Several studies have evaluated the association between plasminogen activator inhibitor-1 (PAI-1) -675 4G/5G polymorphism and sepsis in different populations. However, the available results are conflicting.
A search of Pubmed and EMBASE databases was performed to identify relevant studies for inclusion in the meta-analysis. Odds ratios (ORs) and corresponding 95% confidence intervals (CIs) were determined using a random-effects model.
Twelve case-control studies and three cohort studies were included. Overall, a significant association between 4G/5G polymorphism and sepsis risk was observed for 4G/4G vs. 4G/5G +5G/5G (OR = 1.30, 95% CI 1.08–1.56, P = 0.006). In addition, there was a significant association between PAI-1 4G/5G polymorphism and sepsis-related mortality (OR = 1.72, 95% CI 1.27–2.33, P = 0.0005). In subgroup analyses, increased sepsis risk and mortality risk were found in Caucasians and in patients with sepsis.
Citation: Li L, Nie W, Zhou H, Yuan W, Li W, Huang W (2013) Association between Plasminogen Activator Inhibitor-1 -675 4G/5G Polymorphism and Sepsis: A Meta-Analysis. PLoS ONE 8(1): e54883. https://doi.org/10.1371/journal.pone.0054883
Editor: Devendra Amre, University of Montreal, Canada
Received: October 12, 2012; Accepted: December 17, 2012; Published: January 30, 2013
Copyright: © 2013 Li et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This study was supported by National Natural Science Foundation of China (NO. 81070003) and Natural Science Foundation of Guangdong province (NO. 10151001002000005). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Sepsis is a major public health problem that is responsible for an estimated economic burden of nearly 17 billion dollars annually in the United States. Despite the development of effective antibiotics and supportive care, sepsis remains the leading cause of death in critically ill patients . Therefore, predictive markers to identify high-risk patients are urgently needed for early detection and preventive care. Recently, a number of investigators have begun the search for genetic risk factors that influence clinical outcomes in sepsis, and the plasminogen activator inhibitor-1 (PAI-1) gene has been studied extensively.
PAI-1, a member of the serine protease inhibitor (serpin) family, is associated with the severity and outcome of sepsis . Zeerleder and coworkers reported that PAI-1 levels were significantly higher in septic shock patients than in severe sepsis patients . Furthermore, plasma PAI-1 was significantly higher in septic disseminated intravascular coagulation (DIC) patients than in nonseptic DIC patients, and its elevation was an independent risk factor for mortality in the septic DIC group . In patients with meningococcal sepsis, previous studies showed that concentrations of PAI-1 were markedly elevated and there was a significant correlation between PAI-1 levels and mortality , . Taken together, these results suggest that PAI-1 may play a pivotal role in the pathogenesis of sepsis.
The human PAI-1 gene is located on chromosome 7, which contains a 4G/5G polymorphism located within the promoter region, 675 base pairs upstream of the transcription start site. This polymorphism has a role in the regulation of PAI-1 levels . Recently, a number of research groups have studied this polymorphism as a potential susceptibility factor for sepsis. Several studies assessed the association between PAI-1 -675 4G/5G polymorphism and the risk and outcomes of sepsis , –. However, the results were not consistent and remained inconclusive. As most studies had relatively small sample sizes, we performed a meta-analysis to determine whether PAI-1 -675 4G/5G polymorphism was associated with an increased risk of sepsis or higher sepsis mortality. To our knowledge, this was the first meta-analysis of the association between PAI-1 -675 4G/5G polymorphism and sepsis risk and mortality.
Relevant studies were identified by searching Pubmed and EMBASE databases up to September 2012. The following search terms were used: sepsis and (plasminogen activator inhibitor-1 or PAI-1 or SERPINE1) and (polymorphism or mutation or variant). All the searched studies were retrieved, and their references were also checked for other relevant publications. Review articles were also searched to find additional eligible studies. No publication date or language restrictions were applied.
Inclusion and Exclusion Criteria
Studies fulfilling the following selection criteria were included in this meta-analysis: (1) they evaluated the association between -675 4G/5G polymorphism in PAI-1 gene and risk of sepsis or sepsis mortality; (2) they were case-control or cohort studies; (3) they should report PAI-1 genotype as 4G/4G vs. 4G/5G vs. 5G/5G. Studies were excluded if one of the following criteria existed: (1) the studies were not relevant to -675 4G/5G polymorphism, PAI-1, sepsis, or mortality; (2) they were non-clinical; (3) they were reviews or comments. For overlapping studies, only the one with the largest sample size was included.
