RIFLE Classification: prospective analysis of the association with mortality in critical ill patiants

Abstracts

INTRODUÇÃO: A classificação de RIFLE define três classes de gravidade da lesão renal aguda (LRA): Risco, Injúria e Falência; foi associada à mortalidade conforme a gradação da gravidade da LRA, porém, é pouco avaliada em estudos prospectivos. OBJETIVO: Analisar prospectivamente a associação da classificação de RIFLE com a mortalidade em pacientes criticamente enfermos. MÉTODO: Estudo de coorte prospectiva de 200 pacientes admitidos na unidade de Terapia Intensiva (UTI), no período de julho/2010 a julho/2011. Os pacientes incluídos eram maiores de 18 anos, permaneceram por mais de 24 horas na UTI e assinaram o termo de consentimento livre esclarecido. RESULTADOS: A frequência da LRA na UTI foi de 47% (n = 95), sendo o RIFLEmáximo: Risco 4,5% (n = 9), Injúria 11% (n = 23) e Falência 31,5% (n = 63). A mortalidade geral na UTI foi de 25,5% (n = 51). O RIFLE categorizado em RIFLEmáximo classe Injúria + Falência, apresentou maior mortalidade quando comparado ao subgrupo categorizado sem LRA + com LRA classe Risco (53,3% vs. 4,4). A maior classe de RIFLE alcançado apresentou maior risco relativo em associação à mortalidade: χ2 de Person = 62,2, RR = 7,46 IC: 3,2-17,2; p < 0,001. O RIFLE categorizado em RIFLEmáximo classe Injúria + Falência, o TISS-28 e o escore SOFAmáximo não renal associaram-se independentemente à mortalidade na UTI. CONCLUSÃO: A gravidade da LRA, de acordo com critério de RIFLE foi um marcador de risco para mortalidade nessa população. O grupo com LRA classe Injúria + Falência, foi associado a maior mortalidade quando comparado ao subgrupo sem LRA + com LRA que permaneceu na classe Risco.

apache; lesão renal aguda; mortalidade; prognóstico; unidades de terapia intensiva

ORIGINAL ARTICLE

^IRoyal Spanish Beneficent Society

^IIFederal University of Bahia, Bahia School of Medicine and Public Health

^IIIFederal University of Goiás

Correspondence to

ABSTRACT

INTRODUCTION: The recent RIFLE classification defines three degrees for severity of acute kidney injury (AKI): RISK, INJURY and FAILURE and was associated with mortality according to the grading of the severity of AKI, but little valued at prospective studies.

OBJECTIVE: To evaluate the association of RIFLE score with mortality in critically ill patients and compare the clinical characteristics between them.

METHOD: An observational prospective cohort study of 200 patients admitted to the ICU, from July/2010 to July/2011. Patients included were older than 18 years, had for more than 24 hours in the ICU and signed the Term of informed consent.

RESULTS: The frequence of AKI in the ICU was 47% (n = 95), the maximum RIFLE: Risk 4.5% (n = 09), Injury 11%(n = 23) and Failure 31.5% (n = 63). The ICU mortality was 25.5% (n = 51). The RIFLE categorized into class RIFLE_maximum class Injury + Failure had a higher mortality compared to the subgroup categorized No LRA + AKI Risk class (53.3% vs. 4.4%) and the greater the relative risk of the patient so classified: RR = 3.3 (95%: 2.5 to 4.4) p < 0.001. RIFLE categorized as RIFLE_maximum class Injury + Failure and SOFA_maximum score, independently associated with ICU mortality after adjustment for multiple variables.

CONCLUSION: The severity of AKI according to RIFLE criteria was a risk marker for mortality in this population. The LRA group class Injury + Failure was associated with increased mortality when compared to the subgroup Without AKI + AKI that remained in Risk class even after adjustments for multiple variables.

Keywords: acute kidney injury, apache, intensive care units, mortality, prognosis.

