Background: Sepsis is defined as “life-threatening organ dysfunction, caused by a dysregulated host response to infection”. Neonatal sepsis is the most common cause of morbidity and mortality in the neonatal period. Early diagnosis of neonatal sepsis is difficult. Therefore, this study was conducted with the objective to assess the diagnostic accuracy of procalcitonin as marker of neonatal sepsis and its comparison with C-reactive protein. Methods: The present study was a hospital-based descriptive comparative study. A total of 59 neonates were enrolled. All suspected neonates for the sepsis admitted to NICU were enrolled in study on the basis of inclusion and exclusion criteria. A detailed clinical examination was done. Blood sample was collected for procalcitonin, C-reactive protein and blood culture. Statistical analysis was performed. Results: In our study diagnostic accuracy of procalcitonin in diagnosis of neonatal sepsis was sensitivity (88.46%), specificity (87.88%), positive predictive value (85.19%), negative predictive value (90.63%) and diagnostic accuracy (88.14%). Diagnostic value of C-reactive protein in diagnosis of neonatal sepsis was sensitivity (88.46%), specificity (69.70%), positive predictive value (69.70%), negative predictive value (88.46%) and diagnostic accuracy (77.97%). Diagnostic accuracy of procalcitonin is maximum followed by C-reactive protein. Conclusions: In our study all patient with gram negative organism were procalcitonin positive whereas 50% Staphylococcus aureus were procalcitonin positive and in candida positive cases out of 6 cases, 5 (83.3%) were procalcitonin positive.