HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) All Versions of this Article: CJN.03361006v1 2/3/418    most recent Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Cruz, D. N. Search for Related Content PubMed PubMed Citation Articles by Cruz, D. N.

North East Italian Prospective Hospital Renal Outcome Survey on Acute Kidney Injury (NEiPHROS-AKI): Targeting the Problem with the RIFLE Criteria

Dinna N. Cruz^*, Irene Bolgan^*, Mark A. Perazella, Monica Bonello^*, Massimo de Cal^*, Valentina Corradi^*, Natalia Polanco^*, Catalina Ocampo^*, Federico Nalesso^*, Pasquale Piccinni^*, Claudio Ronco^*; for the North East Italian Prospective Hospital Renal Outcome Survey on Acute Kidney Injury (NEiPHROS-AKI) Investigators

^* Department of Nephrology and Intensive Care, San Bortolo Hospital, Vicenza, Italy; and Section of Nephrology, Department of Medicine, Yale University School of Medicine, New Haven, Connecticut

Address correspondence to: Dr. Claudio Ronco, Department of Nephrology, San Bortolo Hospital, Viale Rodolfi 37, 36100 Vicenza, Italy. Phone: +39-0444-753650; Fax: +39-0444-753973; E-mail: cronco{at}goldnet.it

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion Conclusion Disclosures References

 
Acute kidney injury (AKI) in the intensive care unit (ICU) is associated with an enhanced mortality. The Acute Dialysis Quality Initiative group has proposed the RIFLE (Risk–Injury–Failure–Loss–ESRD) classification to standardize the approach to AKI. This study was performed to estimate the AKI incidence in ICU patients in northeastern Italy and describe clinical characteristics and outcomes of patients with AKI on the basis of their RIFLE class. A prospective multicenter observational study was performed of patients who fulfilled AKI criteria in 19 ICU in northeastern Italy. Data were analyzed using multivariate logistic regression and survival curve analysis. Of 2164 ICU patients who were admitted during the study period, 234 (10.8%; 95% confidence interval 9.5 to 12.1%) developed AKI; 19% were classified as risk (R), 35% as injury (I), and 46% as failure (F). Preexisting kidney disease was present in 36.8%. The most common causes of AKI were prerenal causes (38.9%) and sepsis (25.6%). At diagnosis of AKI, median serum creatinine and urine output were 2.0 mg/dl and 1100 ml/d, respectively. ICU mortality was 49.5% in class F, 29.3% in I, and 20% in R. Independent risk factors for mortality included RIFLE class, sepsis, and need for renal replacement therapy, whereas a postsurgical cause of AKI, exposure to nephrotoxins, higher serum creatinine, and urine output were associated with lower mortality risk. In this study, AKI incidence in the ICU was between 9 and 12%, with 3.3% of ICU patients requiring renal replacement therapy. Sepsis was a significant contributing factor. Overall mortality was between 30 and 42%, and was highest among those in RIFLE class F.

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion Conclusion Disclosures References

 
Acute kidney injury (AKI) affects 5 to 7% of all hospitalized patients, and despite technological advances, it continues to be associated with poor outcomes (1,2). This syndrome is common in the intensive care unit (ICU), with a reported incidence of 1 to 25% depending on the population being studied and the criteria used to define its presence (1–4). AKI in this setting is associated with mortality rates of 50 to 70%, which have remained relatively constant in the past decades (3,5,6).

Although AKI is associated with a high risk for death, its epidemiology and outcome are not well determined, in part as a result of the lack of a standardized definition. More than 35 different definitions have been used in the literature, including one by Bellomo et al. (7), creating conflicting results and making comparisons difficult (8). A practical definition should be based on widely accepted foundations of pathophysiology, clinical behavior, response to treatment, histopathologic features, and prognosis. To define better AKI and its characteristics, the Acute Dialysis Quality Initiative (ADQI) developed a consensus definition for AKI, grading the severity of the disease from "risk" to "injury" to "failure" (Table 1) (9). Recently, a number of single-center studies evaluated the incidence and outcome of AKI using this classification (10–14). It has not been assessed, however, in a multicenter setting. This study estimates the incidence of AKI in ICU patients in northeastern Italy, describes the clinical characteristics of patients with AKI and their outcomes on the basis of their RIFLE (Risk–Injury–Failure–Loss–ESRD) category, and explores the relationship between the two components (serum creatinine and urine output) of the RIFLE criteria.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion Conclusion Disclosures References

