Near Miss

PG is 89 and lives alone; she comes in by ambulance with mild, new confusion that her son noticed when he went to say hello to her as he did every morning. When asked, she did complain of some severe chest pain that had occurred earlier, but there was no syncope, and right now she feels OK. She has no fever or headache, and her pre-hospital course was uneventful. Hypothyroidism was her only stated history, keeping with the maxim that if you show up past the age of 80 with almost no medical problems to the ED, something major must be wrong.

Her exam is pretty unremarkable; T 97.9 F, HR 67, BP 187/85, RR 18 (meaning no one checked, but seemed about right), SpO2 88% on RA. No real pertinent findings; lungs were clear, pulses good, no fever. She has a hard time telling me what day it is, but otherwise is good and conversant. Her pulse ox reverts to normal with only 2 L nasal cannula.

My standard old, sick work-up commences, and she comes back with a troponin of 0.11, which is suggestive of something going on, and with an LBBB on EKG, I suspect NSTEMI; confusion or altered mental status in the elderly is a very common presentation of acute coronary syndrome, and the severe chest pain earlier in the day makes it look like that as well. Lactate normal, not septic, doesn't need an LP, admit, done.

Just before she is taken upstairs, she tries to get up to go to the commode; as soon as she rises and pivots, she complains of central chest pain and relatively severe shortness of breath with discomfort. I attribute this to cardiac causes; she doesn't syncopize or have any other change in vital signs.

She wanders through her hospital course for a few days, and three days later, after a relatively normal echo with only a mildly dilated RV and no change in vital signs, someone gets a D-Dimer. Of course, it's up; and her CT shows a freaking saddle embolus. Not just a l'il baby PE, but a giant monster that extends into both upper lobes. Unfortunately, this is another often overlooked cause of confusion in the elderly; in retrospect, the only tip-off was the pulse ox. Although not insignificant as a vital sign, it's kind of non-specific; my error, such as it was, was to ignore the low pulse ox with no lung findings. If it's cardiac, or COPD undiagnosed, then there probably should've been wheezes. Yet another near miss. I looked at my chart; not bad, as I would suspect on an admitted pt with a relatively straight-forward course. As always, with these, I was so close and yet so far. The near-misses in my career, unfortunately, are never the ones that I suspect; never see coming. It does make me feel a bit better that she walked around on the floor for four days with her saddle PE and did just fine.


Triage Desk

Our urban ED has been undergoing a remodel. This shop is the real deal--busiest in the state, blocks away from war zones, full to the brim on a daily basis with a stew of people gravely injured or sick, shoulder to shoulder with those that have colds or need a refill but don't have a doctor. As with any similar population, even though a majority of patients are well intentioned, violence breaks out on a relatively regular basis.

We have these flashing 'blue lights' in the ceiling--leading to the inevitable comparison to K-Mart--that go off if the front of the hospital isn't safe, usually due to a ghetto ambulance, someone's car dropping off a shooting or stabbing. Prior to the remodel, we had moved towards a plexiglass security corral with security guards in it as a greeting when you walk through the door.

That's all gone. With the remodel, there is a wide, new, fancy triage desk, with warm, wood grain paneling, recessed lighting, and the feel of a spa, consistent with our group's suburban outpatient centers.

The triage nurse is supposed to sit behind the desk and greet new patients, and is exposed to everyone still in the waiting room. There is a security desk, but it's over on the far side of the lobby, unobtrusive, and a great distance from the nurse triage desk.

Leave aside for a second that they aren't doing a thing to the arena, where we actually take care of patients, siding is falling off the cubicles, there are a bunch of curtained stalls where they always place the combative psych patients only three feet from the RN/doc area, and the overhead lights only work sometimes, and ask--was that really a good idea? I mean, taking the plexiglass OFF the front of the ED. This, in a place where security guards have quit mid-shift and left. I asked some of the nurses, and the general comment was 'thank God I don't work triage anymore'.

