Narcan Nebs - Why? New research doesn't say..

When I learned about naloxone years ago, I went through a few emotional stages. First, I was amazed that such a drug, the perfect antidote existed. Hey, we don't have any such drug for cocaine or marijuana, and especially not for alcohol. How cool! On the other hand, the first time I gave an aggressive dose of naloxone to an opiate-addict, I was really regretting it for a long while.

My patient was also a marionette.

So this recent study holds out the promise of a great middle path - treatment of opiate OD, but without the potential over-correction of bolus dosing, whether by IV, IM, SQ, or IN routes. However, I wonder if the results address a question that anyone has actually asked.

Another way to put it - after reading the paper, I'm not sure what was treated, and whether it (whatever it was) required treatment. 

Also, I'm not clear on what the outcome of treatment was.

Can Nebulized Naloxone Be Used Safely and Effectively by Emergency Medical Services for Suspected Opioid Overdose? is a retrospective study of the use of nebulized naloxone by EMS in Chicago. The methodology is, as usual, important to understanding the conclusions.

Methods:
The Chicago FD EMS system had already implemented the use of nebulized naloxone (2mg in 3ml NS) for various indications, excluding patients in shock or who were apneic. The researchers reviewed the EMS patient care reports of any patient who was given nebulized naloxone for any reason, including "suspected opioid overdose, altered mental status, or depressed respirations."

That's the first item that bears some scrutiny - these indications are vague, and reflect some different conditions. Some of these conditions require naloxone, others don't.

Do all patients with suspected opioid overdose require naloxone? Probably not.

For that matter, do all patients with altered mental status require naloxone? Likely no.

Respiratory depression, of course, is a very good reason to give naloxone, but no definition is provided; Respiratory rate? Hypoxia? Hypercarbia?

Next, what was the outcome they were studying?

The researchers looked at the PCRs to determine if the response was "complete, partial, or no improvement." This is an unfortunate choice of an outcome measure, for two reasons.

The first is that it begs the question - what was the desired response? Are we aiming for clarity of speech, or resolution of hypoxia, or reversal of miosis?

Mission accomplished.

The second reason this makes for a poor outcome measure is that the medics are not "blinded" to the treatment the patient got - they all knew they were giving an active agent. Now, if the outcome was intubation, oxygen saturation, or some other "hard" number, this might matter as much. But given the vague inclusion criteria, this matters a lot. It biases the paramedic when they're examining the patient for a response. and would predispose them to to "see" an effect that isn't really there.

Results:
When we look at the results, we can't be sure what to make of them, given the study design. Out of 105 patients, naloxone was given to:

Suspected opioid overdose      70 patients (66.7%);

Altered mental status               34 patients (32.3%);

Respiratory depression              1 patient (0.9%).

This is an odd collection of diagnoses, as only the last item is viewed by toxicologists as an indication for naloxone. "Suspected opioid OD" can be confirmed through urine, serum, or just asking the patient later - it isn't a diagnosis that in itself requires treatment.

"Altered mental status" may be due to the use of opioids, but it may also be coincidental. For example, Peter Canning describes a case in which he gave naloxone to a patient who had a subsequent return to full LOC. It turned out to be entirely coincidental - the patient actually had an intracranial hemorrhage.

http://xkcd.com/552/

There are so many potential causes of altered mentation, that the "knee-jerk" use of naloxone isn't likely to be very helpful. This was shown in a Pittsburgh study from 2009: The routine use of naloxone by EMS for "depressed mental status" didn't help in 92% of cases.
So, what was the response to the nebulized naloxone in this study?

Complete response                    22%

Partial response                         59%

No response                              19%

This is the frustrating part - they don't say what the responses were (Increased RR? Wide pupils? Screaming and vomiting?). Also frustrating is that the majority of response were partial, which can be really subjective. They try to break down the results in the table:

...which doesn't really show any meaningful differences between the complete/partial/none-responders.

In fact, it shows an interesting similarity that all 3 groups have: the average RR was above 12/minute. In many patients, it appears to have been significantly higher- the "No Response" group had an average initial RR of 18.

