Intensive Care Medicine

, Volume 41, Issue 8, pp 1411–1423 | Cite as

Epidemiology of acute kidney injury in critically ill patients: the multinational AKI-EPI study

  • Eric A. J. Hoste
  • Sean M. Bagshaw
  • Rinaldo Bellomo
  • Cynthia M. Cely
  • Roos Colman
  • Dinna N. Cruz
  • Kyriakos Edipidis
  • Lui G. Forni
  • Charles D. Gomersall
  • Deepak Govil
  • Patrick M. Honoré
  • Olivier Joannes-Boyau
  • Michael Joannidis
  • Anna-Maija Korhonen
  • Athina Lavrentieva
  • Ravindra L. Mehta
  • Paul Palevsky
  • Eric Roessler
  • Claudio Ronco
  • Shigehiko Uchino
  • Jorge A. Vazquez
  • Erick Vidal Andrade
  • Steve Webb
  • John A. Kellum
Seven-Day Profile Publication



Current reports on acute kidney injury (AKI) in the intensive care unit (ICU) show wide variation in occurrence rate and are limited by study biases such as use of incomplete AKI definition, selected cohorts, or retrospective design. Our aim was to prospectively investigate the occurrence and outcomes of AKI in ICU patients.


The Acute Kidney Injury–Epidemiologic Prospective Investigation (AKI-EPI) study was an international cross-sectional study performed in 97 centers on patients during the first week of ICU admission. We measured AKI by Kidney Disease: Improving Global Outcomes (KDIGO) criteria, and outcomes at hospital discharge.


A total of 1032 ICU patients out of 1802 [57.3 %; 95 % confidence interval (CI) 55.0–59.6] had AKI. Increasing AKI severity was associated with hospital mortality when adjusted for other variables; odds ratio of stage 1 = 1.679 (95 % CI 0.890–3.169; p = 0.109), stage 2 = 2.945 (95 % CI 1.382–6.276; p = 0.005), and stage 3 = 6.884 (95 % CI 3.876–12.228; p < 0.001). Risk-adjusted rates of AKI and mortality were similar across the world. Patients developing AKI had worse kidney function at hospital discharge with estimated glomerular filtration rate less than 60 mL/min/1.73 m2 in 47.7 % (95 % CI 43.6–51.7) versus 14.8 % (95 % CI 11.9–18.2) in those without AKI, p < 0.001.


This is the first multinational cross-sectional study on the epidemiology of AKI in ICU patients using the complete KDIGO criteria. We found that AKI occurred in more than half of ICU patients. Increasing AKI severity was associated with increased mortality, and AKI patients had worse renal function at the time of hospital discharge. Adjusted risks for AKI and mortality were similar across different continents and regions.


Acute kidney injury Critically ill Renal replacement therapy Epidemiology Kidney function Hospital mortality 


Over the past decades there was an increasing incidence of acute kidney injury (AKI), and given the adverse outcomes especially in the long term, AKI is now a growing concern for health care worldwide [1, 2]. AKI is a frequent complication in patients admitted to the intensive care unit (ICU) and is associated with adverse outcomes including increased length of ICU and hospital stay, development of chronic kidney disease (CKD), and increased short- and long-term mortality risk [3, 4, 5, 6].

Since the publication of the RIFLE consensus classification for AKI, and the modifications by the Acute Kidney Injury Network (AKIN), and Kidney Disease: Improving Global Outcomes (KDIGO), these definitions have been used in the majority of studies reporting on AKI [7, 8, 9]. Establishing an accurate event rate for AKI is important for health policy, quality initiatives, and for design of clinical trials. However, analyzing AKI in the ICU setting from existing databases is often limited by missing data elements needed for application of these definitions (e.g., inclusion of urine output, selection of baseline serum creatinine, etc.). Administrative databases are limited, as billing codes do not capture many cases of AKI [10]. Together with differences in patients’ baseline characteristics, this may explain the wide variation in the occurrence of AKI reported in ICU patients, with an average reported occurrence of 30–40 % [2]. Further, there is no information on the worldwide patterns of AKI using current definitions.

The objective of the Acute Kidney Injury–Epidemiologic Prospective Investigation (AKI-EPI) study was to prospectively investigate the epidemiology of AKI in ICUs worldwide using the latest consensus definition for AKI and a standardized data collection instrument.


The AKI-EPI study is a multicenter international cross-sectional study on the epidemiology of AKI in ICU patients. A convenience sample of investigators was asked to record AKI during the first week of admission in ten or more consecutively admitted ICU patients. When a hospital had several ICUs, the investigators were encouraged to record these as individual cohorts. Investigators were recruited after announcement of the study by the principal investigators at international critical care meetings.

This manuscript reports results according to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines [11].


Patients who were 18 years or older and admitted to the ICU were included. Exclusion criteria were chronic renal replacement therapy (RRT), kidney transplantation within 3 months before ICU admission, anticipated alive ICU discharge within 24 h, and readmission to the ICU during the same hospitalization episode. The number of cases recorded by the investigators determined the sample size of the study.

In order to reduce the risk of information bias, we excluded units that recorded less than 10 patients, and units were allowed to record a maximum of 65 patients.

Data collection

Data were anonymously recorded via a specific password-protected website. The website was opened from 1 April 2009 until 31 December 2010. The data collection included both serum creatinine and urine output details for diagnosis of AKI and assessment of AKI severity. AKI was diagnosed and classified on the basis of the worst serum creatinine or urine output criteria according to the KDIGO classification [9]. For this a reference serum creatinine concentration was recorded. This reference serum creatinine value was defined as a concentration obtained within a 3-month period before ICU admission and considered to be representative of the baseline kidney function according to the judgment of the treating physician. If this reference value was not available, we used the minimum of either serum creatinine at the time of admission to the ICU, or in patients without CKD, a calculated serum creatinine concentration using the MDRD equation as recommended by the Acute Dialysis Quality Initiative (ADQI) [7, 9, 12]. When used, modality of RRT was also recorded.

The Simplified Acute Physiology Score (SAPS) 3 score was used for assessment of severity of illness at the time of ICU admission, and the Sequential Organ Failure Assessment (SOFA) score for organ dysfunction at the time of diagnosis of AKI [13, 14]. Glomerular filtration rate was estimated with the CKD-EPI equation (eGFR), and patients with eGFR less than 60 mL/min/1.73 m2 were classified as CKD class 3 or higher [15]. Outcomes were measured at the time of ICU and hospital discharge and included mortality and the renal outcomes serum creatinine, eGFR, and RRT.

In order to establish eGFR, the investigators scored race and ethnicity of the participants (Caucasian, black, Asian, Hispanic, or other).

We used five different methods to group countries: continents, according to latitude, world zones according to the United Nations geo-scheme classification, according to the World Bank’s classification of income of countries, and to the proportion of the global domestic product (GDP) spend for country health expenditure as reported by the World Health Organization [16, 17, 18].

Local ethics committees approved the study according to the local regulations.

Statistical analysis

Data are presented as median and interquartile range (IQR; 25th and 75th percent quartile) or proportion [95 % confidence interval (CI)]. Comparison of categorical variables was performed with the Chi-square or the Fisher exact test, and continuous variables were compared with the Mann–Whitney U test. In case of missing data that are essential for assessment of AKI, patients were excluded. In case of data missing completely at random, the number of observations included in the analysis is reported. We planned a sensitivity analysis of the occurrence rate of AKI in patients with known reference serum creatinine value.

