm Hg, SBP 80 mm Hg, change of SBP Ͼ50 mm Hg between two measurements, DBP Ͼ130 mm Hg, DBP Ͻ40 mm Hg, change of DBP Ͼ40 mm Hg, pulse pressure Ͻ20 mm Hg or Subjects Ͼ110 mm Hg, or pulse pressure change of Ͼ40 Twenty male and 20 female hypertensive patients mm Hg. No values were censored except for the with office diastolic BP (DBP) 100–114 mm Hg and value preceding a repeat measurement. Repeat SBP р200 mm Hg aged 60–76 years were recruited. values outside these limits were not excluded from Patients with a myocardial infarction within the pre- the statistical analyses in order not to miss true vious 6 months, severe or unstable angina pectoris, values at the extremes. congestive heart failure, non-controlled diabetes According to the study protocol, patients were mellitus, renal artery disease or arrhythmias were classified as responders if the office DBP was р90 not included. mm Hg or decreased by у10 mm Hg during the Eighteen patients, seven men and 11 women, were active treatment phase. randomised to the 3 mg dose, and 22 patients, 13 According to the ambulatory BP measurement the men and nine women, to the 6 mg dose. Baseline patients were classified as hypertensives if either characteristics of the patients are given in Table 1. their average daytime SBP was Ͼ140 mm Hg or aver- The patients did not have either coronary heart dis- age night time SBP Ͼ120 mm Hg or average daytime ease or cerebral artery disease. Other antihyperten- DBP Ͼ90 mm Hg or average night time DBP Ͼ80 sive agents, long-acting nitrates, beta-adrenergic mm Hg (the recommendations of the University of blocking agents, calcium-channel blocking agents, Connecticut Group).10 Patients who fulfilled the diuretics, MAO-inhibitors, tricyclic antidepressants, office criteria for inclusion, but had normal ambulat- tranquilizers, lipid-lowering agents and potassium ory BP measurement curves, were classified as supplements were prohibited. white-coat hypertensives.

Table 1 Baseline characteristics of the patients. Mean (s.d.)

Spirapril 3 mg Spirapril 6 mg Total (n = 18) (n = 22) (n = 40)

Age (yrs) 66.3 (4.6) 65.0 (3.3) 65.6 (4.0) Body weight (kg) 77.8 (11.6) 80.5 (12.2) 79.3 (11.8) Height (cm) 168.2 (9.0) 171.0 (10.0) 169.8 (9.6) SBPa (mm Hg) 161.9 (8.1) 161.0 (10.2) 161.4 (9.2) DBPa (mm Hg) 102.4 (3.1) 104.1 (3.2) 103.3 (3.3) Heart ratea 79.9 (9.7) 78.3 (13.1) 79.0 (11.6)

aAfter 10 min in the sitting position at the end of the placebo period.

Journal of Human Hypertension Nocturnal hypotensive episodes I Kantola et al 875 The deﬁnition of a low BP in this study was MAP Ͻ50 mm Hg (critically low), corresponding to the lower limit of cerebral autoregulation in healthy per- sons; and MAP Ͻ70 mm Hg (low), representing the lower limit of cerebral autoregulation in hypertensive patients.1,11 The MAP for each measurement was automatically calculated by the Accutracker II equipment.

Safety Adverse events were checked on four occasions evenly spread throughout the study period. Any change in the patient’s condition, even a common cold or shoulder pain, was noted as an adverse event, irrespective of whether a casual relationship to drug therapy was plausible or not. Figure 1 Smoothed (three-point moving median) curves showing the placebo-subtracted antihypertensive effect of active treatment with spirapril, 3 mg and 6 mg (all 40 patients). Statistical analyses The circadian BP curves are smoothed by applying the three-point moving medians to them (in each gradually declined until 1 to 2 am, when it had place the middle one in size order of the three con- nearly disappeared (Figure 1). No second peak of secutive measurements is taken). The advantage of treatment effect, during sleeping hours, was found. the moving median is the removing of all large sin- The smoothed MAP curves revealed a sharp decline gle deviations. in MAP, from 110 to 95 mm Hg, between 9 and 12 Proportions were compared with the Fisher’s pm (Figure 2). This decline seemed to be as rapid exact, the Chi-square and the Mc’Nemar tests, and and large in the placebo period as in the period of continuous variables by two-tailed t-test and active therapy. ANOVA. The aim was to recruit 40 patients assuming a stat- Patients responding to therapy istical power of at least 80%. Eight (44%) of the patients in the 3 mg spirapril group and 12 (55%) in the 6 mg group reached the Results treatment goal, ie ofﬁce DBP р90 mm Hg or a fall in у = In total, 6535 ambulatory BP measurements were DBP 10 mm Hg (P NS between the groups). registered, of which 954 (17.1%) were repeats which Among the ofﬁce responders, the frequency of MAP Ͻ70 mm Hg was 65%, compared with 25% among almost equals that found (16%) in the validation Ͻ study.8 The number of uncensored readings aver- non-responders (P 0.05). The response rate was aged 68.6 (range 47–76)/ambulatory measurement. lower according to the 24-h ABPM (10) (11% in the 3 mg group and 32% in the 6 mg group, P = 0.12).

