Journal of Critical Care 30 (2015) 998–1002

Contents lists available at ScienceDirect

Journal of Critical Care

journal homepage: www.jccjournal.org

Sepsis may not be a risk factor for mortality in patients with acute kidney
injury treated with continuous renal replacement therapy
Isao Nagata, MD, MPH a,⁎, Shigehiko Uchino, MD b, Natsuko Tokuhira, MD c, Tetsu Ohnuma, MD d,
Yoshitomo Namba, MD e, Shinshu Katayama, MD f, Hiroo Kawarazaki, MD g, Noriyoshi Toki, MD h,
Kenta Takeda, MD i, Hideto Yasuda, MD j, Junichi Izawa, MD b,
Makiko Uji, MD kfor JSEPTIC (Japanese Society for Physicians Trainees in Intensive Care) Clinical Trial Group
a
Department of Emergency, Kanto Rosai Hospital, Kawasaki-shi, Kanagawa, Japan 211-8510
b
Intensive Care Unit, Department of Anesthesiology, Jikei University School of Medicine, Tokyo, Japan 105-8471
c
Division of Intensive Care, University Hospital, Kyoto Prefectural University of Medicine, Kyoto, Japan 602-8566
d
Intensive Care Unit, Department of Anesthesiology, Saitama Medical Center, Jichi Medical University, Saitama-shi, Saitama, Japan 330-8503
e
Department of Emergency and Critical Care, Showa University Fujigaoka Hospital, Yokohama-shi, Kanagawa, Japan 227-0043
f
Department of Emergency Medicine, Asahi General Hospital, Asahi-shi, Chiba, Japan 289-2511
g
Department of Nephrology and Hypertension, St Marianna University School of Medicine, Kawasaki-shi, Kanagawa, Japan 216-8511
h
Department of Internal Medicine, Tokyo Metropolitan Tama Medical Center, Tokyo, Japan 183-0042
i
Division of Intensive Care Medicine, Hyogo College of Medicine, Nishinomiya-shi, Hyogo, Japan 663-8501
j
Intensive Care Unit, Department of Emergency and Critical Care Medicine, Japanese Red Cross Musashino Hospital, Musashino-shi, Tokyo, Japan 180-8610
k
Intensive Care Unit, Osaka University Hospital, Suita-shi, Osaka, Japan 565-0871

a r t i c l e i n f o a b s t r a c t

Keywords: Purpose: We aimed to study the clinical characteristics, courses, and outcomes of critically ill patients with septic
Acute kidney injury acute kidney injury (AKI) treated with continuous renal replacement therapy (CRRT) in comparison with
Continuous renal replacement therapy nonseptic AKI treated with CRRT.
Sepsis Methods: This is a multicenter retrospective observational study conducted in 14 Japanese intensive care units in
Mortality
2010. All adult patients with severe AKI treated with CRRT were eligible (n = 343), and information on patient
characteristics, variables at CRRT initiation, CRRT settings, and outcomes was collected. Patients were categorized
into the septic AKI group and the nonseptic AKI group according to contributing factors to AKI.
Results: Approximately half of study patients (48.7%) had sepsis/septic shock as a contributing factor to AKI, and
patients with septic AKI treated with CRRT had more serious clinical conditions than patients with nonseptic AKI.
However, no significant difference was observed in intensive care unit mortality (48.5% vs 43.8%; P = .44) and
hospital mortality (61.1% vs 56.3%; P = .42) between patients with septic and nonseptic AKIs treated with
CRRT. Furthermore, sepsis was associated with lower hospital mortality (odds ratio, 0.378; P = .012) in multivar-
iable regression analysis.
Conclusion: Sepsis may not be a risk factor for mortality in patients with AKI whose condition has become severe
enough to require CRRT.
© 2015 Elsevier Inc. All rights reserved.

