Intensive Care Med (2010) 36:915–925 DOI 10.1007/s00134-010-1861-1
Nele Brusselaers Stan Monstrey Kirsten Colpaert Johan Decruyenaere Stijn I. Blot Eric A. J. Hoste
Received: 23 July 2009 Accepted: 2 March 2010 Published online: 24 March 2010 Ó Copyright jointly held by Springer and ESICM 2010 Electronic supplementary material The online version of this article (doi:10.1007/s00134-010-1861-1) contains supplementary material, which is available to authorized users.
N. Brusselaers S. Monstrey K. Colpaert J. Decruyenaere E. A. J. Hoste Burn Unit, Ghent University Hospital, Ghent University, Ghent, Belgium S. Monstrey Department of Plastic Surgery, Ghent University Hospital, Ghent University, Ghent, Belgium K. Colpaert J. Decruyenaere E. A. J. Hoste ()) Department of Intensive Care Medicine, Ghent University Hospital, Ghent University, De Pintelaan 185, 9000 Ghent, Belgium e-mail:
[email protected] Tel.: ?32-9-3324197 Fax: ?32-9-3324995
REVIEW
Outcome of acute kidney injury in severe burns: a systematic review and meta-analysis
N. Brusselaers S. I. Blot Department of General Internal Medicine, Ghent University Hospital, Ghent University, Ghent, Belgium
Abstract Purpose: The main objective of this review was to analyse the prevalence and outcome of acute kidney injury (AKI) in patients with severe burn injury. AKI is a common complication in patients with severe burn injury and one of the major causes of death (often combined with other organ dysfunctions). Several definitions of AKI have been used, but the RIFLE ‘consensus’ classification is nowadays considered the gold standard, enabling a more objective comparison of populations. Methods: We performed a systematic literature search (1960–2009), involving PubMed, the Web of Science, the search engine GoogleTM and textbooks. Reference lists and the Science Citation Index search were also consulted. Attributable mortality was assessed by performing a metaanalysis. Results: This search yielded 57 articles and abstracts with relevant epidemiologic data of AKI in
Introduction Acute kidney injury (AKI) is a frequent complication in patients with severe burn injury, and especially severe AKI, with need for treatment with renal replacement therapy (RRT), is associated with a particularly
the burn population. Of these, 30 contained complete mortality data of the burn and control population, which revealed a 3- to 6-fold higher mortality for AKI patients in univariate analysis, depending on the applied definition. When defined by the RIFLE consensus classification, AKI occurred in one quarter of patients with severe burn injury (median mortality of 34.9%), and when defined by the need for renal replacement therapy (RRT), AKI occurred in 3% (median mortality of 80%). The prevalence of AKI slightly increased, but AKI-RRT decreased. However, the outcome in both groups improved. Conclusion: Despite the wide variation of the analysed burn populations and definitions of AKI, this review clearly showed that AKI remains prevalent and is associated with increased mortality in patients with severe burn injury. Keywords Acute kidney injury Burns Thermal injury Meta-analysis Mortality Systematic review
unfavourable prognosis. The is the first report of patients with burn injury who developed AKI and survived hospital stay dated from the early 1960s [1–6]. Older age, severity of the burn injury assessed by the total burned surface area (TBSA), sepsis and multiorgan dysfunction are well-recognised risk factors for
916
the development of AKI in burn patients [7–12]. The pathophysiology of AKI in burn injury is less well studied. Contributing factors for the development of AKI are decreased renal perfusion and inflammation. Renal ischemia is probably less important in the acute phase of burn injury than originally presumed [13]. Instead, inflammation and apoptosis are probably playing an important role [9, 14]. Excessive volume resuscitation to prevent and treat burn shock may lead to intra-abdominal hypertension and abdominal compartment syndrome, which probably is an underestimated contributor to the development of late AKI after burn shock [9, 15]. An important difference between major burn trauma patients and other types of ICU patients may be the intensity and duration of the inflammatory response, which may last longer than in other trauma patients [16]. The scientific reporting on AKI is heterogenic, with more than 35 different definitions applied to describe an ‘abrupt and sustained decrease in kidney function’ [17, 18]. AKI was formerly only considered important when there was need for RRT (AKI-RRT). A major limitation of this definition is that there is no consensus on when RRT should be initiated [19, 20]. Also, even moderately decreased kidney function has a negative impact on the outcome [9, 17]. To establish a uniform definition, in 2004 the Acute Dialysis Quality Initiative formulated the RIFLE classification, which was later adopted in a slightly modified version into the AKI staging system [21–23]. This classification defines three grades of increasing severity of AKI (Risk, Injury and Failure), based on either an increase in serum creatinine (S-Cr) or decrease in urine output (UO). The aim of this review was to evaluate the prevalence and outcome of AKI in patients with burn injury, by performing a systematic review and meta-analysis.
