Acute kidney injury (AKI) is a common and serious complication in intensive care unit (ICU) patients that effects on mortality rate in them. ¹,²,³,⁴ There are different methods for evaluation of these patients consists of Acute Physiology and Chronic Health Evaluation (APACHE) II and III, Sequential Organ Failure Assessment (SOFA) and Risk, Injury, Failure, Loss, and End-stage (RIFLE) classification. ⁵,⁶,^7] The RIFLE classification is based on changes in serum creatinine or urine output from baseline condition. ⁷ The maximum RIFLE means the criteria evaluation based on the worst glomerular filtration rate (GFR) or the least urine output during admission. ¹ In this study, we tried to characterize AKI defined by the maximum RIFLE classification and compare it with APACHE II and SOFA classification in high risk ICU patients.

We constructed a retrospective study of all ICU patients over 5 years of period from 21 March 2003 to 21 March 2009 at Imam Reza Hospital a university affiliated hospital, in Mashhad, Iran. Patients with a history of chronic hemodialysis were excluded. In total 628 patients were included, but due to incomplete data in 219 patients they were omitted and finally 409 patients enrolled in the study. Those who readmitted to ICU that fulfilled the criteria of study every patient with the occurrence of AKI in hospital with complete data records entered as well. All patients were classified according to the maximum RIFLE class (class R, class I, or class F) that reached during their hospital stay. In addition to the RIFLE classification SOFA, APACHE II are widely used in critically ill patients. ⁸ SOFA scoring is based on assessment of organ failure in each of the respiratory, cardiovascular, hepatic, coagulation, renal, and neurological systems. Twelve physiological variables including the respiratory rate and oxygenation, blood pressure, body temperature, heart rate, etc., during the first 24 h after admission are used in providing of the APACHE II system. ⁹ The total and non-renal APACHE II and SOFA scores were calculated based on the worst variables recorded during 1 ^st day of admission in ICU and discharge along with max RIFLE. ⁸,⁹

Statistical analysis

Data were analyzed using the SPSS 11.5. The central tendency for continuous data is expressed as mean ± standard deviation (SD) or the median (interquartile range). We compared means using t-test or Mann-Whitney tests according to the parametric pre assumptions. In all comparisons P0 < 0.05 was considered significant.

There were 409 eligible adult admissions to study. The mean age of these patients was 41.9 years (SD), 21.8), and 50.37% were males. Primary diagnostic categories of ICU admission included intoxication 98 (24%), pulmonary disease 62 (15.2%), infection 53 (13%), cardiovascular disease 26 (6.4%), neurological disease 22 (5.4%), renal disease 17 (4.2%), gastrointestinal disease 13 (3.2%), others 118 (28.6%).

The mean (SE) of baseline serum creatinine was 1.2 ± 0.41 mg/dl. The mean and SE of baseline GFR was 59.96 ± 2.09. According to RIFLE classification, the AKI and non-AKI groups consisted of 181 (44.3%) and 228 (55.7%) respectively. The AKI group was also classified according to RIFLE ⁷ as follow: Risk (n = 40, 9.8%) injury (n = 66, 16.1%) and failure (n = 64, 15.6%) and reminders were in loss (n = 7, 1.71%) and end stage renal disease (n = 4, 0.98%). In the 1 ^st day of admission, the total and non-renal APACHE II and SOFA scores compared with max RIFLE. The maximum RIFLE means the criteria evaluation based on the worst GFR or the least urine output during admission. The scores of patients in class F and class I were significantly higher than patients in class R and non-AKI groups in all classifications ( P < 0.05) Table 1 and Table 2. We found similar results in comparison total and non-renal APACHE II and SOFA with max RIFLE at the time of discharge as well ( P < 0.05) Table 1 and Table 2. The overall hospital mortality rate of the ICU patients was 49.70%. Although the mortality rate in patients without renal injury was 31.3% (72 in 233), in patients with renal disease mortality rate was significantly higher 67.23% (119 in 177) ( P < 0.001).{Table 1}{Table 2}

We conducted a single-center study with 409 ICU patients to characterize AKI, defined by RIFLE classification. In this study, AKI was occurred in 44.3% in ICU patients. This result is similar to result of Park et al. study. ⁸ Although there are other reports such as study of Bellomo et al. that found a higher incidence of AKI in their results. ⁶,¹⁰ These differences may be due to the difference in type of ICU admissions (surgery, internal) and age of patients. In comparison of different methods of evaluation of ICU patients, there are significant correlation between class I and F in max RIFLE classification with total and non-renal APACHE II and SOFA scores. It means that based on RIFLE criteria, we are able to recognize patients with severe illnesses as well as APACHE II and SOFA criteria. This relation hasn′t been found between classes R with other classifications. May be this result is showing stronger ability of RIFLE criteria for detection of high risk patients to APACHE II and SOFA. Not surprisingly, the occurrence of AKI was associated with higher mortality rate in our patients even in patients in class R that were only at risk for AKI Figure 1. In our study, the ICU mortality rate based on different classes of RIFLE criteria was similar to other reports, such as studies of Park et al. and Hoste et al . results. ¹,⁸Figure 1

Hospital mortality rates for ICU patients without acute kidney injury and ICU patients with increasing risk, injury, failure

In our ICU, the RIFLE classification can be used as non-invasive, quick, and available method to predict the outcome of patients.

This study was the result of a MD student (Sanaz Arian) thesis proposal and financially supported by the research vice chancellor of Mashhad University of Medical Sciences.