Diagnostic Microbiology of the Immunocompromised Host, Second Edition

Host defense from infection depends upon a complex, integrated system of physical barriers (e.g., skin, stomach acid, and mucociliary clearance), innate immunity (e.g., phagocytic cells, natural killer cells, complement), and adaptive immunity (B and T lymphocytes). An individual may have deficiencies of one or more components of host defense, but no individual is defenseless. Because each functional compartment of the immune system plays a specialized role in host defense, defects in specific functions lead to increased susceptibility to specific pathogens. The key to understanding the susceptibility of a particular patient is to understand the specific host defense defects of that patient. This article will briefly review the components of host defense and the types of infections that are most likely to occur with specific defects in those defense mechanisms. Following that will be descriptions of the infections that occur in patients with a variety of primary and secondary immune deficiency disorders, with the intent of providing illustrative examples.

The human immunodeficiency viruses type 1 (HIV-1) and 2 (HIV-2) are enveloped RNA viruses that are members of the genus Lentivirus and the family Retroviridae. HIV-1 is further divided into four groups; three of these originated from chimpanzee simian immunodeficiency virus and are designated group M (major)—responsible for the majority of the worldwide epidemic, group O (outlier), and group N (nonmajor and nonoutlier). Each group phylogenetically correlates with three separate transmission events ( 1 ). The fourth, group P, has been recently identified and is closely related to a gorilla simian immunodeficiency virus ( 2 ). Within group M there are nine different subtypes and more than 40 circulating recombinant forms (CRFs), which together lead to enormous viral genetic diversity. This variability plays an important role in the design and interpretation of viral load and resistance assays, and it provides clues to epidemiology and transmission of HIV-1. HIV-2 is a much smaller epidemic than HIV-1 and has its origins in simian immunodeficiency virus sooty mangabey (Cercocebus atys) with at least eight distinct transmissions to humans. Correct diagnosis of HIV-1 versus HIV-2 is critical as there are distinct differences in response to antiretroviral therapy.

Viral hepatitis contributes significantly to the morbidity and mortality of at-risk individuals, particularly in chronic, untreated infections. Clinical symptoms vary from none (asymptomatic carriers) to fulminant liver failure. Hepatitis viruses A, B, C, D, and E are responsible for the majority of liver disease, with hepatitis A and E viruses being the least common of the culprits.

Cytomegalovirus (CMV), the fifth member of the human herpesvirus family, is one of the largest viruses known to cause clinical disease. It is a double-stranded DNA virus that belongs to the beta-herpesvirus subfamily, along with human herpesviruses 6A, 6B, and 7. CMV was first associated with an infectious mononucleosis-like illness in healthy individuals in 1965 ( 1 ). Currently, it is known to cause a wide range of clinical syndromes, from asymptomatic infection in healthy hosts, to severe and even fatal disease in immunocompromised individuals, such as transplant recipients.

Epstein-Barr virus (EBV), a gamma herpesvirus, is a ubiquitous cause of infection in humans worldwide ( 1 ). Evidence of prior infection is present in adults throughout the world, with over 90% showing a serologic response. Exposure typically occurs early in life, with the majority of children in developing countries becoming seropositive by age 5. While onset of infection is delayed in areas with greater socioeconomic development, adults are almost uniformly positive. EBV is most commonly transmitted by contact with respiratory secretions, which promotes access and entry into the reticuloendothelial cells of the upper respiratory tree. While the primary target cell of EBV is the B lymphocyte, infection of a wider range of cell types can occur in immunocompromised hosts, particularly in those of epithelial lineage. Pharyngeal infection is followed by dissemination of virus throughout the body, with B lymphocytes as the primary target. The immune response to infection mounts steadily, with expansion of EBV-specific cytolytic T-cell clones eventually recognizing and controlling the primary infection. Control of EBV proliferation is signaled by a shift from lytic viral activity (marked by lytic proteins associated with cell destruction, such as BZLF1 and BRLF1) to a latent phenotype in an immortalized B lymphocyte pool, which provides a lifelong source of low-grade reactivation. Development of a serological response, with initial IgM and IgG to viral capsid antigen, followed by antibody to the EBV nuclear antigen developing months after infection, provides reliable markers for acute and chronic infection in immunocompetent hosts.

