Advertisement
Clinical research study| Volume 130, ISSUE 6, P707-719.e8, June 2017

Optimal Systolic Blood Pressure Target After SPRINT: Insights from a Network Meta-Analysis of Randomized Trials

Published:January 18, 2017DOI:https://doi.org/10.1016/j.amjmed.2017.01.004

      Abstract

      Background

      The optimal on-treatment blood pressure (BP) target has been a matter of debate. The recent SPRINT trial showed significant benefits of a BP target of <120 mm Hg, albeit with an increase in serious adverse effects related to low BP.

      Methods

      PubMed, EMBASE, and CENTRAL were searched for randomized trials comparing treating with different BP targets. Trial arms were grouped into 5 systolic BP target categories: 1) <160 mm Hg, 2) <150 mm Hg, 3) <140 mm Hg, 4) <130 mm Hg, and 5) <120 mm Hg. Efficacy outcomes of stroke, myocardial infarction, death, cardiovascular death, heart failure, and safety outcomes of serious adverse effects were evaluated using a network meta-analysis.

      Results

      Seventeen trials that enrolled 55,163 patients with 204,103 patient-years of follow-up were included. There was a significant decrease in stroke (rate ratio [RR] 0.54; 95% confidence interval [CI], 0.29-1.00) and myocardial infarction (RR 0.68; 95% CI, 0.47-1.00) with systolic BP <120 mm Hg (vs <160 mm Hg). Sensitivity analysis using achieved systolic BP showed a 72%, 97%, and 227% increase in stroke with systolic BP of <140 mm Hg, <150 mm Hg, and <160 mm, respectively, when compared with systolic BP <120 mm Hg. There was no difference in death, cardiovascular death, or heart failure when comparing any of the BP targets. However, the point estimate favored lower BP targets (<120 mm Hg, <130 mm Hg) when compared with higher BP targets (<140 mm Hg or <150 mm Hg). BP targets of <120 mm Hg and <130 mm Hg ranked #1 and #2, respectively, as the most efficacious target. There was a significant increase in serious adverse effects with systolic BP <120 mm Hg vs <150 mm Hg (RR 1.83; 95% CI, 1.05-3.20) or vs <140 mm Hg (RR 2.12; 95% CI, 1.46-3.08). BP targets of <140 mm Hg and <150 mm Hg ranked #1 and #2, respectively, as the safest target for the outcome of serious adverse effects. Cluster plots for combined efficacy and safety showed that a systolic BP target of <130 mm Hg had optimal balance between efficacy and safety.

      Conclusions

      In patients with hypertension, a on-treatment systolic BP target of <130 mm Hg achieved optimal balance between efficacy and safety.

      Keywords

      Clinical Significance
      • Data from 204,103 patient-years of follow-up showed that blood pressure (BP) targets of <120 mm Hg and <130 mm Hg ranked #1 and #2 as the most efficacious target, whereas BP targets of <140 mm Hg and <150 mm Hg ranked #1 and #2 as the safest target for the outcome of serious adverse effects.
      • A systolic BP target of <130 mm Hg had optimal balance between efficacy and safety.
      Most major national and international guidelines for management of hypertension recommend a systolic blood pressure (BP) target of <140 mm Hg. However, the optimal BP target in patients with hypertension is controversial, with 2 recent major trials (albeit in different populations) reaching opposite conclusions.
      • Kjeldsen S.E.
      • Narkiewicz K.
      • Hedner T.
      • Mancia G.
      The SPRINT study: outcome may be driven by difference in diuretic treatment demasking heart failure and study design may support systolic blood pressure target below 140 mmHg rather than below 120 mmHg.
      The Systolic Blood Pressure Intervention Trial (SPRINT) of patients without diabetes showed a benefit of an intensive systolic BP target of <120 mm Hg, with reduction in the rate of the composite primary outcome by 25%, and the risk of death by 27%, compared with the standard systolic BP target of <140 mm Hg.
      • Wright Jr., J.T.
      • Williamson J.D.
      • Whelton P.K.
      • et al.
      SPRINT Research Group
      A randomized trial of intensive versus standard blood-pressure control.
      In the Action to Control Cardiovascular Risk in Diabetes Blood Pressure (ACCORD BP) trial, intensive systolic BP target of <120 mm Hg failed to reduce cardiovascular outcomes with the exception of stroke, when compared with a standard BP target of <140 mm Hg in patients with diabetes.
      • Cushman W.C.
      • Evans G.W.
      • Byington R.P.
      • et al.
      ACCORD Study Group
      Effects of intensive blood-pressure control in type 2 diabetes mellitus.
      Aforementioned randomized trials were not powered for individual endpoints. While composite endpoints are used in clinical trials mainly to reduce the sample size, the interpretation of the trial results on individual outcomes (which is more important for physician and patients) is problematic. In addition, prior meta-analyses on this subject have compared broad categories of more intensive vs less intensive BP targets.
      • Xie X.
      • Atkins E.
      • Lv J.
      • et al.
      Effects of intensive blood pressure lowering on cardiovascular and renal outcomes: updated systematic review and meta-analysis.
      However, the more intensive group had a systolic BP target ranging from <150 mm Hg (as was targeted in the UK Prospective Diabetes Study Group [UKPDS]-38 trial) to a target of <120 mm Hg (as in the SPRINT and ACCORD trials).
      • Xie X.
      • Atkins E.
      • Lv J.
      • et al.
      Effects of intensive blood pressure lowering on cardiovascular and renal outcomes: updated systematic review and meta-analysis.
      Although more intensive BP targets were shown to be associated with reduction in cardiovascular outcomes in such analyses, given the heterogeneity of “more intensive” BP group definition in trials (ranging from <120 mm Hg to <150 mm Hg), these studies do not permit identification of an optimal systolic BP target.
      The objective of the study was to evaluate the efficacy and safety of various systolic BP targets using a network meta-analysis (where BP targets can be compared individually without the need to lump them together) and to determine the impact of various BP targets on individual outcomes.

      Methods

      Eligibility Criteria

      We searched PubMed, EMBASE, and Cochrane Central Registry of Controlled Trials (CENTRAL) for randomized clinical trials comparing treating to different BP targets. Trials comparing systolic, diastolic, or mean arterial BP targets were selected. The search terms are outlined in Supplementary Table 1 (Appendix, available online). We checked the reference lists of original studies, review articles, and meta-analyses identified by the electronic searches to find other eligible trials. There was no language restriction for the search.
      Eligible trials had to fulfill the following criteria: 1) trials comparing different BP targets; and 2) trials reporting the outcomes of interest (below). We excluded trials that were not strategy trials of BP targets, such as those testing antihypertensive agents vs placebo or against each other.

      Selection and Quality Assessment

      Two authors (BT and SB) independently assessed trial eligibility, extracted data, and evaluated trial bias risk. Disagreements were resolved by consensus. We used the risk of bias assessment scheme recommended by the Cochrane Collaboration for quality assessment,

      Higgins J, Green S.Cochrane Handbook for Systematic Reviews of Interventions Version 5.0.0 ed: The Cochrane Collaboration; 2008. Available at: www.cochrane-handbook.org. Accessed August 1, 2016.

      which includes adequacy of sequence generation of the allocation, adequacy of allocation concealment, adequacy of blinding of participants, personnel, and outcome assessors; absence of incomplete outcome data; absence of selective outcome reporting; and other sources of bias. Trials with high or unclear risk for bias for any one of the first 3 components were categorized as trials with high risk of bias. Otherwise, they were categorized as trials with low risk of bias.

      Data Extraction and Synthesis

      Treatment arms were categorized into 5 systolic BP targets: 1) <160 mm Hg; 2) <150 mm Hg; 3) <140 mm Hg; 4) <130 mm Hg; and 5) <120 mm Hg. The categorization was done based on the intended BP target in the systolic BP target trials. For example, the intensive BP group in the SPRINT trial targeted a systolic BP of <120 mm Hg and was grouped as <120 mm Hg category even though the achieved mean BP was 121.4 mm Hg. For trials that targeted either the diastolic BP targets or the mean arterial pressure, the mean achieved systolic BP was used to group the treatment arms. Although for simplicity we represent the BP targets as, for example, <160 mm Hg, <150 mm Hg, they are, in essence, BP targets of 150 to <160 mm Hg, 140 to <150 mm Hg, and so on.

