Treatment of hypertension in peripheral arterial disease(review)

Transcription

1 Cochrane Database of Systematic Reviews Treatment of hypertension in peripheral arterial disease (Review) LaneDA,LipGYH LaneDA,LipGYH. Treatment of hypertension in peripheral arterial disease. Cochrane Database of Systematic Reviews 2013, Issue 12. Art. No.: CD DOI: / CD pub3. Treatment of hypertension in peripheral arterial disease(review) Copyright 2013 The Cochrane Collaboration. Published by John Wiley& Sons, Ltd.

2 T A B L E O F C O N T E N T S HEADER ABSTRACT PLAIN LANGUAGE SUMMARY BACKGROUND OBJECTIVES METHODS RESULTS Figure Figure Figure Figure Figure Figure DISCUSSION AUTHORS CONCLUSIONS ACKNOWLEDGEMENTS REFERENCES CHARACTERISTICS OF STUDIES DATA AND ANALYSES Analysis 1.1. Comparison 1 ACE inhibitors versus placebo, Outcome 1 Cardiovascular events Analysis 1.2. Comparison 1 ACE inhibitors versus placebo, Outcome 2 Claudication distance Analysis 1.3. Comparison 1 ACE inhibitors versus placebo, Outcome 3 Maximum walking distance Analysis 1.4. Comparison 1 ACE inhibitors versus placebo, Outcome 4 Ankle brachial pressure index Analysis 2.1. Comparison 2 Calcium antagonists versus placebo, Outcome 1 Degree of diameter stenosis Analysis 2.2. Comparison 2 Calcium antagonists versus placebo, Outcome 2 Ankle brachial pressure index Analysis 2.3. Comparison 2 Calcium antagonists versus placebo, Outcome 3 Arterial intima-media thickness Analysis 3.1. Comparison 3 Thiazide diuretics versus alpha-adrenoreceptor blocking drugs, Outcome 1 Arterial intimamedia thickness (IMT) Analysis 5.1. Comparison 5 Beta-adrenoreceptor blockers versus thiazide diuretics, Outcome 1 Change in intermittent claudication distance Analysis 5.2. Comparison 5 Beta-adrenoreceptor blockers versus thiazide diuretics, Outcome 2 Absolute claudication distance Analysis 5.3. Comparison 5 Beta-adrenoreceptor blockers versus thiazide diuretics, Outcome 3 Change in ankle brachial pressure index Analysis 5.4. Comparison 5 Beta-adrenoreceptor blockers versus thiazide diuretics, Outcome 4 All-cause mortality.. 51 Analysis 5.5. Comparison 5 Beta-adrenoreceptor blockers versus thiazide diuretics, Outcome 5 Non-fatal cardiovascular events Analysis 6.1. Comparison 6 Beta-adrenoreceptor blockers versus beta-adrenoreceptor blockers, Outcome 1 All-cause mortality Analysis 6.2. Comparison 6 Beta-adrenoreceptor blockers versus beta-adrenoreceptor blockers, Outcome 2 Ankle brachial pressure index Analysis 6.3. Comparison 6 Beta-adrenoreceptor blockers versus beta-adrenoreceptor blockers, Outcome 3 Intermittent claudication distance Analysis 6.4. Comparison 6 Beta-adrenoreceptor blockers versus beta-adrenoreceptor blockers, Outcome 4 Absolute claudication distance Analysis 6.5. Comparison 6 Beta-adrenoreceptor blockers versus beta-adrenoreceptor blockers, Outcome 5 Need for revascularisation Analysis 7.1. Comparison 7 Calcium antagonist-based strategy versus beta-adrenoreceptor blocker-based strategy, Outcome 1 Composite endpoint of death, non-fatal MI, or non-fatal stroke Analysis 7.2. Comparison 7 Calcium antagonist-based strategy versus beta-adrenoreceptor blocker-based strategy, Outcome 2 Composite endpoint of death, non-fatal MI or stroke, and revascularisation i

3 APPENDICES WHAT S NEW HISTORY CONTRIBUTIONS OF AUTHORS DECLARATIONS OF INTEREST SOURCES OF SUPPORT DIFFERENCES BETWEEN PROTOCOL AND REVIEW INDEX TERMS ii

