Printer Friendly

Early diagnosis of neonatal sepsis through hematological and biochemical markers.

Byline: Sarah Arif, Ayesha Ehsan, Mizna Arif, Javaid Hussain and Rahila Bano


Background Neonatal sepsis is a common occurrence in our part of the world characterized by signs and symptoms of bacterial infection during first 28 days of life. This study was carried out to evaluate the hematological parameters and C-reactive protein estimation in combination for early diagnosis in patients with neonatal sepsis.

Material and Methods: This cross sectional study was conducted at the Pediatrics Department, District Headquarters Hospital, Dera Ismail Khan, from October 2008 to April 2009. Seventy-five neonates having clinical features of sepsis and 35 clinically normal (asymptomatic) neonates were evaluated with a set of investigations. Total leukocyte count (TLC) absolute neutrophil count (ANC) platelet count (PLT) and C-reactive protein (CRP) estimation were used for diagnosis of neonatal sepsis.

Results: TLC had sensitivity of 75% for group A (proven sepsis) and 76% for group B (probable sepsis); and had a negative predictive value (NPV) of 80% and 65% respectively. The sensitivity of ANC was 65% and 76% in group A and B respectively. For proven sepsis, the sensitivity of CRP was 75% and 76% for probable sepsis. The sensitivity and NPV for the combination of TLC, ANC and CRP were 100% each in group A and 79% and80% in group B.

Conclusion:The combination of TLC, ANC and CRP is more sensitive in detection of culture positive than culture negative cases of neonatal sepsis.

KEY WORDS: Neonatal sepsis; Culture; TLC; CRP.INTRODUCTION

Sepsis is defined as systemic inflammatory response of the body to an infection. When it occurs in first 28 days of life, it is known as "Neonatal Sepsis" (NS).1 Evaluation of a neonate for possible infection is a challenging clinical problem in newborn nurseries.2

Annually five million neonates die in Asia and Africa, out of which 1.6 million (20%) are due to neonatal sepsis.3 The incidence of neonatal sepsis in developed countries is 1-5 per 1,000 live births. In Pakistan, three times higher incidence has been reported. Neonatal sepsis has non-specific symptoms and signs and a delay in diagnosis and initiation of treatment results in high morbidity and mortality rate.4

The causes of NS include, intrauterine infection,5 ascending bacterial infection, and postnatalinfection.5-7 The patient presents with respiratory distress or grunting, lethargy or irritability, fever or hypothermia, hypo or hyperglycemia, acidosis, hypotonia, vomiting, poor feeding, apnoea, cyanotic spells, seizures, persistent pulmonary hypertension, poor perfusion or shock, petechiae or purpura, unexplained jaundice, or a very sick-look.8

Many attempts have been made to develop a set of screening tests, which can rapidly diagnose infected neonates, thus, preventing delay.9 The diagnosis based on culture of blood, cerebrospinal fluid or urine is established after delay of 24 hours. However, many patients with bacterial infection have negative blood cultures.10 It has been suggested that a combination of hematological tests (total leucocyte count (TLC), absolute neutrophil count (ANC), immature to total neutrophil ratio(I/T ratio), platelet count and C-reactive protein (CRP) estimation provide early diagnosis of bacteremia.11

This study was carried out to evaluate the hematological parameters TLC, ANC and CRP estimation in combination for the early diagnosis in patients with neonatal sepsis.MATERIAL AND METHODS

This cross-sectional study was carried out in Nursery Section at Pediatrics Department, District Headquarters Teaching Hospital, Dera Ismail Khan, from October 2008 to April 2009.

Seventy-five patients suspected of having sepsis on the basis of history and clinical examination were included in the study after having informed consent from parents. Thirty-five clinically normal subjects from postnatal ward were included to define the normal ranges of hematological parameters in our population. Patients with history of antibiotics use within 48 hours of admission and those with history of exposure to viral and fungal infections in perinatal period were excluded from the study. Neonates with culture positive for coagulase negative Staphylococcus, Corynebacterium diphtheriae, and Baillus (contaminants) were also excluded. A written consent of the parents of all study groups was obtained for inclusion in the study.

