Background: : In critically ill patients, severe infection and systemic inflammation due to non-infectious causes produce very similar clinical presentations, and traditional infection markers do not always differentiate these two conditions. Both procalcitonin and neopterin have been suggested to aid in the early diagnosis of bacterial infections and in differentiating bacterial infections from systemic inflammatory, non-infectious diseases or from viral infections.
Methods: : Procalcitonin (PCT) and neopterin were analyzed in 208 ICU patients who developed acute fever or septic shock. Blood samples were taken every 8th h within 48 h of the onset of fever or septic shock.
Results: : A total 162/208 of patients had infection, the most common location being the respiratory tract. Mortality was higher in infected patients (31.4% vs. 10.9%; P < 0.01). The optimum cut-off levels in identifying patients with infection of daily peak PCT were 0.8 microg/L on day 1 and 0.9 microg/L on day 2, and both sensitivity (67.7% and 60.9%, respectively) and specificity (47.8% and 63%) were poor. Accordingly, the optimum cut-off values of peak neopterin were 18 and 16 pg/L. The sensitivity was 62.7% on day 1 and 69.3% on day 2, while specificity was correspondingly 78.3% and 67.9%. There were no significant differences between the markers in discriminating between patients with infection or inflammation. Both PCT and neopterin increased with the severity of infection. They were higher in non-survivors.
Conclusion: : PCT and neopterin were equally effective, although not very accurate in differentiating between infection and inflammation in critically ill patients. Neopterin was more specific than PCT, suggesting that neopterin is related to the activity of inflammatory response.