07 September 2009

Bedside Ultrasound in the ED

An interesting take on ultrasound in community medicine from the EM Blog:

Community Hospital: Ultrasound Free Zones
Why is it that every EM resident training in the USA learns Bedside Ultrasonography (BSU) during residency, and then upon graduation goes out into the cold ultrasound-less world of the community hospital ED? Is rapid ultrasound so available in these hospitals that their skills in BSU are really not necessary? Hint; NO, THAT IS NOT IT! Well if that isn’t the explanation, then what is?
Another blogger (Jan Shoenberger) on this site noted the disconnect between academic environments and the “real world” noting that many new grads are unprepared for the realities of practice. One example of this is that they know how to do BSU but will likely never have a chance to use that skill.
The posited reasons why ultrasound hasn't penetrated community hospitals are part right: ED Directors who don't keep up, radiology turf battles, etc.  But I have to say that the reality of bedside ultrasound has never quite lived up to the hype and its purported potential.

For reference: We have had ultrasound capability in our ED for about seven years, I think.  We work in a busy (100,000 visit), high-acuity urban/suburban center without too much trauma.  About two-thirds of our physicians were trained in ultrasound use in residency, including two or three docs who did ultrasound fellowships. The verdict?  About half of our docs never use the machine, and of those of us who do use it, we have infrequent need for it.

For example, in the community, vascular access has not been a terrifically challenging issue.  I rarely have to do central lines, and when I do, a blind approach works fine.   Blunt trauma goes straight to CT scan.  Potential AAA cases are infrequent and almost always are stable enough for CT (and as often as not the vascular surgeon wants the CT before operating anyway). Biliary studies are challenging and unless the results are pretty clear most docs are not comfortable performing these studies without confirmatory follow-up.  I would not be able to rule out torsion with the machine we have (Sonosite Micromax), and so I tend not to do pelvic ultrasounds in non-pregnant patients.  The other "indications" such as retinal detachment (really?) are rare enough and require enough operator skill that it's just not practical or necessary in the typical ED.

Where we do use the ultrasound a lot is for OB exams, especially in the anxious/spotting first-trimester moms.  It takes ten seconds to find the heartbeat, print out a picture, and the patient goes home happy and quickly.  That's awesome -- it's a common presentation and the ability to quickly turn those cases around saves a ton of time and money.  I also use them for paracentesis on liver patients -- which can also be done blind, but since we have the machine, why not?

The bottom line is that these machines are expensive, and the utilization needs to be high enough to justify the cost (especially if you are not billing for the studies).  The "real world" experience is pretty marginal there.

Now don't take away that I am opposed to bedside ultrasound -- I love it.  I recently diagnosed a ruptured cornual pregnancy with it, and having that capability really expedited the patient to the OR.  I don't use it every shift, but pretty frequently.  I was highly instrumental in convincing our hospital to purchase it. But the business plan -- the cost/benefit -- is a challenging argument to make to administration.  "Well, Mr CEO, I'd like you to spend $50,000 to buy a machine that won't bring in any additional revenue, will be used infrequently, and is guaranteed to piss off the radiologists."   We sold it based on quality of care and patient safety, especially for the infrequent cases where it really can be lifesaving, and we are lucky enough to have collaborative radiologists who work well with us on this issue.  So we won that round.  But I can't bring myself to be too critical of the ED director who chooses other matters to expend political capital over, especially in a smaller ED.

Addendum: yes, I know that bedside ultrasound can theoretically generate revenue.  But it's not easy: you need an ED doc who is enthusiastic and committed enough (or compensated) to champion it in radiology, oversee the image archiving and QA, and ensure that the docs performing it are doing it right.  You need a friendly radiology department (or a powerful ally on the medical staff).  And you need compliant payors who actually reimburse for ultrasound, which they do not in many areas.  A lot of stars have to align to make bedside ultrasound a financially valuable service, and that is very challenging in a non-academic setting.


  1. Its been a few years for me- but is FAST ultrasound still being used for blunt abdominal trauma work-ups- I trained in the era when DPL was being swept aside in favor of FAST abdominal U/S...do you do them in your ER?