Introduction

Acute renal failure (ARF) is defined as the abrupt deterioration in kidney function. It is common in hospitalized patients with poor prognosis. As a reflection of the interest and concern of the treating physicians, numerous scientific studies have been published on ARF. The lack of a consensus in a definition of ARF hampers the advances in the scientific research on the topic.¹

In 2004, the Acute Dialysis Quality Initiative (a group of nephrologists) published the RIFLE classification in an attempt to standardize the definition of ARF. The acronym RIFLE refers to Risk (risk of renal dysfunction), Injury (injury/damage to the kidney), Failure (failure of kidney function); Loss (loss of kidney function); and End-stage renal disease.² The first 3 classes of RIFLE have high sensitivity and are related to the degrees of severity of renal dysfunction; they are assessed by relative changes in the serum creatinine (SCr) levels or the glomerular filtration rate (GFR) from a baseline value and in the reduction of the urine flow rate (DU) calculated per kilogram of weight at a specific time. The last 2 classes have high specificity and an evolutionary character and are defined by the duration of the loss of renal function. The severity of the acute kidney injury (AKI) is determined on the basis of the most serious of the following 2 parameters: the relative change in the SCr level or GFR and the DU (Table 1).

In this paper, we use the term AKI to refer to a complex clinical syndrome that causes both structural and functional changes in the kidneys.

The RIFLE classification system was created with the goal of establishing the presence or absence of the disease in a particular patient or situation and to describe the severity of this syndrome. This classification was not created to predict mortality or an adverse clinical course, although it is logical to link that a more severe disease results in a worse clinical course. ^2,3

In 2008, Ricci et al. published a systematic review of 24 studies describing the epidemiology of AKI and assessing its association with the severity of the disease by applying the RIFLE classification. They reported a large heterogeneity in the study methods performed. Notably, different interpretations of the RIFLE criteria may produce different epidemiologic results. However, despite differences in the methodology of the various studies, the results showed an association of RIFLE classification with mortality, i.e., the risk of death increases with the increase in the severity of the disease. Although developed on the basis of prospective studies, the RIFLE classification system has been extensively validated was extensively studied in retrospective models.

Therefore, considering that AKI is a very common, complex, clinical entity, particularly in critical patients, with a significant impact on fatal outcomes and that only a few prospective studies have been conducted in Brazil on the classification criteria of RIFLE, we undertook an observational prospective cohort study aimed to analyze the association of RIFLE classification with mortality in critically ill patients. A prospective design applied in a general ICU enables a closer assessment of the actual clinical situation in ICUs.

Methods

Studied population

The inclusion criteria were age above 18 years and ICU stay for longer than 24 hours at medical facilities in the city of Salvador, Bahia, Brazil, between July 2010 and July 2011. Patients who provided written informed consent were included in the study. The exclusion criteria were a history of chronic renal disease or renal transplantation and length of ICU stay for less than 24 hours. This research protocol was approved by the medical ethics in research committee of the Royal Spanish Beneficent Society.

Study Protocol

The follow up of patients was observational and prospective during the patient's ICU stay until the outcome, discharge, or death. The researcher was not a member of the team assisting the patients in the study and did not take participate in the therapeutic decisions concerning these individuals. Information on patient identification, conditions associated with their hospitalization, clinical progress, and relevant laboratory data were collected from the medical records as per protocol.

The criterion of urine flow measurement was adapted. Although all patients were using a permanent urinary catheter, allowing continuous measurement and hourly recording of urine flow by the nursing staff, only the cumulative urine volume over 24 hours was assessed, and the weight of the patients was estimated at 60 kg, considering the average weight of an eutrophic adult. Patients were grouped into the following categories by urinary flow: urinary flow < 30 ml/h (Risk), < 18 ml/h (Injury), and < 4 ml/h (Failure). Subsequently, analysis of the criteria for diagnosis and classification of AKI were carried out. The clinical endpoint was ICU mortality.

The RIFLE classification was used, according to the prerequisites of the Acute Dialysis Quality Initiative Group (ADQI),² for the definition and classification of acute renal failure. The evolutionary RIFLE criteria were not considered: loss of kidney function and end-stage renal disease. The ADQI recommended the use of the MDRD equation as an alternative in the absence of the serum creatinine measurements. The value of baseline serum creatinine, in this study, was established as the lowest creatinine value measured before ICU admission. In cases where the baseline creatinine value was unknown, this value was calculated by the simplified MDRD formula, assuming approximately 75 ml min/1.73m² as the value of "normal" GFR.

GFR=186x-|Scr|^-1.154x|Age|^-0.203x|0.742 if female|x|1.210 if black|.

The first classification of AKI patients considering any length of stay in the ICU was labeled as RIFLE-1, and RIFLE_maximum (RIFLE_max) was considered as the highest criterion for classification. The studied population was categorized into 2 groups: one, with patients classified as No AKI + AKI RIFLE_max Risk and another group, as AKI RIFLE_max Injury + Failure class.