1. Introduction
Acute kidney injury (AKI) occurs commonly in intensive care units
(ICUs), and the development of AKI increases morbidity and mortality
in critically ill patients [1-5]. Acute kidney injury treated with renal
replacement therapy (RRT) represents the most severe form of AKI and
⁎ Corresponding author. Tel.: +81 45 628 6100; fax: +81 45 628 6101.
E-mail addresses: issaisao@gmail.com (I. Nagata), s.uchino@mac.com (S. Uchino),
natsueteto@yahoo.co.jp (N. Tokuhira), tetsu.fe3@gmail.com (T. Ohnuma),
irowa_nioedo@yahoo.co.jp (Y. Namba), shinsyu_k@hotmail.com (S. Katayama),
hirookawarazaki@yahoo.co.jp (H. Kawarazaki), vandervaart01jp@yahoo.co.jp (N. Toki),
takeda74@hyo-med.ac.jp (K. Takeda), yasudahideto@me.com (H. Yasuda),
jizawa13@gmail.com (J. Izawa), makiko.uj@gmail.com (M. Uji).
has been reported to occur in 15% to 73% of patients with AKI [6-12].
Sepsis is the leading cause of AKI treated with RRT in critically ill patients,
and the frequency has been described to range from 34% to 57% [13-17].
Previous data suggest that septic AKI may be characterized by a dis-
tinct pathophysiology [18-20]. Therefore, septic AKI may be associated
with differences in patient characteristics, response to interventions,
and clinical outcomes compared with nonseptic AKI. Several multicen-
ter observational studies have shown that septic AKI has a higher bur-
den of illness, greater abnormalities in vital signs and blood chemistry
analysis, and worse outcome compared with nonseptic AKI [6,10,21-
23]. However, although continuous RRT (CRRT) has become a common
treatment for severe AKI over the last 2 decades [13,24,25], information
on septic AKI treated with CRRT is limited. Because the hospital

http://dx.doi.org/10.1016/j.jcrc.2015.06.021
0883-9441/© 2015 Elsevier Inc. All rights reserved.
 I. Nagata et al. / Journal of Critical Care 30 (2015) 998–1002 999

