Elevated Plasma Concentrations of IL-6 and Elevated APACHE II Score Predict Acute Kidney Injury in Patients with Severe Sepsis : Clinical Journal of the American Society of Nephrology

Journal Logo

Advanced Search
Acute Renal Failure

Elevated Plasma Concentrations of IL-6 and Elevated APACHE II Score Predict Acute Kidney Injury in Patients with Severe Sepsis

Chawla, Lakhmir S.; Seneff, Michael G.; Nelson, David R.; Williams, Mark; Levy, Howard; Kimmel, Paul L.; Macias, William L.

Author Information

*Department of Anesthesiology and Critical Care Medicine and †Division of Renal Diseases and Hypertension, Department of Medicine, George Washington University Medical Center, Washington, DC; and ‡Lilly Research Laboratories, Eli Lilly and Company, Indianapolis, Indiana

Address correspondence to: Dr. Lakhmir S. Chawla, Department of Anesthesiology and Critical Care Medicine, The George Washington University Medical Center, 900 23rd Street, NW, Room G-105, Washington, DC 20037. Phone: 202-715-4570; Fax: 202-715-4759; E-mail: [email protected]

Accepted October 18, 2006

Received July 18, 2006

Clinical Journal of the American Society of Nephrology 2(1):p 22-30, January 2007. | DOI: 10.2215/CJN.02510706
  • Free

Abstract

Critically ill patients are at high risk for developing acute kidney injury (AKI), and this is associated with increased mortality (1). Critically ill patients who develop AKI and require renal replacement therapy (RRT) have a mortality of 50 to 80% (1–3). Despite many advances in therapeutic and research techniques in the past 20 yr, including the introduction of genomics and proteomics, fundamental changes in the outcome of patients with AKI have not occurred (4,5). The limited progress may be related to many factors, including (1) lack of a consensus definition for AKI, (2) lack of early diagnostic tests that indicate the onset of AKI and renal injury, and (3) the absence of effective therapy for AKI aside from RRT (5).

Animal models of ischemic, toxic, and septic AKI all have suggested multiple therapeutic agents that seem to attenuate renal injury if administered before insult or shortly thereafter (6–8). However, to date, no randomized, controlled studies in humans have shown benefit in treating established AKI (9,10). The vast majority of AKI studies in humans (including the anaritide and the insulin growth factor-1 clinical trials) used increases in serum creatinine concentration (Scr) as the criterion for determining which patients have AKI (9,10). Scr has significant limitations as an early marker of AKI. Creatinine is dissolved in extracellular water, and when patients experience volume expansion (e.g., resuscitation in patients with sepsis), decreases in GFR often are not reflected by an early commensurate increase in Scr. Because of these limitations, using increases in Scr as an early marker of AKI may prevent intervention at the onset of renal injury. In current clinical practice, the recognition of AKI often occurs many hours or even days after the initial insult, when the impact of therapeutic agents may be limited. If patients who are at high risk for developing AKI can be well characterized and identified early, then preventive treatments can be initiated expeditiously. The risk factors for developing AKI have been well characterized in certain subset populations, such as contrast-related AKI and AKI after cardiac surgery (11–13). However, only three studies have evaluated the predictors of AKI in patients with sepsis (sepsis-associated AKI [SA-AKI]), and two of these studies were conducted in the same intensive care unit using two separate cohorts (14–16). The purpose of this study was to investigate and elucidate further the predictors of SA-AKI.