The bible of toxicology describes the goal of naloxone administration as restoring "adequate spontaneous ventilation," not the reversal of slurred speech, sleepiness, or tiny pupils. Frankly, it's also not a medical goal to thwart someone's high. Nonetheless, at least one toxicologist feels that nebulized naloxone is an "attractive alternative."

My take on this study:

You can nebulize anything, and it was shown here that paramedics could nebulize a clear liquid containing salt and naloxone. 

It didn't show us if it worked, because almost 80% of the responses were "partial" or less. 

Even with a 22% rate of "complete" response, we don't know what to conclude as they didn't show us what they were trying to treat.

Review of the protocols:
Naloxone is only mentioned in the altered mental status guideline - there is no indication under the "Overdose/Poisoning" or "Respiratory Arrest" sections.

You generally don't want to reverse opioids unless you find coexisting respiratory insufficiency. Not a whole lot of benefit, and enough potential for adverse effects!

For example...

Enjoy!

Prehospital Analgesia - Recent Research

Enough cardiology - time to come back to a topic that's a lot closer to my heart.

Given the relative safety of morphine, the reversibility, the low cost, as well as the immediate and clinically-apparent benefit, I've always been interested about the hand-wringing and strong feelings that surrounds the use of this medication. Frankly, I'm more worried about giving Lasix to someone who turns out to have pneumonia, or overdosing an atrial fib patient on diltiazem.

I wanted to review some recent results in the literature, highlighting a few aspects. While giving any medication requires thought and attention from the paramedic, I think some of the results should make paramedics feel more comfortable with breaking out the box o' narcs.

Should have a lock on it. Two locks, really.

How much morphine can we give?

The first paper, while not that recent, points out safety and effectiveness of the aggressive use of morphine. Is there an ideal morphine dose for prehospital treatment of severe acute pain? describes a study done in a French EMS system. That system uses physicians in place of paramedics, but I'm not sure that matters so much for this topic.

They randomized patients to 2 different IV morphine dosing strategies:

Group A - Start @ 0.05 mg/kg, then 0.025 mg/kg every 5 min.

Group B - Start @ 0.1 mg/kg, then 0.05 mg/kg every 5 min.

Pretty strong doses; a 220 lb guy in Group B would get an initial bolus of 10 mg of morphine. The endpoint was getting the pain score below a 3/10, and Group B appeared to get there quicker (no surprise).

The interesting part was the rate of adverse events:

Basically no difference. An extra 3 patients in group B vomited, and one person had their sat drop to 92% - nobody got intubated or became hypotensive. Pretty safe, and very effective.

Great, but that was a study setting. What about in the "real world?"

In a follow-up study that was published last year, they looked at the "real-world" use of morphine and sufentanil in the same EMS system. It was a prospective trial, but just observational. Patients got at an average initial dose of morphine of 0.083 mg/kg, but not so many titration doses. Again, they found that this kind of dose was very safe, as can be seen in their Table 3.

Looking at the Greater Bridgeport Sponsor Hospital protocols, with a max of 0.1 mg/kg of morphine (before talking to med control), you can feel pretty confident giving an initial bolus of 5 mg in adults!

Well, morphine is going out of style. How about fentanyl?

If you read a recent post by RogueMedic, you know that fentanyl is supposed to be pretty strong, so we might expect that it has a high rate of hypoxia and hypotension. In fact, though, it appears to be just as safe as morphine.  

(Well, not always - there is one way that fentanyl can cause respiratory failure that you may not have prepared for:)

Fentanyl patch stuck in the left bronchus: Reference

The study Fentanyl in the out-of-hospital setting: variables associated with hypotension and hypoxia looked at the experience of an aeromedical service. They used relatively high doses of fentanyl, and medics could give up to 5 µg/kg total every hour. For those of you who don't have experience with this drug, that's about equivalent to 0.5 mg'kg of morphine every hour, or 35 mg for a "average" weight patient. The researchers checked the vital signs before and after drug administration to catch any problems with hypotension or hypoxemia. 

They gave plenty of opioid: The average inital dose was 1.1 µg/kg,while the average total amount given during transport was 3 µg/kg (or 0.3 mg/kg of morphine-equivalents).

Hypotension? Not so much. 