mortality of patients with severe AKI treated with CRRT is very high, regression analysis was conducted to investigate the impact of sepsis on
that is, more than 50% [13,26,27], the causative factors of AKI may hospital mortality. The following variables were applied to multivariable
have a smaller impact, and the hospital mortality of septic AKI patients logistic regression analysis using a backward elimination approach if the
treated with CRRT may be comparable with that of nonseptic AKI. univariate P value comparing survivors and nonsurvivors was less than
The Japanese Society of Education for Physicians and Trainees in In- .15: contributing factors to AKI (sepsis/septic shock, cardiogenic shock,
tensive Care CRRT database is a multicenter retrospective study that major surgery, hypovolemia, drugs, hepatorenal syndrome, and others),
aims to understand multiple aspects of CRRT [28-30]. As part of the larg- age, sex, weight, SAPS II, chronic kidney disease, postoperative admission,
er study, which was conducted in 14 ICUs in 12 centers in Japan, we duration between hospital admission and ICU admission, vasopressor,
aimed to describe the clinical characteristics and courses and outcomes MAP, mechanical ventilation, PaO2/FIO2 ratio, bicarbonate, lactate, GCS,
of critically ill patients with septic AKI treated with CRRT in comparison platelets, bilirubin, albumin, prothrombin time–international normalized
with patients with nonseptic AKI. ratio, creatinine and BUN at CRRT initiation, urine output, duration be-
tween ICU admission and CRRT initiation, blood flow, and CRRT intensity.
2. Methods As a sensitivity analysis, first, multivariable logistic regression analysis
was repeated for only patients treated with vasopressors at CRRT initia-
2.1. Study population tion, and, second, multivariable logistic regression analysis with cardio-
genic shock forced to remain in addition to sepsis/septic shock was
All patients who were admitted to one of the participating ICUs from repeated. Data are presented as odds ratios with 95% confidence intervals.
January 1, 2010, to December 31, 2010, were retrospectively screened. Pa- All analyses were performed using SAS 9.3 (SAS Institute, Inc, Cary, NC).
tients who were included in the study were 18 years or older and were
treated with CRRT for AKI according to the Risk, Injury, Failure, Loss, End- 3. Results
stage kidney disease (RIFLE) criteria [31]. Patients were excluded if they
were treated with any RRT before admission to the ICU or had end-stage Of 343 patients included in the database, 167 patients (48.7%) had
renal failure. If a patient was admitted to the ICU and was treated with sepsis/septic shock as at least 1 of contributing factors to AKI and were
CRRT more than once during the same hospital admission, only the first categorized as the septic AKI group; the rest were categorized as the
ICU admission was included. The study protocol was reviewed by the nonseptic AKI group.
ethics committee or the investigational review board of each participating Table 1 shows the demographics of study patients. Compared with the
center. Because of the anonymous and retrospective nature of this study, nonseptic AKI group, patients in the septic AKI group had a higher SAPS II
ethical committees of all centers waived the need for informed consent. score (59.6 vs 52.4; P b .001) and a lower postoperative ICU admission rate
(21.6% vs 39.8%; P b .001). The most common contributing factor to AKI in
2.2. Data collection the nonseptic AKI group was cardiogenic shock (41.5%), followed by
major surgery (35.8%) and hypovolemia (31.8%). The primary sources of
A case report form was developed for the purpose of the study, in- sepsis in patients with septic AKI are shown in Table 2; the most common
cluding the following information: age, sex, body weight (measured or was intrathoracic sites (30.0%), followed by intraabdominal sites (28.7%).
estimated at ICU admission), date of hospital admission, premorbid cre- Variables at CRRT initiation and outcomes are shown in Table 3. Va-
atinine, date of ICU admission, the Simplified Acute Physiology Score sopressor requirement was more frequent (80.2% vs 67.1%; P = .0082),
(SAPS) II [32] on the day of ICU admission, and primary diagnosis. Con- PaO2/FIO2 ratio was lower (181 vs 230 Torr; P b .001), and GCS was lower
tributing factors to AKI were identified from a list of possible choices (13.5 vs 14.0; P = .027) in the septic AKI group compared with the
(sepsis/septic shock, cardiogenic shock, drug-induced, hypovolemia, nonseptic AKI group. Furthermore, BUN at CRRT initiation was higher
major surgery, hepatorenal syndrome, and others) according to the (54.0 vs 42.5 mg/dL; P b .001) and urine output was lower (0.29 vs
judgment of the treating clinician. More than 1 contributing factor 0.37 mL/kg per hour; P = .044) in the septic AKI group compared
could be selected in each case. Sepsis was diagnosed by treating clini- with the nonseptic AKI group. During CRRT, CRRT dose was statistically
cian based on the published consensus criteria [33]. At the initiation of different (P = .0041) between the 2 groups, and even after adjusted for
CRRT, the following information was collected: use of vasopressors the body weight, CRRT intensity was still statistically different (15.1 vs
and mechanical ventilation, mean arterial pressure (MAP), PaO2/fraction 13.3 mL/kg per hour; P b .001). Although both the patients' general con-
of inspired oxygen (FIO2) ratio, lactate, Glasgow Coma Scale (GCS), dition and severity of AKI in the septic AKI group tended to be more crit-
platelet count, bilirubin, diuretics use, urine output, creatinine, and ical than those in the nonseptic AKI group, ICU and hospital mortality
blood urea nitrogen (BUN). The following information on CRRT was was not statistically different between the 2 groups.
also collected: date and time of CRRT initiation and discontinuation,
blood flow, prescribed dialysate, and replacement flow rate (milliliters Table 1
per hour). The CRRT dose was defined as the sum of the prescribed dial- Demographics of study patients with AKI treated with CRRT
ysate and replacement flow rate (milliliters per hour), and intensity of Septic Nonseptic P
CRRT was defined as the CRRT dose divided by body weight (milliliters
No. of patients 167 (48.7%) 176 (51.3%)
per kilogram per hour). Information on ICU and hospital mortality and Age (y), mean (SD) 65.6 (14.9) 66.8 (15.4) .50
RRT requirement at hospital discharge was also collected. Male sex (%) 67.1 64.8 .74
Weight (kg), mean (SD) 59.1 (14.0) 62.2 (16.9) .065
2.3. Statistical analysis SAPS II score, mean (SD) 59.7 (19.7) 52.4 (20.5) b.001
Chronic kidney disease 35.9% 47.6% .081
Postoperative admission 21.6% 39.8% b.001
Patients were categorized into a septic AKI group and a nonseptic Hospital to ICU (d), median (IQR) 0.95 (0.67-10.4) 1.11 (0.68-5.96) .76
AKI group according to contributing factors to AKI. Normally or near Contributing factors to AKI
normally distributed continuous variables are presented as means and Sepsis/septic shock 167 (100%) –
Cardiogenic shock 19 (11.4%) 73 (41.5%)
SD and were compared using the t test. Nonnormally distributed contin-
Hypovolemia 24 (14.4%) 56 (31.8%)
uous data are presented as medians and interquartile ranges (IQRs) and Major surgery 15 (9.0%) 63 (35.8%)
were compared using the Wilcoxon rank sum test. Categorical data Drugs 6 (3.6%) 13 (7.4%)
were summarized as percentages with or without counts and were Hepatorenal syndrome 3 (1.8%) 4 (2.3%)
compared using the Fisher exact test or χ2 test. P b .05 was considered Others 6 (3.6%) 39 (22.2%)

statistically significant for all comparisons. Multivariable logistic Hospital to ICU indicates duration between hospital admission and ICU admission.
1000 I. Nagata et al. / Journal of Critical Care 30 (2015) 998–1002