4 [Intervention Review] Treatment of hypertension in peripheral arterial disease Deirdre A Lane 1, Gregory YH Lip 1 1 University of Birmingham Centre for Cardiovascular Sciences, City Hospital, Birmingham, UK Contact address: Gregory YH Lip, University of Birmingham Centre for Cardiovascular Sciences, City Hospital, Dudley Road, Birmingham, B18 7QH, UK. g.y.h.lip@bham.ac.uk. Editorial group: Cochrane Vascular Group. Publication status and date: New search for studies and content updated (no change to conclusions), published in Issue 12, Citation: Lane DA, Lip GYH. Treatment of hypertension in peripheral arterial disease. Cochrane Database of Systematic Reviews 2013, Issue 12. Art. No.: CD DOI: / CD pub3. Background A B S T R A C T Peripheral arterial disease (PAD) causes considerable morbidity and mortality. Hypertension is a risk factor for PAD. Treatment for hypertension must be compatible with the symptoms of PAD. Controversy regarding the effects of beta-adrenoreceptor blockade for hypertension in patients with PAD has led many physicians to stop prescribing beta-adrenoreceptor blockers. Little is known about the effects of other classes of anti-hypertensive drugs in the presence of PAD. This is the second update of a Cochrane review first published in Objectives To determine the effects of anti-hypertensive drugs in patients with both raised blood pressure and symptomatic PAD in terms of the rate of cardiovascular events and death, symptoms of claudication and critical leg ischaemia, and progression of atherosclerotic PAD as measured by ankle brachial index (ABI) changes and the need for revascularisation (reconstructive surgery or angioplasty) or amputation. Search methods For this update the Cochrane Peripheral Vascular Diseases Group Trials Search Co-ordinator searched the Cochrane Peripheral Vascular Diseases Group Specialised Register (last searched March 2013) and CENTRAL (2013, Issue 2). Selection criteria Randomised controlled trials (RCTs) of at least one anti-hypertensive treatment against placebo or two anti-hypertensive medications against each other, with interventions lasting at least one month. Trials had to include patients with symptomatic PAD. Data collection and analysis Data were extracted by one author (DAL) and checked by the other (GYHL). Potentially eligible studies were excluded when the results presentation prevented adequate extraction of data and enquiries to authors did not yield raw data. Main results Eight RCTs were included with a total of 3610 PAD patients. Four studies compared a recognised class of anti-hypertensive treatment with placebo and four studies compared two anti-hypertensive treatments with each other. Studies were not pooled due to the variation of the comparisons and the outcomes presented. Overall the quality of the available evidence was unclear, primarily as a result of a lack of detail in the study reports on the randomisation and blinding procedures and incomplete outcome data. Two studies compared angiotensin converting enzyme (ACE) inhibitors against placebo. In one study there was a significant reduction in the number of 1

5 cardiovascular events in patients receiving ramipril (odds ratio (OR) 0.72, 95% confidence interval (CI) 0.58 to 0.91; n = 1725). In the second trial using perindopril (n = 52) there was a marginal increase in claudication distance but no change in ABI and a reduction in maximum walking distance. A trial comparing the calcium antagonist verapamil versus placebo in patients undergoing angioplasty (n = 96) suggested that verapamil reduced restenosis (per cent diameter stenosis (± SD) 48.0% ± 11.5 versus 69.6% ± 12.2; P < 0.01), although this was not reflected in the maintenance of a high ABI (0.76 ± 0.10 versus 0.72 ± 0.08 for verapamil versus placebo). Another study (n = 80) demonstrated no significant difference in arterial intima-media thickness (IMT) in men receiving the thiazide diuretic hydrochlorothiazide (HCTZ) compared to those receiving the alpha-adrenoreceptor blocker doxazosin (-0.12 ± 0.14 mm and ± 0.13 mm, respectively; P = 0.66). A study (n = 36) comparing telmisartan to placebo found a significant improvement in maximum walking distance at 12 months with telmisartan (median (interquartile range (IQR)) 191 m (157 to 226) versus 103 m (76 to 164); P < 0.001) but no differences in ABI (median (IQR) 0.60 (0.60 to 0.77) versus 0.52 (0.48 to 0.67)) or arterial IMT (median (IQR) 0.08 cm (0.07 to 0.09) versus 0.09 cm (0.08 to 0.10)). Two studies compared the beta-adrenoreceptor blocker nebivolol with either the thiazide diuretic HCTZ or with metoprolol. Both studies found no significant differences in intermittent or absolute claudication distance, ABI, or all-cause mortality between the anti-hypertensives. A subgroup analysis of PAD patients (n = 2699) in a study which compared a calcium antagonist-based strategy (verapamil slow release (SR) ± trandolapril) to a beta-adrenoreceptor blocker-based strategy (atenolol ± hydrochlorothiazide) found no significant differences in the composite endpoints of death, non-fatal myocardial infarction or nonfatal stroke with or without revascularisation (OR 0.90, 95% CI 0.76 to 1.07 and OR 0.96, 95% CI 0.82 to 1.13, respectively). Authors conclusions Evidence on the use of various anti-hypertensive drugs in people with PAD is poor so that it is unknown whether significant benefits or risks accrue. However, lack of data specifically examining outcomes in PAD patients should not detract from the overwhelming evidence on the benefit of treating hypertension and lowering blood pressure. P L A I N L A N G U A G E S U M M A R Y Treatment of high blood pressure for people with peripheral arterial disease When blood pressure is consistently high it can lead to complications such as a heart attack (myocardial infarction) or stroke. Both peripheral arterial disease (PAD), a condition that affects the blood vessels (arteries) carrying the blood to the legs, arms, and stomach area, and high blood pressure (hypertension) are associated with atherosclerosis. This is hardening of the arteries which is caused by deposits of fat, cholesterol and other substances inside the blood vessels. PAD is diagnosed when the blood supply to the legs is restricted causing pain and cramping that limits walking (intermittent claudication). It is measured by the walking distance (on a treadmill) before onset of pain (claudication distance) or ankle brachial index (ABI), the ratio of the blood pressure in the arms to the blood pressure in the legs. If the blood pressure is lower in the legs compared to the arms (ABI of less than 1.0) this indicates blocked arteries in the legs (or PAD). PAD can progress to pain at rest and critical limb ischaemia (sudden lack of blood flow to a limb caused by a blood clot or fatty deposit blockage) that requires revascularisation (restoring the blood flow by opening up the blocked blood vessel) or amputation. Treatment of hypertension to reduce cardiovascular events (heart attack or stroke) and death needs careful consideration in people with PAD. Anti-hypertensive medications may worsen the PAD symptoms by further reducing blood flow and supply of oxygen to the limbs, and may have long-term effects on disease progression. The evidence from randomised controlled trials (RCTs) examining the risks and benefits of various anti-hypertensive drugs on measures of PAD is lacking. We identified eight RCTs with a total of 3610 people with symptomatic PAD where participants were randomised to receive an antihypertensive treatment for at least one month or placebo or no treatment. Four studies compared an anti-hypertensive treatment with placebo and four studies compared two anti-hypertensive treatments with each other. The studies were not combined due to the variation of comparisons and the outcomes presented. One trial with 1725 participants showed that the angiotensin converting enzyme (ACE) inhibitor ramipril was effective in reducing the number of cardiovascular events by 28% compared to placebo. In one other study using an ACE inhibitor (n = 52) the perindopril group showed a small increase in claudication distance but no change in ABI and a reduction in maximal walking distance (MWD). In patients undergoing peripheral arterial angioplasty (a procedure to open narrowed or blocked blood vessels) the results from a trial with 96 participants suggested that the calcium channel blocker verapamil reduced restenosis (new blockage of the artery) at six months. In one small study (n = 80) peripheral arterial wall thickness was similar whether men received the thiazide diuretic hydrochlorothiazide (HCTZ) or the alpha-adrenoreceptor blocker doxazosin. In another small study (n = 36) MWD was improved at 12 months in the angiotensin-ii receptor antagonist telmisartan group compared to the placebo group but there were no significant differences in ABI or arterial wall thickness. Another study (n = 163) found no significant differences in intermittent or absolute claudication distance, ABI, all-cause mortality or non-fatal cardiovascular events 2

