INTRODUCTION — Cryptococcus gattii is a fungal pathogen that is endemic in the tropics and subtropics and temperate climatic regions of Australia. In the last three decades, it has become established in British Columbia, Canada, and the United States Pacific Northwest. C. gattii is genetically and biochemically distinct from Cryptococcus neoformans [1-3]. Together, these two species complexes, C. gattii and C. neoformans, account for most cases of cryptococcal infections in humans, although C. neoformans is more common. Like C. neoformans, infection with C. gattii manifests most often as meningoencephalitis and/or pneumonia.
The microbiology, epidemiology, risk factors, and pathogenesis of C. gattii infection will be reviewed here. The clinical features, diagnosis, and treatment of C. gattii infection are discussed separately. C. neoformans infection is also reviewed elsewhere. (See "Cryptococcus gattii infection: Clinical features and diagnosis" and "Cryptococcus gattii infection: Treatment" and "Microbiology and epidemiology of Cryptococcus neoformans infection" and "Epidemiology, clinical manifestations, and diagnosis of Cryptococcus neoformans meningoencephalitis in patients with HIV" and "Clinical manifestations and diagnosis of Cryptococcus neoformans meningoencephalitis in patients without HIV".)
MICROBIOLOGY — C. gattii is a basidiomycetous fungus, which can be found in the environment (see 'Environmental exposure' below). In clinical specimens, C. gattii is visualized as single or budding yeasts with round to cylindrical cells enveloped in a thick polysaccharide capsule. It was first proposed as a new taxonomic entity, Cryptococcus neoformans var gattii, in 1970 [4], in parallel with the characterization of the capsular antigens of C. neoformans. It is now characterized as a separate species, C. gattii.
Of the four major capsular serotypes of Cryptococcus spp (A, B, C, and D), B and C serotypes are exclusive to what has become known as C. gattii, a species distinct from C. neoformans [5,6]. Recognition of rare hybrids of C. gattii and C. neoformans provide evidence of taxonomic proximity, but not identity, between the two species [7,8].
Microbiologic tests used for the diagnosis of C. gattii infection are discussed separately. (See "Cryptococcus gattii infection: Clinical features and diagnosis", section on 'Culture and histopathology'.)
Molecular types — There are at least four molecular types or genotypes of C. gattii, VGI to VGIV, each containing subtypes, as determined by several genotyping techniques [8,9]. Multilocus sequence typing (MLST), based on sequences of multiple gene fragments, provides particularly reproducible and discriminatory strain differentiation and is helpful for tracking of strains in an outbreak [10-12]. Based on these molecular types and MLST and other genotyping studies, five species within C. gattii have been proposed [13]; however, the clinical relevance of these with respect to pathogenicity and epidemiology remains to be elucidated and hence we use the terminology "C. neoformans species complex" and "C. gattii species complex" for practical purposes, rather than creating more species [14].
Data from studies of molecular types of clinical (and some environmental) strains indicate that genotype distribution and frequency vary in different geographic regions (table 1) [3,15,16]. The reasons for this are unknown but may relate to preferred ecologic niches of different genotypes. (See 'Environmental exposure' below.)
The following geographic distribution of genotypes has been observed [3,16]:
●VGI is the most common genotype of isolates from Australia and Papua New Guinea, where C. gattii is endemic; VGII is much less common, and its occurrence appears to be geographically restricted to the "Top End" of the northern territory of Australia and to the southwestern region of the state of Western Australia [3,17,18].
●Isolates from the outbreak in British Columbia, Canada, and the United States Pacific Northwest are of the molecular type VGII, represented by three clonal lineages, VGIIa, VGIIb, and VGIIc [19,20]. VGIIa has been the most common genotype associated with this outbreak, followed by VGIIb; genotype VGIIc has been the least common genotype and has been isolated only from patients in the United States Pacific Northwest but not from patients in Canada [3].
●Other molecular types have been documented to cause disease in other regions of the United States, including VGI, non-outbreak VGII types, and VGIII.
●VGIV is rare outside Africa [21,22].
●In Asia and Mexico, there is a broad distribution of genotypes (table 1).
Whether the different genotypes affect the site, severity, or outcomes of C. gattii infection remains uncertain. There appear to be differences in fluconazole susceptibility according to molecular type. (See "Cryptococcus gattii infection: Treatment", section on 'Antifungal susceptibilities'.)
