Comparative Study of Diagnostic Markers in Neonatal Sepsis

Address for correspondence: Dr. Vijay Baburao Sonawane, Assistant Professor, Department of Paediatrics, MGM Medical College and Hospital, Navi Mumbai, Maharastra, India. E-mail: vijay_ltm@yahoo.co.in 1Dr. Vijay B. Sonawane, MBBS, MD, Assistant Professor, 2Dr. Sonali U Gaikwad, MBBS, MD, Senior Resident, 3Dr. Nitin N Kadam, MBBS, MD, Professor and Head of Department, 4Dr. Jitendra Gavhane, MBBS, MD, Assistant Professor. All from the Department of Paediatrics, MGM Medical College and Hospital, Navi Mumbai, Maharashtra, India. Abstract


Introduction
N eonatal Sep cemia is the leading cause neonatal mortality and morbidity in India.It is es mated that 20% of all neonates develop sepsis 1 and is responsible for 30-50% of total neonatal death in developing countries 2 .Accurate and mely diagnosis of neonatal sepsis s ll remains a major challenge to the pediatricians and neonatologists.Mortality due to neonatal sepsis is preventable and if diagnosed early the outcome is be er.Several indicators have been evaluated as "sep c screen" for the early diagnosis of neonatal sepsis like CBC, CRP, Micro-ESR, IT ra o, and Blood Culture.There is a constant search for be er and ideal diagnos c marker.Ideal marker should be sensi ve, specifi c and easily available.Its results should be immediate and reproducible.

Materials and Methods
In our study involved 160 babies delivered in the hospital, having risk factors, along with those coming to hospital with signs and symptoms of sepsis up to 28 days of life (Study group) also normal newborns admi ed to the postnatal ward without high risk factors (Control group).Newborns were treated with an bio cs for clinical evidence of sepsis & posi ve sep c score as per neonatal sepsis score system (        A detailed history was taken and examina on was done.Following laboratory tests were done as soon as presump ve diagnosis of sepsis was made based on sep c score system and on clinical grounds.All inves ga ons were done within 24 hours of birth or at presenta on before star ng an bio cs like CBC, CRP with ter, Micro ESR, Immature to Total (IT) ra o, peripheral smear for toxic granules and band forms, blood culture (Bac Talert), x-ray chest, CSF whenever indicated etc.Soon a er birth, 1 ml of venous blood was drawn for blood culture.Also 5 ml of venous blood was collected for TLC, DLC, peripheral blood smear, micro-ESR and CRP.CRP was sent at 12 hours of life in newborns with high risk factors for sepsis.
This study was approved by Ethical Commi ee of this hospital.Informed Wri en Consent was obtained from parents before entry into this study.Data was collected, classifi ed, tabulated and analyzed.Tests of signifi cance were applied at appropriate places and interpreta on was done accordingly.To evaluate the diff erence between the categories, McNemar Chi Square test was used as a test of signifi cance.A p-value of less than 0.05 was considered sta s cally signifi cant.

Results
Among the babies of study group, 48 (60%) were blood culture (BacTalert) posi ve while in control group, only 2 (2.5%) were blood culture (BacTalert) posi ve (Fig. 1).In study group, E-coli comprised the maximum number of cases accoun ng for sepsis i.e. 7 (17.5%)while there were no growths in 16 (40% ) cases.In control group, only 1 case (2.5%) showed growth of Acinetobacter baumanii and 39 cases (97.5%) were sterile (Fig. 2).Among the high risk factors, prematurity cons tuted the maximum number of cases i.e. 55% followed by meconium stained amnio c fl uid in 35%, premature rupture of membranes for >24 hours in 25%, birth asphyxia in 25%, prolonged or diffi cult labor in 22.5%, febrile illness in mother in 20%, more than three vaginal examina on in 20% and other risk factor included home delivery, unsterile handling and cu ng of cord, mul ple gesta on and faulty feeding prac ces in 25% of cases.