  2. We do them fairly rarely. FAST is best used when you have an unstable patient, you know they need to go to the OR, and you just need to know which body compartment to open.

    If they are stable and they need imaging, they get CT. Sometimes for fun we'll do a FAST prior to CT. If they are stable and low-risk, we generally don't image them at all.

  3. I wonder if BSU is used more often at hospitals that don't have 24 hour U/S services? My hospital only has u/s tech on call 10p-7a. And like you said, about 1/2 of our docs do atleast basic u/s. So it seems they are more likely to attempt their own BSU at night, maybe because they don't want to wait the 30 minutes for a tech to show up?

  4. Hmm, my ER sounds similar to yours (community, little trauma)and our newer docs have spent a considerable amount of time training the older docs. In addition to the uses you mentioned, we DO use them a lot for vascular access. Also checking for cardiac motion in full arrests and occasionally for urinary retention issues. In fact, I had a doc call the urologist when our Foley didn't show up in the bladder on U/S (but still drained 300ml of urine. I think they save the cost of calling in a tech in the middle of the night, and sometimes render additional testing unnecessary. They also save time which can be hard to quantify in $$.
    -whitecap nurse

  5. Healthcare Observer9/08/2009 1:26 PM

    Surely you shouldn't be discussing this in the context of 'revenue generation' given the woeful problems the US has with overtreatment and piece-rate fees.

  6. Revenue is a part of the business of medicine. You buy a $50,000 machine, "How's it getting paid for?" is a valid question. If you accept that ultrasound is a good thing to have in the ED, that it enhances the quality of care and patient safety, then the funding -- which in our system is dependent on reimbursements -- needs to be considered.

  7. Healthcare Observer9/08/2009 2:27 PM

    Sure, but are the costs written off over the overall operating budget - which would suggest commitment to clinical need only - or do you make a charge for every scan? And if the latter, do you stop that charge when the machine is paid for?

  8. I think DVT is another great use, and incredibly simple to do.

    Ultrasound looks like it may in the next few years be confirmed as a better test than CXR for pneumothorax.

    And if you've got a first time likely kidney stone, and you don't wanna image, seeing hydro on the affected side with an empty bladder gives you your answer, too.

    AAA I thought had a fairly high miss rate, often because it's nonspecific, sometimes presents like a kidney stone. You can quickly run the aorta in 30 seconds and rule it out.

  9. I know everyone thinks turf wars are a big issue with this, but as an ultrasound tech I LOVE that some of our ER docs can at least attempt a bedside u/s. I get really really tired of coming in on call 3-4 times a night to scan a spotting first trimester pregnancy. And I KNOW gets irked when I have to wake him up 3-4 times a night. I always appreciate the nights when I'm on with a doc who can at least try... then I don't mind so much coming in for something.

  10. my radiologist gets irked... sorry left that important word out :)

  11. Anon,

    Totally right. I almost forgot -- we sold this to radiology on the theory that we might order fewer studies during night hours for their overburdened techs. I don't know that it's really worked out to a huge degree, but every little bit helps.


    Many kidney stones do not cause hydronephrosis, and hydroureter is a delicate finding even for experienced ultrasonographers, so CT remains the confirmatory test of choice, when needed. I would argue that every first time kidney stone deserves imaging. Recurrent kidney stones require neither CT nor U/S IMHO, so long as not septic and a good clinical story.

    And if AAA is in your differential, it's hard to argue that anything less than CT is standard of care. Sure, it they are thin and you are confident in your imaging skills, U/S can do, but I've imaged many a fat old guy in whom the Aorta was fuzzy, if visible at all. This is not to say "It's useless, don't do it," but rather, "ultrasound is a fairly limited screening tool."

    As for DVT, it requires a better machine than we have in our ED.

  12. There is even a push to use ultrasound in EMS. One theory is that it will help to spot the benign appearing abdominal injury and avoid sending them to the community hospital. There is a bunch of research from Iraq that supports it, but I do not see a benefit in civilian EMS.

    The full text of a review of EMS ultrasound is available for free at PubMed Central.