EPIDEMIOLOGY — The epidemiology of C. gattii infection is well described in areas endemic for the organism, such as Australia [1,23-25], and increasingly in other regions as well. However, many medical centers do not routinely identify cryptococcal isolates to the species level, instead reporting all Cryptococcus spp isolates as C. neoformans. Only after recognition of the North American outbreak (beginning in 1999) and the consequent increased interest in species distinction has the complexity of the epidemiology been appreciated.
Geographic distribution — Australia and Papua New Guinea have long been known to be sites of C. gattii endemic disease. Although C. gattii was formerly thought to be geographically restricted to tropical and subtropical regions, an outbreak in the Pacific Northwest of North America has changed our understanding of the epidemiology of C. gattii infection [3,16].
The emergence of clusters of cryptococcosis due to C. gattii in British Columbia, Canada, in 1999, with subsequent clusters documented in the United States Pacific Northwest, has highlighted the ability of this species to exploit new climatic environments and has expanded our knowledge of the epidemiology and clinical manifestations in the outbreak setting [26-33]. This outbreak was first detected on Vancouver Island in British Columbia, Canada, in 1999 [19,34-36]. Between 1999 and 2007, 218 individuals in British Columbia were found to have C. gattii infection [27], and cases continue to occur [37]; the rate of disease in British Columbia has been reported to be 0.3 to 0.5 cases per 100,000 population [37]. Since 2004, over 90 infections have been detected in the United States Pacific Northwest, particularly in Washington and Oregon [3].
Studies have demonstrated that C. gattii causes sporadic disease in multiple regions of North America, including not only the Pacific Northwest but also California, Idaho, Hawaii, and multiple states on the East Coast (eg, Georgia, Rhode Island, Florida) [31,38-43]. Sporadic cases have also been detected in temperate regions of Europe [44-46], as well as in Asia, Africa, Mexico, and South America [3,47,48]. Although the epidemiology is not well understood, it is becoming better defined as the relevance of species distinction has been appreciated. As noted above, the distribution of C. gattii genotypes varies by geographic region. (See 'Molecular types' above.)
Environmental exposure — Isolates of C. gattii have been found in the environment most commonly in tropical and subtropical areas, such as Australia, Hawaii, Brazil, Southeast Asia, and Central and sub-Saharan Africa [49,50], and in temperate areas such as the Pacific Northwest region of Canada and the United States [19,36].
Historically, C. gattii infection has been associated with exposure to certain trees. In particular, an environmental link with two species of Australian eucalypts (or red gum), Eucalyptus camaldulensis and Eucalyptus tereticornis, was first reported in the 1990s [51,52]. The concentration of E. camaldulensis along water courses and the association of rural-dwelling Aborigines with these trees are thought to explain the high prevalence of infection in this population, although host genetic factors in Aborigines and the non-Aboriginal population have not been studied. This link was subsequently confirmed by clinical and molecular epidemiologic studies [25,53] and the association of C. gattii infection with residence or work in a rural or semi-rural domicile has been maintained [54].
C. gattii has also been isolated from eucalypts in countries with small numbers of human cases, including some parts of the United States, Brazil, and Italy [45,52,55]. Cases of C. gattii also occur in regions where eucalypts are not found, such as Malaysia, South America, the Mediterranean basin, and the "Top End" of the northern territory of Australia [3,18,56,57]. In northern Brazil, isolates of C. gattii of the VGII genotype have been isolated from native forests and rivers [58].
In addition to eucalypts, trees native to Vancouver Island in British Columbia, Canada, have been implicated as an environmental niche for C. gattii; such trees include the Douglas fir, coastal western hemlock, alder, Garry oak, grand fir, and cedar [3,19]. In the greater Los Angeles area in the United States, C. gattii has been found in association with trees and soil debris with recovery from the Canary Island pine and American sweetgum, among other species [59]. In other areas of the United States, C. gattii has also been isolated from soil [3].
In contrast with the cases in Australia, of which a large proportion occurred in rural areas, cases from British Columbia, Canada, have occurred predominantly in urban and suburban communities [16]. Many patients in North America have had no apparent exposures other than residing or visiting areas of endemicity.