Study Population
Patients who were older than 18 yr and who were admitted to one of the participating ICU during the study period were considered for study inclusion. From this population, only incident patients who fulfilled criteria for the risk, injury, and failure categories of the RIFLE criteria for AKI were included in the study (9). Either the serum creatinine or urine output criterion was used to diagnose AKI and determine the RIFLE classification of each patient, using the criteria that led to the worst possible classification (Table 1). The RIFLE classification at the time of diagnosis of AKI (also called RIFLE "initial") was considered for analysis. For patients in whom baseline serum creatinine values were not known (22%), the value was estimated using the Modification of Diet in Renal Disease equation (assuming average baseline GFR of 75 ml/min per 1.73 m²), as recommended by the ADQI workgroup (9,15). Patients who had stage 5 chronic kidney disease and were undergoing long-term renal replacement therapy (RRT) were excluded from the study. Two patients were admitted twice to the ICU during the study period for different reasons; in each case, the second admission was counted as a new case.

Data Collection
Multiple data elements were prospectively collected on each patient with AKI at study inclusion and were recorded on a standardized case report form that was developed for this study, which was pilot-tested in one of the participating centers (Ospedale San Bortolo) before dissemination to the other study sites. Requested data included demographics, premorbid renal function, use of potentially nephrotoxic drugs, medical history, urine output, serum creatinine levels, and associated laboratory abnormalities at time of AKI diagnosis. For patients who were treated with RRT, we collected data regarding the initial and subsequent (if any) RRT modality, as well as the type of medical and nursing personnel who initiated and managed the therapy at bedside. Indications, frequency, duration, and operational characteristics of RRT were individualized for each patient by the treating physicians at each site. There were no prespecified criteria for initiation or withdrawal of RRT or for any of its aspects. ICU mortality was the primary outcome of interest. We also collected data on cause of death and ICU length of stay, as well as recovery of renal function, defined as a serum creatinine <1.5 mg/dl (133 µmol/L) and urine output >800 ml/d, or a return to baseline level of renal function at time of ICU discharge or death.

For calculation of the incidence of AKI, summary data on all ICU admissions from April to June 2003 were obtained from each participating center. This included the total number of admissions; frequency distribution of age, gender, and ICU length of stay for all ICU admissions; and the overall ICU mortality for that time period. Individual patient data were not available for patients without AKI.

Statistical Analyses
Continuous data are presented as the mean (95% confidence interval [CI]) or the median (interquartile range), as appropriate, and nominal data are presented as percentages. Comparisons among multiple groups were performed using the Fisher or the ² test for nominal variables, and ANOVA (with Bonferroni correction) or Kruskal-Wallis test was used for numerical variables where appropriate. Kaplan-Meier survival analysis was used to compare 28-d ICU mortality among the RIFLE categories. Patients who were alive at ICU discharge were censored. The log-rank statistic was used to test differences between groups. Exploratory univariate analysis for several variables was performed to identify possible predictors of ICU mortality. P < 0.05 was considered as statistically significant. Serum creatinine levels and urine output were evaluated as both continuous variables (actual values and natural log) and dichotomous variables (above or below a particular level). Previous exposure to potentially nephrotoxic agents was explored in three different forms: "Angiotensin-converting enzyme inhibitors/angiotensin II type 1 receptor blockers (ACEI/ARB) or nonsteroidal anti-inflammatory drugs," "ACEI/ARB and nonsteroidal anti-inflammatory drugs together," or "any nephrotoxic agent." The form "any nephrotoxic agent" had the strongest association with both mortality and need for RRT and was used for the final analysis.

