Hypertension Journal

Show Contents

Will the Systolic Blood Pressure Intervention Trial(SPRINT) change Treatment Targets in Hypertension?
  JOHTN
COMMENTARY
Will the Systolic Blood Pressure Intervention Trial
(SPRINT) change Treatment Targets in Hypertension?
Michael A Weber
Professor
Department of Cardiovascular Medicine, State University of NewYork, Downstate College of Medicine, Brooklyn, New York, USA
Correspondence Author: Michael A Weber, Professor, Departmentof Cardiovascular Medicine, State University of New YorkDownstate College of Medicine, Brooklyn, New York, USA
Phone: +7148157430
e-mail: michaelwebermd@cs.com
 
ABSTRACT
Most of the current guidelines recommend a target systolic bloodpressure (SBP) of < 140 mm Hg when treating hypertension.However, in the recent systolic blood pressure intervention trial(SPRINT) study which compared targets of 140 and 120 mm Hgin patients at high cardiovascular risk and concluded that therewere greater cardiovascular (CV) and mortality benefits at thelower target, including those patients aged above 75 years.The blood pressure (BP) measurement method in SPRINTwas unconventional and the achieved SBP in the intensivelytreated patients corresponds to pressure in an orifice valve ofabout 130 mm Hg. Importantly, SPRINT excluded patients withdiabetes or at low CV risk. These patients should be treated to< 140 mm Hg since CV events may increase at more aggressivetargets. In communities with limited resources it is often not possibleto evaluate CV risk and < 140 mm Hg should be the usualtarget; indeed, in Stage 1 hypertension, drug therapy can bedelayed while lifestyle changes are tried.
Keywords: Cardiovascular risk, Hypertension guidelines, Systolicblood pressure, Systolic blood pressure intervention trial.
How to cite this article: Weber MA. Will the Systolic BloodPressure Intervention Trial (SPRINT) change Treatment Targetsin Hypertension? Hypertens J 2016;2(4):183-188.
Source of support: Nil
Conflict of interest: None
 
 

INTRODUCTION

The results of the systolic blood pressure interventiontrial (SPRINT) have already impacted the field of hypertension.1 This trial compared intensive with standardblood pressure (BP) targets in hypertensive patientsat high cardiovascular (CV) risk and concluded that atarget systolic BP of < 120 mm Hg was superior to a targetof < 140 mm Hg in preventing major CV outcomes andmortality. We will discuss further details of SPRINT laterin this brief commentary, but it was evident as soon asthese results were announced and publicized in the laymedia that the practice of clinical hypertension and thewriting of formal hypertension practice guidelines wouldbe influenced by this authoritative study.

 
BACKGROUND

For several years, at least since the mid-1990s, mostguidelines have recommended 140/90 mm Hg as thethreshold for diagnosing hypertension and that achievingBPs below this threshold should be the appropriatetarget of treatment.2,3 Although this widely used targetwas not based on any compelling randomized trial, theweight of evidence appeared to support a systolic targetof around 140 mm Hg for most adults.

One of the most influential studies in guiding thisrecommendation was the Systolic Hypertension in theElderly Program in which active treatment, when comparedwith placebo, significantly reduced stroke andcoronary events in patients aged 60 or older who hadsystolic BPs ≥ 160 mm Hg.4 In fact, the actively treatedgroup finished with a mean systolic BP of 143 mm Hg,and the placebo group finished with 155 mm Hg. Sincethere appeared to be few if any safety issues with theactive treatment, most experts felt comfortable recommendinga target of < 140 mm Hg. As an interestingsidelight on this assumption, as recently as 2013, theso-called JNC 8 Report reinterpreted these data and recommendedthat, for patients aged 60 or more, a targetof < 150 mm Hg would be adequate.5 Not surprisingly,this recommendation was greeted with considerablequestioning and uncertainty, but since the results ofSPRINT have become available, it is now understoodthat the JNC 8 recommendation has doubtful validity.