Systolic blood pressures did not, on average, change after giving fentanyl. There were some people, about 5% of all the patients, who had hypotension after getting fentanyl, but about half of those patients were hypotensive before they got the drug as well. 

On the other hand, some hypotensive patients (about 50%) had their blood pressure increase after fentanyl! Go figure...

As for hypoxia, none of the non-intubated patients became hypoxic, 0%. 

So, what can I do with this information on my next shift?

Most importantly, remember that morphine is a pretty safe drug. Looking over the doses they were giving in these studies - pretty aggressive. Now, there is always a risk of some rare event happening, and your patient drops their pressure to 70 and stops breathing. But rare events can happen with any of the drugs in your kit, not just the narcs. The rare patient will have an anaphylactic reaction to aspirin, but it is still a good idea to give it (assuming you've asked about drug allergies).

 

Strange case of pediatric syncope

Faints (not feints)
Frankly, a kid who faints is not the most interesting patient. Young ladies have a habit of swooning, especially if Justin Beiber is involved. As for young guys, they are hardly immune:

So it it usually isn't too exciting to be called for a 14 year old girl who had fainted. Usually.


The Patient
During the evening she had been feeling an upset stomach, describing some epigastric discomfort, as well as nausea. Her parents remember that she looked a little pale before the episode ocurred. She was walking into the kitchen when she began to feel faint, and was seen to fall onto the floor. Her parents, frightened, called 911 immediately.


The ALS crew found her sitting up in a chair, a tissue held up to a small cut on her forehead, but otherwise looking well. Vital signs were normal, and the parents were starting to feel like they had been overreacting in calling 911. While sitting in the chair, however, she suddenly stiffened up her whole body, and then went limp. After being put on the floor, she regained consciousness. Although she came around quickly this time as well, everybody decided that calling 911 had been just fine, and she was moved to the ambulance for transport to the ED.


During transport, the medic noticed that her heart rate was slowing down episodically, dipping into the 40's. She still looked pale, and vomited a few times, but her mental status was fine, and she never became hypotensive. A rhythm strip was obtained:

An ECG was also acquired:

The transient episodes of bradycardia did not require pacing or medication. She was brought into room 4 (our trauma/acute resus room), looking a bit intimidated by the number of nurses and doctors around her. She still looked pale, but her speech and mentation were normal. Naturally, we acquired an ECG immediately (about 30 minutes after the EMS ECG):

Vital signs were P-100, R-20, BP-127/59, SaO2-100% RA. Her lab work was uninteresting, and a CT of her head show no fractures or bleeding. The cut on her face needed only some Dermabond.


ECG analysis
The ECG and rhythm strip obtained by EMS were acquired only a minute apart. The rhythm is interesting, showing a sinus rate of around 120, but with a 2:1 AV block, so that the ventricular rate averages 60. On the rhythm strip we see two episodes of more advanced AV block, with 3:1 conduction, while on the 12-lead we see an apparent Wenkebach pattern in complexes 5-9, with a progressively lengthening PR interval. The QTc, fortunately, appears normal, and there are no signs of pre-excitation, Brugada, or arrhythmogenic right ventricular dysplasia.


The ED ECG shows, first of all, how frakkin' important it is for EMS to grab ECGs in the field, since a number of features had changed on the ED tracing. In the ED there is only a mild 1st-degree AV block (214 ms), and no sign of Wenkebach phenomena. The computer read the QTc as 441 ms, "borderline prolonged," and it does appear to violate the "half the RR interval" rule for a normal QT.

https://www.kg-ekgpress.com/

Hospital course
Our patient was continued on cardiac monitoring during her hospital admission, but 24 hours of telemetry revealed only a single further dropped beat, as well as a resolution of her PRI prolongation. Tests for Lyme, as well as rarer infectious causes of AV block or myocarditis, were negative. She was due to start a Holter study upon discharge, and follow up with a pediatric cardiologist.

Turns out HIPPA does not apply to dogs, so I can show you a canine Holter monitor.

