Blood stream infection (BSI) is a life-threatening condition that implies prompt diagnosis and management. With the time-consuming traditional microbiological culture and limited provision of molecular diagnostics, procalcitonin (PCT) has emerged as a rising rapid biomarker for prediction of BSI. We aimed to evaluate the diagnostic performance of PCT in predicting BSI as well as in tentative differentiation of the possible underlying microbial etiology. Blood samples were collected from total 328 patients suspected of having BSI and were measured for levels of procalcitonin, C-reactive protein (CRP) and total leucocyte count (TLC). Concomitant microbiological culture was done for detection and identification of microbial etiology. Microbial cultures were positive in 47.9% and negative in 52.1% of patients. Bacterial and fungal growth accounted for 87.8% and 12.1%, respectively, where bacterial growth was differentiated as Gram-positive and Gram-negative bacteria. Culture-positive group displayed significantly higher median values of PCT (2.69 ng/ml) and CRP (117.2 ng/ml) than culture-negative group. PCT levels were significantly higher among Gram-negative than Gram-positive cultures, however were non-significantly higher among bacterial than fungal positive-cultures. Compared to other biomarkers, PCT displayed the highest diagnostic accuracy in differentiating culture-positive and -negative BSI (ROC-AUC: 0.706, P < 0.001) at cut-off value of 0.675 ng/ml, as well as in discriminating Gram-negative and Gram-positive bacteria (ROC-AUC:0.636, P: 0.004) at cut-off value of 2.135 ng/ml. PCT exhibited higher diagnostic accuracy than other biomarkers in BSI and in differentiating Gram-negative and Gram-positive bacterial etiology, which may offer a modest guidance in sepsis management and optimization of antibiotic intake. Not applicable.
Abbas et al. (Tue,) studied this question.