A number of clinical trials, although not originallyintended for this purpose, have supported the systolictarget of < 140 mm Hg. Figure 1 represents an example ofthese data. In the valsartan antihypertensive long-termuse evaluation trial, where investigators studying newBP medications were encouraged to achieve systolic BPs< 140 mm Hg, an analysis of outcomes in the patientsachieving this goal, compared with those who did not,demonstrated powerful stroke, mortality, and cardiacbenefits of this target.6

A DIRECT COMPARISON OF TARGETS

The action to control cardiovascular risk in diabetes(ACCORD) trial, in some ways designed similarly to thelater SPRINT trial, compared outcomes in hypertensivediabetic patients randomized to systolic BPs of < 120 and< 140.7 A major difference between ACCORD and SPRINT was that ACCORD was performed entirely in hypertensivepatients with type II diabetes, whereas in SPRINT patientswith diabetes were excluded. The primary endpoint inACCORD was the composite of CV mortality and nonfatalmyocardial infarction (MI) or stroke. This outcome was notsignificantly different between the two treatment groups,but - of considerable interest - the single outcome of strokewas significantly lower in the intensively treated group.7This finding, in the eyes of many clinicians, was importantsince stroke is a feared outcome of hypertension, althoughthe failure of other outcomes to benefit from aggressivetherapy did raise some questions about aggressive treatmenttargets in diabetes.

Hypertension Journal, October-December, Vol 2, 2016 183

Michael A Weber

Will the Systolic Blood Pressure Intervention Trial(SPRINT) change Treatment Targets in Hypertension?
Fig. 1: Effective BP control (SBP < 140 mm Hg) reduces cardiovascularrisk (VALUE trial)

To some extent these concerns were amplified bysubsequent analyses of data from other clinical trials.A recent publication based on the avoiding cardiovascularevents through combination therapy in patientsliving with systolic hypertension (ACCOMPLISH) trialincluded roughly equal proportions of diabetic and nondiabeticpatients.8 Using the same composite outcomeas ACCORD, patients with diabetes in ACCOMPLISHclearly had a significant benefit when their systolic BPwas reduced below 140 mm Hg when compared with≥ 140 mm Hg; further reduction to < 130 mm Hg didnot provide further benefit, and in patients achieving< 120 mm Hg there was a trend to a worsening of thisprimary endpoint, driven primarily by a significantincrease in coronary events.8 This finding, together withdata from other studies in diabetes,9 seems to indicate thatthere could well be a J-curve phenomenon for systolic BPin treating patients with diabetes. Interestingly, and consistentwith the findings from SPRINT, the nondiabeticcohort in ACCOMPLISH appeared to get progressivelygreater CV protection with reductions in BP to around120 mm Hg.8 It would be unwise to overinterpret thesesecondary analyses from ACCOMPLISH, but they helpprovide a useful perspective in understanding that thedifferences in results between ACCORD and SPRINT may very well represent true outcomes differences betweendiabetic and nondiabetic patients when treated aggressivelyfor hypertension.
 
SPRINT Findings

As mentioned at the start of this Commentary, SPRINTdemonstrated that a more aggressive systolic BP targetwas superior to a standard target in preventing fatal andnonfatal outcomes in high risk hypertensive patients.This result, however, must be looked at with care. Table 1gives some of the important underlying data in thistrial. For a start, it is critical to understand how BPs weremeasured. In an attempt to minimize the so-called whitecoat effect, the SPRINT investigators used a novel methodfor measuring BP. In brief, patients were seated in a quietroom with the cuff of an automated BP-measuring deviceattached to the patient's arm. At that point, all medicalpersonnel left the room, and after a 5-minute rest periodthe device automatically measured three BPs at 1 minuteintervals. The average of these three BPs became the officialreading. This type of "unobserved" BP measurementhas not previously been used in a major outcomes trial,and so this has created difficulty in accurately interpretingthe SPRINT BP values in terms of conventionaloffice-measured BPs. There is no question that carefullymeasured office BPs, even when using good qualityautomated devices, are higher than the values obtainedby the SPRINT method. It is reasonable to assume thatthis difference is roughly 7 to 10 mm Hg.

