Identification of pediatric septic shock subclasses based on genome-wide expression profiling
© Wong et al; licensee BioMed Central Ltd. 2009
Received: 08 July 2009
Accepted: 22 July 2009
Published: 22 July 2009
Septic shock is a heterogeneous syndrome within which probably exist several biological subclasses. Discovery and identification of septic shock subclasses could provide the foundation for the design of more specifically targeted therapies. Herein we tested the hypothesis that pediatric septic shock subclasses can be discovered through genome-wide expression profiling.
Genome-wide expression profiling was conducted using whole blood-derived RNA from 98 children with septic shock, followed by a series of bioinformatic approaches targeted at subclass discovery and characterization.
Three putative subclasses (subclasses A, B, and C) were initially identified based on an empiric, discovery-oriented expression filter and unsupervised hierarchical clustering. Statistical comparison of the three putative subclasses (analysis of variance, Bonferonni correction, P < 0.05) identified 6,934 differentially regulated genes. K-means clustering of these 6,934 genes generated 10 coordinately regulated gene clusters corresponding to multiple signaling and metabolic pathways, all of which were differentially regulated across the three subclasses. Leave one out cross-validation procedures indentified 100 genes having the strongest predictive values for subclass identification. Forty-four of these 100 genes corresponded to signaling pathways relevant to the adaptive immune system and glucocorticoid receptor signaling, the majority of which were repressed in subclass A patients. Subclass A patients were also characterized by repression of genes corresponding to zinc-related biology. Phenotypic analyses revealed that subclass A patients were younger, had a higher illness severity, and a higher mortality rate than patients in subclasses B and C.
Genome-wide expression profiling can identify pediatric septic shock subclasses having clinically relevant phenotypes.
While septic shock is fundamentally an infection-based disease entity, it is not a singular, homogenous disease in the traditional sense. Rather, septic shock is more akin to a syndrome or a broad, heterogeneous disease classification within which likely exist several disease subclasses. The concept of septic shock subclasses is clinically relevant in that potentially it could have major implications for the design of more specifically targeted therapies .
Physiology-based subclassifications of septic shock are well recognized and have clear implications for hemodynamic management [2–4]. More recently, there have been attempts to biologically subclassify patients with septic shock using blood-derived biomarkers. For example, a previous clinical trial centered on an anti-tumor necrosis factor antibody strategy used serum interleukin-6 concentrations to identify and stratify septic shock patients with a higher severity of illness, ostensibly to select a patient population that could potentially derive a greater benefit from immune modulation therapy [5, 6]. We recently identified a subclass of children with septic shock having a high probability of survival with standard care based on admission serum interleukin-8 levels . While the ease of this type of patient subclassification is clinically appealing, it is overly simplistic from a biological standpoint given the complexity and heterogeneity of septic shock .
A potentially more comprehensive approach to subclassification of septic shock involves genome-wide expression profiling based on microarray technology and bioinformatics . Our previous genome-wide expression studies demonstrated and validated that pediatric septic shock is characterized by early, persistent, and concomitant repression of gene programs corresponding to the adaptive immune system and to zinc-related biology [9–13]. Herein we tested the hypothesis that pediatric septic shock subclasses can be discovered through genome-wide expression profiling.
The study protocol was approved by the individual Institutional Review Boards of each participating institution (N = 11 institutions). Children ≤10 years of age admitted to the pediatric intensive care unit and meeting published, pediatric-specific criteria for septic shock were eligible for the study . Controls were recruited from the ambulatory departments of participating institutions using the following exclusion criteria: a recent febrile illness (within 2 weeks), recent use of anti-inflammatory medications (within 2 weeks), or any history of chronic or acute disease even remotely associated with inflammation. The median age (intra-quartile range (IQR)) for the control cohort (N = 32) was 1.6 years (0.2 to 3.7 years). There were 19 males and 13 females in the control cohort.
Sample and data collection
After obtaining informed consent from parents or legal guardians, blood samples were obtained within 24 hours of initial presentation to the pediatric intensive care unit with septic shock. Severity of illness was calculated using the pediatric risk of mortality (PRISM) III score . Organ failure was defined using pediatric-specific criteria [14, 16]. Annotated clinical and laboratory data were collected daily while in the pediatric intensive care unit. Clinical, laboratory, and biological data were entered and stored using a web-based database developed locally.
