According to recent data, sepsis continues to be a major cause of mortality and morbidity during the neonatal period and is still as a global health challenge (1). Early onset sepsis (EOS) diagnosed ≤72 h after birth is most often related to perinatal factors including chorioamnionitis, prolonged rupture of amniotic membranes, and maternal colonization with group B streptococcus (GBS) (2). In developing countries in Asia, the overall incidence of EOS is 0.62 cases per 1,000 live births or 4.91 per 1,000 admissions, with a mortality rate of 7.0%, whereas in the mainland of China, the incidence of EOS has been as high as 20.03 per 1,000 admissions likely because antepartum antibiotics are not commonly used in China (3, 4). Although regarded as the gold standard for detecting sepsis in adults or children, positive blood culture rates have been reported to be as low as 3.3% among infant cases and the method simply requires too much time (5), which limits the use of blood culture as a basis for initial empirical antibiotic use. Moreover, the clinical signs of sepsis in neonates are non-specific and subtle, and conventional parameters for evaluating sepsis, such as white blood cell count (WBC), immature/total neutrophil (I/T) ratio, and C-reactive protein (CRP) level, do not offer sufficient sensitivity or specificity for EOS. Hence, early biomarkers that can be accurately and rapidly measured are greatly needed for predicting EOS and guiding initial antibiotic use (6).
Interleukin (IL)-27, a heterodimeric cytokine of the IL-12 family, has been identified as a vital modulator of inflammatory immune responses (7–9). Recently, through genome-wide expression analysis, Epstein–Barr virus-induced gene 3 (EBI3), a subunit of IL-27, was shown to have high predictive strength for bacterial infection (10). Additional research demonstrated that IL-27 is a good biomarker for estimating the risk of bacterial infection in the blood in critically ill pediatric patients and adults (11–13). However, no data are available regarding changes in IL-27 levels in neonatal sepsis. Except for the biomarkers mentioned above, procalcitonin (PCT), IL-6, IL-8, and tumor necrosis factor (TNF)-α have been studied and are regarded as some of the most promising marker candidates of EOS. Extensive data also are available for sepsis-related changes in heat shock protein (HSP) 70, macrophage inflammatory protein (MIP)-1α, MIP-1β, granzyme B, and matrix metalloprotease (MMP)-8 in adults or animal models, but information regarding these factors in EOS is limited (14–21). In the present study, using high-throughput multiplex bead arrays, we aimed to compare the performance of these sepsis markers in neonates with EOS and to evaluate whether combined use of any of these markers with CRP and PCT could discriminate neonates who will develop sepsis from those unlikely to develop sepsis.
PATIENTS AND METHODS
This prospective study was carried out from May 2015 to February 2016 in the neonatal diagnosis center of Children's Hospital, Chonqing Medical University, Chongqing, China, which is a national clinical specialty department with 215 beds, including a 40-bed neonatal intensive care unit and a 175-bed intermediate care section. This center founded and operates the Critical Neonates Network of China (http://www.cneonet.com/) and receives more than 10,000 neonates from 15 provinces each year. The study population comprised infants born ≥34 weeks gestation with suspicion of an early-onset infection (≤72 h of birth) as diagnosed by neonatologists with more than 10 years of experience who were blinded to the biomarker results. The study was approved by the review board of Children's Hospital (2013-109, 2015-115), Chongqing Medical University, and informed consent from the infants’ parents or guardians was obtained before inclusion. Cases with major congenital malformations, confirmed intrauterine viral infection, or for which parental consent was not granted were excluded.
Evaluation of sepsis
Patients were assessed immediately after admission and considered at risk for sepsis based on the presence of one or more of the following criteria: antinatal risk factors (prolonged rupture of membranes >18 h, chorioamnionitis or positive evidence of GBS); clinical signs including respiratory dysfunction (distress or apnea), tachycardia (heart rate >190 beats/min), or bradycardia (heart rate <90 beats/min), cardiovascular compromise (e.g., paleness or peripheral cyanosis and mottled skin with capillary refill delayed >3 s), neurological signs (seizures, irritability, lethargy), feeding intolerance, or abdominal distension; positive results on blood or cerebrospinal fluid cultures; and positive results on conventional laboratory tests (WBC <5×109 or >20×109/L, CRP >10 mg/L, I/T ratio >0.12, platelet count <100,000/mm3).