Multivariable logistic regression analysis was then conducted to investigate independent predictors for ICU mortality and need for RRT. The analysis was conducted with backward stepwise method (exclusion probability 0.20; inclusion probability 0.10) for the following covariates: Age, sepsis, postsurgical cause of AKI, previous use of nephrotoxic agents, need for RRT, serum creatinine (natural log), and urine output (natural log). The model's goodness of fit was tested with the Hosmer-Lemeshow statistic. Agreement between the creatinine and urine output criteria of the RIFLE classification was evaluated with the test. For exploration of the contribution of each criterion to the outcomes of interest, multivariable logistic regression analysis was repeated using the RIFLE class based on serum creatinine only (RIFLE-c) and again using the RIFLE class based on urine output only (RIFLE-o). Patients who failed to fulfill criteria for R_c or R_o are labeled as class 0_c or 0_o, respectively. All statistical analyses were performed using SPSS 11.5.1 (SPSS, Chicago, IL). A two-sided P < 0.05 was considered significant.

	Results

Top Abstract Introduction Materials and Methods Results Discussion Conclusion Disclosures References

 
We evaluated a total of 2164 admissions in 19 ICU during the study period. Characteristics of the participating centers are shown in Table 2. Patients who were admitted to these ICU had a mean age of 64.3 yr, and 62.2% were male. The overall ICU mortality was 13.4% (95% CI 11.0 to 16.0%) and ranged from 3.4 to 27.4% in the individual study sites. The mean length of stay for survivors was 5.7 d.

Outcomes
Overall ICU mortality for all patients with AKI was 36.3% (95% CI 30.1 to 42.5%), significantly higher than that in the entire ICU cohort (P < 0.001). The top three reported causes of death were sepsis (35%), cardiac causes (12.8%), and pulmonary causes (4.7%). ICU mortality was highest among class F (49.5%), followed by class I (29.3%) then class R (20%). Kaplan-Meier survival analysis was statistically significant (P = 0.03 by log rank test; Figure 1). The results of the multivariate logistic regression analysis are shown in Table 5. RIFLE classification was the strongest predictor of ICU mortality. With respect to class R, the odds ratio (OR) was 2.2 for class I and 4.9 for class F (P = 0.01). Other variables that were associated with increased risk for ICU mortality were sepsis as the cause of AKI and need for RRT. A postsurgical cause for AKI, previous exposure to any nephrotoxic agent, higher serum creatinine, and urine output were associated with a lower risk for death. Mortality was not significantly different between patients with a prerenal cause of AKI (30 versus 39% [other cause of AKI]; P = 0.17). Thirty-six percent of patients with AKI recovered renal function at the time of death or ICU discharge. Recovery of renal function was seen more frequently among class R (57.8%) and I (43.9%) patients than among F patients (20.6%; P < 0.001). The median ICU length of stay for AKI survivors was 10 d (interquartile range 3 to 24 d), significantly longer than that of the entire ICU cohort (P < 0.001). ICU length of stay was not significantly different among the three RIFLE classes. The analysis was repeated using the RIFLE class only on the basis of serum creatinine criteria (RIFLE-c). With respect to class 0_c, the OR was 3.0 (95% CI 0.2 to 38.3) for R_c, 7.3 (95% CI 0.6 to 88.7) for I_c, and 18.2 (95% CI 1.2 to 267.4) for F_c. However, when the analysis was performed with RIFLE class only on the basis of urine output criteria (RIFLE-o), it did not emerge as an independent predictor of mortality.

View larger version (21K): [in this window] [in a new window] [as a PowerPoint slide]   Figure 1. Kaplan-Meier curve for intensive care unit survival by RIFLE class; log-rank statistic P = 0.03. Solid line, risk; dashed line, injury; dotted line, failure.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion Conclusion Disclosures References

