Validation of serum C-reactive protein for the diagnosis and monitoring of antibiotic therapy in neonatal sepsis

Objective: To evaluate the adequacy of serum C-reactive protein (CRP) in diagnosing neonatal sepsis and role of CRP in determining the duration of antibiotic treatment in neonatal sepsis. Methods: In this validation study, we included 135 neonates with suspected diagnosis of sepsis within duration of nine months from September 2016 to May2017 in Children Hospital Multan. Blood samples were drawn from every neonate for culture sensitivity and measurement of serum C-reactive proteins. In all suspected neonates, empirical antibiotics e.g. Gentamycin or Ampicillin were started after taking blood samples. Serum CRP levels >5 mg/dl were marked as positive results. 2nd blood samples for measurement of serum CRP were taken after 72 hours of the first sample. There were two primary endpoints; one to determine the sensitivity and specificity of CRP against blood culture and second was to determine the negative predictive value of CRP in determining the duration of anti-biotic in neonates presenting with sepsis. Results: Out of these 135 babies, 102 (75.5%) were confirmed to have sepsis using blood culture reports. CRP results were Positive in 85 (62.9%) neonates on first baseline measurement and were positive in 103 (76.29%) neonates after 72 hours of admission. The sensitivity of CRP in diagnosing sepsis was 98.03%, specificity was 91.0%, positive predictive value (PPV) was 97% and negative predictive value (NPV) was 93.7%. The mean duration of antibiotic treatment in CRP guided group was 5.03 days versus 7.02 days in standard treatment duration group (p-value <0.001). The NPV of CRP in determining the duration of antibiotics was 100.0%. Conclusion: Serum CRP level is a reliable test in establishing the diagnosis of neonatal sepsis. It accurately monitors the duration of antibiotic therapy and results in significant reduction in the treatment duration of neonatal sepsis.