So, as shown in Table 1, the achieved systolicBP in SPRINT for the intensive treatment group of121.5 mm Hg becomes approximately 129 mm Hg by"conventional" office criteria; whereas the SPRINT valueof 134.6 mm Hg in the standard treatment group shouldbe regarded as approximately 142 mm Hg in the office.So, most experts are already re-interpreting the BPresults of SPRINT and concluding that a target of around130 mm Hg is probably the desirable goal in high-risk hypertensive patients. Moreover, there is a concern thattreating down to < 120 mm Hg by conventional officereadings may be taking patients to levels of BP that inactuality are lower and with a potential for harm thanthe values reported by SPRINT.

184

Will Recent Trials change the Systolic Treatment Targets?

Table 1: SPRINT: In nondiabetic medium/high-risk hypertensivepatients, is there an outcomes benefit in achieving < 120 vs< 140 mm Hg?
Will the Systolic Blood Pressure Intervention Trial(SPRINT) change Treatment Targets in Hypertension?
*Mean during treatment period using SPRINT unobserved BPmethod
**Adjusted by daytime ABPM difference of approximately 7 mm Hg(other adjustments have also been considered)


One other important note about SPRINT is also inTable 1. On average, patients in the intensively treatedgroup received one more drug than those in the standardgroup. As discussed later, this difference in theintensity of drug therapy could have outcomes effectsindependent of the differences between the two treatmentgroups in BP.

Interpreting SPRINT

The detailed results from SPRINT have been widelydiscussed and I shall not get into much detail here. Theprimary outcome was a composite of coronary, stroke,heart failure, and CV mortality outcomes. This outcomewas reduced by 25% in the intensively treated group, apowerful and compelling result. The two most impressivecomponents were heart failure, which was reducedby 38%, and CV death, which was reduced by 43%. Totalmortality, which was a secondary endpoint, was reducedby a remarkable 27%, a result that in essence put to restany debate about other individual outcomes in SPRINT.These benefits of intensive treatment were just as compellingin the subgroup of patients aged 75 or more, andappeared to benefit these older patients regardless oftheir frailty status.10

Medication Issue

Investigators in SPRINT were free to select whicheverdrugs they preferred in order to help patients achievetheir target systolic BPs of < 120 or < 140 mm Hg in thisopen-label study. Even though patients upon study entrywere randomized to either the intensive or standardtreatment groups, the investigators were immediatelyinformed of each patient's assignment so that appropriateintensities of treatment could be administered. Table 2lists some of the major drug types that were used during the SPRINT study.1 It is quite clear that more drugswere used in the intensive treatment group than in thestandard treatment group. Quite remarkably, about 1,000more patients in the intensive group received diuretics,and a similarly greater number received blockers ofthe renin angiotensin system angiotensin convertingenzyme inhibitors or angiotensin receptor blockers. Evenagents like spironolactone and beta blockers were morecommon. So, as asked in the table, could the differencein heart failure events between the two groups - actuallynumbering just 38 individual patients - be explained bythe far greater use of drugs that, quite apart from theirBP-lowering effects, are known to be highly effective intreating the symptoms and improving the prognosisof patients with heart failure? Although not shown inthis table, roughly 1,000 more patients in the intensivelytreated group received calcium channel blockers which,together with the drug classes already mentioned, couldalso have contributed to the mortality benefits observedwith intensive treatment. These questions deserve to becarefully considered because while there is no questionthat the intensive treatment protocol in SPRINT hadpowerful CV and mortality benefits, it may be importantin future analyses of the SPRINT database, or in furthertrials designed to explore these issues, to learn more aboutoptimal drug combinations that - beyond BP reduction -can optimize the benefits of hypertension therapy.

 
Table 2: SPRINT: Drugs that treat or mask heart failure
Will the Systolic Blood Pressure Intervention Trial(SPRINT) change Treatment Targets in Hypertension?
Question: Can 38 fewer heart failure events be explained by 1,005more patients on thiazides, 998 on RAS blockers, 220 on spiro,and 479 on BBs - independent of blood pressure effects?