In each case 5ml of blood sample was collected once only by venipuncture out of which 1.5ml was used for culture and 3.5ml for estimation of TLC, ANC, Platelet count, peripheral smear and CRP. As blood culture was not required for control group,3.5ml of blood sample was collected only. Blood culture was observed for growth after 48 hours,then sub-cultured for next 72 hours and read after five to seven days.

For description of results the neonates were divided into three groups: Group A: Confirmed cases of neonatal sepsis with positive blood culture. Group B: Symptomatic neonates having clinical diagnosis of sepsis but negative blood culture. Group C: Normal asymptomatic neonates.

Automated hematology analyzer Sysmex KX-21 (Nihon Kohden, Tokyo) was used to study full blood count. CRP was measured by latex agglutination kit (SPAN Diagnostic Ltd, Gujrat).12

The data obtained was analyzed using SPSS version 13.0.Qualitative variables were presented in the form of frequencies and percentages. The sensitivity, specificity, positive and negative predictive values were determined for each quantitative variable individually and in combination.


In the present study, 75 (33 boys and 42 girls) neonateswere admitted with clinical features of neonatal sepsis and 35 (22 boys and 13 girls) normal neonates from postnatal ward were investigated. The common clinical features of sepsis were poor feeding in 60(80%), fever 54 (72%), inactivity 53 (70%), vomiting 50 (66%), diarrhea 34 (45%) and respiratory distress 25 (33%).

Blood culture was positive out of 75 diseased subjects in 11 (15%) cases. Gram-negative organisms contributed 72% of the total number of neonates with proven sepsis. Klebsiell apneumoniae was the commonest organism (58%) followed by E. coli (10%) and Pseudomonas aeruginosa (4%). Among gram-positive coagulase positive Staphlococcus aureus accounted for 28%. No other microorganism was isolated.

TLC showed sensitivity of 75% for blood culture positive cases of NS (group A) ( Table 1) and 76% for

Table 1: Sensitivity, specificity and predictive values of tests in proven sepsis (Group A)

###at defined cut-off values.

Tests###Sensitivity###Specificity###Positive predictive###Negative predic-

###values###tive values

TLC less than 5000 or greater than 20,000/mm3###75%###57%###50%###80%

ANC less than 1750mm3###65%###78%###58%###80.5%

CRP greater than 5 mg/dl###75%###85%###72%###85.8%


Table 2: Sensitivity, specificity and predictive values of tests in probable sepsis (Group B)

###at defined cut-off values.

Tests###Sensitivity###Specificity###Positive predictive###Negative predic-

###values###tive values

TLC less than 5000 or greater than 20,000/ mm3###76%###82%###62%###65%

ANC less than 1750mm3###76%###78%###70.4%###83%

CRP greater than 5 mg/dl###76%###89%###87%###71%


at defined cut-off values.

Table 3: Haematological parameters and Cre

active protein in normal neonates (n=35)###

Parameters###Normal Range###

Total leucocyte count###5,00020,000 /mm3###

Absolute neutrophil count###1750-6000 /mm3###

C reactive protein###less than 5mg /dl

blood culture negative cases (group B). (Table 2) Its specificity was 57% in group A and 82% in group B respectively. The positive predictive value was 50% and 62%, whereas its negative predictive value was80% and 65% for group A and B respectively.

ANC had sensitivity of 65% for group A and76% for group B. Its negative predictive value was80.5% for culture proven sepsis and 83% for culture negative cases of NS.

CRP had sensitivity of 75% in proven sepsis and 76% in probable sepsis. The negative predictive value of CRP was 85.5% and 71% in group A and B respectively.