6 after 24 weeks of treatment in the beta-adrenoreceptor blocker nebivolol group and the HCTZ group. A study comparing two betaadrenoreceptor blockers, nebivolol and metoprolol, found no clear differences in intermittent or absolute claudication distance, ABI, all-cause mortality or revascularisation after 36 weeks of treatment. A subgroup analysis of PAD patients (n = 2699) in the final study revealed no significant differences in the combined endpoints of death, non-fatal myocardial infarction or non-fatal stroke with or without revascularisation between the calcium antagonist-based strategy (verapamil slow release (SR) with or without trandolapril) compared to the beta-adrenoreceptor blocker strategy (atenolol with or without HCTZ). The evidence on the use of various anti-hypertensive drugs in people with PAD is poor so that it is not known whether significant benefits or risks accrue. However, lack of data specifically examining outcomes in hypertensive PAD patients should not detract from the overwhelming evidence on the benefit of treating hypertension and lowering blood pressure. B A C K G R O U N D Description of the condition Peripheral arterial disease (PAD) is a condition that affects the blood vessels (arteries) carrying the blood to the legs, arms, and stomach area. PAD occurs when these arteries start to narrow because of a build-up of fatty deposits inside the blood vessels. PAD mainly affects the arteries in the legs. It can cause discomfort or pain in the lower legs when walking because the narrowing in the blood vessels means that less oxygen-rich blood reaches the legs. This discomfort or pain is called intermittent claudication. Not everyone has these symptoms. PAD is the third leading cause of morbidity from atherosclerotic disease (disease of the arteries caused by the build-up of fatty deposits on the inner lining of the blood vessels) after coronary heart disease and stroke (Fowkes 2013). PAD is the cause of a large number of hospital admissions each year. In the United States of America, 9.6% of cardiovascular events are due to PAD, resulting in 63,000 annual hospital admissions (Kannel 1996). In addition, PAD is associated with significant morbidity and mortality (Criqui 1992; Clement 2007; Fowkes 2013). Patients with PAD, with and without intermittent claudication, have almost three times the risk of dying or having a major cardiovascular event (stroke or heart attack) compared to people without PAD (Pande 2011; Fowkes 2013). Although the risk of non-fatal cardiovascular events (morbidity) among patients with intermittent claudication is reported to be approximately five to 10 in every 100 patients (Dormandy 2000) the risk of death is much higher. For example, the Framingham study showed that two out of every five patients with intermittent claudication died within 10 years of being diagnosed with PAD (Murabito 2005); this high adverse outcome is also reported in several other studies (Reunanen 1982; Bowlin 1997; Dormandy 2000; Feringa 2006) and the risk is similar whether the patient is symptomatic or not (Criqui 1992). Hypertension is the name given to high blood pressure that is consistently at such a level that it can cause disease or lead to complications (for example myocardial infarction (MI) (known as a heart attack) or stroke). Hypertension is a common and important risk factor for all vascular disorders including PAD (Kannel 1974; Fowkes 1992; Clement 2007). At presentation, between 2% and 5% of hypertensive patients have intermittent claudication, and this prevalence increases with age. Similarly, 35% to 55% of patients with PAD at presentation also have hypertension (Johnston 1988; Novo 1992; Binaghi 1994; Violi 1996; Cheng 1999; Hirsch 2001; Clement 2007; Singer 2008). Patients who suffer from either hypertension or PAD have a high risk of MI and stroke and when hypertension and PAD are both present the risk is greatly increased (Makin 2001; Clement 2007; Singer 2008). Hypertension contributes to the pathogenesis (progression of disease) of atherosclerosis (the basic pathological process underlying PAD) (Simon 1986; Bauwens 1989; McGill 1998). Indeed, both hypertension and PAD are associated with abnormal levels of lipid and coagulation factors in the blood (Makin 2002). Description of the intervention Given the co-existence of PAD and hypertension, the overlap in risk factors between the two conditions, the increased risk of nonfatal cardiovascular events in PAD, and the greater risk of death (Criqui 1992; Murabito 2005; Clement 2007; Fowkes 2013) treatment guidelines highlight the importance of reducing the patient s overall cardiovascular risk as well as targeting the individual risk factors (Clement 2007; ESH 2013). Treatment of hypertension in patients with symptoms of PAD is an obvious therapeutic target but this needs careful consideration because of the possibility that the anti-hypertensive medication will cause secondary effects and alter the process(es) causing PAD. While beta-adrenoreceptor blockers (which block the action of noradrenaline and adrenaline on the heart and thus help to reduce blood pressure) are recom- 3