A Need for Caution

Two recent clinical trials provide useful perspectives onthe SPRINT data and serve to warn us that being excessivelyaggressive in treating hypertension may createimportant problems for patients. One important trial wasCLARIFY, a carefully conducted registry study of majoroutcomes in hypertensive patients with stable coronarydisease.11 Like some of the studies already discussed,CLARIFY had a primary outcome of CV mortality, ornonfatal MI, or stroke. When looking at 10 mm Hg systolicBP intervals, the CLARIFY investigators found that the10 mm Hg range below 130 mm Hg was associated withthe lowest event rate. Compared with that reference group, patients with BPs between 140 and 149 mm Hgwere about 50% more likely to have the primary endpoint,and those ≥ 150 mm Hg were almost 2.5-fold more likelyto have events. However, of particular importance in consideringthe SPRINT data, CLARIFY demonstrated thatachieved office systolic BPs < 120 mm Hg were associatedwith a 56% greater event rate than at the < 130 mm Hgtarget. Since these are high-risk patients similar to thosein SPRINT, it does seem important that our interpretationof the SPRINT BPs - in particular, that the intensivelytreated SPRINT group should be regarded as havingachieved 130 mm Hg rather than 120 mm Hg - is of greatimportance. Clinicians who interpret the SPRINT data ina literal way, and aim to achieve office BPs < 120 mm Hg intheir high-risk patients, could be at risk of causing harm.

Hypertension Journal, October-December, Vol 2, 2016 185

Michael A Weber

Another major trial that was announced soon afterSPRINT was the heart outcomes prevention experience-3(HOPE-3) that was performed in a large communitycohort with patients at relatively low CV risk.12 In fact,HOPE-3 excluded patients who had histories of priorCV events, although they did include people with moreconventional CV risk factors. The HOPE-3 trial failed tofind an overall benefit of active treatment (an ARB plus athiazide) compared with placebo. However, in a prespecifiedanalysis, patients in the highest tertile for baselineBPs (roughly > 140 mm Hg) did have a significant 24%reduction in the composite CV outcome; patients in themiddle tertile (roughly 130-140 mm Hg) had a neutraloutcome with similar event rates in the actively treatedand placebo groups. Of concern, patients in the lowesttertile (roughly ≤ 130 mm Hg) showed a trend to increasedCV events in the actively treated group compared withplacebo.12 This finding also provides a warning to cliniciansthat over-interpreting SPRINT and treating patientsin the range 120 to 130 mm Hg to achieve office values< 120 mm Hg could possibly precipitate, rather thanprevent, major events in lower-risk patients.

RECOMMENDED SYSTOLIC BP TARGETS

A summary of evidence-based systolic BP targets isshown in Table 3. Based on the SPRINT trial, it appearsmost appropriate to recommend an office target of 130 mm Hg in nondiabetic hypertensive patients atmedium-to-high CV risk. This recommendation is supportedby other reports as well.13,14 For patients withdiabetes, ACCORD and other trials7-9 indicate that a targetof < 140 mm Hg is justified; perhaps in these patients,it is reasonable to approach 130 mm Hg, although notbelow. For patients at low-to-medium CV risk, the targetshould be < 140 mm Hg, particularly in view of the recentHOPE-3 trial.12

Table 3: Office systolic BP targets
Will the Systolic Blood Pressure Intervention Trial(SPRINT) change Treatment Targets in Hypertension?
Note: Data are strongest in patients aged >55. Lower BP targetsmay be appropriate in younger adults, but evidence is lacking
 
It should be emphasized that there has not been asystematic randomized trial that has prospectively testedoffice systolic BPs of < 120 mm Hg (remember: The targetof < 120 mm Hg in SPRINT utilized a unique BP measurementmethod that corresponds more closely to an officevalue of around 130 mm Hg). Indeed, as discussed earlier,achieving office BPs of < 120 mm Hg have the potentialto cause harm.11,12

In a way, these recommendations could be simplifiedby suggesting a general target of 130 mm Hg, orslightly above, which should satisfy the needs of almostall hypertensive patients. It is also worth adding that, asSPRINT so capably demonstrated, elderly patients havedone very well when treated to office values of around130 mm Hg, even those regarded as "frail." One majoromission from these recommendations is for youngerpatients. Almost every clinical trial cited to support BPtarget recommendations is based on patients aged 55 orolder, with an overall average age in these trials in the midto high 60s. It is probably appropriate to target youngerpatients in their 20s, 30s, and 40s to 130 mm Hg, althoughif the pressure falls closer to 120 mm Hg without the needfor intensive pharmacologic intervention, that level of BPis probably very acceptable.

