For reprint orders, please contact: [email protected]
Future Microbiology
Review
Cryptococcus gattii infections: contemporary aspects of epidemiology, clinical manifestations and management of infection Brendan Joseph McMullan1,2,3, Tania Christine Sorrell4,5 & Sharon Chih-Ann Chen*4,5 Department of Immunology & Infectious Diseases, Sydney Children’s Hospital, Randwick, New South Wales, Australia 2 School of Women’s & Children’s Health, University of New South Wales, Kensington, New South Wales, Australia 3 Westmead Clinical School, University of Sydney, Westmead, New South Wales, Australia 4 Centre for Infectious Diseases & Microbiology, Westmead Hospital, Westmead, New South Wales, Australia 5 Sydney Emerging Infections Biosecurity Institute, University of Sydney, New South Wales, Australia *Author for correspondence: Tel.: +61 2 9845 6255 n Fax: +61 2 9893 8659 n [email protected] 1
Cryptococcus gattii is an important primary and opportunistic pathogen, predominantly causing meningoencephalitis and pulmonary disease with substantial mortality. Initially considered geographically restricted to immunecompetent, highly exposed individuals in the tropics, an apparent epidemic in North America has led to new perspectives on its ecology, epidemiology and clinical associations, which are distinct from its sibling species Cryptococcus neoformans. The role of C. gattii molecular genotypes/subtypes in different settings is under investigation. Diagnostic and treatment strategies are similar to those for C. neoformans in immunocompetent hosts, although data indicate that more prolonged induction, as well as total duration of therapy, is required. Exclusion of CNS involvement is mandator y. Brain cr yptococcomas are characteristic of C. gattii infection, and raised intracranial pressure is common, for which surgery is often required. Immune reconstitution syndrome may occur. Ongoing C. gattii research and greater awareness and availability of specific diagnostic tests are required to improve patient outcomes.
Cryptococcal infections are responsible for approximately 1 million cases of meningo encephalitis annually, principally in HIVinfected individuals, with approximately 625,000 deaths [1]. The genus Cryptococcus con tains at least 70 distinct species [2]. However, two, Cryptococcus neoformans and Cryptococcus gattii, account for the majority of human dis ease. C. neoformans causes the most cases of cryptococcosis worldwide but C. gattii was, until recently, considered a geographically restricted rara avis of tropical and subtropical regions, and to be a variant of C. neoformans. It is now a sepa rate, distinct species, based on genetic and epide miological characteristics [3,4]. Its incursion into new environmental niches and ability to cause infection, including case clusters, in temper ate regions over the past 10–15 years in British Columbia, Canada and the Pacific Northwest (USA) [5–7] and Europe [8] have changed our appreciation of this pathogen. Although tradi tionally associated with a propensity to cause 10.2217/FMB.13.123 © 2013 Future Medicine Ltd
disease in immune-competent hosts, immuno compromised patients are also at risk of infec tion [7,9]. Evolution of disease associations has provided the impetus for contemporary clinical research into C. gattii. This review summarizes the recent ecology, clinical and molecular epidemiology, clinical fea tures and diagnostic and therapeutic aspects of C. gattii infections, highlighting novel develop ments and problematic management aspects rel evant to caring for patients with C. gattii crypto coccosis. Detailed discussion of the organism’s biology, and pathogenesis aspects, including the link thereof with ecological niches, are beyond the scope of this review. History
Cryptococcus was first reported in 1894 as a yeast in peach juice [10], but it was not until 1970 that Gatti and Eeckels described the organism as an unusual variant of C. neoformans causing men ingitis in a 7-year-old boy from the Congo [11]. Future Microbiol. (2013) 8(12), 1613–1631
Keywords antifungal drugs n brain and lung cryptococcomas n Cryptococcus gattii n epidemiology n immune reconstitution syndrome n raised intracranial pressure n
part of
ISSN 1746-0913
1613
Review
McMullan, Sorrell & Chen
Its original consideration as a variety of C. neoformans (C. neoformans var. gattii) was in part based on analysis of 628 clinical and 97 labora tory isolates, which demonstrated low prevalence of C. gattii in northern Europe, northern USA, Canada and Japan, compared with Australia, South America, southeast Asia and Africa [12]. Early reports of C. gattii infections stemmed mainly from Australia and Papua New Guinea (PNG) [13–18]. The emergence of C. gattii out breaks in North America in 1999 prompted reevaluation of these epidemiologic assumptions and investigation into C. gattii genetic diversity and recognition of molecular types (see ‘North America’ section). Ecology