Study group allocation
In the first step, all recruited patients were categorized according to the likelihood of sepsis into four groups as described previously (16): proven sepsis (any abnormal findings with a positive result on blood or cerebrospinal fluid culture); probable sepsis (negative culture results but ≥3 abnormal findings); possible sepsis (negative culture results and two abnormal findings); and low-risk of sepsis (negative culture results and only a single abnormal finding).
Then, in our analysis of the predictive power of these biomarkers, we included cases of proven sepsis and probable sepsis within the infected group and cases of possible or low-risk sepsis in the uninfected group for comparison. In particular, we did not apply a traditional “gold standard” infected group based on positive culture results, because the positive culture rate is extremely low in patients with EOS. All comparisons and analyses in this study were based on this classification. This approach is consistent with the previous study (16) and the clinical protocol in which patients in the infected group usually are given antibiotics for at least 7 days, whereas such treatment is usually ceased after 48 to 72 h in infant cases of uninfected group.
Measurement of biomarkers in blood samples
A 50-μL venous blood sample was obtained from each infant once the infant was sent to the nursery ward for sampling for blood cultures and other routine tests. The sampling time in terms of hours after birth was recorded for each infant. Samples were centrifuged immediately after collection for 5 min at 2,000 g, and the serum obtained was immediately frozen in sterile tubes at −80°C for later measurement of biomarker levels after samples from all patients had been collected.
To accurately evaluate changes in the level of IL-27 within the first 72 h after birth, we also measured IL-27 in the cord blood of 33 healthy infants born beyond 34 weeks gestational age.
Plasma levels of IL-6, IL-8, IL-27, TNF-α, granzyme B, HSP 70, MIP-1α, MIP-1β, and MMP-8 were determined using the Milliplex Map Human Th17 Magnetic Bead Panel and Sepsis Panel (Millipore, Billerica, Mass) according to the manufacturer's instructions. Values were calculated based on a standard curve constructed for each marker, and the results were interpreted using the Bio-plex 200 system based on Luminex-200 technology (Bio-Rad Laboratories Inc, Hercules, Calif). PCT was measured by chemiluminescence immunoassay using the Snibe Maglumi 2000 system (Shenzhen New Industries Biomedical Engineering, Shenzhen, China). All measurements were made by the same person who was blinded to the clinical data.
We checked the normality of the data using the Kolmogorov–Smirnov test. Non-normally distributed data are expressed as median values with interquartile ranges, and comparisons between the infected and uninfected groups were assessed with the Mann–Whitney test. Comparisons among different intervals and baseline data were assessed with the Kruskal–Wallis test. Qualitative or categorical variables are expressed as frequencies and proportions. Proportions were compared using the chi-square test. Correlation analysis among different markers was assessed by Spearman rank test. All statistical tests were two-sided and performed at a significance level of P = 0.05. These analyses were performed using SAS 9.0 (SAS Institute Inc, Cary, NC). Receiver operating characteristic (ROC) curves were constructed for each marker, and the area under the ROC curve (AUC) for each marker was calculated using Medcalc 12.7 (MedCalc Software, Ostend, Belgium). Optimal cut-off points were determined based on ROC curves. The sensitivity, specificity, positive predictive value, negative predictive value (NPV), and positive or negative likelihood ratio for all of the study parameters in predicting EOS were calculated. In addition, multivariate analysis based on a stepwise logistic regression model was used to identify markers that independently predicted EOS.
