Hi, and welcome to this beautiful city of San Francisco for SCCM 2023. I will be talking about how to set up a POCUS quality assurance process, what are the steps somebody needs to take at their own hospital to have a robust POCUS quality assurance process. My name is Siddharth Dugar. I am a staff physician in Cleveland Clinic and director of Point of Care Ultrasound. I have a chest sinusoid grant to study lower extremity venous DVT in COVID-19 patients. So we'll talk about POCUS. What is POCUS? POCUS is an exam that is performed by the bedside clinician who makes a decision to perform an ultrasound exam, who does the image acquisition, interpret the images, and integrate it into clinical practice. So the four roles or three to four roles that are usually required to perform consultative ultrasound or comprehensive ultrasound is done by one person here. It is usually done to answer a clinical question or to guide diagnosis. So the question comes to mind is, do we really need a quality assurance process for POCUS? Because as we just discussed, we are just trying to answer a clinical question or reach a diagnosis. Well, this is a study that was published in Annals of Emergency Medicine in 2020. This is the CMS Medicaid data for reimbursement looking at diagnostic point-of-care ultrasound performed and reimbursed from 2012 to 2016. Now one can see that there were close to 13,000 diagnostic ultrasound that were performed in 2012. It has increased by almost three times to 31,000 by 2016. Number of physicians who are performing point-of-care ultrasound or at least billing for it based on this data were 542. It has grown to over 1,350 physicians in 2016. And this data is only up to 2016. We know during COVID there was an exponential growth in utilization of POCUS to guide management. And if you are performing so much of POCUS, we have to be sure that the studies we are doing are of good quality and can answer the question that we are looking for. How I got involved in point-of-care ultrasound quality assurance process? Definitely, I was worried about the quality of POCUS examination being done and our interpretation as a director of point-of-care ultrasound at Cleveland Clinic. But then Dr. Gomez shared this article with me, which is put out by ECRI, which is Emergency Care Research Institute, a nonprofit organization looking at safety and quality and cost effectiveness in healthcare. And it noted point-of-care ultrasound to be one of the top two health hazard in 2020. And some of the points that they raised were very appropriate. It's growing very fast, which is good because we want a point-of-care ultrasound to be used for providing better care to the patient. But also the concern is there is lack of oversight on when and how we are using it. For example, indications. Somebody is using it for hypoxic respiratory failure, a very appropriate indication or shock. While POCUS may not be the best when we are looking at regional wall motion abnormality with mild troponin elevation with no shock or trying to see if the patient has left atrial appendage clot, again, those are more appropriate indication for comprehensive echocardiogram or TEE even. The training that all the POCUS users are getting is not well standardized. So the image that are being acquired and how they're interpreted it is very variable across the spectrum. And also there is over-reliance on POCUS finding. So the clinical integration of the POCUS image, which is just an extension of our physical finding or physical examination into the clinical picture is lacking. There is a lot more POCUS-driven decision making, which may or may not be true. Is quality assurance keeping pace with it? So these are the publications from 1970 that I searched on NCBI looking at POCUS. As you can see, from 2010, it just exploded, while the number of quality assurance studies that have been published are pretty low. So again, the quality assurance is not keeping up with the growth of point-of-care ultrasound. What is quality assurance process? So I've taken this definition and modified it from quality assurance process in other imaging modalities. So it's a systematic process to ensure that the information that we get by POCUS is timely, appropriate and accurate. Now when we are performing POCUS, each of these decisions to perform an ultrasound, to acquire the images, to interpret and clinically integrate lies with the primary clinician. So for the entire process to work, we have to make sure each of the process works. Indications now again, since I started doing point-of-care ultrasound, which was 2014 to now, the number of indications and the number of organs that we can assess with point-of-care ultrasound has only increased. So there is not a single organ that I can think of that cannot be assessed with ultrasound. The only organ that was like, everybody was worried that we will never be able to assess using ultrasound, which is lung, we are able to assess it and assess it very well. So again, indications are growing. So how do we make sure that the indications are appropriate for your program? So one thing that I have learned is don't start a point-of-care ultrasound program when you are looking at quality assurance, expecting that you will be able to assess all the organs and you will be able to include all the indications the first day you start the program. You have to know your patient population, what are the most common pathologies that you see, focus on those pathologies, be very good at them, know what are the indication for using focus on when do we get a comprehensive study. Also know the skill set of yourself and your team, and that will help you with deciding on what are the most important indication and organ system we should be starting with. Again, remember, change is the only constant, your indications will increase. The organ system you will be assessing using point-of-care ultrasound will increase, but don't start with the entire body. It is very difficult to