Stroke is a major public health problem that contributes significantly to the global disease burden. The stroke incidence rate has increased by over 100% since the 1970s in low- to middle-income countries.1 Although early (21 days to one month) stroke case fatality has declined worldwide since the 1980s, including in low- to middle-income countries, stroke is still one of the most important causes of death in adults,1,2 and is the leading cause of death in China.3,4 Stroke is also a major cause of long-term disability1 and recognition impairment.5 Hypertension is the most important of the prevalent and modifiable risk factors for stroke. All subtypes of hypertension, including isolated diastolic/systolic hypertension and systolic and diastolic hypertension, increase the risk of stroke.6,7
The stroke risk among hypertensive patients seems to be higher in Asian populations. The Asia-Pacific Cohort Studies Collaboration has reported a much stronger relationship between high systolic blood pressure (SBP) level and stroke risk, especially haemorrhagic stroke, in Asia compared with Australia.8 There is also a continuous epidemiological relationship between BP levels and recurrent stroke risk.9 Controlling hypertension is known to be the most important factor in the prevention of stroke. In general, antihypertensives bring about a much greater reduction of stroke than of coronary heart disease (CHD).10 Lowering SBP by 5-6mmHg and/or diastolic BP (DBP) by 2-3mmHg significantly reduces the risk of stroke (relative risk reduction [RRR] 40%).11 Furthermore, controlling hypertension also significantly reduces the stroke mortality rate and the incidence of other vascular deaths.12,13
A recently published meta-analysis underlined the powerful benefit of controlling hypertension in the prevention of stroke in patients without a history of vascular disease (RRR 45%, 95% confidence interval [CI] 35-55%) and in patients with a history of stroke or transient ischaemic attack (TIA) (RRR 34%, 95% CI 21- 44%).10 Therefore, all guidelines worldwide recommend lowering BP in all patients with hypertension. However, there are still some uncertainties in the control of BP and the prevention of stroke, especially for patients with previous cerebrovascular events (secondary prevention of stroke); these uncertainties include the goal BP, who to treat and which class of BP-lowering drugs to use. This article aims to answer these questions by interpreting the guidelines and updated clinical data and then promoting standardised clinical practice for stroke prevention.
What Is the Goal Blood Pressure in Patients with Previous Cerebrovascular Events?
Owing to evidence from many cohort studies and randomised controlled trials (RCTs), it is well accepted that the goal BP for primary prevention of stroke is <140/90mmHg, aiming for normal levels of 120/80mmHg. For patients with congestive heart failure, myocardial infarction, diabetes or chronic renal failure, the goal BP is <130/80mmHg.14-18 However, for secondary stroke prevention, the goal BP remains controversial. In 2003, the hypertension guidelines of the Seventh Joint National Committee (JNC7)17 and the European Society of Hypertension (ESH)18 recommended a goal BP of <140/90mmHg; in 2007, the ESH guideline (ESH 2007) changed the goal BP to <130/80mmHg.15 However, the guidelines for ischaemic stroke/TIA from the American Heart Association (AHA)/ American Stroke Association (ASA) in 2006 (ASA 2006)19 and from the European Stroke Organisation (ESO) in 2008 (ESO 2008)14 declared that the absolute target BP level and reduction were uncertain and should be individualised in patients with previous cerebrovascular events.
This raises the question of which goal BP and which guideline to use in clinical practice. To resolve this problem and to better understand the differences among the guidelines, it is necessary to look at the evidence used to produce the guidelines. First, cohort (prospective observational) studies do not show a lower BP limit below which risk ceases to decline, namely 'the lower, the betterÔÇÖ; this is also true of randomised trials for primary stroke prevention across a wide range of BPs.10 There have been four double-blind, placebo-controlled RCTs that have focused specifically on secondary stroke prevention and BP-lowering drugs: the Post-stroke Antihypertensive Treatment Study (PATS),20 the Perindopril Protection Against Recurrent Stroke Study (PROGRESS),21 Acute Candesartan Cilexetil Therapy in Stroke Survivors (ACCESS)22 and Prevention Regimen for Effectively Avoiding Second Strokes (PRoFESS).23 Briefly, the first three trials showed a clinical benefit of BP-lowering drugs, but PRoFESS did not. The baseline and achieved SBP in the four trials were 149 and 143mmHg, 141 and 132mmHg, 150 and 150mmHg and 140 and 136mmHg, respectively.