      Study Outcomes

      The efficacy outcomes evaluated were stroke, myocardial infarction, death, cardiovascular death, and heart failure. In addition, safety outcome of serious adverse effects was evaluated. The definitions of the outcomes were as per the individual trials and are outlined in Table 1.
      • Wright Jr., J.T.
      • Williamson J.D.
      • Whelton P.K.
      • et al.
      SPRINT Research Group
      A randomized trial of intensive versus standard blood-pressure control.
      • Cushman W.C.
      • Evans G.W.
      • Byington R.P.
      • et al.
      ACCORD Study Group
      Effects of intensive blood-pressure control in type 2 diabetes mellitus.
      • Agodoa L.Y.
      • Appel L.
      • Bakris G.L.
      • et al.
      Effect of ramipril vs amlodipine on renal outcomes in hypertensive nephrosclerosis: a randomized controlled trial.
      • Appel L.J.
      • Wright Jr., J.T.
      • Greene T.
      • et al.
      Intensive blood-pressure control in hypertensive chronic kidney disease.
      • Wright Jr., J.T.
      • Bakris G.
      • Greene T.
      • et al.
      Effect of blood pressure lowering and antihypertensive drug class on progression of hypertensive kidney disease: results from the AASK trial.
      • Schrier R.W.
      • Estacio R.O.
      • Esler A.
      • Mehler P.
      Effects of aggressive blood pressure control in normotensive type 2 diabetic patients on albuminuria, retinopathy and strokes.
      • Schrier R.W.
      • Estacio R.O.
      • Mehler P.S.
      • Hiatt W.R.
      Appropriate blood pressure control in hypertensive and normotensive type 2 diabetes mellitus: a summary of the ABCD trial.
      • Estacio R.O.
      • Coll J.R.
      • Tran Z.V.
      • Schrier R.W.
      Effect of intensive blood pressure control with valsartan on urinary albumin excretion in normotensive patients with type 2 diabetes.
      • Chew E.Y.
      • Ambrosius W.T.
      • Davis M.D.
      • et al.
      ACCORD Study Group, ACCORD Eye Study Group
      Effects of medical therapies on retinopathy progression in type 2 diabetes.
      • Hannson L.
      The BBB Study: the effect of intensified antihypertensive treatment on the level of blood pressure, side-effects, morbidity and mortality in “well-treated” hypertensive patients. Behandla Blodtryck Battre.
      • Verdecchia P.
      • Staessen J.A.
      • Angeli F.
      • et al.
      Usual versus tight control of systolic blood pressure in non-diabetic patients with hypertension (Cardio-Sis): an open-label randomised trial.
      • Asayama K.
      • Ohkubo T.
      • Metoki H.
      • et al.
      Cardiovascular outcomes in the first trial of antihypertensive therapy guided by self-measured home blood pressure.
      • Hansson L.
      • Zanchetti A.
      • Carruthers S.G.
      • et al.
      Effects of intensive blood-pressure lowering and low-dose aspirin in patients with hypertension: principal results of the Hypertension Optimal Treatment (HOT) randomised trial. HOT Study Group.
      JATOS Study Group
      Principal results of the Japanese trial to assess optimal systolic blood pressure in elderly hypertensive patients (JATOS).
      • Klahr S.
      • Levey A.S.
      • Beck G.J.
      • et al.
      The effects of dietary protein restriction and blood-pressure control on the progression of chronic renal disease. Modification of Diet in Renal Disease Study Group.
      • Sarnak M.J.
      • Greene T.
      • Wang X.
      • et al.
      The effect of a lower target blood pressure on the progression of kidney disease: long-term follow-up of the modification of diet in renal disease study.
      • Ruggenenti P.
      • Perna A.
      • Loriga G.
      • et al.
      Blood-pressure control for renoprotection in patients with non-diabetic chronic renal disease (REIN-2): multicentre, randomised controlled trial.
      • Schrier R.
      • McFann K.
      • Johnson A.
      • et al.
      Cardiac and renal effects of standard versus rigorous blood pressure control in autosomal-dominant polycystic kidney disease: results of a seven-year prospective randomized study.
      • Benavente O.R.
      • Coffey C.S.
      • Conwit R.
      • et al.
      SPS3 Study Group
      Blood-pressure targets in patients with recent lacunar stroke: the SPS3 randomised trial.
      • Pergola P.E.
      • White C.L.
      • Szychowski J.M.
      • et al.
      Achieved blood pressures in the secondary prevention of small subcortical strokes (SPS3) study: challenges and lessons learned.
      Tight blood pressure control and risk of macrovascular and microvascular complications in type 2 diabetes: UKPDS 38. UK Prospective Diabetes Study Group.
      • Ogihara T.
      • Saruta T.
      • Rakugi H.
      • et al.
      Target blood pressure for treatment of isolated systolic hypertension in the elderly: valsartan in elderly isolated systolic hypertension study.
      National Institute for Health and Care Excellence
      Hypertension: clinical management of primary hypertension in adults. NICE guidelines [CG127].
      • Wei Y.
      • Jin Z.
      • Shen G.
      • et al.
      Effects of intensive antihypertensive treatment on Chinese hypertensive patients older than 70 years.
      The definition of serious adverse effects, though variable in the trials, included angioedema, hypotension, syncope, bradycardia/arrhythmia, or hypo-/hyperkalemia.
      Table 1Trial Definition of Efficacy and Safety Outcomes
      TrialMyocardial InfarctionStrokeSerious Adverse Effects
      AASK
      • Agodoa L.Y.
      • Appel L.
      • Bakris G.L.
      • et al.
      Effect of ramipril vs amlodipine on renal outcomes in hypertensive nephrosclerosis: a randomized controlled trial.
      • Appel L.J.
      • Wright Jr., J.T.
      • Greene T.
      • et al.
      Intensive blood-pressure control in hypertensive chronic kidney disease.
      • Wright Jr., J.T.
      • Bakris G.
      • Greene T.
      • et al.
      Effect of blood pressure lowering and antihypertensive drug class on progression of hypertensive kidney disease: results from the AASK trial.
      When trials did not define or report serious adverse event, we have defined serious adverse event as “angioedema, hypotension, syncope, bradycardia/arrhythmia or hypo/hyperkalemia.”
      Definitive nonfatal myocardial infarction is defined as a clinical report of myocardial infarction from the investigator and the presence of one of the following: 1) Elevation of CPK > twice the upper limit of normal for the given hospital supported by the elevation of cardiac-specific enzyme above the normal range, such as MB fraction or cardiac troponin; 2) In the absence of cardiac specific enzymes, determination of a typical evolutionary pattern defined as an elevation of CPK to twice the upper limit of normal for the given hospital followed by a fall of at least 50% or the appearance of new pathological Q-waves in 2 or more contiguous leads; 3) The appearance of a R-wave with R/S ratio in lead V1 >1.0 in the absence of another explanation or a loss of progression of R-waves V2 through V5. Probable nonfatal myocardial infarction is defined by a clinical report from the investigator but lacking confirmation of elevated enzymes or ECG changes, or defined by centrally read ECG that documents a new myocardial infarction in comparison with the baseline ECG but without clinical event.Definitive stroke is defined as a permanent neurological deficit of at least 24 h, attributed to stroke by the personal physician, requiring hospitalization and confirmation by radiographic imaging. Probable stroke is defined as above but lacking confirmation by radiographic imaging.Angioedema, hypotension, syncope, bradycardia/arrhythmia, or hypo-/hyperkalemia
      ABCD (Normotensive)
      • Schrier R.W.
      • Estacio R.O.
      • Esler A.
      • Mehler P.
      Effects of aggressive blood pressure control in normotensive type 2 diabetic patients on albuminuria, retinopathy and strokes.
      • Schrier R.W.
      • Estacio R.O.
      • Mehler P.S.
      • Hiatt W.R.
      Appropriate blood pressure control in hypertensive and normotensive type 2 diabetes mellitus: a summary of the ABCD trial.
      ECGs are evaluated for abnormalities consistent for Q-wave myocardial infarctions. The End Point Committee reviewed all hospital admissions that appeared to be related to a cardiovascular event.The End Point Committee reviewed all hospital admissions that appeared to be related to a cardiovascular eventNR
      ABCD-2 Valsartan
      • Estacio R.O.
      • Coll J.R.
      • Tran Z.V.
      • Schrier R.W.
      Effect of intensive blood pressure control with valsartan on urinary albumin excretion in normotensive patients with type 2 diabetes.
      An independent Endpoint Committee, which was blinded to the study intervention arms, reviewed all cardiovascular events. The Endpoint Committee reviewed all hospital admissions that appeared to be related to cardiovascular events.An independent Endpoint Committee, which was blinded to the study intervention arms, reviewed all cardiovascular events. The Endpoint Committee reviewed all hospital admissions that appeared to be related to cardiovascular events.NR
      ACCORD
      • Cushman W.C.
      • Evans G.W.
      • Byington R.P.
      • et al.
      ACCORD Study Group
      Effects of intensive blood-pressure control in type 2 diabetes mellitus.
      • Chew E.Y.
      • Ambrosius W.T.
      • Davis M.D.
      • et al.
      ACCORD Study Group, ACCORD Eye Study Group
      Effects of medical therapies on retinopathy progression in type 2 diabetes.
      The diagnosis of MI is based on the occurrence of a compatible clinical syndrome associated with diagnostic elevation of cardiac enzymes (ie, an increase in troponin T or troponin I to a level indicating myonecrosis or an increase in creatine kinase–myocardial band to a level more than twice the upper limit of normal). Q-wave MI is defined as the development of new significant Q waves. Silent MI is diagnosed when new (compared with the previous 12-lead electrocardiogram) significant Q waves are detected by surveillance electrocardiography performed every 2 y and at study end in all participants.Stroke is diagnosed by a focal neurologic deficit that lasts >24 h, associated with evidence of brain infarction or hemorrhage by computed tomography, MRI, or autopsy.Life-threatening, cause permanent disability, or necessitate hospitalization and were considered by the investigators to be possibly, probably, or definitely related to antihypertensive medications
      BBB
      • Hannson L.
      The BBB Study: the effect of intensified antihypertensive treatment on the level of blood pressure, side-effects, morbidity and mortality in “well-treated” hypertensive patients. Behandla Blodtryck Battre.
      When trials did not define or report serious adverse event, we have defined serious adverse event as “angioedema, hypotension, syncope, bradycardia/arrhythmia or hypo/hyperkalemia.”
      NRNRAngioedema, hypotension, syncope, bradycardia/arrhythmia or hypo/hyperkalemia
      Cardio-Sis
      • Verdecchia P.
      • Staessen J.A.
      • Angeli F.
      • et al.
      Usual versus tight control of systolic blood pressure in non-diabetic patients with hypertension (Cardio-Sis): an open-label randomised trial.
      When trials did not define or report serious adverse event, we have defined serious adverse event as “angioedema, hypotension, syncope, bradycardia/arrhythmia or hypo/hyperkalemia.”
      Myocardial infarction (MI) is defined as follows:

      (1) Q-wave or ST-elevation MI: new significant Q-wave (>0.04 s duration or 3 mm in depth and loss in height of ensuing R wave or new significant R waves in V1–V2) in at least 2 leads on the standard 12-lead ECG. ST elevation is defined as 0.1 mV new ST elevation of 0.1 mV in peripheral leads or 0.2 mV in precordial leads. New onset left bundle branch block during MI is equivalent to a Q-Wave MI. There must be at least one of the following criteria: 1) typical chest pain or increase of CK-MB above the upper limit of normal within 36 h of onset of acute symptoms; 2) serum glutamic oxaloacetic transaminase or LDH at least twice the laboratory upper limit; and 3) elevated troponin T or I level above the normal laboratory range.

      (2) “Non-Q-wave” or “non-ST elevation” MI: new and persistent (>24 h) ST-segment or T wave changes in addition to cardiac enzymes/markers elevation (see above) or typical symptoms of chest pain.

      (3) MI without significant ECG changes: typical symptoms with significant elevation of cardiac enzymes (see above).
      “Stroke” is defined by acute focal neurological deficit thought to be of vascular origin and signs or symptoms lasting >24 h. On the basis of symptoms and laboratory tests (computed tomography/magnetic resonance imaging) or necropsy results, stroke is classified as: 1) definite or probable ischaemic stroke; 2) definite or probable haemorrhagic stroke; or (3) subarachnoid hemorrhage or (4) uncertain or unknown stroke.