Effects on 24-h ambulatory BP The mean 24-h daytime SBP was during the placebo period 161.8 (21.5) mm Hg and 152.9 (22.3) mm Hg during active treatment (P Ͻ 0.001). The corresponding DBP values were 91.2 (9.7) mm Hg and 85.0 (11.1) mm Hg (P Ͻ 0.001). The night time SBP was 139.1 (25.1) mm Hg during the placebo period and 139.0 (27.4) mm Hg during the active treatment (P = NS). The corresponding DBP values were 75.7 (12.5) and 74.4 (12.6) mm Hg (P Ͻ 0.01). The nocturnal SBP fall was 22.7 (14.2) mm Hg during the placebo period and 13.9 (17.6) mm Hg during the active treatment (P Ͻ 0.001). The corresponding DBP falls were 15.5 (9.1) mm Hg and 10.5 (8.0) mm Hg (P Ͻ 0.001). The placebo-subtracted curves showed a maximal treatment effect of approximately 16 mm Hg on SBP Figure 2 Smoothed (three-point moving median) 24-h MAP and 10 mm Hg on DBP, during the ﬁrst hours after curves of the patients treated with either placebo (at 4 weeks) or taking medication. Thereafter, the treatment effect spirapril 3 mg (n = 18) or 6 mg (n ϭ 22) (at 9 weeks).

Journal of Human Hypertension Nocturnal hypotensive episodes I Kantola et al 876 Table 2 The pattern of critically low or low BP during the placebo and the active treatment periods. The percentage ﬁgures within parenthesis refer to the total number of BP measurements or the total number of patients in each group

Spirapril 3 mg (n = 18) Spirapril 6 mg (n = 22) Total (n = 40)

placebo active placebo active placebo active (%) (%) (%) (%) (%) (%)

Episodes of MAP Ͻ50 mm Hg 0 (0) 1 (0.08)a 0 (0) 10 (0.65)a 0 (0) 11 (0.4)b No. patients with MAP Ͻ50 mm Hg 0 (0) 1 (5.6) 0 (0) 3 (13.6) 0 (0) 4 (10.0) Episodes of MAP Ͻ70 mm Hg 11 (0.9) 31 (2.5)c 43 (2.9) 86 (5.8)c 54 (1.9) 117 (4.2)d No. patients with MAP Ͻ70 mm Hg 4 (22.2) 7 (38.9) 8 (36.4) 11 (50.0) 12 (30.0) 18 (45.0)

aP Ͻ0.01, bP Ͻ0.01, cP Ͻ0.0001, dP Ͻ0.001 comparison between placebo and active treatments.