RNA extraction and microarray hybridization
The data and protocols described in this manuscript are compliant with the minimum information about a microarray experiment (MIAME) and are deposited in the National Center for Biotechnology Information Gene Expression Omnibus (GEO) under accession number GSE13904 (GEO, http://www.ncbi.nlm.nih.gov/geo/). All of the controls and 67 of the patients with septic shock have been previously reported in analyses addressing completely different questions than that addressed in the current study [9, 11–13]. An additional 31 patients in the septic shock cohort have not been previously reported. Total RNA was isolated from whole blood samples using the PaxGene™ blood RNA system (PreAnalytiX, Qiagen/Becton Dickson, Valencia, CA, USA) according the manufacturer's specifications. Microarray hybridization was performed by the Affymetrix Gene Chip Core facility at the Cincinnati Children's Hospital Research Foundation as previously described using the Human Genome U133 Plus 2.0 GeneChip (Affymetrix, Santa Clara, CA, USA) [9, 11–13].
Analyses were performed using one patient sample per chip. Image files were captured using an Affymetrix GeneChip Scanner 3000. CEL files were subsequently preprocessed using robust multiple-array average (RMA) normalization and GeneSpring GX 7.3 software (Agilent Technologies, Palo Alto, CA, USA). All signal intensity-based data were used after RMA normalization, which specifically suppresses all but significant variation among lower intensity probe sets . All chips representing patient samples were then normalized to the respective median values of the controls. Lists of differentially regulated genes were generated using a series of expression and statistical filters embedded in the GeneSpring GX 7.3 software. Further details regarding these filters will be provided in the Results section.
Gene lists of differentially expressed genes were primarily analyzed using the Ingenuity Pathways Analysis (IPA) application (Ingenuity Systems, Redwood City, CA, USA) that provides a tool for discovery of signaling pathways and gene networks within the uploaded gene lists as previously described [12, 18]. Adjunct analyses of gene lists were performed using the National Institutes of Health Database for Annotation, Visualization and Integrated Discovery (DAVID) . Both applications are based on the established biomedical literature and use specific approaches to estimate significance (P values) based on non-redundant representations of the microarray chip and to convert the uploaded gene lists to gene lists containing a single value per gene. The P values provide an estimate of the probability that a given enrichment is present by chance alone and are derived using corrections for multiple comparisons.
Initial identification of putative septic shock subclasses
Differential gene expression across septic shock subclasses
To determine if there are significant differences in gene expression across the putative septic shock subclasses, we carried out a statistical test using a three-group analysis of variance (ANOVA), all genes on the microarray (54,641), and the three putative subclasses as the comparison groups. When we applied a Benjamini-Hochberg false discovery rate of 0.1% the resulting gene list consisted of more than 20,000 genes. While this result suggests that the three putative septic shock classes have highly significant differences in gene expression, a working gene list of more than 20,000 genes is excessively large for practical analysis. Accordingly, we applied a more stringent correction for multiple comparisons (Bonferroni; P < 0.05) and thus generated a working list of 6,934 genes differentially regulated between the three putative septic shock subclasses.
Clinical phenotypes of the septic shock subclasses
Demographic and clinical data for the septic shock subclasses indentified in Figure 2.
Number of patients
Median age in years (IQR)
Number of males/females
Number of deaths (%)
Median pediatric risk of mortality (PRISM) score (IQR)
Maximum number of organ failures (IQR)3
Number with co-morbidity (%)4
Number with immune suppression (%)5
Number receiving hydrocortisone (%)6
Number with Gram-positive bacteria (%)7
Number with Gram-negative bacteria (%)
Number with negative cultures (%)
K-means clustering and pathway analysis
Ingenuity Pathways Analysis-derived signaling pathways corresponding to the 10 individual K-means clusters depicted in Figure 4.
