Procalcitonin as a marker of Candida species detection by blood culture and polymerase chain reaction in septic patients
Department of Biopathology, Medical and Forensic Biotechnologies (DIBIMEF), Section of Anaesthesia, Analgesia, Intensive Care and Emergency. Policlinico "P. Giaccone", University of Palermo, Palermo, Italy
BMC Anesthesiology 2014, 14:9 doi:10.1186/1471-2253-14-9Published: 21 February 2014
The aim of our study is to test procalcitonin (PCT) as surrogate marker of identification of Candida spp. by blood culture (BC) and real-time-polymerase chain reaction (PCR), whether alone or in association with bacteria, in septic patients.
We performed a single-centre retrospective study. We reviewed the clinical charts of patients with a diagnosis of severe sepsis or septic shock treated at our general intensive care unit from March 2009 to March 2013. We analysed all diagnostic episodes consisting of BC, real-time PCR assay and dosage of PCT. We registered age, sex, white blood count, sequential organ failure assessment score and type of admission between medical or surgical. When inclusion criteria were met more than once, we registered the new diagnostic episode as subsequent diagnostic episode. The diagnostic performance of PCT to predict Candida spp. identification alone or in mixed infections by either BC or PCR was tested using the receiver-operative characteristic curve. Logistic regression was constructed using presence of Candida spp. as the dependent variable.
A total of 260 diagnostic episodes met the inclusion criteria. According to BC results classification, a significantly lower value of PCT was observed in Candida spp. BSI (0.99 ng/ml, 0.86 - 1.34) than in BSI caused by bacteria (16.7 ng/ml, 7.65 - 50.2) or in mixed infections (4.76 ng/ml, 2.98 - 6.08). Similar findings were observed considering PCR results. A cut-off of ≤ 6.08 ng/ml for PCT yielded a sensitivity of 86.8%, a specificity of 87.4%, a positive predictive value of 63.9%, a negative predictive value (NPV) of 96.3% and an area under the curve of 0.93 for Candida spp. identification by BC. A similar high NPV for a cut-off ≤ 6.78 ng/ml was observed considering the classification of diagnostic episodes according to PCR results, with an AUC of 0.85. A subsequent diagnostic episode was independently associated with Candida spp. detection either by BC or PCR.
PCT could represent a useful diagnostic tool to exclude the detection of Candida spp. by BC and PCR in septic patients.