Research of the risk factors predicting progression and prognosis of acute respiratory distress syndrome

Objective To explore the early diagnosis and risk factors for judging prognosis of acute respiratory distress syndrome (ARDS), and to provide references for clinical intervention. Methods Using the method for prospective cohort study, clinical data were collected from 64 ARDS and 66 high-risk ARDS patients in Department of Respiratory Diseases of Xinqiao Hospital from January 2013 to March 2016. They included patients' demographic data, Acute Physiology and Chronic Health Evaluation system Ⅱ (APACHE Ⅱ) score, oxygenation index, blood routine test, coagulation function and inflammatory markers (procalcitonin, C-reaction protein, tumor necrosis factor and interleukin -6) within 24h and the state of survival or death of the 24th day. Risk factors for predicting progression of the high-risk ARDS patients into ARDS patients and influencing the prognosis of the ARDS patients were analyzed by using logistic regression. Results Univariate logistic regression analysis found that the independent risk factors for progression of ARDS were APACHE Ⅱ score (OR=6.764, P=0.001), hypoproteinemia (OR=10.54, P=0.002), white blood cell count (OR=3.912, P=0.012), fibrinogen (OR=9.953, P=0.064), and D-dimer (OR=4.239, P=0.029). The mortality rate was 43.75% (36/64) in ARDS group, and the oxygenation index (OR=6.573, P=0.014), platelet count (OR=9.376, P=0.003), hypoproteinemia (OR=10.738, P=0.056) were the independent risk factors of death in ARDS patients. Multivariate logistic regression showed that combination of multiple indicators for predicting ARDS improved the specificity, but reduce the sensitivity. APACHE Ⅱ and hypoproteinemia (sensitivity 62.50%, specificity 92.42%) and APACHE Ⅱ and D dimmer (sensitivity 62.07%, specificity 93.33%) had better specificity and sensitivity. The specificity and sensitivity of combining hypoproteinemia and platelet count to predict the risk of death in these patients were 77.78% and 60.71%. Conclusions In high-risk ARDS patients, combining hypoproteinemia or APACHEⅡ score with D-dimer to judge the development of ARDS has good specificity but poor sensitivity. For ARDS patients, low oxygenation index, thrombocytopenia and hypoproteinemia indicate a poor prognosis. DOI: 10.11855/j.issn.0577-7402.2017.05.17