Materials and Methods

We assessed 547 of the 840 patients in the placebo group of the Prospective Recombinant Human Activated Protein C Worldwide Evaluation in Severe Sepsis (PROWESS) study (17). The inclusion and exclusion criteria for this study are outlined in the original report (17). In brief, from July 1998 through June 2000, eligible patients were enrolled in this randomized, double-blind, placebo-controlled trial, which was conducted at 164 centers in 11 countries. The institutional review board at each center approved the protocol, and written informed consent was obtained from all participants or their authorized representatives. The criteria for severe sepsis (SS) were a modification of those defined by Bone et al. (18), which summarized the American College of Chest Physicians and the Society of Critical Care Medicine Consensus Committee Meeting (Table 1). Patients were eligible for the trial when they had a known or suspected infection on the basis of clinical data at the time of screening and when they met the following criteria within a 24-h period: Three or more signs of systemic inflammation and sepsis-induced dysfunction of at least one organ or system that developed within 24 h. Patients had to begin treatment within 24 h after they met the inclusion criteria. Patients were randomly assigned in a 1:1 manner to receive drotrecogin α activated or placebo (0.9% saline with or without 0.1% human serum albumin) at each center. Block randomization stratified according to site was used, and all assignments were made through a central randomization center. Drotrecogin α activated, at a dosage of 24 μg/kg body wt per h, or placebo was administered intravenously at a constant rate from foil-wrapped bags for a total duration of 96 h. The patients, investigators, and the sponsor were unaware of the patients’ treatment assignments. The study protocol did not call for a standardized approach to critical care (17).

Patients

For our study, we excluded placebo PROWESS patients when (1) admission renal Sepsis Organ Failure Assessment (SOFA) (19) score was 2 or more (i.e., baseline creatinine ≥2 mg/dl); (2) RRT was used at the start of the study; (3) there was absence of Scr measures during the first week after baseline (Table 2). This left 547 patients of the 840 patients for assessment. All patients who were enrolled in the PROWESS study had baseline demographic, clinical, and biochemical data obtained. In addition, baseline laboratory tests were obtained: Platelet count, protein C concentration, plasma IL-6 concentration, and level of plasma d-dimer. AKI was defined as an increase in Scr of 25% or 0.3 mg/dl during the first week after baseline. Acute Physiology and Chronic Health Evaluation II (APACHE II) scores were determined in all patients at the time of enrollment (20).

Statistical Analyses

Univariate analyses were used to compare patients without and with AKI with Wilcoxon rank-sum tests (for continuous or ordered categorical variables) or χ2 tests (for unordered categorical variables). Multivariable Cox proportional hazards regression was used to assess risk factors. These methods were used so that patients who died without AKI were included in the analysis but censored at the time of death rather than considered non-AKI cases. Therefore, patients who died without AKI are used for the period of their survival to evaluate risk factors for AKI, but after their death date, they are no longer used as either AKI or non-AKI events. No patients in the cohort were lost to follow-up at 28 d. Variables included generally were the same as those to predict mortality in the article by Ely et al. (21), with the additional variables of baseline PF ratio (partial pressure of arterial oxygen divided by the fraction of inspired oxygen), baseline serum albumin, and history of malignancy (Table 3) (22). There were no significant deviations from the proportional hazards assumptions.

To reduce possible type I error as a result of multiple variables, we performed a method of internal validation to prevent analysis-wide errors. Patients were separated into two randomly selected groups. One group served as the model-building group, in which all the predictor variables were included. Variables in the model-building group with P < 0.10 then would be applied to the second group of the patients, and if they maintained significance in the same direction, then they were included in the “final model.” Using this method, the type I error for a given variable would be 0.10 to be selected in the model-building phase and 0.05 (one tailed) to be selected in the validation stage, so a given variable would have a type I error rate of 0.005.

Log-rank tests were performed to compare Kaplan-Meier curves. Correlations between variables were assessed with Spearman nonparametric rank-based correlations. Because of the relatively rare frequency of hemofiltration, we examined the 28-d window for hemofiltration as compared with the 7-d proximate window for AKI. Because of the rarity of hemofiltration, we used a logistic regression analysis to maximize the power for the data at hand. Statistical analyses were performed with SAS version 8.02 software (SAS Institute, Cary, NC).

