It WAS a trial hailed as world-changing for management of hypertension; a trial that loudly called into question a global trend towards more relaxed blood pressure (BP) targets.
But the Systolic Blood Pressure Intervention Trial (SPRINT), halted a year ahead of schedule because of clear evidence of a benefit with more aggressive BP lowering treatment, has created consternation in primary care about how to apply its findings to real-world patients.
Not everyone agrees that systolic BP <120mmHg should be the new “normal”.
The study of 9300 US adults older than 50 at increased cardiovascular (CV) risk, but without diabetes, polycystic kidney disease or prior stroke, found significantly improved cardiovascular (CV) outcomes after three years in those treated towards a systolic BP goal of <120mmHg compared with those treated to the standard target of <140mmHg.
The landmark study, funded by the US National Institutes of Health, demonstrated that the intensively treated group had a 25% relative reduced risk (1.6% absolute risk reduction) of a first cardiovascular event or death (the composite primary outcome) and a 43% lower relative risk of death from cardiovascular causes.
However, seven months on from its ground-breaking publication in the New England Journal of Medicine1, specialists are divided about the generalisability of the results.
Professor Garry Jennings, Melbourne cardiologist and CEO of the Heart Foundation, says the trial results have been instrumental in questioning a trend towards guidelines with relaxed targets, such as those from Europe and the US in early 2015 which nominated 140mmHg as the trigger for antihypertensive medication in patients with diabetes.
Australian specialists cautioned against this systolic target after a study showed patients with diabetes who achieved a low of 130mmHg had about a 25% lower stroke risk.
With the November 2015 publication of SPRINT trumpeting the benefits of <120mmHG, a working party updating the Heart Foundation’s 2010 clinical guidelines on hypertension went back to the drawing board.
“SPRINT, coming when it did, had a definite impact,” Professor Jennings says. “I can’t pre-empt the guidelines but in patient populations that fit the SPRINT criteria or are similar, it is reasonable to go for a target of 120. But the evidence is for high- risk populations or people with established heart disease.”
Professor Anthony Rodgers, from The George Institute, Sydney, co-author of an editorial2accompanying the SPRINT publication, says the delay in the new Heart Foundation guidelines indicates that the strength of evidence from SPRINT cannot be ignored.
“The key thing that is important about SPRINT is that it is a range of high-risk patients and the benefit of those blood pressure ranges are for those high risk patients,” Professor Rodgers says.
Observational studies have shown progressive cardiovascular risk with BP >115mmHg but SPRINT was born out of a decision by a US National Heart Lung and Blood Institute expert panel in 2007 to test hypertension prevention in patients without diabetes in a randomised controlled trial.
Where few dispute the evidence for more aggressive BP targets for high-risk patients such as those in SPRINT, experts remain divided about how to implement the findings when managing other patient populations, including the frail elderly, those with diabetes and those at intermediate risk.
Professor Jennings says it is important to acknowledge how BP was measured in SPRINT participants — via an automated machine with no health professionals in the room — something he would like to see more of in Australian clinical practice. Taking this into account, he believes the evidence is increasingly shifting towards <120mmHg as “a reasonable target, with various provisos”.
Much of the debate has focused on whether BP demonstrates a linear relationship with future CV events or whether there is a tipping point.
Professor Tom Marwick, director and chief executive of the Baker IDI Heart and Diabetes Institute, says it is important to take note of the exclusions from the study.“People do liken blood pressure levels to fashions. You get a study like this and you get indication creep to other groups.
“SPRINT is a well done and large study. I don’t think we can ignore it but we have to take notice of who was excluded and who was studied.
“[SPRINT participants] are people who are at significant risk. The person who turns up for a routine check-up and is young, this is not that patient.”
In April this year, the HOPE-3 study3 showed lowering BP in an intermediate risk group does not yield a benefit. HOPE-3, however, did demonstrate a benefit when patients at intermediate risk were prescribed a statin or a combination of a statin and an antihypertensive.
All three arms of the trial demonstrated the greatest benefits in those patients who were in the upper third of systolic BP measurements, >143.5mmHg.
“We need to think about these things in relation to absolute risk, not just treating high blood pressure,” Professor Marwick says.
“There might be other ways of reducing risk without the risk of side effects.”
Professor Marwick agrees with Professor Jennings about the importance of reliable measurements when comparing individual patients with the SPRINT population.“If I am a GP and I have someone with a borderline blood pressure I want to make sure that measure is comparable to that of SPRINT.”
At a population level, the focus should be on recognition of hypertension and better treatment of poorly controlled hypertension, he says.
“A third of hypertensive patients are not diagnosed and a third of those are not treated. We need to focus on those rather than getting those who are already being treated down a couple of millimetres.
“We know what to do but we don’t do it. We have a group of people in the community who we are just not accessing.”
The numbers needed to treat in SPRINT were 61 to prevent a primary outcome event (including myocardial infarction (MI), acute coronary syndrome (ACS), stroke, heart failure or death), 90 to prevent death from any cause and 172 to prevent death from CV causes.