Since the discovery of Eucalyptus camaldulensis and other Eucalyptus species by Ellis and Pfeiffer as putative ecological sources of C. gattii [13], the fungus has been recovered from >50 species of tree worldwide, including almond, avocado and cacti plants, soil, pine needles, vegetables and fruits (summarized in [19,20]). C. gattii has also been recovered from the environment of the coastal Douglas Fir and Western Hemlock on Vancouver Island (Canada) [21]. In Australia, seasonal flowering of eucalyptus has been linked with airborne dispersal of C. gattii spores [13], although Kidd et al. found no rela tionship with the flowering of any of the native trees of British Columbia [22]. Plants themselves are not susceptible to C. gattii infection, but are hypothesized to be an environmental reservoir. Environmental samples in the vicinity of human occupational and recreational activities further suggest a human role in C. gattii dispersal [23]. Taxonomy & serotypes
C. neoformans was considered a homogeneous single species until distinguishable serotypes based on capsular antigens were discovered [24]. These serotypes were traditionally character ized using rabbit antisera but monoclonal anti bodies against capsular antigens and molecu lar techniques are also available [25]. None of these methods are used in clinical practice (see ‘Microbiological diagnosis’ section for meth ods that are used). On the basis of serotypes, C. neoformans and C. gattii were divided into three varieties of a single species: C. neoformans var. neoformans and C. neoformans var. grubii (serotypes D and A, respectively) and C. neoformans var. gattii (serotypes B and C). Following further morphological and genetic studies [24], C. gattii was separated from C. neoformans, and 1614
Future Microbiol. (2013) 8(12)
the specific epithet or species name ‘Cryptococcus gattii’ agreed upon in 2002 [4]. However, this two-species paradigm is somewhat controversial in light of recent reports of intraspecies genetic diversity (see ‘C. gattii molecular types’ section). Nonetheless, the considerable genetic evolution ary distinction (time divergence of ~34 million years) between the two species justifies their separate species status [26]. C. gattii molecular types
There are at least four molecular types or geno types of C. gattii, VGI–VGIV, with a time divergence between molecular types of up to 12.5 million years [27,28]. Each genotype contains subtypes, as evidenced by a number of typing techniques, including multilocus sequence typ ing (MLST) analysis [29,30]. MLST, based on comparisons of sequences of multiple gene frag ments, provides reproducible and discriminatory strain differentiation [29], and is the current pre ferred genotyping method, since it is discrimi natory and standardized, allowing objective comparison of results. Under the auspices of the International Society of Human and Animal Mycology, a consensus set of seven genetic loci was selected as standards for MLST of C. gattii: six Cryptococcus housekeeping genes, CAP59, GPD1, LAC1, PLB1, SOD1 and URA5, and the ribosomal IGS1 gene [29]. The role of these genes in cryptococcal biology or pathogenesis, and the gene products they encode for, are summarized in Table 1. Details of primer design, amplification conditions and sequence analyses are given by Meyer et al. [31]. Other genotyping methods that have been proven to be discriminatory and have good clinical utility include PCR fingerprint ing, randomly amplified polymorphic DNA, PCR restriction fragment length polymorphism and amplified fragment length polymorphism (AFLP) [30]. In one large study, AFLP typing was used to investigate the C. gattii outbreak in Vancouver Island, where the vast majority of environmental and clinical isolates were AFLP6 (VGII) [21]. Genotype nomenclature has now been standardized [20]. Readers are referred to comprehensive reviews for further discussion of these other typing methods [30,31]. The fact that MLST has utility for geno typing and strain tracking of C. gattii is illus trated by worldwide studies characterizing the distribution and prevalence of genotypes with geographic region (summarized in [31]). A good example is the ability of the technique to iden tify the British Columbia case clusters as being predominantly caused by isolates of molecular future science group
Cryptococcus gattii infections
Review
Table 1. Recommended consensus multilocus sequence-typing loci† for molecular typing of Cryptococcus gattii isolates. Cryptococcal gene locus
Gene product
Role of gene product in cryptococcal virulence
CAP59
Capsular-associated protein
Protection against phagocytosis by host cells
GPD1
Glyceraldehyde-3-phosphate dehydrogenase
None
LAC1
Laccase
Melanin production, egress of cryptococci from lung
PLB1
Phospholipase
Tissue invasion
SOD1
Cu, Zn superoxide dismutase
Antioxidant function
URA5
Orotidine monophosphate pyrophosphorylase
None
IGS1
rRNA intergenic spacer
None
Consensus loci [29] are for typing for all cryptococci within the Cryptococcus neoformans–Cryptococcus gattii species complex. †
type VGII, subtypes VGIIa and VGIIb, while in neighboring Pacific Northwest, subtypes VGIIa and VGIIc were the most common [19,21,32,33]. The reasons for geographic differences in geno type distribution are unknown, but may relate to preferred ecologic niches for different genotypes and differences in growth conditions and viru lence characteristics. Although VGII genotypes exhibit different susceptibilities to azole anti fungals (see ‘Antifungal susceptibilities’ section) [34], the clinical relevance of different genotypes remains uncertain. Authoritative texts detail the global molecular epidemiology [20,31]. Molecular epidemiology
