Thoughts on Lights and Sirens: Stroke

I hardly ever go lights and sirens to the hospital. I feel so strongly about not going lights and sirens unless absolutely necessary, I wrote what became our statewide policy on lights and sirens. Although it was toned down through the various committees it went though, the gist of it remained the same. You should only go lights and sirens to the hospital if the hospital can do something in the minutes saved by going lights and sirens that you can’t do that will make a difference in the patient’s mortality or morbidity.

Here’s how it was eventually worded:

When transporting the patient utilizing lights and sirens, the need for immediate medical intervention should be beyond the capabilities of the ambulance crew using available supplies and equipment and be documented on the patient care report.

I used to go lights and sirens quite a lot when I first started, but then I began to wonder. Here I am going lights and sirens, making cars veer out of my way, and running the risk of someone slamming into me, or someone else and for what? To get ahead in the triage line? Or to get to a room to turn the call over to a nurse, who I can’t even find to be seen by a doctor who has four other patients to see first?

Sometimes when I worked at night years ago, we went lights and sirens on bullshit calls, just to drop the patient off in the waiting room to clear to take the next holding call. That seemed more reasonable to me than going lights and sirens to wait in the line.

I posted that once on an early internet EMS list serve and got slammed for it — and with good cause. I don’t do that anymore. I recognize now getting into an accident going lights and sirens for a patient with a smashed toe is not good form. Mea culpas galore.

Here’s what I have been going lights and sirens in 2015:

Major trauma with physiological changes. Decreased GCS, penetrating trauma, hypotension or tachycardia, amputations.
Refractory anaphylaxis.
Pale, cool diaphoretic, abdominal pain. Thinking AAA or ischemic bowel or another surgical emergency
Refractory seizure
Refractory CHF not responding to CPAP and NTG
Major stroke*
Others depending on unique circumstances (like someone who I think is going to crash)
Cardiac arrest (only sometimes).
Seriously impending childbirth. (Some people like to deliver babies, but I think the baby deserves an OB team more than just a paramedic, particularly when things go bad.)

I call your attention to itme 7 on the list — major stroke. Why not minor stroke? Good question. I acknowledge that I should be going lights and sirens on these patients, but I have been having a hard time, actually doing it. I make excuses. “Let’s just get going I will tell my partner, but keep it on a two. I’ll do everything on the way.” My partner on this day is new and drives rough and doesn’t always know where he is going. I justify my decision that it is a safety issue, which trumps all.

I have a dual role when it comes to stroke. I work at a stroke center and collect data. One of the data fields I collect is whether or not the ambulance went to the hospital lights and sirens. I was shocked to discover EMS only transports a little more than half of stroke patients on a priority, even though we have all been taught — time is brain. They say you lose 32,000 brain cells every second in stroke. I am not talking about unresponsive Cincinnati 3 here. I am talking about the patient with a mild facial droop and arm weakness, who is hemodynamically stable. I ask medics who have called in stroke alerts why they didn’t go on a priority. 32,000 brain cells a second. I remind them. We are on the clock –the patient is still in the tPA window. They shrug. Maybe it’s the fact it takes hospitals so long to give tpa — the goal benchmark is 60 minutes. The time it takes to give it is not unreasonable — the patient needs to be scanned, thoroughly evaluated by Neurology and then have a conversation on the risks and benefits of tPA (IT MIGHT KILL YOU!). If they get tpA within 52 minutes instead of 50, does it matter?

If you buy into tPA (not everyone does) and you buy into our stroke system (which is build around tPA) then we all should be going lights and sirens on even minor strokes. It was shocking to me that 60% of our EMS patients who were recognized strokes and got tPA, were transported nonpriority.

Like many time sensitive interventions the data shows the sooner people get it the better it works. The longer time passes, the greater the risk until the point at 3-4.5 hours (depending on patient) when the risk exceeds the benefit.

Based on the data I collected (showing area medics reluctance to go on priority) we added the following to our regional stroke guideline:

Try to limit scene time to 15 minutes or less, and transport rapidly. Transport should be equivalent to trauma or acute myocardial infarction calls.

I haven’t had a stroke for awhile, so I haven’t been tested, but other medics keep resisting. It’s odd.

