| Name: | Description: | Size: | Format: | |
|---|---|---|---|---|
| 1.03 MB | Adobe PDF |
Authors
Advisor(s)
Abstract(s)
Background
Legionellosis are infections caused by Legionella spp. The diagnosis of high-risk patients should rely on microbiological tests which allow the establishment of this infection etiology. Cases have to be confirmed through the available diagnostic methods which have different performances, sensitivity, specificity, error causes, limitations, and needs of careful interpretation.
Objectives
Assess the accuracy of urinary antigen detection, direct fluorescent antibody (DFA) staining, serological testing, Protein Chain Reaction (PCR) versus culture (reference standard), in patients suspected to be infected with Leggionella or patients with laboratory confirmed Legionnaire Disease (LD).
Search Methods
We searched the Cochrane Central Register of Controlled Trials (CENTRAL) in The Cochrane Library and
MEDLINE (PubMed). We also handsearched the reference lists of the included studies.
Selection criteria
Observational studies were included, comparing the index tests with culture in patients suspected to be infected with Legionella or patients with laboratory confirmed LD.
Data collection and analysis
Two authors independently assessed the trials and extracted data. Data was analysed using statistical software
Review Manager 5.1.
Main results
Five studies met the inclusion criteria. All studies evaluated PCR and DFA tests to detect Legionella in clinical specimens, comparing it with culture (reference standard) and were included in meta-analysis. PCR sensitivity and specificity ranged from 56% to 100% and from 89% to 100%, respectively. The pooled sensitivity was
74% (95% IC 67%-80%), and the specificity 97% (95%IC 96%-80%). DFA sensitivity varied from 33% to
44% and the specificity from 100% the pooled sensitivity was 40% (95% IC 21%-61%) and the specificity
100% (95% IC 81%-100%).
Author´s conclusions
This review demonstrates that PCR have a high sensitivity and specificity for early diagnosis of LD. However standardization is required for biological samples. Although this, culture is always required for epidemiological studies, strains molecular typing and antibiotic sensibility evaluations if needed