You don't need to be signed in to read BMJ Blogs, but you can register here to receive updates about other BMJ products and services via our site.

Primary Care Corner with Geoffrey Modest MD: SPRINT Trial: Elderly Subgroup Study of Lower Blood Pressure Goal

2 Jun, 16 | by EBM

By Dr. Geoffrey Modest

A subgroup analysis of the SPRINT trial found improved clinical outcomes in community-dwelling patients >= 75yo on intensive blood pressure control (see doi:10.1001/jama.2016.7050). For an overall review/critique of SPRINT, see


  • In brief, the SPRINT trial involved 9361 patients, mean age 68 (but they had a pre-designated subgroup >=75 yo), randomized to SBP goals of <120 vs <140 and achieving SBP of 121 vs 136, then finding a pretty dramatic clinical benefit in those with the more aggressive blood pressure goal. The researchers did not determine the antihypertensives used, but encouraged using those with known cardiovascular benefit (diuretics, calcium blockers, ACE/ARB)
  • The current study is of the 2636 people were over >=75 (mean age 79.9, 38% women, 75% white/17% black/7% Hispanic, baseline BP 142/71, 10% with orthostatic hypotension, creatinine 1.1 mg/dl, eGFR 63, total chol 182/HDL 56, 52% on statin, 10-yr Framingham risk score 25%, mean of 2 antihypertensives, 13% physically fit/55% less fit/32% frail)
  • Results in this older cohort, after 3.14 years (study terminated early):
    • Achieved blood pressure: 123.4/62.0 vs 134.8/67.2; those in the intensive group required average of 1 more antihypertensive (frail patients had SBP difference of 10.8 mm Hg, less fit 11.3 and fit 13.5 mmHg)
    • Primary composite outcome (nonfatal MI, acute coronary syndrome, nonfatal stroke, nonfatal acute decompensated heart failure, death from cardiovascular cause):
      • 102 events in intensive group (2.59%/yr) vs 148 (3.85%/yr) on standard therapy, a 34% reduction [HR 0.66 (0.51-0.85)]. Number-needed-to-treat for 3.14 years was 27
    • Secondary outcome (all-cause mortality):
      • 73 deathsin intensive group vs 107 on standard therapy, a 33% reduction [HR 0.67 (0.49-0.91)].  NNT for 3.14 years was 41
    • Rate of serious adverse events was not statistically different overall, with:
      • Hypotension in 2.4% in intensive vs 1.4% standard, nearly significant trend
      • Orthostatic hypotension at clinic visit in 21.0% vs 21.8%
      • Syncope in 3.0% vs 2.4%
      • Electrolyte abnormalities in 4.0 vs 2.7%, nearly significant trend
      • Acute kidney injury in 5.5 vs 4.0% (leading to hospitalization or were reported in hospital discharge summary), nearly significant trend
      • Injurious falls in 4.9% vs 5.5%
      • And, overall, no increase in the serious adverse events in the frail group

So, several points:

  • The hypertension guideline trend has been for increasingly higher blood pressure goals in the elderly: the European guidelines target initiating treatment only if SBP>160 in those >80 yo; the JNC8 recommendations are for a target of 150 in those >60 yo
  • The SPRINT guidelines did have some really important exclusions: history of diabetes, symptomatic heart failure in past 6 months or LVEF <35%, clinical dementia, prevalent stroke, unintentional weight loss of >10%, an SBP of <110 mmHg after standing for 1 minute, or residing in a nursing home
  • Not surprisingly, the NNT was much lower in this trial than the overall SPRINT trial, given the higher event rate of this older group
  • And, reassuringly re: orthostatic hypotension, this subgroup analysis is pretty convincing that there was no increase, but they did exclude anyone with a standing SBP <110
  • But, a few caveats here:
    • The baseline BP was pretty low: mean 142/71. I would be hesitant to assume the same results would apply to those with baseline BP of 200/100
    • This is a much smaller cohort than the larger study, and the overall SPRINT study did find statistically significant though small increases in adverse effects in the intensive group: for hypotension, syncope, electrolyte disturbances, and acute kidney injury
    • They do not give a breakdown of the age distribution, nor do they do further subgroup analysis to know if the same benefit would apply to those 85 years old or older
    • They did not look at initial orthostatics (see )which may be more prevalent and important than the usual postural changes that they did assess.
    • I still remain concerned that the diastolic pressures are pretty low (62 range), given that it is in diastole that the coronaries perfuse, and this high risk group likely had some obstructive coronary disease. I do understand that these 3 year results did not show an increase in cardiovascular events/mortality, but I remain a tad skeptical
  • However, this issue of blood pressure goal in older people is really important clinically, since studies such as the Framingham Study have found that about 90% of elderly develop hypertension if they live long enough. Another study found that 75% of those 75 years old have hypertension
  • So, what makes sense for clinical practice?
    • I do think this study, despite above caveats, should inform clinical practice
    • As per my last blog, I do think it is really important to make sure the blood pressure reading we are basing decisions on is really accurate (make sure done correctly, use ambulatory/home monitors,…)
    • I would check orthostatics, including initial orthostatics, on all elderly patients given this lower goal
    • But that being said, I would strongly consider slowly decreasing systolic blood pressure in elderly patients to the low 120 range, with slow incremental med changes, in those with no prior history of diabetes, stroke, or standing BP <110 mm Hg (as excluded in the SPRINT trial).

By submitting your comment you agree to adhere to these terms and conditions
You can follow any responses to this entry through the RSS 2.0 feed.

Latest from Evidence-Based Medicine

Latest from EBM