Back in September the NIH tantalized the medical community with a preliminary announcement of the results of a major clinical trial, SPRINT (Systolic Blood Pressure Intervention Trial). The NIH said SPRINT was a “landmark trial” that could “save lives,” but their claims were impossible to evaluate since they only gave the slightest hint of the actual results.
On November 9 the final results of SPRINT will be officially presented at the American Heart Association meeting in Orlando, Florida. Here’s some background perspective along with some of the questions the experts will be asking.
In the trial 9,361 hypertensive patients 50 years of age or older were randomized to standard control of blood pressure, with a systolic blood pressure target of 140 mm Hg, or intensive control of blood pressure, with a lower target of 120 mm Hg. The primary endpoint was the first occurrence of MI, ACS, stroke, heart failure, or cardiovascular death. The main result, as we know from the press release, is that intensive treatment reduced the primary endpoint “by almost a third and the risk of death by almost a quarter.”
But the NIH did not give any information about the reduction in absolute risk. When the results are reported it will be crucial to pay attention to the absolute numbers. And because overall mortality is such a powerful measurement it will be particularly important to look at the total number of deaths and ask whether this is a meaningful finding.
We also know that the trial was stopped early by the Data Monitoring Committee, and that in itself will almost certainly be the subject of a lot of discussion. Stopping a trial early increases the likelihood that the observed result is exaggerated. “We know from prior prematurely terminated trials that the effects size will be larger than what would have been anticipated had it gone through to the finish,” Sripal Bangalore (NYU) told me. Some experts will assume that the actual effect is at least slightly less than the effect seen in the trial.
Another important question will be whether the reduction in the composite endpoint will be caused mostly by reductions in the softer endpoints of ACS and heart failure. I’ve heard at least one rumor that the reduction in the endpoint was due predominantly to the reduction in heart failure hospitalizations.
We know from the press release that there was a reduction in overall death but we don’t know about the reduction in death from CV disease, and we don’t know how large the absolute differences are. If the reduction in mortality is largely due to non-cardiovascular mortality then some may wonder if the finding is due to the play of chance.
There will also almost certainly be extended discussions about how broadly the results of the trial should be applied. Some experts responded to the preliminary results with a call to change guidelines and adopt the lower blood pressure target, but it is not at all clear how the results should be applied to the broad hypertension population. SPRINT enrolled a high risk population, so its results might not apply to the large number of patients who are at lower risk. (But it should be noted that the SPRINT population was not as high risk as the investigators intended, as there were lower percentage of women, elderly patients, patients with chronic kidney disease disease, and patients with CV disease.) For lower risk patients the benefits of intense therapy will be less evident while the risks of adverse events like hypotension and falling may be even more important.
On a related note, John Ryan (University of Utah) wants to know how many patients withdrew from the trial. This will almost certainly have an impact on its generalizability.
Another very important question is why this more intensive strategy worked in SPRINT but not in ACCORD, a smaller NIH trial which studied hypertensive patients who also had type 2 diabetes. Intensive therapy was not shown to be superior to standard therapy in ACCORD, but as the SPRINT investigators point out in an earlier paper, the results of ACCORD “did not exclude a benefit as large as a 27% improvement in the composite cardiovascular disease outcome.”
Franz H. Messerli (Mount Sinai, NY) will be asking several important questions about the details of the trial results. He first wants to know if there were any important differences in outcome or adverse events between patients who started off the trial with only modest blood pressure elevations in the 130-140 mm Hg range and those who started the trial with very high blood pressure levels over 180 mm Hg and who subsequently had a large reduction in their blood pressure.
Messerli is also keen to learn if there is any sort of relationship between the decrease in systolic blood pressure (either absolute or relative) and outcomes and adverse events. In addition, were there important differences in outcome based on the number of drugs taken? Finally, he wants to know if there was any difference in outcome based on the type of diuretic used, either chlorthalidone or hydrochlorothiazide.
- SPRINT website
- Design and Rationale Paper
- SPRINT Protocol
- SPRINT on ClinicalTrials.Gov
- NIH Press Release (September 11, 2015)
- A Premature SPRINT To The Finish Line (My story on how the NIH blew the announcement.)
- Heart Failure Experts SPRINT To An Early Finish (Another news story about premature acclamation of the trial.)