      “Transient ischaemic attack” is defined by focal neurological or monocular defect with associated symptoms lasting <24 h and thought to be due to occlusive (embolic or thrombotic) vascular origin.
      Angioedema, hypotension, syncope, bradycardia/arrhythmia or hypo/hyperkalemia
      HOMED-BP
      • Asayama K.
      • Ohkubo T.
      • Metoki H.
      • et al.
      Cardiovascular outcomes in the first trial of antihypertensive therapy guided by self-measured home blood pressure.
      We coded end points according to the tenth revision of the International Classification of Diseases (ICD-10). An endpoint adjudication committee blind to treatment assignment reviewed all suspected primary and secondary outcomes.We coded end points according to the tenth revision of the International Classification of Diseases (ICD-10). Fatal and nonfatal stroke did not include transient ischemic attacks. An endpoint adjudication committee blind to treatment assignment reviewed all suspected primary and secondary outcomes.NR
      HOT
      • Hansson L.
      • Zanchetti A.
      • Carruthers S.G.
      • et al.
      Effects of intensive blood-pressure lowering and low-dose aspirin in patients with hypertension: principal results of the Hypertension Optimal Treatment (HOT) randomised trial. HOT Study Group.
      A classification of all reported events was made by the Independent Clinical Event Committee based on all available information (hospital records, physician's records, death certificates, and necropsy reports). Silent myocardial infarctions were documented by taking an electrocardiogram (ECG) at randomization and at the final visit.A classification of all reported events was made by the Independent Clinical Event Committee based on all available information (hospital records, physician's records, death certificates, and necropsy reports). Diagnosis of a nonfatal stroke required unequivocal signs or symptoms of remaining neurological deficit, with a sudden onset and a duration of more than 24 h. Diagnosis of a fatal stroke also required the criteria given above. Alternatively, the diagnosis could be given in the hospital records or described in the necropsy report.NR
      JATOS
      JATOS Study Group
      Principal results of the Japanese trial to assess optimal systolic blood pressure in elderly hypertensive patients (JATOS).
      Endpoints were evaluated by the members of the Endpoint Evaluation Committee, who were blinded to the treatment assignments of the subjects.Endpoints were evaluated by the members of the Endpoint Evaluation Committee, who were blinded to the treatment assignments of the subjects.NR
      MDRD
      • Klahr S.
      • Levey A.S.
      • Beck G.J.
      • et al.
      The effects of dietary protein restriction and blood-pressure control on the progression of chronic renal disease. Modification of Diet in Renal Disease Study Group.
      • Sarnak M.J.
      • Greene T.
      • Wang X.
      • et al.
      The effect of a lower target blood pressure on the progression of kidney disease: long-term follow-up of the modification of diet in renal disease study.
      NRNRNR
      REIN-2
      • Ruggenenti P.
      • Perna A.
      • Loriga G.
      • et al.
      Blood-pressure control for renoprotection in patients with non-diabetic chronic renal disease (REIN-2): multicentre, randomised controlled trial.
      NRNRNR
      Schrier et al
      • Schrier R.
      • McFann K.
      • Johnson A.
      • et al.
      Cardiac and renal effects of standard versus rigorous blood pressure control in autosomal-dominant polycystic kidney disease: results of a seven-year prospective randomized study.
      NRNRNR
      SPRINT
      • Wright Jr., J.T.
      • Williamson J.D.
      • Whelton P.K.
      • et al.
      SPRINT Research Group
      A randomized trial of intensive versus standard blood-pressure control.
      The definition includes myocardial infarction that occurred during surgery or a procedure and myocardial infarction aborted by thrombolytic therapy or procedure. Silent myocardial infarction, determined using 12-lead ECG at years 2 and 4 and the close-out visit compared with baseline, is determined centrally in the absence of clinically detected myocardial infarction using the Minnesota ECG classification. The diagnosis of non-myocardial infarction acute coronary syndrome requires hospitalization for evaluation, with documented new or changing cardiac ischemic symptoms. Furthermore, confirmatory evidence of coronary artery disease is required.Stroke is defined as the rapid onset of focal neurologic symptoms, headache, or meningismus not due to other conditions (eg, central nervous system infection), plus a lesion on brain imaging consistent with symptoms except when death occurs within 24 h without resolution of symptoms.Hypotension, syncope, bradycardia, electrolyte abnormalities, injurious fall and acute kidney injury or failure
      SPS3
      • Benavente O.R.
      • Coffey C.S.
      • Conwit R.
      • et al.
      SPS3 Study Group
      Blood-pressure targets in patients with recent lacunar stroke: the SPS3 randomised trial.
      • Pergola P.E.
      • White C.L.
      • Szychowski J.M.
      • et al.
      Achieved blood pressures in the secondary prevention of small subcortical strokes (SPS3) study: challenges and lessons learned.
      Acute myocardial infarct, defined by standard criteria consisting of ECG and cardiac enzyme changes requiring acute hospitalization.Ischemic stroke is clinically defined as a focal neurological deficit persisting for >24 h and is ascertained via clinical evaluation and use of CT or MRI. Ischemic strokes, restricted to worsening neurological deficits associated with the qualifying or other prior stroke, must meet at least one of the 2 following criteria: first, no major metabolic derangement that could “unmask” prior focal deficits and an objective change in neurological examination by an SPS3 neurologist, and second, unequivocally positive DWI indicative of acute ischemia. The CNS hemorrhage is defined as primary intracranial bleeding of any type documented by neuroimaging, CSF examination, or autopsy. These can be spontaneous, traumatic, or associated with underlying lesions (ie, aneurysms, arteriovenous malformations, or amyloid angiopathy) and include intraparenchymal, subdural, epidural, intraventricular, and subarachnoid locations. All possible clinical stroke events are assessed at the clinical site by both the local neurology investigator and a neurologist blinded to the assigned treatment arms.Orthostatic syncope, stroke associated with hypotension, myocardial infarction, fall with injury, other adverse effects related to hypotension
      UKPDS-38
      Tight blood pressure control and risk of macrovascular and microvascular complications in type 2 diabetes: UKPDS 38. UK Prospective Diabetes Study Group.
      WHO clinical criteria with associated ECG/enzyme changes or new pathological Q wave.NRNR
      VALISH
      • Ogihara T.
      • Saruta T.
      • Rakugi H.
      • et al.
      Target blood pressure for treatment of isolated systolic hypertension in the elderly: valsartan in elderly isolated systolic hypertension study.
      End points and adverse events were blindly evaluated according to the prospective, randomized, open-label, blinded end point design
      National Institute for Health and Care Excellence
      Hypertension: clinical management of primary hypertension in adults. NICE guidelines [CG127].
      by the endpoint committee and the safety committee, respectively.
      End points and adverse events were blindly evaluated according to the prospective, randomized, open-label, blinded end point design
      National Institute for Health and Care Excellence
      Hypertension: clinical management of primary hypertension in adults. NICE guidelines [CG127].
      by the endpoint committee and the safety committee, respectively.
      NR
      Wei et al
      • Wei Y.
      • Jin Z.
      • Shen G.
      • et al.
      Effects of intensive antihypertensive treatment on Chinese hypertensive patients older than 70 years.
      Endpoints were evaluated by the members of the Endpoint Evaluation Committee, who were blinded to the treatment assignments and the time course of BP. Acute myocardial infarction was diagnosed using the criteria by 2008 ESC Committee for Practice Guidelines.Endpoints were evaluated by the members of the Endpoint Evaluation Committee, who were blinded to the treatment assignments and the time course of BP. The proper diagnosis of stroke required both neurological examinations and craniocervical computed tomography, or magnetic resonance imaging.NR
      AASK = African American Study of Kidney Disease and Hypertension; ABCD = Appropriate Blood Pressure Control in Diabetes; ACCORD = Action to Control Cardiovascular Risk in Diabetes; BBB = Behandla Blodtryck Battre; BP = blood pressure; Cardio-Sis = Studio Italiano Sugli Effetti CARDIOvascolari del Controllo della Pressione Arteriosa SIStolica; CPK = creatine phosphokinase; CT = computed tomography; DBP = diastolic blood pressure; ECG = electrocardiogram; HOMED-BP = Hypertension Objective Treatment Based on Measurement by Electrical Devices of Blood Pressure; HOT = Hypertension Optimal Treatment; HTN = hypertension; JATOS = Japanese Trial to Assess Optimal Systolic Blood Pressure in Elderly Hypertensive Patients; MAP = mean arterial pressure; MDRD = Modification of Diet in Renal Disease; MRI = magnetic resonance imaging; NR = not reported; REIN = Ramipril Efficacy In Nephropathy; SBP = systolic blood pressure; SPRINT = Systolic Blood Pressure Intervention Trial; SPS3 = Secondary Prevention of Small Subcortical Strokes; T2DM = type 2 diabetes mellitus; UKPDS = UK Prospective Diabetes Study Group; VALISH = Valsartan in Elderly Isolated Systolic Hypertension.
      When trials did not define or report serious adverse event, we have defined serious adverse event as “angioedema, hypotension, syncope, bradycardia/arrhythmia or hypo/hyperkalemia.”