GLOBAL ISSUES

The recommendations made so far in these commentshave been based on communities with reasonable accessto health care. But as shown in Table 4, in many low- ormiddle-income countries resources are severely limited.There is often a shortage of trained health workers,including those with the ability to perform BP measurements.Another problem is that the cost of acquiringand maintaining simple BP measuring devices is often prohibitive. In addition, it becomes very difficult tomeasure or estimate overall CV risk in patients in suchcommunities because many of the standard evaluations,including routine blood tests, X-rays, and electrocardiograms(ECGs) are simply not commonly available. Ontop of that, administrative logistics are limited, makingit difficult to perform essential tasks like record keepingand patient scheduling.

Table 4: Global issues: Limited resources
Will the Systolic Blood Pressure Intervention Trial(SPRINT) change Treatment Targets in Hypertension?

186

Will Recent Trials change the Systolic Treatment Targets?

And finally, it should be emphasized that access todrugs is highly variable. Many such communities dependon donations of drug supplies, which often means thatthere is inconsistency from one patient visit to the nextin the type of therapy available to manage the patient'shypertension.

STRATEGIES IN GLOBAL GUIDELINES

It is probably most reasonable in low-resource settingsto reserve treatment for patients who are at greater risk,which often means those patients with untreated BPsat higher levels are given greater priority. The recommendationof 130 mm Hg that was discussed previouslymay be too ambitious in many areas, so 140 mm Hgshould probably be a more universal target. Anotherfactor that differentiates the care of patients in areaswith limited resource is the types of lifestyle therapiesthat can be recommended. For instance, recommendingweight loss is not practical in low-income areas wheremost patients have a body mass index that is clearly inthe normal range by Western standards. Similarly, inhot and tropical climates, it is hard to manage sodiumintake, particularly in patients engaged in manual workwho may experience a great deal of sodium loss duringthe day.

Based on some of these considerations, the guidelineslisted in Table 5 seem reasonable for communities withlimited resources. Very simply, we should use a systolicBP of 140 mm Hg as the threshold and target for treatment,and at times it may be economically necessaryto delay therapy at levels even higher than that. Sincetraditional office BP measurement techniques are difficultto teach, it is definitely preferable to use automated devices that do not require operator training. We mustrecognize that the cost of these devices, and even the costof the batteries to operate these devices, can be a majorconsideration, but this is the most accurate and consistentway to measure BP.

Table 5: Global hypertension guidelines forlow-resource settings

Will the Systolic Blood Pressure Intervention Trial(SPRINT) change Treatment Targets in Hypertension?
 
Lifestyle modifications must be based on the environment,and in many cases are difficult to enforce. The oneexception may be within those cultures where very highsalt intake is part of the standard diet, and where a communityor country-wide reduction in food salt contentmight produce a broad-based benefit.

In settings where it is not possible to get access tomultiple drug classes, it might be most desirable toselect a dihydropyridine calcium channel blocker as theoverall best choice since these agents appear to workwell regardless of patient ethnicity and age. For moresever hypertension, it would be highly advantageous toobtain single-pill combinations that offer two or eventhree-drug options based on the three-core hypertensiontherapies: Blockers of the renin angiotensin system/calcium channel blockers/thiazide diuretics. It shouldbe emphasized that this relatively simple strategy wasthe one used most commonly in the highly successfulSPRINT trial. Many of these combination products arenow generically available, although their cost may stillbe daunting in many parts of the world. Even so, sinceSPRINT the incentive to upgrade hypertension care hasbecome stronger than ever.