maintain quality when every indication and organ is being scanned at the same time. Image acquisition, so the quality of the image that we get depends on so many factors. There are patients, there are patients with BMI, lungs are hyperinflated, you may have a patient with COPD, bones, external devices, patches on the body that will affect it. So we'll examine her position, knowing your machine, you know, patient is coding, you are thinking about too many things that will affect the image quality. Your competency will definitely affect image quality. Having 10 different machines in the same hospital will affect the image quality. Just knowing the machine, how do you optimize images per machine, using the probes appropriately, and again, if you're trying to get into a very crammy room, your image quality will not be the best. So when we are talking about image acquisition, these are the things that I think every place that does point-of-care ultrasound with the goal that we want to improve the image quality should focus on, position, position, position. I cannot tell you how many times getting in a very good position for yourself and your patient, just trying to get left lateral position if you can for cardiac imaging, trying to get them in left or right position so you can assess the plaques points will help you with getting much better image quality and assessing for all pathologies. Like one view is no view. Again, when we start, we always try to get just peristalne long axis interpreted and move on, but there are a lot of times it will lead to misinterpretation. So having a standardized approach where you put it out there that these are the images that we expect anybody who is doing point-of-care ultrasound in our hospital should be acquiring. Again, I am not trying to say that we should acquire all these images when the patient is in cardiac arrest, but if it is a semi-urgent or patients where we can acquire these images, I will suggest highly to acquire all these images so that, you know, everybody follows a standardized approach. So the quality of the images and misdiagnosis will lessen. Very important. You know, this is one thing I've learned is trying to standardize ultrasound throughout the hospital. You like pick one ultrasound company and stick with it just because it takes a lot of times to learn the proficiency with each ultrasound system. And we have seen that with other devices that we use in critical illness, be it a defibrillator monitor or glidoscope, the more system you have, the more confusion everybody has. Everybody has to learn how to use them appropriately. Using one system, even though if it is not the best quality, we'll still get best images because everybody will know how to optimize the images on that. Again, we struggled with lung ultrasound and DVT scanning when there was not enough annotation. One thing we learned is most of the systems can do annotation now. So just building it in the program so everybody has to annotate. So you studying me or I studying me, our interpretation will remain the same. We don't have to think like, is it right or left? Always compare your images when you're starting to the comprehensive studies. Again, their image quality will be great. All you're trying to do is match 60 to 70 percent of their image quality. So and that is one way to compare to see how they were able to do it. Collaborate with them. Go with the ultrasound tech. All our fellows at our institution has the option to go with an echo and an ultrasound tech so they can learn how to perform more comprehensive examination. And everybody has told me that that has really helped them elevate the quality of images they are getting. So this is a really nice study that was published by Dr. Motat in 2021. It was a web-based survey that was sent to all 81 pediatric emergency fellowship in United States and Canada, just asking them how many times do they have images archived? Do they document the performance of ultrasound and do they document interpretation? So performance is gray and interpretation documentation is black. And you can see interpretation documentation all the time. Only 20 to 30 percent of the program always document the performance of an ultrasound and interpretation. Most of the programs close to 60 percent do it sometimes while 10 to 20 percent don't do it at all. So we have to move this 60 percent where the performance of a point-of-care ultrasound is not documented. Interpretation is not documented to all the time. The reason being, if you perform an ultrasound and we cannot see the images, it never happened. So if you are performing an ultrasound where you are making clinical decision based on that, you have to have some way of saving the images. I have learned this that when I put the probe at bedside, my interpretation at a heart rate of 100 per minute for me and for my patient is different than when I go back and look at it when I'm like just sitting in my office and going through Epic. It really makes a difference and LV that may look abnormal there or RV that may look abnormal there. When you come back and look at it, you're like, maybe it's not that bad. So again, save your images, look at the images, make a decision, come back, look at the images again. You will have a different perspective. There is nothing as good a communicative tool as documentation. We document all our images that we acquire for POCUS. Also, when I find something clinically significant, I talk with the physician and just discuss the case and see like this is my interpretation of the POCUS and that helps a lot. Again, if you are doing a point of care ultrasound or getting a CT of the chest or abdomen and you have performed POCUS, just see what the findings were on that study. That has helped me immensely in learning what my findings were. Sometimes I was not sure when we look at the CT scan, we find this and then next time I will be more confident that this set of images correspond to this diagnosis. Again, collaboration is the best way to grow. And if you are saving images and you're billing for it, money, money is always good. It always helps a lot when you are asking for a new or better ultrasound. To me, POCUS is just an extension of my physical examination. All I'm doing is using sound waves to assess physiology. Hence, you have to know your physiology for you to be able to apply POCUS nicely to your critical ill patient. Know the loading conditions. It is one thing to say the RV is dilated and another thing to say RV is dilated but the patient is on 30 of PEEP. The patient has gotten 20 liters of fluid because then the interpretation is different. Know the limitation and nuances. I was just having this discussion where we put a probe on a patient who has been in the ICU for 10 days and we saw A-line pattern on the abdomen and the concern was, is there a perforated viscous? And again, that data is true for trauma patient. It may not be true in critical ill patients who have been in the ICU just because gastric distention of bowels may cause similar patterns. So examining the belly, seeing if it is showing signs of peritonitis and then integrating them together will be more fruitful just rather than just following the images. Very important to differentiate acute from chronic. Again, as our patient population ages, we'll be seeing a lot of chronic disease, lung, cardiac disease and reacting to it and not able to differentiate acute on chronic will lead to more bad decision making. Remember the natural history of critical illness is very dynamic. I remember a case where there was a patient with acute corporeal malignancy who was started on heparin and then a fellow called me that, hey, the patient RV is normal and the patient has IVC variability. I'm like, no way. This is just six hours. We just have started heparin. We did not even give TPA. And I went and I assessed the patient because I just did not trust the fellow. And the RV was small and the IVC was collapsible. So remember, it's very dynamic. Things change. So if you're making medical decision based on an image that was 24 hours old, you are making decision based on clinical condition that existed 24 hours ago. And it's just a one data point among money, assess the patient, use every information that is available and then make decisions. So the best way to do it, earn while you learn, do the ultrasound, read it with experts, read it independently and then discuss it with the expert. Start generating reports, but also have an expert give you feedback either during the same time or maybe later on. Again, the documentation will help you a lot because the only way I know what your thing is, is when you're documenting it or communicating with me directly. And again, collaboration is the best way to grow. I learned POCUS by pestering my mentors with a lot of questions. That will help you a lot. It will give you an insight of how as you grow, your interpretation of an image changes. And again, it's a process. You are at a current state. Remember, there are people that are processing technology. This is where you want to go, but it's always going to be a continuous improvement. The perfect system does not exist. There is only a road to perfect system. And that's what we are trying to do. Another thing that we did at our institution is to learn from radiology and cardiology where we did everything that we just discussed, but we also added a secondary QA or a meta-QA where we looked at what was the indication of the studies that were performed? How many times did they adhere to the protocol that we have put forward? Did they use documentation? Did they document everything that we asked them to document? And how quickly was the fellow able to document their finding? And how quickly was the staff able to attend the note? We also took 10% of our studies and we took the patient information away from those images, send it to two expert reviewers trying to see what their interpretation of that study is and try to compare it to fellow's interpretation and the staff interpretation. And we learned a lot using this process. What are the things that all of us are doing great? What are the things that some of us are doing great and some are not? What are the things that all of us are struggling with? For example, we found that differentiating mild moderate and severe RV dilatation was very challenging. So we took it away and we just now documented it's RV dilated or not dilated. So again, this helped us a lot to improve, reduce variability between all the POCUS reviewers, but also see what are the things that we are doing good and what are the things that we can work on. These are some of the important study that I come across for anybody who is interested in starting a POCUS QA process. Again, this is a POCUS quality checklist, a POCUS quality scale that you can use to see how good are your images. This is a really nice article that describe how a system went from no archiving to 100% archiving. They like, you know, again, they used a system that helps with image archiving. They did a lot of quality assurance process throughout the year to make sure that the number of studies that are archived increased every time they did an intervention. The important point is we are archiving everything, but we still have to go back and remind them to always put the order in to make sure the images are saved. And this is a study by Dr. Fox, who is not here, who was a moderator. He had a baby, so congratulations. But he wrote a paper where they were able to provide QA to POCUS studies remotely. So again, the field is growing, field is expanding, but I think the future of POCUS is going to be in making sure that the images are of optimal quality and our interpretation and integration is great. So thank you so much and everybody have a great day.

quality assurance process

Point-of-Care Ultrasound

appropriate indications

image acquisition quality

collaborating with experts

clinical decision-making