      Statistical Analyses

      Network Meta-Analysis

      The primary analysis compared each systolic BP target category against the <160 mm Hg group, which was chosen as the reference group. Pairwise comparisons were performed to assess all other possible comparators using a network meta-analysis. Network meta-analysis allows for synthesis of direct and indirect evidence from a network of trials that compare different systolic BP targets by incorporating the indirect comparisons constructed from 2 trials that have one arm in common. Network meta-analysis was performed using a frequentist-based approach with the use of multivariate random-effects meta-analysis.
      • Higgins J.P.T.
      • Jackson D.
      • Barrett J.K.
      • Lu G.
      • Ades A.E.
      • White I.R.
      Consistency and inconsistency in network meta-analysis: concepts and models for multi-arm studies.
      • White I.R.
      • Barrett J.K.
      • Jackson D.
      • Higgins J.P.
      Consistency and inconsistency in network meta-analysis: model estimation using multivariate meta-regression.
      The weight of each direct comparison was proportional to the variance of the observed effect and the network structure. Multivariate meta-analysis was used to estimate the within-study correlations and to estimate the relative treatment effects, their variance and covariance.
      • White I.
      Multivariate random-effects meta-analysis.
      The covariance between 2 estimates from the same study (sharing a common treatment group) was the variance of the data in the shared arm and was calculated using the multivariate meta-analysis method. To account for the different follow-up duration of the individual trials, we used the rate of outcomes per 1000 person-years to obtain the log rate ratios (RR) of one systolic BP target to another. Rates, rather than number of events, were considered the most appropriate outcome for these analyses because they incorporate the duration of the trials. Patient-years of follow-up for each trial were calculated by multiplying the trial sample size with the mean follow-up duration of the trial. Sensitivity analyses were conducted using the number of events. Inconsistency in a network model was estimated using the inconsistency factors (the difference between direct and indirect treatment effect estimates) and their uncertainty (using loop-specific heterogeneity estimates).
      The presence of small-study effect was assessed using a “comparison-adjusted” funnel plot that takes into account different summary effect for each set of studies (measure of precision vs estimated treatment effect).
      • Chaimani A.
      • Higgins J.P.
      • Mavridis D.
      • et al.
      Graphical tools for network meta-analysis in STATA.
      The horizontal axis is the study-specific effect sizes centered to the respective comparison-specific pooled effect size, while the vertical axis is the inverted standard error of the effect sizes as used in a standard funnel plot. If all studies are lying symmetrically around the zero line, the comparison-adjusted funnel plot suggests no evidence of small-study effects in the network.
      • Chaimani A.
      • Higgins J.P.
      • Mavridis D.
      • et al.
      Graphical tools for network meta-analysis in STATA.
      In addition, ranking plots (rankograms) were constructed using probabilities (and their uncertainty in relative treatment effects) that a given treatment had the lowest event rate for each outcome. In addition, a cumulative rankogram, which presents the probabilities that a treatment would be among the best treatments, was also constructed. The surface under the cumulative ranking curve (SUCRA), which is a simple transformation of the mean rank, was used to provide a hierarchy of the treatments.
      • Salanti G.
      • Ades A.E.
      • Ioannidis J.P.
      Graphical methods and numerical summaries for presenting results from multiple-treatment meta-analysis: an overview and tutorial.
      The larger the SUCRA value, the higher the rank of the treatment.
      • Salanti G.
      • Ades A.E.
      • Ioannidis J.P.
      Graphical methods and numerical summaries for presenting results from multiple-treatment meta-analysis: an overview and tutorial.
      A clustered ranking plot was also constructed using SUCRA values for efficacy (stroke, myocardial infarction, death) and safety (serious adverse effects) outcomes to obtain information on meaningful systolic BP targets, which maximizes benefit for both efficacy and safety. The probability was estimated under a Bayesian model with flat priors (ie, similar to frequentist estimates) using 10,000 replicates to minimize Monte Carlo error. Sensitivity analyses were performed based on achieved systolic BP. All analyses were performed using standard software (Stata 14.0; StataCorp LP, College Station, Texas). P <.05 was used to denote statistical significance.

      Results

      Our search yielded 17 randomized trials testing various BP targets (Figure 1, Supplementary Figure 1, available online). The trials enrolled 55,163 patients who were followed up for a mean of 3.7 years for a total of 204,103 patient-years of follow-up. The baseline characteristics of the trials, including baseline BP, achieved BP, and risk of bias assessment of the trials, are outlined in Table 2 and Supplementary Table 2, available online. All included trials used office BP measurements. However, ACCORD, SPRINT, Hypertension Objective Treatment Based on Measurement by Electrical Devices of Blood Pressure (HOMED-BP) trial, Hypertension Optimal Treatment (HOT) trial, Schrier et al,
      • Schrier R.
      • McFann K.
      • Johnson A.
      • et al.
      Cardiac and renal effects of standard versus rigorous blood pressure control in autosomal-dominant polycystic kidney disease: results of a seven-year prospective randomized study.
      Secondary Prevention of Small Subcortical Strokes (SPS3) trial, and UKPDS-38 trials used office BP measurements using an automated BP machine. The VALISH trial did not specify method of measurement of office BP. The rest of the trials used auscultatory method to measure office BP.
      Figure thumbnail gr1
      Figure 1Network of systolic blood pressure target comparisons.
      Table 2Baseline Characteristics of Included Trials
      TrialYearSample SizeCohortBaseline BP (mm Hg)BP Target (Intense vs Standard)Achieved BP (mm Hg) (Intense vs Standard)
      AASK
      • Agodoa L.Y.
      • Appel L.
      • Bakris G.L.
      • et al.
      Effect of ramipril vs amlodipine on renal outcomes in hypertensive nephrosclerosis: a randomized controlled trial.
      • Appel L.J.
      • Wright Jr., J.T.
      • Greene T.
      • et al.
      Intensive blood-pressure control in hypertensive chronic kidney disease.
      • Wright Jr., J.T.
      • Bakris G.
      • Greene T.
      • et al.
      Effect of blood pressure lowering and antihypertensive drug class on progression of hypertensive kidney disease: results from the AASK trial.
      20011094HTN and GFR 20-65 mL/min/1.73 m2 and no other identified causes of renal insufficiency150/95.5MAP ≤92 vs MAP 102-107128/78 vs 141/85
      ABCD (Normotensive)
      • Schrier R.W.
      • Estacio R.O.
      • Esler A.
      • Mehler P.
      Effects of aggressive blood pressure control in normotensive type 2 diabetic patients on albuminuria, retinopathy and strokes.
      • Schrier R.W.
      • Estacio R.O.
      • Mehler P.S.
      • Hiatt W.R.
      Appropriate blood pressure control in hypertensive and normotensive type 2 diabetes mellitus: a summary of the ABCD trial.
      2002354T2DM with normotension (DBP 80-89.9 mm Hg)136.4/84.4DBP reduction by 10 mm Hg vs DBP 80-89128/75 vs 137/81
      ABCD-2 Valsartan
      • Estacio R.O.
      • Coll J.R.
      • Tran Z.V.
      • Schrier R.W.
      Effect of intensive blood pressure control with valsartan on urinary albumin excretion in normotensive patients with type 2 diabetes.
      2006129T2DM with SBP <140 mm Hg, DBP 80-90 mm Hg, and without evidence of overt albuminuria126/84DBP ≤75 vs DBP 80-90118/75 vs 124/80
      ACCORD
      • Cushman W.C.
      • Evans G.W.
      • Byington R.P.
      • et al.
      ACCORD Study Group
      Effects of intensive blood-pressure control in type 2 diabetes mellitus.
      • Chew E.Y.
      • Ambrosius W.T.
      • Davis M.D.
      • et al.
      ACCORD Study Group, ACCORD Eye Study Group
      Effects of medical therapies on retinopathy progression in type 2 diabetes.
      20104733T2DM with age ≥40 y and cardiovascular disease or age ≥55 y with risk factors for cardiovascular disease139.2/76SBP <120 vs SBP <140119.3/64.4 vs 133.5/70.5
      BBB
      • Hannson L.
      The BBB Study: the effect of intensified antihypertensive treatment on the level of blood pressure, side-effects, morbidity and mortality in “well-treated” hypertensive patients. Behandla Blodtryck Battre.
      19942127Treated HTN patients with DBP 90-100 mm Hg155/94.5DBP ≤80 vs DBP 80-100141/83 vs 152/91
      Cardio-Sis
      • Verdecchia P.
      • Staessen J.A.
      • Angeli F.
      • et al.
      Usual versus tight control of systolic blood pressure in non-diabetic patients with hypertension (Cardio-Sis): an open-label randomised trial.
      20091111Nondiabetic patients with SBP 150 mm Hg or greater and at least one additional risk factor163.3/89.6SBP <130 vs SBP <140131.9/77.4 vs 135.6/78.7
      HOMED-BP
      • Asayama K.
      • Ohkubo T.
      • Metoki H.
      • et al.
      Cardiovascular outcomes in the first trial of antihypertensive therapy guided by self-measured home blood pressure.
      20123518Age ≥40 y with mild-to-moderate hypertension151.6/89.95BP <125/80 vs BP 125-134/80-84128.8/76.1 vs 130.4/76.8
      HOT
      • Hansson L.
      • Zanchetti A.
      • Carruthers S.G.
      • et al.
      Effects of intensive blood-pressure lowering and low-dose aspirin in patients with hypertension: principal results of the Hypertension Optimal Treatment (HOT) randomised trial. HOT Study Group.
      199818,790HTN with DBP 100-115 mm Hg169.7/105.4DBP ≤80 vs DBP ≤85 or DBP ≤90139.7/81.1 vs 142.6/84.2
      JATOS
      JATOS Study Group
      Principal results of the Japanese trial to assess optimal systolic blood pressure in elderly hypertensive patients (JATOS).
      20084418Elderly HTN patients aged 65-85 y and SBP >160 mm Hg171.6/89.1SBP <140 vs SBP <160135.9/74.8 vs 145.6/78.1
      MDRD
      • Klahr S.
      • Levey A.S.
      • Beck G.J.
      • et al.
      The effects of dietary protein restriction and blood-pressure control on the progression of chronic renal disease. Modification of Diet in Renal Disease Study Group.
      • Sarnak M.J.
      • Greene T.
      • Wang X.
      • et al.
      The effect of a lower target blood pressure on the progression of kidney disease: long-term follow-up of the modification of diet in renal disease study.
      1994840CKD with serum creatinine 1.4-7 mg/dL in men or 1.2-7 mg/dL in women130.5/80MAP <92 vs MAP <107126.2/76.9 vs 133.8/80.7
      REIN-2
      • Ruggenenti P.
      • Perna A.
      • Loriga G.
      • et al.
      Blood-pressure control for renoprotection in patients with non-diabetic chronic renal disease (REIN-2): multicentre, randomised controlled trial.
      2005335Nondiabetic nephropathy with proteinuria 1-3 g/day and GFR <45 mL/min/1.73 m2 or proteinuria >3 g/day and GFR <70 mL/min/1.73 m2136.7/84.1BP <130/80 vs DBP <90129.6/79.5 vs 133.7/82.3
      Schrier et al
      • Schrier R.
      • McFann K.
      • Johnson A.
      • et al.
      Cardiac and renal effects of standard versus rigorous blood pressure control in autosomal-dominant polycystic kidney disease: results of a seven-year prospective randomized study.
      200275Autosomal dominant polycystic kidney disease patients with HTN, left ventricular hypertrophy, and creatinine clearance >30 mL/min/1.73 m2142.5/95.5BP <120/80 vs BP 135-140/85-90119/77 vs 131/82
      SPRINT
      • Wright Jr., J.T.
      • Williamson J.D.
      • Whelton P.K.
      • et al.
      SPRINT Research Group
      A randomized trial of intensive versus standard blood-pressure control.
      20159361Nondiabetic patients with SBP 130 mm Hg or higher and an increased cardiovascular risk139.7/78.1SBP <120 vs SBP <140121.4/68.7 vs 136.2/76.3
      SPS3
      • Benavente O.R.
      • Coffey C.S.
      • Conwit R.
      • et al.
      SPS3 Study Group
      Blood-pressure targets in patients with recent lacunar stroke: the SPS3 randomised trial.
      • Pergola P.E.
      • White C.L.
      • Szychowski J.M.
      • et al.
      Achieved blood pressures in the secondary prevention of small subcortical strokes (SPS3) study: challenges and lessons learned.
      20133020Age ≥40 y with lacunar stroke143/78.5SBP <130 vs SBP 130–149126.7/69.1 vs 137.4/74.8
      UKPDS-38
      Tight blood pressure control and risk of macrovascular and microvascular complications in type 2 diabetes: UKPDS 38. UK Prospective Diabetes Study Group.
      19981148T2DM with HTN159.3/94BP <150/85 vs BP <180/105144/82 vs 154/87
      VALISH
      • Ogihara T.
      • Saruta T.
      • Rakugi H.
      • et al.
      Target blood pressure for treatment of isolated systolic hypertension in the elderly: valsartan in elderly isolated systolic hypertension study.
      20103260Age ≥70 and ≤85 y with isolated systolic hypertension (SBP >160 mm Hg and DBP <90 mm Hg)169.6/81.4SBP <140 vs SBP ≥140 and <150 mm Hg136.6/74.8 vs 142/76.5
      Wei et al
      • Wei Y.
      • Jin Z.
      • Shen G.
      • et al.
      Effects of intensive antihypertensive treatment on Chinese hypertensive patients older than 70 years.
      2013724Age ≥70 y with BP >150/90 mm Hg or with HTN currently receiving antihypertensive treatment159.5/84.2BP ≤140/90 vs BP ≤150/90135.7/76.2 vs 149.7/82.1
      AASK = African American Study of Kidney Disease and Hypertension; ABCD = Appropriate Blood Pressure Control in Diabetes; ACCORD = Action to Control Cardiovascular Risk in Diabetes; BBB = Behandla Blodtryck Battre; BP = blood pressure; Cardio-Sis = Studio Italiano Sugli Effetti CARDIOvascolari del Controllo della Pressione Arteriosa SIStolica; CKD = chronic kidney disease; DBP = diastolic blood pressure; GFR = glomerular filtration rate; HOMED-BP = Hypertension Objective Treatment Based on Measurement by Electrical Devices of Blood Pressure; HOT = Hypertension Optimal Treatment; HTN = hypertension; JATOS = Japanese Trial to Assess Optimal Systolic Blood Pressure in Elderly Hypertensive Patients; MAP = mean arterial pressure; MDRD = Modification of Diet in Renal Disease; NR = not reported; REIN = Ramipril Efficacy In Nephropathy; SBP = systolic blood pressure; SPRINT = Systolic Blood Pressure Intervention Trial; SPS3 = Secondary Prevention of Small Subcortical Strokes; T2DM = type 2 diabetes mellitus; UKPDS = UK Prospective Diabetes Study Group; VALISH = Valsartan in Elderly Isolated Systolic Hypertension.

