- Similares en SciELO
versión impresa ISSN 0103-507X
Rev. bras. ter. intensiva vol.24 no.1 São Paulo enero/mar. 2012
Experimental Pathophysiology Laboratory and National Institute of Medical Translational Science and Technology, Health Sciences Postgraduate Program, Health Sciences Unit, Universidade do Extremo Sul Catarinense - Criciúma (SC), Brazil
OBJECTIVES: Neonatal sepsis is a complex syndrome involving an uncontrolled systemic inflammatory response associated with an infection. It may result in the dysfunction or failure of one or more organs or even death. Given its high incidence in premature neonates, the identification of prognostic factors to optimize the early diagnosis and therapeutic interventions are highly desirable. This objective study determine the relationship between inflammatory markers and oxidative parameters as prognostic factors in early neonatal sepsis.
METHODS: We conducted a prospective observational study by collecting data from 120 patients in the maternity unit of a university hospital. Preterm (<37 weeks of pregnancy) infants with at least one additional risk factor for neonatal sepsis were included. The levels of interleukin (IL)-6, IL-10, thiobarbituric acid reactive species (TBARS) and protein carbonyls and their association with sepsis severity were determined in the cord blood.
RESULTS: Levels of IL-6 and TBARS, but not IL-10 and protein carbonyls, demonstrated a mild to moderate correlation with the SNAPPE-II severity score (r=0.435, p=0.02 and r = 0.385, p = 0.017, respectively). No correlations were found between these markers and mortality.
CONCLUSION: TBARS and IL-6 have a mild to moderate correlation with SNAPPE-II, but none of the studied markers were able to predict mortality in our sample.
Keywords: Sepsis/blood; Infant, newborn; Fetal blood/chemistry; Intensive care units, neonatal; Oxidative stress; Interleukin-6; Interleukin-10
Sepsis is a complex syndrome related to a systemic inflammatory response with multiple manifestations that may cause the dysfunction or failure of one or more organs or even death.(1) Early neonatal sepsis occurs within the first six days of life, whereas late neonatal sepsis occurs after the first week of life (>6 days). Differentiating early from late neonatal sepsis is clinically important because in early neonatal sepsis, the infectious organisms are acquired during the delivery, whereas in late sepsis the infecting organisms are mostly acquired after birth, from either hospital or community sources.(2,3)
The identification of patients who are at increased risk is challenging in critically ill populations.(4,5) Most of the clinical decisions in caring for patients with severe sepsis are based on clinical and laboratory data,(4) which are often not very accurate. In addition, the pathophysiology of systemic inflammation could be better understood; therefore, identifying more accurate predictors of severity is important.(4,6)
Several parameters have been suggested to be diagnostic of early sepsis, including the levels of some cytokines such as interleukin-1 (IL-1), interleukin-6 (IL-6), interleukin-8 (IL-8) and interleukin-10 (IL-10), tumor necrosis factor (TNF) and its soluble receptor (TNF receptor), acute phase proteins (C-reactive protein) and procalcitonin.(7-9) We have recently shown a correlation between oxidative parameters and the development of neonatal sepsis;(10) however, the medical literature has little information on the early predictors of severity in neonatal sepsis. Therefore, this study aimed to determine the correlation between inflammatory and oxidative parameters in cord blood and the severity of early neonatal sepsis.
This cross-sectional case-control, retrospective study collected cord blood from 120 consecutive premature newborns in the maternity ward of Hospital Nossa Senhora da Conceição, Tubarão, SC, Brazil, from February to September, 2005, as previously described.(10) The data were primarily collected to determine the correlation of biomarkers with the diagnosis of neonatal sepsis.(10) For the present study, the data from the original study were used. This study was approved by the ethics committee of Universidade do Extremo Sul Catarinense - UNESC, under the protocol number 168/2005.
Cord blood samples were collected by pressing the blood from the umbilical cord into a glass tube after the delivery and prior to discarding the placenta. The samples were centrifuged at 1000 g for 10 minutes, and liquid nitrogen was added. The samples were stored at -80ºC until the analysis.
Inclusion and exclusion criteria
Premature neonates (<37 weeks of pregnancy) with at least one additional risk factor for neonatal sepsis (premature rupture of membranes >18 hours before delivery, maternal colonization with group B streptococci, intrapartum fever with an axillary temperature above 37.5ºC, or chorioamnionitis) were included. Some of the parameters assessed in this study may be increased in children born to mothers with preeclampsia and/or diabetes and in babies with restricted intrauterine growth. Therefore, these cases were excluded.