A quick Google search on healthcare workplace violence provides some interesting backstory. A post on KevinMD from 2 years ago illustrates the difference in a primary care setting and an ED setting. The most interesting read is the comments section where things quickly get out of hand. JAMA has a commentary noting that simple assaults are four times more likely to occur in healthcare, with ED nurses at high risk. A fact sheet from the International Council of Nurses notes that nurses are more likely to be assaulted than police officers and prison guards.

More likely to be assaulted than police officers and prison guards? The numbers bear out the sense I have, and that most people have that I talked to, that there is an undercurrent of danger in the healthcare workplace. Now, if you read the comments on KevinMD's post, it seems as though some (probably a minority, and not all the posters) feel that our callous disregard for patients somehow balances that out, or that nurses are somehow abusing patients right back. It's hard not to take offence at this; mainly because almost all of us that work in urban EDs honestly do try and do our best. We get socks, and towels, and blankets, and sandwiches, often without complaint, and often in response to shockingly rude requests (as in 'go get me a sandwich now' as I walk by a room where a very not sick patient is standing outside glaring at staff). We deal with drunk people who routinely slap, kick, and punch people, and high people, and crazy people, all of whom are unpredictable. I've been cussed at, yelled at, spit at, my life threatened by a convicted felon for not giving antibiotics (FYI, if someone says 'do you KNOW who I am?' it's probably not a good idea to actually check the state registry of convictions). And that's last year alone. Somehow I find it hard to believe that I and the staff around me are so callous, cruel, and unfeeling that I deserve this treatment.

I'm not even the vulnerable one. I talked to a number of nurses about that triage desk, putting a shiny, trendy face out to the parking lot in one of the worst neighborhoods in the city. Now, statistics say that she (or he) is no more likely to be the victim of gun violence than any other healthcare worker--which I suppose, statistically, I can take at face value. But what about the torrent of abuse otherwise? What about the verbal abuse, the pushes, and all the rest? Because I'm pretty darn sure almost none of that gets reported. Imagine, if you are an ED nurse, filling out an incident report for every incidence of violence or threat that you are subject to on a regular shift. I'm not sure what the solution is, but I'm pretty sure a new waiting room isn't it; sounds like lipstick on a pig to me.


Controlling Costs

Clinical Vignette: 62 yo F, smoker, presents to ED with sudden onset severe, stabbing lateral neck pain for four hours, non-radiating, with no trauma. Pain does not radiate (i.e., does not move from place to place). She has diaphoresis (spontaneous sweating). She has had muscle spasm before, but never like this. No recent yoga, chiropractic adjustment. No headache. No vision loss or weakness. Vital signs: temp 97.8 F, pulse 102, respiratory rate 20, blood pressure 194/96, pulse ox 91% on room air. Exam, normal except for diaphoreses, and tenderness over the insertion of the paraspinous muscles on the L. There is no ptosis, no dryness of half the face, and eye movements are normal; all pulses are equivalent.

What did you think of for a differential diagnosis?

Most likely, you thought of muscle spasm; if you've been reading recent case reports, the pertinent negatives of yoga and chiropractic adjustment raised the spectre of vertebral artery dissection, and the clue of a smoking history and hypertension with no history of this in the past might have made you a bit uneasy. On exam, diaphoresis may have made you uneasy, but there is no neurologic deficit to lean on, and the tenderness of the muscles of the neck can be used to support spasm.

So, now the rubber meets the road: what do you do? If this case is being presented to you as a case on oral boards, there is an answer, and the heart rate, blood pressure, pulse ox, and respiratory rate must all be addressed and treated, and you can bet that there is a diagnosis waiting to be found. If this is a teaching conference, then most likely it is in the conference because there is a great case to be found as well.

In those cases, 'what do you do' would probably be met with the knee-jerk treatment for a potentially sick patient in the ED--two large bore IVs (or at least one), cardiac monitor, EKG, oxygen for the pulse ox, a full spread of labs, and symptomatic treatment to start, followed by imaging or a procedure to find the answer; either advanced neuroimaging, or, given the neck stiffness, a lumbar puncture for hemorrhage or infection I would guess. However, 'what do you do' in a community ED setting is quite different. Usually, I would venture a guess that this patient would get a dose of strong narcotic pain medication, probably with an IV because it's a richer community hospital and she doesn't have a long history of visits; she's also polite, and white, and not asking for drugs by name. An EKG probably is done, and some labs. Often there's some spice to the narcotic based on experience and training--some valium, or some oral cyclobenzaprine (a muscle relaxant and sedative), or oral tramadol (a useless hypnotic pain medication), or droperidol (another sedative), or some other such medication.