      Stroke

      There was a significant decrease in stroke with systolic BP target of <120 mm Hg (vs <160 mm Hg) (RR 0.54; 95% CI, 0.29-1.00), with the point estimate of the RR favoring lower systolic BP targets (<120 mm Hg, <130 mm Hg) when compared with higher systolic BP targets (<140 mm Hg or <150 mm Hg) (Figure 2A). Sensitivity analysis using achieved BP showed a 72%, 97%, and 227% increase in stroke with systolic BP of <140 mm Hg, <150 mm Hg, and <160 mm Hg, respectively, when compared with systolic BP <120 mm Hg (Figure 2B). Systolic BP targets of <120 mm Hg and <130 mm Hg ranked #1 and #2, respectively, for the outcome of stroke (Figure 2C). There was no evidence of small-study effect (Figure 2D) or statistical inconsistency between direct and indirect estimates (χ2 = 5.18; P = .08).
      Figure thumbnail gr2
      Figure 2(A) Network meta-analysis of various systolic blood pressure targets for the outcome of stroke. (B) Sensitivity analysis based on achieved systolic pressure: Outcome of stroke. (C) Hierarchy of the systolic pressure targets (rankogram) using the rank probabilities for the outcome of stroke. (D) Comparison adjusted funnel plot for the outcome of stroke. (A <160 mm Hg; B <150 mm Hg; C <140 mm Hg; D <130 mm Hg; E <120 mm Hg). CI = confidence interval; RR = rate ratio.

      Myocardial Infarction

      There was a significant decrease in myocardial infarction with systolic BP target of <120 mm Hg (vs <160 mm Hg) (RR 0.68; 95% CI, 0.47-1.00), with the point estimate of the RR favoring lower systolic BP targets (<120 mm Hg, <130 mm Hg) when compared with higher systolic BP targets (<140 mm Hg or <150 mm Hg) (Supplementary Figure 2, available online). Systolic BP targets of <120 mm Hg and <130 mm Hg ranked #1 and #2, respectively, for the outcome of myocardial infarction (Figure 3A). There was no evidence of small-study effect (Supplementary Figure 3, available online) or statistical inconsistency between direct and indirect estimates (χ2 = 0.16; P = .69). Sensitivity analysis using achieved systolic BP yielded largely similar results.
      Figure thumbnail gr3
      Figure 3(A) Hierarchy of the systolic pressure targets (rankogram) using the rank probabilities for the outcome of myocardial infarction. (B) Hierarchy of the systolic pressure targets (rankogram) using the rank probabilities for the outcome of death. (C) Hierarchy of the systolic pressure targets (rankogram) using the rank probabilities for the outcome of cardiovascular death. (D) Hierarchy of the systolic pressure targets (rankogram) using the rank probabilities for the outcome of heart failure. (E) Hierarchy of the systolic pressure targets (rankogram) using the rank probabilities for the outcome of serious adverse effects.

      Death

      There was no difference in death when comparing any of the systolic BP targets. However, the point estimate of the RR favored lower systolic BP targets (<120 mm Hg, <130 mm Hg) when compared with higher systolic BP targets (<140 mm Hg or <150 mm Hg) (Supplementary Figure 4, available online). Systolic BP targets of <120 mm Hg and <130 mm Hg ranked #1 and #2, respectively, for the outcome of death (Figure 3B). There was no evidence of small-study effect (Supplementary Figure 5, available online) or statistical inconsistency between direct and indirect estimates (χ2 = 1.83; P = .61). Sensitivity analysis using achieved systolic BP yielded largely similar results.

      Cardiovascular Death

      There was no difference in cardiovascular death when comparing any of the systolic BP targets. However, the point estimate of the RR favored lower systolic BP targets (<120 mm Hg, <130 mm Hg) when compared with higher systolic BP targets (<140 mm Hg or <150 mm Hg) (Supplementary Figure 6, available online). Systolic BP targets of <120 mm Hg and <130 mm Hg ranked #1 and #2, respectively, for the outcome of cardiovascular death (Figure 3C). There was no evidence of small-study effect (Supplementary Figure 7, available online) or statistical inconsistency between direct and indirect estimates (χ2 = 0.47; P = .79). Sensitivity analysis using achieved systolic BP yielded largely similar results.

      Heart Failure

      There was no difference in heart failure when comparing any of the systolic BP targets. However, the point estimate of the RR favored lower systolic BP targets (<120 mm Hg, <130 mm Hg) when compared with higher systolic BP targets (<140 mm Hg or <150 mm Hg) (Supplementary Figure 8, available online). Systolic BP targets of <120 mm Hg and <130 mm Hg ranked #1 and #2, respectively, for the outcome of heart failure (Figure 3D). There was no evidence of small-study effect (Supplementary Figure 9, available online) or statistical inconsistency between direct and indirect estimates (χ2 = 5.63; P = .06). Sensitivity analysis using achieved systolic BP yielded largely similar results.

      Serious Adverse Effects

      There was a significant increase in serious adverse effects with systolic BP target of <120 mm Hg when compared with <150 mm Hg (RR 1.83; 95% CI, 1.05-3.20) or <140 mm Hg (RR 2.12; 95% CI, 1.46-3.08) (Supplementary Figure 10, available online). Systolic BP targets of <140 mm Hg and <150 mm Hg ranked #1 and #2, respectively, for the outcome of serious adverse effects (Figure 3E). There was no evidence of small-study effect (Supplementary Figure 11, available online). The definitions of the serious adverse effects are outlined in the online supplement (Table 1).

      Efficacy and Safety

      Cluster plots for combined efficacy and safety outcomes (Figure 4A-C) showed that a systolic BP target of <130 mm Hg had the best balance between efficacy and safety.
      Figure thumbnail gr4
      Figure 4(A) Cluster plot for stroke vs serious adverse effects. (B) Cluster plot for myocardial infarction vs serious adverse effects (SAE). (C) Cluster plot for death vs serious adverse effects. The higher the surface under the cumulative ranking curve values of a systolic pressure target, the greater is the efficacy or safety. “Star” indicates the ideal systolic pressure target that maximizes outcomes efficacy and safety endpoints. MI = myocardial infarction; SUCRA = surface under the cumulative ranking curve.
      Sensitivity analyses conducted using number of events yielded largely similar results.

      Discussion

      In this analysis of 17 randomized BP target trials that enrolled 55,163 patients with 204,103 patient-years of follow-up, we observed a significant reduction in stroke and myocardial infarction with lower on-treatment systolic BP targets but a significant increase in serious adverse effects with no differences among other outcomes. Systolic BP targets of <120 mm Hg and <130 mm Hg ranked #1 and #2, respectively, as the most efficacious targets, whereas systolic BP targets of <140 mm Hg and <150 mm Hg ranked #1 and #2, respectively, as the safest targets. Cluster plots for combined efficacy and safety outcomes showed that a systolic BP target of <130 mm Hg achieved the best balance between efficacy and safety.