Multivariate logistic regression analysis was conducted to investigate risk factors for mortality and AKI. A three-level hierarchical logistic regression model was applied using proc glimmix. Variables selected for inclusion in the regression model were those with biological or plausible rationale and a p value of 0.25 or less in bivariate analysis of AKI and no-AKI patients, and survivors and non-survivors. In the three models we used the random effects variables country and center.

The logistic regression model for mortality was used to calculate predicted mortality. Calibration of the models was assessed by the Hosmer–Lemeshow goodness of fit test. The Hosmer–Lemeshow statistic, however, explicitly assumes that all of the observations are independent, and it is not clear if this test is robust when this assumption may be violated, as in this data set. Therefore, the goodness of fit was tested on a non-hierarchical logistic regression model in which we included all fixed covariates and also country. The area under the receiver operating characteristic curve assessed discrimination (c statistic).

Statistical significance was accepted when the p value was less than 0.05 (double sided). Statistical analysis was performed with the software packages SPSS statistics version 22 (IBM Corporation and others, NY, USA) and SAS 9.4 (SAS Institute Inc., Cary, NC, USA); 95 % CI for proportions were calculated in Excel (Microsoft Corporation, USA) [19].


A total of 139 ICUs were registered on the website. After exclusions, the final study cohort consisted of 97 ICUs that reported on 1802 patients, originating from 33 different countries (15 European, 5 South American, 5 Asian, 4 North American, 2 African, Australia and New Zealand) (study flow chart and unit characteristics are in Fig. 1 and Table 1 in the electronic supplementary material).
Fig. 1

Patients flow chart

The median age of the patients included in the study was 63 years, 64.4 % were male (95 % CI 62.2–66.6), and 68.1 % were Caucasian (95 % CI 66.0–70.3) (Table 1). AKI occurred in 1032 ICU patients (57.3 %; 95 % CI 55.0–59.6) at day 1 (1, 2) of ICU stay. In 630 patients (35.0 %; 95 % CI 32.8–37.2) baseline creatinine information was not available, and the reference serum creatinine concentration was estimated on the basis of the MDRD equation (n = 169), or serum creatinine at ICU admission (n = 461). When these patients were excluded, the occurrence of AKI was higher (62.5 %; 95 % CI 59.7–65.3; p = 0.005). Comorbidities such as cancer, hypertension, chronic heart failure, cirrhosis, AIDS, chronic obstructive pulmonary disease (COPD), or diabetes mellitus were present in 71.5 % of patients (95 % CI 69.4–73.5). Two or more co-morbidities were present in 37.6 % of patients (95 % CI 35.4–39.8). The majority of patients had an unplanned ICU admission, coming from the emergency department, operating room, or a hospital ward, and 48.1 % of the patients had surgery before admission (95 % CI 45.8–50.4).
Table 1

Baseline characteristics, outcomes, and adjusted odds ratios for acute kidney injury


All patients




Baseline characteristics

 Number of patients

1802 (100 %)

770 (42.7 %)

1032 (57.3 %)


 Age (years)

63 (52, 73)

61 (49, 70)

65 (54, 75)



1161 (64.4 %)

511 (66.4 %)

650 (63.0 %)


 Race, ethnicity




1228 (68.1 %)

496 (64.4 %)

732 (70.9 %)



100 (5.5 %)

44 (5.7 %)

56 (5.4 %)



246 (13.7 %)

131 (17.0 %)

115 (11.1 %)



156 (8.7 %)

77 (10.0 %)

79 (7.7 %)



72 (4.0 %)

22 (2.9 %)

50 (4.8 %)





  Australia and New Zealand

107 (6.0 %)

33 (4.3 %)

74 (7.2 %)



34 (1.9 %)

4 (0.5 %)

30 (2.9 %)



230 (12.8 %)

128 (16.7 %)

102 (10.0 %)



761 (42.5 %)

304 (39.6 %)

457 (44.6 %)


  North America

416 (23.2 %)

185 (24.1 %)

231 (22.6 %)


  South America

244 (13.6 %

114 (14.8 %)

130 (12.7 %)


 Type of hospital




1300 (72.1 %)

545 (70.8 %)

755 (73.2 %)



502 (27.9 %)

225 (29.2 %)

277 (26.8 %)


  Number of hospital beds

600 (300, 850)

560 (300, 850)

600 (300, 850)


 Number of ICU beds

19 (12, 23)

19 (12, 23)

19 (12, 23)


 Number of ICU beds per nurse


2 (2, 2)

2 (2, 2)

2 (2, 2)



4 (3, 6)

4 (3, 6)

4 (3, 6)


 ICU organization



  Closed ICU

1534 (85.1 %)

650 (84.4 %)

884 (85.7 %)


  Open ICU

216 (12.0 %)

93 (12.1 %)

25 (11.9 %)



52 (2.9 %)

27 (3.5 %)

25 (2.4 %)




357 (19.8 %)

145 (18.8 %)

212 (20.5 %)


  Heart failure (NYHA class IV)

101 (5.6 %)

28 (3.6 %)

73 (7.1 %)



83 (4.6 %)

23 (3.0 %)

60 (5.8 %)



858 (47.6 %)

303 (39.4 %)

555 (53.8 %)



288 (16 %)

124 (16.1 %)

164 (15.9 %)


  Diabetes (n = 1520)

384 (25.3 %)

115 (17.5 %)

269 (31.2 %)


 Reason for ICU admission (n = 1799)

  Unplanned ICU admission

1282 (71.3 %)

517 (67.2 %)

765 (74.3 %)



624 (34.6 %)

193 (25.1 %)

431 (41.8 %)


   Hypovolemic shock

152 (8.4 %)

46 (6.0 %)

106 (10.3 %)


   Septic shock

279 (15.5 %)

62 (8.1 %)

217 (21.0 %)


 Liver failure

71 (3.9 %)

15 (1.9 %)

56 (5.4 %)


 Acute abdomen, other

146 (8.1 %)

47 (6.1 %)

99 (9.6 %)



466 (25.9 %)

214 (27.8 %)

252 (24.4 %)


 Surgical status at admission (n = 1539)


  Scheduled surgery

500 (32.5 %)

241 (37.1 %)

259 (29.1 %)


  Emergency surgery

240 (15.6 %)

105 (16.2 %)

135 (15.2 %)


  No surgery

799 (51.9 %)

304 (46.8 %)

495 (55.7 %)


 Anatomical site of surgery

  Transplantation surgery

15 (0.8 %)

3 (0.4 %)

12 (1.2 %)


  Cardiac surgery

190 (10.5 %)

84 (10.9 %)

106 (10.3 %)



83 (4.6 %)

60 (7.8 %)

23 (2.2 %)


  Other surgery

483 (26.8 %)

211 (27.4 %)

272 (26.4 %)


 In-hospital location before ICU admission (n = 1638)


  Other ward

354 (21.6 %)

270 (38.1 %)

317 (34.1 %)


  Emergency room

587 (35.8 %)

240 (33.9 %)

257 (27.6 %)


  Operation room

497 (30.3 %)

126 (17.8 %)

228 (24.5 %)


  Other ICU

106 (6.5 %)

34 (4.8 %)

72 (7.7 %)



94 (5.7 %)

38 (5.4 %)

56 (6.0 %)


 SAPS 3 score

53 (42, 66)

48 (39, 59)

58 (46, 70)


 Serum creatinine at ICU admission (mg/dL)

