Should Low Diastolic Blood Pressure Level Be a Barrier to Intensive Blood Pressure Lowering?

Randomized controlled trials have demonstrated the benefits of lowering systolic blood pressure (SBP) to reduce the risk for cardiovascular disease (CVD) and all-cause mortality.1 In the Systolic Blood Pressure Lowering Intervention Trial (SPRINT), participants were randomized to SBP targets of 120 or 140 mm Hg and those randomized to the lower SBP goal had a lower risk for CVD and all-cause mortality (hazard ratios 0.75, 95% confidence interval [CI] 0.64–0.89 and 0.73, 95% CI 0.60–0.90, respectively).2 Reductions in CVD and mortality risk with intensive SBP lowering were confirmed in the Strategy of Blood Pressure Intervention in the Elderly Hypertensive Patients, a trial of 8,511 adults randomized to a SBP target of 110 to <130 or 130 to <150 mm Hg, and subsequent meta-analyses.1,3 For example, in a meta-analysis of 48 randomized trials (n = 344,716), lowering SBP by 5 mm Hg reduced the risk for CVD events by 10% (hazard ratio 0.90; 95% CI 0.88–0.92) with no evidence of effect modification by baseline SBP, from <120 to ≥170 mm Hg.1 Despite the impressive CVD risk reduction benefits of lowering SBP, antihypertensive medication lowers diastolic blood pressure (DBP) simultaneously with SBP and the effect of initiating or intensifying antihypertensive medication among individuals with low DBP has been controversial. Randomized controlled trials from the 1960s and 1970s demonstrated the CVD risk reduction benefit of antihypertensive medication among individuals with high DBP. In the second VA Cooperative Study, participants had a baseline DBP ≥90 mm Hg with approximately 25% of participants having a baseline DBP ≥115 mm Hg.4 In this trial, 12% of those randomized to antihypertensive treatment experienced cardiovascular complications compared with 29% of their counterparts randomized to the control group. In the Hypertension Detection Follow-up Program, which recruited adults with a DBP ≥90 mm Hg at baseline, the 5-year mortality rate was 17% lower in participants randomized to stepped care, with a DBP goal <90 mm Hg or a reduction in DBP of 10 mm Hg, vs. their counterparts randomized to referred care, a group which did not have a DBP goal.5 In the Hypertension Detection Follow-up Program, participants in the stepped care achieved a mean DBP between 80 and 90 mm Hg during follow-up. The benefit of lowering DBP below 80 mm Hg was not evaluated in this study. While there is consensus that lowering blood pressure (BP) among adults with DBP ≥90 mm Hg reduces CVD risk, the benefit of treating lower DBP, especially among older adults, is unclear. An increased risk for CVD events and all-cause mortality with low DBP has been reported for decades in observational studies.6,7 From a physiological perspective, it seems plausible for low DBP to be harmful as it may reflect low myocardial perfusion. Also, a reduction in DBP among people with already low levels might result in a perfusion pressure below which autoregulation cannot occur.6 Compromised myocardial perfusion can result in myocardial ischemia. The potential harms of very low DBP have created tension with the decision to initiate and intensify antihypertensive medication among adults with high SBP and low DBP. A high proportion of adults, especially older adults, have high SBP and low DBP. According to NHANES 2017–2020, 15% of US adults with SBP ≥130 mm Hg have DBP <70 mm Hg.8 Among US adults ≥65 years of age with SBP ≥130 mm Hg, 33% have DBP <70 mm Hg. New data on the association of low DBP with CVD and mortality risk from the Women’s Health Initiative Long Life Study are published in the current issue of the American Journal of Hypertension.9 The study population included 7,527 women who completed the Women’s Health Initiative and are being followed prospectively for CVD and mortality outcomes.10 For this analysis, BP was measured following a standardized protocol during a study visit in 2012 and 2013 and participants were followed for CVD and all-cause mortality. Over 5 years of follow-up, both low and high levels of DBP were associated with an increased risk for all-cause mortality. Specifically, after multivariable adjustment, the hazard ratio for all-cause mortality associated with DBP of 50 and 100 mm Hg, each vs. 80 mm Hg, was 1.50 (95% CI 1.18–1.90) and 1.83 (95% CI 1.43–2.35), respectively. For CVD, compared with participants with DBP of 80 mm Hg, the hazard ratio among those with DBP of 50 and 100 mm Hg were 1.14 (95% CI 0.78–1.67) and 1.50 (95% CI 1.03–2.17), respectively. The authors noted that it was unclear whether low DBP was causally associated with increased mortality or if it served as a risk