Four retrospective, single-center studies have reported the use of the RIFLE classification in ICU patient populations (Table 6) (10–13) and one in a general hospital population (14). The incidence of AKI in the ICU, that is, the aggregate of the R, I, and F categories, ranged from 19 to 67% (10,13), and the need for RRT ranged from 2.5 to 4.1% (10,11,13). In contrast to our study, in which we classified patients into RIFLE categories at time of diagnosis of AKI (RIFLE "initial"), others classified the patients on ICU admission (12), at peak creatinine levels (10), or maximum RIFLE class attained (13). Mortality, variably reported as hospital mortality, 30- or 90-d mortality, in class R ranged from 8 to 38.3%, in class I from 11.4 to 50%, and in class F from 26.3 to 74.4%. In all of these studies, mortality was, similar to our findings, significantly higher in class F. Kuitunen et al. (10) analyzed 813 consecutive cardiac surgery patients during an 8.5-mo period. Patients were classified into R, I, or F on the basis of the change between the preoperative and peak levels of plasma creatinine and urine output. Abosaif et al. (12) from the United Kingdom applied the RIFLE criteria to 183 ICU patients who had a serum creatinine level >1.7 mg/dl (150 µmol/L) and compared them with 24 control patients without AKI. They found a greater need for RRT in class F and noted a higher mortality as compared with our own and other studies, regardless of RIFLE class. ICU mortality in their F group was 74.4%. One possible explanation is that patients who already have AKI at the time of ICU admission have a poorer prognosis than those who develop AKI later in their ICU stay. The published literature is not clear on this issue. An earlier study from France showed delayed occurrence of AKI as an independent risk factor for mortality (6); however, a second French study did not find the same association (16). Bell et al. (11) studied 207 patients who were treated with CRRT in the ICU during a 6-yr period, 188 of whom were categorized into R, I, and F. The authors did not specifically state when the patients were classified. The remaining 16 patients were dialysis dependent before their ICU admission (considered as ESRD, RIFLE category E), and three were classified into the loss category. Being the only study that included exclusively patients who were treated with RRT, one would anticipate that this was a group with more advanced AKI and, therefore, a worse outcome. However, their mortality for all three categories was lower than that in the UK study (12) (Table 6). This may be due to a difference in case mix and center-specific practices regarding dialysis. It is possible that the Swedish center (11) may have been more aggressive in initiating RRT (criteria for CRRT not described), or patients with AKI on ICU admission have a worse prognosis. In the multivariate analysis in this study, only the RIFLE category, specifically class F, remained an independent predictor of mortality; class I had a mortality risk equal to that of class R (11). Recently, Hoste et al. (13) reported their experience with a large cohort of patients from seven ICU in a single center. In this retrospective study, they were also able to examine the progression from the initial RIFLE category to the maximum or peak RIFLE class attained by patients. They reported that more than half of R patients go on to progress to I or F, and more than one third of I patients progress to F. In terms of long-term follow-up, the Swedish study showed that only 4.8% of survivors were dialysis dependent at 6 mo (11); likewise, the UK study reported that survivors had significantly improved creatinine after 1 and 6 mo (12).

Aside from the RIFLE category, we noted that sepsis and need for RRT were independent predictors for ICU mortality. Sepsis and septic shock have been associated with increased mortality risk in several studies (1–3,17,18). Our results show that a postsurgical cause for AKI was associated with a lower risk. This is congruent with an Australian study in which patients who came from various surgical units had lower odds for hospital mortality (14). This would also be in agreement with the lower overall ICU mortality reported by Kuitenen et al. (10) in their cardiac surgery patients. Prerenal azotemia is traditionally regarded as a less problematic form of AKI. It is interesting that patients with a prerenal cause of AKI seem to have a mortality similar to those with other causes of AKI, although we may not have sufficient statistical power to detect a small difference. Although serum creatinine levels and urine output at AKI diagnosis are the basis of the RIFLE classification, they remain as independent predictor variables for mortality when placed in a multivariable analysis with RIFLE. Low urine output and anuria have been long known to be associated with poor outcomes (1,19,20). In our analysis, lower creatinine concentrations at the time of AKI diagnosis were associated with an increased risk for death. Other authors have also noted a similar pattern in their cohorts (19–21). They hypothesized that low serum creatinine, particularly after adjustment for age and gender, probably reflects loss of muscle mass. However, it could also be related to volume overload and its diluting effect on the serum creatinine. Previous studies have reported the association between volume overload and mortality, in both adults and children (22,23). We also noted that a previous exposure to nephrotoxic agents was associated with a lower risk for death. One possible explanation for this favorable prognosis is that it is a readily reversible cause of AKI, and definitive management (i.e., withdrawal of the offending agent) is relatively simple to carry out.

In our study, the use of CRRT was high, particularly with respect to the practice in the United States as reported by the Program to Improve Care in Acute Renal Disease (PICARD) study (24). The reason for the difference in center practices is not clear. We found that a collaborative effort between the nephrologist and the intensivist, as well as the ICU and dialysis nurse (data not shown), predominates among our centers, which may differ from the practices in "closed" ICU systems (25).

