Flawed Studies Define This Year's International Stroke Conference

Hans-Christoph Diener, MD, PhD


March 07, 2017

This feature requires the newest version of Flash. You can download it here.

Dear colleagues, I am Christoph Diener, a neurologist from the department of neurology at the University of Essen in Germany. My topic today is the International Stroke Conference, which was held in February in Houston. The congress was well attended with about 4000 participants, many from China and Japan, and almost no one from the Middle East—for obvious reasons.

My bottom line is that you can only attend this conference if you are on antidepressants. Basically, almost all of the presented studies were negative. This is partly due to the medical problem being studied, but many of these trials also had serious design problems or were underpowered. I will give you a few examples.

Results were presented from HeadPoST (Head Position in Acute Stroke Trial),[1] in which more than 11,000 patients were randomized to two head positions after acute ischemic stroke: either 30° up or lying flat. The hypothesis was that if you have a severe stroke, lying down most probably could lead to a brain edema and pneumonia. The trial's endpoint of functional outcome at 90 days was negative. Why? They included the wrong patients. The majority of patients had moderate strokes, and this is a question that should be investigated in severe strokes.

The ASTER (Adapt versus StEnt Retriever) trial[2] was performed in France in 381 patients with occlusions of the internal carotid artery and the middle cerebral artery. They compared contact aspiration with the Penumbra system versus a stent retriever. As you would expect, the recanalization rate was identical.

The ASSORT (Administration of Statin on Acute Ischemic Stroke Patient) trial[3] in Japan looked for secondary stroke prevention at early versus late initiation of statin after stroke in 270 patients. Half of the patients received a statin immediately and the other half after 7 days. With a sample size of 270 patients, you would never expect a difference. This trial was vastly underpowered.

The TARDIS (Triple Antiplatelets for Reducing Dependency in Ischaemic Stroke) trial[4] investigated triple antiplatelet therapy versus standard antiplatelet therapy in people who had an ischemic stroke. The sample size was 4100 patients. I don't think you're surprised to hear that this trial had to be stopped prematurely because of an increase in major bleeding. The trial was negative for the primary outcome of a modified Rankin scale at 90 days, but there was a 2.5 times increase in major bleeding with triple therapy.

We've known for a long time that dual antiplatelet therapy has a much higher bleeding risk than monotherapy. How could we expose patients to triple therapy with this knowledge? We have to stick with monotherapy long-term and perhaps combine it with aspirin, depending on the results of the ongoing POINT trial (Platelet-Oriented Inhibition in New TIA and Minor Ischemic Stroke Trial), which looks at the combination of aspirin and clopidogrel for 90 days.

The SOCRATES (Acute Stroke or Transient Ischaemic Attack Treated with Aspirin or Ticagrelor and Patient Outcomes) trial had an interesting subgroup analysis.[5] Interestingly, this was the only positive result. This was not presented in the last clinical trial session. You might remember that the main SOCRATES trial[6] randomized people with high-risk transient ischemic attack and mild strokes to ticagrelor or aspirin for 90 days. For the primary endpoint, the study was negative. It almost missed significance.

Dr Pierre Amarenco and colleagues[5] looked at the 23% of the patients who had atherosclerosis in the carotid arteries. These patients had a clear benefit from ticagrelor over aspirin. That is logical. If you want to have a more potent antiplatelet work, it should be used in people with atherosclerosis, not in people with lacunar strokes.

There was also a trial that looked at nimodipine in mild cognitive impairment.[7] In this China-based study, 654 patients were treated for 6 months with nimodipine or placebo. As you would expect, the trial was negative. Why? We never could show that nimodipine is effective. The only trial that was ever done in acute ischemic stroke was fake. [Editor's note: According to the author, this information was never made public.] An observation period of 6 months is much too small. Any trial that wants to look at improvement in cognitive impairment needs to run for 4 or 5 years. So, the trial was too small and it was too short.

Finally, there was the SPOTLIGHT (The Spot Sign Selection of Intracerebral Hemorrhage to Guide Hemostatic Therapy) study.[8] There is a so-called spot sign on CT if you have an intracerebral hemorrhage. If you give contrast medium and you can see a small spot of contrast, this basically indicates that the patient has a high risk and the bleeding will increase.

This is why a study in Canada[8] was done where people with positive spot signs were randomized to recombinant factor VIIa or placebo within 8 hours. The trial had only 142 patients. As you would expect again, it was negative, as was another major trial done a few years ago with factor VIIa.[9]

Ladies and gentlemen, in conclusion, there were lots of negative trials, mostly due to problems with the design, the scientific question, and how the studies were conducted.

I am Christoph Diener, stroke neurologist from the University of Essen in Germany. Thank you for listening.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: