THE JOURNAL OF CLINICAL HYPERTENSION VOL. IV NO. II MARCH/APRIL 2002108

The elderly, those 65 years of age and older, will ac-count for 20% of the population by the year 2040.Hypertension affects more than one half of the elderlyand its prevalence continues to increase with age. Thepresence of hypertension confers an increased risk ofstroke, congestive heart failure, coronary heart disease,end-stage renal disease, and death. Although both dias-tolic and systolic blood pressure elevations are inde-pendently associated with increased cardiovascular riskin the younger individual, as vascular compliance be-comes reduced at age 60, an increasing systolic bloodpressure and lower diastolic blood pressure (or widerpulse pressure) increase cardiovascular risk in the el-derly. Isolated systolic hypertension is the most com-mon form of hypertension seen in the elderly. Lifestylemodification, including weight loss and salt restriction,reduces blood pressure, and may decrease the need forpharmacologic therapy. When the systolic blood pres-sure is 160 mm Hg or more and the diastolic bloodpressure is <90 mm Hg, the initial use of a diuretic-based or calcium channel blocker-based regimen im-proves outcome. Alpha blocker therapy should not beused as initial monotherapy. Most elderly hyperten-sives will require two to three drugs to achieve the pre-sent blood pressure goal of <140 mm Hg. Cautionshould be exercised when lowering diastolic pressureto <55 mm Hg in those with isolated systolic hyperten-sion. Although side effects of therapy are no more like-ly to occur in the elderly than in the younger individual

with hypertension, blood pressure reduction should beaccomplished gradually. (2002;4:108–112, 119) ©2002

Le Jacq Communications, Inc.

The elderly, defined as individuals 65 years of ageand older, represent the most rapidly growing seg-

ment of the population. In 1990, they accounted for13% of the US population and are expected to ac-count for 20% of the population by the year 2040.1The percentage of “old elderly” (i.e., those over age85) is also growing, and is projected to reach 16 mil-lion by the middle of the 21st century. Hypertension,defined as a systolic blood pressure (SBP) of 140 mmHg or higher and a diastolic blood pressure (DBP) of90 mm Hg or higher, affects approximately 50 millionadult Americans, 75% of whom are not adequatelycontrolled (i.e., blood pressure (BP) of <140/90 mmHg). Control rates are even worse in those at highestrisk, including the elderly, in whom only one in five(20%) has BP controlled. Hypertension affects morethan one half of those aged 65 and older, and itsprevalence continues to increase with age.2 High BP inthe elderly confers a three- to four-fold increase in riskfor cardiovascular disease, compared to younger indi-viduals. It is a significant risk factor for stroke, conges-tive heart failure, coronary heart disease, end-stagerenal disease, and death.1,2

ISOLATED SYSTOLIC HYPERTENSIONIsolated systolic hypertension (ISH) is defined as a SBPof ≥140 mm Hg and a DBP of <90 mm Hg. It repre-sents the most common form of hypertension in the el-derly and its prevalence increases with age; two thirdsof individuals 60 years of age and older, and threefourths of those over 75 years of age, have ISH.3 More

R e v i e w s

Hypertension in the Elderly: A Review of the Importance of Systolic Blood Pressure Elevation

Jan Basile, MD

From the Ralph H. Johnson VA Medical Center, Med-ical University of South Carolina, Charleston, SCAddress for correspondence: Jan Basile, MD, Ralph H. Johnson VA Medical Center,1090 Bee Street, Charleston, SC 29403Manuscript received May 9, 2001;accepted October 2, 2001

VOL. IV NO. II MARCH/APRIL 2002 THE JOURNAL OF CLINICAL HYPERTENSION 109

than 25% of adults 60 years of age or older have stage1 ISH (SBP of 140–159 mm Hg with DBP of <90 mmHg). It is the predominant form of hypertension in theelderly, and often goes untreated.

