To the Editors:

A recent editorial by 3 eminent musculoskeletal radiologists on the use of musculoskeletal ultrasound (MSUS) in rheumatology correctly highlights the important role that MSUS can play in improving the diagnosis of synovitis, enthesitis, and bony erosions and in guiding local therapy to the benefit of rheumatology patients (1). The editorial also concludes that rheumatologists and radiologists should engage in the development of MSUS by “cooperation and constant communication between the specialties.” We are in full support of these concepts as we enjoy working with our musculoskeletal radiology colleagues, both in our clinical rheumatologic practice, and also in the implementation of national and European MSUS training and research projects. However, this editorial makes a number of strident statements that are unfortunately not correct and that are not helpful in achieving good relations and collaboration between radiologists and rheumatologists in the development of MSUS.

Despite adopting a moderate tone towards the end of the editorial, the authors argue firmly against the practice of MSUS by rheumatologists. One of the principal arguments against training rheumatologists in MSUS is on the grounds that this will lead to inappropriate or self referral of patients for MSUS. They cite the example from other specialties where the impact of the availability of a number of radiographic procedures in primary care led to a higher use of all imaging modalities when compared with physicians who referred to radiologists (2). There are numerous other explanations for this, which include higher patient acceptability for immediate on-site scanning and the fact that a superior working knowledge by physicians of an imaging modality may lead to a greater use of this modality. Indeed the authors of the study quoted eventually concluded that “it is not possible to determine which group of physicians uses imaging more appropriately” (2). Roemer et al then go on to argue that the radiologist is uniquely placed as a “gatekeeper” to “guard patients against greedy self referral” by rheumatologists and they cite the presence of radiologists as providing a “heavier focus on patient care.” This is a very unfortunate and extremely inappropriate misconception that does not respect the decency or professionalism of clinical rheumatologists in how they manage their patients, nor does it reflect the reality of working practices between rheumatologists and radiologists.

In all of our rheumatologic practices, we are trained to routinely select from a wide number of imaging modalities including plain radiography, MSUS, computed tomography (CT), magnetic resonance imaging (MRI), and nuclear medicine to achieve better assessment and diagnosis in our patients. We have not noted that our own radiology colleagues feel it necessary to act as gatekeepers in restricting these activities, and in the minority of cases when we are unsure of the optimal imaging option to perform, we routinely seek the advice of our radiology colleagues. Indeed in the financial model in which they seek to stake out the moral high ground for themselves, there is equal incentive for the radiologist to proceed with inappropriately referred imaging for financial reward. Rheumatologists could equally perform unnecessary procedures (such as joint injection, and the prescription and supervision of unnecessary long-term medications such as anti–tumor necrosis therapy) solely for financial reward, something of which there is absolutely no evidence. We believe that the professional training of both radiologists and rheumatologists renders them equally capable of acting responsibly in the patient's best interests and regret that our radiology colleagues would express an alternative point of view, and worse, that it should be published unchallenged in a leading rheumatology journal. The authors go on to state that radiologists “will choose the most cost-efficient modality to query a rheumatologic disease.” This statement ignores the predominant use of MRI over MSUS for musculoskeletal imaging in the US, in contrast to the wider use of more cost-effective MSUS in musculoskeletal disease in Europe where clinicians have been involved in performing MSUS for over a decade.

We firmly believe that both rheumatologists and radiologists will continue to perform MSUS in the future, and that each speciality has specific advantages over the other in training in MSUS. The advantages of a rheumatologist performing MSUS are not discussed in the editorial, although the majority of references at the end of the article refer to innovative ultrasound research performed by rheumatologists, sometimes in collaboration with their radiology colleagues. Most rheumatologists who use MSUS perform it at the end of the history and examination. This places them in a uniquely time-efficient setting where a diagnosis can be immediately confirmed or refuted, and the patient may receive information and therapy. If the rheumatologist needs to proceed with a joint aspiration or injection, MSUS again provides an immediate aid to improve the accuracy of this, a fact confirmed by many studies (3–5). Alternatively the patient may wait for a radiology appointment and return for a further rheumatologic visit to relay the results. In our practices this can entail a wait of up to 12 weeks due to the demand placed on our radiology colleagues. The wider availability of MSUS among rheumatologists does mean that more of their patients will undergo MSUS as part of their care, with early studies confirming that that this results in significant improvements in diagnosis, therapy, and outcome (3). This is a reason to encourage training of MSUS among rheumatologists, not to discourage it.

The authors then place a final hurdle in front of clinicians who wish to train in ultrasound. They state that physicians should train in all imaging modalities including CT and MRI before they can practice MSUS. This is not borne out by the experience of Dr. Roemer's colleagues in Germany, where MSUS is successfully taught as part of the rheumatology curriculum. Neither is it required by the American Institute of Ultrasound in Medicine, or the Guidelines on the Training of Physicians and Surgeons in Ultrasound recently published by the Royal College of Radiology in the UK (6). Cardiologists and obstetricians have also dealt with this issue, and both specialties have ultimately evolved their ultrasound practices independently of radiologists (7). Rheumatologists would prefer to continue developing MSUS to the benefit of our patients in collaboration with radiologists, but collaboration requires that we consider and respect each other's specialties.


Dr. Kane is an arthritis research campaign Clinical Senior Lecturer in Rheumatology.

David Kane PhD, MRCPI*, George Bruyn MD, PhD†, Erin Arnold MD, FACR‡, Walter Grassi MD?, * University of Newcastle-upon-Tyne, Newcastle-upon-Tyne, UK, † Medisch Centrum Leeuwarden, Leeuwarden, The Netherlands, ‡ Illinois Bone and Joint Institute, Morton Grove, Illinois, ? Cattedra di Reumatologia Osepdale A. Murri, Jesi Ancona, Italy.