Optimal Approach for BP Lowering Debated Amid New J-Curve Signals in Stable CAD

A large, international registry study is providing support for a so-called J-curve phenomenon regarding blood pressure and adverse outcomes in patients with stable coronary disease, with levels on either end of the spectrum—whether too high or too low—being linked to elevated risks.

To TCTMD, however, a SPRINT trial leader cautioned against reading too much into the results, even as investigators from SPRINT and HOPE-3 continue the debate over which trial’s approach to lowering blood pressure may be most appropriate for clinical practice.

The registry study, presented at the European Society of Cardiology Congress 2016 in Rome, Italy, shows that patients with a systolic pressure below 120 mm Hg or a diastolic pressure less than 70 mm Hg had greater risks of a composite of CV death, MI, or stroke, as well as all-cause mortality.

“Together with previous literature, our study suggests caution when treating patients with coronary artery disease with antihypertensive drugs,” write lead author Emmanuelle Vidal-Petiot, MD (Hôpital Bichat, Paris, France), and colleagues in their paper simultaneously published online in the Lancet. “Future randomized controlled trials will be needed to confirm the cutoff blood pressure value below which harm outweighs benefit in this population.”

They add, though, that the results “should not slow down the constant effort that is still needed to improve patient care, because even with the conventional pressure goal of less than 140/90 mm Hg, only about half of the population with hypertension is controlled.”

Commenting for TCTMD, Paul Whelton, MB, MD (Tulane University School of Public Health and Tropical Medicine, New Orleans, LA), chair of the SPRINT steering committee, called the study interesting but pointed out that observational studies can only raise questions that must be answered in randomized controlled trials.

And the central message of trials that have been conducted so far is that lower blood pressure is better, Whelton said, adding that “we’re probably not being as intensive in our treatment in routine practice as we could be and should be.”

That said, the takeaway from this new analysis is something that clinicians should already know, he noted.

“When we’re managing high blood pressure and managing cardiovascular disease in general . . . we have to be cognizant of the clinical situation in which this management is being undertaken,” he said. “For people who have overt comorbid disorders, we have to be careful and we have to use good clinical judgment” regarding how well a patient is tolerating the intervention and whether the intensity needs to be pulled back.

Uncertainty About Blood Pressure Goals

Although there is consensus that lowering blood pressure in patients with hypertension reduces risks of CV events and death, there is lingering uncertainty about optimal levels, something reflected in—and perhaps fed by—inconsistent guideline recommendations.

Earlier randomized trials failed to demonstrate that aiming for goals below 140/90 mm Hg improves outcomes, but the SPRINT trial results reported in November 2015 showed that intensive treatment to a systolic pressure below 120 mm Hg versus a standard goal of less than 140 mm Hg cut rates of adverse outcomes in patients at high cardiovascular risk.

Post hoc analyses of some previous trials, however, suggested that it might be possible to go too low, at which point myocardial perfusion may be impaired and outcomes may be worsened.

To explore that issue in a population that may be particularly susceptible to that concern, Vidal-Petiot et al examined data on 22,672 patients with stable CAD receiving antihypertensive treatment who were enrolled in the CLARIFY registry from 45 countries.

During a median follow-up of 5 years, adverse outcomes were more frequently seen in patients with systolic pressures of 140 mm Hg or diastolic pressures of 80 mm Hg or more, as expected.

But low pressures were also associated with elevated risks of events. The risk of CV death, MI, or stroke (primary outcome) was higher in patients with systolic pressures below 120 mm Hg (adjusted HR 1.56; 95% CI 1.36-1.81) and with diastolic pressures of 60-69 mm Hg (adjusted HR 1.41; 95% CI 1.24-1.61) or less than 60 mm Hg (adjusted HR 2.01; 95% CI 1.50-2.70). Similar associations were seen for all-cause and CV death, MI, and hospitalization for heart failure, but not for stroke.

Conflict With SPRINT?

The study authors say their findings do not conflict with those of SPRINT, despite the fact that the trial aimed for a systolic goal below 120 mm Hg, the range associated with harm in the registry.

That is “because unlike other blood pressure intervention trials, the blood pressure values in SPRINT were measured under unattended conditions to minimize any white coat effect, and so might underestimate casual blood pressure values by between 5 and 16 mm Hg,” they write.

“This finding actually led hypertension experts to warn that the SPRINT target translated into community practice might have deleterious effects because the same targets obtained in routine practice would potentially lie within the left part of the J-curve,” they continue. “Our results, which show a J-curve in patients with casual blood pressure measurements with harmful thresholds very close to the achieved blood pressure obtained in the intensive-treatment group of SPRINT, indeed support this word of caution.”

Senior author Philippe Gabriel Steg, MD (Hôpital Bichat), told TCTMD in an email that the measurement technique used in SPRINT “should be the routine measurement, but never will, and thus remains the perfect but unachievable gold standard.”

He described the situation as a Catch-22. “Had they not used a careful measurement, they would have been criticized for using an unreliable measure,” Steg said. “Because they used the ‘perfect’ measure, they are criticized because their results are difficult to extrapolate to the imperfect measurements used in clinical practice.”