Age-related physiologic changes explain the fre-quent development of ISH. While DBP elevation iscaused by constriction of the smaller arterioles, ISH iscaused by the loss of distensibility of the larger arter-ies, especially the aorta.4 In younger patients, the aortais highly distensible, expanding during systole to mini-mize the rise in BP. The majority of elderly individu-als, however, develop progressive stiffening of theirarterial tree with age, reducing the compliance of theaorta during systole, which leads to a progressive ele-vation in SBP. Because the smaller arterioles are notinvolved in this process, the DBP remains normal ortends to decrease, contributing to a higher pulse pres-sure (SBP-DBP) with age.4 Accordingly, the elevatedsystolic pressure increases both left ventricular workand the risk for left ventricular hypertrophy, while thedecreased DBP may compromise coronary bloodflow, upon which it is dependent.5

SBP is not only easier to determine than DBP inthe elderly, but also it allows more appropriate riskstratification. In a recent analysis of the FraminghamHeart Study,6 knowing only the SBP correctly classi-fied the stage of BP in 99% of adults over the age of60, while knowing only the DBP allowed 66% to becorrectly classified.

In younger populations, both DBP and SBP are in-dependently associated with the cardiovascular eventrisk. At age 60, however, as vascular compliance is re-duced, an increasing SBP and a lower DBP increasescardiovascular risk. Accordingly, the pulse pressure isa stronger predictor of cardiovascular risk than SBP orDBP.7 As there is a lack of trial-based evidence utiliz-

ing pulse pressure narrowing as a means of lesseningrisk, SBP will continue to be targeted for evidence-based event-rate reduction.

APPROPRIATE GOALS OF THERAPY IN THE ELDERLYTreatment Benefits. The optimal BP level in the elder-ly has not yet been conclusively defined.1 In general,BP goals depend on the type of elevation and presenceof concomitant conditions. Several large, prospectiveclinical trials conducted several decades ago, focusingon DBP as well as SBP elevation, demonstrated the benefits of treating hypertension in the elderly (Table I). Based on diastolic entry criteria, theyshowed significant decreases in morbidity and mortal-ity when treating to a DBP goal of less than 90 mmHg. Treatment benefits are more marked in olderthan younger individuals, due to their greater absoluterisk for cardiovascular disease.8 Even though cardio-vascular events have been favorably decreased bytreatment of hypertension in the elderly, the demon-strated benefit has been largest for the prevention ofstroke and stroke-related mortality.1

Several recent randomized, placebo-controlled trialsdemonstrated significant benefit from drug treatment inelderly patients with ISH (Table II). In those with anSBP of ≥160 mm Hg and a DBP of <90–95 mm Hg,they found a 35%–40% reduction in stroke, up to a50% reduction in heart failure, a 30% reduction incoronary events, and a 10%–15% reduction in mortali-ty. In order to achieve this benefit, SBP was reduced byat least 20 mm Hg from baseline, to a level below either150 or 160 mm Hg. In none of the trials was an aver-age SBP of <140 mm Hg achieved (Table III).

Although the vascular risk of stage1 ISH (140–159mm Hg) is well established, no outcome-based trial

Table I. Percentage of Event Reduction in Clinical Hypertension Trials in Older Patients: Average12/5 mm Hg Systolic/Diastolic Reduction

STROKE CAD CHF ALL CVD

Systolic/diastolicAustralian 33 18 -- 31EWPHE 36 20 22 29*STOP 47* 13† 51* 40*MRC 25* 19 -- 17*HDFP 44* 15* -- 16*

Isolated systolicSHEP 33* 27* 55* 32*Syst-Eur 42* 30 29 31*Syst-China 38* 27 -- 25*

CAD=coronary artery disease; CHF=congestive heart failure; CVD=cardiovascular disease; EWPHE=European Work-ing Party on High Blood Pressure in the Elderly; STOP=Swedish Trial in Old Patients with Hypertension; MRC=Med-ical Research Council; HDFP=Hypertension Detection and Follow-up Program Cooperative Group; SHEP=SystolicHypertension in the Elderly Program; Syst-Eur=Systolic Hypertension-Europe trial; Syst-China=Systolic Hypertension-China trial; *statistically significant; † myocardial infarction only

THE JOURNAL OF CLINICAL HYPERTENSION VOL. IV NO. II MARCH/APRIL 2002110

has been completed to test whether treatment reducesclinical event rates. A large trial is currently ongoing toevaluate this question. Nevertheless, the sixth reportof the Joint National Committee on Prevention, De-tection, Evaluation, and Treatment of High BloodPressure (JNC VI)9 and a recent consensus statement3based on epidemiologic and not trial-based data rec-ommend achieving an SBP of <140 mm Hg.