Results

A total of 547 patients were included in our study, the mean age was 59.8 ± 17.0, and 43.3% of the cohort were female. The ethnicity breakdown was as follows: White 83.2%, black 5.9%, and other 11%. Univariate relationships with AKI are presented in Table 3. Of the 547 patients, 127 (23.2%) placebo patients met the AKI criteria. Patients with AKI had a higher incidence of a dependence on the basis of activity of daily living scale (38.6 versus 26.7%; P = 0.01), older age (63.7 versus 58.7 yr; P = 0.008), a lower baseline platelet count (193,000 versus 222,000; P = 0.02), a lower serum albumin (g/L; 18.9 ± 6.8 versus 20.6 ± 6.9; P = 0.03), a lower PF ratio (142 ± 83.1 versus 173 ± 126; P = 0.01), a higher baseline respiratory SOFA score (2.9 versus 2.7; P = 0.02), higher preinfusion APACHE II score (24.8 versus 22.0; P = 0.0002), and higher log of IL-6 (6.6 versus 5.8; P = 0.0006). Of the 127 patients who developed AKI in the first week, 25 (19.7%) eventually required hemofiltration during the 28-d study period and 72 (56.6%) patients met the composite 28-d death or dialysis end point. Of the 420 patients without AKI in the first week, seven (1.7%) eventually required hemofiltration during the 28-d study period and 77 (18.3%) patients met the composite 28-d death or dialysis end point.

Multivariable models indicated that among the placebo patients, baseline IL-6 and APACHE II scores were significant in both the model-building and model-validation groups (Table 4). Baseline IL-6 was significant in the overall placebo population (P = 0.0006), and for every 10-fold increase in IL-6, the risk for AKI increased 16%. APACHE II also was significant in the overall population (P = 0.0008), with a 23% increase in risk for AKI for every five-point increase. We also removed the creatinine points that are part of the APACHE II score to determine whether the relationship was driven solely by the creatinine points. APACHE II without creatinine remained significant (P = 0.002), with a 22% increase in risk for AKI for every five-point increase. The statistical interaction between IL-6 and APACHE II was NS in the multivariable model (P = 0.77). The correlation between IL-6 and APACHE II was positive and significant but not to a level high enough to effect the validity of the model (Spearman correlation, 0.12; P = 0.004). Of the patients who initially were included, 32 required hemofiltration during the 28-d period. A logistic regression of risk for hemofiltration was performed with the two AKI risk factors identified in the multivariate analysis. Baseline IL-6 was a significant risk factor (P = 0.01), whereas APACHE II was NS (P = 0.36).

Of the 547 patients, 180 (32.9%) met the PROWESS renal entry criteria of urine output <0.5 ml/kg body wt per h for 1 h, despite adequate fluid resuscitation. Removal of these patients provides a subset of 367 patients, 70 (19.1%) of whom developed AKI. Within this subset of patients, IL-6 remained a significant (P = 0.02) risk factor for AKI, and for every 10-fold increase in IL-6, the risk for AKI increased 15%. However, in this subset, APACHE II score either with or without creatinine points was not related significantly to AKI risk (P = 0.20 and P = 0.21, respectively). Therefore, IL-6 was the sole risk factor within this patient subgroup that met the significance criteria.

To determine whether a specific component of the APACHE II score was the key component of the overall score’s predictive ability, we analyzed each component of the APACHE II score separately. We found that no single component of the APACHE II score was associated with the development of AKI (data not shown). In patients with an APACHE II score <25, the highest IL-6 quartile had significantly more AKI compared with the first, second, and third quartiles (P = 0.001; Figure 1). Figures 1, 2, and 3 show the development of AKI by IL-6 quartile with APACHE II score ≥ and <25.

Discussion

PROWESS was an international, multicenter, prospective, randomized, controlled, clinical trial that showed that drotrecogin α (activated) improved survival in patients with SS (17). PROWESS represents the largest clinical trial for SS ever completed (17). In analyzing this well-characterized cohort of placebo patients with SS, we have identified by multivariable Cox regression two predictors of AKI: Increasing log IL-6 and increasing APACHE II score. Both log IL-6 and increasing APACHE II score were robust risk factors of AKI given the modeling and validation process that we used.