    Int J Emerg Med. 2008 Dec;1(4):253-9.
    Use of ultrasound by emergency medical services: a review.
    Nelson BP, Chason K.

    In EMS, I think we have far more important things to try to improve, before we start adding toys. It might be nice for EMS, in more places, to consistently intubate at a success rate that reflects competence, rather than a psychotic game of Whack a Mole.

  13. anotherEDdoc12/19/2009 2:17 AM

    I'm an ED doc, started using US about 10 years ago during residency. I teach/ use ultrasound at our academic center, but also work frequently at a freestanding community ED (ED ultrasound, techs on call, but no residents or on site consultants) as well as occasionally working at an urgent visit that doesn't have ultrasound.

    In fact, it is 5am in our freestanding ED and have a bit of a lull. I just used ultrasound on the last two patients -- one was a 33 y.o. who had a metatarsal fusion earlier in the day. He was in excruciating pain, unrelieved by percocet. I used the ultrasound to identify the nerves supplying his foot and injected lidocaine and bupivicaine under US guidance -- he got complete relief without further narcotics, he is now on his way home to sleep. This should last 12 hours at least and hopefully the post-op pain will be tolerable by then.
    The other patient was an obese woman with knee pain x1 week. Works the night shift. Couldn't identify any fluid on physical exam but found a lateral bursa filled with fluid, was able to aspirate and inject anesthetic and a steroid -- complete relief.

    I use it all the time to get peripheral IVs on patients with difficult access.

    Can you do ED work without US? Sure, but if you really know how to use it in my opinion you can be a much more effective doc.

    If you see really sick patients (particularly unexplained hypotension), an integrated US can be invaluable -- ? free fluid, ? pericardial effusion, hydration status, PE, AAA, etc.

    I've found thoracic ultrasound (lung rockets) to be very helpful in identifying CHF/ interstitial syndrome in undifferentiated dyspnea.

    You might be surprised how many significant pericardial effusions are out there if you start looking more frequently. Our hospital just settled a case of a patient admitted with an undiagnosed pericardial effusion who died on the floor prior to us having US.

    Last month I diagnosed a dilated cardiomyopathy in a 20 y.o. who presented with cough asking for antibiotics.

    Have numerous cases of people triaged as non-urgent with internal bleeding, including one a few months ago who presented with abdominal pain after eating a sandwich, but forgot to mention hitting his left side wrestling with his kids earlier in the week. He dropped his pressure and after a surprisingly positive FAST was found to have a ruptured spleen requiring emergent operative intervention.

    Many places I go to teach courses don't have access to US overnight. Recently reviewed a case of a patient who received "prophylactic" lovenox while they awaited DVT US. They didn't have a DVT, but bled into their ruptured baker's cyst requiring an operation for compartment syndrome.

    Have stopped several inappropriate chest tubes when a trauma resident "didn't hear breath sounds" but there was lung sliding on the US.

    I could go on...when I do occasionally practice in a place without bedside US I feel as if my hands are tied. I get correspondence all the time from current and former residents in places near and far who are effectively using US.

    I've always felt it would take a few turnovers of the workforce with good US training in residency for US to really start to reach its potential in the community. We still have a ways to go. Also, as the equipment gets better (cheaper, better resolution, better workflow solutions) it will be more widely used.

  14. anotherEDdoc12/19/2009 2:18 AM

    A few comments on some of those above:

    -Reimbursement. Needed when there are costs for equipment, time, training, QA, image retention; but not sustainable at current US radiology reimbursement rates. We need CPT codes specifically for point of care US that are not over-reimbursed. Reimbursement for bundled care may also make US by the doc much more desirable to the hospital. Does healthcare observer think the radiologist stops charging for the MRI once it is paid off?

    -Renal US and CT. While the ACR would agree with the post that all first time renal colic needs a CT, a kidney stone is unlikely to kill you -- but as recent literature highlighted a CT might. A young patient with flank pain, hematuria and hydronehprosis on bedside ultrasound has a stone, and the vast majority of these will pass without intervention.


Note: Only a member of this blog may post a comment.