Number of genes
B cell receptor signaling
Leukocyte extravasation signaling
Triggering receptor expressed on myeloid cells signaling
Janus kinase (JAK)/signal transducers and activator of transcription (STAT) signaling
Insulin receptor signaling
B cell receptor signaling
Axon guidance signaling
Glycine, serine, and threonine metabolism
One carbon pool by folate
Protein ubiquitination pathway
Glucocorticoid receptor signaling
Glycine, serine, and threonine metabolism
B cell receptor signaling
Nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB) signaling
Retinoic acid receptor activation
Peroxisome proliferator-activated receptor (PPARα)/retinoid × receptor (RXRα) activation
PI3 kinase/Akt signaling
p38 mitogen-activated protein kinase (MAPK) signaling
Toll-like receptor signaling
Ephrin receptor signaling
Antigen presentation pathway
Glyoxylate and dicarboxylate metabolism
B cell receptor signaling
Epidermal growth factor signaling
Epidermal growth factor (ERK)/MAPK signaling
Huntington's disease signaling
Death receptor signaling
B cell receptor signaling
Huntington's disease signaling
T cell receptor signaling
Natural killer cell signaling
Stress-activated protein kinase/c-Jun NH2-terminal kinase signaling
Leave on out cross-validation and top predictor gene derivation
The above data demonstrate the existence of septic shock subclasses, in very broad terms, based on more than 6,000 differentially regulated genes and 10 K-means clusters consisting of between 300 and 1,100 genes. Herein we sought to refine the gene expression patterns that best distinguish the three septic shock subclasses. We assumed that the genes corresponding to the individual signaling and metabolic pathways listed in Table 2 likely have the most biological significance with regard to differentiating the three septic shock subclasses. Accordingly, we extracted the individual genes corresponding to these pathways (307 total genes). These 307 genes were then subjected to a leave one out cross-validation procedure using the support vector machines algorithm and the Fisher's exact test method of gene selection . This cross-validation procedure yielded 89 correct subclass predictions (subclass A, B, or C) out of 98 (91%).
Top five Ingenuity Pathways Analysis-derived signaling pathways corresponding to the top 100 predictor genes identified by leave one out cross-validation procedures.
Number of genes
B cell receptor signaling
T cell receptor signaling
Glucocorticoid receptor signaling
Natural killer cell signaling
Peroxisome proliferator-activated receptorα/retinoid × receptor activation
Forty-four genes corresponding to the signaling pathways in Table 3.
Protein kinase, cAMP-dependent, regulatory, type I, α
Phosphoinositide-3-kinase, class 2, α polypeptide
K(lysine) actetyltransferase 2B
Transforming growth factor, β receptor I
p21 (CDKN1A)-activated kinase 2
Son of sevenless homolog 1 (Drosophila)
POU domain, class 2, transcription factor 2
GNAS complex locus
Mitogen-activated protein kinase kinase kinase 3
Mitogen-activated protein kinase kinase kinase 1
Calcium/calmodulin-dependent protein kinase II δ
Mitogen-activated protein kinase 1
TAF9B RNA polymerase II
Dual adaptor of phosphotyrosine and 3-phosphoinositides
vav 3 oncogene
FYN oncogene related to SRC, FGR, YES
Nuclear receptor coactivator 2
Calcium/calmodulin-dependent protein kinase II γ
Zeta-chain (TCR) associated protein kinase 70 kDa
E1A binding protein p300
Phosphatase and tensin homolog
Natural cytotoxicity triggering receptor 3
Fc fragment of IgG, low affinity IIc, receptor for (CD32)
Nuclear factor of activated T cells, calcineurin-dependent 1
Protein kinase C, beta 1
Protein tyrosine phosphatase, receptor type, C
Mitogen-activated protein kinase kinase kinase 7
cAMP responsive element binding protein 5
Janus kinase 2 (a protein tyrosine kinase)
CD3e molecule, epsilon (CD3-TCR complex)
Phosphoinositide-3-kinase, class 3
Mitogen-activated protein kinase kinase kinase 5
Fc fragment of IgG, low affinity IIa, receptor (CD32)
Mitogen-activated protein kinase kinase 4
B cell CLL/lymphoma 6 (zinc finger protein 51)
Phospholipase C, gamma 1
v-yes-1 Yamaguchi sarcoma viral oncogene homolog
Inositol polyphosphate-5-phosphatase, 145 kDa
SMAD, mothers against DPP homolog 4 (Drosophila)
Mitogen-actvated protein kinase kinase kinase kinase 4
TAF11 RNA polymerase II
V-raf-1 murine leukemia viral oncogene homolog 1
Regulation of zinc biology-related genes across the septic shock subclasses
Multiple clinical trials have been conducted in patients with septic shock and most have been based on strategies targeting various components of the immune or inflammatory system . Despite being well supported by quality preclinical data, the majority of these strategies have failed when tested by way of randomized, placebo-controlled trials. One major reason for these recurrent failures is the broad heterogeneity intrinsic to the syndrome of septic shock . That is, it is unlikely that any single immune or inflammatory modulating therapy will be beneficial to a heterogeneous group of patients with septic shock. Thus, identification of septic shock subclasses could facilitate the design of more specifically targeted clinical trials having a higher likelihood of demonstrating efficacy.