1.10 (0.80, 1.70)

0.90 (0.70, 1.18)

1.40 (0.96, 2.39)


 Reference serum creatinine (mg/dL)

0.95 (0.78, 1.19)

0.90 (0.70, 1.10)

1.00 (0.80, 1.26)


 Reference serum creatinine based upon



1171 (65 %)

439 (57.0 %)

732 (70.9 %)


  ICU admission value

461 (25.6 %)

283 (36.8 %)

178 (17.2 %)


  MDRD recalculated value

169 (9.4 %)

48 (6.2 %)

121 (11.7 %)


 eGFR (mL/min/1.73 m2)

81 (60, 96)

85 (70, 101)

76 (53, 93)


 eGFR <60 mL/min/1.73 m2

446 (24.8 %)

118 (15.3 %)

328 (31.8 %)


 Vasoactive drugs before ICU admission

318 (17.6 %)

100 (13.0 %)

218 (21.1 %)



272 (15.1 %)

92 (11.9 %)

180 (17.4 %)



 ICU outcomes

  Length of stay ICU (days)

5 (3, 9)

4 (3, 6)

6 (4, 12)


  Creatininedischarge (mg/dL) (n = 1165)

0.9 (0.7, 1.4)

0.8 (0.6, 1.0)

1.2 (0.8, 2.0)


  RRT at discharge (n = 1693)

118 (7.0 %)

1 (0.1 %)

117 (12.1 %)


  eGFR (mL/min/1.73 m2)

80 (46, 104)

95 (75, 110)

57 (29, 89)


  eGFR <60 mL/min/1.73 m2

400 (34.3 %)

70 (13.2 %)

330 (52.0 %)



266 (15.7 %)

34 (4.7 %)

232 (24.0 %)


 Hospital outcomes

  Length of stay (days)

14 (8, 26)

12 (7, 22)

15 (9, 29)


  Creatininedischarge (mg/dL) (n = 1057)

0.9 (0.7, 1.3)

0.8 (0.7, 1.0)

1.1 (0.8, 1.8)


  RRT at discharge (n = 1694)

44 (2.6 %)

0 (0 %)

44 (4.5 %)


  eGFR (mL/min/1.73 m2)

79 (50, 103)

94 (72, 110)

61 (33.5, 90)


  eGFR <60 mL/min/1.73 m2

346 (32.7 %)

71 (14.8 %)

275 (47.7 %)



312 (18.4 %)

52 (7.2 %)

260 (26.9 %)



Odds ratio

95 % Confidence interval



Adjusted odds ratios for acute kidney injury

Age (/year)


0.995, 1.018










0.736, 1.340



Race, ethnicity







0.454, 1.536





0.327, 3.063





0.359, 1.299





0.410, 2.157



Number of ICU beds


0.990, 1.005





 Australia and New Zealand





0.186, 13.753





0.056, 1.555





0.131, 1.597



 North America


0.099, 2.159



 South America


0.107, 1.722





 Heart failure (NYHA class IV)


0.489, 1.942





0.470, 2.502





1.291, 2.427





1.261, 2.477



Reason(s) for ICU admission


 Unplanned ICU admission



 Planned ICU admission


0.713, 2.128





1.150, 2.411



 Liver failure


0.797, 5.405



 Acute abdomen, other


0.929, 2.821



Surgical status at admission


 Scheduled surgery



 Emergency surgery


0.579, 2.002



 No surgery


0.546, 1.876



 Transplantation surgery


0.786, 31.091





0.154, 0.680



In-hospital location before ICU admission


 Other ward



 Emergency room


0.543, 1.222



 Operation room


0.589, 1.573



 Other ICU


0.520, 2.038





0.560, 2.519



SAPS 3 score


1.014, 1.042



eGFR (mL/min/1.73 m2)


0.989, 1.001



Vasoactive drugs before ICU admission


0.968, 2.042





0.532, 1.265



Values are presented as n (proportion) or median (interquartile range)

Adjusted odds ratios were generated by a multivariable model for acute kidney injury with random effects variables country and center

Multivariate model: calibration was assessed by the Hosmer–Lemeshow goodness of fit test, χ2 = 6.233, df = 8, p = 0.621. Discrimination assessed by the area under the receiver operating characteristic curve for the model for occurrence of acute kidney injury = 0.728 (95 % CI 0.700–0.756), p = 0.014. Patients included in the analysis, n = 1479

AKI acute kidney injury, OR odds ratio, CI confidence interval, NYHA New York Heart Association functional classification of heart failure, COPD chronic obstructive pulmonary disease, AIDS acquired immune deficiency syndrome, COPD chronic obstructive lung disease, ICU intensive care unit, SAPS 3 Simplified Acute Physiology Score 3, eGFR estimated glomerular filtration rate

Sepsis and hypovolemia were the most frequent reported aetiologies for AKI (Table 2). Nephrotoxic drugs were reported as the aetiology for AKI in 14.4 % (95 % CI 12.0–17.3) of patients. At the time of AKI diagnosis, one-third of patients were treated with diuretics and 11.9 % with non-steroidal anti-inflammatory drugs (95 % CI 9.6–14.5). Aminoglycosides, glycopeptides, and contrast media were administered in less than 10 % of AKI patients. Half of AKI patients were treated with vasoactive therapy at the time of AKI diagnosis, and one-third were mechanically ventilated. A SOFA score of 2 or more for the liver or coagulation component was present in one-fifth of patients.
Table 2

Variables at the time of acute kidney injury (n = 666)

Etiology of AKI


271 (40.7 %)


227 (34.1 %)

 Drug related

96 (14.4 %)

 Cardiogenic shock

88 (13.2 %)

 Hepatorenal syndrome

21 (3.2 %)

 Obstruction of the urine outflow tract

9 (1.4 %)

Predisposing factors for AKI

 Diuretic treatment

216 (32.4 %)

 NSAID administration

79 (11.9 %)

 Aminoglycoside administration

45 (6.8 %)

 Glycopeptide administration

9 (1.4 %)

 Amphotericin administration

0 (0 %)

 Radiocontrast media administration

14 (2.1 %)

Organ dysfunction at time of AKI

 Mechanical ventilation


185 (27.8 %)


33 (5.0 %)


0.4 (0.3, 0.6)


90 (471, 114)


227 (142, 314)

 Vasoactive therapy

325 (48.8 %)


242 (36.3 %)

   Dose (μg/kg/min)

0.26 (0.10, 0.60)


24 (3.6 %)

   Dose (μg/kg/min)

0.25 (0.10, 0.48)


78 (10.4 %)

   Dose (μg/kg/min)

6.6 (5.0, 10.0)

   Dopamine ≤4 μg/kg/min

16 (2.4 %)


69 (10.4 %)

   Dose (μg/kg/min)

5.0 (3.2, 7.5)

 Glasgow coma score

14.5 (10, 15)

 Bilirubin (mg/dL)

0.9 (0.5, 1.8)

 SOFA score for liver ≥2 (bilirubin ≥2mg/dL)

122 (21.3 %)

 Platelets (×103/μL)

170 (111, 235)

 SOFA score for platelets ≥2 (platelets <100 × 103/μL)

135 (21.2 %)

 Urine output (mL/24 h)

955 (450, 1680)

Data are presented as n (%) or median (interquartile range)

AKI acute kidney injury, NSAID non-steroidal anti-inflammatory drug, FiO2 fraction of inspired oxygen, PaO2 arterial oxygen concentration, SOFA score Sequential Organ Failure Assessment score

Comparison of patients who had AKI and patients without AKI

Patients who developed AKI were older, more often Caucasian, more severely ill at the time of ICU admission as illustrated by a higher SAPS 3 score, and had worse kidney function at baseline and at the time of ICU admission (Table 1). A greater proportion of AKI patients were admitted from another hospital ward or ICU, had a medical reason for ICU admission, and had comorbidities such as heart failure, hypertension, diabetes, and cirrhosis. Also, a greater proportion of AKI patients had hypovolemic or septic shock, and were already treated with vasopressor agents at the time of ICU admission. Further, more AKI patients had, at the time of ICU admission, liver failure or acute abdomen.