Although the post hoc analysis of PROGRESS in 2006 showed a progressive reduction in the incidence of stroke recurrence (particularly haemorrhagic stroke) down to an achieved SBP of about 120mmHg,24 there is no direct evidence from these four RCTs for a goal BP of <130/80mmHg for secondary stroke prevention. A similar phenomenon has also been found in CHD trials.25 Based on these RCT data, in 2009 ESH published a reappraisal of European guidelines on hypertension management (Reappraisal ESH 2009), which concluded that "it may be prudent to recommend lowering SBP/DBP to values within the range 130-139/80-85mmHg, and possibly close to lower values in this range, in all hypertensive patients".25 The AHA/ASA 2010 guideline for the management of spontaneous intracerebral haemorrhage (ICH) also recommends a goal BP of <140/90mmHg (or <130/80mmHg in the presence of diabetes or chronic kidney disease), although specific data on the optimal BP for reducing ICH recurrence are not available.26 Second, a J-curve phenomenon may occur in certain patients with ischaemic stroke/TIA. Rothwell et al.27 analysed data from 8,328 cases from the European Carotid Surgery Trial (ECST), the North American Symptomatic Carotid Endarterectomy Trial (NASCET) and the UK-TIA Aspirin Trial, and the findings indicate a J-curve phenomenon in patients with bilateral carotid artery stenosis 70%. In this report, the relative risk (RR) of stroke recurrence was 1.52, 0.98 and 0.50 when SBP was <140, 140-160 and 160mmHg, respectively, in patients with bilateral carotid artery stenosis 70%. On the other hand, post hoc analysis of the Warfarin-Aspirin Symptomatic Intracranial Disease (WASID) study28 showed that in patients with symptomatic intracranial stenosis (50-99%), there was a progressive reduction of stroke recurrence down to an achieved SBP of <120/70mmHg. However, WASID enrolled only 527 cases, and the data seem to show no obvious difference among groups at a BP <140/90mmHg.
Based on the controversial results and insufficient evidence from RCTs, the ASA 2006 and ESO 2008 guidelines stressed that "an absolute target BP level and reduction are uncertain and should be individualized" and that "BP should not be lowered intensively in patients with suspected haemodynamic stroke or in those with bilateral carotid stenosis". In addition, the ESH reappraisal in 200925 concluded that "a J-curve phenomenon is unlikely to occur until lower values are reached, except perhaps in patients with advanced atherosclerotic artery diseases". So, before the determination of goal BP in patients with previous ischemic stroke/TIA, careful neurovascular imaging examinations should be carried out.
Should Blood-pressure-lowering Treatment Be Applied to Patients with Normal Blood Pressure for Secondary Prevention?
The ESO 2008 guidelines recommend BP-lowering treatment after the acute phase of ischaemic stroke, including in patients with normal BP. The ESH 2007 guidelines define patients with established cerebrovascular events as a very-high-risk population when their BP is in the range of |120-140/80-90mmHg, and recommend immediate drug treatment in such patients. Unfortunately, in spite of these strong recommendations, clinicians are still confused and hesitate to use antihypertensive drugs in these patients with normal BP.
To better understand the recommendations and to standardise clinical practice, it is again essential to look at the main evidence: PATS, PROGRESS, ACCESS and PRoFESS.20-23 The inclusion criterion of ACCESS was ischaemic stroke plus hypertension, while in the other three trials the inclusion criterion was ischaemic stroke/TIA with or without hypertension. As PRoFESS showed no benefit of lowering BP and all the patients included in ACCESS had hypertension, these two trials cannot provide any information about this issue. PATS used a BP of >140/90mmHg as the definition of hypertension, and only 16% of the 5,663 cases had normal BP. The authors only reported the benefit for the whole group, so no subanalysis of the 16% of patients with normal BP is available. PROGRESS subgroup analysis showed a significant reduction of stroke recurrence (RRR 24%, 95% CI 9-37%) in 'non-hypertensive' patients. In spite of this, PROGRESS defined non-hypertensive as an SBP <160mmHg and/or a DBP <90mmHg; the mean BP at entry of this subgroup was 136/79mmHg. This is the strongest evidence behind the recommendation of the ESO 2008 and ESH 2007 guidelines.