Herein we have attempted to discover and identify pediatric septic shock subclasses by leveraging the discovery potential of high throughput genomics, a strategy that is now well established in the field of cancer [24–29]. The foundation of our data is a discovery-oriented expression filter, which identified genes having at least two-fold expression difference in between 25% to 50% of the septic shock patients, relative to the median of controls. While this strategy is certainly not the only valid approach to the goals of our study, it allowed us to initially identify three putative subclasses of patients with septic shock based on unsupervised hierarchical clustering. The ability of this strategy to effectively identify subgroups is well demonstrated by the results of direct statistical testing, which mandated the use of a correction for multiple comparisons procedure (Bonferroni) generally thought to be overly stringent for microarray data [30, 31]. The large number of differentially regulated genes identified by stringent, direct statistical testing strongly suggests that the three putative septic shock subclasses are biologically plausible.
The differentially regulated genes identified by direct statistical testing were able to distinguish three broad septic shock subclasses when subjected to hierarchical clustering and principal component analysis, thus further supporting the assertion of biological plausibility. K-means clustering of these differentially regulated genes generated coordinately regulated gene clusters corresponding to multiple signaling and metabolic pathways relevant to the pathobiology of septic shock. Importantly, the cluster maps derived from K-means clustering demonstrate that these signaling and metabolic pathways are differentially regulated across the three subclasses, thus illustrating, at a genomic level, the concept that any one specific immune or inflammatory modulating therapeutic strategy will not be applicable to a heterogeneous cohort of patients with septic shock.
The three septic shock subclasses identified by expression profiling differed significantly with respect to important clinical phenotypes. Specifically, patients in subclass A had a higher level of illness severity, a higher degree of organ failure, and a higher mortality rate compared with the other two subclasses. In addition, patients in subclass A were younger than patients in subclass B. The largest epidemiologic study of pediatric septic shock to date, by Watson et al. , demonstrated that children between the ages of 1 and 12 months had the highest mortality rate (13.5%), when compared with other age groups. Thus, the demonstration that the subclass having the highest mortality rate is composed of younger children is consistent with the existing epidemiologic literature. However, the mortality rate of subclass A (36%) is substantially higher than the overall mortality reported by Watson et al.  (10.3%), as well as the overall mortality rate (17.3%) reported in the largest pediatric septic shock interventional trial to date . In addition, the median ages between patients in subclass A and C were not significantly different. Thus, it is likely that the higher mortality rate in subclass A is, at least in part, a direct manifestation of the gene expression profile that identified the subclass, rather than a simple artifact of having identified a subclass that was significantly younger.
Based on leave one out cross-validation procedures and subsequent extraction of the top 100 predictor genes, the most distinguishing gene expression signature of the subclass A patients was repression of genes corresponding to the adaptive immune system. This pattern of gene repression does not appear to be an artifact of lymphopenia, in as much as the absolute lymphocyte counts were not significantly different across the three subclasses. In addition, subclass A patients were characterized by repression of genes corresponding to glucocorticoid receptor signaling, an intriguing finding given the current controversies surrounding glucocorticoid treatment in septic shock . Finally, our subanalysis focused on repression of genes having zinc biology-related functional annotations demonstrated that repression of zinc biology-related genes was also a distinguishing feature of the subclass A patients. Thus, we conclude that subclass A patients are particularly distinguished from subclass B and C patients by gene repression patterns corresponding to adaptive immunity, glucocorticoid receptor signaling, and zinc-related biology. This pattern of gene repression correlates with higher illness severity, a greater degree of organ failure, and higher mortality in subclass A patients.
Since our data are based on whole blood-derived RNA, it is possible that some of the gene expression patterns that distinguish the three subclasses are a reflection of different distributions of white blood cell populations, rather than within-cell differences in gene expression. As stated above, however, the absolute lymphocyte counts were not significantly different across the three subclasses, thus suggesting that our data are not simply artifacts of differential white blood cell populations. Nonetheless, we are currently in the process of directly addressing this important question by generating expression data from leukocyte subset-specific RNA.
The existing literature supports the biological plausibility of our data at two broad levels. First, our conceptual framework of the pathobiology of septic shock has evolved over the last decade to include the concept of immune paralysis [35–40]. Whereas septic shock has been traditionally viewed as being a reflection of uncontrolled hyper inflammation (an innate immunity problem), it is now thought that septic shock also has a strong, perhaps predominant, anti-inflammatory component that can be manifest as immune suppression and the relative inability to effectively clear an infectious challenge (an adaptive immunity problem). The finding that subclass A patients are characterized by repression of key adaptive immunity genes is well in line with this concept.
