Editor's note: Dr. Hoppmann is the Dorothea H. Krebs Endowed Chair of Ultrasound Education, Professor of Medicine, and Director of the Ultrasound Institute of the University of South Carolina School of Medicine. He founded and served as the first President of the Society of Ultrasound in Medical Education. We spoke with him about safety and usability of point-of-care ultrasound.
Dr. Robert M. Wachter: Tell us about your background as it relates to ultrasound. What got you interested in this in the first place?
Dr. Richard Hoppmann: When I was the Associate Dean of Medical Education, two emergency physicians approached me and asked if I thought there was any place in the curriculum for ultrasound, given the technical advances that had been made and applications people were starting to use. We put together a program, presented it to the Dean, got things started with the support of GE Healthcare, and launched the curriculum in 2006. We decided from the very beginning that we wanted to have it integrated across all 4 years. We are now expanding to the residency programs. We offer CME [continuing medical education] programs on primary care ultrasound. We also have a PA [physician assistant] ultrasound training program.
RW: Did you begin thinking about it more in the procedural realm and then move to physical diagnosis?
RH: Initially, we were thinking in terms of what would be the best application for medical students and medical education—in anatomy, physiology, pathophysiology—and then moved into the clinical clerkships. As we moved into the clerkships, we worked with the simulation center and started teaching guided procedures, and now that has become standard. We introduce students to guided procedures at the end of the second year, before they start the clinical clerkships.
RW: Tell us about the early political and turf challenges that relate invariably to the beginning of the diffusion of this technology that initially lived in a department called radiology.
RH: Well, it was interesting because we also made the decision very early on that we would not have a department own the ultrasound curriculum. It is independent from any particular department. We also made it clear from the beginning that this was going to be about education. One thing we needed to do in the very beginning was to solicit all available ultrasound expertise. We felt strongly that this should be a multidisciplinary effort. Radiology was engaged—cardiology, emergency medicine, OB/GYN, and the basic science departments as well. We did not encounter too much resistance. We also got both adult and pediatric cardiology engaged from the beginning. Of course, some faculty had reservations. Over time, though, when they realized what we were doing and we were not trying to turn everybody into radiologists and cardiologists—they were very much on board with making ultrasound an education tool. In the next phase, we started to move beyond education to look at things like improving patient care.
RW: What has the literature shown about the advantages of using ultrasound in care and of training people in it?
RH: We started the Society of Ultrasound in Medical Education in 2008 and that was a multidisciplinary process. The key was to build the Society over time. We needed to reach a critical mass of supporters of ultrasound in education for a number of reasons. One was to teach, but we also needed to make it evidence-based. We have reached that point now as reflected by the increase in the ultrasound education literature. Obviously, one of the early applications was guided procedures. Even back in the early 2000s, it was recognized that with guided procedures, especially internal jugular line placement, you could decrease complications and the number of sticks. Many societies now recognize it as a standard of care. As more folks were trained in ultrasound, more studies were conducted, and it has become clear from the data that ultrasound can enhance patient safety and quality of care.
Lung ultrasound is becoming a standard of care, especially in the areas of critical care and emergency medicine. The same thing is true in pediatrics. From 2006 to 2016, there has been a tremendous increase in the number of publications related to ultrasound education and point-of-care ultrasound. Take for example pediatric safety and quality around appendicitis—if a patient comes in with physical exam, history, and blood work consistent with appendicitis and ultrasound is positive for appendicitis, then a CT is not needed. These patients can go to the OR. The same thing is true in patients that come in with renal stones with a history consistent with nephrolithiasis. You can always fall back on CT if necessary. These are just two examples of decreasing ionizing radiation exposure, which is especially important in children. Data is accumulating across the board for using ultrasound to decrease radiation exposure in many clinical scenarios.
RW: For areas where somebody is making a judgment about a kidney stone or about appendicitis, how accurate are nonradiologists in their reading when you compare them to radiologists?
