Prone Positioning in Severe Ards Due to Covid-19

ABSTRACT

INTRODUCTION: Studies have shown early application of prone-positioning in ARDS significantly decreased mortality. Our goal is to evaluate the effect of early prone-positioning specifically on COVID ARDS patients. METHOD(S) We performed a multicenter, retrospective observational analysis with a total of 1,335 patients with COVID ARDS that underwent prone positioning. Data was obtained from all HCA facilities within the dates of 1/1/2020- 6/20/2021. ARDS was defined using the Berlin criteria. Logistic regression was used to predict the likelihood of in-hospital all-cause mortality early vs late prone-positioning. Secondary outcomes were the relationship between age of the patient, MAP, days on ventilator and ICU length of stay (ICULOS) likelihood of in-hospital mortality. RESULT(S) From 1/1/2020-6/20/2021, a total of 3,407 patients with COVID ARDS were admitted to the participating facilities. 1,335 patients were included in the final analysis. Patients were mostly between ages 51-80 years old (77%), male (61.5%), white (55.4%), all of them admitted to ICU on mechanical ventilation. In-hospital allcause mortality was significantly lower in the shorter time to prone group (< 16 hours) than the longer time to prone group (>16, >24 hours), (p < 0.001, Exp(B) = 1.119, 95% C.I. [1.088, 1.151]). Mortality rate < 16 hours (46.53%), >16 hours (55%) vs >24 hours (68.1%). Patients that were prone in < 16 hours were less likely to experience an in-hospital mortality than those prone >16 hours (X2 (1, N= 1513) = 19.051, p < 0.001). There was not any statistically significant difference between the 16- and 24-hours group. For each one-day increase in days on the ventilator the likelihood of mortality is 0.978 times as likely. (p < 0.01, Exp(B) = 0.978, 95% C.I. [0.968, 0.989]). Expired rate by time to prone < 16 hours (55.45%) vs >16 hours (79.69%). For each one-year increase in age, patients are 1.045 times as likely to experience an in-hospital mortality (p < 0.001, Exp(B) =1.045, 95% C.I. [1.033,1.056]). CONCLUSION(S) We concluded through logistic regression that the time to prone had a statistically significant relation to in-hospital all-cause mortality and patients with COVID ARDS can benefit from early prone treatment.