The evaluation of C-reactive protein as a screening tool for neonatal sepsis

Abstract

Background: Bacterial sepsis and meningitis continues to be major causes of neonatal mortality and morbidity in low birth weight infants. C - reactive protein (CRP) is a simple investigation which has diagnostic potential but the previous studies have yielded variable results. A combination of screening tests along with the clinical signs is useful to diagnose/rule out neonatal sepsis along with gold standard blood culture. The objective was to study CRP in relation to other haematological parameters in diagnosis of sepsis and to find out whether CRP estimation helps in early treatment interventions like stopping empirical antibiotic therapy in neonatal sepsis. Methods: A prospective study of 62 neonates with suspected sepsis admitted in level 2 neonatal special care unit of TMH, Jamshedpur from 1 st January 2010 to 30 th September 2010 were evaluated after meeting inclusion criteria using clinical criteria, sepsis screen and blood culture, which was taken as gold standard for diagnosis of neonatal sepsis. Results: Total blood culture positive cases were 26 (42%) and CRP was positive in total 24 cases (39%) of total 62 cases, of which, it is increased in 19 cases (79%) of total 26 culture proven cases and in 5 cases (21%) of culture negative cases. The CRP was increased in 14 cases (35%) out of total 40 cases in early onset sepsis (EOS) group as compared to 10 cases (45.5%) out of 22 in late onset group. The sensitivity of CRP is more in late onset sepsis (LOS) (45.5%) than in early onset sepsis (35%). CRP rise was highest with 73% in proven sepsis, where as it is 23% and 8% in probable and no sepsis groups respectively. The CRP in present study is having sensitivity of test: 73%, specificity of test: 86.1%. The PPV and NPV were 79.1 and 81.6% respectively. If CRP is added to septic screen, it has improved the specificity of the test. Whereas, sepsis screen has sensitivity of 88.5%, the specificity of 83.3%, PPV 79.3% and NPV 90.9%. So, totally empirical antibiotic therapy was stopped in 23 cases of “no sepsis” group including both early as well as late onset sepsis. Conclusions: Based on the result of our study, the CRP is an effective parameter for the diagnosis of neonatal sepsis and when used in conjunction with the sepsis screen, which increases the sensitivity and specificity of the test, it can help identify septic neonates and help in appropriate management while also reducing the unnecessary use of antibiotics, thereby helping curb the growing menace of antibiotic resistance.