Characteristics of the study population
During the study period, we treated 166 neonates with suspected EOS. Of these, three cases with viral infection, four cases with congenital malformations, and eight cases for which informed consent was not provided were excluded. Thus, a total of 151 neonates were included in the cohort with 68 in the infected group and 83 in uninfected group. Six patients in the infected group were identified by positive blood culture results (8.82%); these included two cases of Group B Streptococcus infection, two cases of Escherichia coli infection, one case of Enterococcus faecalis infection, and one case of Staphylococcus epidermidis infection. An overview of the basic characteristics of patients in each group is provided in Table 1. Apart from WBC, no significant differences in gestational age, birth weight, age at admission, sex distribution, or mode of delivery were observed between the two groups.
Differences in individual biomarkers between infected and uninfected groups
The ability of biomarkers to predict EOS was evaluated by comparing levels between the infected group (n = 68) and uninfected group (n = 83), as defined in the Materials and Methods. Comparison of levels of individual biomarkers between the infected and uninfected groups (Fig. 1) revealed that the infected group had higher levels of IL-27 (1.25 ng/mL vs. 0.88 ng/mL, P <0.01), IL-6 (136.6 pg/mL vs. 44.1 pg/mL, P <0.01), TNF-α (74.54 pg/mL vs. 64.34 pg/mL, P = 0.02), HSP 70 (605 μg/mL vs. 462 μg/mL, P <0.01), MMP-8 (70.5 μg/mL vs. 56.5 μg/mL, P = 0.02), PCT (8.4 ng/mL vs. 2.8 ng/mL, P <0.01), and CRP (6.0 mg/l vs. 2.0 mg/l, P <0.01). In contrast, plasma levels of IL-8 (643.1 pg/mL vs. 524.8 pg/mL, P = 0.09), granzyme B (9.1 pg/mL vs. 9.9 pg/mL, P = 0.90), MIP-1α (287.8 pg/mL vs. 347.4 pg/mL, P = 0.50), MIP-1β (399.4 pg/mL vs. 317.3 pg/mL, P = 0.19) did not differ significantly between the two groups.
Considering that levels of these cytokines may change considerably during the early postnatal period, we further analyzed differences in the levels of these markers between the infected and uninfected groups within the time intervals of 0 to 24, 24 to 48, and 48 to 72 h based on when blood samples were collected after birth (Fig. 1). All levels of these markers were measured in single samples obtained in different patients while suspected EOS was built. The results show that biomarker levels differed significantly between the infected and uninfected groups during certain intervals and not during others.
Predictive value of individual biomarkers for EOS
ROC curve analyses were performed for all of the individual biomarkers (Fig. 2 and Table 2). IL-27 performed well at distinguishing infected neonates from uninfected neonates, with an AUC of 0.747, and a cut-off point of >1 ng/mL had a sensitivity of 70.59% and a specificity of 71.08%. In addition, IL-6, TNF-α, HSP 70, MMP-8, PCT (for which the cut-off levels changed among different time intervals consistent with the findings of a previous study (22), see Supplementary Table 1, http://links.lww.com/SHK/A482), and CRP were significantly predictive of EOS with AUC values of 0.706, 0.614, 0.661, 0.607, 0.723, and 0.720, respectively. The ROC curve analyses for the remaining markers, IL-8, granzyme B, MIP-1α, and MIP-1β showed that these markers were not significant predictors of EOS.
A stepwise multivariate logistic regression model was applied to identify independent predictors of EOS among the tested biomarkers, excluding CRP, due to its use as a diagnostic criterion for EOS. Only IL-27 and PCT qualified for the model (Table 3), whereas the other biomarkers were not found to be significant predictors of EOS. IL-27 showed the strongest association with EOS with an odds ratio (OR) of 9.55 (3.55–25.67, P <0.01), followed by PCT (OR: 1.02 [1.00–1.04], P = 0.030).
To better understand the changes in IL-27 levels after birth, we measured the IL-27 level in cord blood from 33 healthy neonates, which could be regarded as the baseline level, and compared it with IL-27 levels within the first 72 h after birth (Fig. 3). Comparison of the baseline level to those in blood samples of neonates in the low-risk sepsis group during the different time intervals after birth (only one measurement in one neonate at one time point was outside the normal range) revealed that the initial concentration of 0.74 mg/mL did not change significantly within the first 72 h after birth (median levels: 0–24 h: 0.84 ng/mL; 24–48 h: 0.99 ng/mL; 48–72 h:0.70 ng/mL; P = 0.309).