Second, normal zinc homeostasis seems to be absolutely critical for normal functioning of both the innate and adaptive immune systems [41, 42]. Thus, we have postulated that abnormal zinc homeostasis may be linked to adaptive immune dysfunction in children with septic shock . In support of this assertion, we previously demonstrated that non-survivors of pediatric septic shock had abnormally low serum zinc concentrations compared with survivors . Potential links between altered zinc homeostasis, adaptive immune function, and septic shock are the subject of ongoing work in our basic and translational research programs.
We have demonstrated the existence of three broad subclasses of children with septic shock based on gene expression profiling conducted within the first 24 hours of admission to the intensive care unit with septic shock. Broadly, the three subclasses demonstrate differential regulation of genes corresponding to multiple signaling and metabolic pathways relevant to the pathobiology of septic shock, thereby illustrating the important concepts of patient heterogeneity at a genomic level and the need to design more specifically targeted therapies. On a more specific level, the three subclasses are characterized by differential regulation of genes corresponding to the adaptive immune system and zinc-related biology, and these patterns of gene regulation correlate with distinct and relevant clinical phenotypes. Genome-level subclassification of septic shock may one day allow for the design of more specifically targeted therapies, and our data indicate that the adaptive immune system and zinc homeostasis may be appropriate targets to explore further.
analysis of variance
database for annotation, visualization, and integrated discovery
epidermal growth factor
gene expression omnibus
ingenuity pathways analysis
mitogen-activated protein kinase
minimum information about a microarray experiment
nuclear factor kappa-light-chain-enhancer of activated B cells
peroxisome proliferator-activated receptor
pediatric risk of mortality
robust multiple-array average
retinoid × receptor
signal transducers and activator of transcription.
This work was supported by a grant from the National Institute of General Medical Sciences (RO1 GM064619), and The Amanda Kanowitz Foundation http://www.amandakfoundation.org.
- Marshall JC, Vincent JL, Fink MP, Cook DJ, Rubenfeld G, Foster D, Fisher CJ, Faist E, Reinhart K: Measures, markers, and mediators: toward a staging system for clinical sepsis. A report of the Fifth Toronto Sepsis Roundtable, Toronto, Ontario, Canada, October 25–26, 2000. Crit Care Med. 2003, 31: 1560-1567. 10.1097/01.CCM.0000065186.67848.3A.View ArticleGoogle Scholar
- Ceneviva G, Paschall JA, Maffei F, Carcillo JA: Hemodynamic support in fluid-refractory pediatric septic shock. Pediatrics. 1998, 102: e19-10.1542/peds.102.2.e19.View ArticleGoogle Scholar
- Jellema WT, Groeneveld AB, Wesseling KH, Thijs LG, Westerhof N, van Lieshout JJ: Heterogeneity and prediction of hemodynamic responses to dobutamine in patients with septic shock. Crit Care Med. 2006, 34: 2392-2398. 10.1097/01.CCM.0000233871.52553.CD.View ArticleGoogle Scholar
- Brierley J, Peters MJ: Distinct hemodynamic patterns of septic shock at presentation to pediatric intensive care. Pediatrics. 2008, 122: 752-759. 10.1542/peds.2007-1979.View ArticleGoogle Scholar