Herpes simplex viruses (HSV) are enveloped large DNA viruses (approximately 152,250 to 154,750 base pairs, depending on HSV type, and 90 transcriptional units). HSVs are α-herpesviruses (family Herpesviridae) and are divided into three major clades. There are two HSV types (HSV 1 and 2) that are genetically distinct but that are colinear and share roughly 83% genomic homology ( 1 ). HSV from mucocutaneous lesions, or from asymptomatic shedding in the oral or genital secretions from an infected contact, enters the skin/mucosa of a new host through minor breaks or abrasions to infect the underlying epithelium. Infection at the dermal–epidermal junction produces characteristic skin lesions. These begin as macules and papules that culminate in vesicles that contain infectious virus ( Fig. 1A ). These form pustules that rupture after 2 days; the resulting ulcers and crusts form within 96 hours. Vesicles and pustules are most likely to contain infectious virus, whereas ulcers and crusts reliably contain HSV DNA. Local vesicle formation is the hallmark of HSV infection, but vesicles may not be appreciated on mucosal surfaces because they rupture shortly after forming ( Fig. 1B ). An essential characteristic of α-herpesviruses during primary infection is entry into a permanent latent relationship with sensory neurons, with the potential for subsequent reactivation to cause recurrent local infections ( 2 ).

Human herpesviruses 6A, 6B, and 7 (HHV-6A, HHV-6B, HHV-7), are members of the family Herpesviridae, subfamily Betaherpesvirinae, and genus Roseolovirus. The three viruses are genetically close to each other and to cytomegalovirus (CMV), the type-species of human betaherpesviruses. Like other human herpesviruses, they are ubiquitous and establish a lifelong latent infection which originates further reactivations and reinfections. Early after the recent discovery of the three viruses, these properties, as well as the genetic relationship with CMV, have been a potent rationale for speculating on their potential pathogenicity in the immunocompromised population. The pathogenic role of HHV-6, this term collectively referring to HHV-6A and HHV-6B, is now ascertained in immunocompromised subjects whereas the clinical impact of HHV-7 infection appears less important in this domain. However, numerous questions are still pending, regarding diagnostic as well as therapeutic approaches targeted to these three viruses. Due to their strong similarities, the three viruses will be presented together throughout the chapter and specific details will be given for each of them within the different sections when needed.

Human papillomavirus (HPV), a member of the Papillomaviridae family, is a small (8kb), nonenveloped, double-stranded DNA virus. HPV has a predilection for infecting cutaneous and mucosal epithelial cells ( 1 ). Infection with HPV is associated with a wide range of pathology. HPV is the etiological factor in benign cutaneous warts and juvenile respiratory papillomatosis, as well as low-grade squamous intraepithelial lesions (LSIL), high-grade squamous intraepithelial lesions (HSIL), a precursor to cancer, and invasive carcinoma ( 2 ). Harald zur Hausen, a German virologist, was the first to describe the association of HPV and cervical cancer in the 1970s ( 3 , 4 ). It is now understood that HPV is necessary in the development of cervical cancer and is also associated with carcinoma of the vulva, vagina, penis, anus, and oropharynx ( 5 , 6 ).

