Table 2.

Alternative testing that can be performed with nontissue specimens based on the morphology present in tissue and the suspected fungi

Morphology using H&E, GMS, and PAS staining	Suspect fungus(i)	Alternative test with nontissue samples	Comment(s)
Broad-based budding yeasts (10–15 μm)	Blastomyces dermatitidis	Antigens in urine or serum	EIA antigen assay for urine, serum, BAL fluid, or CSF is available at MiraVista Diagnostics (Indianapolis, IN) (sensitivity has been reported to be around 90%, but this test should be performed simultaneously with Histoplasma testing because of cross-reactivity between the two antigens); EIA can be used to follow response to treatment
		Serology	There is poor sensitivity (9 to 28%) and specificity if using complement fixation and immunodiffusion, radioimmunodiffusion and EIAs have better sensitivity (77 to 83%) and specificity (95%) but are not commercially available
		Cultures of sputum or BAL fluid	Diagnostic yields of 86 to 92%, respectively; however, delays in diagnosis are inherent to the technique
Narrow-based budding yeasts (4–10 μm) with a thick capsule	Cryptococcus spp.	Cryptococcal antigen	Sensitivity of >90% but may have up to 7% false-positive results; latex agglutination and EIA with serum and CSF are FDA approved, but testing in urine is not; laboratories can perform the test with urine samples if they validate for the sample type
		Cultures of blood, CSF, and other fluids	Use of canavanine-glycine-bromthymol blue medium is helpful to distinguish Cryptococcus gattii from C. neoformans
Small yeasts (2–4 μm) with narrow-based budding grouped in clusters inside macrophages	Histoplasma capsulatum	Antigens in urine or serum	Several commercial assays are available; sensitivity is about 90% in cases of disseminated disease and 75% during acute pulmonary disease; false-positive results occur in cases with other endemic mycoses, with blastomycosis being the most important overlap; Aspergillus galactomannan also cross-reacts
		Serology	Accomplished by the use of complement fixation or immunodiffusion with about 80% sensitivity, useful for immunocompetent individuals but may not be useful when the patient is immunosuppressed
		Cultures of blood and other fluids	It is important to use lysis-centrifugation to release the organisms from phagocytic cells; although specific, culture may take up to 6 weeks
Spherules with multiple endospores	Coccidioides immitis/posadasii	Antigens in urine or serum	An EIA for Coccidioides antigen in urine, serum, and CSF is commercially available (MiraVista Diagnostics, Indianapolis, IN); studies using rabbit antibodies against a Coccidioides galactomannan in an EIA showed that 71% of patients with coccidioidomycosis have antigenuria; however, EIA cross-reacts in 10% of patients with other endemic mycoses
		Serology	Measurement of combined IgM and IgG by EIA shows a sensitivity and specificity of >95%, immunodiffusion can also be used to measure IgM and IgG
		Cultures of any sample	Coccidioides is a select agent with stringent culture regulations
		DNA detection in sputum	PCR detection of ITS2 or Ag2/PRA targets has been published, showing good sensitivity, but assays are not commercially available
Small yeasts (3–5 μm) intermingled with pseudohyphae and/or hyphae	Candida spp.	Detection of beta-d-glucan in serum	Sensitivity will vary between 57 to 90% depending on the patient population and cutoff values used for the assay, specificity ranges from 44 to 92%
		Cultures of blood and other fluids	Blood cultures occur in 50 to 70% of cases, and when positive, they indicate invasive disease; peptide nucleic acid fluorescent in situ hybridization assay (Yeast “Traffic Light” PNA-FISH assay; AdvanDx Inc., Woburn, MA) of positive blood culture smears can identify the most frequent Candida spp. without the need for subcultures
		DNA detection in whole blood, serum, or plasma	Several PCRs have been developed and validated; they seem to have increased sensitivity compared to blood cultures, but the clinical significance of these findings is still being researched
Nonpigmented (hyaline), septated hyphae with acute-angle branching	Aspergillus spp., Fusarium spp., Scedosporium spp., Trichoderma spp., and Paecilomyces spp.	Detection of galactomannan in serum or BAL fluid	Galactomannan antigen detection using commercially available Aspergillus EIA (Platelia Aspergillus test; Bio-Rad, Hercules, CA) has sensitivities that range from 40 to 100% and specificities from 56 to 100%, depending on the patient population; false-positive results occur in 50% of patients taking certain antibiotics or plasmalyte or having other fungal infections; other genera in this morphological category do not have specific alternative tests, although galactomannan has been shown to be positive in patients with Paecilomyces infections
		Detection of beta-d-glucan in serum	Measured using commercially available EIA (Fungitell kit; Associates of Cape Cod, East Falmouth, MA) with sensitivities that range from 50 to 100% and specificities of from 44 to 98%, the presence of beta-glucan is not specific for particular fungal genera
		Cultures of blood	Fusarium spp. may have positive blood cultures, other genera are rarely cultured from the blood, even in the presence of invasive disease
Nonpigmented (hyaline), pauciseptate ribbon-like hyphae with right-angle branching	Mucorales genera	Serology	Serologic tests for invasive disease have been attempted but have not been clinically useful
		Culture of blood	Mucorales genera are rarely recovered from blood, even with invasive disease
Pigmented irregular hyphae and yeast-like structures both with septations	Dematiaceous fungi, including Madurella spp., Fonsecaea spp., Cladophialophora spp., Exophiala spp., Curvularia spp., Bipolaris spp., and others	Serology	Not available
		Culture of blood or other fluids	These genera are rarely cultured from the blood, even in the presence of invasive disease