During the last 2 decades, the approach to histopathologic diagnosis has been dramatically transformed by immunohistochemistry (IHC), specifically in the diagnosis and classification of tumors, and more recently in the diagnosis of infectious diseases in tissue samples.
Pathologists play an important role in recognizing infectious agents in tissue samples from patients, providing a rapid morphologic diagnosis and facilitating clinical decisions in patient treatment. In many cases, when fresh tissue is not available for culture, pathologists can provide a rapid morphologic diagnosis and facilitate clinical decisions in patient treatment. In addition, pathologists have played a central role in the identification of emerging and reemerging infectious agents and describing the pathogenetic processes of emerging diseases, such as hantavirus pulmonary syndrome, other viral hemorrhagic fevers, Middle Eastern respiratory syndrome (MERS), severe acute respiratory syndrome (SARS), leptospirosis, and rickettsial and ehrlichial infections, as well as the diagnosis of anthrax during the bioterrorist attacks of 2001.
Traditionally, microbial identification in infectious diseases has been made primarily by using cultures and serologic assays. However, fresh tissue is not always available for culture, and culture of fastidious pathogens can be difficult and may take weeks or months to yield results. Moreover, culture alone cannot distinguish colonization from tissue invasion. In addition, serologic results can be difficult to interpret in the setting of immunosuppression or when only a single sample is available for evaluation. Some microorganisms have distinctive morphologic characteristics that allow their identification in formalin-fixed tissues using routine and special stains. Nevertheless, in many instances, it is difficult or even impossible to identify an infectious agent specifically by conventional morphologic methods. Furthermore, microorganisms may be few or distributed focally in the tissue; thus the absence of demonstrable microorganisms does not entirely rule out infection given the possibility of sampling error. In other instances, more than one microorganism or pathologic process may coexist.
IHC is one of the most powerful techniques in surgical pathology. There has been an increasing interest in the use of specific antibodies to viral, bacterial, fungal, and parasitic antigens in the detection and identification of the causative agents in many infectious diseases. The use of a specific antibody to detect a microbial antigen was first performed by Coons and associates to detect pneumococcal antigen in tissues. The advantages of IHC over conventional staining methods ( Box 3.1 ) and the contributions of IHC in infectious diseases ( Box 3.2 ) are substantial. In many instances, IHC has shown high specificity, allowing the differentiation of morphologically similar microorganisms. IHC is especially useful when microorganisms are difficult to identify by routine or special stains, are fastidious to grow, or exhibit atypical morphology ( Box 3.3 ). It is important to understand that there may be widespread occurrence of common antigens among bacteria and pathogenic fungi, and both monoclonal and polyclonal antibodies must be tested for possible cross-reactivities with other organisms and validated before clinical use. Finally, it is important to emphasize that IHC has several steps and all of them can affect the final result; however, in general, the only limitations are the availability of specific antibodies and the preservation of epitopes.
- 1.
Opportunity for rapid results
- 2.
Reduced risk of exposure to serious infectious diseases by performance on formalin- fixed, paraffin-embedded tissue
- 3.
High sensitivity allowing identification of infectious agents even before morphologic changes occur
- 4.
Opportunities for retrospective diagnosis of individual patients and for in-depth study of the disease
- 5.
Specific identification of infectious agents with many monoclonal antibodies and some polyclonal antibodies
- 1.
Allows identification of new human pathogens
- 2.
Allows microbiologic-morphologic correlation establishing the pathogenic significance of microbiologic results
- 3.
Provides a rapid morphologic diagnosis allowing early treatment of serious infectious diseases
- 4.
Contributes to understanding of the pathogenesis of infectious diseases
- 5.
Provides a diagnosis when fresh tissue is not available or when culture methods do not exist
- 1.
Identification of microorganisms that are difficult to detect by routine or special stains
- 2.
Detection of microorganisms that are present in low numbers
- 3.
Detection of microorganisms that stain poorly
- 4.
Identification of microorganisms that are fastidious to grow or noncultivable
- 5.
Identification of microorganisms that exhibit atypical morphology
Table 3.1 lists some commercially available antibodies for diagnostic use in surgical pathology.
