Morgenstern, J. The FAST exam: overused and overrated?, First10EM, August 30, 2021. Available at:
https://doi.org/10.51684/FIRS.85822
Might the FAST exam be the most overused test in emergency medicine? If not the most overused (it does have to compete with the white blood cell count after all), perhaps it is the most overrated? Considering the sacrosanct position ultrasound holds in emergency medicine, even asking this question might get me in trouble. However, all tests come with risks and benefits. All tests have false positives and false negatives. I have seen numerous patients harmed by seemingly unnecessary FAST exams, and it makes me wonder whether this test (as it is currently being used) is truly providing a net benefit in our patient population.
The official role of the FAST exam in trauma is to identify patients with hemorrhagic shock who require immediate transfer to the operating room.1,2 “The aim is to identify life-threatening intra-abdominal bleeding or cardiac tamponade with a view to expediting definitive surgical management. It does not aim to exclude abdominal or thoracic injury.”1
Despite this clear vision from ultrasound experts, this is not how the FAST exam is actually being used. We don’t reserve it for the sickest patients; we use it on everyone. We scan patients with normal vital signs. We scan patients without abdominal symptoms. The vast majority of patients being scanned clearly don’t require a surgical procedure, let alone “immediate transfer to the operating room”. In contrast to the clearly stated aims, we are scanning to rule out disease, not rule it in.
Unfortunately, the FAST exam is an imperfect test. The sensitivity is reported to be between 41% and 95%, so the FAST exam cannot be used to rule out intra-abdominal injuries.2,3 Furthermore, although the FAST exam is very specific, in hemodynamically stable patients, most trauma experts suggest performing a CT after a positive FAST scan to confirm and further delineate injuries. Thus, in hemodynamically stable patients, a CT scan is recommended whether the FAST scan is positive or negative, and therefore “the use of focused assessment with sonography for trauma in hemodynamically stable blunt trauma patients seems not worthwhile. It should be reserved for hemodynamically unstable patients with blunt trauma.”2 Other experts agree, stating that “FAST is a rule-in triage tool in patients with [blunt abdominal trauma], not a true diagnostic tool…. and we must question whether FAST represents a productive use of health resources in terms of equipment and training.”3
In keeping with its poor diagnostic characteristics, there is no evidence that the FAST exam improves clinical outcomes. A Cochranre review identified 4 studies, and the FAST exam was not associated with any improvement in mortality (RR 1.00; 95% confidence interval 0.5 to 2.0). They conclude that “the experimental evidence justifying FAST‐based clinical pathways in diagnosing patients with suspected abdominal or multiple blunt trauma remains poor.”4
The FAST exam has not been well validated in a pediatric population, but appears to be even less accurate than in adults.5 A 2007 meta-analysis found that the sensitivity of ultrasound for intra-peritoneal fluid in children was only 66%.6 Almost all of the included studies involved formal ultrasounds read by radiologists, so the sensitivity is likely to be much worse in the hands of a community emergency physician. Studies of the FAST exam performed by emergency physicians in a pediatric population reveal a sensitivity between 28% and 52%.7–9 Furthermore, it is not clear that identifying free fluid is helpful in the pediatric population, as children can have free fluid without any injury, and also frequently present with injury in the absence of free fluid.10
In keeping with the poor diagnostic accuracy of the FAST exam in a pediatric population, the clinical evidence does not support a benefit. There is a randomized clinical trial looking at the use of the FAST exam in a pediatric population, and they conclude that “among hemodynamically stable children treated in an ED following blunt torso trauma, the use of FAST compared with standard care only did not improve clinical care, including use of resources; ED length of stay; missed intra-abdominal injuries; or hospital charges. These findings do not support the routine use of FAST in this setting.”9
Real world experience seems to support this conclusion. One large observational study in pediatric trauma concludes that “true positive FAST exams are uncommon and would rarely direct management. While the negative FAST would have potentially reduced CT use due to practitioner reassurance, this reassurance may be unwarranted given the test’s sensitivity.”11
With such underwhelming evidence, how did the FAST exam come to be thought of as a “standard of care”? Aside from the cool factor of being able to peer inside patients’ bodies, I think the overuse of ultrasound arises from our general misunderstanding about the harms of tests. We are all keenly aware of direct harms, such as the radiation from CT, but massively underestimate the harms of being misled by our tests. Humans are generally bad at grasping probabilities, and we don’t properly account for the effects of false positives and false negatives on our patient’s outcomes.