After adjustment, hypertension, diabetes, cardiovascular cause of admission, neurosurgery, and SAPS 3 score were associated with occurrence of AKI. AKI patients had longer lengths of stay in the ICU and hospital, and worse renal outcomes indicated by a higher serum creatinine concentration, lower eGFR, more patients with CKD stage 3 or greater, and a greater proportion of patients who were treated with RRT (non-recovery of renal function) at the time of ICU and hospital discharge (Table 1).

Occurrence rate and mortality of AKI

A maximum AKI severity of KDIGO stage 1 occurred in 331 patients (18.4 %; 95 % CI 16.7–20.2), KDIGO stage 2 in 161 patients (8.9 %; 95 % CI 7.7–10.3), and KDIGO stage 3 in 540 patients (30.0 %; 95 % CI 27.9–32.1).

There was a stepwise increase in mortality with increasing AKI severity [KDIGO stage 1: odds ratio (OR) 2.19; 95 % CI 1.44–3.35, KDIGO stage 2: OR 3.88; 95 % CI 2.42–6.21, and KDIGO stage 3: OR 7.18; 95 % CI 5.13–10.04]. When adjusted for other variables that may explain mortality, KDIGO stage 2 (OR 2.945; 95 % CI 1.382–6.276; p = 0.005) and KDIGO stage 3 (OR 6.884; 95 % CI 3.876–12.228; p < 0.001) were still associated with increased in-hospital mortality (Fig. 2) (comparison alive and death in Table 2 in the electronic supplementary material; adjusted mortality models in Tables 3 and 4 in the electronic supplementary material). This was similar in two sensitivity analyses, one including countries that included 30 or more patients, and another excluding patients where baseline serum creatinine was unknown and assessed by the MDRD equation (data not shown). A third sensitivity analysis of patients without missing variables showed a significant association of AKI stage 1 with mortality (OR 2.09; 95 % CI 1.19–3.67; p = 0.010) (Fig. 2).
Fig. 2

Adjusted odds ratio for mortality per AKI severity grade. The association of severity of AKI and hospital mortality was explored and adjusted in two three-level hierarchical logistic regression models. Variables included in the first model were country and center (random effects), and a set of fixed predictors. Variables included in the first model were country and center (random effects) and a set of fixed predictors (model 1). The second model was similar to model 1, but without the variables serum creatinine at time of ICU admission and location before ICU admission (model 2).  The full models for mortality are reported in the Table 3 in the electronic supplementary material.   Model 1: Calibration assessed by the Hosmer–Lemeshow goodness of fit test, χ2 = 4.266, df = 8, p = 0.832. Discrimination assessed by the area under the receiver operating characteristic curve for the model for occurrence of hospital death = 0.847 (95 % CI 0.819–0.875), p < 0.001. Patients included in the analysis, n = 1232.   Model 2: Calibration assessed by the Hosmer–Lemeshow goodness of fit test, χ2 = 7.873, df = 8, p = 0.446. Discrimination assessed by the area under the receiver operating characteristic curve for the model for occurrence of hospital death = 0.840 (95 % CI 0.814–0.866), p < 0.001.   AKI acute kidney injury

We found a significant difference in the occurrence of AKI and in mortality for patients with AKI between continents and world zones (Fig. 1 in the electronic supplementary material). However, adjusted rates for AKI and mortality were quite similar across different continents (Table 1; Fig. 3). When countries were grouped according to income, proportion of GDP spend for health expenditure, or latitude, the rates of AKI and mortality associated with AKI were also similar (Fig. 1 in the electronic supplementary material).
Fig. 3

Adjusted association of continent and mortality of AKI patients. The association of continent and risk for mortality was explored and adjusted in a three-level hierarchical logistic regression model. Variables included in the model were country and center (random effects) and a set of fixed predictors. The full models for mortality are reported in Table 4 in electronic supplementary material.   Calibration assessed by the Hosmer–Lemeshow goodness of fit test, χ2 = 10.080, df = 8, p = 0.259. Discrimination assessed by the area under the receiver operating characteristic curve for the model for mortality = 0.800 (95 % CI 0.763–0.837), p < 0.001. Patients included in the analysis, n = 703

Renal replacement therapy

During the whole 1-week study period, a total of 243 patients were treated with RRT (13.5 % of all patients; 95 % CI 12.0–15.1, and 23.5 % of AKI patients; 95 % CI 21.1–26.2). The majority of RRT procedures were with a continuous modality (CRRT); CRRT in 615 sessions (75.2 %; 95 % CI 72.1–78.0), intermittent hemodialysis in 197 sessions (24.1 %; 95 % CI 21.3–27.1), and peritoneal dialysis in six sessions (0.7 %; 95 % CI 0.3–1.6).


This is the first cross-sectional study to evaluate the occurrence of AKI defined by the complete KDIGO classification in ICUs worldwide. As such it provides, a decade after the BEST Kidney study, a contemporary update on the occurrence rate and mortality of AKI [20]. We found that AKI was a frequent finding, occurring in over half of ICU patients. Occurrence of AKI and crude mortality from AKI showed significant variation across countries and regions of the world. However, after adjusting for baseline risk, rates of AKI and mortality for patients with AKI were actually quite similar in different continents. Although we were unable to ensure that sites were truly representative of a given country or region, our results suggest that at least for large academic institutions AKI imposes a similar burden on patients worldwide. This finding adds strength to the estimates of AKI rates and outcomes and demonstrates generalizability of our results. Prior studies have found wide variation in the rates of AKI in ICU patients, likely due to variation in the application of AKI criteria (use of urine output, estimates of baseline creatinine, etc.) [5]. While clinicians, researchers, industry, and policy makers require robust estimates of AKI incidence to guide decision-making. These results also have important implications for health care delivery around the world. Indeed they suggest that the level of health care does not influence rate and mortality of AKI. Alternatively, the results may suggest that current strategies to reduce AKI in developed countries are ineffective or have not been implemented in any large-scale way. This may be illustrated by recent findings from the UK that revealed delays in diagnosis and poor assessment of risk in 43 % of AKI cases [21].

We found a greater use of RRT in ICU patients compared to a previous international multicenter study (13.5 vs. 4.3 %) [20]. This may be explained by an increased occurrence of severe AKI during the last decade or by more liberal criteria for initiation of RRT [2].