      Blood Pressure Targets

      The BP targets in patients with hypertension or those at high risk of cardiovascular events are controversial. A meta-analysis of 61 prospective observational studies of one million adults with no previous vascular disease demonstrated a linear relationship between BP and ischemic heart disease mortality and stroke mortality down to a BP of 115/75 mm Hg.
      • Lewington S.
      • Clarke R.
      • Qizilbash N.
      • et al.
      Age-specific relevance of usual blood pressure to vascular mortality: a meta-analysis of individual data for one million adults in 61 prospective studies.
      This led to the “lower the better” hypothesis. However, other observational studies have shown a J-curve phenomenon where the risk of cardiovascular events increased above and below a certain BP threshold.
      • Stewart I.M.
      Long-term observations on high blood-pressure presenting in fit young men.
      • Stewart I.M.
      Letter: hypertension and myocardial infarction.
      • Stewart I.M.
      Relation of reduction in pressure to first myocardial infarction in patients receiving treatment for severe hypertension.
      Few other studies have shown a target organ heterogeneity in that lower was better for cerebrovascular events but not for coronary events.
      • Bangalore S.
      • Messerli F.H.
      • Wun C.C.
      • et al.
      J-curve revisited: an analysis of blood pressure and cardiovascular events in the Treating to New Targets (TNT) Trial.
      • Bangalore S.
      • Qin J.
      • Sloan S.
      • et al.
      What is the optimal blood pressure in patients after acute coronary syndromes?: Relationship of blood pressure and cardiovascular events in the PRavastatin OR atorVastatin Evaluation and Infection Therapy-Thrombolysis In Myocardial Infarction (PROVE IT-TIMI) 22 trial.
      Prior to the SPRINT trial, randomized trials such as the ACCORD BP trial failed to reduce the risk of fatal and nonfatal major cardiovascular events with the exception of stroke, with a systolic BP of <120 mm Hg vs <140 mm Hg.
      • Cushman W.C.
      • Evans G.W.
      • Byington R.P.
      • et al.
      ACCORD Study Group
      Effects of intensive blood-pressure control in type 2 diabetes mellitus.
      In the ACCORD BP, the intensive BP-lowering group also required a greater number of antihypertensive medications (mean number of medications 3.4 vs 2.1) and had higher serious adverse events attributed to these medications (3.3% vs 1.3%; P <.001). A meta-analysis of 37,736 participants in 13 randomized clinical trials of hypertensive patients with type 2 diabetes mellitus or impaired fasting glucose also found that lower was better for stroke, even down to a systolic BP of ≤120 mm Hg, but not for other microvascular or macrovascular events. In addition, lower BP was associated with increased risk of serious adverse effects, including hypotension and hyperkalemia.
      • Bangalore S.
      • Kumar S.
      • Lobach I.
      • Messerli F.H.
      Blood pressure targets in subjects with type 2 diabetes mellitus/impaired fasting glucose: observations from traditional and bayesian random-effects meta-analyses of randomized trials.
      A meta-analysis of 66,504 patients in 15 randomized clinical trials of antihypertensive therapy in coronary artery disease patients found that intensive BP control (achieved systolic BP ≤135 mm Hg compared with ≤140 mm Hg) was associated with lower rates of heart failure (15% decrease) and stroke (10% decrease), but increased risk of hypotension (105%), and showed no difference in death or cardiovascular death.
      • Bangalore S.
      • Kumar S.
      • Volodarskiy A.
      • Messerli F.H.
      Blood pressure targets in patients with coronary artery disease: observations from traditional and Bayesian random effects meta-analysis of randomised trials.
      Although a J-shaped curve was not demonstrated, there was no benefit of lowering BP below 135 mm Hg. Thus, in the pre-SPRINT era, the 2014 JNC8 Hypertension Guideline
      • James P.A.
      • Oparil S.
      • Carter B.L.
      • et al.
      2014 evidence-based guideline for the management of high blood pressure in adults: report from the panel members appointed to the Eighth Joint National Committee (JNC 8).
      for individuals <60 years old, the 2013 ESH/ESC hypertension guideline,
      • Mancia G.
      • Fagard R.
      • Narkiewicz K.
      • et al.
      2013 ESH/ESC Guidelines for the management of arterial hypertension: the Task Force for the management of arterial hypertension of the European Society of Hypertension (ESH) and of the European Society of Cardiology (ESC).
      and the National Institute for Health and Clinical Excellence (NICE) hypertension guideline all recommend a target BP of <140/90 mm Hg.
      National Institute for Health and Care Excellence
      Hypertension: clinical management of primary hypertension in adults. NICE guidelines [CG127].
      The SPRINT trial enrolled 9361 participants aged 50 years and older, including a substantial proportion of elderly (≥75 years), those with chronic kidney disease or preexisting cardiovascular disease, and showed a significant benefit of an intensive systolic BP goal of <120 mm Hg with reduction in the composite primary outcome, death from all causes, death from cardiovascular causes, heart failure but with excess of acute renal failure, electrolyte disturbances, hypotension, and syncope, compared with the target systolic BP of <140 mm Hg.
      • Wright Jr., J.T.
      • Williamson J.D.
      • Whelton P.K.
      • et al.
      SPRINT Research Group
      A randomized trial of intensive versus standard blood-pressure control.
      Surprisingly, there was no difference in stroke or myocardial infarction between the 2 groups. Moreover, the intensive BP target group required more medications than the standard-treatment group (2.8 vs 1.8).
      • Wright Jr., J.T.
      • Williamson J.D.
      • Whelton P.K.
      • et al.
      SPRINT Research Group
      A randomized trial of intensive versus standard blood-pressure control.
      In a meta-analysis of more intensive BP-lowering vs less intensive BP lowering, Xie et al
      • Xie X.
      • Atkins E.
      • Lv J.
      • et al.
      Effects of intensive blood pressure lowering on cardiovascular and renal outcomes: updated systematic review and meta-analysis.
      observed that more intensive BP lowering reduced the risk of major cardiovascular events, myocardial infarction, and stroke, but without clear effect on heart failure, cardiovascular death, or total mortality. However, the limitation of this study is that the “more intensive” BP group consisted of a heterogeneous mix of BP targets ranging from a target of <150 mm Hg (as was in the UKPDS-38 trial) to a target of <120 mm Hg (as was in the ACCORD trials).
      • Xie X.
      • Atkins E.
      • Lv J.
      • et al.
      Effects of intensive blood pressure lowering on cardiovascular and renal outcomes: updated systematic review and meta-analysis.
      Our analysis employed a network meta-analysis, which avoided the lumping of heterogeneous BP targets. The study found a benefit of lower systolic BP targets for the reduction of stroke and myocardial infarction but no statistically significant difference for other outcomes (similar to the meta-analysis by Xie et al
      • Xie X.
      • Atkins E.
      • Lv J.
      • et al.
      Effects of intensive blood pressure lowering on cardiovascular and renal outcomes: updated systematic review and meta-analysis.
      ). However, ranking probabilities indicated that systolic BP targets of <120 mm Hg and <130 mm Hg ranked #1 and #2, respectively, for all of these outcomes. Although lower systolic BP targets were efficacious, they were associated with a significant increase in serious adverse effects in that the systolic BP targets of <140 mm Hg and <150 mm Hg ranked #1 and #2, respectively, as the safest target. Cluster plots for combined efficacy and safety showed that a systolic BP target of <130 mm Hg yielded the best balance between efficacy and safety. Our analyses, therefore, provide a scorecard for efficacy and safety for various BP targets. Unlike other meta-analyses, we excluded trials that were not BP target strategy trials, that is, those that did not test treating to a BP target.

      Study Limitations

      Our study was based on trial-level data and therefore was not able to control for differences between trials, including differences in how BP was measured or the cohort enrolled. We also were not able to control for the medications used or the dosage of medications used. Although the SPRINT trial excluded patients with diabetes, the other trials had adequate representation of patients with diabetes. Moreover, although it is desirable to evaluate individual side effects that constitute serious adverse effect in the composite, the individual side effects were not reported consistently in all of the trials. Although a systolic BP target of <130 mm Hg was found to have the optimal balance between efficacy and safety, the study does not offer insights into desired targets for individual groups (such as the elderly, those with diabetes, or those with chronic kidney disease). In addition, we tested optimal target for systolic BP and not for diastolic BP, as the recent trials such as ACCORD and SPRINT were designed to evaluate systolic BP targets. We observed statistically significant difference only for a few comparisons in clinical outcomes assessed. It is likely that the current available data from randomized trials are underpowered for individual endpoints. Although ranking probabilities can provide a scorecard for efficacy and safety of each BP target, these have to be interpreted cautiously when the treatment effects are, in fact, not different from the null beyond chance, as was seen for most outcomes. The current data from randomized trials of BP targets therefore expose the lack of precision (likely due to being underpowered) to choose a BP target for outcomes other than stroke. In addition, although the SPRINT trial showed a mortality benefit for intensive BP target when compared with <140 mm Hg, our estimate for the outcome of death showed no difference between <120 mm Hg vs <140 mm Hg (RR 0.89; 95% CI, 0.63-1.24) for this outcome when all trials are included, raising concerns about over-interpretation of endpoints not powered for in randomized trials (like SPRINT), especially when they are terminated prematurely.

      Conclusions

      Data from randomized BP target strategy trials suggest a significant reduction in stroke and myocardial infarction with lower systolic BP targets, such that systolic BP targets of <120 mm Hg and <130 mm Hg ranked #1 and #2, respectively, as the most efficacious targets. However, lower systolic BP targets were associated with a significant increase in serious adverse effects such that higher systolic BP targets of <140 mm Hg and <150 mm Hg ranked #1 and #2, respectively, as the safest targets. Cluster plots for combined efficacy and safety outcomes showed that a on-treatment systolic BP target of <130 mm Hg achieved the optimal balance between efficacy and safety. The current body of evidence from randomized trials of BP target trials are likely underpowered for individual outcomes, with nonsignificant difference for most outcomes.