- Reinhart K, Wiegand-Lohnert C, Grimminger F, Kaul M, Withington S, Treacher D, Eckart J, Willatts S, Bouza C, Krausch D, Stockenhuber F, Eiselstein J, Daum L, Kempeni J: Assessment of the safety and efficacy of the monoclonal anti-tumor necrosis factor antibody-fragment, MAK 195F, in patients with sepsis and septic shock: a multicenter, randomized, placebo-controlled, dose-ranging study. Crit Care Med. 1996, 24: 733-742. 10.1097/00003246-199605000-00003.View ArticleGoogle Scholar
- Reinhart K, Menges T, Gardlund B, Harm Zwaveling J, Smithes M, Vincent JL, Tellado JM, Salgado-Remigio A, Zimlichman R, Withington S, Tschaikowsky K, Brase R, Damas P, Kupper H, Kempeni J, Eiselstein J, Kaul M: Randomized, placebo-controlled trial of the anti-tumor necrosis factor antibody fragment afelimomab in hyperinflammatory response during severe sepsis: The RAMSES Study. Crit Care Med. 2001, 29: 765-769. 10.1097/00003246-200104000-00015.View ArticleGoogle Scholar
- Wong HR, Cvijanovich N, Wheeler DS, Bigham MT, Monaco M, Odoms K, Macias WL, Williams MD: Interleukin-8 as a stratification tool for interventional trials involving pediatric septic shock. Am J Respir Crit Care Med. 2008, 178: 276-282. 10.1164/rccm.200801-131OC.View ArticlePubMed CentralGoogle Scholar
- Cobb JP, Mindrinos MN, Miller-Graziano C, Calvano SE, Baker HV, Xiao W, Laudanski K, Brownstein BH, Elson CM, Hayden DL, Herndon DN, Lowry SF, Maier RV, Schoenfeld DA, Moldawer LL, Davis RW, Tompkins RG, Bankey P, Billiar T, Camp D, Chaudry I, Freeman B, Gamelli R, Gibran N, Harbrecht B, Heagy W, Heimbach D, Horton J, Hunt J, Lederer J, Mannick J, McKinley B, Minei J, Moore E, Moore F, Munford R, Nathens A, O'Keefe G, Purdue G, Rahme L, Remick D, Sailors M, Shapiro M, Silver G, Smith R, Stephanopoulos G, Stormo G, Toner M, Warren S, West M, Wolfe S, Young V: Application of genome-wide expression analysis to human health and disease. Proc Natl Acad Sci USA. 2005, 102: 4801-4806. 10.1073/pnas.0409768102.View ArticlePubMed CentralGoogle Scholar
- Wong HR, Cvijanovich N, Allen GL, Lin R, Anas N, Meyer K, Freishtat RJ, Monaco M, Odoms K, Sakthivel B, Shanley TP: Genomic expression profiling across the pediatric systemic inflammatory response syndrome, sepsis, and septic shock spectrum. Crit Care Med. 2009, 37: 1558-1566. 10.1097/CCM.0b013e31819fcc08.View ArticlePubMed CentralGoogle Scholar
- Wong HR: Pediatric septic shock treatment: new clues from genomic profiling. Pharmacogenomics. 2007, 8: 1287-1290. 10.2217/146224188.8.131.527.View ArticleGoogle Scholar
- Wong HR, Shanley TP, Sakthivel B, Cvijanovich N, Lin R, Allen GL, Thomas NJ, Doctor A, Kalyanaraman M, Tofil NM, Penfil S, Monaco M, Tagavilla MA, Odoms K, Dunsmore K, Barnes M, Aronow BJ: Genome-level expression profiles in pediatric septic shock indicate a role for altered zinc homeostasis in poor outcome. Physiol Genomics. 2007, 30: 146-155. 10.1152/physiolgenomics.00024.2007.View ArticlePubMed CentralGoogle Scholar
- Shanley TP, Cvijanovich N, Lin R, Allen GL, Thomas NJ, Doctor A, Kalyanaraman M, Tofil NM, Penfil S, Monaco M, Odoms K, Barnes M, Sakthivel B, Aronow BJ, Wong HR: Genome-level longitudinal expression of signaling pathways and gene networks in pediatric septic shock. Mol Med. 2007, 13: 495-508. 10.2119/2007-00065.Shanley.View ArticlePubMed CentralGoogle Scholar
- Cvijanovich N, Shanley TP, Lin R, Allen GL, Thomas NJ, Checchia P, Anas N, Freishtat RJ, Monaco M, Odoms K, Sakthivel B, Wong HR: Validating the genomic signature of pediatric septic shock. Physiol Genomics. 2008, 34: 127-134. 10.1152/physiolgenomics.00025.2008.View ArticlePubMed CentralGoogle Scholar
- Goldstein B, Giroir B, Randolph A: International pediatric sepsis consensus conference: definitions for sepsis and organ dysfunction in pediatrics. Pediatr Crit Care Med. 2005, 6: 2-8. 10.1097/01.PCC.0000149131.72248.E6.View ArticleGoogle Scholar
