A Comparison Of Different Diagnostic Criteria For Acute Kidney Injury In Critically Ill Patients

Objective:Acute kidney injury(AKI) is very common in the hospital,especially in the intensive care unit(ICU),with higher mortality and longer time staying in hospital.The Kidney Disease: Improving Global Outcomes(KDIGO) proposed a new definition and classification of AKI on the basis of the RIFLE(Risk, Injury, Failure, Loss of kidney function, and End-stage renal failure) and AKIN(Acute Kidney Injury Network) criteria in 2012. Which one is the best? Comparisons of the three criteriain critically ill patients are rare. We explored the diagnostic value of the RIFLE criteria,AKIN criteria and KDIGO criteria for the Patients with AKI in ICU, so as to contribute to early diagnosis and treatment.Methods:We retrospectively analyzed patients admitted to the ICUin the Second Hospital of Hebei Medical University from January 1st, 2013 to December 31 th, 2014.Exclusion criteria:①age<18 years. ②ICUlength of stay≤24 hours. ③patients with chronic kidney disease or patients received renal replacement therapy(RRT).④patients with a history of kidney transplantation.⑤patients with insufficient clinical recordings. We retrieved the basic data and clinical data.Severity of illness was assessed by using the APACHE II score. AKI was defined by the RIFLE, AKIN, and KDIGO criteria and we compared the difference among the three criteria in the diagnosis of acute kidney injury and the value to predict the outcome. Statistical analysis was performed using SPSS 17.0Results:We included 726 critically ill patients, of that,302 AKI patients were diagnosed by RIFLE criteria(Risk 136,Injury75,Failure 91),For AKIN criteria, there were 237 patients diagnosed as AKI(1 stage 124, 2 stage 35,3 stage78). For KDIGO criteria, there were 340 patients diagnosed as AKI(1 stage 157,2 stage 76,3 stage 108). KDIGO criteria identified more patients than did RIFLE criteria(46.8%versus 41.6%, P<0.05) and AKIN criteria(46.8% versus32.6%,P<0.01). RIFLE criteria identified more patients than did AKIN criteria(41.6%:32.6%,P<0.01). Compared with patients without AKI, in-hospital mortality was significantly higher for those diagnosed as AKI by using the RIFLE(27.2% versus 7.3%, P<0.01), AKIN(34.6% versus 6.3%, P<0.01), and KDIGO(27.1% versus 5.4%,P<0.01) criteria, respectively. No significant difference was found among the three criteria in hospital mortality(P>0.05).Meanwhile, AKI incidence was found independent from the mortality, no matter by RIFLE or AKIN or KDIGO criteria(P<0.01). The areas under the receiver operator characteristic curve for in-hospital mortality were 0.727(95%CI:0.672-0.782,P<0.01) for RIFLE, 0.755(95%CI:0.727-0.809,P<0.01) for AKIN,and0.761(95%CI 0.711- 0.811,P<0.01) for KDIGO.Conclusions:KDIGO criteria is more sensitive than RIFLE criteria and AKIN criteria, and AKIN criteria is the worst in the sensitivity of the definition of AKI in critically ill patients. Patients diagnosed as AKI had a significantly higher in-hospital mortality than non-AKI patients, no matter which criteria were used. RIFLE criteria, AKIN criteria and KDIGO criteria were predictive for in-hospital mortality in critically ill patients,but there was no significant difference among them.