RH: We have good data on a number of well-documented examples. One would be for renal stones. In the New England Journal of Medicine a few years ago, a comparison was made of radiologists performing ultrasound and emergency physicians performing point-of-care ultrasound, and they also had a CT imaging group. Overall, results were nonsignificant for differences in the two groups of readers with respect to high-risk diagnoses with complications, serious adverse events, return emergency department visits, or hospitalizations. This study also documented that with ultrasound, patients had decreased exposure to ionizing radiation. Similar outcomes have been reported in other emergency medicine studies as well.
One reason why we emphasize education so much is that ultrasound is operator dependent. Practitioners must be adequately trained, reach a competency level, and know their limitations. Building this approach into all of medical education has been our focus as we see ultrasound as an education continuum, and we feel medical school is the best place to get started. Then you can easily expand into residency and fellowships, etc. It is important to keep in mind that these are focused (not comprehensive) ultrasound examinations we are teaching. They are designed to answer a specific bedside question the practitioner has to help diagnose or manage the care of the patient.
RW: You probably now have people starting your residencies, if they came from your medical school they're already pretty competent, and you have other people coming in who maybe are not very far along in their expertise, how do you deal with this transitional period?
RH: It is truly a transitional period and we are trying to work our way through this transition. We do a number of things to help newcomers get up to speed such as an ultrasound boot camp. We assess where new residents are in their ultrasound baseline and move forward from there.
We have also started an ultrasound fellowship. We can look at the history of how ultrasound evolved in emergency medicine. Back in the early 2000s, emergency medicine physicians started one-year emergency medicine ultrasound fellowships. They had a handful of these programs, maybe 8 or 10. Now they are up to almost 100. An emergency medicine resident graduate would complete the ultrasound fellowship, then go out and train emergency medicine residents in ultrasound and possibly start a new fellowship program as well. So over time, more emergency medicine residents were being trained in ultrasound and as the fellowships expanded, there were more qualified faculty to teach and conduct ultrasound research. This turned out to be a great ultrasound training model.
About 7 years ago, we started a primary care ultrasound fellowship. We take graduates from internal medicine, family medicine, and pediatrics residencies for the fellowship. We started the fellowship because we noted a limiting factor across the country for ultrasound education was having an adequate number of faculty to teach ultrasound.
We're trying to make ultrasound a standard in medical student education. Let me give you an example of how that might be beneficial. There are many ultrasound applications in urology and I had a urology residency director tell me that he tried to introduce hands-on ultrasound into his residency without much success. His residents said that they had so much patient care and so many surgeries to learn that they simply did not have time to learn ultrasound. The director told me that if medical schools send him students with a good foundation in ultrasound, then he could teach them the residency and specialty-specific ultrasound applications without much difficulty. That is what we are trying to do.
Now the American Academy of Family Physicians is moving forward with guidelines for training residents in ultrasound. The American College of Physicians and others are now starting to see the value of ultrasound training and are exploring ways to increase ultrasound training in residency. If medical students and residents progress through medical school and residency with milestones and competencies in ultrasound, they will be very good at point-of-care ultrasound.
RW: When you talk about training primary care doctors, what are the most common uses in an outpatient general practice?
RH: That is a good question. We have completed ultrasound needs assessments for primary care providers. We look at what they want to learn and what scans can be performed relatively quickly and learned well at an outpatient general practice level. We had a Duke Endowment Grant 7 or 8 years ago and trained 12 primary care rural South Carolina practices. We started with the needs assessment. The primary care providers wanted to be able to manage congestive heart failure, look at right upper quadrant pain for gallstones, and look for DVTs [deep vein thrombosis]. If a patient comes in with a swollen right leg on a Friday afternoon and a vascular ultrasound lab is not readily available—what do they do? Now they can perform the exam themselves. They also wanted to learn guided procedures and AAA [abdominal aortic aneurysm] screening. There was also a lot of interest in musculoskeletal ultrasound, because primary care doctors see a lot of muscle and joint problems in their practices. Therefore, we have included basic musculoskeletal ultrasound training as well. We look at what they can learn, what they can learn well, but also want them to appreciate what the limitations might be.
