COMPARISON OF GLUCAN AND GALACTOMANNAN TESTS WITH REAL-TIME PCR FOR DIAGNOSIS OF INVASIVE ASPERGILLOSIS IN A NEUTROPENIC RAT MODEL

The incidence of aspergillosis which has high mortality rates, has increased gradually. Since invasive aspergillosis (IA) is one of the leading causes of death in immunocompromized and neutropenic patients, early and accurate diagnosis of IA is of crucial importance. The aims of this study were to compare the results of culture, real-time polymerase chain reaction (RtPCR), galactomannan (GM) antigen and glucan (GC) antigen detection tests and to evaluate their performances in view of rapid and accurate diagnosis of IA in neutropenic rat model. Female Wistar albino rats were included in the study with the consent of Animal Searching Ethical Committee and classified into three groups as healthy controls (n = 6), neutropenic controls (n = 10) and pulmonary aspergillosis (n = 35) groups. Rats were immuno-suppressed with 5-flourourasil and then Aspergillus fumigatus conidia were inoculated intranasally. On the seventh day of the infection, blood, bronchoalveolar lavage (BAL) and lung tissue samples were collected from the animals, and control and aspergillosis groups were compared in terms of infection markers. All of the tests (culture, RtPCR, GM and BG tests) were found to be negative in controls. At the end of the study 22 rats in aspergillosis group survived. Lung tissue samples from those 22 animals were all positive for the presence of hypha on pathological preparations, while 20 (91%) yielded Aspergillus colonies on the cultures. Aspergillus DNA was detected in 7 of the 12 SAL samples (58.3%), 7 of 19 blood samples (36.8%) and 4 of 22 lung tissue samples (18%) using RtPCR method. GM antigen was detected in 7 of 20 serum samples (35%) with a commercial kit (Platelia (R) Aspergillus ELISA, BioRad, France). Quantitative detection of beta-glucan levels were investigated by using a commercial kit (Fungitell (TM), Cape Cod, USA) in serum and BAL samples and positive results were obtained in 11 of 22 serum (50%) and 9 of 17 BAL (52.9%) samples. In this study it was demonstrated that PCR performed in SAL samples is the most sensitive method compared to the others, for the diagnosis of IA in the rat model. The sensitivity rates were as follows when culture method accepted as the gold standard: 58.3% for BAL-PCR, 41.2% for blood-PCR, 20% for tissue-PCR, 38.9% for serum GM, 55% for serum GC and 52.9% for BAL-GC. It was also concluded that detection of GC activity in serum was more sensitive than GM detection in serum (sensitivity of GM was %38.9, sensitivity of GC was %55, while specificities were 100% for both of the tests), for laboratory diagnosis of IA. The BAL samples were evaluated as the most valuable clinical samples to screen the suspected patients. However, even in proven cases, 41.7% of BAL samples were found negative with PCR, 50% of serum samples were found negative with GC test, and 65% of serum samples were found negative with GM test. Since the pathogenesis of IA has not been completely clarified, the performance of non-culture based diagnostic tests exhibit great variability. Future clinical studies are required to compare the performance of different non-culture based diagnostic methods and the optimal combination of these tests for the most accurate laboratory diagnosis of IA.