Whelton acknowledged that “you have to be careful generalizing [the SPRINT] experience to clinical practice,” but said that the main message of the trial should not be lost as researchers debate various issues surrounding it.

“That is: more intensive treatment, on average with one additional blood pressure medication, lowered blood pressure in the intensive arm by about 15 mm Hg systolic compared to the standard group and it yielded huge apparent benefits,” Whelton said.

Any specific recommendations that can be made based on the results of SPRINT and other trials—HOPE-3, for example—should be left to guideline committees, added Whelton, who is chair of the committee writing the comprehensive hypertension guidelines from the American Heart Association and American College of Cardiology. That guidance is nearing completion and should be released in early 2017, he said.

Is a SPRINT or HOPE-3 Approach Better?

In the meantime, clinicians remain unsure of the best way to manage high blood pressure in their patients, a topic embroiled in ongoing debate. Indeed, in two viewpoints published online September 7, 2016, ahead of print in JAMA Cardiology, investigators from the SPRINT and HOPE-3 trials consider which approach may be more suitable in clinical practice. Unlike SPRINT, HOPE-3 failed to show that blood pressure-lowering reduced major adverse outcomes.

In one viewpoint, Whelton and colleagues say that the discrepant results seen in the two trials could be due to chance, drug choice, or methodological differences. To that last point, they explain that SPRINT:


  • Treated aggressively to a blood pressure goal much lower than recommended in current guidelines, whereas HOPE-3 evaluated a fixed-dose combination pill without specific goals
  • Included a population with a much higher CV risk compared with HOPE-3
  • Included mostly patients (91%) who were being treated with antihypertensives, a higher percentage than in HOPE-3 (22%)
  • Achieved a blood pressure difference between study arms that was twice as high as that seen in HOPE-3


“The HOPE-3 BP trial tested a strategy that has special relevance for a population intervention, where most being treated are at low or intermediate risk for CVD, and fixed-dose combination therapy that requires no titration is an attractive option,” Whelton et al say.

“In contrast, the SPRINT intervention was modeled on a clinically oriented stepped-care approach in which drugs were increased to full recommended doses in participants with high BP who share the high CVD risk characteristics of many patients seen by healthcare professionals,” they continue. “We believe the experience in SPRINT provides a better guide for decision-making in clinical practice.”

In the other viewpoint, the principal investigators of HOPE-3, Salim Yusuf, MBBS, DPhil, and Eva Lonn, MD (McMaster University and Hamilton Health Sciences, Hamilton, Canada), support their trial’s approach while pointing out that SPRINT may have overestimated the benefits of intensive treatment for a variety of reasons, including the early termination of the trial.

They note, too, that intensive treatment in SPRINT was associated with increases in emergency department visits for hypotension, syncope, electrolyte abnormalities, and acute kidney injury, as well as deterioration in renal function, which “suggests that the approach used in SPRINT is not benign and requires careful monitoring.”

Yusuf and Lonn also tout the simplicity of the HOPE-3 strategy, with half as many monthly visits compared with SPRINT (three vs six) and a lack of dose titration or routine laboratory testing.

“Both the HOPE-3 trial and SPRINT show that different approaches can lead to greater benefits for individuals with hypertension,” they write. “However, in the majority of those with hypertension, the simple approach used in the HOPE-3 trial (two BP-lowering drugs at low doses plus statins) is safe and pragmatic and results in substantial benefits.”


Todd Neale is the Associate News Editor for TCTMD and a Senior Medical Journalist. He got his start in journalism at …

Read Full Bio
  • Vidal-Petiot E, Ford I, Greenlaw N, et al. Cardiovascular event rates and mortality according to achieved systolic and diastolic blood pressure in patients with stable coronary artery disease: an international cohort study. Lancet. 2016;Epub ahead of print.

  • Whelton PK, Reboussin DM, Fine LJ. Comparing the SPRINT and the HOPE-3 blood pressure trial. JAMA Cardiol. 2016;Epub ahead of print.

  • Yusuf S, Lonn E. The SPRINT and the HOPE-3 trial in the context of other blood pressure-lowering trials. JAMA Cardiol. 2016;Epub ahead of print.

  • The CLARIFY registry is supported by Servier.
  • Vidal-Petiot reports receiving nonfinancial support from Boston Scientific and Servier.
  • Yusuf and Lonn report serving as principal investigators of the HOPE-3 trial, which was funded by AstraZeneca and the Canadian Institutes of Health Research.
  • Lonn reports receiving research support from AstraZeneca, Amgen, Bayer, GlaxoSmithKline, Eli Lilly, and Sanofi; consulting and/or lecture honoraria from Amgen, Sanofi, and Novartis; and travel support from Sanofi.
  • Yusuf reports receiving research support from Boehringer Ingelheim, AstraZeneca, Bayer, Merck, Bristol-Myers Squibb, Cadila, and Novartis and honoraria and travel expenses from Bayer.
  • Whelton reports no relevant conflicts of interest.