Two authors (Li and Nie) independently assessed the quality of each study. Any disagreement was resolved by consensus. Quality assessment scores of genetic association studies of human sepsis were used to assess the quality of the selected articles . This quality scoring system was based on both traditional epidemiologic considerations and genetic issues. Total scores ranged from 0 (worst) to 9 (best) for cohort studies and 0 (worst) to 10 (best) for case-control studies.
The following variables were extracted from each study if available: first author’s surname, publication year, ethnicity of study participants, study design, age, type of sepsis, numbers of cases and controls, and genotype numbers in cases and controls. Information was carefully entered into predesigned data collection forms, independently by two of the investigators (Li and Nie). The accuracy of the data was verified by comparing collection forms from each investigator. Any discrepancy was resolved by discussion, or a third author (Huang) would assess these articles.
Where the data from at least three similar studies were available, a meta-analysis was performed. OR and 95% CI were employed to evaluate the strength of the association between -675 4G/5G polymorphism and the risk of sepsis and death. ORs were calculated for the genotypes: 4G/4G vs. 5G/5G (OR1), 4G/5G vs. 5G/5G (OR2), and 4G/4G vs. 4G/5G (OR3) for the -675 4G/5G polymorphism. These pairwise differences were used to indicate the most appropriate genetic model as follows: if OR1 = OR3 ≠ 1 and OR2 = 1, then a recessive model was suggested; if OR1 = OR2 ≠ 1 and OR3 = 1, then a dominant model was suggested; if OR2 = 1/OR3 ≠ 1 and OR1 = 1, then a complete overdominant model was suggested; if OR1> OR2>1 and OR1> OR3>1 (or OR1< OR2<1 and OR1< OR3<1), then a codominant model was suggested –. Once the best genetic model was identified, this model was used to collapse the three genotypes into two groups (except in the case of a codominant model) and to pool the results again. A random-effects model, using the Mantel-Haenszel method, was used to calculate the pooled ORs. The statistical significance of OR was determined with the Z test.
Departure from Hardy-Weinberg equilibrium (HWE) in controls was tested by the Chi-square test. The Q statistic and the I2 statistic were used to assess the degree of heterogeneity among the studies included in the meta-analysis. Subgroup analyses were carried out by ethnicity and type of sepsis. Sensitivity analysis was performed through sequentially excluded individual studies to assess the stability of the results. The potential publication bias was examined visually in a funnel plot of log [OR] against its standard error (SE), and the degree of asymmetry was tested using Egger’s test .
All statistical tests were performed using Revman 5.1 software (Nordic Cochrane Center, Copenhagen, Denmark) and STATA 11.0 software (Stata Corporation, College Station, TX, USA). A P value <0.05 was considered statistically significant, except for tests of heterogeneity where a level of 0.10 was used.
The literature search identified a total of 98 records. After removing 24 duplications and reading the abstracts, a further 49 articles were excluded. After reading the full text of the remaining articles, 10 were then excluded and 15 articles remained. The study selection process is shown in Figure 1. Finally, 12 case-control studies and 3 cohort studies were included in our meta-analysis , –. Thirteen studies were conducted in Caucasian populations, and two in Asian populations. Most of the studies comprised adult patients, but five studies focused on pediatric patients. Sepsis was defined as sepsis (eight studies), severe sepsis (two studies), septic shock (two studies), and mixed (two studies). The quality scores of most studies ranged from 5 to 8, suggesting high quality. We did not assess the quality of the study conducted by Lorente et al. , because insufficient information could be extracted from that study. The characteristics of the selected studies are presented in Table 1. Genotype numbers and HWE examination results are shown in Table 2.
