A Systolic BP Below 120 mm Hg for Everyone? I Think NOT.

Guest Perspectives | George L. Bakris, MD

I have changed my blood pressure recommendations for a subgroup of people with criteria similar to those studied in SPRINT based on the early termination and results of this trial. Over the past 25 years a number of clinical outcome trials have evaluated the effectiveness of blood pressure reduction on cardiovascular (CV) outcomes1. All these trials show significant CV risk reduction and especially a reduced incidence of heart failure.1

SPRINT was designed to answer the question: “Does a lower blood pressure level further reduce CV events and all-cause mortality in people with very high CV risk?”

SPRINT was specifically focused on people with higher CV risk than the ACCORD study, primarily driven by advanced chronic kidney disease (CKD), older age, and without diabetes. It was also a much larger trial than ACCORD in order to properly power the study to address this question. The primary analysis of ACCORD did not demonstrate a benefit of a lower BP target when reviewed in the context of aggressive glycemic control. However, the confidence limit did not exclude a benefit as large as a 27% improvement in the composite CV disease outcome. Moreover, a reanalysis of the ACCORD BP trial demonstrated that intensive BP treatment alone with usual glucose control, compared with combined standard treatment, improved major CVD outcomes (hazard ratio 0.67; 95% CI 0.50-0.91).2

SPRINT had a mean age of 68 years, with almost 29% of the 9300+ subjects with Stage 3 or higher CKD (eGFR < 60 ml/min/1.73m2), a known independent CV risk factor.3 Moreover, a little over 28% were over age 75. Given these demographics of very high CV risk, there was a clear benefit of CV risk reduction and all-cause mortality in the group randomized to a systolic BP of < 120 mm Hg.

Reviewing the specific outcomes that comprised the composite primary endpoint was quite revealing, in that the benefit was driven predominantly by heart failure, CV death and non-fatal heart failure. There was NO further benefit on reduction of ischemic type events. This is not surprising, since previous trials such as the Hypertension in Very Elderly (HYVET),4 and the Systolic Hypertension in elderly People (SHEP)5 and the Systolic Hypertension in Europe (SYST- Eur)6 trials all demonstrated the greatest risk reduction on heart failure events when blood pressure was lowered, albeit to much higher levels than SPRINT. In the subgroup analysis of SPRINT, everyone garnered a benefit regardless of age, sex or race, although African Americans tended not to have as great a CV risk reduction.

The interpretation of these outcomes needs careful attention. To be clear, they do NOT indicate that everyone needs to have a systolic BP < 120 mm Hg to reduce events. These data are generalizable only to those who have the same demographic and clinical status as defined in the inclusion criteria of the study. Contraindications to the more aggressive target include:

  • Arterial stiffness. SPRINT like HYVET, recruited people who had relatively good arterial compliance. These individuals can tolerate lower pressures while people with low baseline diastolic BPs at or below 60 mm Hg cannot.
  • People over age 75 who have a history of falls, vertigo, orthostatic hypotension or are frail should not necessarily have their BP reduced to < 120, particularly if they have a pulse pressure of > 70 mm Hg at baseline.

When evaluating SPRINT results, remember that the standard of care in SPRINT was the highest of any trial: cholesterol management was excellent and almost all participants were on a long acting medications. This not only ensures reduction in ischemic events but also volume events, such as heart failure. Perhaps the key element to the success of SPRINT is the ability to maintain BP control over the 24 hour period, a hallmark that is likely associated with better outcomes7 and not present in older trials. Clearly, those with the entry criteria for this study should be encouraged to participate in getting their BP lower to further reduce heart failure related events including death.

As a nephrologist I have to say that while the follow-up was not that long, lower pressure did not make a difference to meaningfully slow CKD progression. SPRINT now makes the fourth randomized BP level trial to show a failure of benefit of lower BP on CKD progression.8 Thus, SPRINT provides meaningful data to help guide aggressiveness of BP lowering in older people with very high CV risk. It does not apply to people with marked arterial stiffness defined by high pulse pressures or those who met the exclusion criteria of the trial. Note that long acting thiazide-like diuretics like chlorthalidone were used in essentially everyone and hence, heart failure benefit drove the response in the lower BP group. The findings related to all-cause mortality are a mystery to me and have no plausible explanation. SPRINT MIND continues as does the follow-up of the CKD group so more data to come and hopefully provide new directions which can further reduce morbidity.

< Back to Listings