The sensitivity, specificity and predictive values of three tests in combination, TLC, ANC and CRP were also calculated for both groups. This combination had sensitivity of 100% for group A and 79% for group B. Its negative predictive value was 100% and 80% for group A and B respectively. (Table 1)


The evaluation of screening tests for neonatal sepsis is essential because the infection may present a very serious threat to the baby. Confirmation of diagnosis by blood culture may take time, and diagnostic tests are used to obtain a rapid indication of the infection status.13

A screening test would detect all the patients with the disease and exclude all the cases, which do not have that specific disease. That means, a screening test should have high sensitivity and high negative predictive value. However, a low specificity and positive predictive value are acceptable, because the risk of missing a patient with a certain infection is greater than the risk of over treatment with antibiotics.14The present study has re-evaluated the markers of sepsis in neonates. No doubt, blood culture is gold standard for the diagnosis and treatment of neonatal sepsis, but there are some setbacks. It can be falsely negative, because of intermittent or low-density bacteremia, suppression of bacterial growth by intrapartum antibiotic administration,15 improper inoculation and incubation of blood culture. Statistaical analysis would be simplified if culture negative neonates were excluded from the study but we cannot ignore these cases because fatal infectionhas been reported in presence of negative blood culture.22 In present study, 15% cases had positive blood cultures. Similarlly, Manuchaet al16 documented 14% blood culture positive cases of NS. Multiple studies have documented blood culture positivity in20%, 42% and 17% of cases. None of the reports exceed 42%. Thus it can be understood that more than half the cases of neonatal sepsis are missed if only blood culture is made the basis of diagnosis.17,18In the present study, when TLC, ANC and CRP were included as the screening tests, it showed the following data.TLC was in the abnormal range (either too low/ too high) in 88% diseased cases. The breakdown of values is shown in table 1. The cut off values of TLC for positive test were taken either less than 5000/mm3 or greater than 20,000/mm3 in the present study as suggested by Ahmed et al.4The current study showed high sensitivity and high negative predictive value for TLC in both group A (75% and 80%) and group B (76% and65%). It was similar to another study conducted by Berger et al20 who reported that leuocopenia (TLCless than 5,000/ mm3) and leucocytosis (TLCless than 20,000/mm3) had high sensitivity i.e. 67% and 74% respectively for the detection of NS.TLC was found to have optimal sensitivity 86% and NPV 96% in a study carried out at Delhi Hospital.16 Contrary to this, Ahmed et al4 reported low sensitivity of TLC 39.3% and 27.8% for blood culture proven sepsis group and culture negative but clinically diagnosed sepsis group.The normal range of absolute neutrophil count had been suggested by Schelonkaet al20 who studied larger number of healthy neonates and found ANC range to be 9500/mm3 to 21,500/mm3. In our study the normal range of ANC was found to be 1750 to6000/mm3 and the cut off value for positive test was taken less than 1750/mm3 which were similar to that of Gerdes et al.21 The present study revealed the sensitivity for ANC 65% in blood culture positive case and76% in blood culture negative cases of NS. Ahmed et al4 also reported moderately high sensitivity for the same parameters.Manuchaet al16 documented optimal sensitivity and NPV of ANC for the diagnosis of NS. Anwer et al22 found ANC to be more sensitivity than TLC as indicator of bacterial sepsis in neonates. The sensitivity was 62% in proven sepsis and 48% in culture negative cases.

There are some factors, which contribute to discrepancy in sensitivity of WBC indices. TLC can be higher in capillary than in arterial or venous blood. To avoid this discrepancy all sampling was carried out from vein. Timing is also important as WBC indices in the septic infant may be normal at the time of initial evaluation, but abnormal 4 to 12 hours later.C-reactive protein (CRP) is a non-specific acute phase protein, released in response to sepsis. Good evidence exists to support the use of CRP measurement in conjunction with other diagnostic tests such as TLC, ANC, platelet count and blood culture to establish or exclude the diagnosis of sepsis in full term or near term infants.26 Benitz and colleagues found that if CRP was greater than 5.0 mg/dl, it was highly suggestive of NS.27

The sensitivity, specificity, positive and negative predictive values of CRP for blood culture positive sepsis, a cut off point greater than 5mg/dl were 100%, 94%,91.6% and 100% respectively.