Quantitative Data Synthesis
Twelve studies determined the association between -675 4G/5G polymorphism and sepsis risk , –, –, . Total sample sizes for sepsis and control groups were 1806 and 2239, respectively. The estimated OR1, OR2 and OR3 were 1.30, 0.97, and 1.30, respectively (Table 3). These estimates suggested a recessive genetic model, and therefore 4G/4G was compared with 4G/5G and 5G/5G. The pooled OR in this analysis was 1.30 (95% CI 1.08–1.56; P = 0.006) (Figure 2). This result suggested that the 4G/4G genotype was significantly associated with sepsis risk. In the subgroup analysis by ethnicity, significant associations were found among Caucasians (OR = 1.24; 95% CI 1.02–1.51; P = 0.03). Subgroup analysis was also performed by sepsis type. Increased risk was found among the patients with sepsis (OR = 1.25; 95% CI 1.07–1.45; P = 0.004).
Twelve studies identified an association between -675 4G/5G polymorphism and sepsis-related mortality risk , , –. The total sample included 2098 patients. Figure 3 shows a significant association between -675 4G/5G polymorphism and sepsis-related mortality with an OR of 1.72 (95% CI 1.27–2.33; P = 0.0005). In the subgroup analysis by ethnicity, significant association was also found among Caucasians with the 4G/4G genotype (OR = 1.59; 95% CI 1.16–2.17; P = 0.004). Stratification by sepsis type showed that patients with sepsis carrying 4G/4G genotype were associated with an increased mortality risk (OR = 2.06; 95% CI 1.38–3.05; P = 0.0004). Summary results of comparisons are listed in Table 3.
For sepsis risk, there was a statistically significant between-study heterogeneity in the recessive genetic model (I2 = 37%). Galbraith plots were used to explain the heterogeneity. The study performed by Sipahi et al.  was the outlier in the recessive genetic model (Figure 4). After excluding this study, the between-study heterogeneity effectively decreased and there was no obvious heterogeneity among the 11 remaining studies (I2 = 0%).
For mortality risk, there was also statistically significant between-study heterogeneity (I2 = 47%). Galbraith plots found one study  as the outlier and the possible major source of heterogeneity (Figure 5). When the study was excluded, the I2 value was 34%.
To evaluate the stability of the results of the meta-analysis, sensitivity analyses were performed through sequentially omitted individual studies. None of the results were materially changed, which suggested the robustness of our results (data not shown).
Cumulative meta-analyses of the two associations were performed via the assortment of studies by publication time. As shown in Figure 6 and Figure 7, inclinations toward significant associations were evident with each addition of more data over time. The results showed that the pooled ORs tended to be stable.
Publication bias was examined by funnel plots qualitatively and estimated by Egger’s test quantitatively. The shapes of the funnel plots seemed slightly asymmetrical (Figure 8 and Figure 9). Egger’s test did not show evidence of publication bias for sepsis risk (P = 0.322). However, publication bias was found for mortality risk (P = 0.024).
Sepsis is a complex clinical syndrome that results from a systemic inflammatory response to bacteria and/or bacterial products . Coagulation and inflammation are reciprocally interdependent and closely related in sepsis . Haemostatic changes, ranging from mild laboratory alterations to DIC, are often observed during sepsis . DIC is an independent predictor of poor outcome in patients with severe sepsis , . The imbalance between fibrin formation and deposition contributes to DIC . PAI-1 is a principal inhibitor of tissue plasminogen activator (t-PA) and can be viewed as a potent inhibitor of fibrinolysis. Recently, Semeraro et al.  suggested that suppression of fibrinolysis due to PAI-1 was an important mechanism in DIC development. In addition, a growing number of studies confirmed that high levels of PAI-1 correlated closely with the severity of sepsis, and were predictive of adverse outcomes , . Consequently, PAI-1 may play a critical role in the development of sepsis, and increased levels of PAI-1 may predict a high mortality risk. An early study reported that a functional mutation in the PAI-1 gene (the 4G/5G polymorphism) could influence the expression of the PAI-1 gene . The 4G/4G genotype has been linked to higher PAI-1 level, compared with the 5G/5G genotype, with the heterozygous genotype associated with intermediate levels . Therefore, we hypothesized that PAI-1 -675 4G/5G polymorphism could influence the susceptibility to sepsis and sepsis-related mortality.
