Abdominal Fat and Muscle Measurement on Routine CT: Is It Useful?

2016 
Defining body composition has always been challenging and a wide range of methods have been developed. Amongthem, computed tomography (CT) is on the rise as a strong correlation exists between tissues at abdominal levels and the whole body composition[1]. Formulas link muscle and fat area at the third lumbar level (L3) with total fat mass and fat-free mass (mostly muscles and organs) in kilograms[2]. An index of muscularity (IMS) was also computed by dividing the muscle area at L3 by the height squared (cm2/m2). This technique is mainly used in cancer studies. Imaging methods such as CT and magnetic resonance imaging (MRI) have shown that muscle and fat behave differently in cancer-related weight loss, part of the well-known cancer cachexia. Here, we confirmed that classic tools such as body weight, body mass index (BMI) and body surface area (BSA) are not sufficient for evaluating body composition in a cohort of 152 colorectal and lung cancer patients. For instance, in normal BMI range (18.5–24.9kg/m2) and overweight/obese category (>25kg/m2), respectively 53% and 30% were sarcopenic. Recent studies have shown that the role of muscle mass and quality is critical. Sarcopenia, used here as a synonym of low muscularity, is strongly linked to prognosis and dose-limiting toxicities of chemotherapy. That is consistent with different types of cancer and therapy[3]. Further more, these associations between CT-defined body composition and cancer out comes were validated in more than 50 studies, including thousands of patients all over the world. CT-based method takes advantage of images that are acquired in standard care for staging or follow-up of cancer patients. It is a prognosis tool that could also improve scaling of treatments. The field is promising and it is highly realistic for routine use in any hospital that has a computer and a CT.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    3
    References
    0
    Citations
    NaN
    KQI
    []