Clinical trials in those with hypertension, includingthe elderly, tend to underestimate the actual benefit oftreatment for several reasons. First, patients with se-vere hypertension, who stand to benefit the most fromtreatment because of their greater absolute risk, areoften excluded from the studies. In addition, manystudy protocols follow an “intention to treat” design,where patients in the placebo group are treated oncetheir BP reaches a designated threshold, thus minimiz-ing the ability to accurately measure treatment benefitsin the active therapy group. In the Systolic Hyperten-sion in the Elderly Program (SHEP), for example, bythe end of the study, 44% of patients in the placebogroup were receiving active treatment. Finally, a shorttrial duration may not allow cardiovascular end pointdifferences to be realized, since cerebrovascular endpoint benefits may occur sooner than coronary heartdisease event reductions.1 Accordingly, the actual re-sults achieved in practice may be greater than thoseshown in the clinical trials.

Lifestyle (Nonpharmacologic) and PharmacologicGoals and Options. Lifestyle changes—in particular,reduced sodium intake and weight loss—are beneficialin controlling BP and are associated with less need forpharmacologic therapy in the elderly hypertensive.The Trial of Nonpharmacologic Interventions in theElderly (TONE),11 showed that restricting salt to 80

mmol per day reduced SBP by 4.3 mm Hg and DBPby 2 mm Hg. Further, the combination of weight lossand salt restriction reduced BP more than either strate-gy by itself. In addition, it was found that weight lossand salt restriction may decrease the need for antihy-pertensive therapy in the elderly.

The majority of elderly patients with hypertensionwill require at least two to three drugs to achieve theSBP goal of <140 mm Hg recommended in JNC VI.In both the SHEP10 and the Systolic Hypertension-Eu-rope (Syst-Eur) trial,12 40%–50% of participants re-quired at least two or more drugs to achieve a finalSBP that was actually >140 mm Hg, while DBP was<80 mm Hg (Table III). As stated, lowering of DBP togoal is generally easier than lowering SBP to goal. Ifone focuses on achieving the SBP goal, the DBP goalwill almost always be reached. For this reason, in theelderly, we need to focus more on SBP to achieve ef-fective BP reduction.

The usual dose of the initial agent is often onehalf that used in the non-elderly. This allows for thealtered pharmacokinetic renal or hepatic metabo-lism that often occurs in the elderly. The dose

Table II. Major Clinical Trials Showing Benefit of Treating Isolated Systolic Hypertension

SHEP (N=4736) SYST-EUR (N=4695) SYST-CHINA (N=2394)

Baseline BP 160–219/<90 160–219/<95 160–219/<95SBP/DBP (mm Hg)BP reduction: 27/9 23/7 20/5SBP/DPB (mm Hg)Drug therapy Chlorthalidone Nitrendipine Nitrendipine

Atenolol Enalapril CaptoprilHCTZ HCTZ

Outcomes (%↓)Stroke 33 42 38CAD 27 30 27CHF 55 29 --All CVD 32 31 25

BP=blood pressure; SBP/DBP=systolic/diastolic BP; CAD=coronary artery disease; CHF=congestive heart failure;CVD=cardiovascular disease; HCTZ=hydrochlorothiazide; SHEP=Systolic Hypertension in the Elderly Program; Syst-Eur=Systolic Hypertension-Europe trial; Syst-China=Systolic Hypertension-China trial

Table III. Blood Pressure Reductions in SHEP andSyst-Eur (mm Hg)

SHEP SYST-EUR

Entry 160–219/<90 160–219/<95Goal (SBP) <160 + ≥21 ↓ <150 + ≥20 ↓Baseline 170/77 174/86Achieved: Rx 143/68 151/79Achieved: Placebo 155/72 161/84

SHEP=Systolic Hypertension in the Elderly Program;Syst-Eur=Systolic Hypertension-Europe Trial; SBP=sys-tolic blood pressure

VOL. IV NO. II MARCH/APRIL 2002 THE JOURNAL OF CLINICAL HYPERTENSION 111

should be up-titrated slowly until the maximum BPreduction occurs at the dose with the fewest side ef-fects. Agents are added until the BP goal is attained.

