To establish a multiplex fluorescent PCR-melting curve method for simultaneous detection of Candida, Aspergillus and Cryptococcus, and to evaluate its diagnostic value.

A multiplex fluorescent PCR-melting curve method was established and its detection system was optimized. Totally 179 clinical specimens including 65 blood specimens, 35 deep sputum specimens, 30 urine specimens, 18 cerebrospinal fluid specimens, 12 hydrothorax and ascites specimens, 10 alveolar lavage fluid specimens and 9 fresh lung tissue specimens for highly suspected invasive fungal infection (IFI) were enrolled in this study. The sensitivity, specificity and repeatability experiments were used to verify the detection performance of this method, and the receiver operating characteristic (ROC) curve was used to evaluate its diagnostic efficiency.

The multiplex fluorescent PCR-melting curve method could simultaneously detect 8 common pathogenic fungi, including 4 kinds of Candida (Candida albicans, Candida tropicalis, Candida glabrata, Candida krusei), 3 kinds of Aspergillus (Aspergillus fumigatus, Aspergillus flavus, Aspergillus niger) and Cryptococcus neoformans. The minimum detection limit was 1 × 10⁴ cfu/mL. Common bacteria had no amplification reaction, and the melting temperature fluctuation of repetitive experiment was less than 0.5 ℃. Among 179 clinical specimens, 112 were positive by "gold standard" methods such as culture, microscopy and pathological diagnosis, and 96 were positive by the multiplex fluorescent PCR-melting curve method. The sensitivity, specificity, positive predictive value and negative predictive value for simultaneous detection of Candida, Aspergillus and Cryptococcus neoformans by this method were 0.857, 0.970, 0.980 and 0.802, respectively. The area under ROC curve was 0.914, and the 95% confidence interval was 0.868 ~ 0.959 (P < 0.001).

The multiplex fluorescent PCR-melting curve method can detect Candida, Aspergillus and Cryptococcus neoformans rapidly and accurately with high throughput, which is of great significance for the early diagnosis of IFI.