Table 2
Primary source of sepsis in patients with septic AKI treated with CRRT

Source of sepsis n (%)

Intrathoracic 50 (30.0%)
Intraabdominal 48 (28.7%)
Urogenital 17 (10.2%)
Endovascular 16 (9.6%)
Skin/soft tissue/bone/joint 11 (6.6%)
Hematologic/immunocompromised 4 (2.4%)
Central nervous system 3 (1.8%)
Pelvis 1 (0.6%)
Source unknown/not specified 17 (10.2%)

Intrathoracic included pneumonia, lung abscess, mediastinitis, pleuritis, empyema, and tu-
berculosis; intraabdominal included biliary disease, perforation of the digestive tract, peri-
tonitis, ischemic colon disease, enterocolitis, and pancreatitis; urogenital included urinary
tract infection, pyelonephritis, and pyonephrosis; endovascular included bacteremia, in-
fective endocarditis, catheter-associated bloodstream infection, and infected aneurysm; Fig. 1. Hospital mortality of each contributing factor to AKI in patients with AKI. Because
skin/soft tissue/bone/joint included cellulitis, necrotizing fasciitis, suppurative arthritis, some patients had more than 1 contributing factor, the total number of patients in this fig-
gas gangrene, and decubitus; hematologic/immunocompromised included neutropenia ure is more than the actual number of the study patients (n = 343).
and post–bone marrow transplant; central nervous system included meningitis, and epi-
dural abscess, and encephalitis.

4. Discussion

The hospital mortality of each contributing factor is shown in Fig. 1.
Patients who had cardiogenic shock (69.6%) and hepatorenal syndrome
(71.4%) had higher hospital mortality than those who had sepsis/septic
shock (61.1%).
The results of multivariable logistic regression analysis for hospital
mortality are shown in Table 4. Sepsis/septic shock was found to be as-
sociated with lower hospital mortality (odds ratio, 0.378; P = .012).
Other variables independently associated with hospital mortality were
SAPS II, bilirubin, and postoperative admission. As a sensitivity analysis,
multivariable logistic regression was repeated for patients treated with
vasopressors at CRRT initiation (n = 252). Although not statistically sig-
nificant, the odds ratio for septic shock was still lower than 1 (0.59; P =
.12). In addition, multivariable logistic regression analysis with cardio-
genic shock forced to remain in addition to sepsis/septic shock was re-
peated. Although the odds ratio for cardiogenic shock was more than
1, it was not statistically significant (1.80; P = .07), and the odds ratio
for sepsis/septic shock was still lower than 1, although not statistically
significant (0.82; P = .50).
4.1. Key findings
In this study, we found that approximately half (48.7%) of patients
with AKI treated with CRRT had sepsis and that patients with septic
AKI treated with CRRT had more serious clinical conditions than pa-
tients with nonseptic AKI (higher SAPS II and vasopressor requirement
and greater aberrations in vital signs and blood chemistry analysis).
However, among the patients with AKI treated with CRRT, no significant
difference was observed in patient outcome between septic and
nonseptic AKIs treated with CRRT, and contrary to expectations, sepsis
was found to have a significantly low odds ratio of 0.378 (P = .012)
for hospital mortality in multivariable regression analysis.
4.2. Relationship to previous studies
Similar to our results, several multicenter observational studies have
reported that septic AKI had greater acuity of illness [6,10,21-23]. These
studies also found that septic AKI was associated with significantly

Table 3
Variables at CRRT initiation and outcomes in patients with AKI treated with CRRT