The initial (preferred) drug or drug class for use inelderly patients with ISH has been addressed in severaloutcome-based trials. The SHEP trial,10 performed in4736 subjects at least 60 years of age with an SBP of≥160 mm Hg and a DBP of <90 mm Hg, showed a fa-vorable reduction in stroke and cardiovascular eventsusing a diuretic-based strategy with or without a βblocking agent (Table II). More recently, the Syst-Eurtrial,12 including 4695 patients older than 60 years ofage over a median follow-up of 2 years, has shown abenefit when using a long-acting dihydropyridine cal-cium channel blocker (CCB) as initial therapy. Ameta-analysis of eight placebo-controlled trials in15,693 patients 60 years of age and older and fol-lowed for an average of 3.8 years found that activetreatment reduced coronary events by 23%, stroke by30%, cardiovascular death by 18%, and total deathby 13%.13 In those patients older than 70 years of age,the absolute benefit was particularly high. Treating 19patients for 5 years prevented one major fatal or non-fatal cardiovascular event. Although most of these tri-als were performed in patients over 60 years of age, arecent meta-analysis14 supported the benefit of antihy-pertensive therapy even in patients over 80 years ofage, as the oldest of the elderly seem to benefit themost from active treatment.

A recent retrospective analysis of SHEP,15 whichproved the benefit of initial diuretic-based therapy,found that 7% of the participants in the active treat-ment group on chlorthalidone had developed hy-pokalemia (K

+of <3.5 mEq/L) by the 1-year visit. The

individuals who developed hypokalemia had an eventrate similar to that of the placebo group, while thosewith levels >3.5 mEq/L had significantly fewer cardio-vascular events, suggesting that K

+should be kept to a

level of ≥3.5 mEq/L. If not used initially, a thiazide di-uretic, based on outcome-based trials, should be in-cluded in most regimens to enhance the efficacy ofother BP-lowering agents as well as to reduce the riskof developing ischemic stroke.16

Although they have been used as single-agenttherapy in no more than 60% of the patientstreated, diuretics and CCBs are the only drugclasses that have so far been tested as initial thera-py in outcome-based trials in elderly patients withISH. If a diuretic is used, potassium levels shouldbe kept as close to normal as possible.15

The Playing Field Appears Level. Although all classesof antihypertensive agents effectively lower combinedSBP/DBP elevation in the elderly, the majority of out-come-based trials showing a reduction in vascular

morbidity and mortality have used diuretics and,when necessary, additive therapy with β blockade.Since JNC VI in 1997, several therapeutic trials in el-derly hypertensives with combined SBP and DBP ele-vation have suggested that the initial agent chosen fortreatment may not have a unique status for event ratereduction and that the BP level achieved appearsmore important.

The open-label Second Swedish Trial in Old Patientswith Hypertension (STOP-2)17 compared the use of an-giotensin-converting enzyme (ACE) inhibitors andCCBs with diuretic and/or β blocker therapy. It includ-ed 6628 elderly hypertensives 70–84 years of age. Withsimilar BP reduction, there was no difference in cardio-vascular mortality (the primary outcome) among thethree randomized groups. The ongoing Antihyperten-sive and Lipid-Lowering Treatment to Prevent HeartAttack Trial (ALLHAT)18 enrolled hypertensives atleast 55 years of age (mean age, 67). It stopped the αblocker (doxazosin) arm early because of a 25% greatercardiovascular event rate as well as a two-fold greaterrisk of heart failure when compared to the diureticchlorthalidone.18 This suggests that although doxazosintherapy can be used as additive therapy, it should not beused as initial monotherapy in the treatment of the el-derly hypertensive.

The open-label Nordic Diltiazem (NORDIL)trial,19 enrolling patients with an average age of 60years, found a similar rate for the primary outcome ofcombined fatal and nonfatal stroke, myocardial infarc-tion, and cardiovascular disease death in comparingthe CCB diltiazem with diuretic and/or β blocker ther-apy. Although fewer strokes occurred in the diltiazemarm, there was a trend toward more myocardial in-farctions and congestive failure with diltiazem. Thedouble-blinded International Nifedipine GITS (IN-SIGHT) trial20 enrolled men and women 55–80 yearsof age, 75% of whom were above the age of 60. Itfound that nifedipine GITS (gastrointestinal treatmentsystem) and diuretic therapy resulted in similar overallcardiovascular disease outcomes.

