267: PERFORMANCE STATUS AS A PREDICTOR OF ICU MORTALITY IN CRITICALLY ILL CANCER PATIENTS

Abstract

Introduction: Frailty and poor performance status (PS) have been associated with low survival in the general intensive care units (ICUs). However, there is limited data focused on critically ill cancer patients. As demand for ICU admission among the cancer population rises, there is a need to prioritize access, especially in resource-limited settings. Therefore, we assessed the association between baseline PS to ICU mortality in critically ill cancer patients. Methods: A retrospective cohort of adult critically ill cancer patients admitted between 12/2020 and 07/2021 to an oncological hospital in Guayaquil, Ecuador. PS deterioration was measured with the Eastern Cooperative Oncology Group scale (ECOG) prior to hospital admission, and it was classified into mild (ECOG 0-1), moderate (ECOG 2), and poor (ECOG 3-4). We conducted a logistic regression to establish associations between characteristics and ICU mortality. Results: A total of 200 patients were included, with a mean age of 57.9 years (SD 16). Of them, there were 168 (84%) patients with solid tumors and 32 (16%) with hematological malignancies. Most patients were post-surgical (52%) and the rest were medical (48%). Regarding PS deterioration, we found 71 (35.5%) with mild PS, 95 (47.5%) with a moderate PS and 34 (17.0%) with a poor PS. Most patients in the poor PS group (70.6%) had a solid tumor. Overall, 96 (48%) patients required invasive mechanical ventilation (IMV) and 88 (44%) vasopressors. Patients with poor PS required more days on IMV (1.5 vs 1 vs 0; p=.001) and days on vasopressors (1vs 1 vs 0; p=.019) than the moderate and mild groups, respectively. The overall ICU mortality was 29%. Poor PS had higher risk of ICU mortality ratio (OR 5.013 [95% CI: 1.6–15.9]) than SOFA (OR 1.3 [95% CI: 1.1-1.6], APACHE II (OR 1.1 [95% CI: 1-1.2], and m-NUTRIC (OR 0.99 [95% CI: 0.7-1.3] scores. Conclusions: A poor performance status was associated with a 5-fold increased risk of ICU mortality. Additionally, poor PS implied a higher requirement of life-sustaining therapies like IMV and vasopressors. The association between ECOG and mortality was superior to the classic ICU scores such as, APACHE II, SOFA and m-NUTRIC. Future research should determine if the PS could be a triage tool in settings with low ICU capacity.