The first polyomavirus was identified in suckling mice by Kilham and Murphy in 1953 ( 1 ) and named the murine pneumotropic virus (MptV). This discovery was ground-breaking because it was demonstrated that infusion of this virus, or of a second, similar murine virus—murine polyomavirus (MpV), discovered the next year—caused adenocarcinomas and leukemias to form when injected into mice ( 2 , 3 ). About 9 years later, two similar viruses, simian virus 40 (SV40) ( 4 , 5 ) and a baboon polyomavirus, SA12, ( 6 ) were identified. These four viruses were all found to be incapable of replication in humans but rather had species-specific replication. It was not until 18 years later, in 1971, that the first two human-associated polyomaviruses, BK and JC, were described ( 7 , 8 ). Both of these human-specific viruses were found to cause disease but were not oncogenic. There was then nearly a 20-year gap before the discovery of another new polyomavirus, GHPyV ( 9 ), isolated from geese in 2000. In contrast to this slow discovery rate using traditional viral methods, the discovery rate for polyomavirus strains dramatically increased in 2006 as a variety of newly available molecular methods for viral discovery became available. As of early 2015, more than 30 additional polyomaviruses have been identified from a variety of mammalian species. It is likely that additional strains are yet to be discovered ( 10 ). Species-specific polyomaviruses have now been identified in humans, apes, monkeys, mice, hamsters, bats, cows, sea lions, horses, raccoons, rabbits, and a variety of different bird species. Of the recently identified polyomaviruses, 11 have been found in human tissues or specimens, including KI ( 11 ), WU ( 12 ), Merkel cell MCPyV ( 13 ), trichodysplasia spinulosa TSPyV ( 14 ), MWPyV/HPyV10 ( 15 – 17 ), human polyomavirus (HpyV) 6 and HPyV7 ( 18 ), HPyV9 ( 19 ), HPyV12 ( 20 ), Saint Louis STLPyV ( 21 ), and New Jersey NJPyV ( 22 ) (see Table 1 ). The 13 human polyomaviruses can be grouped into subgroups based on whole genome sequence alignment ( Fig. 1 ). Human polyomaviruses have been identified in multiple locations throughout the body including blood, respiratory fluids, skin, liver, stool, and gastrointestinal (GI) tract tissues. Although BK, JC, TSPyV, and Merkel cell virus (MCV) have been implicated in human disease, the remaining more recently described human polyomaviruses require further study to elucidate their potential to cause clinical disease.

Adenoviruses are non-enveloped, double-stranded deoxyribonucleic acid (DNA) viruses associated with a wide range of clinical syndromes in humans ( 1 , 2 ). To date, 67 immunologically distinct serotypes of adenoviruses have been described and further classified into one of seven (A–G) species based on hemagglutinin properties, DNA homology, oncogenic potential in rodents, and clinical disease (see Table 1 )( 1 , 3 – 6 ).

Viruses account for more episodes of respiratory tract disease than any other microbial pathogens, and exact a significant toll on society and healthcare systems, leading to increased morbidity and mortality in both pediatric and adult populations throughout the world. Annually, almost 29 to 59 million people in the United States become infected with influenza virus; >200,000 are hospitalized and approximately 36,000 die ( 1 , 2 ). In addition, nearly 500 million people in the U.S. experience two or more noninfluenza-related viral respiratory tract infections each year ( 3 ). The economic burden of these yearly influenza outbreaks and noninfluenza-related viral respiratory infections approaches $87.1 billion and $40 billion, respectively ( 3 , 4 ).

Human enteroviruses (EVs) and parechoviruses (HPeV) belong to the family Picornaviridae ( 1 ). The human enteroviruses are classified into four species based on molecular and biologic characteristics: human enterovirus A (HEV A), HEV B, HEV C, and HEV D, with the traditional names retained for individual serotypes ( Table 1 ). Genetic characterization of two echovirus serotypes, 22 and 23, has resulted in their reclassification into a separate new genus, Parechovirus, and they are termed HPeV 1 and 2, respectively. Subsequent studies have demonstrated that there are now 16 distinct HPeV types, of which HPeV 1–6 are known to cause infections in humans.

In 1975, Yvonne Cossart first identified parvovirus particles by electron microscopy while evaluating tests for hepatitis B ( 1 ). This new virus, detected in serum number 19 in plate B, has since been known as parvovirus B19. The most common diseases subsequently associated with parvovirus B19 include fifth disease or erythema infectiosum (EI) of childhood, polyarthralgias/polyarthritis, fetal hydrops, transient aplastic crisis in individuals with chronic hemolytic anemias, and chronic red cell-aplasia in immunocompromised hosts.