Septic Nonseptic P

Vasopressor (%) 80.2 67.1 .0082

MAP (mm Hg), mean (SD) 71.5 (14.5) 72.8 (14.7) .42
Mechanical ventilation 83.8% 81.3% .63
PaO2/FIO2 ratio 181 (104-281) 230 (145-338) b.001
Bicarbonate (mmol/L), mean (SD) 19.6 (6.1) 20.7 (5.7) .097
Lactate (mmol/L), median (IQR) 2.75 (1.50-5.45) 2.78 (1.60-6.81) .53
GCS, median (IQR) 13.5 (9.0-15.0) 14.0 (9.5-15.0) .027
Platelets (103/μL), median (IQR) 79.0 (47.0-152) 93.0 (59.0-151) .077
Bilirubin (mg/dL), median (IQR) 1.10 (0.60-2.40) 1.10 (0.60-2.40) .78
Albumin (g/dL), mean (SD) 2.38 (0.67) 2.66 (0.67) b.001
PT-INR, median (IQR) 1.42 (1.18-1.85) 1.32 (1.16-1.64) .031
Creatinine at CRRT initiation (mg/dL), median (IQR) 2.85 (1.93-4.00) 2.73 (1.82-3.70) .29
BUN at CRRT initiation (mg/dL), median (IQR) 54.0 (39.9-81.0) 42.5 (30.0-64.0) b.001
Urine output (mL/kg per hour), median (IQR) 0.29 (0.11-0.58) 0.37 (0.13-0.93) .044
Diuretic therapy 35.9% 49.4% .016
ICU to CRRT start (d), median (IQR) 0.71 (0.18-1.98) 0.80 (0.16-1.69) .66
Blood flow (mL/min), median (IQR) 100 (80-100) 100 (80-100) .18
CRRT dose (mL/h), median (IQR) 800 (700-1000) 800 (600-1000) .0041
Intensity (mL/kg per hour), median (IQR) 15.1 (12.0-19.4) 13.3 (10.3-16.7) b.001
CRRT duration (d), median (IQR) 2.87 (1.72-5.59) 3.13 (1.63-5.73) .99
ICU mortality 48.5% 43.8% .44
Hospital mortality 61.1% 56.3% .42
RRT at hospital discharge among survivors 7.8% 14.1% .36