APACHE II score, a measure of severity of illness, reproducibly predicts mortality in patients who are critically ill, including critically ill patients with AKI (16,20,23,24). Although it might be intuitive to expect that overall severity of illness of a patient with sepsis would predict AKI independently, this finding has not been well delineated. Previously, Iglesias et al. (16) showed that APACHE II score predicts acute renal failure (ARF) in patients with septic shock (odds ratio 1.03; 95% confidence interval 1.003 to 3.92; P = 0.062). Although the P value in this multivariate analysis was >0.05, APACHE II score predicted ARF in the univariate analysis. These data taken in the context of our findings suggest that measures of severity of illness are useful in predicting AKI. Because the APACHE II score contains many components, including mean arterial pressure and Scr, we assessed each individual component of the APACHE II score to determine whether a single component primarily was responsible for the score’s overall predictive ability. Because no single component of the APACHE II score predicted AKI (data not shown), we conclude that the APACHE II score’s ability to predict AKI is greater than the sum of its parts, in both its inclusiveness and its ability to score severity on a wider continuum. This finding, if validated further, would provide a valuable and objective measure for identifying patients who are destined to develop AKI.

Our study is not the first to assess predictors of SA-AKI in critically ill patients, but it is the largest. Hoste et al. (14) evaluated the day-of-admission risk factors that were associated with the development of ARF in a cohort of patients who had sepsis in a surgical intensive care unit. In their cohort of 185 patients, 30 (16.2%) developed ARF. These investigators found that low pH (<7.30) and elevated baseline Scr concentration (>1.0 mg/dl) were associated independently with the development of ARF (14). In the same intensive care unit, Yegenaga et al. (15) evaluated a separate cohort of 257 patients with systemic inflammatory response syndrome/sepsis, 29 (11%) of whom developed ARF. These investigators found that in a multivariate analysis, age, Scr level, central venous pressure, and presence of liver failure significantly contributed to a logistic regression model for ARF (15).

Our population differed from these two cohorts. We evaluated 547 patients, 127 (23.2%) of whom developed AKI. We excluded patients with Scr ≥2.0 mg/dl (renal SOFA score of 2 or more) to ensure that patients with advanced AKI at baseline were not included in the cohort. In addition, our population of patients were those who were enrolled in a multinational clinical trial for the treatment of SS. Patients in the PROWESS study had pretreatment blood samples collected and analyzed for plasma IL-6, d-dimer, and protein C concentrations, which provides an assessment of inflammation and coagulation. Moreover, our clinical end point was AKI defined as a 25% rise in Scr or an increase of 0.3 mg/dl compared with ARF, which was defined as an increase in Scr >2.0 mg/dl or oliguria (<400 ml/24 h) as used in the previous two studies. We chose to look at smaller increases in Scr on the basis of the American Society of Nephrology’s research consensus meeting on acute kidney disease and evidence that small increases in Scr predict mortality in patients after cardiac surgery, patients with contrast-induced nephropathy, and hospitalized patients (25–28).

We were able to find only one other study that evaluated proinflammatory cytokines and acute kidney disease in patients with sepsis. Iglesias et al. (16) evaluated the placebo group in the Norasept II study, which was a study that evaluated the use of monoclonal TNF-α antibody in patients with septic shock. Within this group of severely ill patients, these investigators found plasma IL-6 levels to be higher in patients who developed ARF compared with those who did not develop ARF (62 ± 148 ng/ml versus 29 ± 87 ng/ml; P = 0.057, respectively). However, in their study, plasma IL-6 did not predict ARF in their multivariate analysis (16). In contrast to our study, Iglesias et al. (16) evaluated a different population and a different outcome. In our study, we evaluated the development of AKI in patients with SS (Table 1), whereas Iglesias et al. evaluated the outcome of ARF (Scr >3.5 mg/dl or need for dialysis) in patients with septic shock (hypotension despite volume resuscitation, most requiring vasopressors).