Fungi are ubiquitous, eukaryotic organisms found throughout the environment. These microorganisms exist as either saprophytes or parasites, with the former obtaining nutrients from decaying organic matter and the latter sequestering nutrients from a living host. As such, fungi can be both plant and animal pathogens with significant agricultural and medical impact. A limited number of fungi are primary pathogens. These organisms have the ability to cause disease in both immunocompetent and immunosuppressed individuals. In contrast, the majority of filamentous fungi causing disease in humans are opportunistic and require specific host conditions be met before infection can occur. Immunocompromised patients are at highest risk for the development of invasive infection with opportunistic organisms. The focus of this chapter is on the medically important filamentous (i.e., molds) and dimorphic fungi known to cause disease in immunocompromised hosts.

Yeasts represent one of the growth forms of fungi. They are unicellular, and reproduce mostly by budding or less often by fission. Most medically important yeasts originate from the division of Ascomycota or from the division of Basidiomycota. A simple biochemical reaction, the urease test, can usually distinguish the yeasts from each of the divisions. Ascomycetous yeasts are usually urease negative, whereas basidiomycetous yeasts are urease positive.

Many Mycobacterium species are capable of causing infections in the immunocompromised individual, but Mycobacterium tuberculosis and Mycobacterium avium are among the most recognized pathogens in this genus. In broad terms, the mycobacteria are divided into the Mycobacterium tuberculosis complex (MTBC) and the nontuberculous mycobacteria (NTM). M. tuberculosis has long been recognized as a significant human pathogen, and much effort has been applied to the development of diagnostics for the detection of this organism. Historically, the NTMs garnered much less attention than MTBC, but they are now recognized to be important causes of human disease, especially in the immunocompromised host.

The group of Gram-positive bacillary organisms broadly known as “aerobic actinomycetes” consists of heterogeneous and taxonomically divergent genera belonging to the phylum Actinobacteria, in the class Actinobacteria, subclass Actinobacteridae. The majority of human pathogens in this group are placed in the suborder Corynebacterineae, with some in the suborders Streptosporangineae, Streptomycineae, and Micrococcineae ( 1 ). Many of the clinically significant genera are characterized by the presence of at least rudimentary and sometimes longer branching filaments with the ability to produce spores or to fragment the filaments ( 1 ). Except for the Mycobacterium tuberculosis complex, Mycobacterium leprae, and the genus Dermatophilus (the only clinically significant genus in the suborder Micrococcineae, family Dermatophilaceae), most genera in this group are considered saprophytes and few are associated with human pathogenesis ( 2 ). An extensive number of aerobic actinomycete genera have been described, but only a minority of these have been associated with human or veterinary health.

Parasites are an important cause of human disease worldwide. Malaria alone was responsible for 584,000 estimated deaths in 2013, while millions of others died from Chagas disease, African trypanosomiasis, strongyloidiasis, amebiasis, leishmaniasis, ascariasis and schistosomiasis ( 1 ). Eleven parasitic infections have been identified by the World Health Organization (WHO) as neglected tropical diseases because they threaten the health of millions of individuals and disproportionately impact impoverished individuals ( 2 ). The Centers for Disease Control and Prevention (CDC) has also identified five parasitic infections that have significant public health implications for individuals living in the United States, demonstrating that parasitic infections are not limited to the tropical regions of the world.

Aerobic Gram-positive and Gram-negative bacteria can be important pathogens in the immunocompromised host, especially in individuals who have compromised humoral immunity. Emphasis in this section will be placed on the most common bacteria that are associated with the immunocompromised host, although other aerobic bacteria will be discussed in certain sections for completeness. Aerobic bacteria are those that survive in the presence of oxygen, for example, in room air; they may grow in the laboratory under anaerobic conditions as well, but they prefer aerobic conditions. Some are obligate aerobes, such as Mycobacterium tuberculosis and Pseudomonas aeruginosa; others are facultative anaerobes, surviving with and without oxygen, such as Staphylococcus aureus and members of the Enterobacteriaceae (such as Escherichia coli and Salmonella spp.). Obligate anaerobes include Bacteroides fragilis and most Clostridium spp. Some anaerobes, such as Clostridium tertium, are termed aerotolerant anaerobes since their metabolism is anaerobic, but they can exist in the presence of oxygen. Lastly, there are microaerophilic bacteria, such as Campylobacter spp., which grow optimally in an environment in which the amount of oxygen is reduced but not eliminated.

