To the Editors:

We thank colleagues Kane et al, Drs. Thiele and Evangelisto, and Dr. Lindsley for their valuable remarks on the role of musculoskeletal ultrasound in the rheumatologist's hand.

We would like to provide some additional comments to their statements and will be happy to clarify some obvious misunderstandings.

Kane et al agree with us that “superior working knowledge by physicians of an imaging modality may lead to greater use of this modality.” Furthermore, an article discussed in our editorial (Brown AK, O'Connor PJ, Roberts TE, Wakefield RJ, Karim Z, Emery P. Recommendations for musculoskeletal ultrasonography by rheumatologists: setting global standards for best practice by expert consensus. Arthritis Rheum 2005;53:83–92) showed significant differences in agreement scores for indications and anatomic areas between rheumatologists and radiologists concerning the use of musculoskeletal ultrasound, which might further support the above mentioned quote. With regard to the patient's needs, it is irrelevant if an adequately trained rheumatologist or a radiologist performs a procedure such as an ultrasound examination. In a setting where access is limited to a diagnostic center equipped with all possible imaging modalities, we agree that an ultrasound examination performed by the clinician may be preferable to no imaging at all to further guide the therapeutic decision process. We also strongly support the statement that a predominance of one imaging modality over another due to limitations in training or accessibility should be avoided.

However, in the light of rapidly evolving new developments in the field of cross-sectional imaging over the last 10 years such as multislice CT and MRI (including whole-body MRI, clinical high field systems of up to 3T, and in-house MRI with dedicated scanners) and the increasing possibilities of image processing, we would modify our statement to indicate that a radiologist or a trained clinician who is familiar with these novel techniques will be the most able to adequately integrate this knowledge into the diagnostic process. We did not state that one should learn CT and MRI before MSUS can be practiced, but we restate that a working knowledge of all imaging modalities is desirable for a more profound expertise. Limited access to clinical information in many situations is very unfortunate, and we believe that a clinical radiologist should have as much insight as possible into patient history, clinical findings, and serologic studies to improve interpretation of imaging findings.

MSUS certainly is the least expensive of all imaging modalities; however, it is also the most operator-dependent technique, which stresses the need for consensus among all groups performing MSUS concerning training and in the setting of indications. We acknowledge the importance of constant communication between radiologists and their clinical partners and we completely agree with Dr. Lindsley's statement that the benefits of musculoskeletal imaging in daily rheumatologic practice will improve treatment decisions and patient care. It has been through daily communication with clinical colleagues and through continued learning that radiologists have played an important role in developing this technique, and they should serve as consultants and teachers to help bring this knowledge to a broader rheumatologic audience.

Frank W. Roemer MD*, Marnix van Holsbeeck MD†, Harry K. Genant MD‡, * Klinikum Augsburg, Augsburg, Germany, † Wayne State Medical School, Henry Ford Hospital Detroit, MI, ‡ University of California at San Francisco.