Not-so-old Dogs and Not-so-new Tricks: Is There a Middle Ground for ED Ultrasound?

I read with interest the commentaries by the “new pups” and the “not-so-old dog” regarding their take on emergency point-of-care ultrasound (US). As a physician about to cross that self-perceived milestone of 10 years out of residency, nearly 9 years after completing an emergency US fellowship, I have experienced both the initial fascination with US as well as the realities of emergency department (ED) US use over a decade of practice, teaching, and research.

I recently attended a lecture by Peter Rosen, who is certainly (and very respectfully) one of the “old dogs” in emergency medicine—a founder of our specialty. He spoke about how in emergency medicine the majority of patients present with a condition that will get better no matter what we do, other patients will die no matter what we do, while in an important minority of patients, what we do can make a difference. The best physicians use training, experience, and available technology to discern this important group and to intervene without doing harm.

I do believe that proper training and appropriate utilization of point-of-care US can safely expedite care and improve cost-effectiveness, patient satisfaction, and our ability to take care of an ever-burgeoning ED population with fewer resources. However, this is not what makes me an ardent supporter of teaching and practicing bedside US. What comes back to me over my career so far are those patients in whom I was able to make a difference because I was able to effectively utilize point-of-care US. The unstable 18-year-old on her way to the operating room for a suspected hemorrhagic ovarian cyst, where a bedside US revealed severe right heart strain—allowing her to be treated for and survive a massive pulmonary embolism; the woman whose status post a laparoscopic bilateral tubal ligation with a normal hematocrit that the gynecologist urged me to discharge as “vasovagal syncope”—where a positive FAST allowed us to expeditiously address an aorta punctured by a trochar; the patient discharged the day before from a hospital across town that does not regularly utilize ED US with instructions for “constipation”—diagnosed with pericardial tamponade minutes after arrival in our ED and who arrested on the way to the cath lab but was safely resuscitated; the hypotensive patient received at sign-out as “sepsis going to the MICU”—where bedside US showed both right heart strain and a deep venous thrombosis leading to effective treatment with thrombolysis; the asymptomatic third-year medical student in whom we discovered right heart strain during a teaching exercise—who went on to have chest surgery for a large ostial septal defect that the surgeon stated could have taken 20 years off his life if it had gone undiagnosed; the thoracic aortic dissection diagnosed using bedside echo in a patient with a normal chest x-ray on his way to the chest pain center; and many more.

I realize that a collection of anecdotes does not make data, and as someone who has spent a great deal of time and effort in researching US, I am well aware of the limitations of evidence in the field. However, US is not a pill upon which you can do a double-blind, placebo-controlled trial. It is a user-dependent technology that is difficult to study without bias and has not had a home for well-funded trials. I agree that doing 25, 50, or any arbitrary number of scans does not necessarily ensure competency, and current efforts within the US community and among residency educators are working toward addressing this. However, as professionals, the reason we undergo training for so long is to both establish and learn the limitations of our skills, whether in diagnosis or in laceration repair. This is the essence of being a professional. In the situation described where an attending physician is not comfortable with a resident making a diagnosis using US, the attending should absolutely proceed with a confirmatory study or additional consultation where both the resident and the attending can learn whatever the final diagnosis may be.

I have also seen US “impede residents from achieving efficiency,” particularly when they are learning a new examination or using US to guide an unfamiliar procedure. Residents need to learn efficiency, but it should not be the only concern during training, and while bedside US can improve efficiency, it takes time and training to get to the point where this is the case—something I also emphasize to postresident physicians who are taking a course to learn bedside US.

Ultrasound is not a magic wand, and it is extremely important that attending physicians with experience temper the enthusiasm of well-meaning novices. I have seen residents who think that just because they have an US probe they are invincible from puncturing the carotid artery or causing a pneumothorax during an internal jugular line placement—not true. The possibility that one of my trainees will misdiagnose an interstitial pregnancy as an intrauterine pregnancy keeps me up at night and vigilant to make sure that our residents are well trained. While it can be straightforward to diagnose gallstones or a deep vein thrombosis, and expedite care when they are clearly present, it can be more difficult to rule these conditions out. In a well-structured training program (and beyond), there should be quality assurance methods to complement self-monitoring of performance with this user-dependent technology.

Point-of-care US is no more immune from liability than much of what we do as emergency physicians. Yes, excellent training, careful use, and meticulous documentation can help minimize this risk, but it does not go away. However, using US to appropriately guide a procedure, or catch a diagnosis when a formal test is difficult or unobtainable, may help decrease that liability.

In 1999, a letter to Academic Emergency Medicine written by two experienced program directors lamented “misguided residency applicant questions” about emergency US, speculating that this would be “a technology that is probably going to be even less important in the near future.”¹ As of the fall of 2010, there were over 60 emergency US fellowships, up from two when I did mine in 2001. This does not seem to be a passing fad. Regarding a later survey I conducted, it is worth mentioning that in 2004 more than half of community ED directors surveyed had obtained their equipment less than 2 years prior, and more than a third of those without equipment had plans to purchase an US machine in the immediate future.² I have no doubt that as of 2010 many have done so. In 2004, a vast majority of ED directors supported residency training in US while listing “lack of training” as the primary impediment to implementing ED US.

As attending physicians, we need to balance the restraint of our experience with the courage to adapt to evolving practice and technology—all that glitters may not be gold, but do not throw the baby out with the bathwater either.