A paradigm shift in BP management: Results of SPRINT and other recent trials
by Haitham M. Ahmed, MD, MPH; Seamus P. Whelton, MD, MPH; and Roger S. Blumenthal, MD
In 2015, we have seen groundbreaking updates in the treatment of BP that will undoubtedly affect clinical practice and professional society guidelines. On Aug. 20, the NHLBI announced that it would stop the SPRINT trial nearly 2 years earlier than anticipated due to a significant mortality benefit in the intensive BP treatment arm.
Haitham M. Ahmed
In SPRINT, 9,361 patients aged older than 50 years who had a systolic BP between 130 mm Hg and 180 mm Hg older and one additional CV risk factor were randomly assigned intensive BP control with a target systolic BP < 120 mm Hg or standard control with a target systolic BP < 140 mm Hg. Patients with diabetes were excluded from SPRINT as the ACCORD trial was ongoing.
At the end of the trial, the mean systolic BP was 121.5 mm Hg in the intensive-treatment group vs. 134.6 mm Hg in the standard-treatment group. Surprisingly, on average, only one additional BP medication was needed to achieve this 13-mm Hg separation between groups (2.8 medications in the intensive-treatment group vs. 1.8 in the standard-treatment group).
Roger S. Blumenthal
After only 3.26 years, there was a 25% (P < .001) reduction in the composite endpoint (MI, non-MI ACS, stroke, HF or CVD mortality) in the intensive-treatment group compared with the standard-treatment group. Importantly, there also was a significant reduction in CVD mortality (HR = 0.57; 95% CI, 0.38-0.85) and all-cause mortality (HR = 0.73; 95% CI, 0.6-0.9) in the intensive-treatment group. This corresponds to a number needed to treat of 61 patients for the primary outcome and 90 patients for all-cause mortality. There was no significant difference in the treatment effect among the six prespecified subgroups, including patients aged older than 75 years.
Overall, the composite of adverse events was similar between groups. However, the intensive-treatment group had an increase in hypotension, syncope, electrolyte abnormalities and acute kidney injury, but not injurious falls. In patients without chronic kidney disease at baseline (estimated glomerular factor receptor [eGFR] > 60 mL/min/1.73m2), there was a greater than threefold increase in the number of participants with at least a 30% reduction in their eGFR to an eGFR of less than 60 mL/min/1.73m2 (HR = 3.49; 95% CI, 2.44-5.1). Whether this is a reversible result attributable to the preferential use of diuretics, ACE inhibitors and angiotensin receptor blockers or a true reduction in eGFR is unclear. Further detailed analyses of renal outcomes and understanding long-term renal implications will be invaluable.
Interpreting results in ACCORD, SPRINT
How then do we interpret the seemingly negative results found in the BP arm of the ACCORD trial, which used the same systolic BP treatment goals?
Firstly, it is important to note that the ACCORD BP trial showed a nonsignificant 12% (P = 0.2) reduction in its primary outcome (nonfatal MI, nonfatal stroke or CVD mortality) and a significant 41% (P = .01) reduction in stroke. Ultimately, the ACCORD trial was underpowered due to an observed event rate that was half of what was anticipated (2% per year vs. 4% per year) and a sample size half as large as that in the SPRINT trial. The participants in ACCORD may have been at lower CVD risk than those in SPRINT, partially due to a younger cohort (mean age, 62 vs. 68 years) and exclusion of participants with chronic kidney disease (creatinine > 1.5 mg/dL). Differences in the primary composite outcomes used in the trials may also have contributed because HF was used in SPRINT but not ACCORD. However, an analysis of the SPRINT trial showed a 25% reduction in events when the ACCORD primary outcome was applied to the SPRINT cohort, similar to the findings using the SPRINT composite outcome that includes HF.
In a post-hoc analysis of ACCORD, the primary outcome was 26% lower (P = .049) in the intensive BP treatment group that was randomly assigned to standard glycemic goals compared with those randomly assigned to standard BP and standard glycemic goals. The factorial design of the ACCORD trial may also have contributed: In long-term observational follow-up of the ACCORD participants (ACCORDION), there was an interaction between the intensive glucose-lowering and intensive BP-lowering interventions (P = .037). Whether further trials evaluating the effect of intensive BP lowering in patients with diabetes will be funded is uncertain, but ensuring an adequate sample size will be essential.
Adding further evidence in support of treating BP to lower than currently recommended targets is a meta-analysis by Xie and colleagues published ahead of print in The Lancet on Nov. 6. This meta-analysis included randomized controlled trials of at least 6 months comparing the effect of BP treatment to different targets. The mean systolic BP was 133 mm Hg in the treatment groups vs. 140 mm Hg in the control groups. In nearly 45,000 patients followed for a mean duration of 3.8 years, there was a 14% (95% CI, 4-22) reduction for major CVD events, a 13% (95% CI, 0-24) reduction in MI and a 22% (95% CI, 10-32) reduction in stroke in participants randomly assigned to a lower BP goal. Of note, there was no statistically significant difference in the reduction of major CVD events between trials conducted in participants with diabetes compared with those without diabetes (P for heterogeneity = .76).