With regard to sepsis and septic shock, the definition criteria proposed by the 1991 Consensus Conference of the American College of Chest Physicians/Society of Critical Care Medicine Consensus Conference Committee (ACCP/CCM) were used. The ACCP/CCM Consensus Conference: (CHEST 1992, 101:1644-55) considers systemic inflammatory response syndrome (SIRS) as a response of the body to different situations, for example, trauma, major burn, and systemic infections. In the presence of at least 2 conditions among (a) body temperature > 38ºC or < 36ºC; (b) heart rate > 90 breaths per minute (bpm); (c) respiratory rate > 20 bpm or pCO₂ < 32 mmHg; (d) leucocytes > 12,000 cells/mm³ or < 4000 cells/mm³ or presence of 10% rods, sepsis was diagnosed when SIRS is the result of a confirmed infectious process and septic shock, when hypotension or hypoperfusion is present, induced by sepsis refractory to adequate volume resuscitation, thereby necessitating the use of vasoactive drugs. Although all physiological and laboratory data necessary to fulfill the definition criteria were collected by the researchers as part of the database for this study, the definitions of sepsis or septic shock were the responsibility of the medical staff who attended to patients in the ICU and were registered in the respective records as items of the problem list.

The prognostic scores Acute Physiology and Chronic Health Evaluation (APACHE II)⁸ and Sequential Organ Failure Assessment (SOFA)⁹ were calculated on admission, on the day of RIFLE-1 assessment, and on the day of RIFLE_max assessment to avoid time-dependent bias. The Therapeutic Interventions Scoring System (TISS-28)¹⁰ score was calculated only on admission. The scores were calculated with and without the relevant renal dysfunction score for the analysis of disease severity. To calculate the scores, the most altered values of the vital data and laboratory tests were used. In sedated patients, the Glasgow coma scale (GCS) was used to assess the state of consciousness immediately before sedation. All the data needed for the calculation of prognostic scores were collected.

Statistical analysis

Statistical analyses were performed using the SPSS software version 17.0 (SPSS INC. Chicago, IL USA). Descriptive statistics were used to characterize the population. The continuous variables were presented as mean ± standard deviation or minimum valuemaximum value or as median (interquartile range, 25-75%) according to the distribution. The means were compared using the t test when normally distributed and by the Mann-Whitney U test when not. Categorical variables were analyzed by the χ² test or Fisher's exact test when this was not possible.

Univariate analysis and logistic regression were performed to assess the impact of different baseline characteristics that were statistically significant for the occurrence of the most severe AKI and mortality. A priori, the variables analyzed were use of diuretics on ICU admission, TISS-28 and non-renal APACHE II scores calculated on ICU admission, and the nonrenal SOFA score calculated on the RIFLE_max day. The scores calculated with results not consistent with renal dysfunction were chosen to avoid multicollinearity.

Mortality and survival time among the 2 sample groups classified by RIFLE_max were analyzed using duration of stay in the ICU from the day of diagnosis of RIFLE as a variable, by using the χ² test and the Kaplan-Meier survival curve. For the comparative analysis between the 2 groups, the log-rank test was used. Patients who were discharged from the ICU were censored.

Results

During the study period, 200 patients were assessed, of 53.3% (n = 107) were female, with an average age of 66 years (± 16.7). Clinical admissions were more common than surgical admissions (67.5% vs. 32.5%), with 27.3% of the patients being admitted for impairment of the respiratory tract, 26.3%, for neurological injuries; and 21.7%, for cardiac causes. The duration of stay in the ICU was 12 days (interquartile range [IQR]: 4-17 days; Table 2).

The most common causes associated with the development of AKI were as follows: septic shock in 74.2% (n = 42); sepsis in 22.5% (n = 20); low cardiac output, in 17%; and other causes, in 12 (13, 4%).The frequency of AKI on admission to the ICU was 36% (n = 72) and 47.5% (n = 95) on the day of discharge or death.

Nine patients (4.5%) were deemed to have RIFLE_max Risk; 23 (11%), Injury; and 63 (31.5%), Failure (Figure 1).

The overall ICU mortality was 25.5% (n = 51), with 53.3% of the cases being from the group AKI RIFLE_max Injury + Failure and 4.4%, from the group No AKI + RIFLE_max Risk. The category RIFLE_max Injury + Failure was found to be associated with mortality with a relative risk (RR) = 7.46 (95% confidence interval [CI]: 3.2 to 17.2; p < 0.001), Pearson's χ² = 62.2 df.¹

The RIFLE_max Injury + Failure class, the TISS-28 scores, and non-renal SOFA_max were found to be independently associated with mortality, even after adjustment for other variables (Table 3).