7 mended for use in MI to increase survival (Hjalmarson 1997), there has been some concern that they may worsen the symptoms of intermittent claudication for people with PAD although several studies have shown that there is no evidence to support this (Bogaert 1983; Hiatt 1985; Radack 1991). In addition to betaadrenoreceptor blockers, there are other classes of anti-hypertensives that are commonly used to reduce blood pressure including diuretics, calcium channel blockers (CCBs), angiotensin converting enzyme (ACE) inhibitors, angiotensin receptor blockers (ARBs), and alpha-adrenoreceptor blockers. How the intervention might work The mechanism thought to cause the worsening of PAD symptoms with beta-adrenoreceptor blockers is peripheral vasoconstriction (narrowing of peripheral blood vessels) leading to further ischaemia in an already ischaemic limb (reducing blood flow to limbs that already have a poor oxygen supply). There is only scant evidence for this with studies showing either no change in walking distance when one beta-adrenoreceptor blocker is used (Solomon 1991) or a deterioration (Schweizer 1996), particularly when two beta-adrenoreceptor blockers are combined (Solomon 1991). Some beta-adrenoreceptor blockers (for example nebivolol or carvedilol) have intrinsic sympathomimetic activity (ISA) that mimics the effects of the sympathetic nervous system causing the blood vessels away from the heart (for example in the feet) to open up (vasodilation) thus allowing greater blood flow. This might be useful for patients with impaired blood flow to the lower extremities (the legs). If too much sodium is retained by the body, this can lead to fluid overload and increases the blood pressure. Diuretics work by increasing the flow of urine thereby removing sodium and chloride (and in turn water) from the body and can help to lower blood pressure. Control of blood pressure can help reduce the risk of cardiovascular events which are common in PAD patients. Dihydropyridine calcium channel blockers (for example, amlodipine, felodipine, nifedipine) cause vasodilation thereby helping to lower blood pressure. CCBs such as verapamil and diltiazem also vasodilate blood vessels but they can reduce cardiac output by lowering the heart rate and impair cardiac systolic function. As a result, the cardioselective CCBs should not be used in patients with concomitant heart failure. ACE inhibitors inhibit the conversion of the inactive form of the hormone angiotensin to the active form. The active form raises blood pressure, thus blocking its formation with ACE inhibitors helps to reduce blood pressure. There is some evidence that ACE inhibitors increase peripheral perfusion (blood flow) and claudication distance (walking distance before onset of pain in people who have this symptom, known as claudicants) in normotensive PAD patients (Breckenridge 1992; Ahimastos 2006) although concerns about exacerbating renal artery stenosis (blockage) arise (Makin 2001). These agents are also increasingly recognised for their cardioprotective effects (HOPE 1996). Angiotensin receptor blockers (ARBs) have very similar properties to ACE inhibitors. ARBs selectively block the receptor for angiotensin-ii to inhibit vasoconstriction (narrowing of the blood vessels), secretion of aldosterone, and sodium and water retention, thereby helping to lower blood pressure. Cardioselective beta-adrenoreceptor blockers, CCBs, ACE inhibitors and ARBs may potentially help to reduce leg pain by increasing the blood flow to the legs thereby increasing pain-free walking distance. In addition, these drugs have been shown to reduce blood pressure and the risk of cardiovascular events in hypertensive populations and in those with coronary heart disease (ABCD 2003; HOPE 2004; INVEST 2006; VALUE 2006) some of which have included patients with concomitant PAD. Long-acting alpha-adrenoreceptor blockers such as doxazosin also have vasodilating properties (increasing the blood flow) and are recommended as fourth line drugs, in addition to other anti-hypertensive drugs, in people with resistant hypertension (ESH 2013). Why it is important to do this review Current guidelines for treating hypertension recommend an individualised approach using anti-hypertensive agents (ESH 2013). There is scant evidence from randomised controlled trials of the benefit or harm of one class of anti-hypertensive medication over another in PAD patients and subsequently there are no specific national guidelines on the choice of blood pressure medication in patients with PAD. However, a consensus document for the management of PAD (TASC-II 2007) highlights that controlling a patient s total cardiovascular risk is paramount in such patients. Identification of individual cardiovascular risks, of which hypertension is prevalent, is the first step in the recommendations with subsequent optimal management of each risk factor. Addressing overall cardiovascular risk necessitates polypharmacy and a regimen that includes anti-platelet agents, statins and an ACE inhibitor (unless contraindicated) plus lifestyle modification (smoking cessation, reduction of salt intake, weight control and exercise) (Clement 2007; TASC-II 2007). Co-morbidities need to be considered when making decisions about treatment (Clement 2007; Singer 2008) as these may indicate that one anti-hypertensive is preferable to another in a particular patient, for example use of a beta-adrenoreceptor blocker or ACE inhibitor following an acute coronary syndrome or avoidance of the CCB verapamil or diltiazem in those with concomitant heart failure. Anti-hypertensive drugs have been used for many years in the treatment of hypertension although little is known of their effect in PAD. Therefore, before prescribing anti-hypertensive medication for people suffering with PAD, the short-term effects on symptoms and the longerterm effects on disease progression and cardiovascular outcomes should be considered. 4

