This article is more than 5 years old. Information may no longer be current.
Intense drop in BP may confer increased risk for mortality
NEW ORLEANS — Dramatic drops in systolic BP may be linked to an elevated risk for mortality, according to data presented at the American Heart Association Scientific Sessions.
The researchers analyzed the relationship between magnitude of systolic BP reduction and mortality in patients with hypertension but not diabetes.
Peter M. Okin, MD, professor of medicine and director of clinical affairs, division of cardiology, Coleman Center for Electrocardiography, Weill Cornell Medical College, New York, told Cardiology Today that the research was prompted by the SPRINT trial showing reduced CV mortality for patients treated to a systolic BP target < 120 mm Hg, in contrast to an analysis of patients from the LIFE study, which found an increased CV mortality risk in patients achieving systolic BP < 130 mm Hg.
Okin and colleagues said they hypothesized that the difference could be explained at least in part by the SPRINT cohort having a lower baseline systolic BP (mean, 140 mm Hg) than those in the LIFE study.
“Although the inclusion criteria for systolic BP in SPRINT were a baseline of 130 mm Hg to 180 mm Hg, fully one-third of the population had baseline systolic BP < 132 mm Hg, and the second tertile was only 133 mm Hg to 140 mm Hg. So, more than half of the population were already at standard treatment goal for the study before they even went into the study,” Okin said in an interview with Cardiology Today. “What concerned us was that this could be a function of selection bias, where the people enrolled in SPRINT were already a group of people not only able to achieve the standard BP goal, but also able to tolerate [the necessary medications] well. That’s not always the case with hypertensive patients.”
Therefore, he said, it is possible the differences in SPRINT and LIFE “could in part be explained by different behavior according to baseline systolic BP. In other words, we wondered whether patients in LIFE who had lower systolic BPs might be more likely to benefit from lower achieved systolic BPs, whereas those who had higher systolic BPs and had bigger [systolic BP] decreases may not be benefiting as much.”
Okin and colleagues analyzed 7,998 participants from the LIFE study, who were hypertensive, did not have diabetes, had left ventricular hypertrophy confirmed by ECG and, in the original study, had been randomly assigned to a treatment strategy led by losartan or atenolol. Patients were stratified into tertiles based on on-treatment mean systolic BP (< 142 mm Hg, 142 mm Hg to < 152 mm Hg or 152 mm Hg).
Okin and colleagues observed an interaction between baseline systolic BP 164 mm Hg (the 25th percentile of systolic BP in the cohort) and on-treatment BP < 142 mm Hg (χ2 = 15.48; P < .001).
After adjustment for other predictors of mortality and propensity matching, compared with mean on-treatment BP 152 mm Hg, mean on-treatment BP < 142 mm Hg was associated with a 32% increased risk for mortality (HR = 1.32; 95% CI, 1.01-1.65), but mean on-treatment BP 142 mm Hg to < 152 mm Hg was associated with a 24% decreased risk for mortality (HR = 0.76; 95% CI, 0.59-0.98), according to the researchers.
Among patients with baseline systolic BP 164 mm Hg, compared with mean on-treatment BP 152 mm Hg, mean on-treatment BP < 142 mm Hg was associated with reduced risk for mortality (HR = 0.6; 95% CI, 0.36-0.99), as was mean on treatment BP 142 mm Hg to < 152 mm Hg (HR = 0.51; 95% CI, 0.3-0.89), Okin and colleagues found.
“In people who had lower systolic BPs at baseline, achievement of either the middle or lower tertile of BP were associated with reductions in all-cause mortality. In contrast, in patients who had a baseline systolic BP 164 mm Hg, achieving the middle tertile still had a decreased risk for all-cause mortality, however, those in the lowest tertile had a 32% increased risk for all-cause mortality,” Okin said in an interview. “We performed sensitivity analyses looking at 5 mm Hg increment increases in systolic BP, and ... as baseline systolic BP went up ... the benefit went away ... but the increased risk persisted.
“This is a suggestion that maybe we need to think more carefully about this. Our data would suggest that if you start off with very high BP, perhaps achieving a systolic BP in the 140 mm Hg range, the standard therapy range from the past, may be safer than achieving lower systolic BPs,” he said.
Okin said a limitation is that the analysis included achieved BPs, not randomized target BPs, and his team will be conducting more analyses to determine other factors that may have played a role in the findings. – by Erik Swain
Reference:
Okin PM, et al. Presentation 711. Presented at: American Heart Association Scientific Sessions; Nov. 12-16, 2016; New Orleans.
Disclosure: Okin reports no relevant financial disclosures.
Perspective
Back to Top
LIFE enrolled people with thick hearts at baseline. This is usually the result of years of difficult-to-control BP. Thick hearts can be at risk for poor perfusion at the innermost heart layer when BP is reduced, and that may be at play in the LIFE results. These are a selected population, chosen for a specific form of BP-related target organ damage. LIFE was a test of two interventions (losartan vs. atenolol), NOT of two BP goals.
These findings may perhaps have implications in that subset of SPRINT who have ECG evidence of LV hypertrophy at the time of randomization. That is likely to be a relatively small group, but it is probable that the observations of the Cornell group in LIFE may have bearing in the SPRINT population.
An abstract is a “free podium” — it’s a manuscript publication, particularly one validated by independent study in other patient populations, that has the greatest likelihood of influencing clinical practice. Right now, this fits more in the category of “Novel finding – let’s see if it passes peer review and is found in other hypertension cohorts (like SPRINT).”
The SPRINT population was easily managed with BP medications, though it’s not clear that they are directly comparable. I think we’ll see in a few months from now whether there is a significant impact on how to interpret SPRINT results based on this study.