Discussion
Neonatal sepsis is mainly a bacterial infec on in the blood which is caused by bacterial, fungal, parasi c or viral pathogens.It is a condi on found in infants less than one month old and a poten ally serious infec on that requires NICU care.
Our study revealed that males (70%) were reported to be more likely than females to develop sep cemia.Faridi and Gupta (1992) 3 also reported 66.67% males and 33.33% females out of 63 cases of neonatal sep cemia.
In our study of 80 cases, 44 (55%) were preterms while there were no preterm case in the control group.This showed increased risk of neonatal sepsis in premature newborns.This was similar to the observa on made by Washburn and Medeasis 4 where 53.97% preterm and 46.03% full term neonates.
We observed that prematurity and LBW were leading high risk factors for sepsis in 55% of the newborns, other risk factors were meconium stained amnio c fl uid (35%) premature rupture of membrane for >24 hours (25%), birth asphyxia (25%), prolonged or diffi cult labour (22.5%), febrile illness in the mother (20%), >3 per-vaginal examina on (20%) and 25% others like, mul ple gesta on, home delivery, unsterile cord handling and cu ng technique, faulty feeding prac ces, nursing in unhygienic environment, etc.
The most important risk factors for neonatal sepsis were prematurity and low birth weight as reported by Jain and Maheshwari et al 5 in their study.
Blood culture is the gold standard method to diagnose sep cemia.In our study, blood culture (BacTalert) had shown growth in 48 cases i.e.60% (46 bacterial and two fungal ) in the study group, while blood culture (BacTalert) was posi ve in two case (2.5%) in the control group.The most common organism grown on blood culture was E.Coli (17.5%).Jain and Maheshwari et al 5 analyzed the signs and symptoms of neonatal sepsis in 106 neonates with suspected sepsis out of which 30 were culture posi ve.The most common organism was E. coli which is similar to this study.
Neonatal sepsis data from most of the medical centers in the US shows that E. coli and group B streptococcus accounted for 50-70% of the cases.Mirf et al 6 in their study of 50 cases also showed similar results.McCraken 7 , Faridi and Gupta 3 , Kumar GD et al 8 have also reported that gram nega ve sep cemia is more common than gram posi ve sep cemia In this study CBC had low sensi vity (37.50%) and high specifi city (75.00%).Chan and Ho 9 revealed in their study that abnormal CBC had the lowest sensi vity and PPV while abnormal ANC had the lowest specifi city and NPV among them.
In this study CRP was reported most sensi ve (84.21%) but low specifi c (28.57%), Franz AR et al 10 showed that there is generally a delay of up to 24 hours between onset of symptoms of infec on and a rise in serum CRP.Sensi vity of the test at presenta on is only 40% that is, 60% of subsequently proven sepsis episodes will have a normal ini al CRP.
In our study IT ra o was reported NPV of 50.0%.Ghosh et al 11 studied 103 high risk neonates having predisposing perinatal factors or clinical suspicion of sepsis and found that an abnormal immature to total neutrophil (IT) ra o were the most sensi ve indicators in iden fying neonates with sepsis showing high nega ve predic ve value over 94%.
In our study Micro-ESR has shown sensi vity of 50.0%, specifi city of 62.5.K.K. Diwakar and Rosul G 12 studied on 114 term neonates for early neonatal sepsis.The sensi vity and specifi city of the 'revised' Micro-ESR was 62.5% and 60.9% respec vely in diagnosing culture proven sepsis.

Conclusion
CRP was found most sensi ve, while CBC was found to be most specifi c marker in neonatal sepsis. 17

Fig 1 :Fig. 2 :
Fig 1: Distribu on of Cases According to Blood Culture

MGM
Medical College and Hospital, Navi Mumbai, for his invaluable guidance, supervision to complete this study.I thank Mr. Da atraya Parle for helping me in this study.Funding: None Confl ict of Interest: This ar cle is based on the fi nal thesis (part of compulsory thesis submission) submi ed to Department of Paediatrics MGM Medical College and Hospital, University of Mumbai.Permission from IRB: Yes