8 O B J E C T I V E S To conduct a systematic review to determine the effects of antihypertensive drugs in patients with both raised blood pressure and symptomatic peripheral arterial disease (PAD) in terms of: rate of cardiovascular events and death; symptoms of claudication and critical leg ischaemia; progression of atherosclerotic PAD as measured by ankle brachial index changes and the need for revascularisation (reconstructive surgery or angioplasty) or amputation. M E T H O D S and include patients with symptomatic PAD. The intervention in each group had to last for at least one month. Types of outcome measures Primary outcomes 1. All-cause mortality 2. Cardiovascular deaths (stroke, myocardial infarction (MI), pulmonary embolism, peripheral arterial embolism) 3. Non-fatal cardiovascular events (stroke, MI, pulmonary embolism, peripheral arterial embolism) Criteria for considering studies for this review Types of studies Randomised controlled trials (RCTs) comparing at least one recognised class of anti-hypertensive treatment against placebo or two anti-hypertensive medications against each other and including patients with symptomatic PAD as either the study population or a subgroup. The intervention must have been for at least one month in each group. Studies were excluded when the trial design was adequate but the published data were inadequate for the purposes of the review, or if all attempts to contact the authors to obtain further data were unsuccessful. Cross-over trials were excluded as they would not be suitable for measures of disease severity and progression in view of the erroneous assumption that any effects are fully reversible after each treatment. Types of participants Patients with PAD and hypertension. For the purposes of this review PAD was restricted to the effects of PAD thus including those patients with intermittent claudication (IC) and critical limb ischaemia (CLI) (as defined by ischaemic rest pain) and excluding patients with aortic aneurysm, carotid disease, thrombophlebitis, and vasospasm. Hypertension was defined as elevated blood pressure (based on the current definition of hypertension at the time of recruitment) or was assumed in patients receiving anti-hypertensive medication. Types of interventions To be eligible for inclusion, trials had to include an intervention of at least one recognised class of anti-hypertensive treatment against placebo or two anti-hypertensive medications against each other Secondary outcomes 1. Measurements of disease severity, including claudication distance on a treadmill and ankle-brachial index (ABI) 2. Measurements of disease progression (e.g. ABI and arterial intima-media thickness (IMT)) 3. Need for revascularisation and other vascular interventions (e.g. bypass surgery, angioplasty, amputation) Search methods for identification of studies There were no language restrictions and relevant papers that were in languages other than English were translated. Electronic searches For this update the Cochrane Peripheral Vascular Diseases Group Trials Search Co-ordinator (TSC) searched the Cochrane Peripheral Vascular Diseases Group Specialised Register (last searched March 2013) and the Cochrane Central Register of Controlled Trials (CENTRAL) (2013, Issue 2), part of The Cochrane Library ( See Appendix 1 for details of the search strategy used to search CENTRAL. The Specialised Register is maintained by the TSC and is constructed from weekly electronic searches of MEDLINE, EMBASE, CINAHL and AMED, and through handsearching relevant journals. The full list of the databases, journals and conference proceedings which have been searched, as well as the search strategies used, are described in the Specialised Register section of the Cochrane Peripheral Vascular Diseases Group module in The Cochrane Library ( For the original review, the authors searched the National Health Service Database of Abstracts of Reviews of Effectiveness (DARE) (last searched August 2002) to identify potentially eligible studies and review articles. 5