REFERENCES
  1. The SPRINT Research Group, Wright JT Jr, Williamson JD,Whelton PK, Snyder JK, Sink KM, Rocco MV, Reboussin DM,Rahman M, Oparil S, et al. A randomized trial vs. standardblood pressure control. N Engl J Med 2015 Nov 26;373(22):2103-2116.
  2. Chobanian AV, Bakris GL, Black HR, Cushman WC, Green LA,Izzo JL Jr, Jones DW, Materson BJ, Oparil S, Wright JT Jr,et al. The Seventh Report of the Joint National Committeeon Prevention, Detection, Evaluation, and Treatment of HighBlood Pressure: the JNC7 report. JAMA 2003 May 21;289(19):2560-2572.
  3. Mancia G, Fagard R, Narkiewicz K, Redon J, Zanchetti A,Bohm M, Christiaens T, Cifkova R, De Backer G, Dominiczak A,et al. 2013 ESH/ESC guidelines for the management of arterialhypertension. J Hypertens 2013 Jul;31(7):1281-1357.
  4. SHEP Cooperative Research Group. Prevention of strokeby antihypertensive drug treatment in older persons withisolated systolic hypertension. Final results of the SystolicHypertension in the Elderly Program (SHEP). JAMA 1991Jun 26;265(24):3255-3264.
  5. James PA, Oparil, S, Carter BL, Cushman WC, Dennison-Himmelfarb C, Handler J, Lackland DT, LeFevre ML, Mac-Kenzie TD, Ogedegbe O, et al. Evidence based guidelines forthe management of high blood pressure in adults: report fromthe panel members appointed to the Eighth Joint NationalCommittee (JNC8). JAMA 2014 Feb 5;311(5):507-520.

Hypertension Journal, October-December, Vol 2, 2016 187

Michael A Weber

  1. Weber MA, Julius S, Kjeldsen SE, Brunner HR, Ekman S,Hansson L, Hua T, Laragh JH, McInnes GT, Mitchell L, et al.Blood pressure dependent and independent effects of antihypertensivetreatment on clinical events in the VALUE trial.Lancet 2004 Jun 19;363(9426):2049-2051.
  2. ACCORD Study Group, Cushman WC, Evans GW,Byington RP, Goff DC Jr, Grimm RH Jr, Cutler JA, Simons-Morton DG, Basile JN, Corson MA, et al. Effects of intensiveblood pressure control in type 2 diabetes mellitus. N EnglJ Med 2010 Apr 29;362(17):1575-1585.
  3. Weber MA, Bloch M, Bakris GL, Weir MR, Zappe DH, Dahlof B,Velazquez EJ, Pitt B, Basile JN, Jamerson K, et al. Cardiovascularoutcomes according to systolic blood pressure in patientswith and without diabetes: an ACCOMPLISH substudy.J Clin Hypertens 2016 Apr;18(4):299-307.
  4. Brunstrom M, Carlberg B. Effect of antihypertensive treatmentat different blood pressure levels in patients withdiabetes mellitus: systematic review and meta-analyses. BMJ2016 Feb 24;352:i717.
  5. Williamson JD, Supiano MA, Applegate WB, Berlowitz DR,Campbell RC, Chertow GM, Fine LJ, Haley WE, Hawfield AT,Ix JH, et al. Intensive vs standard blood pressure control and cardiovascular disease outcomes in adults aged ≥ 75years: a randomized clinical trial. JAMA 2016 Jun 28;315(24):2673-2682.

 
  1. Vidal-Petiot E, Ford I, Greenlaw N, Ferrari R, Fox KM,Tardif JC, Tendera M, Tavazzi L, Bhatt DL, Steg PG, et al.Cardiovascular event rates and mortality according toachieved systolic and diastolic blood pressure in patientswith stable coronary artery disease: an international cohortstudy. Lancet 2016 Oct;388(10056):2142-2152.
  2. Lonn EM, Bosch J, Lopez-Jaramillo P, Zhu J, Liu L, Pais P, Diaz R,Xavier D, Sliwa K, Dans A, et al. Blood pressure lowering inintermediate risk persons without cardiovascular disease.N Engl J Med 2016 May 26;374(21):2009-2020.
  3. Weber MA, Bakris GL, Hester A, Weir MR, Hua TA, Zappe D,Dahlof B, Velazquez EJ, Pitt B, Jamerson K. Systolic bloodpressure and cardiovascular outcomes during treatment ofhypertension. Am J Med 2013 Jun;126(6):501-508.
  4. Ettehad D, Emdin CA, Kiran A, Anderson SG, Callender T,Emberson J, Chalmers J, Rodgers A, Rahimi K. Blood pressurelowering for prevention of cardiovascular disease anddeath: a systematic review and meta-analysis. Lancet 2016Mar 5;387(10022):957-967.

188