Predictive value of biomarker combinations for EOS
Next, we evaluated the performance of combinations of markers based on multivariate analysis, namely IL-27, PCT, and CRP, the most commonly used biomarkers in clinic for predicting EOS (Table 4). We tested combinations including two and three of these markers. The results showed that the addition of IL-27 to PCT resulted in increased AUC values (from 0.723 to 0.792, P = 0.02). For CRP, PCT, and IL-27 together, the AUC value increased to 0.834, which was also higher than that for CRP together with PCT with AUC of 0.784 (P = 0.02).
EOS is still a leading cause of admission to neonatal wards, because symptoms are often wrongly interpreted due to their nonspecific nature and late occurrence (23). Moreover, EOS is primarily diagnosed based on laboratory tests (24), and effective and early biomarkers could improve early recognition of EOS and patients’ prognosis as well as help to avoid unnecessary exposure of neonates without EOS to antibiotics.
IL-27, composed of the EBI3 and IL-27p28 subunits, is mostly secreted by antigen-presenting cells upon exposure to microbial products and inflammatory stimuli and works as a regulator of T cells, macrophages, and neutrophils (7). Therefore, we hypothesized that IL-27 might be a promising marker for EOS. In this study, we demonstrated that an elevated plasma IL-27 level was strongly correlated with the risk for EOS independent of PCT and other markers and IL-27 had the highest predictive ability in our stepwise multivariate model. The AUC for the ability of IL-27 to distinguish infected and uninfected neonates was 0.747, which is slightly greater than that for PCT (AUC = 0.723). This finding is consistent with the conclusion of Hanna et al. (11) that IL-27 can identify bacterial infection in critically ill pediatric patients with an AUC of 0.75. Interestingly, Hanna et al. also found that IL-27 has a greater predictive value in patients with strictly positive blood culture than in those with infections in other body locations. We were unable to repeat such an evaluation, because a low percentage of neonates with EOS have a positive result on blood culture (5). However, considering that we detected IL-27 using the same method and the AUC value for IL-27 was similar in our study compared to that reported by Hanna et al., we believe that an elevated IL-27 level may strongly suggest bloodstream infection among neonates with suspected EOS. In addition, although the sensitivity and NPV were unsatisfactory for IL-27, the combined use of IL-27 and PCT (AUC = 0.792) showed greater predictive performance than PCT or IL-27 alone. Moreover, the sensitivity was increased to 98.53% and the NPV to 97.14% if either the IL-27 or PCT result was positive, which indicates these markers can reinforce each other for screening EOS.
Another biomarker, IL-6, performed well with an AUC of 0.706 in our study, and this finding is in agreement with previous research (25, 26). The AUC value for IL-6 was similar to that for IL-27, and this is likely because they belong to the same family and share the same receptor, gp130 (9, 27).
Strangely, the median levels of IL-27 observed in our study were lower than those in children with sepsis (11, 12). Several factors may account for the differences. Firstly, the inclusion criteria for infected neonatal and pediatric cohorts differed. The pediatric infected patients were classified as having sepsis based on a positive culture for known bacterial pathogens. In contrast, the neonatal infected patients were primarily classified based on perinatal risk factors, clinical manifestations, and laboratory investigations. Secondly, the immunological system of neonates differs that from the children, which may correspond to differences in the IL-27 response between neonates and children. Although data for IL-27 expression in neonates are limited, a previous mice study demonstrated that expression of the EBI3 and IL-27p28 genes significantly elevated from the fourth day of life, which suggests the IL-27 level may be lower during the neonatal period (28). In addition, we also measured the IL-27 level in cord blood in our study, and the median level of 0.70 ng/mL was similar to that in uninfected neonates from 0 to 72 h after birth. This finding demonstrated that our data and classification protocol are reliable. Also, even when detected using the same method, IL-27 levels in adults, both with and without sepsis, previously were found to be higher than those in children, which confirms levels of IL-27 likely change with age (12). Finally, another study also demonstrated that levels of markers such as pancreatic stone protein can differ between neonates and adults (16).