disease, animal studies, case reports and reviews were excluded. Search strategy of systematic review The first selection of the search was performed by one investigator (N.B.) under supervision of the principal investigator (E.H.), who is a content expert. Assessment of eligibility of the remaining articles (after exclusion of the irrelevant articles) was performed by mutual consideration. Two scientific search engines were used: PubMed and the Web of Science (which also contains abstracts of major congresses) (Fig. 1). The search was performed in January 2010 and considered the period 1960–2009. Reference lists of relevant articles, the Science Citation Index and the search engine GoogleTM were also used to identify relevant articles. Finally, two textbooks were hand-searched to retrieve relevant references [24, 25]. For the PubMed search, we used the following Medical Subject Headings or MeSH terms: ‘‘burns’’, and in refining order ‘‘kidney diseases’’, ‘‘renal insufficiency’’ and ‘‘kidney failure’’. The MeSH ‘‘burns’’ was not explanded (‘exp.’) to chemical, electrical, eye and sun burns and inhalation. Both automatic and manual search strategies were used, for example by automatically selecting the articles dealing with the MeSH term as ‘major’ topic. For the search on the Web of Science, the following terms were used: ‘Burns’, ‘Burn’, ‘Renal’, ‘Renal Failure’ and ‘Kidney’. Quality assessment The internal validity of the included studies was analysed according to the framework proposed by Altman [26] for the assessment of articles dealing with prognosis.
Methods
Data extraction and statistical analysis
Selection criteria
The following data were collected: (1) basic study characteristics: author, year of publication, study period, country, retrospectively or prospectively gathered data; (2) population characteristics: adult or paediatric population, the prevalence and mortality of AKI, in- and exclusion criteria of the study (e.g. TBSA), the definition for AKI and the need for acute and chronic RRT. The age group and type of AKI definition were used to define and analyse subpopulations. Statistical analyses were performed with the software program SPSS for Windows, version 16 (SPSS Inc. USA). The meta-analysis was performed with the software package Review Manager (RevMan) (version 5.0 for Windows, Copenhagen: The Nordic Cochrane Centre, The
Studies about paediatric and adult populations with burn injury, providing epidemiologic data on prevalence and mortality of acute kidney injury in patients with severe burn injury, were included. The considered outcome parameter was in-hospital mortality. Articles without a comparison group (= burned patients without AKI) were only integrated for analysing prevalence and mortality, but not for the meta-analysis. Only articles in English, French or Dutch were included. Specific populations such as patients with electrical and chemical burns were not included because of the different pathophysiology. Populations with severe chronic kidney
917
Fig. 1 Flowchart of the systematic literature search, resulting in 57 included articles
743 potential publications identified in PubMed, Web of Science and Google and reference lists
271 publications were excluded: - Language (n = 87) -Animal studies (n = 132) -Type of burn: electrical or chemical (n = 52)
472 potential publications
359 publications did not meet the inclusion criteria -based on title and abstract (n = 314) -excluding reviews, case reports etc (n = 45)
113 publications retrieved for evaluation
56 publications excluded -no data on prevalence or mortality (n = 43) -only data on deceased patients (n = 11) -1 duplicate publication (n = 1) -overlap of data with other article (n=1)
57 publications included into the review Figure legend: AKI, acute kidney injury
Cochrane Collaboration, 2008), with the Manzel-Haenzel test (risk ratios). Several sub-groups were analysed, based on the definitions of AKI or specific baseline characteristics such as age and total burned surface area (TBSA). The results are reported in accordance with the PRISMA and MOOSE guidelines on reporting of meta-analyses [27, 28]. A random effects model was used to combine the data. We used forest plots to visualise the extent of heterogeneity among studies. Heterogeneity was also assessed with the I2 statistic, a standard test that measures the degree of inconsistency across studies. This test results in a range from 0 to 100%, which describes the proportion of variation between the effect of AKI on mortality due to inter-study variation. Higher values indicate more
heterogeneity. Finally, a funnel plot was constructed for assessment of heterogeneity and publication bias.
Results Quality assessment and study characteristics The systematic literature search resulted in 57 relevant studies (Supplemental Tables 1 and 2) after excluding two articles for duplicate publication [29] or duplication of part of the population data [30]. Only nine studies worked with prospectively gathered data, and almost half of the studies were published after 1999. The median study duration was
918