There's a gap in these two approaches, isn't there? And yet, in the vast majority of cases, the community approach will work. You perform a focused exam and chase definite findings, then chart like crazy that you considered heart attacks, PEs, dissections, and such, and send the patient home with some hydrocodone to help with your Press-Ganey scores. Most of the time--probably 99 out of a hundred or more, in this case--you will get away with this, meaning there will not be a bad outcome at home. Most of the time, this will also represent good care, in that the patient will feel better and not be harmed in any significant way, either by a miss, or by a harm inadvertently committed by the medical community.

But, once in a blue moon, this will be the zebra. We call rare disorders 'zebras' based on a saying often heard in medical training--when you hear hoofbeats, think horses, not zebras. That is, common things are common and most of the time that's what the patient has. Something common. However, more and more of our conferences and training ask us to find the uncommon, especially in Emergency Medicine. The zebra here is a vascular dissection; with neck pain in the back, a vertebral artery dissection, or a tear in the wall of the blood vessel on the left that travels up both sides of the neck in the back. The downside of missing this diagnosis would be stroke or severe neurologic impairment. It is, however, very rare; so rare that it's hard to tell how often it happens. There is an association with recent chiropractic adjustment, minor neck trauma, and even yoga, but it's hard to tell if this is a true association or rather the result of a hard search for causes of something terrible, since these often happen in young people. Smoking is also associated.

I did look for the zebra in this patient. She got a CT of the chest with neck after a discussion with the radiologist. The case will not appear in any presentations or journals; the study was normal. She felt better, and went home, probably with cervical muscle spasm; I did actually give her a bit of hydrocodone/APAP.

I wrestled with the decision for a while (meaning, since I'm an ER doc, about thirty seconds of hard thought). I'm still new-ish, and coming out of residency we often say that it's good to be overly cautious, and if you think of doing a scan, the right thing to do is to get that scan, so you avoid misses. But it also had to do with our doctor-patient relationship; she was very uncomfortable, sweaty, and hypertensive compared to normal. Yes, I'm jaded, but 194/96 and sweaty is a bad combination occasionally. And this is why costs in our system will be hard to contain. I wasn't overly worried about being sued; our state is not bad as far as climates go for malpractice, and I was certain I could chart well in this case. I would say I considered vertebral artery dissection, but the patient had no neurologic signs or symptoms and responded well to medication. But all of those worrisome little signs added up to a CT scan that cost a ton of money and carried with it the risk of radiation and kidney injury, both rare in their own right, but probably about as prevalent as the bad outcome I was trying to avoid. The problem is, you and your patient will never know which rare bad outcome you might get. It's easy to make a decision if you say to a patient the following: 'you could go home now, but I know for certain that you happen to be in the 0.01% of people I will see in my career with neck pain who have a vertebral artery dissection, and in 18 hours, you will suddenly collapse, lose your vision, and be unable to walk, resulting in a long nursing home stay and a lifetime of disability--would you like a CT scan to catch that for you?' I can't say that. Never will be able to. And until I figure out how to deal with that risk, I'll continue to order CT scans based on gestalt, and probably order a lot of 'useless' scans, with a few scans that catch obscure diagnoses and thus reinforce the cycle. It's a lot more complicated than it seems when we say 'stop ordering useless CT scans'. You try it.