“This is a 16 year old who fell off her bike. Her vitals are normal, and she has no pain on exam, but I think I saw some free fluid on the FAST exam, so we are waiting on the CT.” Over the past few years, I have received this general hand-over many times. The CT has been negative in every single case. Clinically, it was obviously that imaging was unnecessary, but indiscriminate use of the FAST exam generates false positives. Young children received unnecessary radiation. Eventually, incidental findings will lead to unnecessary invasive procedures and even more harm. Ultrasound may not directly harm our patients, but our failure to understand its test characteristics, and adjust for pretest probability, certainly does.
What about the FAST exam in non-trauma situations? I recently cared for a woman in her 20s who presented with severe pelvic pain, a positive home pregnancy test, a heart rate of 140, and a systolic blood pressure below 90. Within 30 seconds of arrival, an ultrasound allowed me to identify free fluid in her abdomen, and obstetrics was involved in her care less than 5 minutes after arrival to the hospital. It certainly felt like the FAST exam might have saved her life.
However, the more I reflect on this case, the more I realize that I might be trapped in a technological mirage. Imagine I was caring for this same patient in the era before ultrasound. Would anything have changed? Even before I put the ultrasound on her belly, I had a young pregnant female with severe pelvic pain and hemodynamic instability. I already knew the diagnosis. The ultrasound might have made me or the obstetrician more comfortable, but did it really change the patient’s management? I am fairly certain that even without the ultrasound, I still would have called the obstetrician within 5 minutes.
Many ultrasound results fall into these classic diagnostic ranges of futility. Positives that were obviously positive based on clinical findings before ever turning on the ultrasound machine. Negatives were obviously negative. This exposes the limits of anecdote, and the need for high quality science, when assessing the value and impact of bedside ultrasound in emergency medicine.
Ultrasound is an amazing tool in modern emergency medicine. I use it almost every shift. I think it has made me a better diagnostician. The procedures I perform are quicker, more successful, and safer with an ultrasound in my hand. But I think we have gone too far. When I discuss these cases with colleagues, ultrasound is assumed to be a universal good. When I suggest ultrasound might not be appropriate in hemodynamically stable, asymptomatic patients, I encounter surprise and disbelief. I have been told ultrasound is the standard of care in such patients, despite the clear lack of evidence and potential for harm.
Bottom line for the FAST exam
The evidence of benefit is non-existent, but I think there is clearly a role for the FAST exam in hemodynamically unstable patients. However, even among the hemodynamically unstable, the role may be more limited than you think. It is essential to consider your resources, and how the test will change your management. In a trauma center, when you are deciding between the operating room and interventional radiology, the FAST exam may provide invaluable information. In the community, where most of us work, such treatment options don’t exist, and the FAST exam may just be delaying the transfer the patient needs.
In hemodynamically stable patients, I think the available data is pretty clear: the FAST exam shouldn’t be used. The sensitivity is not high enough to rule out intra-abdominal injuries. If you are concerned about a patient, CT is the imaging of choice, and a negative ultrasound may just falsely reassure you. The specificity of the FAST exam is excellent, but many hemodynamically stable patients will be managed non-operatively, so essentially all of these patients are going for CT as well. In other words, whether your FAST exam is positive or negative, the patient still needs a CT, which is the definition of a useless test.
Other FOAMed
Tagg, A. Thinking FAST, and slow, Don’t Forget the Bubbles, 2018. Available at: https://doi.org/10.31440/DFTB.17324
Carley, S. On a FAST track to nowhere at St.Emlyn’s. St. Emlyn’s Blog, 2013. Available at: https://www.stemlynsblog.org/on-a-fast-track-to-nowhere-at-st-emlyns/
An opposing view is available in this talk from the amazing Vicki noble:
References
1. Rippey JCR, Royse AG. Ultrasound in trauma. Best Pract Res Clin Anaesthesiol 2009;23(3):343–62.
2. Natarajan B, Gupta PK, Cemaj S, Sorensen M, Hatzoudis GI, Forse RA. FAST scan: Is it worth doing in hemodynamically stable blunt trauma patients? Surgery 2010;148(4):695–701.