Discussion
1. Was this a seizure or what?
Although the second episode of syncope this young lady had began with an apparent seizure, it was unlikely to represent  a primary neurologic process. First of all, the seizure was described as a brief whole body stiffening, follow by loss of tone. While some forms of epilepsy may manifest in this manner, it would be unusual. Also, there was no post-ictal confusion described by the paramedic or parents, making a "true" seizure unlikely. Other historical and exam findings also weigh against a seizure. A good illustration of other factors to consider are found in this article "Seizure versus syncope (PDF download), which features a handy guide:

So this was likely a "convulsive" sub-type of syncope (brief myoclonus due to low blood flow to the brain), following the flowchart above.


2. Ok, it's syncope - what kind was it?
Reading the textbooks, there are dozens of terms to describe the various supposed kinds of seizures; micturation, situational, vasovagal, neurocardiogenic, autonomic, reflex... you get the idea. Really, there are only two types of syncope out there, from an EM/EMS point of view; cardiac or non-cardiac. Let's start with the second.


"Non-cardiac" does, of course involve the heart, but it's role comes in at the end of a sequence of events which are, as usual in medicine, not well worked out. The standard explanation sounds so weird, it may even be true.


For whatever reason, the body can pool blood excessively in the lower extremities. To compensate, both vascular tone and cardiac output increase substantially. Unfortunately, various pressure receptors in the heart now think the body is hypertensive, and act as if the patient has high intercerebral pressure; the heart rate drops, and cardiac output falls.

The fall in heart rate and blood pressure appear to be mediated by the vagal nerve, at least in part, so people call this vasovaagal syncope. On the other hand, since the sympathetic pathway is also affected, others call this autonomic syncope. Of course, as the picture above shows, the brain has a role in this, so it also called neurally-mediated syncope. If bardycardia, or even asystole (transient), are the main manifestations, it may be termed cardioinhibitory, but if hypotension is prominent, it can be instead called vasodepressor syncope. 


Whatever. The important point is to distinguish these benign types from cardiac causes. This kind of "syncope" is caused by a primary problem within the heart, and may not produce all the autonomic symptoms (sweating, warmth, nausea) that normal syncope does. It comes on fast, and can occur even if the patient is seated or supine. This table from a Nadas' Pediatric Cardiology reviews the differences with normal, or "neurally-mediated" syncope:

Remember how normal syncope has a bunch of vague terms to describe the same thing? In contrast, there are 3 clear types of cardiac syncope, with clearly defined etiologies.


First, there are structural causes (hypertrophic cardiomyopathy, tetrology of Fallot, pericardial tamponade, amongst others). 


Next, tachyarrythmias can induce syncope. Examples include long QT, Wolf-Parkinson-White, and Brugada. 

I calculated 411 msec. Normal

Last, but certainly not least, we have to worry about bradyarrythmias - sick sinus syndrome, overdose of beta-blockers, and, of course, advanced AV blocks!


3. Finally! So, did the AV block cause the syncope?
Well, that is the main question! 


Let's consider a few factors in answering. First, our patient is young and otherwise healthy - she is old enough to be fairly sure she doesn't have an undiagnosed heart defect, but young enough that she hasn't had a few infarcts take out her conduction system. Right off the bat, it seems unlikely she could have some intrinsic heart disease causing her syncope.


Next, her presentation fits with a vasovagal/neurally-mediated/autonomic syncope. Look back up at the table above - she had some lightheadedness before she syncoped, she felt cold, had no history of syncope during exercise, and it occurred while she was standing (although the second event happened while seated upright). She looked pale afterwards, and felt nauseous for a time afterwards. 


Lastly, we know that her ED ECG was almost  completely normal, and an ECG the next day showed resolution of the PR interval prolongation. For all these reasons, it seems unlikely that she had some cardiac disease - structural or dysrythmic - that produced the AV block. Most of those problems don't disappear overnight.


4. How often do you see AV blocks in syncope?
Not too often, but it happens. 


In a recent study (Electrocardiographic characteristics of atrioventricular block induced by tilt testing.), researchers noted looked at patients who had been put on a tilt-table to investigate the cause of their fainting. The idea is that through putting the patient through a fancier version of doing orthostatic vital signs, we can provoke and replicate a syncopal episode. Now, it's not a perfect test, with plenty of false-negatives and false-positive, but it is lot less hassle than the alternatives.