PT-INR indicates prothrombin time–international normalized ratio; ICU to CRRT start, duration between ICU admission to CRRT initiation.
 I. Nagata et al. / Journal of Critical Care 30 (2015) 998–1002
Table 4
Multivariable logistic regression analysis for hospital mortality in patients with AKI treated
with CRRT
Odds ratio (95% CI)
SAPS II (point) 1.026 (1.003-1.049)
Bilirubin (mg/dL) 1.149 (1.011-1.306)
PaO2/FIO2 ratio 0.997 (0.994-1.000)
Postoperative admission (%) 0.411 (0.180-0.939)
Sepsis/septic shock 0.378 (0.177-0.810)
CI indicates confidence interval.
higher hospital mortality compared with nonseptic AKI. For example,
Bagshaw et al [6] showed that the hospital mortality of patients with
septic AKI was 70.2%, which was higher than that of patients with
nonseptic AKI (51.8%; P b .001). Vaara et al [22] also reported that the
hospital mortality of patients with septic AKI was higher than that of pa-
tients with nonseptic AKI (44.5% vs 31.3%; P b .001). On the other hand,
our study found no significant difference in hospital mortality between
patients with septic and nonseptic AKIs treated with CRRT and that sep-
sis had a low odds ratio for hospital mortality in multivariable regres-
sion analysis. The biggest difference between the previous studies
[6,10,21-23] and ours is that the previous studies included patients
with AKI of any severity and we concentrated on patients who were
treated with CRRT.
Several other possible explanations exist for the difference between
the previous findings and ours. First, the hospital mortality of patients
with AKI treated with CRRT is higher than that of all patients with AKI,
and the impact of sepsis may be reduced once patients have become se-
vere enough to require CRRT. Second, in our study, cardiogenic shock
was the highest contributing factor to nonseptic AKI, which had high
hospital mortality. Marenzi et al [34] reported that the hospital mortal-
ity of AKI due to acute myocardial infarction complicated by cardiogenic
shock and treated with RRT was 62%, which was as high as the rate of
the hospital mortality of septic AKI treated with RRT [14,22]. However,
we conducted multivariable logistic regression analysis with cardiogen-
ic shock forced to remain in addition to sepsis/septic shock as a sensitiv-
ity analysis and showed that the odds ratio for cardiogenic shock was
not statistically significant. Third, outcomes of sepsis have improved
since the Surviving Sepsis Campaign introduced guidelines for the man-
agement of severe sepsis and septic shock [35,36]. Stevenson et al [35]
showed that a 28-day mortality of severe septic patients decreased
from 47% during 1991 to 1995 to 29% during 2006 to 2009 in a compar-
ative meta-analysis. In addition, in our previous study, we found that
septic patients included in the Japanese Society of Education for Physi-
cians and Trainees in Intensive Care database conducted in 2010 had a
significantly lower odds ratio of 0.409 (P = .001) for hospital mortality
compared with septic patients included in the Beginning and Ending
Supportive Therapy for the kidney study conducted in 2001 [13,28].
Contributing factors to AKI that had high mortality (cardiogenic shock
and hepatorenal syndrome) and improved outcome of sepsis over the
last decade may have had a large impact on outcome.
4.3. Significance and implications
Several previous multicenter observational studies have shown that
34% to 53% of patients dying in the ICU had their life support withheld or
withdrawn [37-39]. For example, Esteban et al [38] reported that 36.1%
of patients dying due to sepsis and multiple-organ dysfunction syn-
drome had a decision to have life support withheld or withdrawn. Clini-
cians may believe that patients with septic AKI have poor outcome
based on the findings in the previous studies, and this may motivate
them to make a decision to withhold or withdraw treatment in patients
with septic AKI treated with CRRT more frequently than in CRRT pa-
tients with other etiologies. We found that, although the clinical condi-
tions of septic AKI treated with CRRT were more severe, their outcomes
were not significantly different from those without sepsis. This finding
1001
may impact decision making for withholding or withdrawing life sup-
port in critically ill patients.
P
4.4. Strengths and limitations
.024
.033 To the best of our knowledge, this study is the first to describe
.026
the clinical characteristics and outcomes of critically ill patients with
.035
.012 septic AKI treated with CRRT in comparison with nonseptic AKI treated
with CRRT.
However, our study also had several significant limitations. First, it
was a retrospective study with a relatively small sample size that was
conducted in one country. However, it included patients from multicen-
ters, and all patients who were treated with CRRT in the participating
centers were included in the study, which should have reduced some
of potential biases. Second, we did not define certain criteria for how
to initiate CRRT and how to select contributing factors to AKI; such de-
cisions were left to each attending physician. However, sepsis was de-
fined according to the published consensus criteria [33]. Third, some
of the patients in the nonsepsis group may well have developed sep-