Microorganism | Antibody/Clone | Dilution | Pretreatment | Source |
---|---|---|---|---|
Adenovirus | Mab/20/11 and 2/6 | 1 : 2000 | Proteinase K | Chemicon |
Aspergillus spp. | Mab/WF-AF-1 | 1 : 50 | HIAR | LSBio |
Bartonella henselae | Mab | 1 : 100 | HIAR | Biocare Medical |
BK virus | Mab/BK T.1 | 1 : 8000 | Trypsin | Chemicon |
Candida albicans | Mab/1B12 | 1 : 400 | HIAR | Chemicon |
Chlamydophila pneumoniae | Mab/RR402 | 1 : 200 | HIAR | Accurate |
Cryptosporidium | Mab/Mabc1 | 1 : 100 | HIAR | Novocastra |
CMV | Mab/DDG9/CCH2 | 1 : 50 | HIAR | Novocastra |
Clostridium spp. | Rabbit polyclonal | 1 : 1000 | None | Biodesign |
Giardia intestinalis | Mab/9D5.3.1 | 1 : 50 | HIAR | Novocastra |
Hepatitis B core antigen | Rabbit polyclonal | 1 : 2000 | HIAR | Dako |
Hepatitis B surface antigen | Mab/3E7 | 1 : 100 | HIAR | Dako |
Herpes simplex 1 and 2 viruses | Rabbit polyclonal | 1 : 3200 | HIAR | Dako |
Helicobacter pylori | Rabbit polyclonal | 1 : 40 | Protease I | Dako |
HHV 8 | Mab/LNA-1 | 1 : 500 | HIAR | Novocastra |
Klebsiella pneumoniae | Rabbit polyclonal | 1 : 200 | Proteinase K | Biogenesis |
Listeria monocytogenes | Rabbit polyclonal | 1 : 5000 | Proteinase K | Difco |
Mycoplasma pneumoniae | Mab/1.B.432 | 1 : 25 | HIAR | US Biological |
Parvovirus B19 | Mab/R92F6 | 1 : 500 | HIAR | Novocastra |
Pneumocystis jirovecii | Mab/3F6 | 1 : 20 | HIAR | Novocastra |
Plasmodium falciparum | Mab/BDI400 | 1 : 1000 | Proteinase K | Biodesign |
Prion | Mab/3F4 | 1 : 200 | HIAR | Dako |
Prion | Mab/12F10 | 1 : 1000 | Proteinase K | Cayman Chemical |
Prion | Mab/KG9 | 1 : 1000 | Proteinase K | TSE Center Resource |
RSV | Mab/5H5N | 1 : 200 | HIAR | Novocastra |
Rhizopus arrhizus | Mab/WSS-RA-1 | 1 : 100 | HIAR | LSBio |
Staphylococcus aureus | Rabbit polyclonal | 1 : 500 | Proteinase K | Biodesign |
Treponema pallidum | Rabbit polyclonal | HIAR | Biodesign | |
Toxoplasma gondii | Rabbit polyclonal | 1 : 320 | HIAR | Biogenex |
West Nile virus | Mab/5H10 | 1 : 400 | Proteinase K | Bioreliance |
Viral Infections
IHC has played an important role not only in the diagnosis of a large number of viral infections but also in the study of their pathogenesis and epidemiology. Conventionally, the diagnosis of viral infections has relied on cytopathic changes observed by routine histopathology. Several viral pathogens produce characteristic intracellular inclusions, which allow pathologists to make a presumptive diagnosis of viral infection. However, for some viral infections, the characteristic cytopathic changes are subtle and sparse, requiring a meticulous search. Moreover, only 50% of the known viral diseases are associated with characteristic intracellular inclusions. In addition, formalin, which is the most commonly used fixative in histopathology, is a poor fixative for demonstrating the morphologic and tinctorial features of viral inclusions. When viral inclusions are not detected in hematoxylin and eosin (H&E)–stained sections, or when the viral inclusions present cannot be differentiated from those of other viral diseases, immunohistochemical techniques offer a more reliable approach to reach a specific diagnosis.
Hepatitis B
Hepatitis B virus infection constitutes an important cause of chronic hepatitis in a significant proportion of patients. In many instances, the morphologic changes induced by hepatitis B virus in hepatocytes are not typical enough to render a presumptive diagnosis of hepatitis B viral infection. In other instances, there may be so little hepatitis B surface antigen (HBsAg) that it cannot be demonstrated by techniques such as orcein staining. In these cases, immunohistochemical techniques to detect HBsAg are more sensitive than histochemical methods and are helpful in reaching the diagnosis. Immunostaining for HBsAg has been used in the diagnosis of hepatitis B and in the study of carrier states. Eighty percent or more of cases with positive serologic results for HBsAg demonstrate cytoplasmic HBsAg using IHC. By immunoperoxidase localization, hepatitis B core antigen (HBcAg) can be demonstrated within the nuclei or the cytoplasm of hepatocytes, or both. Predominantly cytoplasmic expression of HBcAg is associated with a higher grade of hepatitis activity, and diffuse immunostaining of nuclei for HBcAg generally suggests uncontrolled viral replication in the setting of immunosuppression. Immunostaining for HBsAg and HBcAg is useful in the diagnosis of recurrent hepatitis B infection in liver allografts, particularly when present with atypical histopathologic features.
Herpesviruses
Histologically, the diagnosis of herpes simplex virus (HSV) infection involves the detection of multinucleated giant cells containing characteristic molded, ground glass-appearing nuclei and Cowdry type A intranuclear inclusions. When there are abundant viral inclusions within infected cells, the diagnosis is usually straightforward. However, the diagnosis of HSV infection can be difficult when the characteristic intranuclear inclusions or multinucleated cells, or both, are absent or when the amount of tissue in a biopsy specimen is small. In these cases, IHC using either polyclonal or monoclonal antibodies against HSV antigens has proven to be a sensitive and specific technique to diagnose HSV infections ( Fig. 3.1 ).
Although polyclonal antibodies against the major HSV glycoprotein antigens are sensitive, they do not allow distinction between HSV-1 and HSV-2 because these two viruses are antigenically similar. In addition, the histologic features of HSV infection are not specific and can also occur in patients with varicella-zoster virus (VZV) infection. Monoclonal antibodies against the VZV envelope glycoprotein gp1 are sufficiently sensitive and specific to allow a clear-cut distinction between HSV and VZV infections.