Mortality at 20 and 30 days of ICU admission was 55% and 80%, respectively, for the group RIFLE_max Injury + Failure; for the group No AKI + RIFLE_max Risk, it was 20% after 20 days of ICU admission and remained stable thereafter. On comparison of the survival curves of the 2 groups, the long-rank test showed p < 0.001 (Figure 2).

The groups No AKI + AKI RIFLE_max Risk and AKI RIFLEmax Injury + Failure differed in age (p < 0.002). The group AKI Injury + Failure class required diuretics (p < 0.001), vasoactive drugs (p < 0.001), mechanical ventilation (p < 0.001), and multiple medications (p < 0.001) on admission to the ICU (Table 4) more frequently than No AKI + AKI RIFLE_max Risk and had higher APACHE II and SOFA scores at ICU admission, at the time of AKI diagnosis, and on the day they reached RIFLE_max, even after the withdrawal of the scores corresponding to alterations in renal function. Similarly, they presented higher TISS-28 scores on admission (Table 5).

Of the patients initially classified into the class Risk, 40% progressed to a more severe stage of the disease, i.e., Injury (6.7%) or Failure (33.3%), while 26.3% had complete recovery of renal function and 14.7% had partial recovery. Among the patients who recovered renal function fully or partially, 18% died.

No patient initially classified into the Risk class and who remained at that stage of classification died.

Discussion

We noted a high frequency (47%) of AKI in critical care units. However, this value was lower than those reported by Hoste et al.¹ (67%) and Piccinni et al. (2010) (65.3%), but well above the 35.8% incidence of AKI published by Ostemann7 and the 36% described by Bagshaw (2007).¹¹ The differences between these values may be explained by the methodological differences between the studies. The frequencies of AKI observed by Santos¹² and De Abreu et al.¹³ in Brazil are closer to that determined in the present study. Both showed that 40.3% of their patients developed AKI.

Our findings indicated that the AKI category AKI Injury + Failure was independently associated with higher mortality and showed a higher RR of death (RR = 7.46; 95% CI: 3.2-17.2 p < 0.001). In 2008, Ricci et al.⁴ observed that the RIFLE classification was associated with the RR of death, progressively with the increase of the score rating. The Risk class was associated with a RR of 2.40 (95% CI: 1.94-2.97) for death compared to patients without AKI, while the classes Injury and Failure were associated with a mortality of 4.15 (95% CI: 3.14-5.48) and 6.37 (95% CI: 5.14- 7.9), respectively. Previous studies have also shown similar results.^1,7,11,14,15 Only 2 studies, by Bell et al.¹⁶ and Macarriello et al.¹⁷, had different results. This may be attributed to the fact that both sought to assess this association in a population of patients receiving renal replacement therapy (RRT), at the start of RRT.

Patients who developed AKI in our study were more severely ill, as represented by the higher prognostic scores of SOFA and APACHE II; these patients also required more aggressive treatment at the ICU, as indicated by the higher TISS-28 score than those without kidney injury (data not shown). This factor was also associated with severity of renal injury, i.e.,

the higher the achieved RIFLE class, the higher were the prognostic scores calculated, both on admission and on the day of development of renal injury and on the day the RIFLE_max was achieved, even after excluding, from the scores, the values related to renal dysfunction. These results are consistent with those of previous studies.^1,18 This demonstrates that the development of AKI in an intensive care setting is probably part of more severe systemic impairment (such as sepsis and septic shock) and that the severity of kidney injury is directly associated with mortality.

The pathophysiological changes inherent to acute renal dysfunction or due to the adverse effects of RRT may be considered as contributing factors, despite the better efficiency of technology in this area, as well as to reduced morbidity.^11,19,20 In this sense, the medical procedures used have extended the hospitalization time without reducing mortality. However, it is unclear whether AKI is an independent predictor of risk of death or a mere adjuvant to a more severe disease stage.^20-23

Clermont et al. (2002),²² in a prospective and multicenter study comparing the impact of acute and chronic renal failure in the evolution of critically ill patients, found that patients with AKI not requiring dialysis had higher mortality than patients without AKI. In turn, the mortality in patients with AKI requiring dialysis was twice as high as that reported in patients with chronic renal failure, thereby suggesting that the increased mortality associated with AKI is not simply due to the loss of function of the organ alone, but due to all the clinical circumstances surrounding this syndrome.