- Pollack MM, Patel KM, Ruttimann UE: The Pediatric Risk of Mortality III – Acute Physiology Score (PRISM III-APS): a method of assessing physiologic instability for pediatric intensive care unit patients. J Pediatr. 1997, 131: 575-581. 10.1016/S0022-3476(97)70065-9.View ArticleGoogle Scholar
- Wilkinson JD, Pollack MM, Ruttimann UE, Glass NL, Yeh TS: Outcome of pediatric patients with multiple organ system failure. Crit Care Med. 1986, 14: 271-274.View ArticleGoogle Scholar
- Irizarry RA, Hobbs B, Collin F, Beazer-Barclay YD, Antonellis KJ, Scherf U, Speed TP: Exploration, normalization, and summaries of high density oligonucleotide array probe level data. Biostatistics. 2003, 4: 249-264. 10.1093/biostatistics/4.2.249.View ArticleGoogle Scholar
- Calvano SE, Xiao W, Richards DR, Felciano RM, Baker HV, Cho RJ, Chen RO, Brownstein BH, Cobb JP, Tschoeke SK, Miller-Graziano C, Moldawer LL, Mindrinos MN, Davis RW, Tompkins RG, Lowry SF: A network-based analysis of systemic inflammation in humans. Nature. 2005, 437: 1032-1037. 10.1038/nature03985.View ArticleGoogle Scholar
- Dennis G, Sherman BT, Hosack DA, Yang J, Gao W, Lane HC, Lempicki RA: DAVID: Database for Annotation, Visualization, and Integrated Discovery. Genome Biol. 2003, 4: P3-10.1186/gb-2003-4-5-p3.View ArticleGoogle Scholar
- Daffertshofer A, Lamoth CJ, Meijer OG, Beek PJ: PCA in studying coordination and variability: a tutorial. Clin Biomech (Bristol, Avon). 2004, 19: 415-428. 10.1016/j.clinbiomech.2004.01.005.View ArticleGoogle Scholar
- Do JH, Choi DK: Clustering approaches to identifying gene expression patterns from DNA microarray data. Mol Cells. 2008, 25: 279-288.Google Scholar
- Noble WS: What is a support vector machine?. Nat Biotechnol. 2006, 24: 1565-1567. 10.1038/nbt1206-1565.View ArticleGoogle Scholar
- Abraham E: Why immunomodulatory therapies have not worked in sepsis. Intensive Care Med. 1999, 25: 556-566. 10.1007/s001340050903.View ArticleGoogle Scholar
- Ebert BL, Golub TR: Genomic approaches to hematologic malignancies. Blood. 2004, 104: 923-932. 10.1182/blood-2004-01-0274.View ArticleGoogle Scholar
- Golub TR, Slonim DK, Tamayo P, Huard C, Gaasenbeek M, Mesirov JP, Coller H, Loh ML, Downing JR, Caligiuri MA, Bloomfield CD, Lander ES: Molecular classification of cancer: class discovery and class prediction by gene expression monitoring. Science. 1999, 286: 531-537. 10.1126/science.286.5439.531.View ArticleGoogle Scholar
- Ramaswamy S, Tamayo P, Rifkin R, Mukherjee S, Yeang CH, Angelo M, Ladd C, Reich M, Latulippe E, Mesirov JP, Poggio T, Gerald W, Loda M, Lander ES, Golub TR: Multiclass cancer diagnosis using tumor gene expression signatures. Proc Natl Acad Sci USA. 2001, 98: 15149-15154. 10.1073/pnas.211566398.View ArticlePubMed CentralGoogle Scholar
- Rosenwald A, Wright G, Chan WC, Connors JM, Campo E, Fisher RI, Gascoyne RD, Muller-Hermelink HK, Smeland EB, Giltnane JM, Hurt EM, Zhao H, Averett L, Yang L, Wilson WH, Jaffe ES, Simon R, Klausner RD, Powell J, Duffey PL, Longo DL, Greiner TC, Weisenburger DD, Sanger WG, Dave BJ, Lynch JC, Vose J, Armitage JO, Montserrat E, Lopez-Guillermo A, Grogan TM, Miller TP, LeBlanc M, Ott G, Kvaloy S, Delabie J, Holte H, Krajci P, Stokke T, Staudt LM: The use of molecular profiling to predict survival after chemotherapy for diffuse large-B-cell lymphoma. N Engl J Med. 2002, 346: 1937-1947. 10.1056/NEJMoa012914.View ArticleGoogle Scholar