No impact on stroke, MI
Commentators have noted the benefit in the primary outcome was apparently driven by reduced risk of heart failure and death as there were no significant differences in rates of stroke, MI, or ACS between the groups.
Professor Murray Esler, a cardiologist and senior director of the Baker IDI, was among those at the 2015 American Heart Association (AHA) Scientific Sessions in Orlando, Florida, when the results of SPRINT were presented.
He subsequently published an editorial in the journal Hypertension,4 suggesting the results in favour of <120mmHg in SPRINT should be interpreted as SPRINT-specific.
“This can’t become the universal target for everyone,” Professor Esler says. “We know some people can’t tolerate that, and, at the [AHA] meeting, even the chief investigator said ‘absolutely not’.
“We do need to be better at managing hypertension and we should be trying to get to a lower level, but many people cannot tolerate a BP of 120mmHg.
“They get fatigue, dizziness and their renal function can deteriorate completely.”Professor Mark Nelson, chair of the discipline of general practice at the University of Tasmania and a guideline author for the Australian National Vascular Disease Prevention Alliance, emphasises that treatment thresholds should be based on absolute risk, not BP alone.
The SPRINT results, he says, are “definitely not a call for treating all patients with blood pressures above 130mmHg with drugs”.
He says: “Absolute risk is the starting point. Treatment is given to those at high absolute risk irrespective of blood pressure level.
“For those with multiple morbidities, consideration can be made for treatment at intermediate level.
“Those with low absolute risk with elevated blood pressure should be treated on lifestyle factors rather than with drugs unless the blood pressure is very high.”
In SPRINT, an open-label trial, patients in the intensive group were taking three BP-lowering medications and those in the standard group two (a mean of 2.8 vs 1.8).
Professors Nelson and Esler both point out that the entry level systolic BP for SPRINT participants was a modest 137mmHg. In the intensive treatment group this equated to a fall of only 18mmHg.
“A lot of our patients start at 180 or 200 and to get someone down 60 or 80mm is too much for some people,” Professor Esler says.
Patients with diabetes
Another area of dispute is how applicable the results are for patients with diabetes, who were excluded from SPRINT and in whom the similar ACCORD study failed to show a benefit for intensive BP-lowering regimes.
Professor Rodgers says subtle issues must be acknowledged when comparing the two trials, including disproportionate statistical power and the inclusion of events in the primary outcome of ACCORD that were not as sensitive to BP reduction.
“It looks as though you get the same benefits in diabetics but diabetics have more coronary disease in the first place,” he says.
“There is a tendency in BP research to require evidence for every single sub-group rather than stepping back and seeing the big picture.
“The evidence suggests a pretty generalisable trend.”
But Professor Esler describes the exclusion of patients with diabetes from SPRINT as a “black hole”, given how often diabetes is comorbid with hypertension in clinical practice.
Another subgroup sparking debate is the elderly. A third of participants in SPRINT were aged 75 or older. These individuals had no history of diabetes or stroke, were not living in a nursing home and had a life expectancy of at least three years.
Professor David Le Couteur, a geriatrician and clinical pharmacologist from The University of Sydney, says the results are applicable to many community-dwelling older patients.
“However, in multi-morbid older people, doctors need to individualise treatments,” he says.
“If a significant adverse event occurs then it is possible that, for that particular patient, the harm that has occurred outweighs the predicted benefits.
“Adverse drug reactions seem to be less of a problem in many clinical trials than in real life, presumably because the subjects of the trials tend to be healthier and very intensively monitored.”
Professor Le Couteur also points out that the effect of intensive treatment on cognitive outcomes in SPRINT is yet to be published “and this will be an important result”.
Selecting the right patients
Serious adverse events in SPRINT occurred more frequently in the intensive, compared with the standard, BP-lowering arm. They included hypotension, electrolyte abnormalities, acute kidney injury and renal failure.
However, somewhat surprisingly, there was no increased risk of falls.
“Falls and fractures have reached mythic proportions when it comes to blood pressure management,” Professor Rodgers says.
“This is the thing about prevention, no patient comes in and says I would have had a stroke yesterday but I didn’t because I am managing my blood pressure.”
The benefit and risks must always be weighed, he says.
However, the adverse event that cannot be dismissed in SPRINT is the significant increase in kidney damage in the intensively treated group.
“If you have moderate to severe kidney disease then BP lowering is a good thing,” Professor Rodgers says.
“However there is this complicated mechanism where you have a short term change in glomerular filtration and it tips over those people who are marginal.”
When it comes to anti-hypertensive therapy, Professor Esler believes outcomes can be “somewhat unpredictable” and targets need to be individualised.
“My take on SPRINT is that we do need to be trying a bit harder. Doctors often don’t try hard enough and tolerate unsatisfactory levels of blood pressure in patients,” he says.
“But we must also be prepared to back off when we see side effects, and not everyone’s goal should be 120mmHg.”
Perhaps no Heart Foundation hypertension guidelines have been so keenly awaited as those now expected to be finalised in July.
1. N Engl J Med 2015; 373:210-16
2. N Engl J Med 2015; 373:2175-78
3.N Engl J Med 2016; online 6 April
4. Hypertension 2016;67:266-267