3. Smith G. Standard deviations: flawed assumptions, tortured data, and other ways to lie with statistics. 2015.
4. Stengel D, Rademacher G, Ekkernkamp A, Güthoff C, Mutze S. Emergency ultrasound-based algorithms for diagnosing blunt abdominal trauma. Cochrane Database Syst Rev 2015;(9):CD004446.
5. McCormick T. Pediatric Major Trauma Assessment [Internet]. In: CorePendium. CorePendium LLC; Available from: https://www.emrap.org/corependium/chapter/recwbCwJMhkskM6Jt/Pediatric-Major-Trauma-Assessment
6. Holmes JF, Gladman A, Chang CH. Performance of abdominal ultrasonography in pediatric blunt trauma patients: a meta-analysis. J Pediatr Surg 2007;42(9):1588–94.
7. Fox JC, Boysen M, Gharahbaghian L, et al. Test characteristics of focused assessment of sonography for trauma for clinically significant abdominal free fluid in pediatric blunt abdominal trauma. Acad Emerg Med 2011;18(5):477–82.
8. Calder BW, Vogel AM, Zhang J, et al. Focused assessment with sonography for trauma in children after blunt abdominal trauma: A multi-institutional analysis. J Trauma Acute Care Surg 2017;83(2):218–24.
9. Holmes JF, Kelley KM, Wootton-Gorges SL, et al. Effect of Abdominal Ultrasound on Clinical Care, Outcomes, and Resource Use Among Children With Blunt Torso Trauma: A Randomized Clinical Trial. JAMA 2017;317(22):2290–6.
10. Butts C. The Speed of Sound: Not So Fast. Emergency Medicine News 2016;38(8):9.
11. Scaife ER, Rollins MD, Barnhart DC, et al. The role of focused abdominal sonography for trauma (FAST) in pediatric trauma evaluation. J Pediatr Surg 2013;48(6):1377–83.
12. Beno, S. Bottom Line Recommendations: Multisystem Trauma (2020) [Internet]. 2020;Available from: https://trekk.ca/system/assets/assets/attachments/500/original/2021-01-08-MST_v_3.0.pdf?1610662473
13. duPont, A. Management of Abdominal Solid Organ Injuries [Internet]. Available from: https://pediatrictraumasociety.org/multimedia/files/clinical-resources/SOI-1-1.pdf
14. Schonfeld D, Lee LK. Blunt abdominal trauma in children. Curr Opin Pediatr 2012;24(3):314–8.
15. Brunetti MA, Mahesh M, Nabaweesi R, Locke P, Ziegfeld S, Brown R. Diagnostic radiation exposure in pediatric trauma patients. J Trauma 2011;70(2):E24-28.
Morgenstern, J. The FAST exam: overused and overrated?, First10EM, August 30, 2021. Available at:
https://doi.org/10.51684/FIRS.85822
15 thoughts on “The FAST exam: overused and overrated?”
Hi !
Emergency physician there, working essentially in pre-hospital situations, and sometimes in intra-hospital emergency unit, in France.
Your article is, as always, very interesting. I am also a big sonography user, and I totally agree with you that FAST and POCUS has to be used as a rule-in tool and not a rule-out, and on patients whose clinical situation isn’t clear enough to be “labelled” only with clinical examination (your example with the EP is very relevant).
However, I would be very interested to read your opinion about FAST as a bed-side tool to assess the evolution of the patient.
I truly believe (no EBM on that…) that perfoming FAST at T0 will not help me rule out my patient, but may help me having a T0 glimpse of my patient’s situation. And that glimpse will be very helpfull for me of for the next ER’s physician, to be compared with later FAST.
Young patients may be tricky, because their CV system may sustain a big loss of blood before starting to collapse ; by big, I mean big enough to be seen with US.
I don’t see FAST as a one-shot exam, but more as a bedside tool for quick and reconductible evaluation, at least in that case.
I’d be glad to hear your thoughts on that !
This is definitely one of the major limitations of the literature on POCUS in general – it almost always treats it as a one time test, when in reality, we almost always make multiple assessments over time.
Multiple assessments should increase sensitivity, and might improve upon clinical judgment and vital signs in identifying deterioration – but I am cautious, as neither of those have been demonstrated scientifically. My main point is to be cautious about the harms of our tests – and to remember that even something as benign as FAST can have significant harms when misused. Therefore, we really want to see proof of benefit before widespread adoption. However, as a clinician practicing now without those studies, using FAST seems reasonable, as long as one is very critical about why they are using the test, and how the test outcomes will change their management.
