If you’ve been keeping up, a couple weeks ago JAMA had a theme issue for Neurology – which nowadays, apparently, is mostly tPA. And, the latest and greatest – concierge Neurology! In which they come to your house to give you lytics.
This is the Prehospital Acute Neurological Treatmentand Optimization of Medical care in Stroke Study (PHANTOM-S), conducted in Berlin, Germany, using the Stroke Emergency Mobile (STEMO) vehicle. They compared time-to-thrombolysis during 46 weeks of standard care with 46 weeks of STEMO period – and, within STEMO period, operation of the vehicle was a week-on/week-off deployment. Unsurprisingly, driving the tPA to the patient shaves 25 minutes off the alarm to tPA time. Success!
MedPage Today, with it’s usual insightful analysis, breaks out a table of glowing secondary outcomes – improvements in in-hospital all-cause mortality, discharge to home, symptomatic intracranial hemorrhage, and overall tPA complications …
… before acknowledging all these improvements occurred even when the STEMO wasn’t deployed, and it was rather general stroke care improvements over the study period reflected in these secondary outcomes. Additional praise is provided by James Grotta, who has started his own mobile stroke unit in Houston. And, finally, Associated Editor Jeff Saver, of endless tPA conflict-of-interest disclosures, chimes in for the Editor’s audio summary.
I think it’s clear, between this and its preceding pilot study, that it is possible to drive a bus around with a stroke neurologist and a CT scanner and rule out intracranial hemorrhage. The main concern might be over-treatment of stroke mimics, but these authors state the same number of patients treated in all observation windows ultimately received non-stroke diagnoses. However, they report a baseline stroke mimic treatment rate of 2.2% – which is line with other literature describing institutions that don’t go looking very hard for non-stroke diagnoses after tPA. Other institutions that require MRI signs of ischemic lesions have stroke mimic rates up to 15.5%, so I wouldn’t place much stock in this specific statistic as a measure of quality.
The last issue – a reasonable case can be made for safety as long as there’s a neurologist riding shotgun in the ambulance. However, you’ll have to find neurologists willing to take such emergency call and support their salaries while they wait for deployment, which will end up being logistically and financially unworkable. The next step, I presume, will be pre-hospital telestroke where paramedics are supervised by a remote neurologist. A bright, or dim, future, depending on your view of tPA.
“Effect of the Use of Ambulance-Based Thrombolysis on Time to Thrombolysis in Acute Ischemic Stroke”
http://jama.jamanetwork.com/article.aspx?articleid=1861800
No wonder can't get a neurology consult they are all driving around in CT equipped vans donated by Genentech
Great comments Ryan.
This study did not look at clinically important outcomes. Their primary outcome showed they could lyse someone about 25 mintues faster when doing it with their fancy ambulance (equipt with a neurologist, paramedic, radiology tech, point of care lab testing, telemedicine to link in to a neuroradiologist and a mobile CT scanner).
Looking at their numbers, it looks like the mobile unit was busy. They were deployed about 6 times per day. But only gave prehospital tPA about ever other day. Based on lets say an enthusiastic NNT of 8, they would get about 2 patients a months get some benefit from their lysis. Of these 2 per month how many get clinically meaningful improvement from an earlier 25 minutes?
I think it is quite obvious that the resource implications and cost for such a dubious return is absolutely prohibitive and could be put to better use elswhere.
What is "obvious" to you, is clearly not obvious to many neurologists – nor the editors of JAMA who thought this article clearly needed worldwide exposure for its exciting innovation.