Looking at the dispatching side, we noticed early on that only 70% of strokes were being dispatched on a priority, but in certain areas, it was even lower. We looked into those towns and found their dispatch centers were dispatching STROKE (Card 28) in Medical Priority Dispatch non lights and sirens. Reading the card was somewhat shocking.

“STROKE must receive an immediate response that is not subject to delay, lights and sirens are not recommended; however there should be a sense of urgency.”

It is important to note that someone having a massive stroke, leading to say, being unconscious, would be coded out under another dispatch card such as UNCONSCIOUS (card 31) and receive automatic lights and sirens response.

The way EMD works is the EMD system makes recommendations but the medical director makes the final decisions. Stroke coded out as a C or Charlie response and our medical director had ambulances going cold on most of the Charlie calls. We had our medical director, who to his credit is very conservative with calls he will have ambulances go lights and sirens on, change the possible stroke response to lights and sirens. We saw an almost immediate change in times. It seems the historic rationale for the non lights and sirens response on hemodynamically stable strokes is that while stroke was time sensitive, the few minutes saved by going lights and sirens was not worth the risk.

I know there are studies that say using lights and sirens only saves a few minutes. True, maybe when all responses are combined, but there are times of the day when without lights and sirens, you wait forever at strings of lights. There are clearly outliers where lights and sirens will save you 10-20 minutes. That is a lot of brain cells when it comes to possible stroke, and can be the difference in whether or not someone gets tPA. Not only that but tPA, if you believe the studies, shows a better effect the sooner it is given. 1 hour is better than 2, 2 hours is better than 3. After that, the considerable risks outweigh the benefits.

Maybe EMS is reluctant to go lights and sirens on milder strokes because EMS doesn’t get the follow-up on stroke patients. With STEMI, you either go up to the cath lab or you learn the door to balloon time. With trauma, you see the response and when you come back from writing your form, they are up in the OR. Maybe if we provide better followup to EMS, we will come in quicker? I have been a medic over twenty years and I cannot point to one of my patients that I know got tPA. I am sure some of them did, but no one told me, and I never followed up. And it is not like every stroke patient gets it. Our best quarter 20% of ischemic strokes got it, which is quite good. Some hospitals only 1-2% of stroke patients get the drug.

And maybe EMS is reluctant to go lights and sirens because the outcomes are so poor. Even if you believe tPA works, all it means is that the patient has a 10-30% chance of being moderately disabled versus severely disabled. The push tPA and there is the sudden hallelujah moment where the patient can walk and see and talk is largely a myth. Sure, some people can suddenly become better, but it is more likely if they are waking at that moment that it is the natural progression of their individual stroke/TIA and their reperfusion is spontaneous. Even the positive tPA studies show no difference at 24 hours between those who get it and those who don’t.

Over the years I’ve had a number of patients who were completely stroked out, who awoke after I had called in my stroke alert. They just had giant TIAs with spontaneous reperfusion. What if they had been given tPA? Would it have been the tPA or the natural reperfusion? And what if, in getting tPA, it caused them to bleed in their brains and die?

Bottom line, despite my reservations about tPA, I am going to try to go lights and sirens on my future stroke patients, because that is how our system is setup, (tPA is considered a Level I AHA intervention) and 32,000 brain cells a second is our responsibility. I say get them to the hospital quick, and let the neurologists do what they think is best. And if they get tPA long after I am on another call, I hope my lights and sirens transport, made a difference, even if small, to a fellow human’s outcome. I also hope I didn’t cause any accidents on the way to the hospital. Some of the newer people I work with are not the most experienced drivers. I have had a number of intercept medics tell me they won’t go lights and sirens strictly because they don’t trust the driver. I admit to being in that situation at times.


Update: I responded the other day to an elderly man with dementia who had suddenly lurched to the right, and then while he did not fall, was observed unable to move his right arm. I palpated all along it for trauma and elicited no response. He could squeeze his left hand on command, but not the right. He failed the pronator drift, but had no facial droop or speech problems. He was elderly and was hypertensive — 180/100. I would have been more confident in my assessment if I could have had a conversation with him, but I was stuck with his limited ability to converse. His watchers said the right arm was completely not normal for him. I was only a couple miles from the hospital, but I did call it in as a stroke alert, and I did go lights and sirens, although I told my partner to make is an “easy 1”. The patient got a quick neurological exam, and then was sent right to CT scan as a possible stroke. I left for another call, and never got any follow-up.