Cheers
Justin
Totally agree with everything
Reminds me of this other great editorial I believe Casey Parker and yourself had on BroomeDocs. “Crager and Hoffman: But It Makes Sense Physiologically…“
Our expertise should be assessment of pretest probability and which tests make sense to shift probabilities to a meaningful post-test probability (I.e to above treatment threshold or below diagnostic threshold, for the ddx in question). PoCUS may have a broader role than the specific way that it is intended to be used , but we need to do the studies…. if we don’t we just end up with this chaotic scenario where it becomes a “screen all” at the entrance and false positives / cascade of care will drown us and the patients . Healthcare creating more harm than good.
On another note: outside of the big centres, I really don’t see why PoCUS became our “thing” and not for instance “compassion and communication” . It is more generally useful for assessing the pretest probability and “easing suffering / healing/helping the patient” (NNT=1) in one swoop .
I feel like every time you leave a comment, I want to copy and paste it into the main article
Thanks!
I would love to hear your opinion on white blood cell count. This is often debated where i work.
I suspect in a a highly resourced environment POCUS probably doesn’t confer mortality benefit but it might accelerate disposition decisions. But what about mass casualty incidents?
https://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.424.146&rep=rep1&type=pdf
I am admittedly a gratuitous user of POCUS. I justify this not for the decisions it helps me make, but to ensure I maintain my skillset for when it really counts.
For sure. This isn’t a black and white “no one should ever use the FAST exam”.
The argument is more that we are massively over-using it, and that over-use subtly transforms into “standard of care” and loss of critical thought.
Well.
In the spirit of healthy debate, allow me to provide you with a view diametrically opposed to your own.
Your position is that “The official role of the FAST exam in trauma is to identify patients with hemorrhagic shock who require immediate transfer to the operating room.” and that deviation from this has led to poorer outcomes. I put it you that whoever subscribes to that premise has exactly the wrong approach to POCUS.
Over the past 20 years, I have taught more than 20,000 physicians worldwide (though primarily in Canada) to use POCUS. I believe they are almost all providing enhanced care to their patients because they have made the paradigm shift necessary to do so. That is to understand that POCUS is NOT something to use only on the critically ill. It is an extension of the physical exam, to be used on very large numbers of patients to rule-out conditions. It yields ONE data point. Said data point must be integrated into the entire diagnostic algorithm, and re-integrated repeatedly in some cases, as the case evolves. This is particularly true in FAST.
Consider the statement you quote approvingly:
“the use of focused assessment … should be reserved for hemodynamically unstable patients with blunt trauma.”
And consider the case you posit, that of a young woman clearly in shock, with a positive pregnancy test.
In both these situations, the pt has so many clues as to their impending demise that no single test will change the management. This would seem to disprove the stance of the “experts” who claim that FAST should be reserved for the critically ill. In my case, I have always argued that the “sexy save” will make up a tiny number of the POCUS scans one does, IF you are using POCUS properly.
So let’s talk about the non-critically ill, those you claim are harmed by POCUS. If you have never seen a perfectly well-appearing pt who has a dramatically positive POCUS and a belly full of blood, you are not doing nearly enough POCUS scans. By doing POCUS on all comers, you detect injuries BEFORE they are hemodynamically unstable. That is clearly a good thing. You talk about pts being harmed by POCUS. There are FAR more who have been harmed by the POCUS not being done. These turn up regularly in the medicolegal reports I am asked to comment on.
Your use of the post-POCUS CT to argue against POCUS also misses several important points. The reason we do CT after POCUS is to define and delineate the lesion, because POCUS cannot do that. It can only detect free fluid. But knowing the fluid is there is HUGELY beneficial. The pt goes to the head of the line for CT. They do not wait in the ED for a few hours, during which time their bleed can worsen, sometimes catastrophically. They go with two big lines instead of one. They are cross-matched for 6 units instead of 2. Surgery is immediately advised of their presence, because they are certain to be admitted for observation, at the minimum. And all staff are on the lookout for the tachycardia that fails to resolve with fluid administration, the first sign of hemodynamic decompensation the usually-young trauma victim.