In our analysis, the serum creatinine criteria of the classification seemed to be a better predictor of mortality than urine output. However, the very wide CI suggest an unstable statistical model that should be interpreted with caution, which may in part be related to the relatively small number of patients in the I_o category. A rise in serum creatinine is an earlier sign of worsening renal function than oliguria. In >50% of our patients with AKI, the creatinine criteria led to a worse RIFLE class than urine output. That the predictive value of the "true" RIFLE class (using both creatinine and urine output criteria) was statistically more stable underlines the clinical utility of using the composite criteria to define the wide range of disease spectra in AKI.

The results of our study support the utility of the RIFLE classification in prognosticating the outcome of patients with AKI. The study included a large cohort of patients from multiple centers, using both serum creatinine and urine output criteria, as was intended by ADQI. To our knowledge, this is the first prospective, multicenter evaluation of the RIFLE classification and the first to evaluate the performance of the creatinine and urine output criteria. We acknowledge that there are several limitations to the data presented here. First, we had a predominantly white population, and our findings may not be applicable to more culturally diverse centers. Because of the study design, we were unable to compare directly patients with and without AKI. However, we were able to compare basic demographics and outcomes, such as ICU mortality and length of stay, between patients with AKI and the ICU population as a whole. We also did not include severity of illness scores in our evaluation. There was a need to assume a baseline creatinine value using the Modification of Diet in Renal Disease (MDRD) equation in patients in whom this value was not known. This assumption can obviously lead to misclassification of some patients. However, a true baseline is often unknown for many patients who are admitted to the ICU (13). The ADQI Committee was well aware of this contingency, and for this reason, they sought to develop a standardized rule for assuming the level of baseline renal function (9). Patient outcome in our study was evaluated at exit from the ICU, which underestimates the impact of AKI on outcome. Bell et al. (11) showed a difference of 19.8% between ICU and 2-mo mortality and 25.1% between ICU and 6-mo mortality. Last, because changes in serum creatinine and urine output lag behind real-time changes in renal function, there are some limits to the utility of the RIFLE classification, although it represents a tool that was hitherto unavailable. Ultimately, serum and urine biomarkers will redefine how we classify, treat, and monitor AKI.

	Conclusion

Top Abstract Introduction Materials and Methods Results Discussion Conclusion Disclosures References

 
The incidence of AKI in our ICU was between 9 and 12%, with 3.3% of ICU patients requiring RRT. Overall mortality was between 30 and 42% and was lowest in RIFLE class R and highest in class F. Our findings support the utility of the RIFLE classification for AKI and demonstrate that even more moderate degrees of kidney function pose a significant risk for death. Perhaps appropriate intervention in earlier stages of AKI may have a positive impact on patient outcomes. Until serum and urine biomarkers for AKI become widely available for clinical use, a simple and clinically applicable method for identifying these patients across different centers, such as the RIFLE classification, would help in recruitment of patients for future trials.

	Disclosures

Top Abstract Introduction Materials and Methods Results Discussion Conclusion Disclosures References

 
None.

	Acknowledgments

 
This work was presented in abstract form at the annual meeting of the American Society of Nephrology; November 14 through 19, 2006; San Diego, CA; and has been made possible through the International Society of Nephrology funded Fellowship of D.N.C.

Participating Centers and Investigators: Drs. Vincenzo Mione and Dario Carmelo Tomasello (Azienda Ospedaliera Santa Maria degli Angeli, Pordenone); Prof. Giorgio Della Rocca (Policlinico Universitario di Udine); Dr. Orlando Margarit (Azienda Ospedaliera Santa Maria della Misericordia, Udine); Dr. Luciano Silvestri (Ospedale di Gorizia, Gorizia); Dr. Lamberto Padovan (Ospedale Civile di Adria, Rovigo); Drs. Antonio Vassalli and Roberto Ferrari (Presidio Ospedaliero di Bassano del Grappa, Vicenza); Drs. Giorgio Zanardo and Maurizio Pegoraro (Presidio Ospedaliero di Castelfranco Veneto, Treviso); Dr. Livio Todesco (Presidio Ospedaliero di Cittadella, Padova); Dr. Antonio Digito (Presidio Ospedaliero di Feltre, Belluno); Dr. M. Luisa Ferron (Presidio Ospedaliero di Legnago, Verona); Dr. Stefania Michelon (Presidio Ospedaliero di Venezia Terraferma Ospedale Umberto Primo, Venezia); Dr. Andrea Bianchin (Presidio Ospedaliero di Montebelluna, Treviso); Drs. Marco Baiocchi, Giuseppina Bonaccorso, Raffaele Bonato (Azienda Ospedaliera di Padova, Padova); Dr. Donato Borreggine (Presidio Ospedaliero di San Donà di Piave, Venezia); Dr. Roberto Nani (Presidio Ospedaliero di Treviso, Treviso); Prof. Aldo Luzzani and Dr. Laura Borin (Azienda Ospedaliera di Verona, Verona); Dr. Pasquale Piccinni (Presidio Ospedaliero San Bortolo di Vicenza, Vicenza).

