OBJECTIVE The present study was aimed (1)to evaluate the ability of the AKIN classification to predict prognoses in critically ill patients,(2)to investigate the influence of the timing of CRRT defined by AKIN classification on patients outcome,(3)to further identify risk factors associated with the prognoses of the critically ill patients and the patients treated with CRRT through univariate and multivariate analysis.METHODS We retrospectively studied 544 adult patients admitted for≥24 h to the ICU in Qilu hospital of Shandong University from January 2008 to December 2009.We applied the AKI classification to 544 adult patients, and the severity of AKI in these patients was classified into NAKI(no AKI), AKIâ… ,â…¡andâ…¢according to the AKIN criteria.(1)The mortality rate in ICU,APACHEâ…¡score, SOFA score, the number of failed organs and their length of hospital stay in ICU were compared among patients with different severity of AKI. Their APACHEâ…¡score, SOFA score, and other clinical data concerned were also compared between the nonsurvivors and survivors.(2)We compared the mortality rate between patients managed with CRRT versus other AKI patients without CRRT treatment.Their APACHEâ…¡score, SOFA score and the number of failed organs were also compared between the tow groups. We further investigated risk factors associated with the prognoses of patients treated with CRRT by comparing their APACHEâ…¡score, SOFA score,mean arterial pressure(MAP), PaO2/FiO2 before CRRT and MAP, PaO2/FiO2 measured 24 hours after CRRT between the nonsurvivors and survivors.RESULTS (1) According to the AKIN criteria, AKI occurred in 35.5%, with a maximum RIFLE category of AKIâ… in 14.8%,AKIâ…¡in 8.2% and AKIâ…¢in 11.9%.(2)The overall mortality rate in ICU was 24.1%.AKI,defined by any AKIN category,was associated with an increase in ICU mortality(OR7.48,95%CI 4.831~11.587,P<0.0001).The mortality rate was 37% for AKI 1,51.1% for AKIâ…¡and 60%for AKIâ…¢.(3)There was a positive correlativity between the number of failed organs and the mortality rate(P<0.001,r=0.512),the number of failed organs and the AKIN category(P<0.0001, r=0.613),also the mortality rate and the AKIN category(P=0.005, r=0.202).(4)In multivariate analysis, each AKIN category was independently associated with ICU mortality(OR:AKIâ… 3.532;AKIâ…¡5.231;AKIâ…¢10.301). The other independent risk factors for ICU mortality included medical diseases,septic shock,pre-existing chronic illness,APACHEâ…¡score,,the number of failed organs, mechanical ventilation and CRRT.(5)The mortality rate for patients managed with CRRT in AKIâ…¢subgroup was higher than the other groups(69.7%),while with the lowest 30-day survivor rate(18.2%for AKIâ…¢,53%for AKIâ…¡and 100%for AKIâ… ).(6)Among the AKI patients treated with CRRT,the nonsurvivors had a significantly lower mean ABP and a lower mean serum bicarbonate level, which were measured 24hours after CRRT,and than the survivors.Multivariate analysis identified age>60 years old(OR 6.854)and the number of failed organs (OR 2.714)were the independent risk factors for mortality,while the serum bicarbonate level measured 24hours after CRRT was the protective factor(OR 0.653,95%CI 0.459-0.931,P=0.018).CONCLUSIONS (1)The AKIN category closely relates to the prognoses in critically ill patients.The AKIN criteria for acute renal injury has some direction significance to the early detection and classification of AKI and to the prediction of clinical outcomes in critically ill patients.(2)The critically ill patients with AKI requiring CRRT continue to have a high mortality rate, which indicates the importance of the timing of RRT. Early detection and early CRRT may improve the prognosis of patients.(3)A higher MAP and a higher serum bicarbonate level measured 24hours after CRRT may predict a more favorable prognosis.
|