BackgroundThe impact of body composition (BC) abnormalities on COVID-19 outcomes remains to be determined. ObjectivesWe summarized the evidence on BC abnormalities and their relationship with adverse clinical outcomes in patients with COVID-19. MethodsA systematic search was conducted up until 26 September, 2022 for observational studies using BC techniques to quantify skeletal muscle mass (or related compartments), muscle radiodensity or echo intensity, adipose tissue (AT; or related compartments), and phase angle (PhA) in adults with COVID-19. Methodological quality of studies was assessed using the Newcastle-Ottawa Scale. A synthesis without meta-analysis was conducted to summarize the prevalence of BC abnormalities and their significant associations with clinical outcomes. ResultsWe included 62 studies (69.4% low risk of bias) with 12–1138 participants, except 3 studies with ≤490,301 participants. Using CT and different cutoff values, prevalence ranged approximately from 22% to 90% for low muscle mass, 12% to 85% for low muscle radiodensity, and 16% to 70% for high visceral AT. Using BIA, prevalence of high FM was 51%, and low PhA was 22% to 88%. Mortality was inversely related to PhA (3/4 studies) and positively related to intra- and intermuscular AT (4/5 studies), muscle echo intensity (2/2 studies), and BIA-estimated FM (2/2 studies). Intensive care unit (ICU) admission was positively related to visceral AT (6/7 studies) and total AT (2/3 studies). Disease severity and hospitalization outcomes were positively related to intra- and intermuscular AT (2/2 studies). Inconsistent associations were found for the rest of the BC measures and hospitalization outcomes. ConclusionsAbnormalities in BC were prevalent in patients with COVID-19. Although conflicting associations were observed among certain BC abnormalities and clinical outcomes, higher muscle echo intensity (reflective of myosteatosis) and lower PhA were more consistently associated with greater mortality risk. Likewise, high intra- and intermuscular AT and visceral AT were associated with mortality and ICU admission, respectively.This trial was registered at PROSPERO as CRD42021283031.