Hosts — Species-specific differences in the epidemiology of cryptococcosis are well defined and are largely, but not solely, due to differences in host immune status [24,25]. Unlike C. neoformans, which typically causes disease in patients with compromised cell-mediated immunity, most cases of C. gattii have been detected in persons with apparently normal immune systems [23-25,43]. However, it has been hypothesized that some patients with C. gattii infection may have subclinical defects in immunity [60]. In case series of C. gattii in endemic regions, including Australia, 72 to 100 percent of patients have been immunocompetent [24,54,61,62]. The majority of patients affected by the Pacific Northwest outbreak in North America have not had medical conditions or therapies classically associated with immunocompromise and have had pulmonary involvement [19,33,35]. (See "Cryptococcus gattii infection: Clinical features and diagnosis", section on 'Clinical features'.)
However, as C. gattii infection has been studied more intensely, a change in the understanding of the epidemiology has become apparent, with cases of C. gattii meningoencephalitis reported in patients with human immunodeficiency virus (HIV) infection, solid organ transplantation, and other causes of immunodeficiency, as illustrated by the following studies:
●Although C. gattii commonly affects people without apparent immunocompromise, up to 50 percent of patients in some studies [31,63], but not others [19,33,35], from the Pacific Northwest of North America have been immunocompromised. During the outbreak in British Columbia, C. gattii patients were more likely than the general population to be infected with HIV [63] and, in the United States Pacific Northwest case clusters, 4 of 59 patients (5 percent) were HIV positive [31]. It is important to note that only subsets of isolates were identified to the species level in most of these studies; thus, our understanding of the impact of C. gattii in HIV-infected individuals remains incomplete.
●Although population-based studies in Australia and the Gauteng Province of South Africa have reported that only 2 percent of cases of HIV-associated cryptococcal meningoencephalitis are ascribed to C. gattii [21,25,50], other studies have reported a higher prevalence, ranging from approximately 4 to 17 percent. As an example, the prevalence of C. gattii as a cause of cryptococcal disease was 12.3 percent in a retrospective study of HIV-infected patients in southern California [64]. In the outbreak in British Columbia, 4.5 percent of all patients had underlying HIV infection [33]. The highest prevalence of C. gattii, 16.7 percent, was reported in HIV-infected patients with cryptococcosis in Zimbabwe [22]. The prevalence may continue to change as laboratories are increasingly able to characterize cryptococcal isolates.
In addition to HIV infection, other immunosuppressive conditions that appear to increase the risk for C. gattii disease include organ transplantation, various malignancies, and receipt of glucocorticoids [31,40,54,63,65]. Idiopathic CD4+ lymphopenia in the absence of HIV infection, a known risk factor for cryptococcosis due to C. neoformans [66,67], has also been associated with C. gattii infection [54]. Although rare, donor-derived transmission of C. gattii through infected allografts has been described in kidney transplant recipients [40,68].
Other, more subtle host factors may also increase the risk for C. gattii infection. A proportion of patients with this infection in North America had a history of chronic lung disease. Changes in lung function associated with smoking may contribute to this risk since, in the British Columbia outbreak, more patients had underlying lung disease and were current smokers than was the case in the general population [63].
PATHOGENESIS — C. gattii infection is typically acquired by inhalation from the environment, although direct inoculation into the skin has also been reported [1,69]. Both yeast cells and basidiospores probably cause infection since, in nature, C. gattii reproduces both asexually (by budding) and sexually.
Studies of the pathogenesis of C. gattii infection are limited. In animal models, the outcome of infection is determined by a complex set of interacting pathogen and host factors; these include inoculum size, the cryptococcal strain, and the innate susceptibility or resistance (genetic background) of the host [70-72]. As an example, in studies of the VGII molecular type that caused the outbreak in British Columbia, Canada, the predominant strain (strain R265) was more virulent in mice than the less common strain (strain R272) [19,73]. Subsequent research has shown that the increased virulence of the outbreak strain may be mediated through pathogen-derived extracellular vesicles that affect cell-to-cell signaling [74].
Unlike C. neoformans, C. gattii infection typically causes infection in immunocompetent hosts and is more likely than C. neoformans to cause cryptococcomas of the brain and/or lungs (see "Cryptococcus gattii infection: Clinical features and diagnosis", section on 'Comparison of C. gattii and C. neoformans infection'). Although the reasons why cryptococcomas are larger and more common with C. gattii infection remain poorly understood, there are increasing data that the immune response to infection among immunocompetent hosts plays a role. In one study, the cytokine profile of peripheral blood mononuclear cells of healthy individuals was evaluated after in vitro stimulation with heat-killed Cryptococcus isolates [75]. Isolates of C. gattii induced higher concentrations of the proinflammatory cytokines, interleukin (IL)-1-beta, tumor necrosis factor-alpha, and IL-6 and the T helper 17/22 cytokines, IL-17 and IL-22, compared with C. neoformans. Toll-like receptor (TLR)-4 and TLR-9, but not TLR-2, also contributed to the host's cytokine response to C. gattii.
