[Value of a model based on PIRO conception in predicting the prognosis in critical patients].

OBJECTIVE

To investigate the values of factors based on PIRO conception in predicting the prognosis of critical patients.

METHODS

The clinical data of critical patients admitted to Hainan Branch of PLA General Hospital from December 2011 to August 2013 were retrospectively analyzed. The patients were randomly divided into non-survivors and survivors groups according to 28-day outcome. Predisposition (P), injury (I), response (R) and organ dysfunction induced by injury (O) were compared between two groups. The indexes with statistical significance (P<0.2) by univariate analysis were included in multivariate logistic regression analysis, and the receiver operating characteristic curve (ROC curve) was plotted to evaluate the values of factors based on PIRO conception in predicting the prognosis of critical patients.

RESULTS

One hundred and eighty-seven critical patients were enrolled, and among them 75 (40.1%) patients died. Univariate analysis showed that the age, underlying disease scores, history of cardiovascular disease, diabetes mellitus, and cerebrovascular disease, positive blood culture, whether or not complicated with acute respiratory distress syndrome (ARDS) or severe sepsis/septic shock, procalcitonin (PCT), acute physiology and chronic health evaluation II (APACHE II), acute pathophysiology score (APS) and sequential organ failure assessment (SOFA) were found to be the factors related with the prognosis (all P<0.2). Multivariate logistic regression analysis showed that the underlying disease scores [odds ratio (OR)=1.874, 95% confidence interval (95%CI) 1.138-3.084, P=0.014], whether patients occurrence of severe sepsis/septic shock (OR=0.167, 95%CI 0.064-0.435, P=0.000) and SOFA scores (OR=1.498, 95%CI 1.283-1.750, P=0.000) were independent factors for predicting 28-day mortality. The new model combined with above factors had more prognostic value in predicting the mortality than a single variable. The area under ROC curve (AUC) for PIRO model based on indexes with statistical significance by univariate analysis was 0.877 (0.821-0.934), P=0.000. AUC for PIRO model based on underlying disease scores, severe sepsis/septic shock, SOFA scores was 0.871 (0.814-0.928), P=0.000. AUC for SOFA was 0.762 (0.687-0.837), P=0.000. AUC for APS was 0.726 (0.647-0.805), P=0.000. AUC for underlying disease scores was 0.678 (0.593-0.763), P=0.000. AUC for PCT was 0.636 (0.548-0.724), P=0.004. AUC for age was 0.618 (0.532-0.705), P=0.013].

CONCLUSIONS

The multivariate regression analysis based on PIRO system may help to predict 28 days mortality in critical patients.