In our meta-analysis, we found that the 4G/4G genotype was a modest risk factor for developing sepsis in the overall study population. The results revealed that carriers of the 4G/4G genotype had a 30% increased sepsis risk compared with individuals carrying the 5G allele (4G/5G +5G/5G). In the subgroup analysis, we noted that Caucasians carrying the 4G/4G genotype had an increased sepsis risk. There were only two studies on Asians for this polymorphism , . Therefore, subgroup analysis was not performed in the Asians subgroup. More studies in Asian populations are needed to evaluate the effect of -675 4G/5G polymorphism on sepsis risk. In addition, we carried out subgroup analysis by sepsis type. We found that patients in sepsis subgroup who carrying 4G/4G genotype had an increased disease risk. Since there were only three studies performed in patients with severe sepsis or septic shock, subgroup analyses could not be conducted and more studies should be designed to analyze these conditions. A significant association was found between PAI-1 -675 4G/5G polymorphism and sepsis-related mortality. We found that septic patients with the 4G/4G genotype had a 72% increased mortality risk compared to patients with 4G/5G genotype or 5G/5G genotype. Similarly, significant results were also noted in the Caucasian subgroup and sepsis subgroup. Since our meta-analysis included no more than two Asian studies, severe sepsis, or septic shock populations, any positive association between these conditions and sepsis-related mortality could not be ruled out, because a small sample size may have insufficient statistical power to detect a slight effect. These associations require further study.
There were modest heterogeneities in the overall comparisons for PAI-1 -675 4G/5G polymorphism. Galbraith plots were used to explore the sources of heterogeneity. We found that all I2 values were decreased after excluding the outliers. The results suggested that the two outlying studies ,  might be the major source of the heterogeneity. However, heterogeneity did not seem to influence the results, because the significance of the result was not altered after excluding the outliers. Moreover, we carried out sensitivity analyses. Removal of each study did not alter the associations with sepsis risk and mortality risk, suggesting the reliability of these results. The cumulative meta-analyses showed a trend of more marked associations between PAI-1 -675 4G/5G polymorphism and increased risk of sepsis and mortality as data accumulated each year. This procedure also proved that our results were robust.
Salanti et al.  suggested that false-negative results may be suppressed or false-positive results magnified. Thus, the results of meta-analyses might be influenced by publication bias. Although Egger’s test did not show significant publication bias for sepsis risk, we found the shape of the funnel plot was slightly asymmetrical. In addition, significant publication bias was observed for mortality risk. Thus, the results should be interpreted cautiously and more studies are still needed to confirm the findings from this meta-analysis.
Some limitations of this meta-analysis should be pointed out. First, the number of included studies in our meta-analysis was moderate. Second, most of the studies were conducted in Caucasian populations. Therefore, our results may be applicable only to this ethnic group. Third, sepsis is a complex disease, and many genes are associated with it , . However, we could not address gene-gene interactions in this meta-analysis due to the lack of the related information. Fourth, the overall outcome was based on unadjusted data, whereas a more precise analysis could be performed if individual data were available to allow adjustment. Finally, because only the studies that were indexed by the selected databases were included in our meta-analysis, some relevant published studies may not have been included, which may have biased our results.
In conclusion, this meta-analysis suggested that PAI-1 -675 4G/5G polymorphism may represent a risk factor for sepsis and sepsis-related mortality. Well-designed studies with large sample sizes are needed to further evaluate the associations between this polymorphism and clinical outcomes of sepsis in various ethnic populations. Moreover, gene-gene interactions should also be considered in future studies.
We thank Dr Greet Van den Berghe (Head of Intensive Care Medicine; Head Laboratory of Intensive Care Medicine; Department Cellular and Molecular Medicine; University Hospitals Leuven (UZ Leuven)) for providing relevant information.
Conceived and designed the experiments: LL WN. Performed the experiments: LL WN HFZ WFY WFL. Analyzed the data: LL WN HFZ. Contributed reagents/materials/analysis tools: LL WN HFZ WFY WFL. Wrote the paper: LL WN HFZ WJH.
- 1. Dombrovskiy VY, Martin AA, Sunderram J, Paz HL (2007) Rapid increase in hospitalization and mortality rates for severe sepsis in the United States: A trend analysis from 1993 to 2003. Crit Care Med 35: 1244–1250.
- 2. Hermans P, Hazelzet JA (2005) Plasminogen activator inhibitor type 1 gene polymorphism and sepsis. Clin Infect Dis 41: S453–S458.
- 3. Zeerleder S, Schroeder V, Hack CE, Kohler HP, Wuillemin WA (2006) TAFI and PAI-1 levels in human sepsis. Thromb Res 118: 205–212.