In NS, the range of reported statistical outcomes is as follows, sensitivity 70-93%, specificity41-98%, positive predictive value 6-83% and negative predictive value 97-99%.28,29

In this study, sensitivity and NPV of CRP were75% and 85.8% in blood culture positive cases (group A). In blood culture negative cases of NS (group B), these were 76% and 71% respectively. Likewise, Ahmed et al4reported 85.7% sensitivity and95.9% NPV for CRP in group A, whereas, 80.5%sensitivity and 87.1% NPV for CRP in group B. Similarly, Arshad et al27 documented 86.7% and 80.6% sensitivity of CRP for blood culture positive and negative cases of NS respectively.

In the present study, we also analyzed the combination of TLC, ANC and CRP and found that the sensitivity and NPV were improved to 80-100% each. This was very significantly high sensitivity and NPV as compared to those of the individual parameters. Therefore, we report that none of the tests used alone were as reliable, as when, employed in combination to diagnose NS. Individual tests with low sensitivity and low NPV were not considered for combination tests because of their poor value as screening tests. Except for cytoplasmic vacuolization, I/T ratio and toxic granulation in neutrophils were found to be poor indicators of sepsis. The results obtained were almost similar to those of a study by Ahmad et al.4

The tests used, were readily available, reliable and in combination found to be highly sensitive for the detection of NS. It is recommended that when the screening tests show a high NPV, the neonate could be discharged immediately from the hospital, after stopping the antibiotics, therapy reducing the hospital stay, potential exposure of neonate to hospital acquired infections, unnecessary antibiotic exposure and anxiety of the family.

It is suggested that the combination of TLC, ANC and CRP with the new modality like Interleukein-6 (IL-6) be further analyzed to improve the sensitivity and NPV to even higher values for blood culture negative cases of NS. CONCLUSION

The combination of TLC, ANC and CRP is more sensitive in detection of culture positive than culture negative cases of neonatal sepsis.


1. Adams CI, Stoll BJ. Systemic inflammatory response syndrome. Pediatr Infect Dis 2001;12:15-6

2. Movahedian AH, Moniri R, Mosayebi Z. Bacterial culture of neonatal sepsis. Iranina J Publ Health2006; 35: 84-9.

3. Aurangzeb B, Hameed A. Neonatal sepsis in hospital born babies: bacterial isolates and antibiotics susceptibility patterns. J Coll Physicians Surg Pak2003; 13:629-32.

4. Ahmed Z, Ghafoor T, Waqar T, Ali S, Shahid A, Mahmud S. Diagnsoitc value of C-reactive protein and haematological parameters in neonatal sepsis.JColl Physicians Surg Pak 2005; 15:152-6.

5. Goldenberg RL, Hauth JC, Andrews WW. N EnglJ Med 2000; 342:1500-7.

6. Yoon BH, Romero R, Moon JB. Clinical significance of intra-amniotic inflammation in patients with preterm labor and intact membranes. Am J Obstet Gynaecol 2001; 185:1130-6.

7. Sohn AH, Garrett DO, SinkowitzRL.Prevalence of nosocomial infections in neonatal intensive care unit patients. J Pediatr 2001; 139:821-7.

8. Robertson CM, Coopersmith CM. The systemic inflammatory response syndrome. Microbes Infect 2006; 8:13829.

9. Ghosh S, Mittal M, Tajanathan G. Early diagnosis of neonatal sepsis using a hematological scoring system. Indian Pediatr 2000; 54:495-500.

10. Escobar GJ. The neonatal sepsis workshop personal reflections on the development of an evidence based approach towards newborn infection in a managed care organization. Paediatrics 1999; 103:360-73.

11. Levy MM, Fink MP, Marshall JC, Abraham E, Angus D, Cook D, et al. International Sepsis Definitions Conference. Crit Care Med 2003; 31;12506.