However, we do not know exactly how many truly hypertensive patients (SBP 140-160mmHg) were enrolled in the non-hypertensive subgroup in PROGRESS. The statement in the 2009 EHS reappraisal that "Trial evidence concerning antihypertensive drug treatment in patients with previous cardiovascular events in absence of hypertension is controversial, and further trials must be completed before firm recommendations can be given" is much more helpful for current clinical practice.
What Kind of Blood-pressure-lowering Drugs Should Be Chosen for Stroke Prevention?
Five commonly used classes of BP-lowering drugs are available for clinical practice: diuretics, beta-blockers (BBs), calcium-channel blockers (CCBs), angiotensin-converting enzyme (ACE) inhibitors and angiotensin-receptor blockers (ARBs). Many RCTs over the last few decades have proved that the benefit of BP-lowering drugs in preventing stroke comes mainly from BP lowering itself. There are also claims of additional non-BP-lowering (pleiotropic) effects of these drugs, although these effects are relatively small.
A recently published meta-analysis that included 147 trials from 1966 to 2007, with a total of 958,000 cases, compared the efficacious of the five classes of BP-lowering drugs.10 Compared with placebo, diuretics, BBs, CCBs and ACE inhibitors all reduce the risk of stroke and other major vascular events. Comparing the different classes against each other, BBs perform worse (RR 1.18, 95% CI 1.03-1.36) and ARBs seem to be better (RR 0.90, 95% CI 0.71-1.13), while diuretics and ACE inhibitors are almost equal. CCBs are the only superior drugs to others (RR 0.91, 95% CI 0.84-0.98).
From the results of RCTs looking at secondary prevention, we can conclude that diuretics (PATS20), ACE inhibitors plus diuretics (PROGRESS21) have been proved to be more effective than placebo. However, ARBs remain controversial. PRoFESS found that telmisartan had no greater effects than placebo,23 with an average reduction of BP by 3.8/2.0mmHg, while the benefit of candesartan was studied in ACCESS,22 with no difference in BP level. Another head-to-head comparison trial for secondary stroke prevention, Morbidity and Mortality after Stroke, Eprosartan Compared with Nitrendipine for Secondary Prevention (MOSES), showed that eprosartan was better than nitrendipine, a CCB regimen.29
How should these conflicting clinical data be used to guide clinical practice? First, there is strong evidence that diuretics, or a combination of ACE inhibitors with diuretics, are effective for both primary and secondary prevention of stroke. Second, it can be concluded that ARBs are, at least currently, of dubious utility for stroke prevention.
In addition to the controversial results mentioned above, two other recently published trials, the Telmisartan Randomized Assessment Study in ACE-intolerant Subjects with Cardiovascular Disease (TRANSCEND)30 and Nateglinide and Valsartan in Impaired Glucose Tolerance Outcomes Research (NAVIGATOR),31 also showed no benefit of ARBs in reducing the rate of cardiovascular events. Third, CCBs are superior to other classes of BP-lowering drugs, and the exception of the MOSES results29 may be due to the compared drug, nitrendipine, a CCB with a short half-life, which might affect BP variability.
More recently, BP variability has become the hot topic in stroke prevention. In March 2010, The Lancet published three reports and a systematic review and meta-analysis.32-35 The main findings from the four publications are as follows: mean BP has minimal effect on stroke outcome and no effect on CHD outcome; various measures of visit-to-visit BP variability are powerful predictors of both stroke and CHD outcomes; other measures of variability (such as within-visit variability) also predict cardiovascular outcomes but to a lesser extent than visit-to-visit variability; amlodipine (a CCB with a long half-life) reduces variability compared with atenolol; and adjusting for blood variability completely explains the differences in stroke and CHD outcomes between amlodipine-based and atenolol-based treatment in the Anglo-Scandinavian Cardiac Outcomes Trial-Blood Pressure Lowering Arm (ASCOT-BPLA). These factors may give clinicians more confidence to choose a CCB with a long half-life for stroke prevention: in addition to the powerful effect of lowering BP and a better effect on reducing the progression of arthrosclerosis,36 reducing long-term BP variability is another mechanism. ASCOTBPLA and Avoiding Cardiovascular Events Through Combination Therapy in Patients Living with Systolic Hypertension (ACCOMPLISH)37 also provide strong evidence for using almodipine-based combination regimens for preventing stroke.