Diagnostic accuracy of procalcitonin for bacterial infection in the Emergency Department: a systematic review

Abstract

Introduction and objective

The care of patients with a suspected infectious process in hospital emergency departments (ED) accounts for 15%–35% of all daily care in these healthcare areas in Spain and Latin America. The early and adequate administration of antibiotic treatment (AB) and the immediate making of other diagnostic-therapeutic decisions have a direct impact on the survival of patients with severe bacterial infection. The main objective of this systematic review is to investigate the diagnostic accuracy of PCT to predict bacterial infection in adult patients treated with clinical suspicion of infection in the ED, as well as to analyze whether the different studies manage to identify a specific value of PCT as the most relevant from the diagnostic point of view of clinical decision that can be recommended for decision making in ED.

Method

A systematic review is carried out following the PRISMA regulations in the database of PubMed, Web of Science, EMBASE, Lilacs, Cochrane, Epistemonikos, Tripdatabase and ClinicalTrials.gov from January 2005 to May 31, 2023 without language restriction and using a combination of MESH terms: “Procalcitonin”, “Infection/Bacterial Infection/Sepsis”, “Emergencies/Emergency/Emergency Department”, “Adults” and “Diagnostic”. Observational cohort studies (diagnostic performance analyses) were included. The Newcastle-Ottawa Scale (NOS) was used to assess the quality of the method used and the risk of bias of the included articles. Observational cohort studies were included. No meta-analysis techniques were performed, but results were compared narratively.

Results

A total of 1,323 articles were identified, of which 21 that met the inclusion criteria were finally analyzed. The studies include 10,333 patients with 4,856 bacterial infections (47%). Eight studies were rated as high, 9 as moderate, and 4 as low. The AUC-ROC of all studies ranges from 0.68 (95% CI: 0.61–0.72) to 0.99 (95% CI: 0.98–1). The value of PCT 0.2–0.3 ng/ml is the most used and proposed in up to twelve of the works included in this review whose average estimated performance is an AUC-ROC of 0.79. If only the results of the 5 high-quality studies using a cut-off point of 0.2–0.3 ng/ml PCT are taken into account, the estimated mean AUC-COR result is 0.78 with Se:69 % and Es:76%.

Conclusions

PCT has considerable diagnostic accuracy for bacterial infection in patients treated in ED for different infectious processes. The cut-off point of 0.25 (0.2–0.3) ng/ml has been positioned as the most appropriate to predict the existence of bacterial infection and can be used to help reasonably rule it out.