Diagnosis of sepsis
Based on the 2002 Consensus Conference on Pediatric Sepsis: definitions for sepsis and organ dysfunction in pediatrics, sepsis was considered to be proven when a blood culture or culture from a normally sterile site was positive for a likely pathogen, and sepsis was considered to be likely if the bacterial cultures were negative but a clinical and biological infection-associated syndrome was present.
The clinical and biological evidence of sepsis includes the following: hyper-leukocytosis, defined as a blood leukocyte count >25,000/mm3; leukopenia, defined as a blood leukocyte count < 4,000/mm3; neutropenia, defined as a blood neutrophil count < 1,300/mm3; thrombocytopenia, defined as a platelet count < 150,000/mm3; hyperglycemia, defined as blood glucose > 7 mmol/L and a C-reactive protein > 20 g/L. Forty children met the inclusion criteria (proven sepsis, n = 20; likely sepsis, n = 20) and were included in the sepsis group.
The severity of sepsis was determined based on five variables measured at the time of diagnosis: C-reactive protein, lactate, bicarbonate, blood glucose and SNAPPE-II severity score.(11) In addition, the patients were followed until death or discharge from the ICU to determine the mortality rate in the intensive care unit.
Inflammatory and oxidative parameters in cord blood
Oxidative stress was determined based on the levels of thiobarbituric acid reactive substances (TBARS) and protein carbonyl. To measure the TBARS, the samples were mixed with 1 mL 10% trichloroacetic acid and 1 mL 0.67% thiobarbituric acid (Sigma, St. Louis, MO) and then heated in a boiling water bath for 15 minutes. Malondialdehyde equivalents were determined by the absorbance at 535 nm, using 1,1,3,3-tetramethoxypropane (Sigma, St. Louis, MO) as the external standard. BHT was added to the samples to prevent malondialdehyde production during the analysis.(12) The results were expressed as malondialdehyde equivalents (nmol) per protein milligram.
Oxidative protein damage was assessed by the determination of carbonyl groups based on the dinitrophenylhydrazine reaction. In summary, proteins were precipitated by adding trichloroacetic acid 20% and dissolved in dinitrophenylhydrazine (Sigma, St. Louis, MO). Absorbance was read at 370 nm,(13) and the results are shown as nmol/mg protein.
The analysis of inflammatory parameters was performed using commercial enzymatic immunosorbent kits for the detection of IL-6 and IL-10 (Quantikine, R & D Systems Europe, Abingdon, UK); the results are expressed as pg/mg protein.
The correlation between the parameters and mortality was performed using the Student's t test. The correlation between the parameters was assessed with Pearson's correlation. The normal distribution of the variables was tested with the Kolmogorov-Smirnov test. P values of 0.05 or less were considered statistically significant. The sample's statistic power was calculated a posteriori (as the study was originally designed for a different purpose). A sample with 10 patients in the non-surviving group, with an alpha = 0.05 and 0.80 power, would allow the detection of a mean 20% difference, projecting a sigma of 0.15 for the assessed parameters.
A total of 40 patients were included, 20 with proven neonatal sepsis and 20 with likely neonatal sepsis. Of these patients, 30 patients survived and 10 died. The relevant clinical variables have been previously published.(10)
The TBARS level had a mildly significant correlation with the SNAPPE-II severity score (r = 0.385 / p = 0.017) but not with the serum blood glucose (r = 0.288 / p = 0.08), C-reactive protein (r = 0.027 / p = 0.87), lactate (r = -0.233 / p = 0.49) or bicarbonate (r = 0.10 / p = 0.95). IL-6 had a moderate correlation with the SNAPPE-II severity score (r = 0.435 / p = 0.02) but not with bicarbonate (r = 0.341 / p = 0.09), C-reactive protein (r = 0.124 / p = 0.21), lactate (r = 0.221 / p = 0.25) or blood glucose (r = 0.12 / p = 0.32). No significant correlation was found between carbonylated protein levels and any of the assessed severity parameters, as was found for the IL-10 levels.
None of the assessed parameters were associated with mortality in these patients (Figure 1).
The quest for an ideal marker that is highly sensitive and specific is a challenge for healthcare professionals. This study was unable to determine a relevant role for the assessed parameters in terms of prognosis and severity in patients with early neonatal sepsis. Although a significant correlation was found for IL-6 and TBARS with the SNAPPE-II score, this correlation was mild to moderate, and none of the markers were able to predict the patients' outcome.