variable, and in one of them UO was used as a criterion to initiate RRT [5, 42, 70, 71]. In three studies there was no precise definition of decrease of UO [39, 42, 72]. In 13 studies, patients met the criteria for AKI when UO was less than 0.5 ml/kg/h, and in one \0.4 ml/kg/h for a certain period. Five studies defined AKI when the daily UO was less than 350–500 ml/day) [4, 40, 43, 54, 73]. Finally, two studies defined AKI as anuria [5, 70]. (4) Blood urea nitrogen (BUN): elevated blood urea nitrogen concentration or ‘azotemia’ was used as the Definitions for AKI single criterion in 3 older studies [35, 74, 75], and in combination with other variables in 13 studies. Nine A total of 23 different definitions of AKI were mentioned. different levels for BUN were used, defined as BUN/ In six studies the exact diagnostic criteria were not S-Cr ratios under 20 [31, 53, 54], BUN increases specified [6, 32–37]. above 25–200 mg/dL [4, 35, 38, 40, 43, 52, 53, 73– 77] or doubling of the ‘normal for age’ BUN (1) In 14 studies AKI was defined by the need for RRT concentration [67]. The precise increase of BUN (AKI-RRT). In seven of these, the indications for was not specified in one study [72]. RRT were not mentioned [38–44]. Two studies used the SOFA classifications 2 and 4, and 7 studies used combinations of different laboratory tests and clinical Classifications of the included studies symptoms [45–51]. (2) Serum creatinine: 32 studies defined AKI by an increase of S-Cr as one or the only diagnostic Several of the 57 studies used different diagnostic criteria criterion. S-Cr was defined by a cutoff value in 19 and/or described several sub-populations, e.g. ‘AKI’ and studies (Fig. 1a ESM), with S-Cr equal to or greater ‘AKI-RRT’. Seventeen studies used several sub-categothan 2 mg/dl as the most frequent cutoff [31, 52–57]. ries such as AKI-RRT or various (more strict) diagnostic The lowest cutoff was 1.1 mg/dl [12, 58] and the criteria. Because an increased S-Cr concentration was the highest 5.0 mg/dl [43]. Relative increases of S-Cr most commonly used diagnostic criterion, the studies were used in 11 studies, of which 8 studies used the applying this variable were divided in two groups defined RIFLE classification [59–69] (Table 1) (Fig. 1b by relative increases or fixed cutoff values (studies ESM). RIFLE classifies patients into three categories applying more than one diagnostic variable were also of severity of AKI, based on a relative increase of included) (Fig. 1a, b ESM). For the analyses, the 57 serum creatinine or a period of oliguria. Of all studies studies were grouped by the diagnostic criteria as descrithat used the RIFLE classification, only Coca et al. bed in Table 1. The studies using S-Cr cutoff values (N = 19) were grouped according to the sequential organ [59] did not use urine output criteria. (3) Urine output: 24 studies used UO as the diagnostic failure score (SOFA) classification for AKI [78]. The criterion for AKI. In four UO was used as a single studies using relative increases in S-Cr (N = 11) used the 5 years (IQR: 2–11 years), with a maximum of 32 years [31]. The total number of study patients was 34,868, of which 1,969 patients had AKI (AKI-RRT: n = 543). Seven studies only reported on AKI patients (n = 97 patients) and could therefore not be used for analysis of the prevalence of AKI. The median number of patients included per study was 174 (IQR: 45.0–695; range 4– 5,000). Eight studies only considered children.
Table 1 Median prevalence and mortality of acute kidney injury as defined by the most sensitive diagnosis Diagnosisa
Fixed S-Cr cutoff (SOFA) Relative increase S-Cr (RIFLE) Only urine output Only BUN AKI-RRT Otherb Totala
N
19 11 4 3 14 9 60a
Cohort
Prevalence
Mortality
Total (n)
AKI (n)
% of patients
Median % (IQR)
N
Median % (IQR)
N
8,675 2,111 463 468 9,443 12,785 34,868
688 608 69 74 334 316 1,969
34.9 30.9 3.5 3.8 17.0 16.0 100
20.7 26.6 16.0 3.2 3.2 2.0 14.5
18 9 3 3 10 8 51a
73.3 34.9 82.0 88.2 80.0 85.7 77.3
17 11 4 3 13 6 54a
(9.5–30.9) (18.4–47.4) (7.5– 64.1) (3.1–8.1) (1.6–11.6) (0.9–14.1) (3.3–24.4)
(48.8–88.2) (28.4–52.6) (55.9–96.0) (77.3–95.5) (72.0–88.6) (67.9–100) (48.9–88.0)
N number of included studies, n number of included patients, AKI b Definition not or inadequately described, or not within one of the acute kidney injury, S-Cr serum creatinine, AKI-RRT acute kidney other categories (= defined by urine output and BUN) injury requiring renal replacement therapy, BUN blood urea nitrogen a Three studies were analysed in two subgroups (AKI-RRT and one of the the first subgroups). The total number of patients (n) includes each study once
919
same multiplication levels as defined by the RIFLE clas- 48.9–87.8%) (Table 1). In the control population (without sification: 91.5 (risk), 92 (injury) and 93 (failure) [79]. AKI), the median mortality was 13.0% (IQR: 5.6–29.4%). The median mortality of AKI defined by the RIFLE classification was 34.9% (mean 41.9%) and in AKI-RRT patients 80.0% (mean: 41.9 and 80.0% respectively). Of Prevalence of AKI the children with severe burn injury (N = 7 studies), AKI was present in 1,872 out of 34,771 patients with 50.5% of the AKI patients died (101/200). In 7 of the 47 severe burn injury (5.4%). The median prevalence of AKI studies mentioning mortality, all AKI patients died [5, 32, in all studies was 14.5% (IQR 3.3–28.0) (Table 1), and in 36, 46, 51, 56, 80]. It is noteworthy that of these seven the paediatric studies only, this was 3.1% (IQR 1.6–22.6). studies, six were published before 1990. The mortality correlated inversely with the sensitivity The prevalence of AKI varied considerably among all studies (range 0.2–64.1%). This can be explained by the of the diagnosis of AKI, illustrated by the stepwise diagnostic criteria for AKI (Table 1), with lower preva- increase in mortality in the different S-Cr categories lence for more specific and less sensitive definitions. But (Fig. 1a, b ESM). The Pearson correlation coefficient the prevalence is also influenced by the inclusion criteria showed a clear decrease in mortality over time for AKI of the studies. For instance, the study with the highest (r = -0.498; p \ 0.001) and AKI-RRT (r = -0.616; prevalence of AKI included only patients with a TBSA of p = 0.001) (Fig. 2a, b). This decreasing trend in the AKIC60% [71]. When defined by the RIFLE classification, RRT group was still present when the three oldest studies AKI occurred in one quarter of the patients (Table 1) with (with 100% mortality) were excluded from this analysis the highest prevalence in the RIFLE-risk category (Fig. 1b (r = -0.400; p = 0.072) (Fig. 2 ESM). Although the ESM). RRT was performed in almost 