IHC has also been useful in demonstrating the association of human herpesvirus 8 (HHV-8) with Kaposi sarcoma, primary effusion lymphoma, and multicentric Castleman disease. The diagnosis of Kaposi sarcoma may be problematic due to its broad morphologic spectrum and similar appearance to other benign and malignant neoplastic vascular lesions. Immunostaining of Kaposi sarcoma latent associated nuclear antigen-1 (LANA-1) is useful to confirm the diagnosis of Kaposi sarcoma, particularly in difficult early lesions that can closely resemble the appearance of interstitial granuloma annulare and when the neoplasm presents in an unusual location, and allows distinction of Kaposi sarcoma from several morphologically similar vasoproliferative lesions. Immunostaining is restricted to the nuclei of spindle cells and endothelial cells of the slit-like vascular spaces ( Fig. 3.2 ). IHC has also demonstrated expression of HHV-8 LANA-1 in mesothelial cells of human immunodeficiency virus (HIV)-associated recurrent pleural effusions.
Cytomegalovirus (CMV) continues to be an important opportunistic pathogen in immunocompromised patients; it is estimated that 30% of transplant recipients experience CMV disease. The range of organ involvement in posttransplant CMV disease is wide; hepatitis occurs in 40% of liver transplant recipients, and pneumonitis is more frequently seen in heart and heart-lung transplant patients. Other organs that are commonly affected are the gastrointestinal tract and the peripheral and central nervous systems. Histologic diagnosis of CMV in fixed tissues usually rests on the identification of characteristic cytopathic effects, including intranuclear or cytoplasmic inclusions, or both. However, histologic examination lacks sensitivity, and in some cases atypical cytopathic features can be confused with reactive or degenerative changes. In addition, up to 38% of patients with gastrointestinal CMV disease fail to demonstrate any inclusions. In these cases, IHC using monoclonal antibodies against early and late CMV antigens allows the detection of CMV antigens in the nucleus and cytoplasm of infected cells ( Fig. 3.3 ). The sensitivity of IHC for detecting CMV infection ranges from 78% to 93%. In addition, IHC may allow detection of CMV antigens early in the course of the disease when cytopathic changes have not yet developed. For example, CMV early nuclear antigen is expressed 9 to 96 hours after cellular infection and indicates early active viral replication. IHC has been useful in the detection of CMV infection in patients with steroid refractory ulcerative colitis, leading to recommendation for the routine use of IHC for the detection of CMV in the evaluation of these patients. CMV immunostaining has been used in detecting occult CMV infection of the central nervous system in liver transplant patients who develop neurologic complications. It has also been used to demonstrate a high frequency of CMV antigens in tissues from first trimester abortions. CMV is the most common opportunistic organism found in liver biopsies from transplant patients; nonetheless, the incidence of CMV hepatitis appears to be decreasing due to better prophylactic treatments. Although CMV hepatitis presents with characteristic neutrophilic aggregates within the liver parenchyma, atypical features suggestive of acute rejection or changes indistinguishable from those of any other viral hepatitis are occasionally observed. In addition, parenchymal neutrophilic microabscesses have been described in cases with no evidence of CMV infection. It is in these cases where immunostaining for CMV antigens is most useful in determining the diagnosis of CMV infection.
The sensitivity of IHC is better than light microscopic identification of viral inclusions and compares favorably with culture and in situ hybridization (ISH). In addition, immunohistochemical assays can be completed faster than the shell vial culture technique, allowing for rapid results that are important for early anti-CMV therapy.
Other herpesvirus infections that have been diagnosed using immunohistochemical methods include human herpesvirus 6 infection and Epstein-Barr viral infection. IHC has been used to identify Epstein-Barr virus (EBV) latent membrane protein-1 in cases of Hodgkin lymphoma and posttransplant lymphoproliferative disorder ( Fig. 3.4 ).
Adenoviruses
Adenovirus is increasingly recognized as a cause of morbidity and mortality among immunocompromised patients owing to transplant and congenital immunodeficiency. Rarely, adenovirus infection has been described in HIV-infected patients. Characteristic adenovirus inclusions are amphophilic, intranuclear, homogeneous, and glassy. However, in some cases, the infection may contain only rare cells showing the characteristic cytopathic effect. In addition, other viral inclusions, including CMV, human papilloma virus, HSV, and VZV, can be mistaken for adenovirus inclusions and vice versa. In these circumstances, immunohistochemical assay may be necessary for a definitive diagnosis. A monoclonal antibody that is reactive with all 41 serotypes of adenovirus has been used in an immunohistochemical technique to demonstrate intranuclear adenoviral antigen in immunocompromised patients ( Fig. 3.5 ). However, many other antibodies react with only a limited number of adenovirus serotypes. Histologic diagnosis of adenovirus colitis is difficult, and it is usually underdiagnosed. Moreover, in immunosuppressed patients, the incidence of coinfection with other viruses is high, and the presence of adenovirus tends to be overlooked. Immunohistochemical staining has been of value in differentiating adenovirus colitis from CMV colitis.
