Letter to the Editor: “Imaging of sarcopenia: old evidence and new insights”
by Maria L Brun-Vergara, David F Torres-Cortés (firstname.lastname@example.org)Imaging of sarcopenia: old evidence and new insights
To the editor:
We read with great interest the review article by Albano et al.  published in the April 2020 issue of European Radiology. The authors revised current concepts in imaging of sarcopenia, mainly focusing in measurements methods using different imaging modalities. They also addressed topics, such us, definition, clinical considerations and some reporting recommendations. As well said, sarcopenia is currently a hot research topic and merits the writing of literature review articles. Thus, we would like to congratulate on such an excellent work.
However, to contribute to the information of the article by Albano et al. when addressing clinical considerations, we would like to introduce the use of the bioelectrical impedance phase angle (IPA) in the clinical assessment of sarcopenia. IPA has been gaining attention in recent years and is considered the most clinically established impedance parameter nowadays . It mainly represents changes in quantity and quality of soft tissue and normally ranges between 5° and 7° . It has also demonstrated to be meaningful in patient prognosis.
Additionally, in “How to measure sarcopenia” section, authors specified cutoff values for dual-energy X-ray absorptiometry (DXA) and for computed tomography (CT). Nevertheless, authors did not mention ultrasound (US) cutoff values to consider when evaluating sarcopenia. There is still no consensus in US thresholds to be used to establish diagnosis of sarcopenia. However, some recent study results, has shown that a cutoff value for US measurement of thigh muscle thickness for assessment of sarcopenia, should be < 36 mm for males and < 34 mm for females (sensitivity 72% and specificity 74%, and sensitivity 72% and specificity 72%, respectively) .
Finally, yet importantly, authors suggest that non-contrast abdominal or thigh CT measurements are reliable to evaluate sarcopenia. However, authors consider CT muscle segmentation is not feasible to perform. In the era of artificial intelligence, in recent years has appeared a lot of machine-learning and deep-learning based methods for muscle segmentation and quantification. One of the most recent published articles, by Burns et al. , shows a fully-automated system which has the ability to detect and analyze truncal musculature at multiple lumbar vertebral levels and muscles groups on abdominal CT scans. Thus, making muscle segmentation faster and easier. There are already some automated software available on industry markets of Radiology for detection of sarcopenia and muscle segmentation, not only for osteoporosis detection as stablished in the “future perspectives” section.
In brief, for future review publications, authors could also consider assessment of sarcopenia in atypical settings, such us, patients with history of abdominal trauma, in which there are no sufficient literature published.