Morphine for MI - Benefit and risk

Should we be using morphine to treat chest pain? It seems like an odd question - we're supposed to treat pain, right? - but there some interesting wrinkles to this issue when it comes to heart problems; i.e. acute coronary syndrome (ACS).

At the outset, let me emphasize that the Sponsor Hospital Council guidelines suggest that morphine be used, up to 0.1 mg/kg IV, if chest pain of suspected cardiac origin is not relieved with 3 tabs/sprays of nitroglycerin. That hasn't changed!

The benefit of morphine in ACS
Angina and infarction hurt, and morphine can treat that. There are other supposed benefits (reducing ischemia, reducing "stress," blood pressure reduction), but these are mostly theoretical, and often can be accomplished with other agents.

For example!  (Source)

Okay, perhaps not bourbon, but there are others.

The risk of morphine in ACS
Of course, morphine causes (infrequently) respiratory depression, hypotension, or depressed mental status. These are true for any patient, not just those with ACS, and are generally avoidable with careful administration.

The patient with ACS may face specific risks with morphine though, leading to worse outcomes with their ischemia. This is a controversial area, and there is little good evidence. Nonetheless, breathless headlines like this came out after publication of a study in 2005 (PDF link).

"News" link

This study did not prove a cause & effect relationship with morphine - the study design could only show an association - but people have proposed 3 ways that morphine could potentially harm cardiac patients; 1) direct harm to the myocardum, 2) the side effects of orphine harm the myocardium, and 3) the "masking" of ischemic pain, such that it delays definitive therapy, such as angiography/angioplasty.

1. Does morphine directly harm the myocardium in ACS?
Some people are concerned that morphine, during an episode of ACS, may directly harm the ischemic heart. The authors of the "morphine increases death risk" study remark on the discussion section that "in animal studies, morphine has been demonstrated quite conclusively to actually increase myocardial infarction size." 

They cite, as "quite conclusive," a study from 1982, where rats where given a subcutaneous dose of morphine at 3 mg/kg. (That would be 210 mg for an adult human!) They then performed open-heart surgery, and tied-off a coronary artery - just blocked it off. When they looked at the hearts 2 days later, the rats who had morphine had MIs that involved 10% more myocardium than the morphine-free rats.

Rat open-heart surgery, w/ ligation of LAD. source

On the other hand, North Carolina researchers studied rats who underwent a limited period of coronary occlusion, and found that the morphine pre-treated rats had smaller infarct extent.

So it's an open question whether any of these animal studies represent great (or even mediocre) evidence that can guide our clinical management. It's not conclusive, that much is clear.

2. Do the adverse effects of morphine (hypotension, hypoxia) cause harm in ACS?
Perhaps morphine doesn't have a direct toxic effect on the myocardium, but the known adverse effects can certainly cause problems. Generally, though, these are not common if morphine is administered cautiously, and in practice hypotension or hypoxia are rare.

One study that has been cited as demonstrating the potential for hypotension in MI patients was conducted 1969. In The effect of morphine on blood pressure and cardiac output in patients with AMI the authors gave 15 mg morphine, intramuscularly, to 10 patients with an MI. They found a "slight tendency to development of orthostatic hypotension."

So if you give 15 mg IM of morphine, you might want to be ready to give them some fluids, or just not have them stand, but this doesn't seem persuasive as a severe adverse effect.

 3. Does morphine "mask" the ischemic pain, rather than treat it?
 This is really the most interesting question - should we avoid using opiod medications that could "mask" recurrent or continuing ischemic pain, so that we can make better decisions about further interventions?  I.e., does morphine delay angiography?

Spoiler alert: We have no idea. There is no evidence here, only opinion. 



Chest pain today = abdominal pain 50 years ago?
An analogy with abdominal pain is appealing. In the past physicians relied on the severity and evolution of abdominal pain to aid them in the decision to pursue surgery. If a patient's pain was "masked," the physicians had few means to understand the evolution of the disease. Nowadays, given the accuracy of blood tests and CT scans, most clinicians feel they may safely treat the abdominal pain prior to full evaluation of the patient.

A number of people would argue that we are still at the "pre-CT scan" point now for ACS. The argument is that, absent angiography, we have few means to establish the need for invasive treatment. Like the surgeons of ye olden days, we may be led to a "false sense of security" after morphine reduces the pain.