You claim that the lack of positive outcomes, measured by decreased mortality, shows the uselessness of POCUS. Again this is because you have failed to make the paradigm shift described above. Taking a vital signs at triage does not affect outcomes either. The lives saved are far too few to show up in statistical analysis. But if you stopped doing so, people would die because some very sick people would be told to stay in the waiting room rather than being hustled into the treatment area. More to the point, a series of negative POCUS scans through a night shift allows you to defer CT till the morning, or cancel it altogether. THAT is a huge savings of health care resources, not to mention a huge decrease in MD stress, be it from watching a potentially seriously injured pt or from arguing with radiology to do a middle-of-the-night scan. But again, you need to think of POCUS as part of the physical exam.
You state that you have accepted many transfers that go along the lines of:
“This is a 16 year old who fell off her bike. Her vitals are normal, and she has no pain on exam, but I think I saw some free fluid on the FAST exam, so we are waiting on the CT.”
Were I the transferring physician, you would hear me say:
“This is a 16 year old who fell off her bike. Her vitals are normal, and she has no pain on exam, but I thought I saw some free fluid on my first FAST exam at the extreme of one of my sweeps*. I immediately put her in Trendelenburg and log rolled her to the right to send whatever fluid was in the abdomen to the RUQ**. That FAST was clearly positive. I have advised surgery that the pt is coming to them for observation, but have recommended that serial FAST scans be done rather than exposing the pt to radiation, since the majority of these pts do not end up needing surgery.”
Or you would hear me say:
“This is a 16 year old who fell off her bike. Her vitals are normal, and she has no pain on exam, but I thought I saw some free fluid on my first FAST exam. I immediately put her in Trendelenburg and log rolled her to the right to send whatever fluid was in the abdomen to the RUQ. That FAST was clearly negative. I repeated the FAST 15 minutes later. It was till negative. I repeated it an hour later. After an hour in TBurg, it was still negative. She is still pain free and stable. I suggest you scan her again at the 8 hour post-injury mark, after having placed her in TBurg for 15 minutes. If the FAST is still negative and she is still pain free and stable, she could be safely discharged with instructions to return if she experiences any pain or dizziness.”
If you are CT’ing this pt, it is because of the defensive medicine you are forced to practice in the USA.
You state that you “have seen numerous patients harmed by seemingly unnecessary FAST exams”. That is incorrect. You have seen people harmed by a lack of understanding of POCUS.
Regards,
Dr. Ray Wiss
Creator of the EDE – ED Echo course
* Do you sweep the area of interest when you do a FAST? Many POCUS practitioners do not, which leads to vastly worse sensitivity.
** I am sure you know the RUQ is the most high-yield area for FAST, but others reading this may not
Thank you for the in-depth comment.
Unfortunately, I think you are falling into a very common cognitive bias that occurs in medicine. You posit that scanning a large number of well appearing patients will occasional find a belly full of blood, but such a use of the FAST exam is just not supported by its test characteristics, nor by the available clinical evidence. These positive cases may stand out in your mind, but the many false positives and false negatives will not. The only way you can account for the true impacts of a test is to take a step back and study its clinical impacts, and thus far, the clinical trials of the FAST exam have all failed to show a patient benefit. (I would love to see an RCT published of your more nuanced approach to the FAST exam. It would be great to have some evidence that this test helps, but right now we don’t.)
Your analogy to vital signs almost certainly fails when it comes to the FAST exam. There is an inherent technological bias that shapes how we respond to ultrasound results. We all ignore elevated heart rates many times per shift. We don’t ignore ultrasound findings in the same way. So although vital signs are used reasonably well as a screening tool, ultrasound is currently used very poorly when it is adopted as a screening tool, instead of being used as a specific test to ask a specific question with a reasonable pretest probability.
Although I love your approach to the false positive FAST scan, it is not the approach of the vast majority of emergency physicians. I would guess that less than 1% of physicians would approach the scenario as you describe, and that is exactly the problem with the FAST exam as it is currently used. (And you make an incorrect assumption – although I would never have CT’d this patient – the cases I have seen were all in Canada and New Zealand, so this is a world wide problem that has nothing to do with defensive American healthcare.)
Although I think it is fair to quibble between “harm from the FAST exam” and “harm from poor understanding of the FAST exam”, that is a semantic argument that completely fails to account for the fact that the use of the FAST exam is entirely based on the understanding of the FAST exam. I think it is very clear that the general understanding and implementation of the FAST exam in modern emergency medicine is very poor. Therefore, it is very reasonable to state that patients are being harmed by unnecessary FAST exams.