marker. Clarifying this is important given the high risk for mortality and CVD that continues to be identified among individuals with low DBP in some population-based studies. The decision to initiate or intensify antihypertensive medication should weigh the potential benefits and harms of doing so in this population. The Women’s Health Initiative Long Life Study results are important to consider in the context of guidelines that recommend intensive SBP lowering. Several recent guidelines note a knowledge gap related to the level of DBP at which antihypertensive medication should be initiated and the goal DBP level. The 2017 American College of Cardiology/American Heart Association (ACC/AHA) BP guideline recommends adults with hypertension achieve a SBP goal of <130 mm Hg and a DBP goal of <80 mm Hg.11 Older adults, defined by the 2017 ACC/AHA BP guideline as 65 years or older, without high CVD risk, are recommended to achieve a SBP goal <130 mm Hg without a DBP goal. The 2017 American College of Physicians/American Academy of Family Physicians BP guideline notes a lack of data to recommend a DBP goal for older adults while the 2018 European Society of Cardiology/European Society of Hypertension arterial hypertension guideline called for more outcome studies of the optimal DBP treatment target.12,13 Secondary analyses of randomized trials have provided some insight into the potential benefit of antihypertensive medication at low DBP levels. In SPRINT, the hazard ratio for the primary CVD outcome and all-cause mortality with randomization to the SBP goal <120 vs. <140 mm Hg among participants in the lowest quintile of baseline DBP, <68 mm Hg, was 0.78 (95% CI 0.57–1.07) and 0.88 (95% CI 0.60–1.29), respectively.14 Although these estimates did not reach statistical significance, there was no evidence of differences in the effect of the intensive vs. standard SBP goal across quintiles of DBP in SPRINT (P value for interaction = 0.92 for CVD and P value for interaction = 0.57 for all-cause mortality). The investigators concluded that low DBP should not be an impediment to intensive antihypertensive treatment. There is a need to reconcile the results from observational studies showing potential increased CVD and mortality risk with low DBP and the results of SPRINT showing a potential benefit of antihypertensive medication regardless of DBP level. Observational studies, such as Women’s Health Initiative Long Life Study, are useful for identifying high-risk populations and generating hypotheses that can be further investigated in randomized trials. In the Women’s Health Initiative Long Life Study, the stronger association with all-cause mortality vs. CVD suggests low DBP may be a risk marker as there is no clear mechanism linking low DBP with non-CVD events. The randomized design of SPRINT provides some assurance that lowering DBP among adults with DBP <68 mm Hg may be beneficial and is unlikely to be harmful. While randomized trials provide a high level of evidence on the benefits of an intervention, they often do not provide information on who will benefit.15 A single trial has not been designed, nor powered, to investigate subgroups that may be susceptible to low DBP including the oldest-old and those with a history of heart disease. The Blood Pressure Lowering Treatment Trialists Collaboration has data from 52 randomized trials of antihypertensive medication vs. placebo, more vs. less intensive treatment and antihypertensive comparative effectiveness and has reported a benefit of BP lowering regardless of DBP.16,17 The Blood Pressure Lowering Treatment Trialists Collaboration has statistical power to address BP lowering in subgroups; analyses that would be useful for identifying those with low DBP who may benefit or possibly be harmed by BP lowering. In conclusion, the results of the Women’s Health Initiative Long Life Study published in the current issue of the American Journal of Hypertension provide important data on the CVD and mortality risk across a range of DBP levels. The results of this study indicate both low and high DBP are associated with increased mortality. In the context of randomized evidence, it seems unlikely that the relationship for low DBP is causal. However, the results from the Women’s Health Initiative Long Life Study are important in emphasizing the excess risk in those with low DBP and the need for more research into whether there are some subgroups in whom BP lowering might cause more harm than benefit. Drs Muntner and Rahimi have proposed a study to the US National Institutes of Health investigating the effect of blood pressure lowering on cardiovascular disease and all-cause mortality by level of diastolic blood pressure.