Presently, the degree of BP reduction—not the indi-vidual drug used for lowering BP—appears more im-portant for improving outcome, even in the olderindividual.21 The still-to-be-completed ALLHAT trial,comparing the ACE inhibitor lisinopril, the CCB am-lodipine, and the diuretic chlorthalidone, will help toclarify, in the 42,448 high-risk hypertensives, whethernewer-generation agents are superior, similar, or infe-rior to the diuretic chlorthalidone in older individualswith hypertension.22

J-Curve Phenomenon. The J-curve hypothesis describesthe observational concern that lowering DBP below acertain critical value increases the risk of cardiovascular

THE JOURNAL OF CLINICAL HYPERTENSION VOL. IV NO. II MARCH/APRIL 2002112

continued on page 119

death in elderly hypertensives. Prospective data validat-ing this hypothesis are lacking. It is important to re-member that the bulk of the evidence available is fromretrospective trials, which are associated with inherentobservational bias. As this phenomenon has beenprospectively observed in both placebo and activelytreated patients, a low DBP is thought to serve more asa marker than as the cause of events in those with un-derlying coronary disease.

The Hypertension Optimal Treatment (HOT)study23 was designed to test this question. Thisprospective, randomized, open with blinded endpoint evaluation (PROBE) trial randomly assigned18,790 hypertensive patients (mean age, 61.5years) in 26 countries to a target DBP of 80 mmHg, 85 mm Hg, or 90 mm Hg and followed themfor an average of 3.8 years. It found no increasedrisk for the DBP goal of <80 mm Hg vs. <90 mmHg.

A retrospective analysis of the SHEP trial,24

however, suggested that in the few patients whoseDBP was lowered to <55 mm Hg, there was nobenefit in outcome, compared to the placebogroup. Although this may reflect a higher risk inthose with the widest pulse pressure, it suggestswe exercise caution in lowering DBP to <55 mmHg when treating older individuals with ISH.

General Guidelines for Antihypertensive Treatment. • Hypertension therapy in older individuals

should begin with lifestyle modification. • Weight loss and sodium restriction may de-

crease the need for antihypertensive medica-tion in this population.

• The starting dose of medication should beone half of that used in younger patients.

• In the uncomplicated elderly patient with hyper-tension, a diuretic with or without a β blocker isa reasonable approach, based on outcome data.

• In the elderly patient with ISH, a diuretic ispreferred as initial therapy; however, a long-acting dihydropyridine CCB may be used asalternative therapy.

• ACE inhibitor therapy continues to be recom-mended in the elderly hypertensive with diabetesor systolic heart failure, and after myocardial infarction.

• Although not specifically evaluated in elderlyhypertension trials, an angiotensin-receptorblocker may be recommended when a coughor other adverse effect precludes the use of anACE inhibitor. There is, at present, no evi-dence to support the use of an ACE inhibitorand angiotensin receptor blocker together.

• Alpha blocker therapy should not be used as

initial monotherapy in the elderly hyperten-sive. It can be used as additive therapy to fur-ther reduce BP.

• The treatment goal should be determined bythe underlying concomitant hypertensive dis-order. In systolic/diastolic hypertension, thegoal should remain <140/90 mm Hg, withlower goals in diabetics and subjects withrenal disease. Patients with ISH should have aminimal SBP goal of <140 mm Hg.

• With the risk of more side effects, smallerdoses of two tolerated agents are more desir-able than high-dose monotherapy.

• Although BP reduction should occur more gradu-ally in elderly patients, the goals of therapyshould be similar to those in younger patients.

This paper was presented at the National Heart, Lung, andBlood Institute Symposium on Hypertension; April 25, 2001;Atlanta, GA.

REFERENCES1 Ooi HL, Cileman PL, Duggan J, et al. Treatment of hyper-

tension in the elderly. Curr Opin Nephrol Hypertens.1997;7:504–509.

2 Freis E, Papademetriou V. Current drug treatment and treat-ment patterns with antihypertensive drugs. Drugs.1996;52(1):1–16.

3 Izzo J, Levy D, Black HR. Importance of systolic blood pres-sure in older Americans. Hypertension. 2000;35:1021–1024.

4 Tonkin A, Wing L. Management of isolated systolic hy-pertension. Drugs. 1996;51(5):738–749.

5 Madhaven S, Ooi HL, Cohen H, et al. Relation of pulse pres-sure and blood pressure reduction to the incidence of myocar-dial infarction. Hypertension. 1994;23:395–401.

6 Lloyd-Jones DM, Evans JC, Larson MG, et al. Differentialimpact of systolic and diastolic blood pressure level onJNC-VI staging. Hypertension. 1999;34:381–385.

