Извадок EVALUATION OF ( 1 , 3 )-β-D-GLUCAN ASSAY IN DIAGNOSIS OF INVASIVE FUNGAL INFECTIONS WITH ASPERGILLUS

ЕВАЛУАЦИЈА НА (1,3)-β-D-ГЛИКАН ЕСЕЈ ВО ДИЈАГНОЗА НА ИНВАЗИВНИ ИНФЕКЦИИ СО ASPERGILLUS Gordana Mirchevska1, Zaklina Cekovska1, Ana Kaftandzieva1, Zorica Zafirovik2, Elena Trajkovska -Dokic1 1 Ss. Cyril and Methodius University in Skopje, Faculty of Medicine, Institute of Microbiology and Parasitology, Republic of North Macedonia 2 University Clinic for Dermatology; Ss. Cyril and Methodius University in Skopje, Faculty of Medicine, Republic of North Macedonia


EVALUATION OF (1,3)-β-D-GLUCAN ASSAY IN DIAGNOSIS OF INVASIVE FUNGAL INFECTIONS WITH ASPERGILLUS
Гордана Мирчевска 1 , Жаклина Цековска 1 , Ана Кафтанџиева 1 , Зорица Зафировиќ 2 , Елена Трајковска-Докиќ 1 Invasive fungal infections caused by Aspergillus are a significant problem in immunocompromised and critically ill patients and associated with increased morbidity and mortality. Early diagnosis of invasive aspergillosis is still a big clinical and diagnostic challenge. Conventional methods are not sensitive enough, and therefore, there is a need for rapid, more sensitive methods for early diagnosis of invasive fungal infections with Aspergillus.

Introduction
Invasive fungal infections are significant causes of morbidity and mortality, especially in immunocompromised patients undergoing steroid treatment, chemotherapy resulting in severe neutropenia, hematopoietic stem cell and solid organ transplantation. 1 AIDS and malignant diseases can also contribute to development of this opportunistic fungal infection. Aspergillosis usually affects the respiratory system and manifests as a broad-spectrum of diseases including aspergilloma, chronic pulmonary aspergillosis, allergic bronchopulmonary aspergillosis and invasive aspergillosis, which is the most aggressive and rapidly spreading form of infection to the brain, heart, liver, and kidneys, with a very high mortality rate. 2 Criteria for diagnosis of invasive aspergillosis have greatly benefited from the European Organisation for the Research and Treatment of Cancer (EORTC) and Mycoses Study Group (MSG) recommendations for defining invasive fungal infections including invasive aspergillosis. 3 To achieve a favorable prognosis of these life-threatening fungal infections, an early initiation of an antifungal therapy is necessary. It relies on a timely and accurate diagnosis, which in turn is still a big laboratory challenge, because clinical symptoms and signs as well as radiological signs are often non-specific. Histopathologic demonstration of microorganisms in tissue specimens or growth of fungal agents in culture media is still the "gold standard" method. 4 However, invasive procedures for specimen collection may be sometimes contraindicated, especially in patients with profound respiratory insufficiency. Conven-tional methods are time-consuming and relatively insensitive, since they are positive in less than 30% of all invasive Aspergillus infections, and they depend on the quality of the specimen submitted. Also, some fungal pathogens require prolonged incubation, which could further delay the mycological diagnosis. 5 Due to all these limitations, a lot of work has been done in recent years for development of alternative nonculture-based diagnostic assays for detection of invasive fungal infections, like detection of fungal biomarkers. Serum (1,3)-β-D-glucan (BDG) is a panfungal marker which is a cell wall polysaccharide, found in many pathogenic fungi including Aspergillus species, that can be present early in the blood and body fluids in patients suffering from invasive fungal infections. Serum β-D-glucan concentrations show a constant rise even before manifestation of clinical signs, and then start to decrease, and eventually become negative if patients respond well to antifungal treatment. 6  is run in duplicate, is read, and optical density values recorded every 30 seconds over a 40-minute period. Findings from 4 different metaanalyses performed over the years have shown that in patients with a higher risk of development of invasive fungal infections, single positive β-D-glucan testing is associated with sensitivity and specificity generally ranging between 60 and 90%. 6 Other studies, performed primarily in patients with hematologic malignancies, have shown that the presence of two consecutively positive β-Dglucan results increase specificity of the assay to almost 99%, suggesting that these results may be used as a diagnostic marker for the presence of an invasive fungal infection. 8 The aim of this study was to evaluate the diagnostic performance, sensitivity and specificity of serum (1,3)-β-D-glucan BDG marker in comparison with conventional methods (culture) for diagnosis of invasive infections with Aspergillus species.