30% of AKI p-value is not ‘significant’ as classically defined as\0.05, patients, or 3% of the total population with burn injury. the decreasing trend is clear. This higher p-value might be The Pearson correlation test showed an increasing ascribed to the limited number of studies. This decrease prevalence of AKI over time (r = 0.312; p = 0.027), but was less apparent in the control population (r = -0.333; also noteworthy is the decreasing trend in the prevalence p = 0.890). of AKI-RRT (r = -0.368; p = 0.054). In 30 studies, mortality rates of both AKI and control populations were available, which allowed calculating the risk ratio (RR) for mortality. AKI was clearly associated Mortality of AKI patients with increased risk for mortality (aggregate RR = 4.85, 95% CI 3.77, 6.24). However, as mentioned above, there In total, 55.2% of all AKI patients died (959/1,737), was considerable clinical heterogeneity among study with a median mortality in the studies of 77.3% (IQR: cohorts. The I2 statistic of 85% also indicates an
Fig. 2 Evolution of mortality in time, grouped by the year of publication: a all included studies reporting mortality of patients with acute kidney injury (AKI): the regression and correlation analyses indicate a decrease in mortality over time (Pearson’s correlation coefficient = -0.498; p \ 0.001). b All included
studies reporting mortality of patients with acute kidney injury requiring renal replacement therapy (AKI-RRT): The regression and correlation analyses indicate a decrease in mortality over time (Pearson’s correlation coefficient = -0.616; p = 0.001)
920
Fig. 3 Funnel plot of all studies included in the meta-analysis (N = 27). The more pronounced the asymmetry, the more likely it is that the amount of heterogeneity will be substantial
important statistical heterogeneity among the studies, though the funnel plot is rather symmetrical (Fig. 3). To diminish the effect of clinical heterogeneity, we grouped the studies based on RIFLE, SOFA, anuria/oliguria and AKI-RRT. Subgroup analyses according to the type of AKI definition all demonstrated that AKI patients had a three- to six-fold increased risk for in-hospital mortality. The lowest RRs were seen in the three studies with the highest inclusion TBSA (between 50 and 70%). In the five paediatric studies included in the meta-analysis, the RR was 3.2 (95% CI 1.2–6.1). A more sensitive and less specific definition for AKI, as in the RIFLE classification, correlated clearly with increasing mortality risk, although no significantly increased mortality was found in the RIFLE-R category (Fig. 4b). Other outcome parameters Where reported, all AKI-RRT patients had recovery of kidney function in all but one study [31, 38, 42–44, 66, 81], where two patients were described (33%) who required long-term dialysis [45].
Discussion Since 1960, 57 studies have been published containing epidemiological data of prevalence and mortality of acute kidney injury in patients with severe burn injuries. More than 20 different definitions for AKI were used, and the severity of the burn injury varied considerably. This resulted in a wide range of reported prevalence and mortality rates. In general, the prevalence of AKI in patients with burn injury was lower compared to that in a general ICU population (30–66% when AKI is defined by RIFLE classification and 5% for AKI-RRT) [82]. We
found that AKI defined by the RIFLE classification occurred in one quarter of the patients with burn injury and was associated with a 35% mortality rate (median). RRT was used in 3% of patients with burn injury, and was associated with a median mortality rate of 80%. There was a decreasing trend in mortality over the study period. The prevalence of AKI defined by the RIFLE classification was lower in the population with burn injury compared with a general ICU population (27% vs. 30– 66%), but mortality was comparable or even higher (35% vs. 17–36%), especially for the RIFLE failure group (75% vs. 26–57%) [17, 18, 20, 83, 84]. The lower prevalence may be explained by an age difference between populations with burn injury and general ICU populations, wherein populations with burn injury have been described to be younger, with consequently a lower prevalence of co-morbidities. We cannot rule out that the prevalence of AKI based on serum creatinine criteria was overestimated or underestimated. Muscle injury, as a consequence of burn injury, may have led to elevations in serum creatinine concentration by increased release of creatinine in the circulation, while the glomerular filtration rate was unaffected. On the other hand, serum creatinine may already be increased on admission, leading to a false low prevalence of AKI when defined on relative increases of serum creatinine. This problem is generic to this kind of AKI definition and not specific to patients with burn injury. The higher mortality may suggest that burn injury itself, and other factors such as the extent and depth of the burn injury, inhalation injury and infection in particular, have an important impact on outcome. It is also noteworthy that increasing severity of AKI is associated with a stepwise increase in mortality. This suggests that there is indeed a cause-effect relationship between severity of AKI and outcome. This is similar to findings in other patient cohorts, such as in general ICU patients and sepsis patients [17, 85]. In these cohorts, this finding was reinforced by multivariate analysis that demonstrated a persisting effect even after correction for other covariates that may explain mortality [20]. However, the data in this review do not allow such a multivariate analysis. Therefore, we cannot discern whether increased mortality is indeed caused by the greater severity of AKI or is only due to confounding factors such as an increased baseline severity (e.g. the extent of the burn). Nevertheless, the increased mortality was apparent in all study populations, even those including only the most severely burned patients, which justifies the presence of an attributable mortality risk of AKI. The prevalence of AKI-RRT is lower compared to two large multi-centre studies of Uchino et al. and Metnitz et al. [86, 87] in general ICU populations (3% vs. respectively 4.9 and 4.3%). AKI-RRT in patients with severe burn injury is associated with a mortality of 80%, which is again higher than in general ICU populations (50–60%) [18, 86]. The relative low prevalence of AKI-
