RESUMEN
BACKGROUND & AIMS: Evidence shows that CT-derived sarcopenia can predict adverse outcomes in COVID-19 patients. However, discrepancies exist as to which vertebral level can be used to calculate sarcopenia which can effectively serve as a prognostic tool. Thus, we aim to investigate the difference in sarcopenia calculated at the Thoracic and Lumbar vertebral levels. METHODS: An online literature search was conducted on Electronic databases such as PubMed, Cochrane CENTRAL, and Google scholar. Meta-analysis was performed by using Revman 5.3 software. RESULTS: A total of 14 articles were selected for meta-analysis. The prevalence of sarcopenia calculated at the Thoracic level was 31% (95%CI 24%-37%; p < 0.00001; I2 = 86%), while sarcopenia calculated at the Lumbar vertebral level was 63% (95%CI 51%-75%; p < 0.00001; I2 = 88%). Meanwhile, sarcopenia calculated at the Upper thoracic level was a significant predictor of mortality OR 3.47 (95%CI 1.74-6.91; p = 0.0004; I2 = 56%)as compared to sarcopenia calculated at the lower thoracic OR 1.74 (95%Cl 0.91-3.33; p = 0.10; I2 = 60%)or lumbar level OR 2.49 (95%CI 0.45-13.72; p = 0.30; I2 = 57%). In addition to this sarcopenia calculated at the Upper thoracic level was also a significant predictor of severe illness OR 3.92 (95%CI 2.33-6.58; p < 0.00001; I2 = 0%) as compared to lower thoracic OR 1.40 (95%CI 0.78-2.53; p = 0.26; I2 = 67%) or lumbar level OR 1.64 (95%CI 0.26-10.50; p = 0.60; I2 = 81%) CONCLUSIONS: Sarcopenia calculated at the thoracic vertebrae and lumber level has different prognostic values. Sarcopenia is prevalent at the lumbar level. Sarcopenia at the thoracic level has a higher mortality and severity rate.