      Appendix

      This Appendix has been provided by the authors to give readers additional information about their work.
      Supplementary Table 1Details of the Search Terms
      Search Terms Used
      Hypertension; Antihypertensive agents; Angiotensin-Converting Enzyme Inhibitors; Angiotensin receptor antagonists; Diuretics; Thiazides; Calcium channel blockers; Adrenergic alpha-antagonists; Adrenergic beta-antagonists; Vasodilator agents; Ganglionic blockers; Randomized controlled trials; Humans
      Supplementary Table 2Characteristics of the Included Clinical Trials
      TrialAge, yMale (%)Black (%)DM (%)CKD (%)CAD (%)Follow-Up (y)Quality
      Represents risk of bias based on: sequence generation of allocation; allocation concealment and blinding. + represents low bias risk, and ± unclear bias risk.
      AASK
      • Agodoa L.Y.
      • Appel L.
      • Bakris G.L.
      • et al.
      Effect of ramipril vs amlodipine on renal outcomes in hypertensive nephrosclerosis: a randomized controlled trial.
      • Appel L.J.
      • Wright Jr., J.T.
      • Greene T.
      • et al.
      Intensive blood-pressure control in hypertensive chronic kidney disease.
      • Wright Jr., J.T.
      • Bakris G.
      • Greene T.
      • et al.
      Effect of blood pressure lowering and antihypertensive drug class on progression of hypertensive kidney disease: results from the AASK trial.
      55611000100NR3-6.4+±+
      ABCD (Normotensive)
      • Schrier R.W.
      • Estacio R.O.
      • Esler A.
      • Mehler P.
      Effects of aggressive blood pressure control in normotensive type 2 diabetic patients on albuminuria, retinopathy and strokes.
      • Schrier R.W.
      • Estacio R.O.
      • Mehler P.S.
      • Hiatt W.R.
      Appropriate blood pressure control in hypertensive and normotensive type 2 diabetes mellitus: a summary of the ABCD trial.
      59557100NR245.3±±+
      ABCD-2 Valsartan
      • Estacio R.O.
      • Coll J.R.
      • Tran Z.V.
      • Schrier R.W.
      Effect of intensive blood pressure control with valsartan on urinary albumin excretion in normotensive patients with type 2 diabetes.
      566781002151.9±±+
      ACCORD
      • Cushman W.C.
      • Evans G.W.
      • Byington R.P.
      • et al.
      ACCORD Study Group
      Effects of intensive blood-pressure control in type 2 diabetes mellitus.
      • Chew E.Y.
      • Ambrosius W.T.
      • Davis M.D.
      • et al.
      ACCORD Study GroupACCORD Eye Study Group
      Effects of medical therapies on retinopathy progression in type 2 diabetes.
      625224100NR344.7+++
      BBB
      • Hannson L.
      The BBB Study: the effect of intensified antihypertensive treatment on the level of blood pressure, side-effects, morbidity and mortality in “well-treated” hypertensive patients. Behandla Blodtryck Battre.
      6053NRNRNR04.9±±+
      Cardio-Sis
      • Verdecchia P.
      • Staessen J.A.
      • Angeli F.
      • et al.
      Usual versus tight control of systolic blood pressure in non-diabetic patients with hypertension (Cardio-Sis): an open-label randomised trial.
      6741000122+±+
      HOMED-BP
      • Asayama K.
      • Ohkubo T.
      • Metoki H.
      • et al.
      Cardiovascular outcomes in the first trial of antihypertensive therapy guided by self-measured home blood pressure.
      6050NR15NR35.3+±+
      HOT
      • Hansson L.
      • Zanchetti A.
      • Carruthers S.G.
      • et al.
      Effects of intensive blood-pressure lowering and low-dose aspirin in patients with hypertension: principal results of the Hypertension Optimal Treatment (HOT) randomised trial. HOT Study Group.
      6253NR8NR83.8+±+
      JATOS
      JATOS Study Group
      Principal results of the Japanese trial to assess optimal systolic blood pressure in elderly hypertensive patients (JATOS).
      7439NR121032+±+
      MDRD
      • Klahr S.
      • Levey A.S.
      • Beck G.J.
      • et al.
      The effects of dietary protein restriction and blood-pressure control on the progression of chronic renal disease. Modification of Diet in Renal Disease Study Group.
      • Sarnak M.J.
      • Greene T.
      • Wang X.
      • et al.
      The effect of a lower target blood pressure on the progression of kidney disease: long-term follow-up of the modification of diet in renal disease study.
      526185100NR2.2±±±
      REIN-2
      • Ruggenenti P.
      • Perna A.
      • Loriga G.
      • et al.
      Blood-pressure control for renoprotection in patients with non-diabetic chronic renal disease (REIN-2): multicentre, randomised controlled trial.
      5475NRNR100NR1.6±±±
      Schrier et al
      • Schrier R.
      • McFann K.
      • Johnson A.
      • et al.
      Cardiac and renal effects of standard versus rigorous blood pressure control in autosomal-dominant polycystic kidney disease: results of a seven-year prospective randomized study.
      4155NRNRNRNR7+±+
      SPRINT
      • Wright Jr., J.T.
      • Williamson J.D.
      • Whelton P.K.
      • et al.
      SPRINT Research Group
      A randomized trial of intensive versus standard blood pressure control.
      686430028203.3+±+
      SPS3
      • Benavente O.R.
      • Coffey C.S.
      • Conwit R.
      • et al.
      SPS3 Study Group
      Blood-pressure targets in patients with recent lacunar stroke: the SPS3 randomised trial.
      • Pergola P.E.
      • White C.L.
      • Szychowski J.M.
      • et al.
      Achieved blood pressures in the secondary prevention of small subcortical strokes (SPS3) study: challenges and lessons learned.
      63631736NR113.7+±+
      UKPDS-38
      Tight blood pressure control and risk of macrovascular and microvascular complications in type 2 diabetes: UKPDS 38. UK Prospective Diabetes Study Group.
      56557100NRNR8.4±++
      VALISH
      • Ogihara T.
      • Saruta T.
      • Rakugi H.
      • et al.
      Target blood pressure for treatment of isolated systolic hypertension in the elderly: valsartan in elderly isolated systolic hypertension study.
      7637NR13NR52.9+±+
      Wei et al
      • Wei Y.
      • Jin Z.
      • Shen G.
      • et al.
      Effects of intensive antihypertensive treatment on Chinese hypertensive patients older than 70 years.
      7766NR23NR84+±+
      AASK = African American Study of Kidney Disease and Hypertension; ABCD = Appropriate Blood Pressure Control in Diabetes; ACCORD = Action to Control Cardiovascular Risk in Diabetes; BBB = Behandla Blodtryck Battre; BP = Blood pressure; Cardio-Sis = Studio Italiano Sugli Effetti CARDIOvascolari del Controllo della Pressione Arteriosa SIStolica; DBP = Diastolic blood pressure; HOMED-BP = Hypertension Objective Treatment Based on Measurement by Electrical Devices of Blood Pressure; HOT = Hypertension Optimal Treatment; HTN = hypertension; JATOS = Japanese Trial to Assess Optimal Systolic Blood Pressure in Elderly Hypertensive Patients; MAP = Mean arterial pressure; MDRD = Modification of Diet in Renal Disease; REIN = Ramipril Efficacy In Nephropathy; SBP = Systolic blood pressure; SPRINT = Systolic Blood Pressure Intervention Trial; SPS3 = Secondary Prevention of Small Subcortical Strokes; T2DM = type 2 diabetes mellitus; UKPDS = UK Prospective Diabetes Study Group; VALISH = Valsartan in Elderly Isolated Systolic Hypertension; CKD = chronic kidney disease; NR = not reported.
      Represents risk of bias based on: sequence generation of allocation; allocation concealment and blinding. + represents low bias risk, and ± unclear bias risk.
      Figure thumbnail fx1
      Supplementary Figure 1Study selection. Records identified through database search using terms “hypertension,” “antihypertensive agents,” and individual antihypertensive classes; limited to randomized controlled trial, Humans (n = 22,476). DBP = diastolic blood pressure; MAP = mean arterial pressure.
      Figure thumbnail fx2
      Supplementary Figure 2Network meta-analysis of various systolic blood pressure targets for the outcome of myocardial infarction. CI = confidence interval; RR = rate ratio.
      Figure thumbnail fx3
      Supplementary Figure 3Comparison adjusted funnel plot for the outcome of myocardial infarction. (A <160 mm Hg; B <150 mm Hg; C <140 mm Hg; D <130 mm Hg; E <120 mm Hg).
      Figure thumbnail fx4
      Supplementary Figure 4Network meta-analysis of various systolic blood pressure targets for the outcome of death. CI = confidence interval; RR = rate ratio.
      Figure thumbnail fx5
      Supplementary Figure 5Comparison adjusted funnel plot for the outcome of death. (A <160 mm Hg; B <150 mm Hg; C <140 mm Hg; D <130 mm Hg; E <120 mm Hg).
      Figure thumbnail fx6
      Supplementary Figure 6Network meta-analysis of various systolic blood pressure targets for the outcome of cardiovascular death. CI = confidence interval; RR = rate ratio.
      Figure thumbnail fx7
      Supplementary Figure 7Comparison adjusted funnel plot for the outcome of cardiovascular death. (A <160 mm Hg; B <150 mm Hg; C <140 mm Hg; D <130 mm Hg; E <120 mm Hg).
      Figure thumbnail fx8
      Supplementary Figure 8Network meta-analysis of various systolic blood pressure targets for the outcome of heart failure. CI = confidence interval; RR = rate ratio.
      Figure thumbnail fx9
      Supplementary Figure 9Comparison adjusted funnel plot for the outcome of heart failure. (A <160 mm Hg; B <150 mm Hg; C <140 mm Hg; D <130 mm Hg; E <120 mm Hg).
      Figure thumbnail fx10
      Supplementary Figure 10Network meta-analysis of various systolic blood pressure targets for the outcome of serious adverse events. CI = confidence interval; RR = rate ratio.
      Figure thumbnail fx11
      Supplementary Figure 11Comparison adjusted funnel plot for the outcome of serious adverse effects. (A <160 mm Hg; B <150 mm Hg; C <140 mm Hg; D <130 mm Hg; E <120 mm Hg).

