Validity of C-Reactive Protein (Crp) for Diagnosis of Neonatal Sepsis in Comparison With Blood Culture

Abstract

BACKGROUNDSepsis is a recognised cause of morbidity and mortality in the new-borns in the developing countries. Sepsis is well-defined as a "clinical syndrome characterised by systemic signs/ symptoms and bacteraemia during the 1 st month of life." Neonatal sepsis, a serious illness but curable if identified early. Non-specific signs/ symptoms make it very challenging to formulate a timely clinical diagnosis. Neonatal physicians are looking for a test that would help in neonatal sepsis diagnosis, quickly confirming it and decisively ruling it out. Diagnostic tests like blood cultures are time consuming, so correct diagnosis gets delayed. CRP is very sensitive in detecting negative cases of neonatal sepsis. It is simple, easy and gives early results. Aim: To find out validity of C-reactive protein in the diagnosis of neonatal sepsis in comparison with blood culture in terms of-1. Sensitivity, 2. Specificity, 3. Positive predictive value and 4. Negative predictive value. MATERIALS AND METHODSAn observational (validation) study was conducted in the NICU and SCNU at Institute of Child Health, Government Medical College, Kottayam to assess validity of CRP for diagnosis of neonatal sepsis in comparison with blood culture. After taking informed consent from parents and permission from Ethical Committee, neonates brought to neonatology unit were selected by purposive sampling technique. Sample size was 90. All neonates having suspected neonatal sepsis were included. Suspected neonatal sepsis was considered if neonates had features of perinatal risk factors, i.e. maternal pyrexia (within 1 week prenatal and/ or 48 hours postnatal), prolonged rupture of membranes (18 hours), foul smelling vaginal discharge or maternal urinary tract infection in last month. Neonates having unexplained hypothermia/ hyperthermia, lethargy, irritability, poor feeding, respiratory dysfunctionapnoea (> 10 secs), tachypnoea (> 60 breaths/minute), cardiovascular dysfunction-persistent tachycardia (> 160 beats/min) or bradycardia (< 100 beats/min), poor peripheral circulation, hypotonia or circumoral cyanosis or pallor were also included. Babies with birth asphyxia, birth weight < 1500 grams, < 32 weeks gestation and who already had taken antibiotics were excluded. All neonates included in the study were started on empirical antibiotics after drawing samples for blood cultures and CRP. CRP was read as negative when less than 10 mg/dL and positive when level was more than 10 mg/dL. Blood culture was followed for growth for 7 days. Data collection tool was a pre-tested proforma. Analysis was done using SPSS 23. RESULTSThe analysis included a total number of 100 infants. Data was analysed by SPSS version 23. The baseline characteristics of the study group were comparable in terms of birth weight, gestational age, postnatal age, type of delivery and gender. In this study, CRP was compared to blood culture and was found to have good validity with sensitivity 72.27%, specificity 74.15%, positive likelihood ratio 2.75, negative likelihood ratio 0.37, posi...