- 4. Madoiwa S, Nunomiya S, Ono T, Shintani Y, Ohmori T, et al. (2006) Plasminogen activator inhibitor 1 promotes a poor prognosis in sepsis-induced disseminated intravascular coagulation. Int J Hematol 84: 398–405.
- 5. Kornelisse RF, Hazelzet JA, Savelkoul HFJ, Hop WCJ, Suur MH, et al. (1996) The relationship between plasminogen activator inhibitor-1 and proinflammatory and counterinflammatory mediators in children with meningococcal septic shock. J Infect Dis 173: 1148–1156.
- 6. Hermans PWM, Hibberd ML, Booy R, Daramola O, Hazelzet JA, et al. (1999) 4G/5G promoter polymorphism in the plasminogen-activator-inhibitor-1 gene and outcome of meningococcal disease. Lancet 354: 556–560.
- 7. Eriksson P, Kallin B, Van't Hooft FM, Båvenholm P, Hamsten A (1995) Allele-specific increase in basal transcription of the plasminogen-activator inhibitor 1 gene is associated with myocardial infarction. P Natl Acad Sci USA 92: 1851–1855.
- 8. Westendorp RG, Hottenga JJ, Slagboom PE (1999) Variation in plasminogen-activator-inhibitor-1 gene and risk of meningococcal septic shock. Lancet 354: 561–563.
- 9. Menges T, Hermans PWM, Little SG, Langefeld T, Böning O, et al. (2001) Plasminogen-activator-inhibitor-1 4G/5G promoter polymorphism and prognosis of severely injured patients. Lancet 357: 1096–1097.
- 10. Haralambous E, Hibberd ML, Hermans PWM, Ninis N, Nadel S, et al. (2003) Role of functional plasminogen-activator-inhibitor-1 4G/5G promoter polymorphism in susceptibility, severity, and outcome of meningococcal disease in Caucasian children. Crit Care Med 31: 2788–2793.
- 11. Geishofer G, Binder A, Müller M, Zöhrer B, Resch B, et al. (2005) 4G/5G promoter polymorphism in the plasminogen-activator-inhibitor-1 gene in children with systemic meningococcaemia. Eur J Pediatr 164: 486–490.
- 12. Zhan Z, Wang H, Chen D, Cheng B, Wang H, et al. (2005) Relationship between sepsis and 4G/5G polymorphism within the promoter region of plasminogen activator inhibitor-1 gene. Zhonghua Yi Xue Za Zhi 85: 2404–2407.
- 13. Sipahi T, Pocan H, Akar N (2006) Effect of various genetic polymorphisms on the incidence and outcome of severe sepsis. Clin Appl Thromb Hemost 12: 47–54.
- 14. García-Segarra G, Espinosa G, Tassies D, Oriola J, Aibar J, et al. (2007) Increased mortality in septic shock with the 4G/4G genotype of plasminogen activator inhibitor 1 in patients of white descent. Intensive Care Med 33: 1354–1362.
- 15. Jessen K, Lindboe S, Petersen A, Eugen-Olsen J, Benfield T (2007) Common TNF-α, IL-1β, PAI-1, uPA, CD14 and TLR4 polymorphisms are not associated with disease severity or outcome from Gram negative sepsis. BMC Infect Dis 7: 108–115.
- 16. Wei Y, Zeng Q, Zeng H (2008) 4G/5G promoter polymorphism in the plasminogen activator inhibitor 1 gene in children with sepsis. Chin J Evid Based Pediatr 3: 280–285.
- 17. Henckaerts L, Nielsen KR, Steffensen R, Van Steen K, Mathieu C, et al. (2009) Polymorphisms in innate immunity genes predispose to bacteremia and death in the medical intensive care unit. Crit Care Med 37: 192–201.
- 18. Peres Wingeyer S, de Larranaga G, Cunto E, Fontana L, Nogueras C, et al. (2010) Role of 4G/5G promoter polymorphism of Plasminogen Activator Inhibitor-1 (PAI-1) gene in outcome of sepsis. Thromb Res 125: 367–369.
- 19. Lorente L, Martin M, Labarta L, Diaz C, Sole-Violan J, et al. (2009) Role of 4G/5G plasminogen activator inhibitor-1 (PAI-1) gene polymorphism and the PAI-1 plasma levels in the outcome of patients with severe sepsis. Crit Care 13: S143–S144.
