By Dr. Geoffrey Modest
There was a thoughtful editorial in the NY Times by Eric Topol and Harlan Krumholz about the early release of results from the recent NHLBI-sponsored SPRINT study (Systolic Blood Pressure Intervention Trial), a trial which found that the target systolic pressure should be lower than in recent guidelines (see NIH release at http://1.usa.gov/1RbAdyh ). The editorialists comment that this study as released does not have enough details to be useful, since it does not include specifics on the methodology and results to allow for adoption into practice (see http://nyti.ms/1KqZrmH ). Here are the details of the study in brief, just for reference (I will not review this study until the full report becomes available).
- 9300 participants >50yo with at least one extra cardiovascular risk factor (presence of clinical or subclinical cardiovasc disease other than stroke, eGFR 20-59, Framingham Risk Score with10-yr risk of >=15%, or age>75), randomized to a systolic blood pressure <120 vs <140 mmHg. Patients were recruited from 100 medical centers in the US and Puerto Rico
- Study was expected to run 8 years until 2017
- Study did not include diabetics, those with prior stroke, albuminuria> 600 mg/day or eGFR<20 (and a few other exclusions)
- On average those with target systolic <140 were on 2 meds ; those with target of <120 were on 3 meds
- The study was terminated early after only 6 years of the planned 8 years because of a 30% reduction in heart attacks and strokes, and a 25% reduction in deaths.
- Primary results will be published “within a few months”. And more data will be collected over the next year to assess kidney disease, cognitive function and dementia
This is obviously an important study, since hypertension is so prevalent (79M adults in US, or 1 in 3, have high blood pressure and ½ of those treated for it still have systolic >140). However, I have real concerns about the early release of the above information, from which I think it is impossible to draw real conclusions for clinical practice. For example, there are no data on the actual numbers of people who benefit from the aggressive intervention (they include only relative risk reduction with no details on the absolute numbers of patients who benefited), we don’t even know what the achieved blood pressures were for the 2 groups, we don’t know what meds were used (i.e., were those on the 140 group on more b-blockers, which may be less effective clinically?), and perhaps most importantly they do not comment on the adverse effects of aggressive treatment (I have seen lots of people, especially elderly, who have significant orthostatic hypotension despite systolic blood pressures above 140, likely attributable to autonomic neuropathy. And I have real concerns about falls, esp. if the weather is hot, they do not drink enough, or they have diarrhea, etc. leading to even lower orthostatic blood pressures).
Topol and Krumholz comment that in this electronic age, it is essential for important study results to be available as soon as possible, but that the method of release needs to change: the preliminary details/raw data should be available for review to allow for a fuller understanding of the actual study and its clinical relevance. But, I think this suggestion should be a lot stronger, not just releasing the data (which may be hard for us in primary care to interpret fully). So, I would add a few points:
- There actually has been an expedited process for decades, which worked quite well. The high-profile, high-relevance studies were released within 1-2 months (to my memory) in a complete form in the prestigious medical journals, along with editorials. It is quite striking to me the SPRINT study may not be released in a usable form for many months. I personally would be willing to wait the month or two to see the study in a clinically useful form. Perhaps this existing expedited process (now mostly as articles released “online first”) should be tweaked to accelerate the process a bit.
- I am also very concerned about these high profile studies making it prematurely into the popular media, along with hyperbolic quotes from prominent physicians (in the SPRINT case, Dr Mark Creager, president of American Heart Assn, stated “it is outstanding news” in the NY Times on 9/12 — see http://nyti.ms/1YKksTO). My concerns are:
- Patients who see these stories are likely to ask their providers for more aggressive blood pressure treatment, perhaps pushing us to treat some people inappropriately.
- Providers may themselves think we should change our practice, given the enthusiasm of the “academic leaders” in promoting the trial, without our really knowing what happened.
- Another side issue, especially since the vast majority of these “blockbuster medical studies” are funded by drug companies (the SPRINT study, at least from what I can discern, was not) is that sometimes the excessively long release could be used to the advantage of the drug companies. I don’t think I am too paranoid to wonder why the IMPROVE-IT study took 7 months for the final release. It was presented as the block-buster highest profile study of the American Heart Assn annual meeting (which, in every other case I know, comes out simultaneously in the New England Journal or similar journal); had even more initial superlative reviews in the general news media prior to its publication than the SPRINT study, supported by many of the big names in cardiology as a huge breakthrough; and led to a huge profit for this drug-company sponsored study by getting large numbers of people on ezetimibe, and once on it is likely to have these patients continue on this mediocre and expensive/profitable drug. See https://stg-blogs.bmj.com/bmjebmspotlight/2015/06/23/primary-care-corner-with-geoffrey-modest-md-improve-it-trial-ezetimibe/ for my blog critiquing the study.
So, my bottom line is that important studies need to be released in a complete, interpretable form and as quickly as possible. The above NIH release, I think, does us all a disservice. It adds opacity and not clarity to an important and complex issue. We’ll see what the study actually found in a few months…..