Detection of Microorganisms in Granulomas That Have Been Formalin-Fixed: Review of the Literature Regarding Use of Molecular Methods

Table 3

Methods that have been used to define nonmycobacterial infectious agents that can cause granulomatous inflammation.

Disease	PCR	Serology	Culture	Visualization of organism in histology or other	Ref.
Bartonella henselae	From lesion, 12 of 14 (86%); in another series that studied 33 granulomatous lymphadenitis 7 cases positive.	23 of 23 positive.	Recovery in cultures approaches 50%.	Requires bacterial silver staining for visualization, was detected in 12 of 19 (63%).	[5, 51]
Brucella	From bone marrow aspirates, serum and blood. Can be used after antibiotics have been given.	Several methods with varying sensitivity (agglutination, ELISA).	Sensitivity depends on stage of disease and sample type, requires extended culture time as the bacteria grows slowly. Laboratory acquired cases can occur, careful handling is needed.	Requires IHC or immune-fluorescence for visualization.	[26, 57]
Tularemia	RT-PCR from blood and tissues in 39% of 101 patients. 14 of 15 cases in tissues.	Microagglutination, immunofluorescence for detection of IgG and IgM in 94% of 101 patients.	Culture in 21% of 101 patients.	Requires bacterial silver staining or IHC for visualization	[58, 59]
Yersinia enterocolitica	Not done.	ELISA testing is not useful.	Most cases diagnosed this manner. MALDI-FOF has been used for specific identification.	Not commented.	[51, 60, 61]
Syphilis	PCR 39 to 69% of biopsies.	Varies depending on stage.	Not available.	Requires bacterial silver stain (Dieterle or Steiner) for visualization (25%); increases sensitivity with IHC (49 to 51%).	[62, 63]
Leishmania	Used for detection and defining species, in meta-analysis very high diagnostic odds ratio. Using skin formalin-fixed, paraffin-embedded samples detection rate is 97%.	Limited sensitivity which depends on assay used, type of disease (cuteanous versus visceral).	Can be done but not routine with 58% sensitivity.	Visualization with H&E, but when numbers of parasites are low, they may be difficult to diagnose. Enhanced with IHC to 88% sensitivity.	[64–67]
Toxoplasma	Can be performed in tissues and blood, different primers, good for diagnosis of congenital disease.	Variety of assays available. IgG indicates past infection, IgM can remain increased up to 2 years after acute infection.	By inoculating mice (in reference laboratories).	Visualization with H&E in placenta and other tissues. Enhanced with IHC.	[68–70]
Acanthamoeba	Can reach 89% sensitivity in specimens from patients with keratitis. Cyst formation decreases sensitivity. Species can be defined using PCR.	Not available.	Is done primarily for diagnosis of keratitis using a lawn of Escherichia coli.	Amoeba are visualized with H&E in patients with granulomatous meningo encephalitis or keratitis; enhanced diagnosis by use of immune assays (DFA or IHC).	[71–74]
Schistosoma	PCR in urine can have up to 100% sensitivity and 91% specificity.	Detection of antibodies: may cross-react with other helminth infections, IgM may persist long after acute infection. Detection of antigen: in urine and serum. Sensitivity ranges from 41–78% and specificity between 76–100%.	Not available.	Detection of ova in stool or urine; quantification in a fixed amount of urine or stool allows to determine intensity of infestation. Adult worms and eggs can be seen in tissue sections.	[75–78]
Fasciola	PCR from stools and eggs from adult worms obtained from humans and animals.	By ELISA the sensitivity is 95% and specificity 95%. No correlation with number of ova in stool.	Not available.	Detection of ova in stool. Adult worms and eggs can be seen in tissue sections.	[79–81]
Anisakis	Case detected by PCR from ileocecal formalin-fixed, paraffin-embedded tissue.	Seroprevalence using IgE ELISA has been documented to be 6% in Korea.	Not available.	Visualized with H&E.	[82, 83]
Pneumocystis jirovecii	PCR has been used in bronchioalveolar lavage for diagnosis and to define colonization in transplant patients. Can be used for genotyping.	(1 → 3) β-D-glucan has a diagnostic sensitivity of 95% but specificity is 86% since there is cross-reactivity with other fungal infections.	Not available.	Visualized with H&E or fungal silver stain. Fluorescent antibodies can be used.	[12, 84–87]
Histoplasma	Sensitivity of PCR on formalin-fixed, paraffin-embedded tissues 89%. From fresh specimens sensitivity is 100% and specificity 73%.	Urine and serum antigen cross-react with Blastomyces. Antibodies can also be measured.	Can take up to 4 weeks to grow. Can be found in blood cultures.	Visualized with H&E, PAS and fungal silver stain (these are small yeasts with narrow based budding).	[84, 88, 89]
Blastomyces	From fresh specimens sensitivity is 99% and specificity 86%.	Urine and serum antigen cross-react with Histoplasma. Antibodies can also be measured.	Grows well but may take several weeks.	Visualized with H&E, PAS and fungal silver stain (these are 6–15 micron yeasts with broad based budding).	[84, 89, 90]
Coccidioides	PCR on fresh respiratory specimens has a sensitivity of 75 and specificity of 99%.	Antibodies can be measured using several methods. IgG response can be abrogated if treatment is started early.	Grows within 2 to 3 weeks. It is a select agent and it needs to be handled with care due to potential laboratory transmission.	Visualized with H&E, PAS and fungal silver stain (diagnostic structures are spherules with endospores; endospores on their own can be confused with Blastomyces and other yeasts).	[84, 91–93]
Paracoccidioides	PCR has been performed in tissues, including fine needle aspirates. There are commercial kits available.	Several methods available, including Western blots. Some have important cross-reactivity with other yeasts, particularly Histoplasma.	Growth may take up to 2 months.	Visualized with H&E, PAS and fungal silver stain (yeasts with multiple—more than 3 buds, mariner's wheel, are diagnostic).	[84, 94–96]

IHC: immunohistochemistry, H&E: hematoxylin and eosin, DFA: direct fluorescent antibody, PAS: periodic acid shift stain.