Doctors often need to know if a patient has a serious blood infection called community-acquired bacteraemia. This condition can be life-threatening if not treated quickly. A new analysis looked at a test called procalcitonin to see if it could help make this diagnosis faster and more accurately. The researchers combined data from 40 different studies involving over 192,000 patients. They compared the procalcitonin test results against blood cultures, which are the standard way to confirm this infection.
The results showed that a low procalcitonin level is very good at telling doctors when the infection is NOT present. When the test showed a low level of 0.10 ng/mL, it correctly identified the absence of infection in 93% of cases. This means the test is useful for ruling out the problem. However, the test was less reliable at confirming the infection was present. It only correctly identified the infection in 36% of cases where it was actually there.
This mix of results means the test is not perfect. The analysis found that 32 of the 40 studies had a high risk of bias, which suggests some of the data might be less reliable. Because of this, doctors should not rely on the test alone. They must combine the test result with other signs and symptoms the patient shows. A low score can help safely reduce unnecessary blood cultures, but a high score does not guarantee an infection is present. Clinicians need to weigh the trade-offs carefully before making decisions.