7 Franklin SS, Khan SA, Wong ND, et al. Is pulse pressure usefulin predicting risk for coronary heart disease? The FraminghamHeart Study. Circulation. 1999;100:354–360.

8 Black HR. New concepts in hypertension: focus on the el-derly. Am Heart J. 1998;135:S2–S7.

9 Joint National Committee. The Sixth Report of the JointNational Committee on Prevention, Detection, Evaluation,and Treatment of High Blood Pressure. Arch Intern Med.1997;157:2413–2446.

10 SHEP Cooperative Research Group. Prevention of stroke byantihypertensive drug treatment in older persons with isolatedsystolic hypertension. JAMA. 1991;265(4):3255–3264.

11 Whelton PK, Appel LJ, Espeland MA, et al. Sodium restrictionand weight loss in the treatment of hypertension on older per-sons: a randomized controlled Trial of Non-Pharmacologic In-terventions in the Elderly (TONE). Tone CollaborativeResearch Group. JAMA. 1998;279:839–846.

12 Staessen JA, Fagard R, Thijs L, et al. Morbidity and mor-tality in the placebo controlled European trial on isolatedsystolic hypertension (Syst-Eur) in the elderly. Lancet.1997;350:757–764.

13 Staessen JA, Gasowski J, Wang JC, et al. Risk of untreated andtreated isolated systolic hypertension in the elderly: meta-analy-sis of outcome trials. Lancet. 2000;355:865–872.

14 Gueyffier F, Bulpitt C, Goissel JP, et al. Antihypertensivedrugs in very old people: a subgroup meta-analysis of ran-domized controlled trials. Lancet. 1999;353:793–796.

15 Franse LV, Pahor M, Di Bari M, et al. Hypokalemia associated

VOL. IV NO. II MARCH/APRIL 2002 THE JOURNAL OF CLINICAL HYPERTENSION 119

Basile (continued from page 112)

with diuretic use and cardiovascular events in the SystolicHypertension in the Elderly Program. Hypertension.2000;35:1025–1030.

16 Klungel O, Heckbent ST, Longstreth WT Jr, et al. Antihy-pertensive drug therapies and the risk of ischemic stroke. ArchIntern Med. 2001;161:37–43.

17 Hansson L, Lindholm LH, Ekbom T, et al., for the STOP-Hypertension-2 study group. Randomized trial of old andnew antihypertensive drugs in elderly patients: cardiovas-cular mortality and morbidity in the Swedish Trial in OldPatients with Hypertension-2 (STOP-2) study. Lancet.1999;354:1751–1756.

18 The ALLHAT Officers and Coordinators for the ALLHATCollaborative Research Group. Major cardiovascular events inhypertensive patients randomized to doxazosin vs. chlorthali-done: the Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT). JAMA.2000;283:1967–1975.

19 Hansson L, Hedner T, Lund-Johansen P, et al., for theNORDIL Study Group. Randomised trial of effects of calciumantagonists compared with diuretics and β-blockers on cardio-vascular morbidity and mortality in hypertension: the Nordic

Diltiazem (NORDIL) study. Lancet. 2000;356:359–365.20 Brown MJ, Palmer CR, Castaigne A, et al. Morbidity and mor-

tality in patients randomised to double-blind treatment with along-acting calcium-channel blocker or diuretic in the Interna-tional Nifedipine GITS study: Intervention as a Goal in Hyper-tension Treatment (INSIGHT). Lancet. 2000;356:366–372.

21 Staessen J, Wang JG, Thijs L, et al. Cardiovascular protec-tion and blood pressure reduction: a meta-analysis. Lancet.2001:358:1305–1315.

22 Grimm RH, Margolis KL, Papademetriou V, et al., for theALLHAT Collaborative Research Group. Baseline characteris-tics of participants in the Antihypertensive and Lipid LoweringTreatment to Prevent Heart Attack Trial (ALLHAT). Hyper-tension. 2001;37:19–27.

23 Hansson L, Zanchetti A, Carruthers SG, et al. Effects of in-tensive blood-pressure lowering and low-dose aspirin in pa-tients with hypertension: principal results of the HypertensionOptimal Treatment (HOT) randomized trial. Lancet.1998;351:1755–1762.

24 Somes G, Pahor M, Shorr RI, et al. The role of diastolicblood pressure when treating isolated systolic hypertension.Arch Intern Med. 1999;159:2004–2009.