921
Fig. 4 Meta-analysis: risk ratio for mortality of patients with acute kidney injury (AKI) compared with a control population without AKI. a AKI defined by RIFLE. b AKI defined by SOFA. c AKI defined by anuria/oliguria. d AKI defined by treatment with RRT. e AKI in the paediatric burn population
922
Fig. 4 continued
RRT may indicate reluctance to initiate RRT because of the potential adverse events, such as bleeding or catheterrelated bloodstream infections, or because of anticipated therapeutic futility. It seems plausible that when the initiation of RRT is postponed, AKI-related morbidity (e.g. volume overload, acidosis, electrolyte abnormalities, azotemia, etc.) will be more pronounced, and subsequently, mortality will be higher. Further, we found that mortality in the populations with burn injury complicated with AKI decreased significantly during the study period of more than 40 years. When we excluded the three oldest studies where AKI was associated with 100% mortality, this trend was still clear, however not statistically significant. The decrease of mortality in populations with burn injury not complicated by AKI was less apparent. This effect was even larger in patients with AKI-RRT (which can be considered the most severe grade of AKI). An explanation for this could be the progress in patient care. However, the use of more sensitive diagnostic criteria for AKI may also explain that mortality is decreasing and prevalence increasing. More early diagnosis and therapy of AKI and AKI-RRT, leading to the effect of stage migration, may have resulted in better outcomes. This review has several limitations. First, treatment of patients was probably not comparable among different studies because this review covers a period of over 40 years. During this time period important changes in management of patients with burn injury have taken
place (e.g. the creation of specialised burn units). This may have important implications for the epidemiology of AKI and can (to some extent) explain the decrease in mortality in burn populations over the study period. Apart from changes in management over the years, there may also be important differences in treatment protocols among different centres. Second, the absence of a standardised definition for AKI makes comparisons on the epidemiology of AKI difficult. This was in part resolved by analysis of subgroups of studies reporting on AKI defined by the same definition, such as the consensus definition for AKI (the RIFLE classification), and on AKI patients who were treated with RRT. However, even when AKI was defined by treatment with RRT, study cohorts may have had important differences in severity of AKI as indications and timing of initiation may differ among units and over time. Third, there may have been important differences in baseline characteristics of the patients. Also, admission criteria may differ among burn units and may have also changed over time. In general, there was insufficient reporting on baseline characteristics, such as TBSA, inhalation injury, age and chronic conditions such as diabetes, to include these data into this review. Finally, despite including abstracts and book chapters in our review, there may have been reporting bias and under-reporting of studies with low RR as suggested by the funnel plot. All these factors contribute to heterogeneity of the included studies. This is an inherent problem when systematic reviews and
923
meta-analyses are performed, especially when observational studies are used [88–90]. The strength of this review is that it presents a comprehensive overview of all studies reporting on AKI in patients with severe burn injury. It highlights the need for uniform reporting of AKI, and also demonstrates that burn injury patients with AKI have a worse prognosis, which is also almost linearly correlated with severity of AKI. Important is also the finding that even discrete changes of kidney function, e.g. as defined in the RIFLE classification, are associated with worse outcomes in patients with burn injury. Although there was a considerable heterogeneity among the study populations, we may conclude that AKI is an important and frequently occurring complication in patients with severe burn injury. When defined according to the RIFLE consensus classification, AKI occurs in one quarter of patients with severe burn injury, and when
defined by the need for RRT, AKI occurs in 3%. This is lower compared to a general ICU cohort; however, mortality was comparable or even higher. Greater severity of AKI was associated with greater mortality, and importantly, even when AKI is defined by a sensitive definition, outcome is worse. Finally, although AKI and AKI-RRT remain prevalent in populations with severe burn injury, the outcome improved. Acknowledgments We would like to thank Prof. Matthias Egger (Professor of Epidemiology, Department of Social and Preventive Medicine at the University of Berne, Switzerland) and Dr. Thomas Schioler (Senior Medical Officer, The National Board of Health, Denmark) for their valuable contributions to an earlier draft of this article. Stijn Blot is supported by a grant from the European Society of Intensive Care Medicine and iMDsoft Patient Safety Research Award 2008. Eric Hoste is Senior Clinical Investigator of the Research Foundation-Flanders (Belgium) (FWO).
