ABSTRACT
ABSTRACT Objective To verify whether the presence of related nutritional risk indicators prior to COVID-19 diagnosis is associated with poor survival in patients with cancer. Methods We retrospectively analyzed the data of hospitalized cancer patients who tested positive for COVID-19 between March 2020 and February 2021. Nutritional risk was defined as the presence of one of the following characteristics: body mass index <20kg/m 2 , scored Patient-generated Subjective Global Assessment ≥9 points or classification B, albumin level <3.5g/dL, and C-reactive protein level ≥10mg/L, evaluated between 7 and 60 days prior to the date of patient inclusion. The endpoint measure was all-cause mortality within 30 days of COVID-19 diagnosis. Results A total of 253 patients were included, most of whom were elderly (62.4%) and female (63.6%). Overall, 45.4% of the patients were at nutritional risk. Survival was significantly lower in patients at nutritional risk (8 days; interquartile range [IQR]: 3-29) than in patients not at nutritional risk (16 days; IQR: 6-30) (p<0.001). The presence of prior nutritional risk was associated with increased 30-day mortality (HR: 1.42; 95%CI: 1.03-1.94), regardless of age, gender, tumor site or stage, and other risk factors, and the model had good discrimination accuracy (concordance statistic: 0.744). Conclusion The presence of prior nutritional risk indicators is related to poor prognosis in patients with cancer and COVID-19, emphasizing the importance of nutritional care, notably during this pandemic.