RW: Let's say a primary care doctor does an ultrasound for a DVT in the office, how does the billing work?
RH: It is still somewhat variable. Reimbursement is fairly straightforward for some scans like guided procedures or an AAA screen, when the published indications and guidelines are clearly met. We have found that different physicians want to use ultrasound in their practices in different ways. When I did follow-up with physicians who went through our program a number of years ago, some were performing a limited number of ultrasound exams and billing for them. Others were using ultrasound daily to complement their physical exam and patient assessment but were not billing for ultrasound. Then there are others who live in rural settings who have made arrangements with hospitals and insurance carriers to perform ultrasound in their offices, and they bill for many procedures in which they have been trained. For all who plan to bill, we emphasize the importance of following established billing guidelines. Probably the best website for billing is the American College of Emergency Physicians. The site explains the various components of the process such as selection of appropriate billing codes, the need to save images, and other requirements. Billing for ultrasound can vary by state, insurance carriers, etc. Again we are in a transition period, and we hope that many of the billing issues will be resolved in the near future. Personally, I would like to see the use of ultrasound and reimbursement tied to quality of care and cost savings.
RW: What are the largest safety concerns you have about bedside ultrasound?
RH: It is worthwhile commenting about the safety of ultrasound itself. The AIUM, American Institute of Ultrasound in Medicine, has done a great job over the years with evaluating the safety of diagnostic ultrasound with periodic consensus conferences of experts and publishing summaries of the evidence and recommendations. Diagnostic ultrasound is exceptionally safe, but because of potential bioeffects it should always be performed as recommended with equipment that is approved for the study to be performed and the principle of as low as reasonably achievable (ALARA) should be followed. This means that the lowest physical parameters of ultrasound and the least exposure time that allows a clinically useful image to be obtained should be used whenever possible. This will help ensure the thermal and mechanical bioeffects of ultrasound are minimized. To my knowledge, there have not been any recorded cases of patients injured from ultrasound. As the use of ultrasound spreads and new equipment and procedures are introduced, it will be important to continue to monitor ultrasound use for patient safety. There are also other categories of potential toxicity related to ultrasound such as the use of contrast agents that warrant monitoring as well.
From a safety standpoint, ultrasound itself is safe. On the user/operator side, ultrasound is safe and of great value in the hands of well-trained competent practitioners of many different specialties. There will always be concern of errors of overdiagnosing and misdiagnosing conditions with ultrasound in all users, but especially in those new to ultrasound or those not appropriately trained. This is why the full spectrum of education, competency assessment, and ongoing quality control needs to be implemented for ultrasound as it has for many areas of patient safety and quality of care. I see this beginning to happen in medical schools, residencies, medical and professional organizations, and hospital systems. I am optimistic it will continue to evolve into an area of medical practice that is appropriately taught and regulated in the best interest of learners and patients.
With all the interest in primary care ultrasound as well as point-of-care ultrasound, a sizeable market of ultrasound users had emerged that is driving manufacturers to invest in research and development. So today, we now have laptop ultrasound devices, handheld devices, and cloud-based devices. In addition, artificial intelligence (AI) is being added to all systems, including the handheld devices. For example, if you are scanning the heart and capturing an apical four-chamber view, the ultrasound device will compute the ejection fraction for you, but only if you have an acceptable quality image of the heart to ensure an accurate measurement of the ejection fraction.
RW: I was interested in the AI piece. Do you see us getting to a day where the individual operator competency is less important, because it can tell you if there are gallstones or kidney stones or if there's appendicitis?
RH: I think we are going to see tremendous advances in that area with the addition of artificial intelligence and deep learning. I just finished your book, The Digital Doctor, and I loved it.
RW: Thank you.
