Indian Journal of Pathology and Microbiology
Home About us Instructions Submission Subscribe Advertise Contact e-Alerts Ahead Of Print Login 
Users Online: 5539
Print this page  Email this page Bookmark this page Small font sizeDefault font sizeIncrease font size

ORIGINAL ARTICLE Table of Contents   
Year : 2010  |  Volume : 53  |  Issue : 3  |  Page : 494-497
Serum tumor necrosis factor α and C-reactive protein in pediatric patients with sepsis and its correlation with microbiologic findings

1 Department of Microbiology, Maulana Azad Medical College, New Delhi-110 002, India
2 Jawaharlal Nehru Medical College, AMU, Aligarh, India

Click here for correspondence address and email

Date of Web Publication22-Oct-2010


Objective: To study the association of tumor necrosis factor-a (TNF-a) and C - reactive protein (CRP) with microbiologically documented cases of sepsis versus clinically documented cases of sepsis. Materials and Methods: Seventy nine pediatric patients with sepsis were studied. Relevant specimens were processed for bacterial or fungal etiology. TNF-a was detected by enzyme immunoassay and CRP was detected by latex agglutination. Thirty healthy cases were included in the study to establish baseline TNF-α levels. Results: Forty two (53.2%) patients had a microbiologically documented sepsis. Among Gram negative bacilli Escherichia coli was the most common isolate followed by Klebsiella spp. Staphyloccus aureus and Streptococcus pneumoniae predominated among the Gram positive cocci. Patients with a positive culture had significantly higher TNF-α levels than patients with a negative culture (70pg/ml vs. 33 pg/ml P < 0.01). Further, pure gram negative infection correlated with significantly higher TNF-α levels than pure (P < 0.01) gram positive infection. The CRP values did not highlight these differences significantly. Conclusions: TNF-α level was significantly raised in patients with sepsis. TNF-a levels were raised significantly in culture positive cases in general and in Gram negative infections in particular. Serum TNF-α was a more sensitive marker for different categories of sepsis compared to CRP and microbiology culture.

Keywords: C - reactive protein, sepsis, tumor necrosis factor-α

How to cite this article:
Kumar S, Rizvi M. Serum tumor necrosis factor α and C-reactive protein in pediatric patients with sepsis and its correlation with microbiologic findings. Indian J Pathol Microbiol 2010;53:494-7

How to cite this URL:
Kumar S, Rizvi M. Serum tumor necrosis factor α and C-reactive protein in pediatric patients with sepsis and its correlation with microbiologic findings. Indian J Pathol Microbiol [serial online] 2010 [cited 2021 Dec 9];53:494-7. Available from: https://www.ijpmonline.org/text.asp?2010/53/3/494/68290

   Introduction Top

The sepsis syndrome is a complex systematic inflammatory condition associated with infection. In spite of great strides on the therapeutic and technological fronts in the care of these patients, the average mortality rate of sepsis syndrome remains approximately 40% and the incidence of this disorder continues to increase. [1],[2] It is not the infective pathogen that directly causes the sepsis syndrome and corresponding high mortality associated with severe sepsis but the host response to the pathogen. Bacteremia and endotoxemia elicit the production of a cascade of endogenous mediators that cause metabolic and immunologic host systemic inflammatory response. Both Gram negative and Gram positive bacteria as well as fungi and exogenous microbial components can initiate this cascade. [3],[4] However, its diagnosis continues to remain a major challenge. Microbiological diagnosis of sepsis is often difficult and can take approximately three days at the earliest and seven days at the latest. Demonstration of certain elevated cytokines can indirectly point to sepsis and aggressive treatment can be instituted rapidly. Cytokine levels can be monitored by enzyme linked immunosorbent assay (ELISA) and results can be made available on the same day leading to informed patient care. In this work, we studied the relevance of serum tumor necrosis factor a (TNF-α) and C-reactive protein (CRP) as diagnostic markers of sepsis in pediatric patients with septicaemia. Variations in TNF -a levels in culture positive cases as against culture negative cases were assessed along with variations with etiology.

   Materials and Methods Top

One hundred and ten cases in the pediatric age group (18 days-13 years) were enrolled in the study, of which 80 cases with septicemia constituted the study group and 30 healthy cases belonged to the control group. Children in the control group were age and sex matched with those in the study group to be able to compare the serum TNF alpha levels in similar age groups.