- 20. Madach K, Aladzsity I, Szilagyi A, Fust G, Gal J, et al. (2010) 4G/5G polymorphism of PAI-1 gene is associated with multiple organ dysfunction and septic shock in pneumonia induced severe sepsis: prospective, observational, genetic study. Crit Care 14: R79–88.
- 21. Wingeyer SP, Cunto E, Nogueras C, Juan JS, Norberto G, et al. (2012) Biomarkers in sepsis at time zero: Intensive care unit scores, plasma measurements and polymorphisms in Argentina. J Infect Dev Ctries 6: 555–562.
- 22. Clark MF, Baudouin SV (2006) A systematic review of the quality of genetic association studies in human sepsis. Intensive Care Med 32: 1706–1712.
- 23. Thakkinstian A, McElduff P, D'Este C, Duffy D, Attia J (2005) A method for meta-analysis of molecular association studies. Stat Med 24: 1291–1306.
- 24. Nie W, Fang Z, Li B, Xiu Q (2012) Interleukin-10 promoter polymorphisms and asthma risk: A meta-analysis. Cytokine 60: 849–855.
- 25. Nie W, Chen J, Xiu Q (2012) Cytotoxic T-Lymphocyte Associated Antigen 4 Polymorphisms and Asthma Risk: A Meta-Analysis. PLoS ONE 7: e42062.
- 26. Egger M, Smith GD, Schneider M, Minder C (1997) Bias in meta-analysis detected by a simple, graphical test. BMJ 315: 629–634.
- 27. Esmon CT (2004) Crosstalk between inflammation and thrombosis. Maturitas 47: 305–314.
- 28. Semeraro N, Ammollo CT, Semeraro F, Colucci M (2011) Sepsis, thrombosis and organ dysfunction. Thromb Res 129: 290–295.
- 29. Fourrier F, Chopin C, Goudemand J, Hendrycx S, Caron C, et al. (1992) Septic shock, multiple organ failure, and disseminated intravascular coagulation. Compared patterns of antithrombin III, protein C, and protein S deficiencies. Chest 101: 816–823.
- 30. Mann KG (1999) Biochemistry and physiology of blood coagulation. Thromb Haemost 82: 165–174.
- 31. Horrevoets AJG (2004) Plasminogen activator inhibitor 1 (PAI-1): in vitro activities and clinical relevance. Br J Haematol 125: 12–23.
- 32. Semeraro N, Ammollo CT, Semeraro F, Colucci M (2010) Sepsis-associated disseminated intravascular coagulation and thromboembolic disease. Mediterr J Hematol Infect Dis 2: e2010024.
- 33. Mesters R, Flörke N, Ostermann H, Kienast J (1996) Increase of plasminogen activator inhibitor levels predicts outcome of leukocytopenic patients with sepsis. Thromb Haemost 75: 902–907.
- 34. Biemond B, Levi M, Ten Cate H, Van der Poll T, Büller H, et al. (1995) Plasminogen activator and plasminogen activator inhibitor I release during experimental endotoxaemia in chimpanzees: effect of interventions in the cytokine and coagulation cascades. Clin Sci 88: 587–594.
- 35. Dawson S, Wiman B, Hamsten A, Green F, Humphries S, et al. (1993) The two allele sequences of a common polymorphism in the promoter of the plasminogen activator inhibitor-1 (PAI-1) gene respond differently to interleukin-1 in HepG2 cells. J Biol Chem 268: 10739–10745.
- 36. Festa A, D’Agostino R, Rich SS, Jenny NS, Tracy RP, et al. (2003) Promoter (4G/5G) plasminogen activator inhibitor-1 genotype and plasminogen activator inhibitor-1 levels in blacks, Hispanics, and non-Hispanic whites. Circulation 107: 2422–2427.
- 37. Salanti G, Sanderson S, Higgins JPT (2005) Obstacles and opportunities in meta-analysis of genetic association studies. Genet Med 7: 13–20.
- 38. Sutherland AM, Walley KR (2009) Bench-to-bedside review: Association of genetic variation with sepsis. Crit Care 13: 210–220.
- 39. Wong HR (2012) Genetics and genomics in pediatric septic shock. Crit Care Med 40: 1618–1626.