RH: My feeling is that artificial intelligence is going to play a huge role in all of medicine, including medical education. We are working with some manufacturers to encourage them to do two things. Obviously, one is to put AI into ultrasound systems to help with diagnosis, but also to have an on–off switch so you will gather the data yourself, make the interpretation, then activate artificial intelligence, and match your image, your interpretation, your diagnosis with that from AI. AI is going to be a tool we will use, and we should apply the technology to both education and clinical practice. However, there will always be the need for physician input, even with AI, because there are so many subtleties in medicine. I do think artificial intelligence is going to play a tremendous role in diagnosing and improving patient care in the future.
RW: One error that we saw here a few months ago was after an emergency department ultrasound of the heart. It made a call on the valve and said either no endocarditis or thrombosis, and that turned out to be wrong. In the discussion, it became clear that the ER doctor went beyond what someone doing point-of-care ultrasound should opine on. Do you see that? Where people are in some ways going beyond what either they or the technology are capable of doing? People read the result and assume it's the same as if the echocardiographer said that?
RH: We have to monitor users to make sure they understand the limitations of point-of-care ultrasound and their own limitations. I have even noticed with my medical students, they get overconfident and go beyond what they have been taught. In all of medicine, we need to know our limitations and what our competency and comfort levels are. This is where ultrasound needs to be multidisciplinary. This is where we work with the specialists and subspecialists and ask what is appropriate for medical students? What is appropriate for residents? What is going to take more extensive training? We must rely on quality assurance, competencies, and milestones in the practice of ultrasound. We know some practitioners are going to extend their practice beyond their knowledge and skill levels.
RW: Talk a little bit about the medicolegal aspects. If I do an ultrasound of the right upper quadrant, does that end up in the medical record? Should it? And what happens if it's wrong?
RH: Again, we are in a transitional period and practice varies. Emergency medicine has been performing point-of-care ultrasound for more than 20 years, and they have looked at the legal issues and lawsuits having to do with ultrasound being used by nonradiologists and nonsonographers. There have been hardly any lawsuits from emergency medicine practitioners performing point-of-care ultrasound. Some lawsuits have come from not using ultrasound when it should have been used, and that may end up being the biggest issue in the future
Guided procedures is a great example of that. It makes me nervous to see practitioners do procedures without ultrasound guidance. I do not think there's any place right now for blind sticks whether it be thoracentesis, paracentesis, or central lines. Therefore, in this transition period we need to establish definite ultrasound guidelines and determine a reasonable timeline for implementation. We do get questions quite often from the physicians we train. They get anxious about potential lawsuits and ask, "If I look at the right upper quadrant and I don't see a gallstone but I miss a metastatic lesion to the liver, am I going to be liable for that?" You have to be practical and look at the overall situation and provide guidance. Because learners are facing these issues, and as a multidisciplinary group we need to look for reasonable solutions in the best interest of practitioners trying to provide better care and also in the best interest of their patients.
We encourage practitioners learning ultrasound to do scans they feel comfortable and confident performing, and that is what they should use in their practice and document in their records. Ongoing practice and additional training is important because ultrasound skills can decay if not used.
As a valuable training exercise, I encourage new learners to compare their results with official ultrasound studies. Let us say they are learning how to do ultrasound for AAA. We tell them if a formal AAA screen is indicated in one of their patients, they should perform the scan in their office first and then compare their results to the formal radiology study. Over time, they get a sense of their skill level. We also want them to appreciate when they might be over their heads with a difficult-to-scan or inadequately prepped patient. For example, if they cannot get a good image, they should not try to interpret the image. In the long run, there's no question in my mind that appropriately training and assessing practitioners on the front lines of patient care in point-of-care ultrasound will be a huge benefit for both patient safety and quality of care.
RW: When you say they store the records, are people generally setting up systems where the actual ultrasound itself ends up in the electronic medical record?
RH: Yes. A good many are doing that. Putting images in the electronic health record has become easier. One of the beauties of ultrasound is the ability to compare images over time. Saving images also creates an opportunity for ongoing education. We schedule remote multistation image reviews and provide feedback to learners. Learners will send in images, and we will review them together. If they are billing for the ultrasound study, then they are required to save images.