      References

        • Kjeldsen S.E.
        • Narkiewicz K.
        • Hedner T.
        • Mancia G.
        The SPRINT study: outcome may be driven by difference in diuretic treatment demasking heart failure and study design may support systolic blood pressure target below 140 mmHg rather than below 120 mmHg.
        Blood Press. 2016; 25: 63-66
        • Mancia G.
        The SPRINT Trial: Cons.
        2015 (Available at:) (Accessed January 12, 2016)
        • Wright Jr., J.T.
        • Williamson J.D.
        • Whelton P.K.
        • et al.
        • SPRINT Research Group
        A randomized trial of intensive versus standard blood-pressure control.
        N Engl J Med. 2015; 373: 2103-2116
        • Cushman W.C.
        • Evans G.W.
        • Byington R.P.
        • et al.
        • ACCORD Study Group
        Effects of intensive blood-pressure control in type 2 diabetes mellitus.
        N Engl J Med. 2010; 362: 1575-1585
        • Xie X.
        • Atkins E.
        • Lv J.
        • et al.
        Effects of intensive blood pressure lowering on cardiovascular and renal outcomes: updated systematic review and meta-analysis.
        Lancet. 2016; 387: 435-443
      1. Higgins J, Green S.Cochrane Handbook for Systematic Reviews of Interventions Version 5.0.0 ed: The Cochrane Collaboration; 2008. Available at: www.cochrane-handbook.org. Accessed August 1, 2016.

        • Agodoa L.Y.
        • Appel L.
        • Bakris G.L.
        • et al.
        Effect of ramipril vs amlodipine on renal outcomes in hypertensive nephrosclerosis: a randomized controlled trial.
        JAMA. 2001; 285: 2719-2728
        • Appel L.J.
        • Wright Jr., J.T.
        • Greene T.
        • et al.
        Intensive blood-pressure control in hypertensive chronic kidney disease.
        N Engl J Med. 2010; 363: 918-929
        • Wright Jr., J.T.
        • Bakris G.
        • Greene T.
        • et al.
        Effect of blood pressure lowering and antihypertensive drug class on progression of hypertensive kidney disease: results from the AASK trial.
        JAMA. 2002; 288: 2421-2431
        • Schrier R.W.
        • Estacio R.O.
        • Esler A.
        • Mehler P.
        Effects of aggressive blood pressure control in normotensive type 2 diabetic patients on albuminuria, retinopathy and strokes.
        Kidney Int. 2002; 61: 1086-1097
        • Schrier R.W.
        • Estacio R.O.
        • Mehler P.S.
        • Hiatt W.R.
        Appropriate blood pressure control in hypertensive and normotensive type 2 diabetes mellitus: a summary of the ABCD trial.
        Nat Clin Pract Nephrol. 2007; 3: 428-438
        • Estacio R.O.
        • Coll J.R.
        • Tran Z.V.
        • Schrier R.W.
        Effect of intensive blood pressure control with valsartan on urinary albumin excretion in normotensive patients with type 2 diabetes.
        Am J Hypertens. 2006; 19: 1241-1248
        • Chew E.Y.
        • Ambrosius W.T.
        • Davis M.D.
        • et al.
        • ACCORD Study Group, ACCORD Eye Study Group
        Effects of medical therapies on retinopathy progression in type 2 diabetes.
        N Engl J Med. 2010; 363: 233-244
        • Hannson L.
        The BBB Study: the effect of intensified antihypertensive treatment on the level of blood pressure, side-effects, morbidity and mortality in “well-treated” hypertensive patients. Behandla Blodtryck Battre.
        Blood Press. 1994; 3: 248-254
        • Verdecchia P.
        • Staessen J.A.
        • Angeli F.
        • et al.
        Usual versus tight control of systolic blood pressure in non-diabetic patients with hypertension (Cardio-Sis): an open-label randomised trial.
        Lancet. 2009; 374: 525-533
        • Asayama K.
        • Ohkubo T.
        • Metoki H.
        • et al.
        Cardiovascular outcomes in the first trial of antihypertensive therapy guided by self-measured home blood pressure.
        Hypertens Res. 2012; 35: 1102-1110
        • Hansson L.
        • Zanchetti A.
        • Carruthers S.G.
        • et al.
        Effects of intensive blood-pressure lowering and low-dose aspirin in patients with hypertension: principal results of the Hypertension Optimal Treatment (HOT) randomised trial. HOT Study Group.
        Lancet. 1998; 351: 1755-1762
        • JATOS Study Group
        Principal results of the Japanese trial to assess optimal systolic blood pressure in elderly hypertensive patients (JATOS).
        Hypertens Res. 2008; 31: 2115-2127
        • Klahr S.
        • Levey A.S.
        • Beck G.J.
        • et al.
        The effects of dietary protein restriction and blood-pressure control on the progression of chronic renal disease. Modification of Diet in Renal Disease Study Group.
        N Engl J Med. 1994; 330: 877-884
        • Sarnak M.J.
        • Greene T.
        • Wang X.
        • et al.
        The effect of a lower target blood pressure on the progression of kidney disease: long-term follow-up of the modification of diet in renal disease study.
        Ann Intern Med. 2005; 142: 342-351
        • Ruggenenti P.
        • Perna A.
        • Loriga G.
        • et al.
        Blood-pressure control for renoprotection in patients with non-diabetic chronic renal disease (REIN-2): multicentre, randomised controlled trial.
        Lancet. 2005; 365: 939-946
        • Schrier R.
        • McFann K.
        • Johnson A.
        • et al.
        Cardiac and renal effects of standard versus rigorous blood pressure control in autosomal-dominant polycystic kidney disease: results of a seven-year prospective randomized study.
        J Am Soc Nephrol. 2002; 13: 1733-1739
        • Benavente O.R.
        • Coffey C.S.
        • Conwit R.
        • et al.
        • SPS3 Study Group
        Blood-pressure targets in patients with recent lacunar stroke: the SPS3 randomised trial.
        Lancet. 2013; 382: 507-515
        • Pergola P.E.
        • White C.L.
        • Szychowski J.M.
        • et al.
        Achieved blood pressures in the secondary prevention of small subcortical strokes (SPS3) study: challenges and lessons learned.
        Am J Hypertens. 2014; 27: 1052-1060
      2. Tight blood pressure control and risk of macrovascular and microvascular complications in type 2 diabetes: UKPDS 38. UK Prospective Diabetes Study Group.
        BMJ. 1998; 317: 703-713
        • Ogihara T.
        • Saruta T.
        • Rakugi H.
        • et al.
        Target blood pressure for treatment of isolated systolic hypertension in the elderly: valsartan in elderly isolated systolic hypertension study.
        Hypertension. 2010; 56: 196-202
        • National Institute for Health and Care Excellence
        Hypertension: clinical management of primary hypertension in adults. NICE guidelines [CG127].
        (Available at:) (Accessed January 1, 2016)
        • Wei Y.
        • Jin Z.
        • Shen G.
        • et al.
        Effects of intensive antihypertensive treatment on Chinese hypertensive patients older than 70 years.
        J Clin Hypertens (Greenwich). 2013; 15: 420-427
        • Higgins J.P.T.
        • Jackson D.
        • Barrett J.K.
        • Lu G.
        • Ades A.E.
        • White I.R.
        Consistency and inconsistency in network meta-analysis: concepts and models for multi-arm studies.
        Res Synth Methods. 2012; 3: 98-110
        • White I.R.
        • Barrett J.K.
        • Jackson D.
        • Higgins J.P.
        Consistency and inconsistency in network meta-analysis: model estimation using multivariate meta-regression.
        Res Synth Methods. 2012; 3: 111-125
        • White I.
        Multivariate random-effects meta-analysis.
        Stata J. 2009; 9: 40-56
        • Chaimani A.
        • Higgins J.P.
        • Mavridis D.
        • et al.
        Graphical tools for network meta-analysis in STATA.
        PLoS One. 2013; 8: e76654
        • Salanti G.
        • Ades A.E.
        • Ioannidis J.P.
        Graphical methods and numerical summaries for presenting results from multiple-treatment meta-analysis: an overview and tutorial.
        J Clin Epidemiol. 2011; 64: 163-171
        • Lewington S.
        • Clarke R.
        • Qizilbash N.
        • et al.
        Age-specific relevance of usual blood pressure to vascular mortality: a meta-analysis of individual data for one million adults in 61 prospective studies.
        Lancet. 2002; 360: 1903-1913
        • Stewart I.M.
        Long-term observations on high blood-pressure presenting in fit young men.
        Lancet. 1971; 1: 355-358
        • Stewart I.M.
        Letter: hypertension and myocardial infarction.
        Br Med J. 1974; 3: 251
        • Stewart I.M.
        Relation of reduction in pressure to first myocardial infarction in patients receiving treatment for severe hypertension.
        Lancet. 1979; 1: 861-865
        • Bangalore S.
        • Messerli F.H.
        • Wun C.C.
        • et al.
        J-curve revisited: an analysis of blood pressure and cardiovascular events in the Treating to New Targets (TNT) Trial.
        Eur Heart J. 2010; 31: 2897-2908
        • Bangalore S.
        • Qin J.
        • Sloan S.
        • et al.
        What is the optimal blood pressure in patients after acute coronary syndromes?: Relationship of blood pressure and cardiovascular events in the PRavastatin OR atorVastatin Evaluation and Infection Therapy-Thrombolysis In Myocardial Infarction (PROVE IT-TIMI) 22 trial.
        Circulation. 2010; 122: 2142-2151
        • Bangalore S.
        • Kumar S.
        • Lobach I.
        • Messerli F.H.
        Blood pressure targets in subjects with type 2 diabetes mellitus/impaired fasting glucose: observations from traditional and bayesian random-effects meta-analyses of randomized trials.
        Circulation. 2011; 123: 2799-2810
        • Bangalore S.
        • Kumar S.
        • Volodarskiy A.
        • Messerli F.H.
        Blood pressure targets in patients with coronary artery disease: observations from traditional and Bayesian random effects meta-analysis of randomised trials.
        Heart. 2013; 99: 601-613
        • James P.A.
        • Oparil S.
        • Carter B.L.
        • et al.
        2014 evidence-based guideline for the management of high blood pressure in adults: report from the panel members appointed to the Eighth Joint National Committee (JNC 8).
        JAMA. 2014; 311: 507-520
        • Mancia G.
        • Fagard R.
        • Narkiewicz K.
        • et al.
        2013 ESH/ESC Guidelines for the management of arterial hypertension: the Task Force for the management of arterial hypertension of the European Society of Hypertension (ESH) and of the European Society of Cardiology (ESC).
        J Hypertens. 2013; 31: 1281-1357