In light of the ACCORD BP arm and a mandate to base guidelines on randomized trial data, the panel members appointed to the Eighth Joint National Committee (JNC 8) relaxed the BP goal from less than 130/80 mm Hg to less than 140/90 mm Hg for patients with diabetes and from less than 140/90 mm Hg to less than 150/90 mm Hg for patients older than 60 years. However, the addition of the aforementioned high-quality randomized controlled trial data showing a benefit with more intensive BP control represents a paradigm shift in clinical practice as well as future BP management recommendations.
Case vignette and authors' perspective
“Mr. Jones is a 45-year-old man with hypertension, dyslipidemia, ongoing smoking and a family history of premature CAD but excellent dietary and exercise habits. His mean BP after three separate checks is 139/77 mm Hg. He is on one agent for management of his BP. Recent BP measurements at home show an average of 140/80 mm Hg. His Framingham risk score is 21% and his atherosclerotic CVD risk score (by 2013 pooled cohort equation) is 10%. Do you consider the patient to be "at-goal" or do you add a second agent for better BP control?"
The patient in this example does not meet the SPRINT inclusion criteria due to his age of younger than 50 years, but otherwise would, based on his systolic BP between 130 mm Hg and 180 mm Hg and a Framingham risk score of at least 15%. As always, we must be cautious to not over extrapolate the results of these studies among patient groups not part of the trials. However, we should also be mindful not to use the absence of randomized trial evidence as the evidence of absence. Patients not meeting the SPRINT eligibility criteria who have a high CVD risk based on risk factors not used in SPRINT and those with a high CVD risk who are younger than 50 years are two groups in which it would be reasonable for clinicians to consider more intensive BP lowering.
Therefore, we believe it is reasonable to intensify the systolic BP treatment goal for this patient based on results from SPRINT. In doing so, we would recommend the use of chlorthalidone, an ACE inhibitor/angiotensin receptor blocker, or dihydropyridine calcium channel blocker, which were the mainstay medications used in SPRINT, in addition to further optimization of lifestyle along with complete smoking cessation. However, it should be noted that the treatment of patients outside of the SPRINT inclusion criteria who have a lower CVD risk will likely increase the number needed to treat. Further studies are needed to confirm the presence and exact magnitude of benefit with intensive BP control in younger adults and others not included in SPRINT.
Currently, the American Heart Association and American College of Cardiology are in the process of updating the BP guidelines, which are expected to be published in late 2016. These recent data underscore the importance of a timely revision of the upcoming guidelines with inclusion of this newly published data. We anticipate that the systolic BP treatment goal for older individuals will be lowered back to at least less than 140 mm Hg and that patients meeting the SPRINT inclusion criteria will also have significantly reduced BP targets.
Furthermore, the SPRINT authors broach the practical issues of attaining a systolic BP < 120 mm Hg in clinical practice. As SPRINT showed, the mean systolic BP in the intensive-treatment group was 121 mm Hg on an average of 2.8 antihypertensive agents. This will necessitate an increase in the number of follow-up visits, costs associated with these visits and the need for patients to take additional medications. However, these BP titration follow-up visits could be conducted by other skilled health care professionals such as nurse practitioners. It also is important to note that the BP measurements in SPRINT were the mean of three values measured after 5 minutes of rest. In-office, single BP measurements after less than 5 minutes of rest may result in higher values and lead to overtreatment. The use of home BP logs may be a more representative alternative for busy clinical practices.
Overall, there are several major points to take away from the above studies. First, there was a clear and early reduction in CVD outcomes with intensive BP lowering in nondiabetic patients at high risk for CVD in SPRINT. Even more impressive was a significant reduction in all-cause mortality after a median 3.26 years of follow-up. Second, this effect was present in participants older than 75 years, a group in whom the benefit of intensive BP lowering was previously uncertain. Third, on average, only one additional medication was needed in the intensive-treatment group to achieve a mean systolic BP of 121 mm Hg.
Fourth, in post hoc analyses of the ACCORD trial and in The Lancet meta-analysis of nearly 50,000 patients by Xie and colleagues, there was a benefit for lowering systolic BP to < 140 mm Hg in individual with diabetes. However, follow-up studies of intensive BP lowering in patients with diabetes would be required to definitively determine whether the differences between ACCORD BP and SPRINT were methodological or biological in basis. Thus, as has been recommended with the 2013 atherosclerotic CVD lipid guidelines, the physician–patient risk discussion should continue to serve as a cornerstone in all treatment decisions.
Cushman, WC. Long-term cardiovascular effects of 4.9 years of intensive blood pressure control in type 2 diabetes mellitus: The Action to Control Cardiovascular Risk in Diabetes Follow-on Blood Pressure study. Presented at: American Heart Association Scientific Sessions; Nov. 7-11, 2015; Orlando, Fla.
For more information:
Haitham M. Ahmed, MD, MPH, and Seamus P. Whelton, MD, MPH, are clinical fellows at the Johns Hopkins Ciccarone Center for the Prevention of Heart Disease and members of the Cardiology Today Fellows Advisory Board. Roger S. Blumenthal, MD, is director of the Ciccarone Center for the Prevention of Heart Disease and is the CHD and Prevention Section Editor of Cardiology Today.
Disclosure: The authors report no relevant financial disclosures.