J-CURVE FOR BP IN HIGH RISK HYPERTENSIVES

Aiming for the lowest blood pressure possible with antihypertensive drug therapy is not necessarily the best approach, at least in patients with coronary disease, diabetes, or other conditions that put them at increased cardiovascular risk, suggests an analysis of two hypertension trials that supports J-shaped curves for outcomes in such patients^[1].

The research, published online in the Lancet on April 5, 2017 indicated that achieving a systolic BP of less than 120 mm Hg was associated with an increased risk of several cardiovascular outcomes, although not including MI or stroke, in the ONTARGET and TRANSCEND randomized trials. A similar pattern was seen for a diastolic BP of less than 70 mm Hg.

The findings seem to show a more nuanced relationship between achieved BP and outcomes than often appreciated, according to the authors, led by Prof Michael Böhm (Universität des Saarlandes, Hamburg, Germany). They "indicate that achieved blood-pressure values have diverse benefit for different outcomes, and this probably differs according to baseline risk in hypertensive patients."

Consequently, "people with a particular risk for a specific outcome—eg, stroke—might benefit from lower blood pressure than those who are more prone to develop myocardial infarction or cardiovascular death," they write. "The challenge, however, is how to predict who is most likely to develop each of these events."

Relevant to SPRINT

Notably, the findings contrast with those based on the SPRINT trial (among others), in which individuals with a systolic BP of ≥130 mm Hg were randomized to intensive or standard BP management. As reported by heartwire from Medscape, SPRINT showed that a "below-guidelines" target systolic BP of 120 mm Hg was associated with significant reductions in CV events and all-cause mortality, by 30% and 25% respectively^[3].

However, questions have been raised as to whether the SPRINT findings should mean that all patients should be at a target of 120 mm Hg; and there is other research supporting a risk-based approach to BP management rather than using strict BP targets.

Overall, Böhm told heartwire when interviewed, the current analysis shows that, if a patient's blood pressure "is dropping too much, it could be a disadvantage," adding that, in the future, the guidelines "should give a range of blood pressures rather than giving only the upper boundary at which the risk is going to be increased."

On one hand, he said, "blood pressure above 140 mm Hg is always associated with an increased risk, so every blood pressure above 140 mm Hg definitively needs to be treated." But there should be closer monitoring, and clinicians should react to a "particularly low blood pressure."

"When you are dropping to 110 mm Hg or so, then there is a signal for safety and therefore blood pressure should not be reduced further, or people might be even taken off drugs to achieve the appropriate goals."

ONTARGET and TRANSCEND 

Böhm and his colleagues conducted a secondary analysis of data from the ONTARGET and TRANSCEND studies, which included a total of 30,937 patients aged at least 55 years from 733 centers in 40 countries. The participants were all deemed at high cardiovascular risk.

Overall, 25,127 patients from ONTARGET were randomized to oral ramipril 10 mg/day (n=8402), telmisartan 80 mg/day (n=8386), or the combination of both (n=8334). The 5810 ACE-inhibitor tolerant patients from TRANSCEND were randomized to oral telmisartan 80 mg/day (n=2903) or placebo (n=2907).

After a median follow-up of 56 months, there was no difference in the occurrence of the primary composite outcome of cardiovascular death, MI, stroke, and hospital admission for heart failure between the treatment groups.

However, the current analysis revealed that a baseline systolic BP ≥140 mm Hg was associated with a greater incidence of all outcomes than a systolic BP of 120 mm Hg to <140 a="" also="" baseline="" bp="" diastolic="" found="" greater="" hg.="" hg="" linked="" majority="" mm="" of="" outcomes="" p="" risk="" team="" than="" that="" the="" to="" was="">

Patients with a systolic BP <120 1.14="" 1.28="" 1.29="" 120="" 140="" a="" adjusted="" admission="" all-cause="" an="" and="" association="" bp="" cardiovascular="" compared="" composite="" death="" failure.="" for="" had="" hazard="" heart="" hg="" hospital="" increased="" mi="" mm="" mortality="" no="" of="" on="" or="" outcome="" p="" ratio="" risk="" significant="" stroke="" systolic="" there="" those="" to="" treatment="" was="" with="">

Mean achieved systolic BP was found to predict outcomes more accurately than baseline systolic BP or time-updated systolic BP, with the lowest risk seen at approximately 130 mm Hg. At an  systolic BP of 110 to 120 mm Hg, there was an increased risk of the combined outcome, cardiovascular death, and all-cause death, but not stroke.

The results also showed that mean diastolic BP <70 70="" 75="" 80="" a="" admission="" all-cause="" and="" approximately="" associated="" bp="" composite="" diastolic="" during="" failure="" for="" greater="" heart="" hg="" hospital="" linked="" lowest="" mi="" mm="" mortality="" of="" on="" outcome="" p="" pretreatment="" primary="" risk.="" risk="" than="" the="" to="" treatment="" was="" with="">

"According to our analysis, in higher-risk patients, achieving a systolic BP less than 130 mm Hg but not lower than 120 mm Hg should be safe for most and result in improved outcomes," the group writes.

"The findings suggest that, in some patients at low systolic BP on treatment, blood-pressure medication might have to be reduced to avoid adverse outcomes because treat to target does not mean treat under target."

In an accompanying commentary^[3], Prof Thomas Kahan (Karolinska Institutet, Stockholm, Sweden) said that while the findings indicate that variations in blood pressure may be associated with prognosis, "how blood-pressure measurements are performed in different clinical settings should also be considered."

He noted that, "in SPRINT, seated blood pressure was recorded in an office free from staff by an automated oscillometric device," adding: "Such unattended automated blood pressure recordings could yield approximately 10/5-mm-Hg-lower values than those obtained in more usual clinical practice."

Another issue Kahan raises is that many patients treated for hypertension do not reach currently recommended target BP, which may be associated with both healthcare provider– and patient-associated factors, including poor medication adherence.

He therefore concludes: "We need better models for increased engagement from both patients and care providers, as improved drug adherence reduces morbidity and death."