Parvovirus B19 Infection
Parvovirus B19 has been associated with asymptomatic infections, erythema infectiosum, acute arthropathy, aplastic crisis, hydrops fetalis, and chronic anemia and red cell aplasia. In addition, Parvovirus B19 infection has been recognized as an important cause of severe anemia in immunocompromised leukemic patients receiving chemotherapy.
The diagnosis of parvovirus infection can be achieved by identifying typical findings in bone marrow specimens, including decreased or absent red cell precursors, giant pronormoblasts, and eosinophilic or amphophilic intranuclear inclusions in erythroid cells. Because intravenous immunoglobulin therapy is effective, a rapid and accurate diagnostic method is important. IHC with a monoclonal antibody against VP1 and VP2 capsid proteins has been used as a rapid and sensitive method to establish the diagnosis of parvovirus B19 infection in formalin-fixed, paraffin-embedded tissues. IHC is of particular help in detecting parvovirus B19 antigen in cases with sparse inclusions to study cases not initially identified by examination of routinely stained tissue sections or in cases of hydrops fetalis where there is advanced cytolysis ( Fig. 3.6 ). Several studies have found a good correlation between morphologic, immunohistochemical, ISH, and polymerase chain reaction (PCR) results.
Viral Hemorrhagic Fevers
Since the 1960s, numerous emerging and reemerging agents of viral hemorrhagic fevers have attracted the attention of pathologists. These investigators have played an important role in the identification of these agents and in supporting epidemiologic, clinical, and pathogenetic studies of the emerging viral hemorrhagic fevers. Viral hemorrhagic fevers are often fatal, and in the absence of bleeding or organ manifestations, these diseases are clinically difficult to diagnose and frequently require handling and testing of potentially dangerous biological specimens. In addition, histopathologic features are not pathognomonic, and they can resemble other viral, rickettsial, and bacterial (e.g., leptospirosis) infections. IHC is also essential and has been successfully and safely applied to the diagnosis and study of the pathogenesis of these diseases.
Several studies have established the utility of IHC as a sensitive, safe, and rapid diagnostic method for the diagnosis of viral hemorrhagic fevers such as yellow fever ( Fig. 3.7 ), dengue hemorrhagic fever, Crimean-Congo hemorrhagic fever, Argentine hemorrhagic fever, Venezuelan hemorrhagic fever, and Marburg disease. In addition, a sensitive, specific, and safe immunostaining method has been developed to diagnose Ebola hemorrhagic fever in formalin-fixed skin biopsies ( Fig. 3.8 ). IHC demonstrated that Lassa virus targets primarily endothelial cells, mononuclear inflammatory cells, and hepatocytes ( Fig. 3.9 ).
Polyomaviruses
BK virus infections are frequent during infancy; in immunocompetent individuals, the virus remains latent in the kidneys, central nervous system, and B-lymphocytes. In immunocompromised patients, the infection reactivates and spreads to other organs. BK virus nephropathy is also an important cause of graft failure in patients with renal transplant with prevalence varying from 2% to 4.5% in different transplant centers. Because specific clinical signs and symptoms are lacking in BK virus nephropathy, the diagnosis can only be made histologically in a graft biopsy. In the kidney, the infection is associated with mononuclear interstitial inflammatory infiltrates and tubular atrophy, findings that can be difficult to distinguish from acute rejection. Furthermore, the cytopathic changes observed in BK virus infection are not pathognomonic and can be observed in other viral infections. Moreover, in early BK virus infection, there are minimal or no histologic changes, although IHC can identify viral antigen. In this setting, IHC with an antibody against the large T antigen of SV40 virus has been effective in demonstrating BK virus infection ( Fig. 3.10 ).
The human polyomavirus John Cunningham (JC) is a double-stranded DNA virus that causes progressive multifocal leukoencephalopathy (PML). This fatal demyelinating disease characterized is by cytopathic changes in oligodendrocytes and bizarre giant astrocytes. In addition to detection by antibodies to SV40-T antigen, IHC using a polyclonal rabbit antiserum against the protein VP1 is a specific, sensitive, and rapid method for confirming the diagnosis of PML. JC virus antigen is usually seen within oligodendrocytes ( Fig. 3.11 ) and occasional astrocytes, and antigen-bearing cells are more commonly seen in early lesions.
Other Viral Infections
IHC has also been used to confirm the diagnosis of respiratory viral diseases such as influenza A, H1N1 (swine flu), H5N1 (avian flu) virus, and respiratory syncytial virus infections ( Fig. 3.12 ) when cultures were not available.
The diagnosis of rabies relies heavily on histopathologic examination of tissues to demonstrate the characteristic cytoplasmic inclusions (Negri bodies). In an important percentage of cases, Negri bodies may be inconspicuous and so few that confirming the diagnosis of rabies may be extremely difficult. Furthermore, in nonendemic areas, the diagnosis of rabies is usually not suspected clinically or the patient can present with ascending paralysis. In these settings, immunohistochemical staining is a very sensitive, safe, and specific diagnostic tool for rabies ( Fig. 3.13 ). Other viral agents that can be diagnosed using immunohistochemical methods include enteroviruses, eastern equine encephalitis virus, and rotavirus.