IL-8 and TNF-α have also been studied as potential biomarkers for EOS (15, 17, 25, 26). In the present study, TNF-α was significantly associated with EOS, and although IL-8 performed less well, the P value (P = 0.09) for the comparison of IL-8 levels in infected and uninfected neonates was almost significant. Given that the level of IL-8 was shown to vary tremendously in a previous study, this result is not surprising. Two markers rarely used in EOS patients, HSP 70 and MMP-8, showed trend-wise associations with EOS with AUCs of 0.661 and 0.607, respectively. HSP 70 has been shown to play a protective role in mice with sepsis and to be correlated with sepsis in adults (19, 29, 30). Our results also demonstrated that HSP 70 was a good biomarker for EOS in neonates. MMP-8 is an endopeptidase known to function as a collagenase. MMP-8 was found to be highly expressed both in genome-wide expression profiling and protein analysis in children with septic shock (31, 32). However, given that the AUC for MMP-8 was relatively low in the present study, further research is needed to explore the predictive value of MMP-8 in EOS. MIP-1α and MIP-1β regulate leukocyte activation and trafficking and are increased in adult patients with sepsis or septic shock (33). The absence of elevated MIP-1α and MIP-1β levels in the infected group in our study may be due to the fact that septic shock is a rare event in neonates with EOS compared with adults. Similarly, granzyme B, a kind of cytolytic effector molecule, is elevated in adults with severe sepsis but failed to predict EOS in our study (21).
The physiological increase in PCT during the first several days of life requires adjustment of the cut-off point with time as observed in a previous study (34), and we also observed similar physiological changes for IL-6, granzyme B, HSP 70, MMP-8, PCT, and CRP in our study. However, in the present study, we found that such adjustment of the cut-off levels for IL-27 at different time intervals may not be necessary.
The main strength for our study is that multiplex cytokine profiling improves the ability to compare the value of biomarkers because these markers were measured simultaneously in the same protocol and plate and could be easily measured by the same person. One major limitation of this study is that the number of cases of culture-proven sepsis was limited. However, this finding is in accordance with previous studies showing that the incidence of positive culture results in EOS patients ranges from 2% to 3.3% (5, 35). Additional limitations of this study include the relatively lower sensitivity and specificity of IL-27 and the relatively small level differences between the levels of IL-27 in the two groups in this small single center study. Thus, the value of IL-27 as a biomarker for EOS needs to be confirmed in an independent multicenter validation cohort. Finally, the study did not include blood samples from a healthy control group, because blood collection from healthy babies (except for the cord blood from healthy babies) did not meet the requirements for research ethics.
In conclusion, in this preliminary study, multiplex cytokine profiling identified IL-27 as a novel biomarker for EOS in neonates and offers significantly increased predictive ability when used in combination with PCT.
1. Liu L, Oza S, Hogan D, Perin J, Rudan I, Lawn JE, Cousens S, Mathers C, Black RE. Global, regional, and national causes of child mortality in 2000-13, with projections to inform post-2015 priorities: an updated systematic analysis. Lancet
2015; 385 9966:430–440.
2. Ganatra HA, Stoll BJ, Zaidi AK. International perspective on early-onset neonatal sepsis. Clin Perinatol
2010; 37 2:501–523.
3. Al-Taiar A, Hammoud MS, Cuiqing L, Lee JK, Lui K-M, Nakwan N, Isaacs D. Neonatal infections in China, Malaysia, Hong Kong and Thailand. Arch Dis Child Fetal Neonatal Ed
2013; 98 3:F249–F255.
4. Ohlsson A, Shah VS. Intrapartum antibiotics for known maternal Group B streptococcal colonization. Cochrane Database Syst Rev
2009; 3: CD007467.