References 1. Stephens FO, Stewart JH (1965) Burns complicated by anuria: recovery after radical surgery and haemodialysis. Lancet 15–17 2. Stewart JH, Stephens FO (1965) Burns complicated by anuria. Lancet 286:1131 3. Goldsmith HJ, Nakamoto S, Kolff WJ (1960) Expanding the indications for treatment with the artificial kidney: burns, cerebral oedema, sulphonamide intoxication and congestive heartfailure. Lancet 111–112 4. Creyssel J, Deleuze R, Gate A, Caillard B (1962) Renal function in a burned patient. Study of 430 cases. Lyon Chir 58:137–141 5. Hockmuth RE, Ziffren SE (1963) Results of the treatment of severe burns. Surg Gynecol Obstet 117:540–546 6. Proyard G, Cuypers Y (1963) Renal insufficiency in burned patients. Acta Chir Belg 62:510–516 7. Fitzwater J, Purdue GF, Hunt JL, O’Keefe GE (2003) The risk factors and time course of sepsis and organ dysfunction after burn trauma. J Trauma 54:959–966 8. Blot S (2009) Development and validation of a model for prediction of mortality in patients with acute burn injury. Br J Surg 96:111–117 9. Colpaert K, Hoste EA (2008) Acute kidney injury in burns: a story of volume and inflammation. Crit Care 12:192
10. Brusselaers N, Hoste EA, Monstrey S, Colpaert KE, De Waele JJ, Vandewoude KH, Blot SI (2005) Outcome and changes over time in survival following severe burns from 1985 to 2004. Intensive Care Med 31:1648–1653 11. Brusselaers N, Juhasz I, Erdei I, Monstrey S, Blot S (2009) Evaluation of mortality following severe burns injury in hungary: external validation of a prediction model developed on belgian burn data. Burns 35(7):1009– 1014. doi:10.1016/j.burns.2008. 1012.1017 12. Cumming J, Purdue GF, Hunt JL, O’Keefe GE (2001) Objective estimates of the incidence and consequences of multiple organ dysfunction and sepsis after burn trauma. J Trauma 50:510– 515 13. Langenberg C, Wan L, Egi M, May CN, Bellomo R (2006) Renal blood flow in experimental septic acute renal failure. Kidney Int 69:1996–2002 14. Wan L, Bellomo R, Di Giantomasso D, Ronco C (2003) The pathogenesis of septic acute renal failure. Curr Opin Crit Care 9:496–502 15. Tuggle D, Skinner S, Garza J, Vandijck D, Blot S (2007) The abdominal compartment syndrome in patients with burn injury. Acta Clin Belg Suppl 1:136–140 16. Jeschke MG, Mlcak RP, Finnerty CC, Norbury WB, Gauglitz GG, Kulp GA, Herndon DN (2007) Burn size determines the inflammatory and hypermetabolic response. Crit Care 11:R90
17. 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:R73 18. Kellum JA, Hoste EA (2008) Acute kidney injury: epidemiology and assessment. Scand J Clin Lab Invest Suppl 241:6–11 19. Palevsky PM (2006) Dialysis modality and dosing strategy in acute renal failure. Semin Dial 19:165–170 20. Hoste EA, Schurgers M (2008) Epidemiology of acute kidney injury: how big is the problem? Crit Care Med 36:S146–S151 21. 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–R212 22. 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:R31 23. Bagshaw SM, George C, Dinu I, Bellomo R (2008) A multi-centre evaluation of the RIFLE criteria for early acute kidney injury in critically ill patients. Nephrol Dial Transplant 23:1203–1210
924
24. Mariano F, Gangemi EN, Stella M, Tedeschi L, Gregoretti C, Triolo G (2009) Critical care nephrology: Chapter 57: Burns and acute renal failure, 2nd edn. Saunders-Elsevier, Philadelphia 25. Herndon DN (2007) Total burn care, 3rd edn. Saunders-Elsevier, Galveston 26. Altman DG (2001) Systematic reviews of evaluations of prognostic variables. BMJ 323:224–228 27. Moher D, Liberati A, Tetzlaff J, Altman DG (2009) Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. PLoS Med 6:doi:10/1371/journal.pmed.1000097 28. Stroup DF, Berlin JA, Morton SC, Olkin I, Williamson GD, Rennie D, Moher D, Becker BJ, Sipe TA, Thacker SB (2000) Meta-analysis of observational studies in epidemiology: a proposal for reporting. Meta-analysis of observational studies in epidemiology (MOOSE) group. JAMA 283:2008–2012 29. Sabry A, Wafa I, El-Din AB, El-Hadidy AM, Hassan M (2009) Early markers of renal injury in predicting outcome in thermal burn patients. Saudi J Kidney Dis Transpl 20:632–638 30. Chung KK, Juncos LA, Wolf SE, Mann EE, Renz EM, White CE, Barillo DJ, Clark RA, Jones JA, Edgecombe HP, Park MS, Albrecht MC, Cancio LC, Wade CE, Holcomb JB (2008) Continuous renal replacement therapy improves survival in severely burned military casualties with acute kidney injury. J Trauma 64:S179–S185 (discussion S185–177) 31. Jeschke MG, Barrow RE, Wolf SE, Herndon DN (1998) Mortality in burned children with acute renal failure. Arch Surg 133:752–756 32. Cason JS (1965) Treatment of renal failure in burns. In: Transactions of 2nd international congress on research in burns Edinburgh, Livingstone. 12 33. Gudaviciene D, Rimdeika R, Adamonis K (2004) Influence of enteral nutrition on the frequency of complications in case of major burns. Medicina (Kaunas) 40:957–961 34. O’Keefe GE, Hunt JL, Purdue GF (2001) An evaluation of risk factors for mortality after burn trauma and the identification of gender-dependent differences in outcomes. J Am Coll Surg 192:153–160 35. Rohan JE, Beaufils H, Chapman A (1972) Acute renal failure during the first 3 months of evolution of burns in 315 adult patients. Sem Hop 48:3389– 3393 36. Tweddell JS, Waymack JP, Warden GD, Law EJ, Jenkins M, Alexander JW (1987) Hematuria in the burned child. J Pediatr Surg 22:899–903