9 Searching other resources Reference lists of relevant articles were screened. Data collection and analysis Selection of studies Both review authors (DAL, GYHL) independently selected trials for inclusion in the review. Disagreements were resolved by discussion. Data extraction and management DAL extracted the data, which were checked by GYHL. The data extracted included information relating to the complexities of the topic area, such as patient characteristics and concomitant treatments, as well as data relating to study eligibility, quality and outcomes. Disagreements between review authors on data extraction were resolved by discussion. Assessment of risk of bias in included studies For the updated review, risk of bias was assessed independently by the review authors using the criteria outlined in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). We assessed the domains listed below. There were three possible judgements: low risk of bias, high risk of bias, and if insufficient detail was reported the judgement on risk of bias was judged to be unclear. We assessed the new studies included in the updated review and we reassessed the studies already included in the previous versions of the review using these criteria. Sequence generation Low risk of bias, if the allocation sequence was generated using techniques such as a random number table; a computer random number generator; coin tossing; shuffling cards or envelopes; throwing dice. High risk of bias, if the allocation sequence was generated using techniques such as odd or even date of birth; date (or day) of admission; hospital or clinic record number. Unclear risk of bias, if there was insufficient information about the sequence generation process to permit judgement. Allocation concealment Low risk of bias, if the allocation concealment used methods such as central allocation (including telephone, web-based, and pharmacy-controlled randomisation); sequentially numbered drug containers of identical appearance; sequentially numbered opaque, sealed envelopes. High risk of bias, if the participants or investigators enrolling participants could possibly foresee assignments and thus introduce selection bias, such as allocation based on using an open random allocation schedule (for example a list of random numbers); assignment envelopes used without appropriate safeguards (for example if envelopes were unsealed or non-opaque, or not sequentially numbered); alternation or rotation; date of birth; case record number. Unclear risk of bias, if there was insufficient information to permit judgement of low or high risk of bias. This was usually the case if the method of concealment was not described or not described in sufficient detail to allow a definite judgement (for example if the use of assignment envelopes was described but it remained unclear whether envelopes were sequentially numbered, opaque and sealed). Where the method of allocation was unclear, we planned to contact study authors for them to provide further details. Blinding Low risk of bias, if there was no blinding but the review authors judged that the outcome and the outcome measurement were not likely to be influenced by lack of blinding; blinding of participants and key study personnel was ensured and it was unlikely that the blinding could have been broken; either participants or some key study personnel were not blinded but outcome assessment was blinded and the non-blinding of others was unlikely to introduce bias. High risk of bias, if there was no blinding or incomplete blinding and the outcome or outcome measurement was likely to be influenced by lack of blinding; blinding of key study participants and personnel was attempted but it was likely that the blinding could have been broken; either participants or some key study personnel were not blinded and the non-blinding of others was likely to introduce bias. Unclear risk of bias, if there was insufficient information to permit judgement of low or high risk of bias or the study did not address this outcome (for example where the blinding was described only as double-blind without any other details). Incomplete data assessment (loss of participants, for example with withdrawals, dropouts, protocol deviations) Low risk of bias, if there were no missing outcome data; reasons for missing outcome data were unlikely to be related to the true outcome; missing outcome data were balanced in numbers across intervention groups, with similar reasons for missing data across groups; for dichotomous outcome data, the proportion of missing outcomes compared with observed event risk was not enough to 6

10 have a clinically relevant impact on the intervention effect estimate; for continuous outcome data, plausible effect size (difference in means or standardised difference in means) among missing outcomes was not enough to have a clinically relevant impact on observed effect size; missing data were imputed using appropriate methods. High risk of bias, if the reasons for missing outcome data were likely to be related to true outcome, with either imbalance in numbers or reasons for missing data across intervention groups; for dichotomous outcome data, the proportion of missing outcomes compared with observed event risk was enough to introduce clinically relevant bias in the intervention effect estimate; for continuous outcome data, plausible effect size (difference in means or standardised difference in means) among missing outcomes was enough to introduce clinically relevant bias in observed effect size; as-treated analysis done with substantial departure of the intervention received from that assigned at randomisation; potentially inappropriate application of simple imputation. Unclear risk of bias, if there was insufficient reporting of attrition or exclusions to permit judgement of low or high risk of bias (for example numbers randomised were not stated, no reasons for missing data were provided); or the study did not address this. Selective outcome reporting Low risk of bias, if the study protocol was available and all of the study s pre-specified (primary and secondary) outcomes that were of interest in the review were reported in the pre-specified way; the study protocol was not available but it was clear that the published reports included all expected outcomes, including those that were pre-specified. High risk of bias, if not all of the study s pre-specified primary outcomes were reported; one or more primary outcomes were reported using measurements, analysis methods or subsets of the data (for example subscales) that were not pre-specified; one or more reported primary outcomes were not pre-specified (unless clear justification for their reporting was provided, such as an unexpected adverse effect); one or more outcomes of interest in the review were reported incompletely so that they could not be entered in a meta-analysis; the study report failed to include results for a key outcome that would be expected to have been reported for such a study. Unclear risk of bias, if there was insufficient information to permit judgement of low or high risk of bias. Other sources of bias Low risk of bias, if the study appeared to be free of other sources of bias. High risk of bias, if there was at least one important risk of bias (for example the study had a potential source of bias related to the specific study design used; stopped early due to some datadependent process (including a formal stopping rule); had extreme baseline imbalance; had been claimed to be fraudulent; had some other problem). Unclear risk of bias, if there was either insufficient information to assess whether an important risk of bias existed or if there was insufficient rationale or evidence that an identified problem would introduce bias. Measures of treatment effect Statistical analyses were undertaken as follows. For continuous variables (for example the change in IC distance between baseline and follow-up) where the data were presented as mean ± standard deviation (SD) the weighted mean difference with 95% confidence interval (CI) was used. Where continuous variables were presented as median (95% CI) the original data from the study were reported as other data. As a summary measure of effectiveness (for example reduction in all-cause mortality) the odds ratio (OR) with 95% CI was calculated for dichotomous variables. Unit of analysis issues In each of the included studies, participants were individually randomised to one of two intervention groups and the analyses were conducted with each patient as the individual unit of analysis, that is with a single measurement for each outcome collected for each patient and analysed. There were no cross-over trials, cluster-randomised trials or any other non-standard designs. Dealing with missing data For unpublished studies, or where data were incomplete in published papers, attempts were made to contact authors or researchers to obtain further details. For the original review, no reply was received from six authors who were approached (Bostrom 1986; Novo 1986; Rouffy 1989; Solomon 1991; Sutton-Tyrell 1995; Casiglia 1997). A reply was received from one (Bogaert 1983) saying that the original data were no longer available. For the 2009 review update, five large RCTs of anti-hypertensive treatment in hypertensive patients where there were significant numbers of patients with PAD were also identified (LIFE 2002; INVEST 2003; ASCOT 2005; VALUE 2006; ONTARGET 2008). The steering committees for each of these trials were contacted to obtain the raw data for those hypertensive patients with PAD. However, these committees were waiting for publication of the subgroup data before making the data available to us. For this updated review (2013), a subgroup analysis of the INVEST 2003 trial participants with hypertension and PAD was identified (Bavry 2010a) and the data were included. Data on the subgroup of hypertensive patients with PAD from the remaining four large RCTs were still unavailable for this updated review (LIFE 2002; ASCOT 2005; VALUE 7