5. Guerti K, Devos H, Ieven MM, Mahieu LM. Time to positivity of neonatal blood cultures: fast and furious? J Med Microbiol
2011; 60 (pt 4):446–453.
6. Chiesa C, Panero A, Osborn JF, Simonetti AF, Pacifico L. Diagnosis of neonatal sepsis: a clinical and laboratory challenge. Clin Chem
2004; 50 2:279–287.
7. Hunter CA, Kastelein R. Interleukin-27
: balancing protective and pathological immunity. Immunity
2012; 37 6:960–969.
8. Cao J, Xu F, Lin S, Song Z, Zhang L, Luo P, Xu H, Li D, Zheng K, Ren G. IL-27
controls sepsis-induced impairment of lung antibacterial host defence. Thorax
2014; 69 10:926–937.
9. Hunter CA. New IL-12-family members: IL-23 and IL-27
, cytokines with divergent functions. Nat Rev Immunol
2005; 5 7:521–531.
10. Wong HR, Cvijanovich NZ, Hall M, Allen GL, Thomas NJ, Freishtat RJ, Anas N, Meyer K, Checchia PA, Lin R, et al. Interleukin-27
is a novel candidate diagnostic biomarker
for bacterial infection in critically ill children. Crit Care
2012; 16 5:R213.
11. Hanna WJ, Berrens Z, Langner T, Lahni P, Wong HR. Interleukin-27
: a novel biomarker
in predicting bacterial infection among the critically ill. Crit Care
12. Wong HR, Lindsell CJ, Lahni P, Hart KW, Gibot S. Interleukin 27 as a sepsis diagnostic biomarker
in critically ill adults. Shock
2013; 40 5:382–386.
13. Samraj RS, Zingarelli B, Wong HR. Role of biomarkers in sepsis care. Shock
2013; 40 5:358–365.
14. Pierrakos C, Vincent JL. Sepsis biomarkers: a review. Crit Care
2010; 14 1:R15.
15. Prashant A, Vishwanath P, Kulkarni P, Sathya Narayana P, Gowdara V, Nataraj SM, Nagaraj R. Comparative assessment of cytokines and other inflammatory markers for the early diagnosis of neonatal sepsis-a case control study. PLoS One
2013; 8 7:e68426.
16. Schlapbach LJ, Graf R, Woerner A, Fontana M, Zimmermann-Baer U, Glauser D, Giannoni E, Roger T, Muller C, Nelle M, et al. Pancreatic stone protein as a novel marker for neonatal sepsis. Intensive Care Med
2013; 39 4:754–763.
17. Zhou M, Cheng S, Yu J, Lu Q. Interleukin-8 for diagnosis of neonatal sepsis: a meta-analysis. PLoS One
2015; 10 5:e0127170.
18. Buhimschi CS, Bhandari V, Hamar BD, Bahtiyar M-O, Zhao G, Sfakianaki AK, Pettker CM, Magloire L, Funai E, Norwitz ER. Proteomic profiling of the amniotic fluid to detect inflammation, infection, and neonatal sepsis. PLoS Med
2007; 4 1:e18.
19. McConnell KW, Fox AC, Clark AT, Chang NY, Dominguez JA, Farris AB, Buchman TG, Hunt CR, Coopersmith CM. The role of heat shock protein 70
in mediating age-dependent mortality in sepsis. J Immunol
2011; 186 6:3718–3725.
20. Arias MA, Jimenez de Bagues MP, Aguilo N, Menao S, Hervas-Stubbs S, de Martino A, Alcaraz A, Simon MM, Froelich CJ, Pardo J. Elucidating sources and roles of granzymes A and B during bacterial infection and sepsis. Cell Rep
2014; 8 2:420–429.
21. Napoli AM, Fast LD, Gardiner F, Nevola M, Machan JT. Increased granzyme levels in cytotoxic T lymphocytes are associated with disease severity in emergency department patients with severe sepsis. Shock
2012; 37 3:257–262.
