Risk factors of the occurence and death of acute respiratory distress syndrome: a prospective multicenter cohort study

Abstract

To explore the risk factors of the occurence and 28-day death of acute respiratory distress syndrome (ARDS) in intensive care unit (ICU). A prospective multicentral cohort study was conducted. The patients from five ICUs of grade A tertiary hospitals in Beijing from July 2009 to March 2014, including sepsis, septic shock, trauma, pneumonia, aspiration, massive blood transfusion, bacteremia and pulmonary contusion, were enrolled. Researchers in each center reported the records with uniform tables, which included demographic, systemic conditions, the primary disease, and the severity within 24 hours, past history and so on. According to the admission diagnosis in ICU, these patients were divided into ARDS group and other severe disease control group. The risk factors of occurence and prognosis of ARDS were analyzed by univariate analysis, multivariate logistic regression and multivariate COX regression analysis. Kaplan-Meier method was applied to draw the 28-day survival curves of the two groups. There were 343 critical patients included in this prospective multicenter cohort study, of which 163 patients who developed ARDS were considered as ARDS group (2 case lost to follow-up, and 49 died) and 180 patients who did not developed ARDS regarded as severe control group (1 case lost to follow-up, and 34 died). The 28-day mortality of ARDS group was significantly higher than that of severe control group [30.43% (49/161) vs. 18.99% (34/179), χ² = 6.013, P = 0.014]. Multivariate logistic analysis showed that aspiration [odds ratio (OR) = 6.390, 95% confidence interval (95% CI) =2.046-19.953, P = 0.001], history of alcohol (OR=4.854, 95% CI = 1.730-13.617, P = 0.003), sepsis (OR = 2.859, 95% CI=1.507-5.425, P = 0.001), pneumonia (OR = 2.822, 95% CI = 1.640-4.855, P<0.001), acute physiology and chronic health evaluation II (APACHEII) score (OR=1.050, 95%CI=1.007-1.094, P=0.022) were significantly associated with increased risk of ARDS occurence. When respiratory rate>30 beats/min (OR=3.305, 95%CI = 1.910-5.721, P<0.001), heart rate>100 beats/min (OR = 2.101, 95% CI = 1.048-4.213, P = 0.037) happened in critically ill patients, it highly suggested ARDS would happen. The proportion of the patients whose serum creatinine>176.8 μmol/L in ARDS group was lower than that in control group (OR = 0.387, 95% CI = 0.205-0.733, P = 0.004). Multivariate COX regression analysis showed that old age and septic shock were significantly associated with the increased risk of in 28-day death of ARDS [advanced age: hazard ratio (HR)=1.040, 95%CI=1.018-1.064, P<0.001; septic shock: HR=3.209, 95% CI = 1.676-6.146, P<0.001]. Kaplan-Meier showed that the survival patients in ARDS group was significantly lower than those in severe control group (χ² = 7.032, P = 0.008). Among critical ill patients, aspiration, history of alcohol, sepsis, pneumonia, increased APACHEII score were the risk factors of ARDS development. Respiratory rate>30 beats/min and heart rate>100 beats/min could predict the occurrence of ARDS in critical patients. Old age and septic shock were the risk factors of 28-day death of ARDS.