The criteria for establishing sepsis in the study group were (1) objective signs of acute infection (2) >3 recognized signs of systemic inflammatory response and (3) evidence of ≥ 2 organ dysfunctions. [3] 0 Cultures of blood and specimens from other relevant sites were obtained to identify a microbial cause of sepsis. All samples were cultured on blood agar and MacConkey agar and isolates were identified by standard microbiologic procedure. [5]

Blood samples were collected prior to initiation of antibiotic therapy. Using aseptic technique, 0.5ml to 2ml of blood was collected from children by venipuncture and inoculated into blood culture bottles containing 5 ml brain heart infusion broth. The bottles were incubated at 37oC for seven days and examined daily for bacterial growth. Subcultures were performed on the first, second, third, fifth and seventh days of incubation on chocolate agar, 5% sheep blood agar and MacConkey agar.

Three micro liters of blood for estimation of TNF alpha and CRP was obtained from patients at the same time. The sample was collected in a clean dry vial and allowed to clot for 15 minutes. The clot was then broken and the sample centrifuged for separation of serum. The serum was stored at -70oC till the test was put up.

Tumor Necrosis Factor-α Assay

Levels of TNF-α in the sera were estimated by enzyme immunoassay [Titerzyme TNF-α EIA kit Perceptive Biosystems Inc., Framingham, MA 01701] following the manufacture's instructions. Standard curve was prepared against which the sample readings were plotted. The lowest detectable value for TNF-α was 3.72 pg/ml and the highest detectable limit was 1650 pg/ml.

Assay For C-Reactive Protein

CRP analysis was performed by latex agglutination.

Ethical Clearance

Informed consent was obtained from the parents of the children.

Statistical Procedures

Comparison of cytokine concentrations were made using analysis of variance techniques employing log transformed data. The distribution of residual of each model was tested for normality by Shapiro Wilkes Test. When the normality assumption was violated, a non parametric test (Wilcoxon) was applied to confirm the results.

   Results Top

The mean age of the patients in the study group was two years (8 days-13 years). Thirty nine children were below one year of age, 25 were between one to five years and the remaining 15 were beyond five years of age. Sixty seven (61.4%) were male children

CRP and TNF-α in the Control Group

CRP was undetectable in 29/30 (96.6%) of the healthy children. TNF-α levels (mean ± SD) were 20.6 ± 6.1 in the control group, the range being 3.72 pg/ml to 26.4 pg/ml. Cut off value of positive TNF-α was 27 pg/ml.

CRP and TNF-α in the study group

Forty two of the 79 (53.2%) samples exhibited positive growth in bacterial or fungal cultures. Sensitivity of blood culture was 53.1%, specificity 100%, positive predictive value (PPV) was 100% and negative predictive value (NPV) was 44.7%. CRP levels were found to be elevated in 53/79 (67%) cases. Sensitivity of CRP was 67%, specificity 96.6%, PPV 98.1% and NPV was 52.7%. TNF-α levels were elevated in 67/79 (84.8%) cases. p < 0.05. Sensitivity of TNF-α was 84.8%, specificity 100%, PPV 100% and NPV 70.7%. Thus TNF-α had the highest sensitivity for diagnosis of sepsis with 100% specificity while blood culture had the lowest sensitivity but with a comparable specificity. Sensitivity of CRP was higher than that of blood culture with slightly lower specificity . All three markers (CRP, TNF-α and culture) were positive in forty one cases (51.8 %). TNF-α and CRP together identified 73 cases of sepsis (92.4%).

Microbiologic Findings

Overall, 47 patients (58.75%) had microbiologically documented infection while no infectious cause was identified in 33 patients (41.25 %). Among the culture positive patients, 32 cases (68%) had gram negative bacterial infection. 12 (26.3%) had gram positive infection and 3(4.7%) had fungal infection. Age distribution according to type of infection is given in [Table 1]. Among Gram negative bacteria, E. coli was clearly the single, most common isolate, (37.9%) followed by Klebsiella, Enterobacter, Serratia and Citrobacter spp. in descending order. Staphylococcus aureus was the most common gram positive isolate (45%) followed by Streptocccus pneumoniae (27.2%). Distribution of microbial isolates is given in [Table 2]. Clinically, majority of the patients with Gram negative infection had more severe sepsis, 21 (72.4%) compared to those with gram positive infection, 5 (45.4%) P< 0.05 when the following parameters fever, leucocytosis or leucopenia, hypotension, increased cardiac output and decreased peripheral vascular resistance were compared.
Table 1 :Age distribution according to culture results

Click here to view
Table 2 :Distributi on of microbial isolates in patients with sepsis

Click here to view

Of the five patients of bacteriologically confirmed cases of sepsis who expired, four had Gram negative infection.