11 2006; ONTARGET 2008). Should the data become available, these trials will be included in the next update of this review. Assessment of heterogeneity We were not able to perform any meta-analysis as included studies compared different types of anti-hypertensive drugs, to each other or to placebo, and outcomes measured were not common among all studies. However, if meta-analysis had been possible, we would have applied tests of heterogeneity between trials, as appropriate, using the I 2 statistic. Where high levels of heterogeneity among the trials (I 2 > 50%) were identified, we would have explored it using subgroup analysis and by performing sensitivity analysis. We would have used a random-effects model meta-analysis for an overall summary, if this had been considered appropriate. Assessment of reporting biases There were not enough studies in this review to test for reporting bias, thus the review is discussed as a narrative review. Data synthesis It was not possible to perform any meta-analysis as included studies compared different types of anti-hypertensive drugs, to each other or to placebo, and outcomes were not common among all studies. Results of the individual studies were combined into a narrative review. Subgroup analysis and investigation of heterogeneity There were insufficient studies to carry out subgroup analyses. Future revisions of the review may examine different medications within one class of drug (for example atenolol, metoprolol, nebivolol) where the comparator is identical, the duration of the treatment, or treatment effects in men versus women. This will be dependent upon the availability of such data in the included study reports. Sensitivity analysis There were insufficient studies to carry out sensitivity analyses. However, future revisions of the review may employ sensitivity analyses to examine factors that may lead to differences between the results of individual trials including poor quality versus good quality trials. R E S U L T S Description of studies Results of the search For this 2013 update, following screening of titles and abstracts full texts of 77 references to 57 trials were obtained and considered. Four new studies were included in this update and seven new studies were excluded. Included studies See the table Characteristics of included studies. In total, eight trials were finally included in this review (Overlack 1994; Schweizer 1998; DAPHNE 2002; INVEST 2003; HOPE 2004; Zankl 2010; Diehm 2011; NORMA 2011). A subgroup analysis of the PAD patients enrolled in the INVEST 2003 trial was published in 2010 (Bavry 2010a), therefore this study was moved from the excluded studies to the included studies for this update. Interventions Four studies compared anti-hypertensive therapy against placebo (Overlack 1994; Schweizer 1998; HOPE 2004; Zankl 2010), with two of these studies comparing angiotensin converting enzyme (ACE) inhibitors with placebo (Overlack 1994; HOPE 2004), another (Schweizer 1998) comparing a calcium channel blocker (CCB) with placebo, and one (Zankl 2010) comparing an angiotensin-ii receptor antagonist with placebo. Three studies compared two anti-hypertensive agents against each other (DAPHNE 2002; Diehm 2011; NORMA 2011); one study compared a thiazide diuretic with an alpha-adrenoreceptor blocking agent (DAPHNE 2002), one a beta-adrenoreceptor blocking agent with a thiazide diuretic (Diehm 2011) and another compared two beta-adrenoreceptor blocking agents against each other (NORMA 2011). Another study, a subgroup analysis of the INVEST 2003 trial (Bavry 2010a), compared a calcium antagonist-based strategy with a beta-adrenoreceptor blocking strategy. Description of PAD The description of how PAD was diagnosed differed between the included studies. In the INVEST (INternational VErapamil-SR/ Trandolapril) study (INVEST 2003), PAD was defined as a documented history of peripheral vascular disease on the baseline form. Schweizer et al (Schweizer 1998) used the Fontaine Stage IIb classification (Fontaine 1954), that is, a pain-free walking distance of less than 200 m without ischaemic rest pain or trophic changes (changes in skin and muscle related to chronic ischaemia, although how this was quantified was not explained); in addition to arterial angiography and colour-coded duplex ultrasound occlusion (diameter 5 cm) or subtotal stenoses in the distal superficial femoral artery that were present for > six months. Overlack et al (Overlack 1994) included patients with a history of IC for 8