sis/septic shock during the course of AKI and CRRT treatment. However,
the definition for “septic AKI” used in this study was the same as that
used in the previous studies [6,10,21,23]. Fourth, we could not decide
only 1 main contributing factor to AKI. Therefore, contributing factors
to AKI of some patients overlapped in Fig. 1. Fifth, the CRRT intensity
in our study (13-15 mL/kg per hour) was lower than that recommended
by the guidelines (20-25 mL/kg per hour) [40]. However, it is common
to conduct CRRT with such a low intensity in Japan, and our previous
study showed that low intensity did not affect on hospital outcome
[28]. Last, we did not take into account how frequently our study popu-
lation had life support withheld or withdrawn. The fact that our findings
were different from previous findings for septic AKI (similar mortality to
nonsepsis and low odds ratio for mortality in multivariable analysis)
could at least partly be explained by these limitations.
5. Conclusions
Among the patients with AKI treated with CRRT, we have shown that
approximately half of patients with AKI treated with CRRT had sepsis/
septic shock as a contributing factor to AKI, and that septic AKI treated
with CRRT had more serious clinical conditions and greater abnormali-
ties in vital signs and blood chemistry. However, no significant differ-
ence was observed in patient outcome between patients with septic
and nonseptic AKI treated with CRRT, and sepsis was found to be asso-
ciated with lower hospital mortality in multivariable regression analy-
sis. Sepsis may not be a risk factor for mortality in patients with AKI
who have become severe enough to require CRRT. However, our study
was a multicenter retrospective study conducted in 1 country, and larg-
er international studies are needed to confirm our findings.
References
[1] Garzotto F, Piccinni P, Cruz D, Gramaticopolo S, Dal Santo M, Aneloni G, et al. RIFLE-
based data collection/management system applied to a prospective cohort multicen-
ter Italian study on the epidemiology of acute kidney injury in the intensive care
unit. Blood Purif 2011;31:159–71.
[2] Clec'h C, Gonzalez F, Lautrette A, Nguile-Makao M, Garrouste-Orgeas M, Jamali S,
et al. Multiple-center evaluation of mortality associated with acute kidney injury
in critically ill patients: a competing risks analysis. Crit Care 2011;15:R128.
[3] Fuchs L, Lee J, Novack V, Baumfeld Y, Scott D, Celi L, et al. Severity of acute kidney
injury and two-year outcomes in critically ill patients. Chest 2013;144:866–75.
[4] Hoste EA, Clermont G, Kersten A, Venkataraman R, Angus DC, De Bacquer D, et al.
RIFLE criteria for acute kidney injury are associated with hospital mortality in criti-
cally ill patients: a cohort analysis. Crit Care 2006;10:R73.
[5] Bagshaw SM, George C, Dinu I, Bellomo R. A multi-centre evaluation of the RIFLE
criteria for early acute kidney injury in critically ill patients. Nephrol Dial Transplant
2008;23:1203–10.
[6] Bagshaw SM, Uchino S, Bellomo R, Morimatsu H, Morgera S, Schetz M, et al. Begin-
ning and Ending Supportive Therapy for the Kidney (BEST Kidney) Investigators:
septic acute kidney injury in critically ill patients: clinical characteristics and out-
comes. Clin J Am Soc Nephrol 2007;2:431–9.
1002 I. Nagata et al. / Journal of Critical Care 30 (2015) 998–1002
[7] 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, inci-
dence, comorbidity, and outcome. J Am Soc Nephrol 2003;14:1022–30.
[8] Medve L, Antek C, Paloczi B, Kocsi S, Gartner B, Marjanek Z, et al. Epidemiology of
acute kidney injury in Hungarian intensive care units: a multicenter, prospective,
observational study. BMC Nephrol 2011;12:43.
[9] Uchino S, Kellum JA, Bellomo R, Doig GS, Morimatsu H, Morgera S, et al. Beginning
and Ending Supportive Therapy for the Kidney (BEST Kidney) Investigators: acute
renal failure in critically ill patients: a multinational, multicenter study. JAMA
2005;294:813–8.
[10] Neveu H, Kleinknecht D, Brivet F, Loirat P, Landais P. Prognostic factors in acute renal
failure due to sepsis. Results of a prospective multicentre study. Nephrol Dial Trans-
plant 1996;11:293–9.
[11] 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. Crit Care Med 1996;24:192–8.
[12] Cruz DN, Bolgan I, Perazella MA, Bonello M, de Cal M, Corradi V, et al. North East Ital-
ian Prospective Hospital Renal Outcome Survey on Acute Kidney Injury (NEiPHROS-
AKI) Investigators: North East Italian Prospective Hospital Renal Outcome Survey on
Acute Kidney Injury (NEiPHROS-AKI): targeting the problem with the RIFLE criteria.
Clin J Am Soc Nephrol 2007;2:418–25.
[13] Uchino S, Bellomo R, Morimatsu H, Morgera S, Schetz M, Tan I, et al. Continuous
renal replacement therapy: a worldwide practice survey. The beginning and ending
supportive therapy for the kidney (B.E.S.T. kidney) investigators. Intensive Care Med
2007;33:1563–70.
[14] Yasuda H, Kato A, Fujigaki Y, Hishida A. Shizuoka Kidney Disease Study Group: inci-
dence and clinical outcomes of acute kidney injury requiring renal replacement
therapy in Japan. Ther Apher Dial 2010;14:541–6.
[15] VA/NIH Acute Renal Failure Trial NetworkPalevsky PM, Zhang JH, O'Connor TZ,
Chertow GM, Crowley ST, Choudhury D, et al. Intensity of renal support in critically
