Early diagnosis of candidemia in critical ill patients: comparative analysis of real-time PCR, MALDI-TOF and detection of 1,3-beta-D-glucan assay

Abstract

Systemic infections caused by Candida spp. represent an important cause of morbidity and mortality in patients hospitalized in ICU. In addition, these infections are associated with increase in time and costs of hospitalization. Despite the growing importance of this yeast as the etiological agent of severe infections, particularly in patients at risk, early diagnosis remains a challenge in clinical practice. Blood cultures, still considered as a gold standard method in the diagnosis of systemic candidiasis, has limited sensitivity and requires long time to identify the infecting species. The use of real time PCR for the detection of Candida directly in clinical samples has been reported in several publications, showing promising results. However, the lack of methodologic standardization and clinical validation has limited the use in routine clinical laboratories. Therefore, additional studies should be conducted to incorporate recent advances in PCR standardization and to ensure reliability of the results. MALDI-TOF mass spectrometry presents high accuracy for micro-organisms identification. In recent years, the method has been evaluated in detecting and identifying pathogens directly on positive blood culture bottles, allowing results in few hours at low cost. In order to evaluate the accuracy of molecular methods, we propose to employ an in house PCR real-time for detection and identification of five Candida species in clinical samples from 150 adult patients with suspicion of candidemia, attended at ICUs of two public hospitals located in São Paulo. According to EORTC/MSG criteria, patients will be classified with possible, probable or proved candidemia. The 1,3-²-D-glucan test, a serologic marker of fungal infections, will be performed on serum samples of patients, and it can confirm PCR positive results in patients with negative blood cultures. The results of both methods will be compared with blood cultures performance (gold standard method). Sensitivity, specificity, positive predictive and negative predictive values will be determined for the three assays. The performance of MALDI-TOF technique for the correct pathogen identification directly on positive culture bottles will be evaluated. (AU)