With 42 in the waiting room, I'm almost relieved to hear an ambulance call that could actually bring someone who needs me. JZ is a 72 year-old male brought in by ambulance with a chief complaint of syncope--while raking leaves, no less--and a heart-rate of 220. Somehow, the nurse taking the call converts this to 'SVT', or hears 'SVT' somewhere, and thus has no concern when giving me the report at all. I have doubts, for many reasons, most of which I arrive at after playing a game of 'which of these is not like the other': old man, fainting, and SVT. SVT is for young, healthy women who had too much caffeine. We get ready in the normal way for a sick patient; suction, monitors, waiting in the room, gloves on, airway cart and code cart at the ready. It's a preparatory routine to ward off the certain death that comes from complacency; if I had decided to finish my charts first, he would, of course, be coding on arrival.

I am right to doubt the report; the paramedic report is V-tach as they are moving him over to the gurney, not SVT. They had a rate of 220 on the monitor, and tried adenosine to no avail, seeing a couple of 'runs of SVT' which I can't really comment on as I didn't see them. The 2005 ACLS guidelines, though, do allow for adenosine in a wide-complex rhythm, so it's appropriate, it just doesn't get me off the hook. At any rate, he does have the life-threatening arrhythmia one would expect in a syncopal 72 year-old man with a heart rate over 200. The ugly wide kind. Regular and sawtooth.

In boards fashion, what you as the reader 'see and hear' from the bedside is not re-assuring; his eyes are closed, but he opens them to voice and does respond in a faint voice to name. He is pale, diaphoretic--ashen, really--with only a palpable carotid pulse, and just a lot of reinflation and beeping from the automated BP cuff. In vain I check for a ICD--an implanted device that would self-shock the patient if he went into this rhythm. Nothing. He is 'unprotected', so to speak. Despite this he is not writhing or moaning; he is in fact eerily quiet and still, as if concentrating on existing. I have to continually ask him if he's still with us. This is slightly more worrisome than the complaining I've been listening to all night long.

ACLS and resuscitation in general are pretty clear on this one--he needs electricity. Specifically, he needs about 200 J of electricity from some nice pads on his chest. But, he's awake, and responds, knows his surroundings. He has no pain, only trouble breathing. I'm about to change that, or at least the pain part. To sedate, or not?

These days I think it's approaching standard of care to provide sedation for this poor gentleman. He's just minding his own business, it would be poor form to shock him without some sort of medicine to at least help him forget the experience. So, I ask for the absolute best drug in this case--etomidate.

Etomidate is trustworthy, used for RSI all the time, has a rapid onset, produces amnesia and sedation, but is relatively easy on blood pressure. It's main side-effect, myoclonus (jerking of the extremities) really doesn't matter in this case. My other alternatives are poor. Propofol has a strong effect on blood pressure, especially in the elderly, and lower than palpable carotid is way too close to zero. Ketamine, alternatively, is a sympathomimetic, and will accelerate his blood pressure but will also increase the drive to his heart, which is already in overdrive, and will probably constrict the blood vessels to his heart, probably not a good choice in a man potentially having an MI, or at the very least a fair amount of strain with known coronary artery disease. Mixing the two is possible and currently in vogue (via the 'Ketafol' prepartion) but any use of ketamine could be dangerous. Versed and Fentanyl are slow and have blood pressure effects in combination similar to propofol.

I turn to our pharmacist and ask for 10 mg of etomidate, only to get the answer 'we don't have any'. 'Not any, or just not a lot?' 'None in the whole hospital'. What I said next under my breath is not quite reproducible. And so, the drug shortages progress from annoying in the case of no compazine for the routine migraine to downright dangerous; I now have no comparatively safe way to sedate this poor fellow.

Someone finally auscultates a pressure of about 70/40, but that's not making me feel better; fluids are started, and I'm about to go ahead with the lightning bolt/horse kick to the chest when he magically converts to a slow atrial fibrillation and improves his pressure to 100 systolic.

Bullet dodged. What could have been done otherwise? As I haven't used it much in the last year, I forgot about push-dose pressors; a small, peripheral IV push of phenylephrine is comparatively safe and will buy me enough space to use propofol in this patient. But seriously, what are we to do about these shortages? There is a short list of drugs for which there really is not a good substitute. Could I use an alternate? Sure. But etomidate is a drug I've used hundreds of times and had the exact profile I wanted. Why compromise in this case?