Procalcitonin test for diagnosis of sepsis or septic shock: Systematic review of literature and metanalysis

Abstract

Background: Procalcitonin (PCT) is a prohormone of calcitonin, produced by cells C of the thyroid gland and intracellurarly cleaved by proteolytic enzymes into the active hormone. The production of PCT during inflammatory process, is linked with a bacterial endotoxin and with inflammatory cytokines. Mortality due to sepsis, depends to a large extent on a early detection and early start of adecuade therapy, that includes giving appropriate antibiotics. It´s no clear if the PTC diagnostic performance is the same in the context of the definition of the third consensus as in the previous definition. Methods: Studies describing the use of PCT within the first 24 hours of admission as a diagnostic test for sepsis were included. We searched the Medline (Pubmed) and Embase databases. The methodological quality was evaluated according to the Cochrane Collaboration in the development of Systematic Reviews on Analysis Test for the QUADAS-II tool. The publication bias was studied with the Deeks Asymmetry Test. The MIDAS module of STATA 14 was used for the univariate analysis and the construction of the ROC Curve. Results: 2076 records were obtained (783 from Medline and 1293 from Embase). Of the 12 selected studies, a total of 1353 patients were included, with a prevalence in the studies reviewed between 9% and 88%, with an average of 47%. The pooled sensitivity was 0.83% (CI 95% (0.74-0.89)) and the Specificity was 0.84% (CI 95% (0.76-0.89)). The area under the Curve was 0.90 (CI 95% (0.87-0.92)). Heterogeneity between the studies is important I2 88% (CI 95%(77-100)). There is a publication bias according to the Deek test, with a result of P = 0.04. In the analysis on the post test Probability according to the Fagan nomogram, it is 56%, taking into account a pretest probability of 20% according to the positive LR 5. Conclusions: PCT is a diagnostic test with good performance for sepsis or septic shock, in adult patients, not pregnant. Although there is publication bias and great heterogeneity in the results, the test is considered adequate for the sepsis setting according to the new definitions.