The pathophysiology of acute kidney diseases in patients who are critically ill with sepsis is not well understood. Animal models of sepsis and acute kidney disease have identified multiple potential injury pathways, which include but are not limited to endothelin-induced renal vasoconstriction, sepsis-induced endothelial dysfunction, disseminated intravascular coagulation, neurohormonal renal vasoconstriction, and direct effects of endotoxemia on renal vascular tone via cytokine activation (29,30). Most of these injury mechanisms suggest renal arteriole vasoconstriction as a result of decreased renal perfusion pressure as the primary pathophysiology in SA-AKI, which may explain the common continued use of therapeutic agents that cause renal artery vasodilation (e.g., dopamine, fenoldopam) despite data showing lack of efficacy for the treatment of AKI with these agents (31–33). We believe that the data presented here do not support the concept that decreased renal perfusion pressure as a consequence of systemic hypotension is the primary pathologic pathway of SA-AKI. Namely, increased log IL-6 predicted AKI in our study, whereas none of the measures of deranged hemodynamics did. Specifically, mean arterial pressure, presence or dosage of vasopressors, or cardiovascular SOFA score failed to predict the development of AKI. The view that deranged hemodynamics, specifically hypotension, is not the sole cause of SA-AKI in not new and is supported by the only animal model (cecal ligation and puncture) that performs similarly to human SA-AKI. In their pioneering study, Miyaji et al. (8) showed that in the face of volume resuscitation and appropriate antibiotics, evidence of acute kidney disease occurred. In addition, when the animals were killed, the renal histology did not show tubular necrosis, as would be expected if decreased perfusion, local or systemic, was the primary cause of injury (8). The absence of tubular necrosis in multiple animal models of SA-AKI have led some investigators to opine that SA-AKI represents acute tubular apoptosis instead of acute tubular necrosis (8,34,35). Holly et al. (35) showed that in a cecal ligation and puncture rat model of SA-AKI, serum concentrations of IL-6 were higher in animals that developed AKI as compared with animals that did not develop AKI, and this elevation of IL-6 preceded histologic evidence of kidney injury. These observations are congruent with the data that we have presented. If additional studies confirm our observations and those of Holly et al., then IL-6 may prove to be a useful tool in identifying patients for preventive and therapeutic intervention (35).

We believe that SA-AKI is more complex than a common pathway of renal arteriole vasoconstriction, hypoperfusion (local or systemic), and ischemia that lead to acute tubular necrosis. SS is defined by the combination of sepsis plus an organ failure (Table 1) (18). These organ failures (e.g., cardiovascular, pulmonary, metabolic) are direct manifestations of the sepsis syndrome within these organ systems. Similarly, instead of viewing SA-AKI as a consequence of the deranged systemic or local hemodynamics of the sepsis syndrome, we view SA-AKI as the renal manifestation of the sepsis syndrome.

The role of proinflammatory cytokines in acute kidney injury is not well understood. Kwon et al. (36) showed that increased levels of IL-6 and IL-8 predict sustained kidney injury in patients who receive renal allografts, reinforcing the role of inflammation in acute kidney disease. However, to date, direct proinflammatory cytokine manipulation (e.g., TNF-α and IL-1 pathway interruption) in patients with established sepsis has not been fruitful (37–39). Increased plasma levels of IL-6 predict mortality in patients with sepsis and in patients with sepsis and acute kidney disease (40,41). We have shown that IL-6 also predicts the development of AKI and need for RRT in patients with SS. These data suggest that early inflammation as manifested by elevated levels of proinflammatory cytokines forewarns poor outcomes related to acute kidney disease and death in patients with SS. Therapies that mitigate this inflammatory response and are undertaken early in the course of SS may prove life saving to the subset of patients who have increased levels of circulating proinflammatory cytokines in the face of SS.

Analyzing placebo data sets of clinical trials has advantages and disadvantages as compared with prospective observational studies. The advantages are that the cohort of patients is well defined and excludes patients who have low likelihood for survival (i.e., moribund patients). The major disadvantage of analyzing a placebo group from a clinical trial is that the results are not broadly applicable to all patients with that disorder. In addition, our study lacks the ability to measure the local renal hemodynamic effects that may play a critical role in the pathophysiology of SA-AKI.

Conclusion

Measures of severity of illness (APACHE II score) and increased plasma concentration of IL-6 independently predict AKI in patients with SS. Further, given the fact that cardiovascular parameters failed to predict AKI supports the notion that inflammation is a component of SA-AKI.

Disclosures

M.W., H.L., and W.L.M. are employees of Eli Lilly and Company.