12. Mustafa S, Farooqui S, Waheed S, Mahmood K.Evaluation of C-reactive protein as early indicator of blood culture positivity in neonates. Pak J Med Sci 2005; 21:69-73.

13. Chiesa C, Panero A, Osborn JF, Simonetti AF, Pacifico L. Diagnosis of neonatal sepsis: A clinical and laboratory challenge. Clin Chem 2004;50:279-87.

14. Polin RA. The ins and outs of neonatal sepsis. J Pediatr 2003; 143:3-4.

15. Pourcyrous M, Bada HS, Korones SB, Baselski V, Wong SP. Significance of serial C-reactive protein responses in neonatal infection and other disorders. Pediatrics 1993; 92:431-5.

16. Manucha V, Rusia U, Sikka M, Faridi MM, Madan N. Utility of hematological parameters and C-reactive protein in the detection of neonatal sepsis. J Paediatr Child Health 2002; 38:459-64.

17. Sharma A, Kutty CV, Sabharwal U, Rathee S, Mohan H. Evaluation of sepsis screen for diagnosis of neonatal septicemia. Indian J Pediatr 1993;60:559-63.

18. Bataineh HA, Al-Rashed KM. C-reactive protein in neonates with suspected septicemia. Rawal Med J 2007; 32:24-6.

19. Berger C, Uehlinger J, Ghelfi D, Blau N, Fanconi S.Comparison of C-reactive protein and white blood cell count with differential in neonates at risk for septicemia. Eur J Pediatr 2005;154:138-44.

20. Schelonka RL, Bradley YA, Desjardins SE. Peripheral leukocyte count and leukocyte indexes in healthy newborn term infants. J Pediatr 1994;125:603-6.

21. GerdesJS, Polin RA. Newonatal septicemia. In: Burg FD, Ingelfinger JR, Polin RA. Editors. GellisandKagan's Current Pediatric Therapy. Volume17. Philadelphia: WB Saunders 2002;347-51.

22. Anwer SK, Mustafa S. Rapid identification of neonatal sepsis. J Pak Med Assoc 2000; 50:94-8.

23. Christensen RD, Rothstein G, Hill HR. Fatal early onset group B streptococcal sepsis with normalleukocyte counts. Pediatr Infect Dis J 1985; 4:242-5.

24. Arshad A, Asghar I, Tariq MA. Role of serum C-reactive protein in the rapid diagnosis of neonatal sepsis. Pak Armed Forces Med J 2003; 53:178-82.

25. Benitz W, Han M, Madan A. Serial serum C-reactive protein levels in the diagnosis of neonatal infection. Pediatrics 1998; 102: e41.

26. Hengst JM. The role of C-reactive protein in the evaluation and management of infants with suspected sepsis. Adv Neonatal Care 2003; 3:3-13.

27. Nuntnarumit P, Pinkaew O, Kitiwanwanich S.Predictive values of C-reactive protein in neonatal sepsis.J Med Assoc Thai 2002;85:1151-8.

28. Russell G, Smyth A, Cooke R. Receiver operating characteristic curves for comparison of serial neutrophil band forms and C-reactive protein in neonates at risk for infection. Arch Dis Child1993;67:808-12.

29. Chiesa C, Pellegrini G, Panero A. C-reactive protein,postnatal period: influence of illness severity, risk status, antenatal and perinatal complications and infection. Clin Chem 2003; 49:51-9.
COPYRIGHT 2014 Asianet-Pakistan
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2014 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Arif, Sarah; Ehsan, Ayesha; Arif, Mizna; Hussain, Javaid; Bano, Rahila
Publication:Gomal Journal of Medical Sciences
Article Type:Report
Geographic Code:9PAKI
Date:Jul 3, 2014
Previous Article:The pattern of hearing loss as seen at the University of Benin Teaching Hospital, Benin City, Nigeria.
Next Article:Diagnostic value of c-reactive protein and haematological markers in neonatal sepsis.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters