Some thoughts on endoscopic ultrasound training

Abstract

See article in J. Gastroenterol. Hepatol. 2006; 21: 1231–1235. The paper by Dr Ho reassures us that endoscopic ultrasound (EUS) use should continue to grow in East Asia, but that access to adequate training remains a problem. I agree completely. EUS is clearly here to stay, but unless we can train more endosonographers (more specifically, endosonographers who can train other endosonographers), the situation will be slow to improve. For some reason, certain techniques gain acceptance as useful clinical tools before they actually prove that they provide useful information. Remember, just because a test is accurate does not make it is useful. I am biased, but I believe magnetic resonance cholangiopancreatography (MRCP) is an example. How many MRCP impact studies or cost-effectiveness studies were required before authors began jumping on the MRCP bandwagon? None. Early on, numerous studies proved that EUS is accurate for various indications (e.g. diagnosing bile duct stones, staging lumenal cancers). The EUS literature has also been inundated with EUS ‘impact’ studies and cost-effectiveness studies. Apparently, all this has been insufficient to justify mobilization of the resources needed to make EUS available universally. Why has it been so hard for EUS to gain acceptance? In my opinion, it all boils down to inadequate access to EUS training. Let’s assume that, in a defined geographic area, at least one center decides to purchase EUS equipment and manages to find an individual willing to spend the time needed to learn how to use it to its full capabilities (i.e. for both diagnostic and interventional EUS). Adequate training should lead to expertise; expertise should result in examinations that provide useful and reliable information; this should lead to increased acceptance and demand (from physicians and patients); and this should lead to increased priority for funding required to buy equipment in other centers in the same or surrounding geographic area. An informal survey of a few expert endosonographers would appear to confirm this. When they arrived home after their EUS training and the referring physicians in their region obtained adequate access to quality EUS, they soon could not live without it and others then began to show interest in acquiring the technology. Let’s be clear, the key terms are ‘access’ and ‘quality’. Physicians who have never had access to EUS can easily continue to live without it. It is human nature to resist change. For physicians, this is especially true if it requires them to modify their tried and true management algorithms and/or to accept that they have been missing or misdiagnosing certain diseases for years. After all, if you want to ‘win friends and influence people’, one of the cardinal rules is to ‘never tell a man he is wrong’. For surgeons, a test that shows a patient has advanced (i.e. inoperable) disease may be just as unwelcome for him or her as it is for the patient. Therefore, for a new EUS service to successfully forge a place for itself and continue to grow, it must be efficient and effective. It is easier to get people ‘hooked’ on EUS if they can get them done fast, well and when there are multiple indications available to them (i.e. imaging, EUS fine-needle aspiration [FNA] and EUS therapy [e.g. cyst drainage, celiac plexus neurolysis]). Obviously, several variables can influence the efficiency and effectiveness of an endoscopic service, but I believe that the foundation is a well trained operator. And this is the problem. They are hard to come by. However, this is not because EUS is more difficult to learn than other procedures (e.g. endoscopic retrograde cholangiopancreatography [ERCP]). In my opinion, this is a myth that must be dispelled. It is simply because EUS is more difficult to learn alone. EUS is obviously different to other procedures. The ultrasound images are foreign to many but one must also learn to do the opposite of what we do normally with regular endoscopy. That is, remove all the air, press up against the intestinal wall and use ultrasound landmarks instead of endoscopic landmarks. The movements are finer and slower and acquiring the hand-eye coordination required to trace out structures can be challenging because the traditional up/down orientation does not always hold. It is therefore much easier to learn if, initially at least, a trainer is present to help orient and guide the trainee. Thus, increased access to trainers is crucial. However, like other procedures, there are hard cases and easy cases. Interestingly, some of the easiest EUS indications provide the most powerful information (e.g. EUS-FNA of large subcarinal lymph nodes in non-small cell lung cancer is very easy and, if positive, immediately makes the patient inoperable). Therefore, it may be reasonable to expect some endosonographers to focus on learning basic EUS while others with more advanced training concentrate on more difficult cases (just as some ERCPs are reserved for referral centers). I believe that with adequate supervision and by using a ‘station approach’ a novice can, in a relatively short time, attain a level where they can begin to progress on their own and successfully perform basic EUS (including some more straightforward EUS-FNA procedures). This approach systematizes the examination and helps identify core structures that need to be evaluated (e.g. pancreatic body, pancreatic head, uncinate process, celiac axis, subcarinal space). If the landmarks associated with a particular station are identified (or not identified), the operator can then be more confident as to what they have (or have not) seen. As an example, one cannot say they have excluded a pancreatic body mass if they cannot be sure that they have seen the pancreatic body! Using this approach, I believe that basic EUS can be learned in a reasonably short time if there is adequate access to hands on training in patients. However, there is no reason to believe that advanced EUS takes any less time to learn than any other advanced procedure. Therefore, for full proficiency in all aspects of diagnostic and interventional EUS (particularly pancreatico-biliary EUS), a dedicated 1–2 year EUS fellowship is essential. I say this because there seems to be a focus on trying to learn EUS by many ‘artificial’ means (e.g. simulators, phantoms, videos, animal models) instead of by simply performing an adequate number of supervised procedures in real patients. This focus on substitutes for true hands on training is unfortunate in my opinion. It does not make sense to me that fully trained gastroenterologists cannot get licensing for supervised short-term hands on training in EUS. The medico-legal risks involved in such training are surely no greater than those involved in teaching endoscopy to a first-year gastroenterology fellow. I am convinced that quality EUS will stimulate a demand for more EUS. We clearly need to train more trainers. But, to accelerate the current growth of EUS, we must find ways to provide supervised, hands on training in basic EUS to physicians outside of a full year EUS fellowship.