番茄社区

ORLANDO -- When the SPRINT trial's were reported Monday, the big question was why did that trial succeed with a systolic blood pressure target under 120 mm Hg when the ACCORD trial failed with intensive treatment to the same target in diabetes.

Both were large, NHLBI-funded trials comparing treatment with a target of less than 120 mm Hg to that aiming for under 140 mm Hg in higher risk populations.

But SPRINT showed a 25% relative reduction in myocardial infarction (MI), other acute coronary syndromes, stroke, heart failure, or death from cardiovascular causes (P<0.001) in a population excluding diabetes, prior stroke, and polycystic kidney disease; whereas ACCORD showed a nonsignificant 12% relative reduction in its primary endpoint of nonfatal MI, nonfatal stroke, and death from cardiovascular causes (P=0.20).

Secondary endpoints also diverged. SPRINT showed benefits to the lower target for all-cause and cardiovascular mortality and heart failure but no stroke reduction; ACCORD found no significant differences in outcomes aside from stroke, both nonfatal and overall.

A Power Problem?

One of the most popular explanations for the differences was that ACCORD was underpowered and would have agreed with SPRINT had it been larger.

"The ACCORD trial, if you look at it carefully, trended toward similar findings in a slightly different patient population but a smaller sample size," said , medical director of the Lee Memorial Health System Heart and Vascular Institute in Fort Myers, Fla., and president elect of the American College of Cardiology.

"One might hypothesize, although it's purely hypothetical, if the ACCORD trial was as large as SPRINT, perhaps there wouldn't be any discussion of a difference in findings," he told 番茄社区.

The blood pressure portion of ACCORD included 4,733 type 2 diabetes patients with clinical or subclinical cardiovascular disease or at least two risk factors, along with systolic pressures of 130 to 160 mm Hg while on up to three antihypertensives, 161 to 170 mm Hg if on up to two medications, or 171 to 180 mm Hg if on no more than one blood pressure drug.

SPRINT included 9,361 hypertensive adults age 50 or older with an average baseline systolic pressure of 130 to 180 mm Hg, with the upper limit for inclusion dropping stepwise from 180 mm Hg on one or fewer medications to 150 mm Hg on up to four medications. They also had to have additional risk for cardiovascular disease, with the intent to enroll patients with the equivalent of a Framingham 10-year cardiovascular disease risk score of 20%, although the actual risk level turned out somewhat lower.

ACCORD, though, had about half the predicted event rate, leading to fewer events and less sensitivity to detect a difference in outcomes than hoped for, , of University Hospitals Case Medical Center in Cleveland, said in an interview. He was first author on the SPRINT paper online in the New England Journal of Medicine in conjunction with presentation at the in Orlando Monday.

An accompanying the NEJM paper -- one of nearly a dozen such simultaneously released articles spread across several journals -- also suggested that statistical power was the reason for variance. Editorialists Vlado Perkovic, MBBS, PhD, and Anthony Rodgers, MBChB, PhD, both of the University of Sydney, Australia, also noted that the ACCORD primary outcome included a higher proportion of events less sensitive to blood-pressure reduction.

In a pooled analysis, they showed that stroke, heart failure, nonfatal MI, and primary composite outcomes as defined in each trial were directionally similar with significant effects of more intensive targets for all but nonfatal MI and no significant heterogeneity between the trials.

The SPRINT researchers likewise noted in an that "the 95% confidence interval for the primary outcome in ACCORD BP included the possibility of a 27% reduction, which is consistent with the 25% cardiovascular disease benefit observed in SPRINT."

"It is possible, but we believe unlikely, that there is an inherent difference in the cardiovascular disease benefits of intensive systolic blood pressure lowering in diabetic and non-diabetic adults," they added. "Since the ACCORD BP trial was not definitive, it would be ethical to conduct another trial of intensive blood pressure-lowering on major cardiovascular disease outcomes in diabetics."

Generalizability in Practice

"I really think this is a global finding," commented , chief of hypertension and clinical pharmacology at the University of Connecticut in Farmington and immediate past president of the American Society of Hypertension. "I would not say if somebody had a stroke in the past or has had diabetes it would not be reasonable to treat them. That clinically would make no sense to me."

However, an independent commentary in Hypertension by Murray Esler, PhD, of Australia's Baker IDI Heart and Diabetes Institute Melbourne, .

"Given the evidence from ACCORD, SPRINT must not be taken to provide a target BP also for diabetic patients with hypertension," he wrote, adding:

"SPRINT did not confirm the disputed therapeutic principle that hypertensive patients particularly at risk, through coexisting illness, will benefit most from adopting a low target blood pressure. Hypertensive patients with coronary artery disease and chronic kidney disease did no better in the intensively treated group than hypertensive patients without these adverse features. By trial design, the knowledge base for treating hypertension in diabetes was not advanced."

Wright said he would personally have no problem generalizing the SPRINT results to his patients with diabetes. "We don't have the clear evidence that less than 120 provides less benefit in diabetics than nondiabetics. We can't say conclusively that the 120 target does not provide benefit in diabetics at this point."

When asked whether that would take another giant trial in diabetes, he said, "That would be the ideal. The practicality of doing a larger ACCORD trial is well beyond my pay grade."

"In the meantime, guideline committees and the medical community will have to decide whether the SPRINT results should be generalized to patients with hypertension and diabetes," the researchers' Hypertension commentary suggested, while cautioning against turning under 120 mm Hg into a performance target.

A NEJM perspective article noted the "major challenges" for clinicians, as even the more conservative 140/90 mm Hg threshold leaves half to a third of hypertensive Americans uncontrolled and many clinicians and patients are "reluctant to go beyond two different antihypertensive drugs" to the average of three averaged in SPRINT, , of Boston University, wrote.

White agreed. "Physicians will have to be fairly creative in ways to get people's blood pressure that low. It will be frustrating for some clinicians and it will be frustrating for patients as well."

"The fact that the study group was able to use the longest-acting and most potent diuretic as well as angiotensin blocker as well as amlodipine, a calcium channel blocker, helped them achieve this goal more easily," he noted.

Azilsartan (Edarbi) and its combination with chlorthalidone (Edarbyclor) were donated for use in SPRINT by Takeda and Arbor Pharmaceuticals but might not help real-world use as much because no U.S. insurer will cover the more expensive branded agents first line, White cautioned. The two branded agents accounted for 5% of all medications used in SPRINT.

However you slice it, "the SPRINT study should generate substantial renewed interest in hypertension as a risk factor -- and that is probably a very good thing, since it has taken a back seat to cholesterol, obesity, and glucose during the past few years and blood pressure remains the most important risk factor for stroke and heart failure events," White concluded.

Comment