Immunohistochemical staining has been used in the histopathologic diagnosis of viral hepatitis C; however, IHC for this virus is not superior to serologic assays and detection of hepatitis C virus (HCV) RNA in serum.
Bacterial Infections
Among bacterial infections, the greatest number of immunohistochemical studies have been performed in the investigation of Helicobacter pylori . A few studies have evaluated the use of IHC in other bacterial, mycobacterial, rickettsial, and spirochetal infections.
Antigen retrieval is generally not required for the immunohistochemical demonstration of bacteria in fixed tissue. However, interpretation of the results can be complicated by the fact that many of these antibodies cross-react with other bacteria. Moreover, antibodies may react with only portions of the bacteria, and they may label remnants of bacteria or spirochetes when viable organisms are no longer present.
Helicobacter pylori Infection
Gastric infection by H. pylori results in chronic active gastritis and is strongly associated with lymphoid hyperplasia, gastric lymphomas, and gastric adenocarcinoma. Heavy infections with numerous organisms are easily detected on routine hematoxylin and eosin–stained tissues; however, the detection rate is only 66% with many false-positive and false-negative results. Conventional histochemical methods such as silver stains are more sensitive than hematoxylin and eosin in detecting H. pylori . Nonetheless, for the detection of scant numbers of organisms, IHC has proved to be highly specific and sensitive, less expensive when all factors are considered, superior to conventional histochemical methods, and has low interobserver variation ( Fig. 3.14 ). Treatment for chronic active gastritis and H. pylori infection can change the shape of the microorganism, thus making difficult its identification and differentiation from extracellular debris or mucin globules. In these cases, IHC improves the rate of successful identification of the bacteria even when histologic examination and cultures are falsely negative. Helicobacter heilmannii belongs to the family Helicobacteraceae and is a less common causative agent of chronic gastritis found in a few gastric biopsies. The bacterium has been associated with mild chronic gastritis, peptic ulceration, and, rarely, gastric adenocarcinoma and MALT-lymphoma. Commercially available polyclonal antibodies against H. pylori cross-react with H. heilmannii antigens allowing for detection of this microorganism in gastric biopsies with a paucity of bacteria ( Fig. 3.15 ). Recently, polyclonal antibodies against Treponema pallidum have been described to immunostain gastric biopsies with H. heilmannii gastritis.
Whipple Disease
Whipple disease affects primarily the small bowel and mesenteric lymph nodes and less commonly other organs such as the heart and the central nervous system. Numerous foamy macrophages characterize the disease, and the diagnosis usually relies on the demonstration of periodic acid-Schiff (PAS)-positive intracytoplasmic bacteria. Nevertheless, the presence of PAS-positive macrophages is not pathognomonic; they can be observed in other diseases such as Mycobacterium avium infections, histoplasmosis, infections due to Rhodococcus equi, and macroglobulinemia. Tropheryma whipplei is a rare cause of endocarditis that shares many histologic features with other culture-negative endocarditides such as those caused by Coxiella burnetii and Bartonella sp. The development of specific antibodies against these microorganisms has significantly enhanced the ability to detect them in heart valves from patients with culture-negative endocarditis. Immunohistochemical staining with a rabbit polyclonal antibody provides a sensitive and specific method for the rapid diagnosis of intestinal and extraintestinal Whipple disease and for follow-up of treatment response.
Rocky Mountain Spotted Fever
Confirmation of Rocky Mountain spotted fever (RMSF) usually requires the use of serologic methods to detect antibodies to spotted fever group (SFG) rickettsiae; yet most patients with RMSF lack diagnostic titers during the first week of disease. IHC has been successfully used to detect SFG rickettsiae in formalin-fixed tissue sections, and it is superior to histochemical methods ( Fig. 3.16 ). Several studies have illustrated the value of IHC in the diagnosis of suspected cases of RMSF using skin biopsies with high specificity and sensitivity and in confirming fatal cases of seronegative RMSF. Owing to cross-reactivity, Rickettsia rickettsii cannot be distinguished from other SFG rickettsiae such as Rickettsia parkeri or Rickettsia conorii . Similarly, polyclonal antibodies against typhus group organisms, Rickettsia typhi and Rickettsia prowazekii , or monoclonal antibodies against typhus group lipopolysaccharide are useful to detect organisms in skin biopsy or postmortem tissues of patients with murine typhus or louse-borne epidemic typhus.
Bartonella Infections
Bartonella are slow-growing, fastidious gram-negative, Warthin-Starry-stained bacteria associated with bacillary angiomatosis, peliosis hepatis, cat-scratch disease, trench fever, relapsing bacteremia, and disseminated granulomatous lesions of liver and spleen. Bartonella are important agents of blood culture-negative endocarditis. Traditional techniques such as histology, electron microscopy, and serology have been used to identify the agents of culture-negative endocarditis. However, Bartonella sp., C. burnetii , and Tropheryma whipplei endocarditis share many morphologic features that do not allow for a specific histologic diagnosis. Besides, serologic tests for Bartonella sp. may show cross-reactivity with C. burnetii and Chlamydia sp. Immunostaining has been successfully used to identify Bartonella henselae and Bartonella quintana in heart valves from patients with blood culture-negative endocarditis and has significantly enhanced the ability to establish a specific diagnosis in these cases. This polyclonal rabbit antibody that does not allow differentiation between B. henselae and B. quintana has also been used in the detection of these microorganisms in cat-scratch disease ( Fig. 3.17 ), bacillary angiomatosis, and peliosis hepatis. A commercially available monoclonal antibody specific for B. henselae is also available and has been used to demonstrate the organism in a case of spontaneous splenic rupture caused by this bacterium.