Association of Serum Tnf-a Concentrations with Microbiologic Findings

Patients with a positive culture had significantly higher mean TNF-α levels than those from whom no pathogen was isolated: 83 pg/ml ± 12.1 versus 33 pg/ml ± 9.17 P = 0.001 [Table 3]. Among patients with a positive bacterial culture, TNF-α levels were significantly higher in those with Gram negative infection (83 pg/ml + 11.12) than in patients with Gram positive infection (52pg/ml ± 7.21) P < 0.01 or fungal infection (47pg/ml ± 3.42). The area under the receiver operating characteristic (ROC) curve (AUC) for TNF-α was 0.98.
Table 3 :TNF-α¿ levels in patients with culture positive and negative results

Click here to view

The mean CRP value for Gram negative infection was slightly higher than for Gram positive infections, 59 mg/ml versus 47 mg/ ml, the difference not being statistically significant. The mean CRP for culture negative cases was 45 mg/ml.

   Discussion Top

Diagnosis of neonatal sepsis may be difficult because clinical presentations are often nonspecific, bacterial cultures are time-consuming and other laboratory tests lack sensitivity and specificity. In this study, we investigated the role of CRP and tumor necrosis factor-alpha in diagnosing sepsis in pediatric patients. Evaluation of cytokines as early markers for sepsis has been well documented. [6] Silveira and Procianoy [7] reported that TNF-α and IL-6 can be utilized for early diagnosis of neonatal septicaemia. In the present study, we have used a two pronged approach to evaluate the association of TNF-α and CRP, one with both clinically and microbiologically documented cases of sepsis and the other with only clinically documented cases of sepsis.

TNF-α proved to be a far more sensitive marker than CRP and culture for early diagnosis of sepsis and disease severity. Similar results have been reported by Kocabas et al. [8] Culture was not only the least sensitive of the three methods but also had the longest turn around time.

Further, varying TNF-α levels can be utilized to distinguish between infection by gram positive and gram negative infection and thus a more informed empirical treatment can be instituted.

The highest TNF-α values were seen in patients with Gram negative infections. Gram positive infections also produce a marked TNF-α response, although not to the same extent as gram negative infections, the difference being statistically significant. Thus, TNF-α levels at the time of hospitalization of the patient may be used to distinguish gram negative from Gram positive infections. There is growing appreciation that the host recognition pathways for Gram negative and Gram positive microbial products are markedly different and therefore likely to induce a variant host response. Unlike Gram negative bacteria, which are recognized primarily by their shed or membrane associated endotoxin or LPS, the Gram positive microbes are recognized by the host through contact with membrane peptidoglycans, lipoteichoic acid, or soluble extracellular toxins. [9] In this study patients with Gram negative infection had a far more severe sepsis than patients with Gram positive infection when parameters like fever, leucocytosis or leucopenia, hypotension, increased cardiac output, and decreased peripheral vascular resistance were compared.

Patients with unexplained septicemia also demonstrated raised TNF-α levels-well above the cut off value. Thus TNF-α levels were raised in all groups of patients with septicemia and proved to be a more sensitive marker than CRP and culture in diagnosis of sepsis. Using the receiver operative characteristics curve, it was found that TNF-α cut-off value of 27 pg/ml was excellent in identifying true patients with sepsis with a sensitivity of 84.8% and specificity of 100%. These results were significantly better than blood culture and much better than CRP results. However, if TNF-α and CRP are performed in tandem, the sensitivity increases to 92.4%.

In the present study, 53.2% of patients had a microbiologically documented infection. TNF-α levels were significantly higher in patients with microbiologically documented cases than in patients with clinically documented septicemia, a finding in conformity with other reports. [10],[11] Thus, it appears that the distinction between clinically documented and microbiologically documented septicemia is of far greater import than has been previously appreciated. The higher TNF-α levels in microbiologically documented septicemia point to greater morbidity and mortality in patients and such patients with should be classified in a high risk category. TNF-α appears to be an excellent surrogate marker of sepsis as even in clinically documented cases of sepsis the TNF-α levels were much higher than the levels observed in healthy controls.

The significant difference in TNF-α levels between patients with Gram negative and Gram positive infections again stresses the differences in host responses between Gram negative and Gram positive infection and this should be borne in mind while formulating supplementary and adjunctive therapeutic agents. TNF-α levels can be estimated on the very day of onset of sepsis while culture reports can take 48 to 72 hrs. This is an unacceptably long turn around time in critically ill patients. TNF-α values can provide useful information regarding sepsis and nature of infecting microbe, thus leading to more informed empiric management of such cases. This is a baseline study. Studies with larger sample size will shed more light on the issue.