AKI developed in patients who had hypertension or diabetes, were admitted for a cardiovascular reason, and who were more severely ill on admission. Similar to others, we found that AKI was associated with higher mortality, but also with increased length of stay, greater rates of renal non-recovery, and higher serum creatinine at discharge [12, 22, 23]. These outcomes suggest the important economic and social impact of AKI in this patient population [1, 2, 6]. Even after adjusting for a large number of possible confounders, we found that AKI defined by KDIGO was strongly associated with mortality. It is difficult to determine whether mortality can be attributed to AKI or to unmeasured confounders, especially as ICU patients often develop AKI as a consequence of an underlying illness that itself could also explain mortality. However, a strong association between AKI and mortality persists after controlling for presence of underlying disease and severity of illness. Proposed mechanisms that may explain why AKI leads to increased mortality are related to the consequences of AKI and the therapies used [24, 25]. For instance, oliguria and volume resuscitation will lead to volume overload, acidosis, and electrolyte disorders, conditions associated with increased mortality [26, 27, 28]. The inflammatory response to AKI may also lead to systemic consequences and organ cross talk, leading to, e.g., ARDS [29]. Similar to CKD, it is likely that uremic toxins in AKI may impair immunity and increase infection rates [30, 31, 32]. RRT may be life-saving for some patients with RRT, but does not entirely reverse the hazards associated with AKI [33, 34]. Finally, dosing of various drugs in AKI patients is complicated and often leads to treatment failures and adverse drug events [35, 36].

There are several limitations of this study. First, we specifically focused on patients who were admitted to ICU. Second, this study represents a snapshot in time. Especially in centers where a limited number of patients were included, this may have led to sampling bias. Third, participation of centers was on voluntary basis. It is therefore uncertain if cohorts are representative of other centers in the same country. Also, the number of patients in certain countries, continents, and regions is too low to draw firm conclusions as to differences between them. However, the similarity seen in adjusted event rates and outcomes across countries suggests that our results are widely generalizable. Fourth, the reference serum creatinine concentration was unknown in 35.0 % of patients, and estimated by either admission creatinine or back calculated by the MDRD equation, as recommend by the ADQI and KDIGO [7, 9]. Siew et al. have shown that an MDRD-derived baseline serum creatinine leads to increased AKI occurrence, whereas we previously found that this leads to very little misclassification, especially with more severe stages [37, 38]. When these patients were excluded the rate of AKI was actually higher but the impact of AKI on mortality was unchanged. Fifth, we did not collect long-term outcome data. Sixth, the multivariate models were based on a smaller number of patients as a result of missing data, thereby limiting the power of these data. Finally, we used the CKD-EPI equation for calculation of the eGFR. This equation was established in non-ICU patients with CKD, and the values calculated may therefore be less precise in our patients [39, 40].


This is the first multinational cross-sectional study where the epidemiology of AKI in ICU patients was explored using the complete KDIGO criteria. We found that AKI occurred in more than half of ICU patients. Approximately one-fifth of ICU patients had a maximum AKI stage 1, one-tenth AKI stage 2, and one-third 3 AKI stage 3. RRT was used in 13.5 % of ICU patients (23.5 % of patients with AKI). AKI therefore represents an important burden for health care. We found that increasing AKI severity was associated with increased mortality, and this association remained after correction for covariates that may explain mortality. After adjusting for baseline risk there was little variation in AKI occurrence and mortality between different regions in the world.



This study was not funded by an external source, and the authors were not paid to write this article. The European Society of Intensive Care Medicine/European Critical Care Research Network (ESICM/ECCRN) and the Acute Dialysis Quality Initiative (ADQI) endorsed this study. ESICM/ECCRN and ADQI had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript.

The AKI-EPI Study Group

Argentina: Hospital Britanico, Buenos Aires (F. Ballestero, M.L. Caivano Nemet, E. Soloaga), Hospital Santojanni, Buenos Aires (D. Chiacchiara), HZGA Simplemente Evita, Buenos Aires (M. Anchorena, P. Centeno, H. Cabrera, M. Casalis, M. Arzel), Clinica Modelo de Lanus, Lanus (J. Vazquez), Hospital Espanol de Mendoza, Godoy Cruz (R. Fernandez, W. Vazquez), Luis Lagomaggiore, Mendoza (J.M. Pina, M.J. Marengo, G. Zakalik, A. Sanchez), Sanatorio Parque, Rosario (M. Carassai), Sanatorio San Carlos, San Carlos de Barilo (G. Alvarez, S. Benitez).

Australia: Austin Health, Heidelberg, Melbourne (R. Bellomo, G. Eastwood, L. Peck), Royal Perth Hospital, Perth (S. Webb).

Austria: University Hospital Innsbruck, Innsbruck (J. Hasslacher, M. Joannidis, R. Reindl-Schwaighofer).

Belgium: Onze Lieve Vrouw Ziekenhuis, Aalst (N. De Neve), AZ Sint-Jan Brugge-Oostende, Brugge (M. Bourgeois), UZ Brussel, Brussels (P. Honoré, H. Spapen, R. Jacobs), Clinique Europe-St-Michel, Brussels (V. Collin), CHU Charleroi, Charleroi (P. Biston, P. Defrance, M. Piagnerelli), Ziekenhuis Oost-Limburg, Genk (F. Jans), Ghent University Hospital, Ghent (D. Benoit, K. Colpaert, J. Decruyenaere, I. Herck, E. Hoste, L. De Crop, C. Clauwaert, A. Verbeke), Maria Middelares, Ghent (Y. Vandormael, J. Heerman, H. Vanoverschelde), AZ St. Lucas, Ghent (D. Rijckaert, K. Leleu), CHU de Liege, Liege (D. Ledoux, P. Wiesen), CHU UCL Dinant-Godinne, Yvoir (P. Evrard, S. Bouhon).

Brazil: Hospital do Servidor Publico Estadual de Sao Paulo, Sao Paulo (B. Ribeiro de Almeida, L.K. Ramos Pereira de Araujo, S.M. Rodrigues Laranja).

Canada: University of Alberta Hospital, Edmonton (S. Bagshaw, A. Parmar), Hopital Maisonneuve Rosemont, Montreal (J. Harvey, M. Leblanc).

Chile: Clinica Alemana de Santiago, Santiago (M. Espinoza).

China: Peking University People’s Hospital, Beijing (Y. Jinsong), Beijing Friendship Hospital C, Beijing (M. Duan), Zhejiang Provincial People’s Hospital, Hangzhou (Q. Li).

Colombia: Gestion Salud Sa, Cartagena (L.M. Carcamo, C. Espinosa, A. Llama Cano, J.A. Rojas Suarez), Corbic Institute, Envigado (N.J. Fonseca), Clinica Universitaria Bolivari, Medellin (F. Molina, A. Ochoa).

Cuba: Héroes del Baire, Nueva Gerona (J.L. Vazquez Cedeno).

Egypt: El Shefaa, Alexandria (I. Sherif), Wadi El Nile Hospital, Cairo University, Cairo (S. Badawyi), Dar Alfouad hospital, Cairo (A. Alansary).

Finland: Meilahti Hospital, Helsinki (A.-M. Korhonen, S. Nisula, V. Pettilä).

France: Hôpital Antoine Béclère APHP, Clamart (F. Jacobs), Centre Hospitalier de Dieppe, Dieppe (J.-C. Chakarian), Hôpital Raymond Poincaré, Garches (F. Fadel), CHU Michallon, Grenoble (G. Dessertaine), Eduoard Herriot Hospital, Lyon (T. Rimmelé), Bordeaux University Hospital, Pessac (O. Joannes-Boyau), CHU Hôpital Jean Bernard, University of Poitiers, Poitiers (R. Robert).

Germany: Charité University Medicine, Berlin, and Magdeburg University Clinic, Magdeburg (M. Haase, A. Haase-Fielitz), Klinikum Nürnberg, Nuremberg (S. John, J Nentwich, Th. Schrautzer).