Study design
A prospective diagnostic study was performed at the Institute of Microbiology and Parasitology, Faculty of Medicine, Skopje, Republic of Macedonia, during a 2-year period (2014-2016).

Group of patients and mycological investigations
In this study, clinical specimens (from mucosal surfaces of respiratory tract and blood cultures) from 125 patients divided into 4 groups, according to clinical diagnosis and risk factors for invasive aspergillosis, were analyzed at the Laboratory for diagnosis of fungal infections of the Institute of Microbiology and Parasitology, Faculty of Medicine, Skopje, Republic of North Macedonia. These groups included patients with primary immune deficiency, critically ill patients treated in intensive care units, patients with chronic aspergillosis and cystic fibrosis patients. Invasive fungal infection was defined according to the revised definitions by the EORTC/MSG (European Organization for Research and Treatment of Cancer/Mycoses Study group) consensus group, with the necessary modification that (1,3)-β-D-glucan panfungal marker was not included in the microbiological criteria. 3 The specimens were investigated with conventional mycological methods, by inoculation of specimens on culture media for support of fungal growth (Sabouraud and chromogenic CALB medium (Oxoid)). Blood culture was performed with automated BacT/Alert system (bioMerieux, France), Gram stain and culture on Sabouraud and selective chromogenic CALB medium (Oxoid). Identification of Aspergillus on species level was performed with macroscopic analysis of grown mold colonies and further microscopic analysis of the reproductive elements (conidia) with lactophenol cotton blue method. Detection of (1,3)-β-D-glucan panfungal marker was made by Fungitell assay (Associates of Cape Cod). 7 Table 6. Descriptive statistics for the concentration of the BDG marker in serum Table 7. Diagnostic performances of conventional (blood culture and BAL culture) and serological methods in the group with immune deficiency Table 8. Diagnostic performances of conventional (BAL culture) and serological methods in the group with prolonged ICU stay Table 9. Diagnostic performances of conventional (BAL culture) and serological methods in the group with chronic aspergillosis Comparative diagnostic performances of conventional (blood culture and BAL culture) and panfungal BDG marker for diagnosis of invasive in-fections with Aspergillus in the group with immune deficiency are presented in Table 7. Comparative diagnostic performances of conventional (BAL culture) and serological methods for diagnosis of invasive infections with Aspergillus in the group with prolonged ICU stay in critically ill patients are presented in Table 8. Comparative diagnostic performances of conventional (BAL culture) and serological methods for diagnosis of invasive infections with Aspergillus in the group with chronic aspergillosis are presented in Table 9.

Method Se(%) Sp(%) PPV(%) NPV(%) LR+(%) LR-(%)
In the group with cystic fibrosis, only BAL culture was analyzed, and this method had the following diagnostic performances: sensitivity 100%, specificity 54.17%, positive predictive value 35.29%, negative predictive value 100%, likelihood ratio for positive finding was 2.18%, likelihood ratio for negative finding was 0.  12 In a similar medical center, during a 17year follow-up, fungemia with Aspergillus was registered in 4% of all cases with fungemia. Still, in this study, non-transplant patients with hematological malignancies were also included. 10 In the study of Simoneau and collaborators, only one of 19 cases of fungemia with Aspergillus was confirmed as true fungemia. All cases of aspergillemia were detected during a period of 11 years, with a system based on lysis-centrifugation. 12 Out of 23,000 blood cultures analyzed, only 0.2% demonstrated positivity with growth of Aspergillus. Despite the fact that all blood cultures were investigated with a biosafety cabinet, still, contamination with conidia of filamentous fungi couldn't have been prevented. During recent years, many studies have analyzed true aspergillemia with automated systems, and none of these documented aspergillemia. 12,13 In the study of Simoneau, experimental inoculation of blood culture bottles was performed, with BacT/Alert system, and growth with Aspergillus was confirmed, which additionally adds to the capability of the system to support growth of filamentous fungi. 12 According to Lopes-Bezerra, vascular endothelial cells exposed in vitro to kill hyphae of Aspergillu were continuously destroyed. 14 Probably, viability of endocytosed hyphae of Aspergillus species is deeply compromised, which contributes to small chances for recovery of fungi by blood culture. Although A. fumigatus can grow in blood culture bottles, still, blood cultures from patients with invasive aspergillosis are usually negative, and reasons for this are still unclear. 15 Girmenia et al. presented a small number of positive blood cultures (10 %) in patients with invasive aspergillosis, which contributed to the general perception of a very low sensitivity of blood cultures for diagnosis of invasive aspergillosis. 10 Most scientists agree that positive blood cultures with Aspergillus are very rare, even in high-risk patients, like transplant patients with hematopoetic stem cells, hence most positive blood cultures are actually pseudofungemia, and are not connected with real invasive aspergillosis. Also, some studies suggest that DNA of Aspergillus is free in the blood, so most likely that is the reason for the low sensitivity of blood cultures for diagnosis of invasive aspergillosis. 16 As previously discuissed, clinical and radiological presentation, as well as the number of positive blood cultures and the system of blood cultures used, should be taken into consideration when analyzing the significance of positive blood cultures with Aspergillus. Ussully only one positive blood culture with the automated system means pseudofungemia.