Syphilis
Syphilis continues to be a public health problem caused by Treponema pallidum , a fastidious organism that has not been cultivated. The diagnosis of syphilis has relied on serology and the identification of T. pallidum by dark-field microscopy. However, these methods have low sensitivity and specificity, and serologic methods can be negative in early stages of the disease and in immunosuppressed patients such as those coinfected with HIV. In tissue sections, the usual method for detecting spirochetes is through silver impregnation stains (Warthin-Starry or Steiner). These stains, however, can technically be difficult to perform and interpret, are nonspecific, and frequently show marked background artifacts, because silver stains also highlight melanin granules and reticulin fibers. Detection rates of spirochetes using silver stains vary from 33% to 71%. Immunostaining of biopsy specimens with anti– T. pallidum polyclonal antibody ( Fig. 3.18 ) has been shown to be more sensitive and specific than silver stains with sensitivities ranging from 71% to 94%. As mentioned previously, anti– T. pallidum polyclonal antibody has been reported to cross-react with H. heilmannii and therefore cannot be used to differentiate syphilitic gastritis from H. heilmannii –associated chronic gastritis.
Mycobacterium tuberculosis Infection
Identification of M. tuberculosis is routinely achieved by acid-fast bacilli (AFB) stains and/or culture of biopsy specimens. Nevertheless, staining for AFB has low sensitivity and is not specific because it does not allow differentiation of mycobacterial species. Furthermore, cultures may take several weeks, and sensitivity is low in paucibacillary lesions. IHC with anti–bacillus Calmette-Guérin (BCG) polyclonal antibody has been used in the histologic diagnosis of mycobacterial infections, showing better sensitivity than AFB stains though it is not superior to AFB stains in paucibacillary lesions and does not allow for differentiation between M. tuberculosis and other mycobacteria. Recently, a polyclonal antibody against the M. tuberculosis –secreted antigen MPT64 was used in cases of mycobacterial lymphadenitis showing good sensitivity (90%) and specificity (83%) and performing better than AFB stains in cases of paucibacillary disease and comparably to nested PCR.
Other Bacterial Infections
Other bacterial diseases that can be identified by IHC in formalin-fixed tissue include leptospirosis, a zoonosis that usually presents as an acute febrile syndrome but occasionally can have unusual manifestations such as pulmonary hemorrhage with respiratory failure or abdominal pain. Rabbit polyclonal antibodies have been used in IHC to detect leptospiral antigens in the gallbladder and lungs from patients with unusual presentations ( Fig. 3.19 ).
Lyme disease has protean clinical manifestations, and Borrelia burgdorferi is difficult to culture from tissues and fluids. In addition, cultures are rarely positive before 2 to 4 weeks of incubation. B. burgdorferi can be identified in tissues by immunostaining with polyclonal or monoclonal antibodies. Although IHC is more specific than silver impregnation stains, the sensitivity of immunostaining is poor, and the microorganisms are difficult to detect due to the low numbers present in tissue sections.
Q fever is a zoonosis caused by C. burnetii characterized by protean and nonspecific manifestations. Acute Q fever can manifest as atypical pneumonia or granulomatous hepatitis, frequently with characteristic fibrin ring granulomas. This microorganism has been recognized as one of the agents causing blood culture-negative chronic endocarditis. A monoclonal antibody has been used to specifically identify C. burnetii in cardiac valves of patients with chronic Q fever endocarditis.
Recently, IHC has been successfully used to identify Streptococcus pneumoniae in formalin-fixed organs with an overall sensitivity of 100% and a specificity of 71% when compared with cultures. Immunohistochemical assays are useful in the identification of Clostridium sp., Staphylococcus aureus , and Streptococcus pyogenes . Haemophilus influenzae , Chlamydia species, Legionella pneumophila, Legionella dumoffii , Listeria monocytogenes , S almonella , and rickettsial infections other than RMSF such as boutonneuse fever, epidemic typhus, murine typhus, rickettsialpox, African tick bite fever, and scrub typhus.
Fungal Infections
The great majority of fungi are readily identified by hematoxylin and eosin staining alone or in combination with histochemical stains (PAS, Gomori methenamine silver [GMS]). However, these stains cannot distinguish morphologically similar fungi with potential differences in susceptibility to antimycotic drugs. In addition, fungal elements may appear atypical in tissue sections because of several factors including steric orientation, age of the fungal lesion, effects of antifungal chemotherapy, type of infected tissue, and host immune response. Currently, the final identification of fungi relies on culture techniques; however, culture may take several days or longer to yield a definitive result, and many times, surgical pathologists have no access to fresh tissue.