This is an important question since, in the UA/ACS patient, we use "recurrent ischemic pain" as one factor determining how urgently the patient is taken to angiography. Dr. Stephan Smith, of Dr. Smiths ECG Blog points out one case, for example, where he believes that the heavy-handed use of opiods contributed to a delay in angiography. 
  
But how much do we depend on the character or presence of recurrent/refractory pain, as opposed to other factors? We have the ability to check troponins, perform sequential (or even continuous) ECGs, and we can look "directly" at the myocardium with a bedside echo. We are not dependent on the unimpaired report of symptoms - technology has provided a few tools!

Two patients
Let's consider two extremes, two different patient presentations.

First, how should we consider a patient who adamantly denies any chest pain/pressure (or back pain, shoulder aches, or jaw discomfort...), but has an ECG that looks like this:
 

True story. Yeah, it was an RCA.

This patient was in serious denial, but after intensive interviewing (the Geneva conventions were observed!) he admitted to "chest congestion" that improved with NTG, and consented to cath.

Second, how should we proceed with a patient who describes "crushing" chest pain, with left arm radiation, an sensation of doom, etc., and looks like this...

... but whose ECG looks like this?

Stop looking, it's normal.

Yes, pain means something, but so do the troponins, the echo, and of course the ECG. 

In the end, it's worth pointing out that Dr Smith's "missed MI"  had an initial ECG that actually showed a STEMI, albeit an uncommon (but not rare) pattern. 


Bottom line: Still a recommended therapy!
This post contains more opinion than I usually like to write. There just isn't the level of evidence to help us. I think my observations are pretty mainstream, however. Talking with my colleagues, most of them tend to agree with the use of morphine in this situation. One emergency physician remarked "That's what morphine is for - to mask pain."

Despite the hypothesis-generating evidence provide by the CRUSADE trial, there is little other evidence that suggests that morphine is harmful, or that analgesia must be deferred while patients are in pain. Despite the inflammatory press coverage of the CRUSADE results, the AHA continues to recommend the use of morphine in ACS.


But hold the bourbon!


 

Gender, EMS, and STEMI - new study

Does our care of the patient change when the patient is a woman?

If you're like most EMS providers, your response is a strong "no." People who go into EMS generally share strong ideals, and are motivated to provide the best care possible to everyone who needs it. 

With that in mind, though, a recent study found results that are difficult to explain, and ought to prompt us to reflect on our practices. It's not proof, but it deserves discussion.

Is on-scene time longer for women with chest pain?
Time is muscle - you know the drill. With chest pain, and especially in a STEMI, we have the ability to save lives. This is what EMS is for; we identify the sick person, and we get them to the right place, fast.

Too fast! (jk - no injuries.)

With all the emphasis on reducing the door-to-ballon time, we have to identify any reasons that would delay definitive care. What if being female is one of those reasons? 

The authors of Gender differences in scene time, transport time, and total scene to hospital arrival time determined by the use of a prehospital electrocardiogram in patients with complaint of chest pain. (PDF link) looked at the EMS service in San Diego, and analyzed run-forms of chest-pain patients. They looked at two time periods, before & after EMS started obtaining prehospital ECGs. They then looked at the various time intervals, as well as the computer interpretations of STEMI (a required element in their system for field-activation of the cath lab).

They found over 21,000 patients who had been transported for chest pain. About half of the patients were men, but the women were, on average, significantly older (65 vs 59 years of age). Only 3% of the patients (in the later time period) had a STEMI, with most of those being men.

When they looked at the scene times and transport times for chest pain patients, nothing changed between the two time periods overall. In the second period, however, they found that patients with a "STEMI" interpretation on the ECG had shorter scene and transport times.

It gets more interesting, however, when they broke things down by gender. It seems that that women had longer scene times than men, both for those with and without a STEMI. Specifically, women with a STEMI, on average, had scene times about 3 minutes longer than men, while women with chest pain (but no STEMI) had scene times about a 1.5 minutes longer than men. Transport times were the same, roughly, for men and women.

Interpretation
This is a small, but provocative result, and it isn't clear what it demonstrates, let alone proves. 

Did paramedics feel less "urgency" with the female chest-pain patients? This seems unlikely, since the actual transport times were similar. Everyone got driven to the ED at the same speed.

The study design can't answer what accounted for the difference in scene time interval, unfortunately. The generally older age of the female patients could suggest that evaluation was more complex, and accordingly required more time. It's already well-known from other studies that women with ACS generally have more comorbid conditions (such as hypertension and diabetes) than men.And it's also hard to interview older patients quickly; some things are hard to rush.