	Footnotes

 
Published online ahead of print. Publication date available at www.cjasn.org.

Received October 6, 2006. Accepted February 1, 2007.

	References

Top Abstract Introduction Materials and Methods Results Discussion Conclusion Disclosures References

 

1. Lamiere N, Van Biesen W, Vanholder R: Acute renal failure. Lancet365 :417 –430,2005[Medline]
2. Esson ML, Schrier RW: Diagnosis an treatment of acute tubular necrosis. Ann Intern Med137 :744 –752,2002[Abstract/Free Full Text]
3. Uchino S, Kellum JA, Bellomo R, Doig GS, Morimatsu H, Morgera S, Schetz M, Tan I, Bouman C, Macedo E, Gibney N, Tolwani A, Ronco C; Beginning and Ending Supportive Therapy for the Kidney (BEST Kidney) Investigators: Acute renal failure in critically ill patients: A multinational, multicenter Study. JAMA294 :813 –818,2005[Abstract/Free Full Text]
4. De Mendonca A, Vincent JL, Suter PM, Moreno R, Dearden NM, Antonelli M, Takala J, Sprung C, Cantraine F: Acute renal failure in the ICU: Risk factors and outcome evaluated by the SOFA score. Intensive Care Med26 :915 –921,2000[CrossRef][Medline]
5. Liano F, Junco E, Pascual J, Madero R, Verde E: The spectrum of acute renal failure in the intensive care unit compared with that seen in other settings. The Madrid Acute Renal Failure Study Group. Kidney Int Suppl66 :S16 –S24,1998[CrossRef][Medline]
6. Brivet F, Kleinknecht D, Loirat P, Landals P: Acute renal failure in intensive care units-causes, outcome, and prognosis factors of hospital mortality: A prospective, multicenter study. French Study Group on Acute Renal Failure. Crit Care Med24 :192 –198,1996[CrossRef][Medline]
7. Bellomo R, Kellum J, Ronco C: Acute renal failure: Time for consensus. Intensive Care Med27 :1685 –1688,2001[CrossRef][Medline]
8. Kellum J, Levin N, Bouman C, Lameire N: Developing a consensus classification system for acute renal failure. Curr Opin Crit Care8 :509 –514,2002[CrossRef][Medline]
9. Bellomo R, Ronco C, Kellum J, Mehta R, Palevsky P; the ADQI 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 Care8 :R204 –R212,2004[CrossRef][Medline]
10. Kuitunen A, Vento A, Suojaranta-Ylinen R, Pettila V: Acute renal failure after cardiac surgery: Evaluation of the RIFLE classification. Ann Thorac Surg81 :542 –546,2006[Abstract/Free Full Text]
11. Bell M, Liljestam E, Granath, Fryckstedt J, Ekbom A, Martling C: Optimal follow-up time after continuous renal replacement therapy in acute renal failure patients stratified with the RIFLE criteria. Nephrol Dial Transplant20 :354 –360,2005[Abstract/Free Full Text]
12. Abosaif N, Tolba Y, Heap M, Russell J, Nahas A: The outcome of acute renal failure in the intensive care unit according to RIFLE: Model application, sensitivity, and predictability. Am J Kidney Dis46 :1038 –1048,2005[CrossRef][Medline]
13. Hoste EA, Clermont G, Kersten A, Venkataraman R, Angus DC, De Bacquer D, Kellum JA: RIFLE criteria for acute kidney injury are associated with hospital mortality in critically ill patients: A cohort analysis. Crit Care10 :R73 –R82,2006[CrossRef][Medline]
14. 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 Med34 :1913 –1917,2006[CrossRef][Medline]