Looks more like a catapult. Just sayin'.

Out of 786 patients who they tested, 31 developed an AV block, about 4%. And not just some minor first-degree stuff, mind you. They summarize the results in this table:

The AV blocks didn't last long, less than a minute, on average. The majority of these patients had some form of second- or even third-degree block, as well as having significant periods of ventricular aystole! 


A few key results from this study suggest that our patient had a faint causing her AV block, rather than the other way around. 


First, they found that patients in their study had PP intervals that varied significantly. This is subtle, but can be appreciated on the EMS rhythms strip.

If you use the interval between the 2 non-conducted P waves as the standard (5th arrow), it's clear there is some PP variability on this strip.


Next, they found that these AV block developed gradually, usually after PR interval prolongation. We can see this variable on the EMS 12-lead:

5. What did the cardiologists diagnose her with?
They felt this was all likely due to "hypervagal tone," perhaps provoked by some mild hypovolemia. The results of the Holter study are pending.


Wrapping it up
EMS is in a privileged position, able to evaluate these patients within minutes after their episodes. You may be able to evaluate cardiac events that we might not ever see otherwise. This was a healthy young lady who had some pretty rare features of a common condition. It would have been easy to "BLS it in," and not get an ECG, or even a rhythm strip. Without a thoughtful medic in the back of the rig it may have missed entirely.

The IV placed by EMS: Too much, yet not enough.

If I see one more blog post about whether or not paramedics should be intubating...

Ok, that's not fair. RogueMedic, in particular, makes strong arguments about the role of ETT placement, and JEMS has a great website as a resource for discussing airway skills and research.

And airway cartoons!

But how often do we intubate? Unless your day job is as a CRNA, not often. Given all the ink and electrons that have been spilled writing about whether or not the ET tube "defines" ALS practice, it isn't a daily activity.

On a day-to-day level, however, placing an IV defines ALS care. Note that I did not say that it should be the defining act, just that from the perspective of your BLS partner, as well as the billing department, that is the point at which ALS care starts. It doesn't have to have fluids running, or even an immediate need for a medication push. But woe to those who bring in a stable chest pain or dyspnea without popping in a 20g!

Take your work seriously. Very, very, seriously.

A couple of recent papers discuss IV placement by EMS, and the results probably won't provoke the sort of teeth-gnashing discussion that EMS intubation research generates, but it is both far more relevant to your daily work, as well as pretty interesting.

The author of the first paper, Dr Seymour, is not an EMS doctor, or even in emergency medicine. He's a critical-care physician, but has been studying how EMS can improve care of sepsis in the field.

The paper, Intravenous Access During Out-of-Hospital Emergency Care of Noninjured Patients: A Population-Based Outcome Study, was conducted in Washington state. The researchers had an interesting idea - they wanted to see if there was a relationship between EMS placing an IV, and the patients' in-hospital mortality. They did this all retrospectively, looking over run reports and hospital records.

They looked at over 56,000 patients who were transported by King County EMS by ALS to the hospital. About half of the patients got an IV. The researchers then found two results:

• Patients who got an IV were more likely to die in the hospital.
• Patients who got an IV were less likely to die in the hospital, if you accounted for a bunch of other stuff.

The "bunch of other stuff" included the patient's "age, sex, out-of-hospital location, initial out-of-hospital vital signs, receiving hospital, year in cohort, mode of transport, and EMS interventions, diagnosis, and severity code."

Soooo, that's it. You can't really pull too much more out of this study. It has the advantage of having a huge number of subjects in the study, but being retrospective and superficial, you can't figure out why the simple act of placing an IV saves lives. EM Literature of Note concluded about this study that it "could end up telling us something, or nothing."

Now, while it can't really answer the question of why shoving a small piece of plastic under the skin (and occasionally into the vein) saves lives, it serves as a useful response to critics who would prefer that EMS place fewer lines.

A former partner, placing an IV like a boss!

For example, the authors of Unnecessary intravenous access in the emergency setting conclude that a "significant percentage of IVs initiated in the emergency setting are used inappropriately." One of their results that they base their conclusion on was that "only" 38% of EMS-placed IVs were used in the ED, while 46% of ED-placed IVs were used.