Greece: Hygeia Hospital, Maroussi, Athens (K. Edipidis), DTCA Hygeia, Athens (J. Droulias), Corfu General Hospital, Corfu (D. Arsenis, C. Psarakis), Papanikolaou General Hospital, Thessaloniki (A. Lavrentieva).

Hong Kong Special Administrative Region of the People’s Republic of China: Prince of Wales Hospital, Shatin (G. Choi, C. Gomersall).

India: Kalinga Hospital, Bhubaneswar (S. Sahu), Artemis Health Institute, Gurgaon (D. Govil), AMRI, Kolkata (A. Kundu), P. D. Hinduja National Hospital, Mumbai (S. Singh, O. Sundrani), Tata Memorial Centre, Mumbai (B. Trivedi), Ruby Hall Clinic, Pune (S. Prachee).

Italy: Vittorio Emanuele, Catania (G. Castiglione), San Martino, Genova (R. Pinzani), Policlino Umberto I, Rome (M. Cellie, E. Alessandri), San Bortolo, Vicenza (D. Cruz, C. Ronco).

Japan: Kyoto Prefectural University Hospital, Kyoto (F. Amaya), Jikei University School of Medicine, Minatoku (S. Uchino), Osaka City General Hospital, Osaka (T. Natsuko, H. Shimaoka), The University of Tokyo Hospital, Tokyo (K. Doi, T. Yoshida, E. Noiri), Kyorin university Hospital, Tokyo (K. Moriyama), Yokohama City Minato Red Cross Hospital, Yokohama (T. Takei).

Mexico: Medical Center ABC, Mexico City (J. Aguirre), Hospital Espanol de Mexico, Mexico City (E. Vidal, Z.R. Martinez), Angeles Lomas Hospital, Mexico (J.P. Vazquez Mathieu, C. Abascal Caloca), Hospital Angeles Lindavista, Mexico City (C.A. Aguirre Serrato, E. Vidal), Centro Medico ISSEMYM, Metepec (E. Vidal).

The Netherlands: Martini Hospital, Groningen (B. Loef).

New Zealand: Auckland City Hospital, Auckland (R. Parke, C. Simmonds, L. Newby), Middlemore Hospital, Auckland (J. Tai), Christchurch Hospital New Zealand, Christchurch (J. Mehrtens), Waikato Hospital, Hamilton (M. La Pine), Palmerston North Hospital, Palmerston North (G. Cloughley).

Paraguay: Hospital de Clinicas, Asuncion (N. Rivas).

Peru: Hospital San Gabriel, Lima (I. Ramos Palomino).

Portugal: Hospital Garcia de Orta, Almada (S. Lanca), Instituto Portugues de Oncologia, Lisboa (M.J. Bouw), Santo Antonio Hospital, Porto (C. Teixeira, S. Fontes Ribeiro).

Russia: Center for Cardiovascular Surgery, Moscow (M. Yaroustovsky).

Serbia: Institute for Pulmonary Diseases, Sremska Kamenica (U. Batranovic).

South Korea: Konkuk University Hospital, Seoul (K.-M. Lee).

Spain: General Hospital of Castellon, Castellon (S. Mas, S. Altaba), Complejo Universitario de Leon, Leon (I. Gonzalez), ICU, Hospital Universitario de Malaga, Malaga (M.E. Herrera-Gutierrez, R. Olalla-Sanchez, G. Sellez-Perez, L. Chimali-Cobano), Clinica Universidad Navarra, Pamplona (A. Ferrer-Nadal, P. Monedero, J.R. Pérez-Valdivieso), Hospital Virgen Del Camino, Pamplona (M. Garcia-Montesinos), Hospital Universitario De Valme, Sevilla (D. Herrera), Hospital Torrevieja, Torrevieja (E. Herrero), Hospital Xeral de Vigo, Vigo (J.C. Diz, B.M. Jimenez).

Switzerland: Klinik Im Park Hirslanden, Zürich (T. Gaspert).

Tunisia: Abderrahmane Mami, Ariana (M. Besbes).

Turkey: University of Kocaeli, Kocaeli (N. Baykara).

Ukraine: Institute of Nephrology National Academy of Medical Sciences of Ukraine, Kyiv (M. Kolesnyk).

UK: UCL Center for Nephrology, Royal Free Hospital, London (A. Davenport), Guy’s & St Thomas Foundation Hospital, London (M. Ostermann), Western Sussex Hospital Trust, Worthing (Y. Syed, L. Forni).

USA: Albany Medical Center Hospital, Albany (J. Cerda), Cleveland Clinic, Cleveland (S. Demirjian), Geisinger Medical Center, Danville (M. Craft), UPMC McKeesport, McKeesport (A. Uppalapati), Bruce W. Carter Department of Critical Care Medicine, Miami (C. Cely), Ruby Memorial Hospital, Morgantown (R. Schmidt), Hospital of the University of Pennsylvania, Philadelphia (K. Markelz, M. Shashaty), University of Pittsburgh Presbyterian, Pittsburgh (N. Kannan, J Kellum), Mayo Clinic, Rochester (M. Selby, K. Banaei-Kashani, J. Steuernagle), Strong Memorial Hospital, Rochester (D. Kaufman), University of California San Francisco Moffitt-Long Hospital, San Francisco (K. Kordesch, K. Liu).

Conflicts of interest

EH received speakers fee from Astute Medical, and an Industrial Research Fund (IOF) from Ghent University for a validation study on a biomarker for AKI. SMB has consulted for and received honoraria from Gambro-Baxter. CG has received sponsorship for an academic conference from Gambro. OJB has received grants and non-financial support from Gambro, BBraun, Fresenius, and Astute Medical. MJ has consulted for and received honoraria from Baxter, Gambro, Fresenius, CLS Behring, BBraun, AM Pharma, Sanofi, Astute. JK has received grant support and/or consulting fees from Fresenius, Gambro, Baxter, Astute Medical, Alere, AM Pharma, Spectral, Grifols, Cytosorbents, Alung, Atox Bio, Bard, Kaneka. RM has received grants from the International Safety Adverse Events Consortium and Thrasos, he has options in Astute Medical and served in the scientific advisory board for trials for Abbvie, AM Pharma, and Eli Lilly, and consulted for CSL Behring, GSK, Baxter, Sova, Astellas, and Sanofi-Aventis. PP has consulted for Complexa Inc. SW is Director and Shareholder of Aalix Healthcare Services Consulting, which has provided services related to AKI.