Early markers of neonatal sepsis have been studied in recent years, and cord blood interleukins are promising. Increased cord blood levels of IL-6 and IL-1β were neither able to confirm nor preclude, with a high level of confidence, neonatal sepsis.(10,14-17) Recently, an algorithm including IL-6 levels was shown to be highly effective in identifying newborns at a high probability of developing early neonatal sepsis.(18) IL-6 is considered an important systemic inflammatory response marker and is characterized as an "alarm" cytokine as its serum concentrations quickly peak during infections. IL-6 levels increases up to two to four days before symptoms manifest, and its levels remain stable during the first hours, dropping 24 hours after sepsis is established.(17) Maternal IL-6 levels are able to predict premature delivery and unfavorable outcomes in this population, especially intraventricular hemorrhage.(19) Although cord blood IL-6 is apparently promising as a diagnostic marker,(10) our results do not support its use as an outcome predictor. As some of the parameters have up to a moderate correlation with classical severity markers, they may be used to anticipate the disease severity even before outright clinical or blood chemistry signs of infection.
IL-10 is the main negative regulator of the innate immune response. It acts as an inflammatory cytokine by negative feedback, i.e., by inhibiting proinflammatory cytokine synthesis. High IL-10 concentrations reduce monocyte production of TNF-α, IL-6 and IL-8 while activating IL-1Ra production (positive feedback). IL-10 suppression results in increased serum TNF-α and IL-6 levels, suggesting that this cytokine is highly anti-inflammatory; indeed, when endogenous IL-10 is neutralized by monoclonal antibodies, mortality is increased.(14,16,17) The magnitude of the IL-10 response apparently correlates with the severity of the inflammatory process and the concentration of proinflammatory cytokines, leading to TNF-α activation. Increased IL-10 and IL-6 levels in a newborn are indicative of an infection, and IL-10 remains high for 48 hours after the infection manifests itself.(18) However, our results do not point to a role of cord blood IL-10 determination as a prognostic marker in neonatal sepsis.
Additionally, sepsis promotes the unbalanced production of oxidant and anti-oxidant substances, causing an excess of free oxygen radicals. These molecules are meant to destroy the infective agent but are not specific substances and may instead harm the patient.(20) The production of reactive oxygen species (ROS) is increased during sepsis, which may lead to tissue damage.(21) The production of ROS during neonatal sepsis has been documented, with an evident imbalance between pro- and anti-oxidants,(10,20,22) and the levels of these molecules may be related to the development of neonatal sepsis.(10) Notwithstanding, we were unable to demonstrate a relationship between the severity of neonatal sepsis and oxidative stress markers.
The interpretation of our results should take into consideration some limitations. We had no information on the time between the delivery and the first signs of sepsis. As these parameters can change quickly, this timing may have influenced the interpretation of our results. Second, this sample included patients with both proven and likely sepsis; therefore, the results should be analyzed in the light of this limitation. As we have previously shown,(10) the studied population had significant differences in birth weight and one-minute APGAR scores; therefore, this is a heterogeneous population, which may have influenced the results. Third, our sample had the power to detect a 20% difference in the assessed parameters; because the differences detected were below this level, this sample size could explain our negative results in terms of the outcome. However, we believe that changes less than 20% in these parameters lack biological meaning; therefore, we consider this limitation to be of a minor importance.
Thiobarbituric acid reactive substances and IL-6 have mild and moderate correlations with the SNAPPE-II severity score, respectively, but not with mortality. In this context, these factors can be early warning markers of severity but are not useful for determining prognosis, even before the manifestation of the infection.
1. Carvalho PRA, Trotta EA. Avanços no diagnóstico e tratamento da sepse. J Pediatr (Rio J). 2003;79 (Supl 2):S195-204. [ Links ]
2. Goldstein B, Giroir B, Randolph A; International Consensus Conference on Pediatric Sepsis. International pediatric sepsis consensus conference: definitions for sepsis and organ dysfunction in pediatrics. Pediatr Crit Care Med. 2005;6(1):2-8. Review. [ Links ]