22. Stocker M, Fontana M, El Helou S, Wegscheider K, Berger TM. Use of procalcitonin
-guided decision-making to shorten antibiotic therapy in suspected neonatal early-onset sepsis: prospective randomized intervention trial. Neonatology
2010; 97 2:165–174.
23. Qazi SA, Stoll BJ. Neonatal sepsis: a major global public health challenge. Pediatr Infect Dis J
2009; 28 1:S1–S2.
24. Polin RA. Committee on F, Newborn: management of neonates
with suspected or proven early-onset bacterial sepsis. Pediatrics
2012; 129 5:1006–1015.
25. Berner R, Niemeyer CM, Leititis JU, Funke A, Schwab C, Rau U, Richter K, Tawfeek MS, Clad A, Brandis M. Plasma levels and gene expression of granulocyte colony-stimulating factor, tumor necrosis factor-α, interleukin (IL)-1β, IL-6, IL-8, and soluble intercellular adhesion molecule-1 in neonatal early onset sepsis
. Pediatr Res
1998; 44 4:469–477.
26. Kurt A, Aygun AD, Godekmerdan A, Kurt A, Dogan Y, Yilmaz E. Serum IL-1β, IL-6, IL-8, and TNF-α levels in early diagnosis and management of neonatal sepsis. Mediators Inflamm
27. Kastelein RA, Hunter CA, Cua DJ. Discovery and biology of IL-23 and IL-27
: related but functionally distinct regulators of inflammation. Annu Rev Immunol
28. Kraft JD, Horzempa J, Davis C, Jung JY, Pena MM, Robinson CM. Neonatal macrophages express elevated levels of interleukin-27
that oppose immune responses. Immunology
2013; 139 4:484–493.
29. Delogu G, Bosco LL, Marandola M, Famularo G, Lenti L, Ippoliti F, Signore L. Heat shock protein (HSP 70) expression in septic patients. J Crit Care
1997; 12 4:188–192.
30. Singleton KD, Wischmeyer PE. Effects of HSP70
. 1/3 gene knockout on acute respiratory distress syndrome and the inflammatory response following sepsis. Am J Physiol Lung Cell Mol Physiol
2006; 290 5:L956–L961.
31. Solan PD, Dunsmore KE, Denenberg AG, Odoms K, Zingarelli B, Wong HR. A novel role for matrix metalloproteinase-8 in sepsis. Crit Care Med
2012; 40 2:379–387.
32. Wong HR, Cvijanovich N, Allen GL, Lin R, Anas N, Meyer K, Freishtat RJ, Monaco M, Odoms K, Sakthivel B. Genomic expression profiling across the pediatric systemic inflammatory response syndrome, sepsis, and septic shock spectrum. Crit Care Med
2009; 37 5:1558.
33. O’Grady NP, Tropea M, Preas HL, Reda D, Vandivier RW, Banks SM, Suffredini AF. Detection of macrophage inflammatory protein
)-1( and MIP
-( during experimental endotoxemia and human sepsis. J Infect Dis
1999; 179 1:136–141.
34. Chiesa C, Natale F, Pascone R, Osborn JF, Pacifico L, Bonci E, De Curtis M. C reactive protein and procalcitonin
: reference intervals for preterm and term newborns during the early neonatal period. Clin Chim Acta
2011; 412 (11–12):1053–1059.
35. Gómez B, Mintegi S, Benito J, Garcia D, Astobiza E. Blood culture and bacteremia predictors in infants less than three months of age with fever without source. Pediatr Infect Dis J
2010; 29 1:43–47.
Biomarker; early onset sepsis; interleukin-27; multiplex cytokine profiling; neonates; AUC; area under the curve; CI; confidence interval; CRP; creactive protein; EBI3; Epstein–Barr virus-induced gene 3; EOS; early onset sepsis; HSP70; heat shock protein 70; IL-27; interleukin-27; MIP; macrophage inflammatory protein; MMP-8; matrix metalloproteinase 8; PCT; procalcitonin; ROC; receiver operating characteristic; WBC; white blood cell count
Supplemental Digital Content
© 2017 by the Shock Society