More recently, it was reported that anti-granulocyte-macrophage colony-stimulating factor (GM-CSF) autoantibodies were detected in plasma derived from immunocompetent patients with C. gattii infection but not in the plasma of patients infected with C. neoformans [76].
Virulence determinants — C. gattii has similar virulence characteristics as C. neoformans [8]. These include the polysaccharide capsule (important in evasion and suppression of the host immune response), the ability to grow at 37°C, production of melanin (via laccase activity), which functions as an oxidative stress protectant, the invasins, phospholipase B (Plb1) and urease, and the antioxidant, superoxide dismutase (SODp1) [77]. In animals with C. neoformans infection, Plb1 and laccase are essential for egress of cryptococci from the lung and dissemination to the central nervous system (CNS) [78,79], whereas Plb1 and urease are required for cryptococci to cross the blood-brain barrier [79,80].
Neurologic disease — To cause neurologic disease, cryptococci must traverse the lung, disseminate, and cross the blood-brain barrier. In C. neoformans models, cryptococci are transported in the blood in mononuclear phagocytes and as free cells [79,80]. There is also evidence for transendothelial cell transport of free cryptococci and paracellular transport within the phagolysosome of mononuclear phagocytes (the Trojan horse mechanism) [81]. Cryptococci can be expelled from macrophage phagolysosomes without loss of viability [82,83]. Cryptococci transported within mononuclear phagocytes in blood could therefore be released in cerebral microvessels and cross the blood-brain barrier as free cells. In human immunodeficiency virus (HIV)-associated cryptococcal meningoencephalitis, HIV infection of macrophages facilitates their translocation into the CNS [84].
Whether C. gattii differs from C. neoformans in its ability to cause CNS disease is unclear. Among immunocompetent hosts, isolated CNS disease with C. gattii, or CNS plus pulmonary disease, is more common in endemic areas such as Australia, whereas in the outbreak setting in British Columbia, Canada, pulmonary disease is more common [27].
In rats with C. gattii genotype VGII infection, VGIIa and VGIIb strains from the British Columbia, Canada, outbreak caused fatal infection confined to the lung, whereas VGIIa strains from Colombia and Australia caused dissemination to the CNS [85].
SUMMARY
●Cryptococcus gattii is an important fungal pathogen that is endemic in the tropics and subtropics, but is moved outside of these regions; it has caused an outbreak that is ongoing in British Columbia, Canada, and the United States Pacific Northwest. C. gattii is genetically and biochemically distinct from Cryptococcus neoformans. (See 'Introduction' above.)
●Information about the epidemiology and clinical syndromes caused by C. gattii in nonendemic regions is relatively limited due to the failure of many microbiology laboratories to distinguish between C. neoformans and C. gattii. (See 'Epidemiology' above.)
●There are at least four molecular types or genotypes of C. gattii, VGI to VGIV, each containing subtypes. Data from studies of molecular types of clinical and environmental strains indicate that genotype distribution and frequency vary in different geographic regions (table 1). The reasons for this are unknown but may relate to preferred ecologic niches for different genotypes. (See 'Molecular types' above.)
●Australia and Papua New Guinea have long been known to be sites of C. gattii endemic disease. Although C. gattii was formerly thought to be geographically restricted to tropical and subtropical regions, an outbreak that began in Vancouver Island, Canada, in 1999 and that has spread to the United States Pacific Northwest, and emergence of cases in Europe, has changed our understanding of the epidemiology of C. gattii infection. (See 'Geographic distribution' above.)
●The primary difference in host distribution between C. gattii and C. neoformans is that, at least in regions endemic for C. gattii, this species to a large extent causes infection in people with no apparent immunocompromise, although it has been hypothesized that some patients may have subclinical defects in immunity. C. gattii has also been detected in patients with human immunodeficiency virus (HIV) infection, solid organ transplantation, and other causes of immunodeficiency. (See 'Hosts' above.)
●The outcome of infection is determined by a complex set of interacting pathogen and host factors, including inoculum size, the cryptococcal strain, and the innate susceptibility or resistance of the host. (See 'Pathogenesis' above.)
ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges Kieren Marr, MD, who contributed to an earlier version of this topic review.