I wish we could will in a world where everyone had as nuanced an understanding of this test as you. Unfortunately, we do not, and therefore we need to discuss the significant limitations and harms of the FAST exam as it is currently being used.
All the best
Justin
I hope you are having fun, because I am. 😉 I love debates like this and believe this is how medicine advances.
Being older than dirt, it never occurred to me to Google you before replying, and casting aspersions on your nationality. As it happens, my course did go to your center in 2010, so if you can honestly say that “less than 1%” of your colleagues would have scanned your hypothetical case the way I would have, perhaps my teaching is not as effective as I would like it to be.
But there are still flaws in your logic. You state:
“These positive cases may stand out in your mind, but the many false positives and false negatives will not.”
MANY false positives and negatives?? Think about that for a second. If that were true, I should have seen a large number of my POS scans be deemed FPs by a CT, and a large number of my NEG scans bounce back with free fluid in the abdomen. The only instance of the former I have had was a FAST that was declared POS by one of my residents, where the CT came back as “normal”. But when I enquired further, it turned out the rad had felt that the amount of FF was trivial, and he declared it NEG. So my resident’s FAST was, in fact, accurate. Don’t get me going about rads calling images without the clinical context. As for an FN, I have never had one. Admittedly, this is “that I know of”, but in a city where there is a single ED, it is very unlikely).
I would disagree that my approach is “more nuanced”. Rather, it is based on a solid understanding of the strengths and limitations of FAST. That is why I have always emphasized that POCUS, and FAST in particular, does NOT save lives in a statistically significant manner. But can you honestly argue that the day I sent five (5!) first trimester bleeds home in 15 minutes because I detected LIUP in all five did not have a major impact on the patient care in the ED that day? Similarly, to ignore the impact of serially negative FAST scans on the emotional well-being of the physician ignores the reality of those who work in minimal-resource settings. To cite an extreme example, during my two tours in combat in Afghanistan, I was able to cancel several helicopter evacuations because a soldier with numerous shrapnel wounds to the chest had no hemo/pneumo at first contact, and remained so over multiple scans. Not asking a crew to risk their lives flying into a non-permissive environment was one of the most impactful things I did over there. POCUS made that possible.
As for the integration of POCUS into a diagnostic algorithm, my mantra for the past 20 years has been that it is a single data point, one that in the FAST application is extremely time-sensitive, hence the axiomatic need for a re-scan before anyone goes home. And the same goes for vital signs. Do we really “ignore” elevated heart rates? No. We put that single data point in the diagnostic algorithm and decide that, on the balance of things, we should not pursue whatever the elevated heart rate suggested because there is too much countervailing evidence. You agree with this position when you state that:
“If you are concerned about a patient, CT is the imaging of choice, and a negative ultrasound may just falsely reassure you.”
The first two clauses are correct. “If you are concerned about a patient, CT is the imaging of choice.” But the last clause is not. See my earlier reply for the benefits of the NEG FAST in this situation. This is why the First and Last of the “10 Commandments of EDE” is “Clinical skills rule”.
Finally, much of your argument is based on the statement that we “massively underestimate the harms of being misled by our tests”. In other words, you think that FAST causes huge harm. But if anything, you make the case for the contrary when you state that, in the hemodynamically unstable pt, FAST does not alter anything because there is so much pushing you to intervene*. And the statement:
“…the FAST exam may just be delaying the transfer the patient needs.”
begs the question: how long does it take to do a FAST? Especially in a hemodynamically unstable patient, the chances of getting a true POS in about 3 seconds are very high. And even in a stable pt in whom you really want to go around the block, you should be done in under 30 seconds. And that is WITH proper sweeping.
In other words, you are looking in the wrong place for benefit, and you have not proven harm.
Bottom line: we are closer to agreement than you might think. But that comes from my endlessly-repeated position that an intro course is NOT enough to get you out there using POCUS effectively. For that, you need to have Independent Practitioner (IP) status, as per the Canadian Point of Care Ultrasound Society (CPOCUS, the successor to the Canadian Emergency Ultrasound Society (CEUS) that I created in 2002). I am willing to believe that some, maybe even many, of the 20,000 physicians who have taken my course are not following my mantras to the letter. But I am quite certain that IPs practice much more stringently.
Back to you,
Ray
*The “OR or IR” exception is an important point not enough people have fully integrated, so good on you for mentioning that.
