Clot busting drugs are an established part of the tool kit in reducing damage related to strokes. Understanding exactly when to administer them is key to making sure they are used in the most effective way possible. This release describes a study of one mobile stroke unit’s experience administering clot-busting drugs — or tissue plasminogen activator (tPA).
These study findings have not been peer reviewed for publication in a journal and appear to have major caveats around them. We won’t know all the details until the full study is published, and so we were pleased to see some caution expressed in the news release. A few more additions, like cost information and a better explanation of the benefits, would have helped make this release even better.
[Editor’s note: A reference to clot-busting drugs “saving lives” was removed from the review. As noted in the comments section, the scope of benefits offered by tPA is controversial and the American Heart Association withdrew “life-saving” claims about tPA as discussed in these BMJ and Mother Jones stories.]
The recent deployment of a handful of mobile stroke units across the country staffed with specialists and CT scanners makes this a timely release, particularly amidst the backdrop of unsustainable health care costs. The cost of outfitting all US cities with these types of units would be very great, and most of the new mobile units have been funded by private donors. Mobile stroke units could be most useful in remote areas where there is a long transport time from the scene to a hospital that could give thrombolytic therapy, but those areas are the least likely to have enough volume of patients to sustain the expertise of the mobile stroke unit’s staff.
The release briefly mentions the issue of costs with this statement: “Its better outcomes could offset the increased costs of a mobile stroke unit.”
We thought the release would have been better if it had noted the nominal cost of clot busting drugs as well as the cost of a CT scanner and additional personnel essential to mobile stroke units, which would be considerable.
The release does describe the benefits using numbers, but it does so in such a way that readers could be misled. The entire study only included 658 stroke patients. Half received a clot-busting drug at the hospital and half received it in the ambulance. And yet when presenting the benefits, the release presents them as a rate, saying that “three months after stroke, 182 out of every 1,000 patients treated before arriving at hospital were less disabled, including 58 patients who had zero disability, compared to people who received treatment after reaching the hospital.”
How a rate like this based on just 329 patients (half of the total cohort) was derived is never explained, and the description makes it sound like there were thousands of patients in the study.
The release does raise risks in a meaningful way. It says:
“But tPA is not indicated and could hurt a patient’s chances if they are having a hemorrhagic stroke, in which a blood vessel ruptures. That is why patients need a CT scan to confirm the type of stroke before receiving tPA.”
However, that statement could have gone further and noted that some patients with a negative CT scan may have a hemorrhage after getting tPA.
The news release does a reasonable job explaining the basics of the study, though this description falls below some earlier language that has the potential to mislead. It says:
Researchers analyzed results from 658 stroke patients who were treated with tPA — a drug that dissolves blood clots. About half of the participants received the clot-busting drug at the hospital, and half received it while still in the ambulance.
It also notes that the findings were “presented at the American Stroke Association’s International Stroke Conference 2017.” What’s missing from this is the fact that the findings likely only received limited peer review, since they were not published in a journal.
Very low in the release — but at least it was included — it says that the findings did not quite achieve statistical significance. At least that’s how we’re interpreting it. The statement itself is a bit vague, saying,
“The findings bordered on statistical significance, suggesting that future clinical trials with a greater number of patients are needed to show similar benefits.”
The lack of statistical significance, even with caveats, merits a Not Satisfactory for evidence. Moreover, the release doesn’t tell us why in this apparent re-analysis of the results, only about two-thirds of the patients from the original study were included.
There is no disease mongering in the release.
The release explains the funding source and points to the abstract, which notes that one of the authors received a modest grant from a pharmaceutical company for a clinical trial for a clot-busting drug.
The release does an OK job discussing how the approach being studied compares with the standard of care. It says:
“Every second counts, in the current standard of care, patients who experience stroke-like symptoms and call 9-1-1 arriving to the hospital by ambulance are assessed by clinical examination and imaging (CT or MRI scan) in the Emergency Department. This takes a certain amount of time from the patient’s symptom onset which prompted the 9-1-1 call.”
Availability is not addressed beyond UCLA’s Mobile Stroke Rescue Program, which collected the data for the study.
It would have been good to mention that specialist-staffed mobile stroke units have been popping up in different parts of the country over the last 3-4 years. They are mainly affiliated with larger medical centers and funded by private donations.
The release does not adequately establish the novelty of the study or the findings. It does briefly allude to previous research: “Past research showed the sooner that a clot-busting tPA drug is given after an ischemic stroke – one in which a clot is blocking blood flow – the better patients fare.”
There have been around a dozen trials on mobile stroke units in recent years, appearing primarily in neurology journals. Some examples:
The language is generally cautious enough in the release. The errors are mostly of omission.