3. Bizarro MJ, Raskind C, Baltimore RS, Gallagher PG. Seventy-five years of neonatal sepsis at Yale: 1928-2003. Pediatrics. 2005;116(3):595-602. [ Links ]
4. Rubin LG, Sánchez PJ, Siegel J, Levine G, Saiman L, Jarvis WR; Pediatric Prevention Network. Evaluation and treatment of neonates with suspected late-onset sepsis: a survey of neonatologists' practices. Pediatrics. 2002;110(4):e42. [ Links ]
5. Holmes A, Doré CJ, Saraswatula A, Bamford KB, Richards MS, Coello R, Modi N. Risk factors and recommendations for rate stratification for surveillance of neonatal healthcare-associated bloodstream infection. J Hosp Infect. 2008;68(1):66-72. [ Links ]
6. Bozza FA, Salluh JI, Japiassu AM, Soares M, Assis EF, Gomes RN, et al. Cytokine profiles as markers of disease severity in sepsis: a multiplex analysis. Crit Care. 2007;11(2):R49. [ Links ]
7. Marshall JC, Reinhart K; International Sepsis Forum. Biomarkers of sepsis. Crit Care Med. 2009;37(7):2290-8. Review. [ Links ]
8. Miura E, Silveira RC, Procianoy RS. Sepse neonatal: diagnóstico e tratamento. J Pediatr (Rio J). 1999;75(Supl 1):S57-62. [ Links ]
9. Cianciarullo MA, Ceccon MEJ, Yamamoto L, Del Negro GMB, Okay TS. Mediadores pró-inflamatórios e antinflamatórios na sepse neonatal: associação entre homeostase e evolução clínica. Rev Bras Crescimento Desenvolv Hum. 2008;18(2):135-47. [ Links ]
10. Cancelier AC, Petronilho F, Reinke A, Constantino L, Machado R, Ritter C, Dal-Pizzol F. Inflammatory and oxidative parameters in cord blood as diagnostic of early-onset neonatal sepsis: a case-control study. Pediatr Crit Care Med. 2009;10(4):467-71. [ Links ]
11. Richardson DK, Corcoran JD, Escobar GJ, Lee SK. SNAP-II and SNAPPE-II: Simplified newborn illness severity and mortality risk scores. J Pediatr. 2001;138(1):92-100. [ Links ]
12. Esterbauer H, Cheeseman KH. Determination of aldehydic lipid peroxidation products: malonaldehyde and 4-hydroxynonenal. Methods Enzymol. 1990;186:407-21. [ Links ]
13. Levine RL, Williams JA, Stadtman ER, Shacter E. Carbonyl assays for determination of oxidatively modified proteins. Methods Enzymol. 1994;233:346-57. [ Links ]
14. Ng PC. Diagnostic markers of infection in neonates. Arch Dis Child Fetal Neonatal Ed. 2004;89(3):F229-35. [ Links ]
15. Malamitsi-Puchner A, Protonariou E, Boutsikou T, Makrakis E, Sarandakou A, Creatsas G. The influence of the mode of delivery on circulating cytokine concentrations in the perinatal period. Early Hum Dev. 2005;81(4):387-92. [ Links ]
16. 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(6):3765-72. [ Links ]
17. Zeitoun AA, Gad SS, Attia FM, Abu Maziad AS, Bell EF. Evaluation of neutrophilic CD64, interleukin 10 and procalcitonin as diagnostic markers of early- and late-onset neonatal sepsis. Scand J Infect Dis. 2010;42(4):299-305. [ Links ]
18. Buhimschi CS, Bhandari V, Dulay AT, Nayeri UA, Abdel-Razeq SS, Pettker CM, et al. Proteomics mapping of cord blood identifies haptoglobin "switch-on" pattern as biomarker of early-onset neonatal sepsis in preterm newborns. PLoS One. 2011;6(10):e26111. [ Links ]
19. Sorokin Y, Romero R, Mele L, Wapner RJ, Iams JD, Dudley DJ, et al. Maternal serum interleukin-6, C-reactive protein, and matrix metalloproteinase-9 concentrations as risk factors for preterm birth <32 weeks and adverse neonatal outcomes. Am J Perinatol. 2010;27(8):631-40. [ Links ]
20. Saugstad OD. Oxidative stress in the newborn--a 30-year perspective. Biol Neonate. 2005;88(3):228-36. [ Links ]
21. Ritter C, Andrades M, Frota Júnior ML, Bonatto F, Pinho RA, Polydoro M, et al. Oxidative parameters and mortality in sepsis induced by cecal ligation and perforation. Intensive Care Med. 2003;29(10):1782-9. [ Links ]
22. Mrozek JD, Georgieff MK, Blazar BR, Mammel MC, Schwarzenberg SJ. Effect of sepsis syndrome on neonatal protein and energy metabolism. J Perinatol. 2000;20(2):96-100. [ Links ]
Corresponding author: Submitted on February 9, 2012 Conflicts of interest: None. This study was developed at the Hospital Nossa Senhora da Conceição - Tubarão (SC), Brazil and the Experimental Pathophysiology Laboratory, Universidade do Extremo Sul Catarinense - Criciúma (SC), Brazil.
Av. Universitária, 1105
Zip Code: 88806-000 Criciúma (SC), Brazil.
Accepted on March 6, 2012
*These authors contributed equally to this article.
Submitted on February 9, 2012
Conflicts of interest: None.
This study was developed at the Hospital Nossa Senhora da Conceição - Tubarão (SC), Brazil and the Experimental Pathophysiology Laboratory, Universidade do Extremo Sul Catarinense - Criciúma (SC), Brazil.