I absolutely enjoy a good debate
I imagine our thoughts on POCUS are actually not all that different – but I think there is a huge gap between the ideal we would both strive for, and how the tests are currently being used.
This post is only about FAST, and not POCUS in general. I love POCUS. Like you say, I think being able confirm an IUP and disposition a young pregnant patient in 5 minutes is phenomenal.
I am not sure why our experiences are so different, but I have had multiple hand-overs in just the last few years where a false positive FAST scan led to a completely normal CT. Perhaps I have just been unlucky? (Comments on twitter and facebook are mixed – some say they have never seen this, and others seem to think it is a common occurrence.) Perhaps people are more careful handing over to you? Or you double check ultrasounds that have been done? They won’t be all that common, because the specificity FAST is excellent, but when done in a 0 risk population, false positives are inevitable.
More concerning to me is the false reassurance of negative scans. Even you admit that you make major changes in management based off negative FAST exams (although presumably in combination with clinical factors, and with serial exams), but the sensitivity in the published literature simply doesn’t support such an approach. In your expert hands, with the combination of serial exams and clinical judgement, I am not surprised this has worked. (Although I would guess that other aspects of the clinical picture play a much bigger role that the FAST exam, and that you may still be just as good if I made you work a shift without the ultrasound machine.) However, I don’t believe this is the way the FAST exam is being used in most cases. (And there seems to be a reasonable amount of support for that position in the social media discussion).
As for how long it takes, I absolutely agree that scan itself takes 15 seconds. But finding the machine (usually with dead batteries) adds time. Getting it into a COVID airborne rooms takes time. And once it is there, it becomes the priority, but it often shouldn’t be. Nurses are often moved out of the way to facilitate a FAST scan, when the IV start clearly should be a priority. So I think there are multiple logistical harms that also come with the widespread thoughtless adoption of FAST that epitomizes emergency medicine
Now part of my argument could be seen as a strawman. I am not really talking about the idealized FAST exam. I am talking about it as it is actually used. However, even the idealized FAST exam has significant limits. In my mind, it among the least important POCUS modalities we have. (Again, I am a huge proponent of ultrasound. Ultrasound guided procedures have revolutionized emergency medicine, and I think I have saved lives with echo, the RUSH exam for hypotension, and lung scans. It is just the FAST exam I am focusing on here.)
And I have never said it shouldn’t be used. It has a role. I just think it is overused, and I am concerned that people suggest it is a standard of care without any evidence to suggest it actually helps.
I truly think it needs to be used less. But more importantly, this post is written to increase understanding of the limitations of the FAST exam. With those limitations in mind, I think people would likely utilize the FAST exam much like you do, and that would be a win for everyone.
Cheers
Justin
For some reason, there are some problems with adding comments to the site right now. Dr. Wiss had another response, that I will copy and paste below from his email:
Great. Let’s keep the debate going.
I avoid social media like the plague, so cannot comment on whatever you are seeing there. See previous comment about being older than dirt…
You are entirely correct when you say that, of the basic applications, FAST is where the clinician most needs to stay true to rigorous methodology. Global cardiac activity, AAAs, IUPs, Pneumo/Hemo-TX are all much more straightforward. PCEs are somewhere in between. But a number of your other statements remain problematic.
“I have had multiple hand-overs in just the last few years where a false positive FAST scan led to a completely normal CT.”
If that is so, that means that you frequently accepted a handover where multiple data points would have suggested a CT was not needed. The only data point that argued for CT was the FAST, done sometime before you accepted the handover. Why did you not immediately repeat the FAST after 15 minutes of TBurg? As for the prevalence of FPs, I again ask about whether sweeping the AOI is common in your center.
Then you state that I change my management based on a NEG FAST
“….although presumably in combination with clinical factors…”
“presumably”?? Please re-read my earlier comments. FAST is a single, time-sensitive data point. Clinical skills rule. Never let POCUS take you in one direction when your clinical skills are pointing elsewhere. That is the crux of my argument. I make major changes in management based on SERIAL neg scans that, in the setting of ongoing hemodynamic and overall improving clinical picture, reduce the chances of serious lesions to almost zero.
“Nurses are often moved out of the way to facilitate a FAST scan , when the IV start clearly should be a priority.”
WOW! Diagnosis before resuscitation?? The problem there is not with FAST, it is with the ABCs of EM…
And finally:
“I truly think it needs to be used less.”