Discussion
In our study, the culture of BAL specimens demonstrated growth of Aspergillus most frequently in the group of chronic aspergillosis (63.33%), followed by 56.67% of patients with cystic fibrosis, 51.43% of patients with primary immune deficiency, and 43.33% of patients with prolonged ICU stay. Sensitivity and specificity of BAL culture was: 64.29% and 100%, 59.09% and 100%, 54.55% and 12.5%, 100% and 54.17%, in I, II, III and IV group respectively. In the study of Tashiro et al., 165 isolates of Aspergillus species were detected in culture of respiratory tract of 139 patients. Of these, 62 (45%) were colonized with Aspergillus, but didn't demonstrate clinical symptoms of aspergillosis, and the other 77 patients (55%) had some type of pulmonary aspergillosis classified as chronic (48%), aspergilloma (29%), invasive (13%), or АBPA (10%). In the study of Tashiro, patients with chronic necrotizing aspergillosis or aspergillom, most frequently had COPD, tuberculosis or cancer of the lungs. Some of them had received systemic immunosupresive drugs for a prolonged period, or had some chronic diseases like diabetes, cancer or hepatic chirrosis. 17 In patients with invasive aspergillosis, the main predisposing factor had been hematological malignancy, and they were subsequently treated with immunosupresive drugs. Patients with ABPA frequently demonstrated signs of bronchial asthma (88%) or other atopic diseases (63%  21 In our study, we did not isolate A. niger in BAL cultures of our patients. Although A. fumigatus is considered as the most pathogenic species, still this species can frequently be a colonizer of the respiratory tract without any clinical manifestation of invasive aspergillosis, which was also registered in our study, especially in those patients categorized as possible infections according to EORTC/ MSG criteria. Diagnostic value of Aspergillus identification in respiratory specimens is sometimes questionable, since it is very difficult for the clinican to differentiate between colonization and infection. According to Ader, discovery of the same species of Aspergillus in more specimens during an antibiotic treatment, without favorable pharmacological response, in patients with a high risk, should raise a concern for the development of invasive aspergilosis. 22 Therefore, isolation of Aspergillus from respiratory tract specimens in critically ill patients with high risk and clinical signs of pneumonia requires a faster decision for a prompt initiation of antifungal treatment. 23,24 Although in some cases colonization is transient in the respiratory tract, still it could present as a serious warning sign of an infection with Aspergillus. 25  In this analysis, sensitivity and specificity of the BDG assay was 16% and 94%, respectively. Sensitivity was 33% in patients with disseminated aspergillosis, but only 3% in patients with localized infection. Lower sensitivity of the assay was registered among patients with localized infection with Aspergillus compared to those with invasive form. This difference was statistically significant (p<0.0001). 32 In the group of cystic fibrosis, no elevated values of the panfungal marker was registered. Theel et al., evaluated the performance of the BDG assay in serum, for identification of invasive fungal infections in immunocompromised patients with proven, probable and possible aspergillosis according to EORTC/MSG criteria. 33

Conclusions
The results of this study have indicated that no single method could provide definite etiological diagnosis of invasive fungal infection caused by Aspergillus. When using a conventional method, it is neccessery to provide more specimens from each patient, in frequent time intervals, and cautiously interpret the results obtained, since colonisation with fungi without clinical signs of infection is possible. Still, clinicians should be aware that these methods are timeconsuming, with low sensitivity, and depend on the quality of the specimen submitted.
Analysis of the serological panfungal (1,3)-beta-D-glucan marker has demonstrated that this assay could be an additional useful diagnostic tool for screening of invasive fungal infections, but results should be interpreted alongside other clinical and laboratory findings.
In conclusion, implementation and analysis of different microbiological methods, as well as appropriate interpretation of results, in collabora-tion with clinicians, is the most important aspect towards accurate and precise etiological diagnosis of invasive aspergillosis and earlier start of antifungal treatment in order to achieve favorable clinical outcome.