RW: Will we be carrying stethoscopes in 5 or 10 years?
RH: Good question. It is really interesting, because we have been talking for the last decade or so about point-of-care and the pocket ultrasound device being the stethoscope of the 21st century. Because of the recent technological advances and the dramatic drop in cost for hand-held ultrasound devices—a couple of pocket devices are on or coming on the market that are going to be under $2000—we may finally be looking at the stethoscope of the 21st century. There are high-frequency to low-frequency single probes, so they can scan pretty much everything. There are also web-based devices for easy cloud storage, remote consultation, and easy upgrades to software.
My bias is I like heart sounds as well as ultrasound images. Unfortunately, studies have shown that physicians are not very good at auscultation and we may be getting worse at it. I think this is partly because we do not emphasize it enough and we don't have many of those master diagnosticians around teaching any more. However, I am a big believer in the value of heart sounds in teaching. We have recorded simultaneous heart sounds and ultrasounds to teach medical students auscultation. Students can look at ultrasound loops of aortic insufficiency, mitral regurgitation, and other heart pathology and hear the sounds of valves closing and murmurs at the same time. It is a great teaching tool. I know some work is underway on portable devices to pick up body sounds and ultrasound as well. I would like to keep the heart sounds, but in general, ultrasound is probably going to be the core feature of the stethoscope of the 21st century.
RW: The transition zone is partly because you have this death spiral where people no longer learn very much about heart sounds and they learn to only trust the echo. Now, as they're beginning to carry around the equipment, the emphasis on training goes down even further. I imagine some of the time that we used to spend training for heart sounds now gets replaced with this, so you hit that tipping point. How far away do you think we are before we hit that?
RH: This is a difficult question. Probably 10 years ago, I thought that ultrasound would be standard in medical education in just a few years. We are getting there, but it has been a slow process. Fortunately, the pace has accelerated rapidly in the last few years. Because the user market has been stimulating R&D and devices are getting better, cheaper, and more user friendly, I now think that within 5 years pocket ultrasound will be almost as common as today's stethoscopes.
RW: Whenever a technology replaces something people used to do with their hands, palpating and all that, there's always this lamentation of "This is the end of medicine and we're turning doctors into technicians." Of course, that was said when the stethoscope came out as well. What is your feeling about that?
RH: Well, one of the greatest advantages of ultrasound is that it is actually bringing us back to the bedside. Over the past few decades, we have gradually moved away from the bedside and bedside rounds. We now do something called "Gel Rounds," where one of the faculty will meet with three or four students, and round at the bedside with a portable ultrasound machine. The group will perform a history, a physical, and an ultrasound. Then they talk about the role ultrasound might play in the diagnosis and management of this particular disease and review the image with the patient. Ultrasound is a fabulous teaching tool to explain to patients why, for example, they have to take three medications for heart failure or what you mean by an aortic aneurysm that is 3.5 centimeters? We will likely be performing less of the traditional physical exam in the future, but there will always be the need to be with the patient at the bedside. Ultrasound is great for getting us back to the bedside and physically examining the patient.
One of the things we do in our physical diagnosis course is use the immediate feedback from ultrasound to help students learn the traditional physical exam. Let's say you are teaching a student to palpate for the liver edge. We have students work in pairs. One will be using the ultrasound and capture an image just below the liver edge. They both are looking at the screen. They will ask the patient to take a deep breath, down comes the liver, and hits the examiner's fingers. "Ah, I felt it." Ultrasound can be used to improve palpation of the liver, percussion of the lung, listening to the heart sounds, and other traditional physical exam components. We are trying to keep ultrasound closely tied to both the physical examination and to the bedside interaction with the patient.
RW: That's a wonderful answer and quite encouraging to think that it will actually bring us back to the bedside. Because, if you think about it, we used to send the patient away to radiology (or to the echo department) to do all of this stuff. Now, at least in theory, we can do a lot of it ourselves.