In past years, IHC has been used to identify various fungal elements in paraffin-embedded, formalin-fixed tissue. Immunohistochemical methods have the advantage of providing rapid and specific identification of several fungi and allowing pathologists to be able to identify unusual filamentous hyphal and yeast infections and accurately distinguish them from confounding artifacts. In addition, IHC allows pathologists to correlate microbiologic and histologic findings of fungal infections and to distinguish them from harmless colonization. IHC can also be helpful when more than one fungus is present; in these cases, dual immunostaining techniques can highlight the different fungal species present in the tissue. An important limitation of IHC in the identification of fungi is the well-known, widespread occurrence of common antigens among pathogenic fungi that frequently results in cross-reactivity with polyclonal antibodies and even with some monoclonal antibodies. Therefore, assessment of cross-reactivity using a panel of fungi is a very important step in the evaluation of immunohistochemical methods.
Candida species are often stained weakly with hematoxylin and eosin, and sometimes the yeast form may be difficult to differentiate from Histoplasma capsulatum, Cryptococcus neoformans , and even Pneumocystis jirovecii. Polyclonal and monoclonal antibodies against Candida genus antigens are sensitive and strongly reactive and do not show cross-reactivity with other fungi tested. In particular, two monoclonal antibodies against Candida albicans mannoproteins show high sensitivity and specificity. Monoclonal antibody 3H8 recognizes primarily filamentous forms of C. albicans , whereas monoclonal antibody 1B12 highlights yeast forms.
Identification of C. neoformans usually is not a problem when the fungus produces a mucicarmine-positive capsule. However, infections by capsule-negative strains are more difficult to diagnose, and the disease can be confused with histoplasmosis, blastomycosis, or torulopsis. Also, in longstanding infections, the yeast often appear atypical and fragmented. Polyclonal antibodies raised against C. neoformans yeast cells are sensitive and specific. More recently, monoclonal antibodies have been produced that allow identification and differentiation of varieties of C. neoformans in formalin-fixed tissue. The antibodies are highly sensitive (97%) and specific (100%) to differentiate C. neoformans var. neoformans from C. neoformans var. gattii .
Sporothrix schenckii may be confused in tissue sections with Blastomyces dermatitidis and fungal agents of phaeohyphomycosis. In addition, yeast cells of S. schenckii may be sparsely present in tissues. Antibodies against yeast cells of S. schenckii are sensitive but demonstrate cross-reactivity with Candida species; however, after specific adsorption of the antibody with Candida yeast cells, the cross-reactivity of the antibodies is eliminated.
Invasive aspergillosis is a frequent cause of fungal infection with high morbidity and mortality rates in immunocompromised patients. The diagnosis is often difficult and relies heavily on histologic identification of invasive septate hyphae and culture confirmation. Nevertheless, several filamentous fungi such as Fusarium species, Pseudallescheria boydii , and Scedosporium species share similar morphology with Aspergillus species in hematoxylin and eosin–stained tissues. A precise and rapid diagnosis of invasive aspergillosis is important because an earlier diagnosis is associated with improved clinical response, and it allows the correct duration and choice of antimycotic therapy to be planned. It has been shown that the yield of cultures in histologically proven cases is low, ranging from 25% to 50%. Several polyclonal and monoclonal antibodies against Aspergillus antigens have been tested in formalin-fixed tissues with variable sensitivities, and most of them cross-react with other fungi. More recently, monoclonal antibodies (WF-AF-1, 164G, and 611F) against Aspergillu s galactomannan have shown high sensitivity and specificity in identifying Aspergillus fumigatus, Aspergillus flavus, and Aspergillus niger in formalin-fixed tissues without cross-reactivity with other filamentous fungi. Mucormycosis is the third most common fungal infection after candidiasis and aspergillosis. It is clinically important to differentiate between invasive aspergillosis and mucormycosis because antifungal agents with anti- Aspergillus activity may not be active against Mucor infections. A commercially available mouse monoclonal antibody that recognizes Rhizopus arrhizus and other members of the family Mucoraceae reacts strongly with the cytoplasm of hyphae (clone WSSA-RA-1) and shows good sensitivity and specificity, although weak cross-reactivity with Aspergillus species has been described.
Cysts and trophozoites of P. jirovecii can be detected in bronchoalveolar lavage specimens using monoclonal antibodies that yield results that are slightly more sensitive than GMS, Giemsa or Papanicolaou staining ( Fig. 3.20 ). Antibodies are most helpful in cases of extrapulmonary pneumocystosis or in the diagnosis of P. jirovecii pneumonia when atypical pathologic features are present such as the presence of hyaline membranes or granulomatous pneumocystosis where the microorganisms are usually very sparse.
Penicillium marneffei can cause a disseminated infection in immunocompromised patients. Morphologically the organisms must be differentiated from H. capsulatum , C. neoformans , and C. albicans . The monoclonal antibody Ep-CAM (EBA-1) against the galactomannan of Aspergillus species cross-reacts with and detects P. marneffei in tissue sections. IHC has also been used to detect Blastomyces, Coccidioides, and Histoplasma . However, the antibodies have significant cross-reactivity with several other fungi.