37. Chung KK, Wolf SE, Cancio LC, Alvarado R, Jones JA, McCorcle J, King BT, Barillo DJ, Renz EM, Blackbourne LH (2009) Resuscitation of severely burned military casualties: fluid begets more fluid. J Trauma 67:231–237 (discussion 237) 38. Leblanc M, Thibeault Y, Querin S (1997) Continuous haemofiltration and haemodiafiltration for acute renal failure in severely burned patients. Burns 23:160–165 39. Davies MP, Evans J, McGonigle RJ (1994) The dialysis debate: acute renal failure in burns patients. Burns 20:71– 73 40. Holm C, Horbrand F, von Donnersmarck GH, Muhlbauer W (1999) Acute renal failure in severely burned patients. Burns 25:171–178 41. Haberal A, Ozdemir U (1993) Gastrointestinal and renal complications in burned patients. Eur J Plastic Surgery 16:17–21 42. Hubsher J, Olshan AR, Schwartz AB, Zoranski B, DeClement F, Bendlin A, Hensell D, Brezin JH, Krevolin LE, Chinitz JL (1986) Continuous arteriovenous hemofiltration for the treatment of anasarca and acute renal failure in severely burned patients. ASAIO Trans 32:401–404 43. Sun IF, Lee SS, Lin SD, Lai CS (2007) Continuous arteriovenous hemodialysis and continuous venovenous hemofiltration in burn patients with acute renal failure. Kaohsiung J Med Sci 23:344–351 44. Tremblay R, Ethier J, Querin S, Beroniade V, Falardeau P, Leblanc M (2000) Veno-venous continuous renal replacement therapy for burned patients with acute renal failure. Burns 26:638– 643 45. Abdel-Rahman E, Moorthy V, Helgerson RB, Omari B (1997) Acute renal failure (ARF) requiring dialysis in patients with burns: 16 years experience in one center. Clin Nephrol 8:A576– A0576 46. Dorricott NJ, Dawson-Edwards P, Blainey JD (1968) Surgical disasters and renal failure. Proc R Soc Med 61:215–216 47. Garner WL, Reiss M (2005) Burn care in Los Angeles, California: LAC ? USC experience 1994–2004. Burns 31(Suppl 1):S32–S35 48. Gueugniaud PY (1993) Apport de l’he´mofiltration continue au traitement des bruˆle´s graves. Elsevier, Amsterdam 49. Holm C, Horbrand F, Mayr M, von Donnersmarck GH, Muhlbauer W (2004) Acute hyperglycaemia following thermal injury: friend or foe? Resuscitation 60:71–77
50. Maharaj R (2008) Acute renal injury in patients with major thermal injury. In: World Congress of Anaesthesiologists: 2–7 March 2008, Cape Town, South Africa 51. Marshall V, Yoffa D (1965) The diagnosis and management of renal failure in surgical patients. Aust N Z J Surg 35:20–29 52. Aikawa N, Shinozawa Y, Ishibiki K, Abe O, Yamamoto S, Motegi M, Yoshii H, Sudoh M (1987) Clinical analysis of multiple organ failure in burned patients. Burns Incl Therm Inj 13:103– 109 53. Barrow RE, Jeschke MG, Herndon DN (2000) Early fluid resuscitation improves outcomes in severely burned children. Resuscitation 45:91–96 54. Chrysopoulo MT, Jeschke MG, Dziewulski P, Barrow RE, Herndon DN (1999) Acute renal dysfunction in severely burned adults. J Trauma 46:141–144 55. Hoste E, Vandeveire B, Monstrey S, Colardyn F: Acute renal failure and renal dyfunction in burn patients. In: ISBI: 1998; Jerusalem, Israel; 1998: 79 56. Kim GH, Oh KH, Yoon JW, Koo JW, Kim HJ, Chae DW, Noh JW, Kim JH, Park YK (2003) Impact of burn size and initial serum albumin level on acute renal failure occurring in major burn. Am J Nephrol 23:55–60 57. Spies M, Herndon DN, Rosenblatt JI, Sanford AP, Wolf SE (2003) Prediction of mortality from catastrophic burns in children. Lancet 361:989–994 58. Cooper AB, Cohn SM, Zhang HS, Hanna K, Stewart TE, Slutsky AS (2006) Five percent albumin for adult burn shock resuscitation: lack of effect on daily multiple organ dysfunction score. Transfusion 46:80–89 59. Coca SG, Bauling P, Schifftner T, Howard CS, Teitelbaum I, Parikh CR (2007) Contribution of acute kidney injury toward morbidity and mortality in burns: a contemporary analysis. Am J Kidney Dis 49:517–523 60. Chung KK, Lundy JB, Matson JR, Renz EM, White CE, King BT, Barillo DJ, Jones JA, Cancio LC, Blackbourne LH, Wolf SE (2009) Continuous venovenous hemofiltration in severely burned patients with acute kidney injury: a cohort study. Crit Care 13:R62 61. Mosier MJ, Pham TN, Klein MB, Gibran NS, Arnoldo BD, Gamelli RL, Tompkins RG, Herndon DN (2010) Early acute kidney injury predicts progressive renal dysfunction and higher mortality in severely burned adults. J Burn Care Res 31:83–92