Lastly, the difference is small in absolute terms. Although the authors suggest that the difference in scene time for STEMI could result in a 0.25% - 1.6% increased mortality, this is based on a questionable extrapolation.

How does this fit with prior studies?
It is still possible that there is bias in the care of female patients that was not captured in the data here. A prior EMS study showed that 7.5% fewer women than men got ECGs for chest pain (Is there gender bias in the prehospital management of patients with acute chest pain?). We don't know if this was the case in the current study, since the authors "assumed that all patients," men or women, got ECGs, since "it was the protocol." Some basic QA about ECG completion rates, by gender, would have strengthened the study.

Another study examined the differences in prehospital intervals  among men & women who ended up being diagnosed with an MI. The authors of Myocardial Infarction: Sex Differences in Symptoms Reported to Emergency Dispatch also found that women had longer on-scene times, by about 1 minute. When they took age into account, whoever, that difference disappeared. Hopefully the next San Diego EMS study will obtain the data to make these sorts of adjustments.

The Bottom Line
Like all studies that are able to "dredge" through a large data-base of run-forms, we end up with more questions than answers. The large number of patients enables researchers to find some statistical results, but the interpretation gets muddy.

Moving forward, the key will probably be in QA; making sure that all the appropriate patients get ECGs, that transport is expedited in STEMI, and that feedback is obtained from the ED and cardiology. I'm hoping that future, and better, studies will demonstrate that EMS is taking acre of everyone to the same high standards. 

In order to protect the c-spine, should we stop helping?

There have been some interesting recent evolutions in long-held beliefs about managing the possible cervical-spine fracture, however, and a recent study adds an interesting development.
The Study
The new study, Cervical Spine Motion during Extrication, makes an interesting contribution to the research, uses intriguing methods, and is very relevant to EMS. 

It is usual practice for EMS to go through great efforts to maintain cervical spine immobilization after an MVC. The patient will often have both a cervical collar and a "short board," or KED, applied. The EMS crew will then go through elaborate efforts to move the patient onto a long board right from the car seat, avoiding any active participation by the patient.

The ideal

The somewhat messy reality

This is a lot of effort, but the hope is that, by avoiding motion of the cervical spine, there will be no further, or "secondary," trauma to the spine. The actual risk of secondary injury is controversial, but immobilization is the current usual practice. 

Well, a natural question to ask is: Does it work? Do these multiple devices and great physical efforts avoid motion of the cervical spine? 

Well, this question has been approached in a number of ways in the past, but Missouri researchers took a new approach. They used a simulated extrication scenario, and tracked motion of the cervical spine using video motion capture, the technique used to, among there things, render Tom Hanks into a cartoon character.

Looks kind of creepy to me, but Hanks can get away with anything.

Note the little dots on Tom's head, hands, and shoulders. By tracking the movement of these points, a computer can compare degrees of flexion, etc., of the neck.

For the simulated extrication scenarion, they used a mock-up of a Corolla that had been in a bad head-on MVC. They recreated the postions of all the posts with PVC, replaced the surfaces with chicken wire, and replaced the seatback with Plexiglass. This way, they could record the extrication from all angles.

The participants were all paramedics with more than 5 years of experience - some of them also played the victim, along with some non-EMS people.

The victims exited the vehicle in 4 different ways:

1. They were instructed to get out on their own, and walk over to the backboard and lie down.
2. They first had a cervical collar placed, but still had to get out on their own.
3. Collar was placed, and then the EMTs proceeded to maneuver the patient onto the board while "holding c-spine," instructing the victim to not offer assistance.
4. A collar was placed, as well as a KED, before being moved onto the board.

(For what it's worth, method #4 is what I was taught by the good folks at SOLO in Conway, New Hampshire, when I first got my wilderness EMT. It took me awhile before I felt comfortable with method #3, that everyone in the "real world" was using!)

So how much neck motion did they see? A lot! 
 

Looking at the graphs, you can see how much motion, in each plane, they observed with each extrication method. 

Now here's the weird thing: method #2 (c-collar and "get out on your own") showed significantly less cervical motion than both methods #1 and #4! 

For example, take a look at "Graph C" from just above:

In other words, "Here's a collar, now please step out of the vehicle" produced less neck motion than "Don't move! We'll do all the moving for you."