F1-8
Figure 1:
Kaplan-Meier estimates of remaining free of acute kidney injury (AKI) on the basis of IL-6 quartile.
F2-8
Figure 2:
PROWESS placebo patients with Acute Physiology and Chronic Health Evaluation II (APACHE II) <25. Kaplan-Meier estimates of remaining free of acute kidney injury (AKI) on the basis of IL-6 quartile.
F3-8
Figure 3:
PROWESS placebo patients with APACHE II ≥25. Kaplan-Meier estimates of remaining free of AKI on the basis of IL-6 quartile.
T1-8
Table 1:
Summary of inclusion criteriaa
T2-8
Table 2:
Sepsis Organ Failure Assessment (SOFA) Score (18)a
T3-8
Table 3:
Demographics and clinical characteristics and univariate associations with AKIa
T4-8
Table 4:
Multivariable Cox proportional hazards regression to determine risk factors for development of AKI within groups of patients randomly split and factors that were significant in both sets included in the overall final modela

L.S.C.'s effort is supported in part by Satellite Research: Norman S. Coplon Extramural Research Grant.

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

Chawla et al. have associated elevated levels of IL-6 in sepsis with subsequent acute kidney injury. In this month’s issue of JASN, Hoke et al. draw attention to pulmonary injury after bilateral nephrectomy associated with cytokines (pp. 155–164). This suggests that cytokines may be important in the outcome of patients with sepsis and acute renal failure.