12 six months and Fontaine Stage IIb disease who had a proven occlusion on angiography. Zankl et al (Zankl 2010) included patients with documented PAD and at least Stage Fontaine IIa. The NORMA (Nebivolol or Metoprolol in Arterial Occlusive Disease) trial (NORMA 2011) defined PAD as stable IC (Fontaine Stage II) for six months and an ABI < 0.9 (measured as the ratio of systolic blood pressure in the ankle and arm). The HOPE trial (HOPE 2004) defined PAD as either a history of IC with an ABI < 0.9 (measured by manual palpation of the foot pulse and not ultrasound doppler) or previous vascular intervention or limb amputation for PAD. Diehm et al (Diehm 2011) included patients with one or more of the following: a history of typical IC for at six months; actual proven PAD by objective means such as haemodynamics and non-invasive imaging or angiography; history of previous vascular intervention; or ABI of the worse leg < 0.9. In the DAPHNE (Doxazosin Atherosclerosis Progression in Hypertension in the NEtherlands) study (DAPHNE 2002) PAD was defined as IC or peripheral vascular surgery because of atherosclerosis. Definition of hypertension The studies used a variety of different criteria to define hypertension. Schweizer (Schweizer 1998) and Overlack (Overlack 1994) included patients with mild essential hypertension, which was quantified in one study (Overlack 1994) as a sitting diastolic blood pressure of 95 to 104 mmhg on three different occasions. The DAPHNE study defined hypertension on the basis of diastolic blood pressure alone, 95 to 115 mmhg in two out of three supine readings (DAPHNE 2002). The HOPE study defined hypertension as blood pressure > 160/90 mmhg or on the basis that people were on anti-hypertensive treatment (HOPE 2004). Zankl et al (Zankl 2010) included patients with arterial hypertension which was not defined but those with uncontrolled hypertension (systolic blood pressure 170 mmhg and diastolic blood pressure 95 mmhg or both) were excluded. The NORMA study (NORMA 2011) included patients with stage I arterial hypertension (systolic blood pressure (SBP) 140 to 159 mmhg and diastolic blood pressure (DBP) 90 to 99 mmhg) or a previous diagnosis of stage I arterial hypertension who were under current treatment. At the time of inclusion, SBP had to be > 100 mmhg and < 160 mmhg, with DBP < 100 mmhg (NORMA 2011). Diehm et al (Diehm 2011) included patients with a SBP between 140 and 179 mmhg and a DBP 90 to 109 mmhg with or without treatment with antihypertensive drugs at screening. After a four-week run-in period, the second eligibility criterion for hypertension was a SBP > 130 mmhg and DBP pressure > 85 mmhg (NORMA 2011). The INVEST study (INVEST 2003) included patients with essential hypertension defined as SBP > 140 mmhg and DBP > 90 mmhg (if diabetes or renal impairment was present hypertension was defined as SBP > 130 mmhg and DBP > 85 mmhg) requiring treatment. Excluded studies See the table Characteristics of excluded studies. Fifty-nine studies were excluded for the following reasons. 1. Ten studies were excluded because the treatment period was shorter than the month specified in the review protocol (Reichert 1975; Smith 1982; Lepantalo 1984; Hiatt 1985; Lepantalo 1985a; Klieber 1986; Bernardi 1988; Natali 1989; Kalus 1995; Liu 1997). 2. Seven studies were eligible for inclusion but the data presentation in the papers was inadequate and attempts to obtain the specific data failed. For six of these trials the authors were contacted but did not respond (Bostrom 1986; Novo 1986; Rouffy 1989; Solomon 1991; Sutton-Tyrell 1995; Casiglia 1997). For one study (Bogaert 1983) the author was successfully contacted but the data were not available as the study had been conducted several years ago. 3. Nine studies did not include patients with PAD (Lepantalo 1983; Panzner 1992; Weibull 1992; Stumpe 1993; Leeman 1995; Stumpe 1995; OPERA 2002; POISE 2008; Takeda 2010). 4. Six studies were not RCTs (Ingram 1982; Winterfeld 1984; Lepantalo 1985b; Novo 1985; Coto 1991; Van de Ven 1994). 5. Eight studies did not use anti-hypertensive medication (Larsen 1969; Domschky 1977; Jageneau 1977; Nelson 1978; Staessen 1978; Coto 1989; Branchereau 1995; Lievre 1996). 6. Two studies were excluded because none of the outcome measures matched those predefined for this review (Siniscalchi 1993; Brown 1998). 7. Four studies were excluded because the patients were not hypertensive at the outset of the trial, or hypertensive patients were excluded (Spence 1993; Laurent 1994; Schweizer 1996; Ahimastos 2006). 8. Two trials (Svendsen 1986; Roberts 1987) were cross-over trials with a washout period at the start of the trial; the postintervention data from both or all phases were presented lumped together and not separately. Another trial (Roberts 1992) was a cross-over trial with no washout phase between treatments; the data from both phases were presented lumped together. 9. Another trial compared captopril against ticlopidine but did not compare captopril against placebo (Novo 1996). In addition, this trial (Novo 1996) only used ABI to confirm diagnosis although the exact value used was not mentioned. 10. One study involved open-label administration of vasodilating agents to control blood pressure, if necessary, thus making the results of dubious value (Diehm 1993). 11. One publication (Mann 1998) was excluded because it was a protocol for an RCT with no data. Another study was excluded as it was a comment on the results of the ONTARGET trial and did not present any data (Liakishev 2008). One publication was excluded because it was only published as an abstract although the full abstract was not available (Gastmann 1987). 12. The data for PAD patients were not available from five RCTs (LIFE 2002; ABCD 2003; ASCOT 2005; VALUE 2006; 9

13 ONTARGET 2008). Risk of bias in included studies See the table Characteristics of included studies. For the updated review, risk of bias was assessed independently by two review authors using the criteria outlined in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). The results are displayed in Figure 1 and Figure 2. Figure 1. Methodological quality graph: review authors judgements about each methodological quality item presented as percentages across all included studies. 10

14 Figure 2. Methodological quality summary: review authors judgements about each methodological quality item for each included study. 11