Supplementary material

134_2015_3934_MOESM1_ESM.pdf (408 kb)
Supplementary material (PDF 408 kb)


  1. 1.
    Lameire NH, Bagga A, Cruz D, De Maeseneer J, Endre Z, Kellum JA, Liu KD, Mehta RL, Pannu N, Van Biesen W, Vanholder R (2013) Acute kidney injury: an increasing global concern. Lancet 382:170–179PubMedCrossRefGoogle Scholar
  2. 2.
    Siew ED, Davenport A (2015) The growth of acute kidney injury: a rising tide or just closer attention to detail? Kidney Int 87:46–61PubMedCentralPubMedCrossRefGoogle Scholar
  3. 3.
    Bellomo R, Kellum JA, Ronco C (2012) Acute kidney injury. Lancet 380:756–766PubMedCrossRefGoogle Scholar
  4. 4.
    Hoste EA, Schurgers M (2008) Epidemiology of acute kidney injury: how big is the problem? Crit Care Med 36:S146–S151PubMedCrossRefGoogle Scholar
  5. 5.
    Murugan R, Kellum JA (2011) Acute kidney injury: what’s the prognosis. Nat Rev Nephrol 7:209–217PubMedCentralPubMedCrossRefGoogle Scholar
  6. 6.
    Chawla LS, Eggers PW, Star RA, Kimmel PL (2014) Acute kidney injury and chronic kidney disease as interconnected syndromes. N Engl J Med 371:58–66PubMedCrossRefGoogle Scholar
  7. 7.
    Bellomo R, Ronco C, Kellum JA, Mehta RL, Palevsky P (2004) 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 8:R204–R212PubMedCentralPubMedCrossRefGoogle Scholar
  8. 8.
    Mehta RL, Kellum JA, Shah SV, Molitoris BA, Ronco C, Warnock DG, Levin A (2007) Acute Kidney Injury Network: report of an initiative to improve outcomes in acute kidney injury. Crit Care 11:R31PubMedCentralPubMedCrossRefGoogle Scholar
  9. 9.
    Kidney Disease: Improving Global Outcomes (KDIGO) Acute Kidney Injury Work Group (2012) KDIGO clinical practice guideline for acute kidney injury. Kidney Int 2:1–138CrossRefGoogle Scholar
  10. 10.
    Grams ME, Waikar SS, Macmahon B, Whelton S, Ballew SH, Coresh J (2014) Performance and limitations of administrative data in the identification of AKI. Clin J Am Soc Nephrol 9:682–689PubMedCentralPubMedCrossRefGoogle Scholar
  11. 11.
    von Elm E, Altman DG, Egger M, Pocock SJ, Gotzsche PC, Vandenbroucke JP (2007) The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Lancet 370:1453–1457CrossRefGoogle Scholar
  12. 12.
    Hoste EA, Clermont G, Kersten A, Venkataraman R, Angus DC, De Bacquer D, Kellum JA (2006) RIFLE criteria for acute kidney injury are associated with hospital mortality in critically ill patients: a cohort analysis. Crit Care 10:R73PubMedCentralPubMedCrossRefGoogle Scholar
  13. 13.
    Moreno RP, Metnitz PG, Almeida E, Jordan B, Bauer P, Campos RA, Iapichino G, Edbrooke D, Capuzzo M, Le Gall JR (2005) SAPS 3—from evaluation of the patient to evaluation of the intensive care unit. Part 2: development of a prognostic model for hospital mortality at ICU admission. Intensive Care Med 31:1345–1355PubMedCentralPubMedCrossRefGoogle Scholar
  14. 14.
    Vincent JL, Moreno R, Takala J, Willatts S, De Mendonca A, Bruining H, Reinhart CK, Suter PM, Thijs LG (1996) 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 Med 22:707–710PubMedCrossRefGoogle Scholar
  15. 15.
    Levey AS, Stevens LA, Schmid CH, Zhang YL, Castro AF, Feldman HI, Kusek JW, Eggers P, Van Lente F, Greene T, Coresh J (2009) A new equation to estimate glomerular filtration rate. Ann Intern Med 150:604–612PubMedCentralPubMedCrossRefGoogle Scholar
  16. 16.
    World Health Organisation (2014) Global health expenditure database: quick reports: table of key indicators for all the Member States. Accessed 28 Apr 2014
  17. 17.
    The World Bank (2014) Countries and economies. Accessed 28 Apr 2014
  18. 18.
    United Nations (2013) Composition of macro geographical (continental) regions, geographical sub-regions, and selected economic and other groupings. Accessed 28 Apr 2014
  19. 19.
    Altman DG, Machin D, Bryant TN, Gardner MJ (2000) Statistics with confidence. Confidence intervals and statistical guidelines, 2nd edn. BMJ Books, BristolGoogle Scholar
  20. 20.
    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 (2005) Acute renal failure in critically ill patients: a multinational, multicenter study. JAMA 294:813–818PubMedCrossRefGoogle Scholar
  21. 21.
    NCEPOD (2009) Acute kidney injury: adding insult to injury (2009). Accessed 2 Sept 2014
  22. 22.
    Mandelbaum T, Scott DJ, Lee J, Mark RG, Malhotra A, Waikar SS, Howell MD, Talmor D (2011) Outcome of critically ill patients with acute kidney injury using the Acute Kidney Injury Network criteria. Crit Care Med 39:2659–2664PubMedCentralPubMedGoogle Scholar
  23. 23.
    Nisula S, Kaukonen KM, Vaara ST, Korhonen AM, Poukkanen M, Karlsson S, Haapio M, Inkinen O, Parviainen I, Suojaranta-Ylinen R, Laurila JJ, Tenhunen J, Reinikainen M, Ala-Kokko T, Ruokonen E, Kuitunen A, Pettila V (2013) Incidence, risk factors and 90-day mortality of patients with acute kidney injury in Finnish intensive care units: the FINNAKI study. Intensive Care Med 39:420–428PubMedCrossRefGoogle Scholar
  24. 24.
    Hoste EA, De Corte W (2011) Clinical consequences of acute kidney injury. Contrib Nephrol 174:56–64PubMedCrossRefGoogle Scholar
  25. 25.
    Singbartl K, Kellum JA (2012) AKI in the ICU: definition, epidemiology, risk stratification, and outcomes. Kidney Int 81:819–825PubMedCrossRefGoogle Scholar
  26. 26.
    Vaara ST, Korhonen A-M, Kaukonen K-M, Nisula S, Inkinen O, Hoppu S, Laurila JJ, Mildh L, Reinikainen M, Lund V, Parviainen I, Pettila V, Finnaki SG (2012) Fluid overload is associated with an increased risk for 90-day mortality in critically ill patients with renal replacement therapy: data from the prospective FINNAKI study. Crit Care 16:R197PubMedCentralPubMedCrossRefGoogle Scholar
  27. 27.
    Kellum JA, Song M, Li J (2004) Science review: extracellular acidosis and the immune response: clinical and physiologic implications. Crit Care 8:331–336PubMedCentralPubMedCrossRefGoogle Scholar
  28. 28.
    Uchino S, Bellomo R, Ronco C (2001) Intermittent versus continuous renal replacement therapy in the ICU: impact on electrolyte and acid–base balance. Intensive Care Med 27:1037–1043PubMedCrossRefGoogle Scholar
  29. 29.
    Li X, Hassoun HT, Santora R, Rabb H (2009) Organ crosstalk: the role of the kidney. Curr Opin Crit Care 15:481–487PubMedCrossRefGoogle Scholar
  30. 30.
    Hoste EA, Blot SI, Lameire NH, Vanholder RC, De Bacquer D, Colardyn FA (2004) Effect of nosocomial bloodstream infection on the outcome of critically ill patients with acute renal failure treated with renal replacement therapy. J Am Soc Nephrol 15:454–462PubMedCrossRefGoogle Scholar
  31. 31.
    Reynvoet E, Vandijck DM, Blot SI, Dhondt AW, De Waele JJ, Claus S, Buyle FM, Vanholder RC, Hoste EA (2009) Epidemiology of infection in critically ill patients with acute renal failure. Crit Care Med 37:2203–2209PubMedCrossRefGoogle Scholar
  32. 32.
    Mehta RL, Bouchard J, Soroko SB, Ikizler TA, Paganini EP, Chertow GM, Himmelfarb J (2011) Sepsis as a cause and consequence of acute kidney injury: Program to Improve Care in Acute Renal Disease. Intensive Care Med 37:241–248PubMedCentralPubMedCrossRefGoogle Scholar
  33. 33.
    Elseviers MM, Lins RL, Van der Niepen P, Hoste E, Malbrain ML, Damas P, Devriendt J, Sharf Investigators (2010) Renal replacement therapy is an independent risk factor for mortality in critically ill patients with acute kidney injury. Crit Care 14:R221PubMedCentralPubMedCrossRefGoogle Scholar
  34. 34.
    Schneider AG, Uchino S, Bellomo R (2012) Severe acute kidney injury not treated with renal replacement therapy: characteristics and outcome. Nephrol Dial Transplant 27:947–952PubMedCrossRefGoogle Scholar
  35. 35.
    Udy AA, Roberts JA, Lipman J (2013) Clinical implications of antibiotic pharmacokinetic principles in the critically ill. Intensive Care Med 39:2070–2082PubMedCrossRefGoogle Scholar
  36. 36.
    De Waele JJ, Lipman J, Akova M, Bassetti M, Dimopoulos G, Kaukonen M, Koulenti D, Martin C, Montravers P, Rello J, Rhodes A, Udy AA, Starr T, Wallis SC, Roberts JA (2014) Risk factors for target non-attainment during empirical treatment with beta-lactam antibiotics in critically ill patients. Intensive Care Med 40:1340–1351PubMedCrossRefGoogle Scholar
  37. 37.
    Siew ED, Matheny ME, Ikizler TA, Lewis JB, Miller RA, Waitman LR, Go AS, Parikh CR, Peterson JF (2010) Commonly used surrogates for baseline renal function affect the classification and prognosis of acute kidney injury. Kidney Int 77:536–542PubMedCentralPubMedCrossRefGoogle Scholar
  38. 38.
    Zavada J, Hoste E, Cartin-Ceba R, Calzavacca P, Gajic O, Clermont G, Bellomo R, Kellum JA (2010) A comparison of three methods to estimate baseline creatinine for RIFLE classification. Nephrol Dial Transplant 25:3911–3918PubMedCrossRefGoogle Scholar
  39. 39.
    Carlier M, Dumoulin A, Janssen A, Picavet S, Vanthuyne S, Van Eynde R, Vanholder R, Delanghe J, De Schoenmakere G, De Waele JJ, Hoste EA (2015) Comparison of different equations to assess glomerular filtration in critically ill patients. Intensive Care Med 41:427–435PubMedCrossRefGoogle Scholar
  40. 40.
    Schetz M, Gunst J, Van den Berghe G (2014) The impact of using estimated GFR versus creatinine clearance on the evaluation of recovery from acute kidney injury in the ICU. Intensive Care Med 40:1709–1717PubMedCrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg and ESICM 2015