- Sorlie T, Perou CM, Tibshirani R, Aas T, Geisler S, Johnsen H, Hastie T, Eisen MB, Rijn van de M, Jeffrey SS, Thorsen T, Quist H, Matese JC, Brown PO, Botstein D, Eystein Lonning P, Borresen-Dale AL: Gene expression patterns of breast carcinomas distinguish tumor subclasses with clinical implications. Proc Natl Acad Sci USA. 2001, 98: 10869-10874. 10.1073/pnas.191367098.View ArticlePubMed CentralGoogle Scholar
- Vijver van de MJ, He YD, van't Veer LJ, Dai H, Hart AA, Voskuil DW, Schreiber GJ, Peterse JL, Roberts C, Marton MJ, Parrish M, Atsma D, Witteveen A, Glas A, Delahaye L, Velde van der T, Bartelink H, Rodenhuis S, Rutgers ET, Friend SH, Bernards R: A gene-expression signature as a predictor of survival in breast cancer. N Engl J Med. 2002, 347: 1999-2009. 10.1056/NEJMoa021967.View ArticleGoogle Scholar
- Allison DB, Cui X, Page GP, Sabripour M: Microarray data analysis: from disarray to consolidation and consensus. Nat Rev Genet. 2006, 7: 55-65. 10.1038/nrg1749.View ArticleGoogle Scholar
- Nadon R, Shoemaker J: Statistical issues with microarrays: processing and analysis. Trends Genet. 2002, 18: 265-271. 10.1016/S0168-9525(02)02665-3.View ArticleGoogle Scholar
- Watson RS, Carcillo JA, Linde-Zwirble WT, Clermont G, Lidicker J, Angus DC: The epidemiology of severe sepsis in children in the United States. Am J Respir Crit Care Med. 2003, 167: 695-701. 10.1164/rccm.200207-682OC.View ArticleGoogle Scholar
- Nadel S, Goldstein B, Williams MD, Dalton H, Peters M, Macias WL, Abd-Allah SA, Levy H, Angle R, Wang D, Sundin DP, Giroir B: Drotrecogin alfa (activated) in children with severe sepsis: a multicentre phase III randomised controlled trial. Lancet. 2007, 369: 836-843. 10.1016/S0140-6736(07)60411-5.View ArticleGoogle Scholar
- Annane D, Fan E, Herridge MS: Pro-con debate: steroid use in ACTH non-responsive septic shock patients with high baseline cortisol levels. Crit Care. 2006, 10: 210-10.1186/cc4856.View ArticlePubMed CentralGoogle Scholar
- Felmet KA, Hall MW, Clark RS, Jaffe R, Carcillo JA: Prolonged lymphopenia, lymphoid depletion, and hypoprolactinemia in children with nosocomial sepsis and multiple organ failure. J Immunol. 2005, 174: 3765-3772.View ArticleGoogle Scholar
- Heidecke CD, Hensler T, Weighardt H, Zantl N, Wagner H, Siewert JR, Holzmann B: Selective defects of T lymphocyte function in patients with lethal intraabdominal infection. Am J Surg. 1999, 178: 288-292. 10.1016/S0002-9610(99)00183-X.View ArticleGoogle Scholar
- Hotchkiss RS, Tinsley KW, Karl IE: Role of apoptotic cell death in sepsis. Scand J Infect Dis. 2003, 35: 585-592. 10.1080/00365540310015692.View ArticleGoogle Scholar
- Hotchkiss RS, Karl IE: The pathophysiology and treatment of sepsis. N Engl J Med. 2003, 348: 138-150. 10.1056/NEJMra021333.View ArticleGoogle Scholar
- Perry SE, Mostafa SM, Wenstone R, Shenkin A, McLaughlin PJ: Is low monocyte HLA-DR expression helpful to predict outcome in severe sepsis?. Intensive Care Med. 2003, 29: 1245-1252. 10.1007/s00134-003-1686-2.View ArticleGoogle Scholar
- Frazier WJ, Hall MW: Immunoparalysis and adverse outcomes from critical illness. Pediatr Clin North Am. 2008, 55: 647-668. 10.1016/j.pcl.2008.02.009.View ArticlePubMed CentralGoogle Scholar
- Fraker PJ, King LE: Reprogramming of the immune system during zinc deficiency. Annu Rev Nutr. 2004, 24: 277-298. 10.1146/annurev.nutr.24.012003.132454.View ArticleGoogle Scholar
- Ibs KH, Rink L: Zinc-altered immune function. J Nutr. 2003, 1452S-1456S. Suppl 1
- The pre-publication history for this paper can be accessed here:http://www.biomedcentral.com/1741-7015/7/34/prepub
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.