References

1. Brivet FG, Kleinknecht DJ, Loirat P, Landais PJ: Acute renal failure in intensive care units: Causes, outcome, and prognostic factors of hospital mortality; a prospective, multicenter study. French Study Group on Acute Renal Failure. Crit Care Med24 :192– 198,1996
2. 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
    3. Mehta RL, McDonald B, Gabbai FB, Pahl M, Pascual MT, Farkas A, Kaplan RM: A randomized clinical trial of continuous versus intermittent dialysis for acute renal failure. Kidney Int60 :1154– 1163,2001
    4. Liano F, Pascual J: Epidemiology of acute renal failure: A prospective, multicenter, community-based study. Madrid Acute Renal Failure Study Group. Kidney Int50 :811– 818,1996
    5. Star RA: Treatment of acute renal failure. Kidney Int54 :1817– 1831,1998
    6. Kohda Y, Chiao H, Star RA: alpha-Melanocyte-stimulating hormone and acute renal failure. Curr Opin Nephrol Hypertens7 :413– 417,1998
    7. Masuda Y, Nakayama N: Protective action of diethyldithiocarbamate and carbon disulfide against renal injury induced by chloroform in mice. Biochem Pharmacol32 :3127– 3135,1983
      8. Miyaji T, Hu X, Yuen PS, Muramatsu Y, Iyer S, Hewitt SM, Star RA: Ethyl pyruvate decreases sepsis-induced acute renal failure and multiple organ damage in aged mice. Kidney Int64 :1620– 1631,2003
      9. Allgren RL, Marbury TC, Rahman SN, Weisberg LS, Fenves AZ, Lafayette RA, Sweet RM, Genter FC, Kurnik BR, Conger JD, Sayegh MH: Anaritide in acute tubular necrosis. Auriculin Anaritide Acute Renal Failure Study Group. N Engl J Med336 :828– 834,1997
      10. Hirschberg R, Kopple J, Lipsett P, Benjamin E, Minei J, Albertson T, Munger M, Metzler M, Zaloga G, Murray M, Lowry S, Conger J, McKeown W, O’Shea M, Baughman R, Wood K, Haupt M, Kaiser R, Simms H, Warnock D, Summer W, Hintz R, Myers B, Haenftling K, Capra W: Multicenter clinical trial of recombinant human insulin-like growth factor I in patients with acute renal failure. Kidney Int55 :2423– 2432,1999
      11. Fortescue EB, Bates DW, Chertow GM: Predicting acute renal failure after coronary bypass surgery: Cross-validation of two risk-stratification algorithms. Kidney Int57 :2594– 2602,2000
      12. Thakar CV, Arrigain S, Worley S, Yared JP, Paganini EP: A clinical score to predict acute renal failure after cardiac surgery. J Am Soc Nephrol16 :162– 168,2005
        13. Weisbord SD, Palevsky PM: Radiocontrast-induced acute renal failure. J Intensive Care Med20 :63– 75,2005
        14. Hoste EA, Lameire NH, Vanholder RC, Benoit DD, Decruyenaere JM, Colardyn FA: Acute renal failure in patients with sepsis in a surgical ICU: Predictive factors, incidence, comorbidity, and outcome. J Am Soc Nephrol14 :1022– 1030,2003
        15. Yegenaga I, Hoste E, Van Biesen W, Vanholder R, Benoit D, Kantarci G, Dhondt A, Colardyn F, Lameire N: Clinical characteristics of patients developing ARF due to sepsis/systemic inflammatory response syndrome: Results of a prospective study. Am J Kidney Dis43 :817– 824,2004
        16. Iglesias J, Marik PE, Levine JS: Elevated serum levels of the type I and type II receptors for tumor necrosis factor-alpha as predictive factors for ARF in patients with septic shock. Am J Kidney Dis41 :62– 75,2003
        17. Bernard GR, Vincent JL, Laterre PF, LaRosa SP, Dhainaut JF, Lopez-Rodriguez A, Steingrub JS, Garber GE, Helterbrand JD, Ely EW, Fisher CJ Jr: Efficacy and safety of recombinant human activated protein C for severe sepsis. N Engl J Med344 :699– 709,2001
        18. Bone RC, Balk RA, Cerra FB, Dellinger RP, Fein AM, Knaus WA, Schein RM, Sibbald WJ: Definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis. The ACCP/SCCM Consensus Conference Committee. American College of Chest Physicians/Society of Critical Care Medicine. Chest101 :1644– 1655,1992
        19. Vincent JL, Moreno R, Takala J, Willatts S, de Mendonca A, Bruining H, Reinhart CK, Suter PM, Thijs LG: The SOFA (Sepsis-related Organ Failure Assessment) score to describe organ dysfunction/failure. On behalf of the Working Group on Sepsis-Related Problems of the European Society of Intensive Care Medicine. Intensive Care Med22 :707– 710,1996
        20. Knaus WA, Draper EA, Wagner DP, Zimmerman JE: APACHE II: A severity of disease classification system. Crit Care Med13 :818– 829,1985
        21. Ely EW, Laterre PF, Angus DC, Helterbrand JD, Levy H, Dhainaut JF, Vincent JL, Macias WL, Bernard GR: Drotrecogin alfa (activated) administration across clinically important subgroups of patients with severe sepsis. Crit Care Med31 :12– 19,2003
        22. Chawla LS, Abell L, Mazhari R, Egan M, Kadambi N, Burke HB, Junker C, Seneff MG, Kimmel PL: Identifying critically ill patients at high risk for developing acute renal failure: a pilot study. Kidney Int68 :2274– 2280,2005
