P0376 / #562: RISK FACTORS AND MORTALITY FOR PEDIATRIC ACUTE RESPIRATORY DISTRESS SYNDROME (PARDS) IN CRITICALLY ILL CHILDREN WITH CANCER IN THE BRAZILIAN AMAZON: AN OBSERVATIONAL STUDY

Aims & Objectives: To determine the incidence and risk factors associated with ARDS and mortality. Methods: Retrospective cohort study, from an oncological hospital in the Brazilian Amazon, over a 18 months period (2016-2017). Only mechanically ventilated children for more than 24h was included. Those intubated for reasons other than critical illness, long PICU stay (> 90d) and admitted on CRP were excluded. The variables worst values were collected at the first 24h.The primary outcome was 28-day mortality. Results: Among 459, 220 were ventilated and 181 were included, 56(30.9%) without and 125(69.1%) with ARDS, according to the PALICC criteria. Mortality with and without ARDS was 37.6% versus 25%(p: 0.097). There was difference between the two groups with regard to key clinical variables such as leukemia(48% vs 28%; p: 0.016), maintenance chemotherapy phase (76.1% vs 45.4%; p: 0.0001), febrile neutropenia (60.8% vs 35.7%;p: 0.002), septic shock(33.6% vs 14.3%; p: 0.005), PICU stay longer than 10d (61.6% vs 19.7%; p <0.0001), ventilation time longer than 10d (63.2% vs 0%; p <0.0001) and PRISMIV greater than 20 (61.6 % vs 28.6%; p <0.0001). There were lower levels of paO2, pCO2 and ph in the ARDS group (101.8, 41.6 and 7.32 versus 7.38, 34.2 and 182.4; p <0.0001, 0.032 and 0.007, respectively), despite of higher levels of MAP, PEEP and FiO2 (12.7, 7.3 and 79.4% versus 8.7, 5.7 and 49.3%; p <0.0001). Conclusions: There was a higher incidence and mortality into ARDS group. The main risk factor was elevated MAP for ARDS, while for mortality was acidosis.