Pattern of critically low blood pressures MAPs were almost exclusively observed during night time (11 pm to 6 am). During active treatment, No episodes of MAP Ͻ50 mm Hg were seen during MAP values Ͻ70 mm Hg were also seen during day- the placebo phase, instead 11 (0.4% of measure- time (11 am to 5 pm). The probability of MAP Ͻ70 ments, P Ͻ 0.01 compared to placebo) episodes were mm Hg increased during night time, from 5% in the observed in one patient in the 3 mg spirapril group placebo period to 10% during active therapy. Con- and three patients in the 6 mg spirapril group (P cerning the critically low values (MAP Ͻ50 mm Hg), Ͻ 0.01 between the two groups) (Table 2). four of the 11 episodes occurred during the early MAP was Ͻ70 mm Hg in 1.9% of the measure- sleeping hours (11 pm to 1 am), while the remaining ments (12 patients) during the placebo phase and in seven episodes were evenly distributed (7 am to 5 4.2% of the measurements during the treatment pm). phase (18 patients) (P Ͻ 0.001) (Table 2). Twelve The pulse pressure was significantly higher during (67%) of those 18 patients had episodes of two or the placebo period, according to office measurements, more successive low MAPs. The number of episodes in patients with MAP Ͻ70 mm Hg than in other of MAP Ͻ70 mm Hg was higher in the spirapril 6 mg patients 62.6 (12.5 s.d.) mm Hg vs 58.3 (10.9) mm Hg, group than in the 3 mg group, both during the pla- (P Ͻ 0.01). During the placebo period office SBP did cebo period and during the period of active therapy not differ between the patients with MAP Ͻ70 mm Hg (P Ͻ 0.0001). and those with MAP у70 mm Hg (157.2 (13.0) mm Hg Forty (23.4%) of 171 MAPs below 70 mm Hg were vs 157.7 (11.6) mm Hg). Instead DBP was significantly repeat values, ie they were registered by an auto- lower in the hypotensive patients than in the non- matic repeat measurement. Such repeats followed hypotensive patients (94.6 (6.5) mm Hg vs 99.4 (6.5) extreme values registered at regular time intervals. mm Hg, P Ͻ 0.001). Heart rate was a little higher (81.3 The diurnal distributions of MAP Ͻ70 mm Hg (9.6) bpm and 78.1 (12.9) bpm, respectively, P Ͻ 0.05). during the placebo and the active treatment periods are shown in Figure 3. In the placebo period low White-coat effect Two patients in both spirapril groups were normotensive in the placebo phase according to the ABPM (criteria of the University of Connecticut Group).10 Three of them had MAP values Ͻ70 mm Hg during active treatment. Twenty-one of the 36 true hypertensive patients did not have critically low BP during spirapril treatment.

Clinical safety Adverse events were seen in 14 (78%) of the patients with MAP Ͻ70 mm Hg and in 13 (59%) of the other patients (NS). The number of adverse events (some of which occurred in the same patient) was 14 in the group with low BP and 18 in the other group. Figure 3 Frequency diagram of registered MAP Ͻ70 mm Hg in The adverse events were rather well tolerated, and relation to wall clock time. Registrations made at the end of the no patient had to discontinue medication because placebo period and at the end of active treatment. of them.

