Although invasive mycoses have long been recognized for a long time as being caused by significant pathogens, particularly in immunocompromised patients, the frequency of infection with opportunistic fungi is increasing with time and the spectrum of the infectious agent of invasive mycoses is changing [1–4]. In this regard, advances in therapeutic technologies and in particular the development of novel immunosuppressive therapies, have prolonged the period of risk for many individuals [3, 4]. A large number of invasive mycoses have been identified because of the increase in the number of patients at risk [24–26]. Most of the invasive mycoses are caused by filamentous fungi, rarely by other fungi [6, 27–29]. In this regard, a number of IFFD has been reported in renal transplant patients, particularly aspergillosis and mucormycosis . Other IFFD have been more rarely described in immunocompromised hosts, such as scedosporiosis .
A confirmed diagnosis of IFFD is made on the demonstration of hyphae in the diseased tissue. Pathological identification of the hyphae is based on different well-described criteria: diameter, presence of septa, and branching angle (right or acute branching), pigmentation [32, 33]. It is important to consider the characteristics of the infiltrate at the inflammatory region that is in contact with hyphae: presence of angioinvasion, of giant cells and of eosinophilic necrosis . Mycological analysis made on non fixed-tissue specimens, associating microscopic examination and culture allows identification of the species and the study of the sensitivity to anti-fungal drugs. However, the different results of these mycological analyses can be negative in the following circumstances: 1) When there is an error in sampling, the specimen sent to the mycology laboratory does not contain the different fungi, 2) When the specimen containing the hyphae is totally necrotic. In addition, when two filamentous mycoses are present in the same organ, the mycological results are even more inconstant, since only one fungus might be present on the specimen sent to the mycology laboratory. In these different situations, it is essential to compare the culture results and the morphological analysis of the tissue sections. However, histological analyses have some limits, even after histochemical staining.
The distinction between aspergillosis and scedosporiosis, between aspergillosis and fusariosis and certain mucormycosis, such as those caused by Cunninghamella from tissue sections may be difficult or impossible [31, 35, 36]. Other potential pitfalls are more easily eliminated [37, 38]. Ancillary methods using specific antibodies raised against filamentous fungi can be helpful for diagnosis. However, the different antibodies are rarely commercially available [33, 36, 39]. Finally, in situ hybridization has been developed for fungal detection in fixed-tissue sections, but the probes used for this method are rarely available in the pathology laboratory 
In recent years, PCR methods performed with deparaffinized tissue sections have been used to try to improve the detection and identification of pathogens in fixed specimens [39, 41]. This methodology needs to use strict positive and negative controls and exact comparison with histological features and mycological analysis is required. However, false-negative and false-positive results due to artefacts in amplification, or sampling, and possible exogenous contamination by microorganisms, still are very important pitfalls in molecular diagnostic pathology [42, 43]. In the present case, there was no evidence of contamination by external sources and strict laboratory precautions were applied to avoid carry over and false positive results.
The present case report underlines the usefulness of the PCR method performed with embedded paraffin fixed specimens, which allows simultaneous detection on the same tissue of two types of IFFD. Moreover, the pathogenic status of these pathogens can be confirmed by comparing histopathological and molecular biology results obtained from the same embedded paraffin tissue. In the present case, the absence of mycological diagnosis of aspergillosis from non-fixed biopsy specimen was probably due to an error in sampling. In this regard, histological analysis showed only a few areas with the two associated mycoses. Filaments of aspergillosis, but not of mucorales, were noted by cytopathological analysis of the BAL and identified as A. fumigatus in culture. This can be due to an error in sampling, but also because at that time mucorales was undetectable, since it grew subsequently in the patient treated with voriconazole. However, the clinical distinction between aspergillosis and mucormycosis is crucial since there is an increased incidence of mucormycosis in patients treated with voriconazole for suspected aspergillosis . In this regard, a couple of cases of mucormycosis occurring in patients with voriconazole-treated aspergillosis have been reported [9, 44]. The diagnosis of IFFD based on hyphae isolated from the BAL is uncertain. In this case, different associated clinical features and predisposing host factors must be present before giving specific treatment, which can be toxic, ineffective, and favor the development of another mycotic disease [45, 46]. Finally, for most clinicians, the identification of aspergillosis or mucormycosis in the BAL is not sufficient to confirm the diagnosis of IFFD . The utility of performing transbronchial or transparietal biopsies in debilitated patients should be discussed [48, 49].