Interpretation
The results are seemingly paradoxical, but it squares with some of my experience. Placing the KED and moving the patient to the backboard always seemed to go smoothly if the patient was slender, short, and driving a large car. 

Smaller car, bigger person - it doesn't go so smoothly...

E.g. Try putting a KED on Klump.

Also, it's worth highlighting the change in protocols that a few EMS agencies have already rolled out, actually prior to the publication of this study.

For example, in the last year,  Xenia FD in Ohio, as well as New Haven, CT, have started using method #2 for certain lower-risk trauma. A number of other agencies are said to be following suit. 

And although it's not a prehospital policy, Bridgeport Hospital is now encouraging nurses and EMTs to remove backboards as soon as patients arrive in the ED - no doctor involved! This shift in practice was prompted both by the risk to patients that backboards pose, and the scant evidence that supports their use once the patient has arrived at the hospital.

And you get your board back right away!

This study suggests that those policies may not be as risky as previously imagined, and may, in fact, be more conservative than the "usual practice!"

The Bottom Line
This study adds to the growing body of research that suggest that our approach to spinal care may be quite different in the near future!

(In the meantime, of course, please stick to your local protocols and guidelines.)

Bradycardia - Traumatic etiology?

When the patient has bradycardia, you run through a short differential. Electrolytes, MI, drugs,.... trauma? How does that work?

The patient
A trauma alert was rolled into room 5 at the 'Port. A 45 year-old male, restrained driver in a roll-over, who had been ambulatory on EMS arrival. He was mildly intoxicated, denied any significant medical problems. His vitals were normal both prehospital and in the ED. (The trauma alert was due to mechanism, not his condition.) 

His exam was notable for a rather large laceration to his scalp, with a correspondingly large blood clot at the head of the backboard.

"He probably smells my dog!"

The bleeding was controlled, and the evaluation continued. 

While we were getting the chest x-ray, however, we noted that his heart rate, which had intially been around 70-80, started slowing down. 60, 50, 40, down to the 30s! 

His blood pressure, which had also been normal, plunged down to a systolic BP of around 50 mmHg. His ECG showed this:

An old ECG was entirely normal. Interestingly, the patient, in a supine position the whole time, denied any symptoms whatsoever, and was moderately amused by our concern.

Discussion
So what happened to his heart? 

Well, we should worry first about ischemia- or infarction-related bradycardia. An inferior MI is notorious for causing 1° and 2° AV blocks, most of which resolve on their own. These bradycardias manifest with a narrow QRS, since the block involves the AV node, but not the bundle of His, etc. They usually get better in a few days, on their own.

By contrast, anterior MIs may involve an infarct of part of the ventricular conduction pathways. The patient will have a wide QRS and a high-grade 2° or 3° AV block. These are bad, and need permanent pacemakers!

Although you should think of hyperkalemia when you see bradycardia, there was little else to suggest it. He denied any medications, so digoxin, beta-blockers, and calcium-channel blockers seemed unlikely, especially given how quickly the rhythm had developed after normal prehospital vital signs.

Perhsps I should be clearer about how the bleeding from the scalp was controlled. Since the wound was large, and the bleeding brisk, a number of staples were rapidly placed.

NOT an approved wound closure technique. (credit)

Let's take a closer look at the ECG. Like I said before., he gradually slowed down to 34. 

Sinus activity is almost extinguished - I can only find 2 P-waves, preceding beats #1 and #3. The QRS is narrow, suggesting a junctional rhythm, albeit much slower than you would expect (usual junctional rate is 40-60). You can't really call it complete heart block, since there is so little atrial activity; instead, it's just called AV dissociation.

Putting all this together, it appears he had a cardioinhibatory/reflex syncopal episode; i.e. he fainted. The placement of the staples likely triggered a strong vagal reflex, which inhibited both the sinus node (almost no P-waves) and the AV node (junctional bradycardia). 

Fortunately he was already supine when it occurred!

Inappropriate stapling technique.

The treatment
He got atropine 0.5 mg IV, and a liter of NS on a pressure bag. His heart rate corrected quickly, coming up to about 80, but his blood pressure took a few more minutes to come up! He was admitted, and did not have any more bradycardic episodes. 

Bottom line
 No harm, no foul, as they say. But I do think we're going to be more enthusiastic about using lidocaine in the trauma bay, however!

How do patients view our care?