925
62. Palmieri T, Lavrentieva A, Greenhalgh D (2009) An assessment of acute kidney injury with modified RIFLE criteria in pediatric patients with severe burns. Intensive Care Med 35:2125– 2129 63. Palmieri T, Lavrentieva A, Greenhalgh DG (2009) Acute kidney injury in critically ill burn patients. Risk factors, progression and impact on mortality. Burns 36(2):205–211 64. Lopes JA, Jorge S, Neves FC, Caneira M, da Costa AG, Ferreira AC, Prata MM (2007) An assessment of the RIFLE criteria for acute renal failure in severely burned patients. Nephrol Dial Transplant 22:285 65. Mariano F, Cantaluppi V, Stella M, Romanazzi GM, Assenzio B, Cairo M, Biancone L, Triolo G, Ranieri VM, Camussi G (2008) Circulating plasma factors induce tubular and glomerular alterations in septic burns patients. Crit Care 12:R42 66. Steinvall I, Bak Z, Sjoberg F (2008) Acute kidney injury is common, parallels organ dysfunction or failure, and carries appreciable mortality in patients with major burns: a prospective exploratory cohort study. Crit Care 12:R124 67. Williams DM, Sreedhar SS, Mickell JJ, Chan JC (2002) Acute kidney failure: a pediatric experience over 20 years. Arch Pediatr Adolesc Med 156:893– 900 68. Hu S, Sheng Z, Zhou B, Guo Z, Lu J, Xue L, Jin H, Sun X, Sun S, Li J, Lu Y (1998) Study on delay two-phase multiple organ dysfunction syndrome. Chin Med J (Engl) 111:101–108 69. Rossiter K, Kulayat N, Anderson J (1995) Acute renal failure, burns & death: patterns and probabilities in the burn center (1991–1993). Clin Nephrol 6(3):474 70. el Danaf A (1995) Burn variables influencing survival: a study of 144 patients. Burns 21:517–520 71. O’Mara MS, Caushaj P, Goldfarb IW, Slater H (2000) Treatment and mortality trends among massively burned patients. Annals of burns and fire disasters 13(2):73–76
72. Nguyen NL, Gun RT, Sparnon AL, Ryan P (2002) The importance of initial management: a case series of childhood burns in Vietnam. Burns 28:167–172 73. Huang PP, Stucky FS, Dimick AR, Treat RC, Bessey PQ, Rue LW (1995) Hypertonic sodium resuscitation is associated with renal failure and death. Ann Surg 221:543–554 (discussion 554–547) 74. Cameron JS, Miller-Jones CM (1967) Renal function and renal failure in badly burned children. Br J Surg 54:132–141 75. Vertel RM, Knochel JP (1967) Nonoliguric acute renal failure. Jama 200:598–602 76. Saffle JR, Sullivan JJ, Tuohig GM, Larson CM (1993) Multiple organ failure in patients with thermal injury. Crit Care Med 21:1673–1683 77. Sabry A, El-Din AB, El-Hadidy AM, Hassan M (2009) Markers of tubular and glomerular injury in predicting acute renal injury outcome in thermal burn patients: a prospective study. Ren Fail 31:457–463 78. 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 22:707–710 79. Hoste EA, Kellum JA (2006) RIFLE criteria provide robust assessment of kidney dysfunction and correlate with hospital mortality. Crit Care Med 34:2016–2017 80. Schiavon M, Di Landro D, Baldo M, De Silvestro G, Chiarelli A (1988) A study of renal damage in seriously burned patients. Burns Incl Therm Inj 14:107– 112 81. Mustonen KM, Vuola J (2008) Acute renal failure in intensive care burn patients (ARF in burn patients). J Burn Care Res 29:227–237
82. Hoste EA, De Waele JJ (2008) Acute kidney injury in the intensive care unit: it’s the gene, stupid!. Crit Care Med 36:3266–3267 83. Bagshaw SM, George C, Gibney RT, Bellomo R (2008) A multi-center evaluation of early acute kidney injury in critically ill trauma patients. Ren Fail 30:581–589 84. Ostermann M, Chang RW (2007) Acute kidney injury in the intensive care unit according to RIFLE. Crit Care Med 35:1837–1843 quiz 1852 85. Joannidis M, Metnitz B, Bauer P, Schusterschitz N, Moreno R, Druml W, Metnitz PG (2009) Acute kidney injury in critically ill patients classified by AKIN versus RIFLE using the SAPS 3 database. Intensive Care Med 35:1692– 1702 86. 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 294:813–818 87. Metnitz PG, Krenn CG, Steltzer H, Lang T, Ploder J, Lenz K, Le Gall JR, Druml W (2002) Effect of acute renal failure requiring renal replacement therapy on outcome in critically ill patients. Crit Care Med 30:2051–2058 88. Vandenbroucke JP (2004) When are observational studies as credible as randomised trials? Lancet 363:1728– 1731 89. 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 370:1453–1457 90. Higgins JP, Thompson SG, Deeks JJ, Altman DG (2003) Measuring inconsistency in meta-analyses. BMJ 327:557–560