A diverse community makes up the normal healthy microbiota in humans, and anaerobic bacteria are the primary component ( 1 – 4 ). A wide variety of different environments in the human body support complex microbial communities comprising both obligate and facultative anaerobes ( 3 , 4 ). At all body sites where anaerobes are part of the indigenous microbiota, obligate anaerobes greatly outnumber facultative anaerobes by a factor of 10 up to 1,000 times ( Table 1 ) ( 5 – 9 ). Obligate anaerobes are therefore the predominant type of bacteria present in humans at skin and mucosal surfaces. Microbial-community analysis of healthy human intestinal microbiota also reveals a rich and diverse array of anaerobes including Lactobacillus spp., members of the former Bacteroides fragilis group (i.e., B. fragilis, B. distasonis, B. thetaiotaomicron, B. ovatus, and B. vulgatus), other Bacteroides species, and Clostridium spp., as well as a wide array of anaerobes that are less clinically encountered ( 2 , 10 , 11 ). Under normal circumstances, intestinal anaerobes are not pathogenic but are essential for preventing overgrowth of opportunistic organisms or infection with pathogenic bacteria. Colonization resistance against acquisition of enteric pathogens (i.e., Salmonella and Shigella) and hospital-acquired antibiotic-resistant bacteria (e.g., vancomycin-resistant enterococci and Clostridium difficile) is provided by the presence of healthy normal bacteria that prevent gastrointestinal colonization by exogenous bacteria ( 1 , 10 , 12 – 18 ).

Decades of advances in cancer treatments and transplantation immunology have expanded the population of severely immunocompromised patients. In addition, new therapies for the management of rheumatologic, autoimmune, and acquired immune diseases have reduced mortality among these patient groups. Pulmonary infections are the most common syndromes contributing to morbidity and mortality among immunosuppressed patients ( 1 – 3 ). Virtually any potential pathogen can result in significant illness, and pulmonary infiltrates may be caused by a variety of noninfectious syndromes as well. Management of pulmonary syndromes in these vulnerable populations is a challenge for both clinicians and microbiologists, as prompt diagnosis can prevent irreversible pulmonary complications and/or allow withdrawal of potentially toxic empiric therapies. Diagnostic approaches should consider the tempo of the pulmonary process, the extent of immunosuppression, and the radiographic patterns. In addition, the likelihood of a specific infection may be affected by recently administered prophylaxis or empiric therapies.

Infections of the genitourinary tract are a common cause of morbidity in both the general population and in patients with compromised immune systems. Genitourinary tract infections can be broadly classified into a) urinary tract infections (UTI) including cystitis, pyelonephritis and prostatitis; and b) genital tract infections including urethritis, cervicitis, epididymitis, genital ulcerative diseases, endometritis, and pelvic inflammatory disease.

Gastrointestinal (GI) infections in the immunocompromised host include the common bacterial, viral, fungal, and parasitic agents that also cause infections in the immunocompetent host ( Table 1 ). Of special consideration is that immunocompromised patients may be at increased risk for, or experience, increased disease length and severity caused by many common GI pathogens. As the stomach is a major barrier to colonization by some enteric pathogens, it is not surprising that there may be changes in gastric microbiota in the immune-compromised patient that could be involved in increasing the risk of infection ( 1 ). Patients with human immunodeficiency virus (HIV) infection, recipients of solid-organ and bone-marrow transplant, patients with hematologic or other malignancies, or with diabetes, or patients receiving immunosuppressive chemotherapy and corticosteroid therapy, as well as patients with poor nutritional status, and/or at the extremes of age, are all at risk for conventional and opportunistic infections of the GI tract. Patients with primary immunodeficiencies may have GI lesions that are similar to other noninfectious diseases ( 2 ). Additionally, certain microorganisms cause infection in the compromised host but rarely, if ever, are observed in the normal, immunocompetent population ( Fig. 1 ). Some infections that may indicate an underlying immunodeficiency include Candida spp. esophagitis or microsporidial enteritis, which may be the first acquired immunodeficiency syndrome (AIDS)-defining illness experienced by patients with HIV infection.