The RIFLE classification criterion allows the evaluation of the progression of renal injury.¹¹ AKI showed dynamic clinical applicability where milder categories progress to more serious conditions.¹ It has been shown, in this study, that 40% of the patients classified as RIFLE class Risk progressed to the more severe classes of Injury or Failure and that such progression carried important prognostic implications. However, none of the patients initially classified with AKI class Risk and who remained in this classification stage died.

Hoste et al.¹ were the first to assess progression of AKI in a large sample of critically ill patients. In their study, Hoste classified patients with AKI according to the RIFLE classification with the maximum score (RIFLE_max). The concept of RIFLE_max for Hoste, as in this study, was the highest classification score shown during stay in the ICU. Hoste showed that 50% of patients who developed AKI and were classified as class Risk by RIFLE had progressed in class Injury or Failure. Another study by Piccinni et al. (2011)¹⁸, aimed to determine the epidemiology of AKI 10 ICUs in Italy, demonstrated that from the patients initially classified as Risk, 38% progressed to class Injury or Failure and had a poor outcome. Meanwhile, among those who recovered their renal function fully or partially, 18% of patients in the present study and 14.7% in study by Piccinni et al. died.

In this study, old age showed no association with the severity of AKI. In a study by Hoste, the association of old age and a high score of disease severity with the occurrence of AKI did not apply to groups of patients with AKI, classified by RIFLE, who progressed to the more severe degrees of severity; in other words, patients who progressed to the RIFLE classes of Injury and Failure were not older, nor were their disease severity scores higher than those who remained in the Risk class, even after excluding the values corresponding to renal dysfunction. This confirmed that the severity of AKI is associated with mortality. It was not possible to confirm these findings in the present study because of the small size of the sample of patients classified into RIFLE class Risk.

This study has its limitations. The patients evaluated had received treatment at the same research center. Therefore, caution is necessary when extrapolating the data collected to other services. Another limitation was that the population mainly comprised patients with severe clinical conditions and consequently the recruitment of more number of such patients, was not possible. The ADQI group recommended the use of the MDRD equation in to overcome the limitation for the lack of serum creatinine. Since this equation is an estimation of creatinine levels, published results on its accuracy are conflicting. Although the estimated value of serum creatinine was used in only 20% of the sample in this study, the use of this determination may have contributed to misclassification. It is known that calculated serum creatinine levels cannot be substituted for actual creatinine values; however, the validation of the MDRD was not the aim of this study.

On the other hand, few Brazilian studies involve a prospective analysis of the RIFLE score and adhere to the rules of the ADQI. A real-time analysis approach will enable the assessment of the actual clinical situation in ICUs and ensure reliability of the data collected and the statistical analysis performed.

Our results indicate that the frequency of AKI among ICU patients remains high. Critically ill patients developing AKI are severely ill and require aggressive treatment. The RIFLE classification is an easily applicable instrument and useful for the definition and classification of AKI in critical patients. It is important to understand that the severity of kidney injury is associated with the severity of disease and mortality in this population and that acting in cases of AKI in which the RIFLE _max, has been reached can mean lost time and effort. Further studies should be directed towards the analysis of the population in the risk class that has a high probability of progressing to the more severe classes and, often, to irreversible and fatal outcomes.