15. National Kidney Foundation: K/DOQI clinical practice guidelines for chronic kidney disease: Evaluation, classification and stratification. Am J Kidney Dis39[Suppl 1] :S76 –S92,2002[CrossRef]
16. Guerin C, Girard R, Selli J, Perdrix J, Ayzac L: Initial versus delayed acute renal failure in the intensive care unit. Am J Respir Crit Care Med161 :872 –879,2000[Abstract/Free Full Text]
17. Ronco C, Bellomo R, Homel P, Brendolan A, Dan M, Piccinni P, La Greca G: Effects of different doses in continuous venovenous hemofiltration on outcomes of acute renal failure. Lancet356 :20 –30,2000[CrossRef]
18. Schiffl H, Lang S, Fischer R: Daily hemodialysis and the outcome of acute renal failure. N Engl J Med346 :305 –310,2002[Abstract/Free Full Text]
19. Mehta R, Pascual M, Gruta C, Zhuang S, Chertow G: Refining predictive models in critically ill patients with acute renal failure. J Am Soc Nephrol13 :1350 –1357,2002[Abstract/Free Full Text]
20. Paganini EP, Halstenberg WK, Goormastic M: Risk modeling in acute renal failure requiring dialysis: The introduction of a new model. Clin Nephrol46 :206 –211,1996[Medline]
21. Lassnigg A, Schmidlin D, Mouhieddine M, Bachmann LM, Druml W, Bauer P, Hiesmayr M: Minimal changes of serum creatinine predict prognosis in patients after cardiothoracic surgery: A prospective cohort study. J Am Soc Nephrol15 :1597 –1605,2004[Abstract/Free Full Text]
22. Simmons RS, Berdine GG, Seidenfeld JJ, Prihoda TJ, Harris GD, Smith JD, Gilbert TJ, Mota E, Johanson WG Jr: Fluid balance and the adult respiratory distress syndrome. Am Rev Respir Dis35 :924 –929,1987
23. Goldstein SL, Somers MJ, Baum MA, Symons JM, Brophy PD, Blowey D, Bunchman TE, Baker C, Mottes T, McAfee N, Barnett J, Morrison G, Rogers K, Fortenberry JD: Pediatric patients with multi-organ dysfunction syndrome receiving continuous renal replacement therapy. Kidney Int67 :653 –658,2005[CrossRef][Medline]
24. Mehta RL, Pascual MT, Soroko S, Savage BR, Himmelfarb J, Ikizler TA, Paganini EP, Chertow GM; Program to Improve Care in Acute Renal Disease: Spectrum of acute renal failure in the intensive care unit: The PICARD experience. Kidney Int66 :1613 –1621,2004[CrossRef][Medline]
25. Silvester W, Bellomo R, Cole L: Epidemiology, management and outcome of severe acute renal failure of critical illness in Australia. Crit Care Med29 :1910 –1915,2001[CrossRef][Medline]

This article has been cited by other articles:

				S. M. Bagshaw, S. Uchino, D. Cruz, R. Bellomo, H. Morimatsu, S. Morgera, M. Schetz, I. Tan, C. Bouman, E. Macedo, et al. A comparison of observed versus estimated baseline creatinine for determination of RIFLE class in patients with acute kidney injury Nephrol. Dial. Transplant., April 6, 2009; (2009) gfp159v1. [Abstract] [Full Text] [PDF]

				R. A. Fowler, N. K. J. Adhikari, D. C. Scales, W. L. Lee, and G. D. Rubenfeld Update in Critical Care 2007 Am. J. Respir. Crit. Care Med., April 15, 2008; 177(8): 808 - 819. [Full Text] [PDF]

				Z. Ricci and C. Ronco Kidney diseases beyond nephrology: intensive care Nephrol. Dial. Transplant., March 1, 2008; 23(3): 820 - 826. [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: CJN.03361006v1 2/3/418    most recent Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Cruz, D. N. Search for Related Content PubMed PubMed Citation Articles by Cruz, D. N.