ill patients with acute kidney injury. N Engl J Med 2008;359:7–20.
[16] Silvester W, Bellomo R, Cole L. Epidemiology, management, and outcome of severe
acute renal failure of critical illness in Australia. Crit Care Med 2001;29:1910–5.
[17] Charbonney E, Saudan P, Triverio PA, Quinn K, Mentha G, Martin PY. Prognosis of
acute kidney injury requiring renal replacement therapy in solid organ transplanted
patients. Transpl Int 2009;22:1058–63.
[18] Langenberg C, Wan L, Bagshaw SM, Egi M, May CN, Bellomo R. Urinary biochemistry
in experimental septic acute renal failure. Nephrol Dial Transplant 2006;21:3389–97.
[19] Langenberg C, Wan L, Egi M, May CN, Bellomo R. Renal blood flow in experimental
septic acute renal failure. Kidney Int 2006;69:1996–2002.
[20] Brenner M, Schaer GL, Mallory DL, Suffredini AF, Parrillo JE. Detection of renal blood
flow abnormalities in septic and critically ill patients using a newly designed in-
dwelling thermodilution renal vein catheter. Chest 1990;98:170–9.
[21] Bagshaw SM, George C, Bellomo R. ANZICS Database Management Committee: early
acute kidney injury and sepsis: a multicentre evaluation. Crit Care 2008;12:R47.
[22] Vaara ST, Pettilä V, Reinikainen M, Kaukonen KM. Finnish Intensive Care Consor-
tium: Population-based incidence, mortality and quality of life in critically ill pa-
tients treated with renal replacement therapy: a nationwide retrospective cohort
study in Finish intensive care units. Crit Care 2012;16:R13.
[23] Mehta RL, Bouchard J, Soroko SB, Ikizler TA, Paganini EP, Chertow GM, et al. Program
to Improve Care in Acute Renal Disease (PICARD) Study Group: sepsis as a cause and
consequence of acute kidney injury: program to improve care in acute renal disease.
Intensive Care Med 2011;37:241–8.
[24] Bell M, SWING, Granath F, Schön S, Ekbom A, Martling CR. Continuous renal replace-
ment therapy is associated with less chronic renal failure than intermittent
haemodialysis after acute renal failure. Intensive Care Med 2007;33:773–80.
[25] Legrand M, Darmon M, Joannidis M, Payen D. Management of renal replacement
therapy in ICU patients: an international survey. Intensive Care Med 2013;39:101–8.
[26] Allegretti AS, Steele DJ, David-Kasdan JA, Bajwa E, Niles JL, Bhan I. Continuous renal
replacement therapy outcomes in acute kidney injury and end-stage renal disease: a
cohort study. Crit Care 2013;17:R109.
[27] Cerdá J, Cerdá M, Kilcullen P, Prendergast J. In severe acute kidney injury, a higher
serum creatinine is paradoxically associated with better patient survival. Nephrol
Dial Transplant 2007;22:2781–4.
[28] Uchino S, Toki N, Takeda K, Ohnuma T, Namba Y, Katayama S, et al. Validity of low-
intensity continuous renal replacement therapy. Crit Care Med 2013;41:2584–91.
[29] Kawarazaki H, Uchino S, Tokuhira N, Ohnuma T, Namba Y, Katayama S, et al. Who
may not benefit from continuous renal replacement therapy in acute kidney injury.
Hemodial Int 2013;17:624–32.
[30] Yasuda H, Uchino S, Uji M, Ohnuma T, Namba Y, Katayama S, et al. JSEPTIC Clinical
Trial Group. The lower limit of intensity to control uremia during continuous renal
replacement therapy. Crit Care 2014;18(5):539.
[31] Bellomo R, Ronco C, Kellum JA, Mehta RL, Palevsky P. Acute Dialysis Quality Initiative
Workgroup. Acute renal failure—definition, outcome measures, animal models, fluid
therapy and information technology needs: the Second International Consensus
Conference of the Acute Dialysis Quality Initiative (ADQI) Group. Crit Care 2004;8:
R204–12.
[32] Le Gall JR, Lemeshow S, Saulnier F. A new Simplified Acute Physiology Score (SAPS
II) based on a European/North American multicenter study. JAMA 1993;270:
2957–63.
[33] Levy MM, Fink MP, Marshall JC, Abraham E, Angus D, Cook D, et al. 2001 SCCM/
ESICM/ACCP/ATS/SIS International Sepsis Definitions Conference. Crit Care Med
2003;31:1250–6.
[34] Marenzi G, Assanelli E, Campodonico J, De Metrio M, Lauri G, Marana I, et al. Acute
kidney injury in ST-segment elevation acute myocardial infarction complicated by
cardiogenic shock at admission. Crit Care Med 2010;38:438–44.
[35] Stevenson EK, Rubenstein AR, Radin GT, Wiener RS, Walkey AJ. Two decades of mor-
tality trends among patients with severe sepsis: a comparative meta-analysis. Crit
Care Med 2014;42:625–31.
[36] Shiramizo SC, Marra AR, Durão MS, Paes ÂT, Edmond MB, Pavão dos Santos OF. De-
creasing mortality in severe sepsis and septic shock patients by implementing a sep-
sis bundle in a hospital setting. PLoS One 2011;6:e26790.
[37] Prendergast TJ, Claessens MT, Luce JM. A national survey of end-of-life care for crit-
ically ill patients. Am J Respir Crit Care Med 1998;158:1163–7.
[38] Esteban A, Gordo F, Solsona JF, Alía I, Caballero J, Bouza C, et al. Withdrawing and
withholding life support in the intensive care unit: a Spanish prospective multi-
centre observational study. Intensive Care Med 2001;27:1744–9.
[39] Ferrand E, Robert R, Ingrand P, Lemaire F, French LATAREA Group. Withholding and
withdrawal of life support in intensive-care units in France: a prospective survey.
Lancet 2001;357:9–14.
[40] Prowle JR, Schneider A, Bellomo R. Clinical review. Optimal dose of continuous renal
replacement therapy in acute kidney injury. Crit Care 2011;15:207.