   References Top

1.Bone RC, Fisher CJ Jr, Clemmer TP, Slotman GJ, Metz CA, Balk RA. Sepsis syndrome: A valid clinical entity. Crit Care Med 1989;17:389-93.  Back to cited text no. 1      
2.Parrillo JE, Parker MM, Natanson C, Suffredini AF, Danner RL, Cunnion RE, et al. Septic shock in humans: Advances in the understanding of pathogenesis, cardiovascular dysfunction and therapy. Ann Intern Med 1990;113:227-42.  Back to cited text no. 2      
3.Bone RC. Pathogenesis of sepsis. Ann Intern Med 1991;115:457-69.  Back to cited text no. 3      
4.Abraham E, Glauser MP, Butler T, Garbino J, Gelmont D, Leterre PF, et al. p55tumor necrosis receptor fusion protein in the treatment of patients with severe sepsis and septic shock: A randomized controlled multicenter trial, JAMA 1997; 277:1531-8.  Back to cited text no. 4      
5.Collee JG, Miles RS, Watt B. Tests for the identification of bacteria in Mackie and McCartney Practical Medical Microbiology. In: Collee JG, Fraser AG, Marmion BP, Simmons A, editors. 24 th ed. Churchill Livingstone, Edinburgh, Ch. 7, pp 131-49.  Back to cited text no. 5      
6.de Bont ES, Martens A, van Raan J, Samson G, Fetter WPF, Okken A, et al. Diagnostic value of plasma levels of tumor necrosis factor-a (TNF-a) and interleukin-6 (IL-6) in newborns with sepsis. Acta Pediatr 1994;83:696-9.  Back to cited text no. 6      
7.Silveira RC, Procianoy. Evaluation of interleukin-6, tumor necrosis factor and interleukin I for early diagnosis of neonatal sepsis. Acta Pediatr 1999;88:647-50.  Back to cited text no. 7      
8.Kocabaώ E, Sarikηioπlu A, Aksaray N, Seydaoπlu G, Seyhun Y, Yaman A. Role of procalcitonin, C-reactive protein, interleukin-6, interl eukin -8 and tumor necrosis factor-alpha in the diagnosis of neonatal sepsis. Turk J Pediatr 2007;49:7-20.  Back to cited text no. 8      
9.De Kimpe SJ, Kengatharan M, Thiemermann C, Vane JR. The cell wall components peptidoglycan and lipoteichoic acid from Staphylococcus aureus act in synergy to cause shock and multiple organ failure. Proc Natl Acad Sci USA 1995;92:10359-63.  Back to cited text no. 9      
10.Cohen J, Abraham E. Microbiologic findings and correlation with serum tumor necrosis factor-a in patients with severe sepsis and septic shock. J Infect Dis 1999;180:116-21.  Back to cited text no. 10      
11.Lemm G, Carlet J, Cohen J. Cytokine levels in patients with the sepsis syndrome treated with a monoclonal antibody to human TNF-a (INTERSEPT trial). Clin Intensive Care 1995;6:68.  Back to cited text no. 11      

Correspondence Address:
Meher Rizvi
Department of Microbiology, Jawaharlal Nehru Medical College, AMU, Aligarh
Login to access the Email id

Source of Support: None, Conflict of Interest: None

DOI: 10.4103/0377-4929.68290

Rights and Permissions


  [Table 1], [Table 2], [Table 3]

This article has been cited by
1 Biomarkers for Sepsis: A Review with Special Attention to India
George E. Nelson,Vidya Mave,Amita Gupta
BioMed Research International. 2014; 2014: 1
[Pubmed] | [DOI]
2 Diagnostic and prognostic values of high sensitive C- Reactive Protein, Tumor Necrosis Factor and Interleukin-1 β in Neonatal Sepsis
El-Wakeel, M.A., El-Kassas, G.M., Fathy, G.A., El Wakkad, A.S.E., Sebaii, H.M.R., El Zayat, S.M.R., Mohamed, S.I.
Australian Journal of Basic and Applied Sciences. 2012; 6(3): 224-228
3 Study on the influence factors of neonatal CRP
Hua, S., An, S., Wu, Z., Lu, Y., Feng, Z.
HealthMED. 2012; 6(1): 131-135


    Similar in PUBMED
   Search Pubmed for
   Search in Google Scholar for
 Related articles
    Email Alert *
    Add to My List *
* Registration required (free)  

    Materials and Me...
    Article Tables

 Article Access Statistics
    PDF Downloaded130    
    Comments [Add]    
    Cited by others 3    

Recommend this journal