        23. Chen YC, Hsu HH, Kao KC, Fang JT, Huang CC: Outcomes and APACHE II predictions for critically ill patients with acute renal failure requiring dialysis. Ren Fail23 :61– 70,2001
        24. Wang IK, Wang ST, Chang HY, Lin CL, Kuo HL, Chen TC, Lee CH, Chuang FR: Prognostic value of acute physiology and chronic health evaluation II and organ system failure in patients with acute renal failure requiring dialysis. Ren Fail27 :663– 669,2005
        25. Chertow GM, Burdick E, Honour M, Bonventre JV, Bates DW: Acute kidney injury, mortality, length of stay, and costs in hospitalized patients. J Am Soc Nephrol16 :3365– 3370,2005
        26. 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
          27. Rihal CS, Textor SC, Grill DE, Berger PB, Ting HH, Best PJ, Singh M, Bell MR, Barsness GW, Mathew V, Garratt KN, Holmes DR Jr: Incidence and prognostic importance of acute renal failure after percutaneous coronary intervention. Circulation105 :2259– 2264,2002
            28. American Society of Nephrology Renal Research Report. J Am Soc Nephrol16 :1886– 1903,2005
            29. Lameire N, Van Biesen W, Vanholder R: Acute renal failure. Lancet365 :417– 430,2005
            30. Schrier RW, Wang W: Acute renal failure and sepsis. N Engl J Med351 :159– 169,2004
            31. Bellomo R, Chapman M, Finfer S, Hickling K, Myburgh J: Low-dose dopamine in patients with early renal dysfunction: A placebo-controlled randomised trial. Australian and New Zealand Intensive Care Society (ANZICS) Clinical Trials Group. Lancet356 :2139– 2143,2000
            32. Stone GW, McCullough PA, Tumlin JA, Lepor NE, Madyoon H, Murray P, Wang A, Chu AA, Schaer GL, Stevens M, Wilensky RL, O’Neill WW: Fenoldopam mesylate for the prevention of contrast-induced nephropathy: A randomized controlled trial. JAMA290 :2284– 2291,2003
              33. Tumlin JA, Finkel KW, Murray PT, Samuels J, Cotsonis G, Shaw AD: Fenoldopam mesylate in early acute tubular necrosis: A randomized, double-blind, placebo-controlled clinical trial. Am J Kidney Dis46 :26– 34,2005
              34. Wan L, Bellomo R, Di Giantomasso D, Ronco C: The pathogenesis of septic acute renal failure. Curr Opin Crit Care9 :496– 502,2003
              35. Holly MK, Dear JW, Hu X, Schechter AN, Gladwin MT, Hewitt SM, Yuen PS, Star RA: Biomarker and drug-target discovery using proteomics in a new rat model of sepsis-induced acute renal failure. Kidney Int70 :496– 506,2006
              36. Kwon O, Molitoris BA, Pescovitz M, Kelly KJ: Urinary actin, interleukin-6, and interleukin-8 may predict sustained ARF after ischemic injury in renal allografts. Am J Kidney Dis41 :1074– 1087,2003
              37. Abraham E, Anzueto A, Gutierrez G, Tessler S, San Pedro G, Wunderink R, Dal Nogare A, Nasraway S, Berman S, Cooney R, Levy H, Baughman R, Rumbak M, Light RB, Poole L, Allred R, Constant J, Pennington J, Porter S: Double-blind randomised controlled trial of monoclonal antibody to human tumour necrosis factor in treatment of septic shock. NORASEPT II Study Group. Lancet351 :929– 933,1998
              38. Abraham E, Laterre PF, Garbino J, Pingleton S, Butler T, Dugernier T, Margolis B, Kudsk K, Zimmerli W, Anderson P, Reynaert M, Lew D, Lesslauer W, Passe S, Cooper P, Burdeska A, Modi M, Leighton A, Salgo M, Van der AP: Lenercept (p55 tumor necrosis factor receptor fusion protein) in severe sepsis and early septic shock: A randomized, double-blind, placebo-controlled, multicenter phase III trial with 1,342 patients. Crit Care Med29 :503– 510,2001
                39. Opal SM, Fisher CJ Jr, Dhainaut JF, Vincent JL, Brase R, Lowry SF, Sadoff JC, Slotman GJ, Levy H, Balk RA, Shelly MP, Pribble JP, LaBrecque JF, Lookabaugh J, Donovan H, Dubin H, Baughman R, Norman J, DeMaria E, Matzel K, Abraham E, Seneff M: Confirmatory interleukin-1 receptor antagonist trial in severe sepsis: A phase III, randomized, double-blind, placebo-controlled, multicenter trial. The Interleukin-1 Receptor Antagonist Sepsis Investigator Group. Crit Care Med25 :1115– 1124,1997
                40. Groeneveld AB, Tacx AN, Bossink AW, van Mierlo GJ, Hack CE: Circulating inflammatory mediators predict shock and mortality in febrile patients with microbial infection. Clin Immunol106 :106– 115,2003
                41. Simmons EM, Himmelfarb J, Sezer MT, Chertow GM, Mehta RL, Paganini EP, Soroko S, Freedman S, Becker K, Spratt D, Shyr Y, Ikizler TA: Plasma cytokine levels predict mortality in patients with acute renal failure. Kidney Int65 :1357– 1365,2004
                Copyright © 2007 The Authors. Published by Wolters Kluwer Health, Inc. All rights reserved.