Back when I worked on a rural private ambulance, a pair of EMT-B coworkers tipped a stretcher over in the hospital parking lot - with the patient on it. The patient was not seriously harmed, fortunately, aside from a huge bruise on the side of her face. 

"Right here. Not here or here so much. Right here. .."

Now, the two EMTs were petrified about the near-certain complaint, firing, and lawsuit, but the patient just brushed it off - it turns out she had a real soft spot for those "ambulance girls."

On the other hand, I also have seen patients complain about some "trivial" matter, right after receiving high-level ALS care, true life-saving stuff. 

What happens after these complaints? The EMS crew becomes discouraged, the supervisor gets an earful, and the patient tells all their friends about their "terrible care." These sorts of complaints can be very frustrating, seemingly coming out of nowhere. 

Researchers in the UK decided to tackle this issue with a novel approach, aiming to explain the disconnect between the medics intentions and the patients perceptions.

                                                              *** Spoiler alert***
  Your patients may be judging the quality of their care based on non-technical aspects.

The study

PubMed

In Patients' and ambulance service clinicians' experiences of prehospital care for acute myocardial infarction and stroke: a qualitative study, the authors carried out interviews with both patients and with "ambulance service staff." 

Researchers in this UK study chose to study patients who had been transported by the city EMS agency for a suspected MI or CVA. They also selected a groups of the "clinicians" (likely on the level of U.S. paramedics) who worked for the ambulance. In separate interviews, they asked these patients and providers various questions. (Unfortunately, they weren't able to link up the exact medic-patient pairs. That would have been difficult to arrange, I can imagine.)

So, after sifting through hours of recordings of interviews, the researchers found some interesting common ideas.

Communication
The patients placed a good deal of emphasis on how well (or badly) the medic was able to establish rapport, and explain their medical condition. The medics, for the most part, explained that they usually tried to provide reassurance and contact . Of course, a few medics explained that it wasn't their role to diagnose or explain the patient's condition to them, as "it’s up to doctors to make the ultimate diagnosis."

Pain management
This really was the only area in which the patients commented on the medical care itself. They may not know much about adenosine versus amiodarone, but they sure understand if their pain isn't acknowledged, much less treated! Of course, it proved to be a frustrating topic for the medics, as pain management often ends up near the bottom of the list of priorities during transport.

Speed
Again, patients may not know if the medic knows how to treat an MI, but they attach a good deal of importance to how quickly the ambulance arrives! Prehospital folks know this, and understand that a long response time "means that you are already on the back foot." More frustration - many patients viewed delays at ED triage as part of the EMS experience, if not precisely their fault.

Are these results expected? 
Yes & no. No one has done an academic study like this before, but that doesn't mean that people haven't been thinking about it. 

Allen Johnson, a paramedic-turned-hospital-CEO, wrote an article for JEMS back in 2010 about patient satisfaction in EMS. In The Customer's Always Right: Steps you can take to ensure customer satisfaction, he described most of the same issues that were raised in the new study. The article is not short, but I really like his summary:

Patients usually aren't qualified to evaluate the proficiency of our life-saving skills, but they are more than qualified to fully assess our interpersonal skills. And frankly, life-saving skills may be used on less than 1% of patients; our interpersonal skills are assessed 100% of the time.

A colleague once summarized the EMS patient's expectations to four simple points: 1) Get there quickly; 2) Be nice to me; 3) Tell me what you're doing, and 4) Take away my pain.

Leave it to a paramedic to boil down the research to this level!

So why do we care?
Because your bosses care. And their bosses. And regulatory agencies, Medicare, insurers,  the New York Times, as well as Yelp!

That is unlikely to be a surprise to anyone who has worked in healthcare for more than one day. In fact, large segments of the medical industry are devoted to measuring and reporting patient satisfaction, without any regard for the technical quality of the care. Everybody is being judged for patient satisfaction in medicine these day - including helicopter-EMS!

Critical (but also satisfied!) patients: Baptist LifeFlight

Money is also on the line now. Not only do emergency departments worry about patients choosing a different hospital, but they may also be payed less by Medicare if they have poor patient-satisfaction scores.

This attitude is now as dated as that nurse's cap.

Can EMS be far behind in this trend?

The bottom line
Be personable with your patient, communicate with them, and address their pain. 

Of course, you need to get that ECG, treat the hypoxia, and recheck the vitals signs. But those are, increasingly, not the only things that we are judged on these days!