Journal of Human Hypertension Nocturnal hypotensive episodes I Kantola et al 877 Discussion left ventricular hypertrophy and functional and structural alterations in the small coronary ves- According to our study antihypertensive medication sels.3,5,18,19 with the ACE-inhibitor spirapril significantly The relationship between CBF and MAP has been increased hypotensive episodes in these elderly established by invasive BP measurements.1,20 In gen- hypertensive patients. eral, the non-invasive approach tends, when com- The increasing amount of hypotensive episodes pared with the invasive approach, to underestimate during antihypertensive treatment seen in our study SBP and overestimate DBP,21 but the differences are may expose the elderly patients to the hazard of a small. The accuracy of non-invasive ABPM readings critical fall in cerebral blood flow (CBF) because in daily life is rather poor,22 but the readings of an hypertension leads to a shift of the lower limit of Accutracker II have been shown not to deviate more cerebral autoregulation towards higher pressure than 4 mm Hg from intra-arterial readings, both at with an impaired tolerance to pressure decrease.1 rest and during exercise.23 Therefore, it seems This functional haemodynamic adaptation is prob- reasonable to assume that our low MAP values data ably caused by structural hypertensive vascular indicate real values close to the lower limit of cer- changes.12 In middle-aged effectively treated hyper- ebral autoregulation. tensive patients CBF autoregulation may be read- In ABPM there is a risk of missing important infor- apted towards normal.1 However, in elderly hyper- mation about readings on the border of physiologi- tensive patients non-reversible hypertensive cally acceptable ranges. The automatic data cen- vascular structural changes would be expected to soring introduced to reduce movement artifacts may prevail and hence functional vascular readaptation induce bias. It tends to reject low-level outliers and is not likely to take place when BP is lowered. Fortu- night-time readings to a greater extent than high- nately the brain can extract additional oxygen when level outliers.22 The device used in this study perfusion is reduced. (Accutracker II) does not reject readings automati- The effect of spirapril on cerebral autoregulation cally but repeats the measurement. We censored has not been studied. Captopril given intravenously only primary extreme readings to avoid artifacts, but has shortened the autoregulatory plateau shifting no repeat readings. It seems improbable that low both the lower and the upper limit of autoregulation MAP values seen were only caused by technical towards lower BP in acute studies in rats.13 During reasons. chronic treatment the shift of the lower limit of auto- The hypotensive episodes were more common regulation towards lower BP is retained, while the among the patients with lower achieved BP level. less desirable shift of the upper limit is not Thus it might be rational not to set the goal of BP present.14 Also in heart failure patients captopril too low in elderly patients although the HOT- similarly maintained a normal CBF as very low study24 and the former studies in the elderly25 have BP.15,16 The effect of spirapril is probably compara- not shown increased cardiovascular risk even in ble to the effect of captopril because it has been sug- quite low DBP levels. gested that the ACE-inhibitors exert their effect by In our study the treatment of white coat patients inhibiting angiotensin II-dependent tone in the led to frequent hypotensive episodes which suggests larger cerebral resistance vessels.17 Thus it is poss- that white coat patients should be picked up to ible that the change towards more hypotensive epi- avoid unnecessary antihypertensive medications. sodes seen in our study is alleviated by the use of The clinical importance of our observations with the ACE-inhibitor treatment. However, we did not regard to the risk of developing hypotensive tissue measure in our study the ischaemia of target organs damage is unclear because we did not measure tar- which leads only to speculation of the effect of the get organ ischaemia. In patients with glaucoma, a hypotensive episodes seen. correlation between progressive visual field deterio- All the investigations concerning the effects of ration and nocturnal hypotension has been antihypertensive medication on CBF have been per- observed.26 In healthy humans, responses of CBF to formed during daytime, not during sleeping hours, sudden changes in arterial BP have been found to when most of the hypotensive episodes in this study be extremely rapid. Also, the brain can maintain an were found. Therefore, little or nothing is known unchanged metabolism by extracting more oxygen about regional CBF in any clinical situation other from the blood, when CBF falls.27,28 However, these than that of the awake and supine patient. compensatory mechanisms are assumed to be weak- In the study of Pierdomenico et al6 myocardial ened in the elderly.2 ischaemia increased during the night time in the The higher pulse pressure seen in patients with treated overdippers, and the change was associated more hypotensive episodes was caused by lower with the lowest BP values. Thus it seems that drug- DBP, and may indicate stiffer vessel walls and more induced excessive BP lowering leads to hypoperfu- widespread changes in cerebral resistance vessels.29 sion in the presence of significant coronary artery According to our results the question of critically stenoses. Moreover, in hypertensive patients also low BP during antihypertensive medication deserves coronary autoregulation in the subendocardium is attention especially in the elderly patients. We pro- shifted toward higher pressure, probably owing to pose that ABPM should, at least in some patients, be