Infections of the central nervous system (CNS) are associated with significant morbidity and mortality, with outcome often dependent on the rapidity with which a pathogen is identified and appropriate therapy is begun. Immunocompromised hosts are at risk for CNS infections with both common microorganisms, which may present atypically in this population, and with opportunistic pathogens. The microbiologic causes of infection vary by syndrome (e.g., meningitis versus encephalitis) and specific impairment in immunity (e.g., deficits in T-cell function versus hypogammaglobulinemia). This chapter provides a framework for the assessment and laboratory evaluation of immunocompromised patients presenting with CNS infections. Detailed discussions of the individual agents are provided in the corresponding pathogen-specific chapters.

Bloodstream infections (BSIs) and sepsis rank among the top reasons for human mortality for hospitalized patients. The full breadth of sepsis syndromes are particularly dangerous to any immunocompromised patient. Sepsis is one of the oldest of medical syndromes and dates back to the time in which Hippocrates characterized it as a clinical syndrome ( 1 ). Worldwide estimates of sepsis prevalence exceed 19 million cases per year, with over 750,000 in the United States ( 2 ). While only 2% of patients are admitted to the hospital with severe sepsis, they represent at least 10% of all ICU admissions in the U.S. ( 3 ).

Skin and soft tissue infections (SSTIs) encompass a variety of conditions, account for a large percentage of infections requiring hospitalization, and are associated with substantial morbidity ( 1 – 3 ). SSTIs are clinical entities of variable presentation, etiology, and severity that involve microbial invasion of the layers of the skin and underlying soft tissues, ranging from mild infections, such as impetigo or ecthyma, to serious, life-threatening infections, such as necrotizing fasciitis ( 2 – 5 ).

Prosthetic devices are frequently used in the management of patients with underlying immune deficiencies. Examples of prosthetic devices include a variety of joints, ventricular-assist devices, intravenous (IV) catheters, reservoirs for drug delivery, and postcancer reconstructive implants. The spectrum of underlying immunocompromised conditions spans the gamut of innate and acquired deficiencies of antibodies and complement, iatrogenic cellular-immune suppression to maintain transplanted stem cells and solid organs, neutropenia complicating chemotherapy, and biologic manipulation of interleukins and tumor-necrosis factor in numerous disease states. Dysfunction can occur in any and all arms of the immune system with sometimes predictable, but often surprising, infecting organisms and presentations. Prosthetic devices are especially prone to infection, given that skin provides a platform for bacterial colonization and immune evasion after violation of the skin by insertion of the device. Prosthetic-device infections can pose many clinical challenges, in particular the diagnosis and treatment of associated infections.

In their 2007 guideline for isolation precautions, the United States Health Care Infection Control Practices Advisory Committee (HICPAC) defines hospital-associated infections (HAIs) as infections acquired in any setting in which the delivery of healthcare occurs, including, but not limited to, hospitals (referred to as nosocomial infections) and nonacute-care settings, such as longterm care or ambulatory-care facilities and home care ( 1 ). This definition reflects the current trend of patients receiving care in a variety of environments, even for procedures, such as stem-cell transplants, which have traditionally taken place in an inpatient environment ( 2 , 3 ). In addition, an infection can be defined as an HAI if the infection was not present or incubating at admission, and if the patient was admitted for reasons other than the infection ( 4 ). For example, infections that develop either within 48 to 72 hours after admission or within 10 days of discharge from a healthcare facility are classified as HAIs ( 4 ).

Surgical pathology results can play a crucial role in the management of immunocompromised patients. A number of factors contribute to the complexity of histologic diagnosis in these patients. The most obvious difference is that the inflammatory response to pathogens differs from that of immunocompetent hosts. Pathologists must remember to perform special stains for microorganisms, even if there is little or no inflammation in a biopsy specimen. Noticing the lack of inflammatory cells, such as the lack of plasma cells in patients with common-variable immunodeficiency, in tissues where they are normally present can lead to a diagnosis of immunodeficiency.