Authors and Affiliations

  • Eric A. J. Hoste
    • 1
    • 2
    • 3
  • Sean M. Bagshaw
    • 4
  • Rinaldo Bellomo
    • 5
  • Cynthia M. Cely
    • 6
  • Roos Colman
    • 7
  • Dinna N. Cruz
    • 8
  • Kyriakos Edipidis
    • 9
  • Lui G. Forni
    • 10
  • Charles D. Gomersall
    • 11
  • Deepak Govil
    • 12
  • Patrick M. Honoré
    • 13
  • Olivier Joannes-Boyau
    • 14
  • Michael Joannidis
    • 15
  • Anna-Maija Korhonen
    • 16
    • 17
  • Athina Lavrentieva
    • 18
  • Ravindra L. Mehta
    • 19
  • Paul Palevsky
    • 20
    • 21
    • 22
  • Eric Roessler
    • 23
  • Claudio Ronco
    • 24
  • Shigehiko Uchino
    • 25
  • Jorge A. Vazquez
    • 26
  • Erick Vidal Andrade
    • 27
  • Steve Webb
    • 28
  • John A. Kellum
    • 3
    • 22
  1. 1.Department of Intensive Care Medicine, Ghent University HospitalGhent UniversityGhentBelgium
  2. 2.Research Foundation-FlandersBrusselsBelgium
  3. 3.The Clinical Research, Investigation, and Systems Modelling of Acute Illness (CRISMA) Laboratory, Department of Critical Care MedicineUniversity of Pittsburgh, School of MedicinePittsburghUSA
  4. 4.Division of Critical Care Medicine, Faculty of Medicine and DentistryUniversity of AlbertaEdmontonCanada
  5. 5.Department of Intensive CareAustin HospitalMelbourneAustralia
  6. 6.Division of Pulmonary, Critical Care, and Sleep MedicineUniversity of Miami Miller School of MedicineMiamiUSA
  7. 7.Department of Public HealthGhent UniversityGhentBelgium
  8. 8.Division of Nephrology-Hypertension, Department of MedicineUniversity of California San DiegoSan DiegoUSA
  9. 9.Hygeia Medical CenterAthensGreece
  10. 10.Department of Intensive Care Medicine and Surrey Peri-operative Anaesthesia Critical Care Research Group (SPACeR), Royal Surrey County Hospital NHS Foundation Trust, Faculty of Health and Medical ScienceUniversity of SurreyGuildfordUK
  11. 11.Department of Anaesthesia and Intensive Care, Prince of Wales HospitalThe Chinese University of Hong KongHong KongChina
  12. 12.Institute of Critical Care and AnesthesiaMedanta-The MedicityGurgaonIndia
  13. 13.Intensive Care Department, Universitair Ziekenhuis BrusselVUB UniversityBrusselsBelgium
  14. 14.Service d’Anesthésie-Réanimation 2Centre Hospitalier Universitaire (CHU) de BordeauxBordeauxFrance
  15. 15.Division of Intensive Care and Emergency Medicine, Department of Internal MedicineMedical University InnsbruckInnsbruckAustria
  16. 16.Intensive Care Unit, Division of Anaesthesia and Intensive Care Medicine, Department of SurgeryMeilahti University Hospital Central HospitalHelsinkiFinland
  17. 17.Department of Clinical SciencesUniversity of HelsinkiHelsinkiFinland
  18. 18.Burn ICUPapanikolaou General HospitalThessalonikiGreece
  19. 19.Department of Medicine, UCSD Medical CenterUniversity of California San DiegoSan DiegoUSA
  20. 20.Renal SectionVA Pittsburgh Healthcare SystemPittsburghUSA
  21. 21.Renal-Electrolyte Division, Department of MedicineUniversity of Pittsburgh School of MedicinePittsburghUSA
  22. 22.Center for Critical Care Nephrology, Department of Critical Care MedicineUniversity of Pittsburgh, School of MedicinePittsburghUSA
  23. 23.Department of Nephrology, Faculty of MedicinePontificia Universidad Catolica de ChileSantiagoChile
  24. 24.Department of Nephrology, Dialysis, and Transplantation, International Renal Research InstituteSan Bortolo HospitalVicenzaItaly
  25. 25.Intensive Care Unit, Department of AnesthesiologyThe Jikei University School of MedicineTokyoJapan
  26. 26.Department of Critical Care MedicineClinica Modelo de LanusBuenos AiresArgentina
  27. 27.Department of Critical Care MedicineHospital Angeles LomasMexico CityMexico
  28. 28.Department of Critical Care MedicineUniversity of Western Australia and Royal Perth HospitalPerthAustralia

Personalised recommendations