Journal of Human Hypertension Nocturnal hypotensive episodes I Kantola et al 878 included in the decision making of starting antihy- 14 Pedersen EV, Bobkiewicz-Kozlowska T, Waldemar G, pertensive medication and in the follow-up of antihy- Barry DI. The renin angiotensin of cerebral arteries pertensive treatment in the elderly, and that auto- contributes to cerebrovascular resistance. Eur J Clin matic censoring of data should be abolished. Invest 1987; 17: A38. 15 Rajagopalan B, Raine AEG, Cooper R, Ledingham JGG. Changes in cerebral blood flow in patients with severe Acknowledgements congestive cardiac failure before and after captopril This study was supported by Novartis Ltd, Basel, treatment. Am J Med 1984; 76:86–90. Switzerland and Novartis Finland, Espoo, Finland. 16 Kincaid-Smith P, McMichael J, Murphy EA. The clinical course and pathology of hypertension with papil- loedema (malignant hypertension). Quart J Med New References Series 1958; 27: 117–153. 17 Barry DI. Cerebrovascular aspects of antihypertensive 1 Strandgaard S. Autoregulation of cerebral blood flow in hypertensive patients. The modifying influence of treatment. Am J Cardiol 1989; 63: 14C–18C. prolonged antihypertensive treatment on the tolerance 18 Polese A et al. Upward shift of the lower range of cor- to acute, drug-induced hypotension. Circulation 1976; onary autoregulation in hypertensive patients with 53: 720–727. hypertrophy of the left ventricle. Circulation 1991; 83: 2 Kario K et al. Nocturnal fall of blood pressure and 845–853. silent cerebrovascular damage in elderly hypertensive 19 Flora JS. Antihypertensive treatment, myocardial patients. Hypertension 1996; 27: 130–135. infarction, and nocturnal myocardial ischemia. Lancet 3 Harrison DG et al. The effect of hypertension and left 1989; 2: 251–254. ventricular hypertrophy on the lower range of coron- 20 Kuriyama Y, Nakamura M, Kyougoku I, Sawada T. ary autoregulation. Circulation 1988; 77: 1108–1115. Effects of carvedilol on cerebral blood flow and its 4 Mansour et al. Low blood pressure levels and signs of autoregulation in previous stroke patients. Eur J Clin myocardial ischaemia: importance of left ventricular Pharmacol 1990; 38 (Suppl 2): 120–121. hypertrophy. J Hum Hypertens 1993; 7:13–18. 21 Graettinger W, Lipson J, Cheung D, Weber M. Vali- 5 Bostrom PA et al. Blood pressure and myocardial per- dation of portable blood pressure monitoring devices: fusion in hypertensive patients with and without left comparison with intra-arterial and sphygmoman- ventricular hypertrophy. J Hum Hypertens 1995; 9: ometer measurements. Am Heart J 1988; 116: 1155– 969–974. 1160. 6 Pierdomenico SD et al. Circadian blood pressure 22 Berardi L et al. Ambulatory blood pressure monitor- changes and myocardial ischemia in hypertensive ing: a critical review of the current methods to handle patients with coronary artery disease. J Am Coll Car- outliers. J Hypertens 1992; 10: 1243–1248. diol 1998; 31: 1627–1634. 23 White W, Lund-Johanssen P, Omvik P. Assessment of 7 Shuaib A. Alteration of blood pressure regulation and four ambulatory blood pressure monitors and cerebrovascular disorders in the elderly. Cerebrovasc measurements by clinicians versus intra-arterial blood Brain Metab Rev 1992; 4: 329–345. pressure at rest and during exercise. Am J Cardiol 8 Gosse P, Promax H, Durandet P, Clementy J. White 1990; 65:60–66. coat hypertension. No harm for the heart. Hyperten- 24 Hansson L et al, for the HOT Study Group. Effects of sion 1993; 22: 766–770. intensive blood-pressure lowering and low-dose 9 Taylor R et al. Accutracker II (version 30/23) ambulat- aspirin in patients with hypertension: principal results ory blood pressure monitor: clinical validation using of the Hypertension Optimal Treatment (HOT) ran- the British Hypertension Society and Association for domised trial. Lancet 1998; 351: 1755–1762. the Advancement of Medical Instrumentation stan- 25 MacMahon S, Rodgers A. The effects of blood pressure dards. J Hypertens 1993; 11: 1275–1282. reduction in older patients: an overview of five ran- 10 White W. Analysis of ambulatory blood pressure data domized controlled trials in elderly hypertensives. in antihypertensive drug trials. J Hypertens 1991; 9 Clin Exp Hypertens 1993; 15: 967–978. (Suppl 1): S27–S32. 26 Hayreh SS, Zimmerman MB, Podhajsky P, Alward WL. 11 Edvinsson L, MacKenzie E, McCulloch J. Disturbed Nocturnal arterial hypotension and its role in optic cerebral autoregulation. In: Edvinsson L, MacKenzie E, nerve head and ocular ischemic disorders. Am J McCulloch J (eds). Cerebral Blood Flow and Metab- Opthalmol 1994; 117: 603–624. olism. Raven Press: New York, 1993, pp 599–609. 27 Lassen NA. Cerebral blood flow and oxygen consump- 12 Folkow B. The haemodynamic consequences of adapt- tion in man. Phys Rev 1959; 39: 138–238. ive structural changes in the resistance vessels in 28 Wasserman AJ, Patterson JL. The cerebral vascular hypertension. Clin Sci 1971; 41:1–12. response to reduction in arterial cardon dioxide ten- 13 Barry DI et al. Effects of captopril on cerebral blood sion. J Clin Invest 1961; 40: 1297–1303. flow in normotensive and hypertensive rats. Am J Med 29 Strandgaard S, Paulson OB. Cerebrovascular conse- 1984; 76:79–85. quences of hypertension. Lancet 1994; 344: 519–521.

Journal of Human Hypertension