Procalcitonin: All the rave?

Anthony J. Hackett, DO

There seems to be an awful lot of talk lately about the utility of procalcitonin in prognosticating, detecting and tracking bacterial infections. This article review looks at a fairly recent meta-analysis in The Lancet Journal of Infectious Disease which aims to look at all of the data regarding procalcitonin as a marker of infectious disease.

The basics: Multi-year, multi-database meta-analysis that looks at the usefulness of procalcitonin in differentiating SIRS secondary to non-infectious causes from Sepsis/Severe Sepsis/Septic Shock.

Clinical Question: What is the accuracy of using procalcitonin as a diagnostic marker in sepsis

Design:
-Looks at 3487 reports (3244 patients) with sepsis Vs. SIRS. Based on inclusion criteria 136 papers were selected and investigated
-Studies had to include a standard definition of sepsis and causes of sepsis had to be attributed to a documented infection.
-Had to have sufficient data to accurately construct a 2X2 contingency table for sensitivity and specificity

Outcome:
-Conglomerate results demonstrating a sensitivity of 0.77 and a specificity of 0.79 for SIRS secondary to sepsis versus SIRS of other etiologies
-At this time cannot define a cutoff for a “negative test,” The authors noted most studies used 1.1-2.0 but no difference in effect was appreciated between the two.
-Not a perfect marker but needs to be interpreted in the context of clinical suspicion

Strengths:
-Better statistical methods and more rigorous exclusion and inclusion criteria than prior meta-analyses

-Supports the idea that procalcitonin is useful in differentiating sepstic from non-septic causes of SIRS

Weaknesses:
-Heterogeneity in patient characteristics was noted and couldn’t be eliminated in a subgroup analysis
-Excluded neonates (<28 days)
-Heterogeneous assays were used across papers and study groups
-Only three language databases were used (English, French, German)

The Future:
-Prospective clinical trial validation of a cutoff of between 1.1 and 2.0 for differentiating SIRS /Sepsis. Vs. non-septic causes of SIRS
-Correlate numerical values of the level of procalcitonin with the severity of sepsis and possibly outcomes.

Article link:

http://www.thelancet.com/journals/laninf/article/PIIS1473-3099%2812%2970323-7/fulltext