No. It needs to be used properly. That is what your observations suggest, and what you should be arguing for. And so I ask again, this time directly: how many people in your center are CPoCUS IPs?
Regards,
Ray
PS Nice rebuttal regarding abandoning the pediatric FAST exam here. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5863108/
In general, I think people mistakenly put ultrasound into a special class of diagnostic test.
POCUS is a very useful tool, but it is still a diagnostic test. It is still subject to the basic rules of diagnostic math, which means that when you use a test in the wrong population, you are bound to cause harm. (I would love to see someone perform a test-threshold calculation for the FAST exam. Perhaps I can work on that for a future publication.) For more on these generally misunderstood problems with diagnostic tests, see this post: https://first10em.com/the-fast-exam-overused-and-overrated/
I think having multiple viewpoints available to readers is very valuable, but at this point logic and semantic are not going to give us an answer to the question. If treated like any other diagnostic test, the poor sensitivity of the FAST exam means that it shouldn’t be used in low risk populations. However, sensitivity and specificity can be very misleading. (https://first10em.com/the-sensitivity-and-specificity-are-lying-to-you/) The only way that we can really settle this debate is will clinical trials. Although it has been around for a long while, the FAST exam is still a new test, and the burden of proof lies with those arguing that this new technology should be implemented. If it is as good as many think, this should be an easy hypothesis to prove. However, so far all of the clinical trials of the FAST exam show no benefit. Rather than arguing about the value of the FAST, I would suggest that the proponents of the FAST exam do the clinical trials to prove its benefit. As of right now, the best scientific answer we have is that it doesn’t help. I will rapidly change my mind and publish a new blog post if people can demonstrate a patient-oriented clinical benefit in well done clinical trials. I am just very doubtful, given its test characteristics, that that is possible.
I wholeheartedly agree that the fast is overused and that there are limitations that require thoughtful integration when using US to guide decision making.
I think there are a few other considerations that should be mentioned though. Pre US we would avoid CT scans in low risk patients and do serial abdominal exams (or even hemoglobins) prolonging their LOS in the ED to see if their abd exam evolved. I have had more than one patient where we opted for serial fast exams and the initial was negative or unclear with a subsequent exam being positive. I’m all for avoiding CT scans in low risk patients and using serial US may increase provider comfort with an earlier disposition, decrease LOS, and screen in the rare positive fast exam in a low risk patient. (Admittedly more research needed in this specific use of US).
i also think that there is value in having more confidence in your diagnosis. In the example you gave of the hypotensive pregnant patient, what if she just had a mildly lower BP with pelvic pain and a positive pregnancy test? That’s a patient we see all the time and a fast exam can be useful in triaging that patient before they get unstable if positive or just giving fluids and reevaluating if negative. Early diagnosis of free fluid has value when making big decisions such as going to the OR. And having the confidence of negative fast exams might save you an OB consult decreasing resource utilization in a beneficial way.
I can think of other anecdotal cases where the fast exam truly changed management (including on my last shift!), but you are right that we need high quality literature to figure out if there is measurable value in this tool. But its important that we don’t just look at morbidity and mortality. as you mentioned, we need to look at harm (negative CTs and procedures) and look at other outcomes like resource utilization, length of stay, time to diagnosis, or time to intervention. These studies are admittedly hard to do when you broaden the outcomes.
Lastly, there is an under appreciated value in repetition and familiarity. Saving US for the most unstable trauma patients when it is used as it was initially intended is helpful only when the provider has the confidence to perform the fast exam under pressure. The best way to develop that is to use US regularly and be familiar with the positives, negatives and limitations of US as a tool.
I agree with your thoughtful approach to any testing. I just wanted to share an alternate opinion as I’m writing a pro US talk for trauma grand round and found your editorial. Thanks for your perspective and for giving me additional things to consider.
Justin,
I’m late in reading this article, but I really do appreciate it!
One other phenomenon that I’ve seen in my residents and younger colleagues is them reaching for the ultrasound machine nearly immediately—before even completing the primary survey, never mind the secondary survey.
It disturbs me tremendously that these physicians forego a complete physical exam before reaching for the ultrasound. This prevents them from having a complete understanding of the patient (which impairs the ability to develop a pre-test probability) before using a test that, in my opinion, has a narrower use case than it currently enjoys.