References

• 1. Hoste EA, Clermont G, Kersten A, Venkataraman R, Angus DC, De Bacquer D, et al. RIFLE criteria for acute kidney injury are association with hospital mortality in critically ill patients: a cohort analysis. Crit Care 2006;10:R73.
• 2. Bellomo R, Ronco C, Kellum JA, Metha RL, Palevsky P; Acute Dialysis Quality Initiative workgroup. Acute renal failure- definition, outcome measures, animal models, fluid therapy and information technology needs: the Second international Consensus Conference of the Acute Dialysis Quality Initiative (ADQI) Group. Crit Care 2004;8:R204-12.
• 3. Hoste EA, Kellum JA. Acute kidney Injury: epidemiology and diagnostic criteria. Curr Opin Crit Care 2006;12:531-7.
• 4. Ricci Z, Cruz D, Ronco C. The RIFLE criteria and mortality in acute kidney injury: a systematic review. Kidney Int 2008;73:538-46.
• 5. Zappitelli M, Parikh CR, Akcan-Arikan A, Washburn KK, Moffett BS, Golsdtein SL. Ascertainment and epidemiology of acute kidney injury varies with definition interpretation. Clin J Am Soc Nephrol 2008;3:948-54.
• 6. Murugan R, Kellum JA. Acute kidney injury: what´s the prognosis? Nat Rev Nephrol 2011;7:209-17.
• 7. Ostermann M, Chang RW. Acute kidney injury in the intensive care unit according to RIFLE. Crit Care Med 2007;35:1837-43.
• 8. Knaus WA, Draper EA, Wagner DP, Zimmerman JE. APACHE II: a severity of disease classification system. Crit Care 1985;13:818-29.
• 9. Vincent JL, Moreno R, Takala J, Willatts S, De Mendonça A, Bruining H, et al. The SOFA (Sepsis-related Organ Failure Assessment) score to describe organ dysfunction/failure. On behalf of the Working Group on Sepsis-Reated Problems of the European Society of Intensive Care Medicine. Intensive Care Med 1996;22:707-10.
• 10. Miranda DR, Rijk AP, Schaufeli W. Simplified Therapeutic Intervention Scoring System: the TISS-28 items--results from a multicenter study. Crit Care Med 1996;24:64-73.
• 11. Bagshaw SM, George C, Dinu I, Bellomo R. A multi-centre evaluation of the RIFLE criteria for early acute kidney injury in critically ill patients. Nephrol Dial Transplant 2008;23:1203-10.
• 12. Santos ER. RIFLE: association with mortality and length of stay in critically ill acute kidney injury patients. Rev Bras Ter Intensiva 2009;21:359-68.
• 13. de Abreu KL, Silva Júnior GB, Barreto AG, Melo FM, Oliveira BB, Mota RM, et al. Acute kidney injury after trauma: prevalence, clinical characteristics and RIFLE classification. Indian J Crit Care Med 2010;14:121-8.
• 14. Lopes JA, Fernandes P, Jorge S, Gonçalves S, Alvarez A, Costa e Silva Z, et al. Acute kidney injury in intensive care units patients: a comparison between the RIFLE and the Acute Kidney Injury Network classifications. Crit Care 2008;12:R110.
• 15. Uchino S, Bellomo R, Goldsmith D, Bates S, Ronco C. An assessment of the RIFLE criteria for acute renal failure in hospitalized patients. Crit Care Med 2006;34:1913-7.
• 16. Bell M, Liljestam E, Granath F, Fryckstedt J, Ekbom A, Martling CR. Optimal follow-up time after continuous renal replacement therapy in actual renal failure patients stratified with the RIFLE criteria. Nephrol Dial Transplant 2005;20:354-60.
• 17. Maccariello E, Soares M, Valente C, Nogueira L, Valença RV, Machado JE, Rocha E. RIFLE classification in patients with acute kidney injury in need of renal replacement therapy. Intensive Care Med 2007;33:597-605.
• 18. Metha RL, Pascual MT, Soroko S, Savage BR, Himmelfarb J, Ikizler TA, et al. Spectrum of acute renal failure in the intensive care unit: the PICARD experience. Kidney Int 2004;66:1613-21.
• 19. Hoste EA, Schurgers M. Epidemiology of acute kidney injury: how big is the problem? Crit Care Med 2008;36:S146-51.
• 20. Ponce D, Zorzenon CPF, Dos Santos NY, Teixeira UA, Balbi AL. Acute kidney injury in intensive care unit patients: a prospective study on incidence, risk factors and mortality. Rev Bras Ter Intensiva 2011;23:321-6.
• 21. Bucuvic EM, Ponce D, Balbi AL. Risk factors for mortality in acute kidney injury. Rev Assoc Med Bras 2011;57:158-63.
• 22. Clermont G, Acker CG, Angus DC, Sirio CA, Pinsky MR, Johnson JP. Renal failure in the ICU: comparison of the impact of acute renal failure and end-stage renal disease on ICU outcomes. Kidney Int 2002;62:986-96.
• 23. Park WY, Hwang EA, Jang MH, Park SB, Kim HC. The risk factors and outcome of acute kidney injury in the intensive care units. Korean J Intern Med 2010;25:181-7.

RIFLE Classification: prospective analysis of the association with mortality in critical ill patiants

Katia de Macedo Wahrhaftig^I; Luis Cláudio Lemos Correia^II; Carlos Alfredo Marcílio de Souza^III

Publication Dates

• Publication in this collection
  07 Jan 2013
• Date of issue
  Dec 2012

History

• Received
  07 June 2012
• Accepted
  09 Nov 2012