Procalcitonin (PCT) accurately predicts bloodstream infection and differentiates Gram-negative from Gram-positive bacteria

Bloodstream infection (BSI) is a critical, life-threatening condition demanding swift diagnosis and management. Traditional microbiological cultures are time-consuming, and molecular diagnostics are often limited, leaving a significant gap for rapid, reliable biomarkers. Procalcitonin (PCT) has emerged as a promising candidate due to its rapid kinetics in response to bacterial infections, offering potential for quicker prediction and guidance on microbial etiology, thereby improving initial therapeutic decisions.

Researchers conducted a prospective study involving 328 patients suspected of having bloodstream infection (BSI) at an Egyptian tertiary care hospital. Blood samples were collected from all participants to measure levels of procalcitonin, C-reactive protein (CRP), and total leucocyte count (TLC). Concurrently, microbiological culture was performed to detect and identify the specific microbial etiology. The study aimed to evaluate the diagnostic performance of PCT in predicting BSI and differentiating potential microbial causes.

Microbial cultures were positive in 47.9% of patients and negative in 52.1%. Among positive cultures, bacterial growth accounted for 87.8% and fungal growth for 12.1%, with bacterial growth further differentiated into Gram-positive and Gram-negative types. The culture-positive group exhibited significantly higher median values for procalcitonin (2.69 ng/ml) and CRP (117.2 ng/ml) compared to the culture-negative group. PCT levels were significantly higher in Gram-negative cultures than in Gram-positive cultures, although the difference between bacterial and fungal positive cultures was not statistically significant. Compared to CRP and TLC, PCT demonstrated the highest diagnostic accuracy for differentiating culture-positive from culture-negative BSI.

PCT achieved an ROC-AUC of 0.706 (P < 0.001) at a cut-off value of 0.675 ng/ml for predicting BSI. Furthermore, PCT also showed superior performance in discriminating between Gram-negative and Gram-positive bacteria, with an ROC-AUC of 0.636 (P: 0.004) at a cut-off value of 2.135 ng/ml.