* Now modified to stroke


For True to Life EMS Fiction, order Mortal Men Today.

Mortal Men: Paramedics on the Streets of Hartford


  • Just a Medic says:

    You’re right to acknowledge the dismal data and the controversy surrounding tPA. I’ll tell you a bit about the numbers in my area of rural southern Indiana. Most patients who roll into the hospital as a “stroke alert” were dispatched first as an “elderly patient, altered mental status.” For fairness, we need to consider the set of all Elderly-AMS patients as potential “tPA candidates.” Among the Elderly-AMS runs, the EMT/medic weeds out about 70% of the “candidates” by identifying a specific pathology other than stroke (hypoglycemia, trauma, cardiac, sepsis, etc.) Then at the hospital, the triage nurse weeds out the next 90% or so of “candidates” by last-seen-normal time, either unknown (most common) or too long ago. A CT scan of the brain weeds out another 20% or so of “candidates” by identifying “foci that might constitute a tiny hemorrhage in the correct clinical setting,” in the words of the radiologist. Then the emergency physician gets a crack at the patient; applying the lengthy exclusion criteria from NINDS and contraindications printed on the drug label weeds out another 90% of “candidates.” Next, the emergency physician calls a neurologist for a telemedicine consult. (None of our area hospitals have neurologists on staff.) The neurologist advises “risks exceed benefits” about 70% of the time, thus weeding-out another 70% of “candidates.” Finally it’s time for the family discussion. The ED nurse looks for a family member/loved-one who is present, of legal age, and sober, and so weeds out the next 50% of “candidates.” The ED physician sits down with said family member, discusses risks and benefits, including the possibility the patient might get better with no treatment, and the risk tPA “MIGHT KILL YOU,” and the family member’s decision culls another 50% of “candidates.” Now tPA is administered to the “candidates” who remain and we wait 30 days for re-assessment. At 30 days, 18% of the population is doing better thanks to tPA, while 82% is no better (maybe worse) thanks to tPA.

    So how do the odds stack up? Among the Elderly-AMS calls my crew sees every day, 0.0032% of them will benefit from tPA, actually experiencing brain cells saved. That’s maybe 1 patient among the 30,000 patients or so I expect to see during my lifelong career as a medic. It’s tough to get excited about those odds. This is the reason I choose not to endanger the safety of myself, my crew, and other motorists by transporting “stroke alerts” emergently.

  • medicscribe says:

    Well put. It is hard to argue with your logic. This is an issue where I find myself split in two. I sit at my coordinator desk in front of my spreadsheets and want to put the checks in the lights and sirens category to show we are following the AHA guidelines. I want to be able to show that our door to needle times are decreasing and every minute counts. Then I get in the ambulance the rest of the week and have a hard time telling my brand new partner to “fire up the lights and sirens” when I am even more worried that they don’t even know how to get to the hospital in the first place. As far as tPA, I wish they had done better studies so we would really know if it helps or hurt, or know better who it helps and who it hurts, Thanks for the great comment! Peter

  • Denica says:

    Thanks for the article. I agree as a 23 year Paramedic about what’s appropriate and not appropriate for lights & sirens. I read new drivers the riot act when they are driving for me (lights & sirens) because those few times you need to, that means your extra busy in the back & not strapped in (scary).
    At our service we do take stroke into the stroke center lights & sirens (usually BLS takes them).
    On a person note, my mom suffered a major stroke & got tPA. She when from classic right sided paralysis, no speech, facial droop etc, to getting tPA and being 100% better in the ER. She asked if she could go home now & everyone laught. Unfortunately 3 hours later in ICU she re-stroked but at least kept her speech that time & has been in a wheelchair since. So this is close to my heart.
    Thanks again

  • Denica says:

    Sorry, I just saw all my typos in my post. I couldn’t see what I was typing… No need to post it

  • StrokeMD says:

    Lights and sirens should always be used. Every minute counts (1.9 million neurons estimated lost/minute). The evidence for tPA is strong, but the evidence for interventional therapies is even stronger. There are several large RCTs (published January 2015) showing that these interventions not only reduce morbidity, but also mortality — it literally saves lives. Also, minor strokes can become larger ones when untreated; would leave the decision re: interventions to the discretion of Neurologists — bring them in as fast as you can to give them the best chance at a good outcome!