In another study (Out-of-hospital intravenous access: unnecessary procedures and excessive cost.), researchers determined that 56% of the IVs placed during their study were "over-treatment" per their criteria.

On the other hand, the new study helps bolster the argument that, sometimes, EMS needs to be more aggressive in establishing IV access, particularly in children. In the last study mentioned, while they found that adults were often "over-treated," kids were far more often to be "under-treated" than the adults - 33% vs 3% rates of under-treatment.

Likewise, the authors of Prehospital intravenous access in children found that while medics were pretty good at getting IVs in kids, they held off on doing so; in children < 6 years of age who the ED thought needed an IV, only half the medics had done so; this was true even if the child was in cardiac arrest or a trauma patient.

 So, sometimes an IV is too much, and sometimes it is just right.

And, speaking of too much...

How about when it's not enough?

A recent study compared the IO against the IV during EMS resuscitative efforts in cardiac arrest patients. I'm not going to belabor the obvious - placing an IO in the tibia is both easier

and faster

than getting an IV, no question. Science just proved it!

"Well, we have something that's faster and easier, sir."

Rogue Medic already discussed this in some depth, and I would not necessarily disagree with his analysis. Trenchant and witty! Nonetheless, establishing vascular access is the current practice in cardiac arrest, and using the IO likely accomplishes all the desired elements. I don't like that the equipment costs more - it costs too much to die already - but the simplicity and speed are appealing.

The Bottom Line

Placing an IV may save lives, but we don't know how. Sometimes unneeded IVs are placed, and sometimes medics don't put in a line when they ought to. And sometimes you should place something that's faster and easier.

You guessed it - more study is needed!

Just a little burning... and a ton of bricks.

To get back in the flow after the holidays, I got a small one. No big mysteries either - it's an MI, and there's some atrial fibrillation. But it's a nice example of ALS gettin 'er done. Thanks to Tyler, from AMR, for taking care of this lady.

So this 60-ish woman with an "extensive cardiac history" calls 911 for a little burning in her epigastrium, and a feeling of palpitations. Upon further questioning, she admits that she also feels like she has a "ton of bricks" on her chest.
First ECG:

Okay, got some work to do here. A bit of diltizem, get the aspirin on board, and a little nitro. This all brings down the load her chest to just a few bricks. Time for another ECG!

Hey, where did the  *** ACUTE MI SUSPECTED *** label go? To me, the ECGs look pretty similar, but some element in the interpretation algorithm - the depression in aVL? The STE in aVF? - changed enough such that the computer didn't want to make the call. Now, on a gut level, the ST segments in the inferior leads "feel" the same as before, but it sure looks like we have lost the strict "1mm of elevation" criteria that's been drilled into our heads.


We got an ECG in the ED which looked basically the same as the EMS tracing.

 Hmmm.

This is a tough one. It looks like we just fixed a potential cath lab patient, but you should be dubious. You can change a lot of ST depression with NTG and diltiazem, maybe even re-flip some inverted T-waves. But not too many things (at least until we get tenectaplase in the protocols) "fix" ST elevations.

So, the answer to confusion about new ECGs is usually an old ECG. Tough to do at 75 mph, but somewhat easier in room 5.

Our lady's ECG from the recent past:

About 4 months ago

Tyler and I compared the two, and agreed on our interpretation - activate the cath lab. After a brief conversation with one of the great cardiologists at Cardiac Specialists, our patient was whisked up to the cath lab.

Good thing too. Turns out she had thrombosed a stent that had been placed some years back in her RCA. Real bad luck for her, despite being on Coumadin for her a fib, as well as aspirin!

So what is the larger point here? 

If the paramedic had just looked at the rhythm strip and (correctly) treated the rapid a fib, the less-prominent ST elevations might not have been noticed in the field. Heck, it might have slipped past me in the ED! On the other hand, a second ECG may have shown growing elevations - you never know. That's why in some regions EMS is now obtaining up to 3 ECGs in the field as a matter of protocol, and they are catching more STEMIs as a result.

By the way, when I talked to the cardiologist during the cath report, he had one request - that we send the patient up with the prehospital ECGs!