Protozoal Infections
Protozoa usually can be identified in tissue sections stained with hematoxylin and eosin or Giemsa stain; however, because of the small size of the organisms and the subtle distinguishing features, an unequivocal diagnosis cannot always be made. The role of IHC in the detection of protozoal infections has been particularly valuable in cases in which the morphology of the parasite is distorted by tissue necrosis or autolysis. In addition, in immunocompromised patients, toxoplasmosis can have an unusual disseminated presentation with numerous tachyzoites without bradyzoites ( Fig. 3.21 ). IHC has also been useful in cases with unusual presentation of the disease.
The diagnosis of leishmaniasis in routine practice usually is not difficult; however, in certain circumstances, the pathologic diagnosis may be problematic, as is the case in chronic granulomatous leishmaniasis with small numbers of parasites, when the microorganism presents in unusual locations or when necrosis distorts the morphologic appearance of the disease. In these cases, immunohistochemical staining has been a valuable diagnostic tool. The highly sensitive and specific monoclonal antibody p19-11 recognizes different species of Leishmania and allows differentiation from morphologically similar microorganisms ( Toxoplasma , Trypanosoma cruzi , and P. marneffei ).
Immunohistochemical assays using polyclonal antibodies specific for Balamuthia mandrillaris, Naegleria fowleri , and Acanthamoeba sp. are useful to demonstrate amebic trophozoites and cysts in areas of necrosis and allows their differentiation from macrophages in cases of amebic meningoencephalitis.
IHC has also been used to identify Cryptosporidium , Entamoeba histolytica , T. cruzi , babesia, Giardia lamblia , Plasmodium falciparum, and Plasmodium vivax in fatal cases of malaria in formalin-fixed, paraffin-embedded tissue samples.
Emerging Infectious Diseases
In 1992, the Institute of Medicine defined emerging infectious diseases (EIDs) as caused by new, previously unidentified microorganisms or those whose incidence in humans has increased within the past 2 decades or threatens to increase in the near future. The list of pathogens newly recognized since 1973 is long and continues to increase, and recognizing emerging infections is a challenge, with many new infectious agents remaining undetected for years before emerging as identified public health problems. EIDs are a global phenomenon that requires a global response. The Centers for Disease Control and Prevention (CDC) has defined the strategy to prevent and detect EIDs. The anatomic pathology laboratory plays a critical role in the initial and rapid detection of EIDs. IHC, besides assisting in the identification of new infectious agents, has contributed to the understanding of the pathogenesis and epidemiology of EID.
Hantavirus Pulmonary Syndrome
In 1993 several previously healthy individuals died of rapidly progressive pulmonary edema, respiratory insufficiency, and shock in the southwestern United States. IHC was central in the identification of viral antigens of a previously unknown hantavirus. Immunohistochemical analysis was also important in detecting the occurrence of unrecognized cases of hantavirus pulmonary syndrome before 1993 and in showing the distribution of viral antigen in endothelial cells of the microcirculation, particularly in the lung ( Fig. 3.22 ).
West Nile Virus Encephalitis
West Nile virus (WNV) was originally identified in Africa in 1937, and the first cases of WNV encephalitis in United States were described in 1999. The clinical picture is variable and nonspecific, ranging from subclinical infection to flaccid paralysis and encephalitis characterized morphologically by perivascular mononuclear cell inflammatory infiltrates, neuronal necrosis, edema, and microglial nodules, particularly prominent in the brainstem, cerebellum, and spinal cord. The diagnosis of WNV encephalitis is usually established by identification of virus-specific IgM in cerebrospinal fluid (CSF) and/or serum and demonstration of viral RNA in serum, CSF, or other tissue. Immunostaining with either monoclonal or polyclonal antibodies has been successfully used to diagnose WNV infection in immunocompromised patients who lacked an adequate antibody response ( Fig. 3.23 ).
Enterovirus 71 Encephalomyelitis
Enterovirus 71 (EV71) has been associated with hand, foot, and mouth disease; herpangina; aseptic meningitis; and poliomyelitis-like flaccid paralysis. More recently EV71 has been associated with unusual cases of fulminant encephalitis, pulmonary edema and hemorrhage, and heart failure. Severe and extensive encephalomyelitis of the cerebral cortex, brainstem, and spinal cord has been described. Immunohistochemical staining with monoclonal antibody against EV71 has played a pivotal role in the linking of EV71 infection to fulminant encephalitis ( Fig. 3.24 ). Viral antigen is observed within neurons, neuronal processes, and mononuclear inflammatory cells.
Nipah Virus Infection
Nipah virus is a recently described paramyxovirus that causes an acute febrile encephalitic syndrome with high mortality rates. Pathology played a key role in identifying the causative agent. Histopathologic findings include vasculitis with thrombosis, microinfarctions, syncytial giant cells, and viral inclusions. Syncytial giant endothelial cells albeit characteristic of this disease are seen only in 25% of cases, and viral inclusions of similar morphology can be seen in other paramyxoviral infections. Immunostaining provides a useful tool for unequivocal diagnosis of the disease, demonstrating viral antigen within neurons and endothelial cells of most organs ( Fig. 3.25 ).