C. gattii comprises approximately 11% of C. neoformans species complex isolates globally and has a higher prevalence in Australia and PNG but also comprises significant disease burden in Asia, South America and, more recently, North America, Europe and Africa [8,20]. Understand ing the true global epidemiology is limited by the absence of systematic environmental sampling and lack of routine speciation of Cryptococcus isolates in many laboratories [33]. Oceania
In Australia and PNG, where infection is endemic, molecular type VGI is the most com mon C. gattii genotype; VGII genotype is much less common and its occurrence is geographically restricted [35,36]. The annual incidence of C. gattii in Austra lia is estimated to be 0.61/million people per year, with the majority of infections reported in healthy hosts, although, in a recent study, future science group
immunocompromised individuals comprised 25% of patients [9]. In PNG, the disease is reported in immunocompetent adults and chil dren, with an incidence of up to 42 cases/million people per year [19,37]. Molecular type VGIII strains have been reported in PNG and New Zealand [20]. South America
C. gattii is well-reported in Brazil, predomi nantly in HIV-uninfected patients and of the VGII molecular type [20]. In Colombia, isolates of genotypes VGI–VGIV have all been reported, while in Argentina, molecular type VGI pre dominates. Of note, infections in children due to C. gattii are particularly reported from the Brazilian eastern Amazon areas [38]. North America
Previously not considered endemic for C. gattii, during the emergence of infection in British Columbia, C. gattii cases occurred in humans as well as companion animals and wildlife on Van couver Island (Canada; reviewed in [19,21,39–41]). C. gattii was also identified in soil and environ mental samples [21,40]. The majority of isolates were of the molecular subtype VGIIa or VGIIb [21,40]. Infections in the Pacific Northwest fol lowed (>90 cases), predominantly caused by iso lates of subtypes VGIIa and VGIIc; the latter is a novel subtype not found outside the USA [33]. Epidemiological factors particular to the North American cases included a high proportion of respiratory infection in immunocompromised individuals (38–59%) [6,7]. The case clusters led to much speculation on the mechanism of www.futuremedicine.com
1615
Review
McMullan, Sorrell & Chen
introduction of this species. The high frequency of common genotypes suggested that this repre sented a clonal outbreak, with potential contrib uting factors, such as climate change (average: 0.8°C warming) [42] and anthropogenic activities [23]. Potential ‘parent’ genotypes of the North American molecular profiles have been found in Australia and Colombia but, as yet, with no clear lineage [43–45]. Europe
C. gattii is uncommonly reported in Europe, and mainly appears to have been acquired through travel to endemic regions, although apparently autochthonous cases are reported, and C. gattii has been isolated from Eucalyptus trees in Italy and Carob trees in Spain [8,46]. Asia
The epidemiology of C. gattii in Asia varies with locality. In Vietnam and China, infections are not uncommon in HIV-uninfected individuals (predominantly VGI genotype). In Thailand, C. gattii was reported in 12/18 (66%) isolates before the AIDS epidemic, but since then, has accounted for only six of 169 (4%) isolates [47]. In India, C. gattii molecular types VGI, VGII and VGIV are reported, although in one recent review, the majority were VGI. Contrary to find ings in Australia, Eucalyptus trees were not the predominant environmental niches [48]. VGI and VGII genotype isolates have been recovered in Malaysia, and VGII and VGIII isolates in Korea, although very uncommonly [20]. Africa
Although C. gattii infection in Africa is uncom mon, of note, isolates of genotype VGIV, rare elsewhere, are reported from Botswana, the Democratic Republic of Congo, Kenya, Malawi, Rwanda, Zimbabwe and South Africa in individuals with AIDS (reviewed in [19, 49–52]). Molecular type VGI strains have been noted in the Democratic Republic of Congo and those of type VGII in Senegal [43]. C. gattii environmen tal niches have included almond trees in Tunisia and Eucalyptus camaldulensis in Egypt [20]. Epidemiology & host risk groups
C. gattii is an important primary patho gen. In comparison with C. neoformans, data regarding clinical risk factors and manifes tations of C. gattii infections are relatively sparse, partly due to the fact that, until recently, many laboratories did not distinguish between C. gattii and C. neoformans [19]. In 1616
Future Microbiol. (2013) 8(12)
Australia, where C. gattii is endemic, the two species have been routinely distinguished using canavanine–glycine–bromothymol blue agar [53] since the early 1990s [Chen S, Pers. Comm.]. Until recently, C. gattii infections have been thought to primarily affect apparently healthy hosts [14,16–18]. However, in recent studies, only 62–72% of patients were previously healthy [9,41]. Furthermore, patient immunocompro mise was associated with increased mortality (29 vs 5%; p
Future Microbiology
Review
Cryptococcus gattii infections: contemporary aspects of epidemiology, clinical manifestations and management of infection Brendan Joseph McMullan1,2,3, Tania Christine Sorrell4,5 & Sharon Chih-Ann Chen*4,5 Department of Immunology & Infectious Diseases, Sydney Children’s Hospital, Randwick, New South Wales, Australia 2 School of Women’s & Children’s Health, University of New South Wales, Kensington, New South Wales, Australia 3 Westmead Clinical School, University of Sydney, Westmead, New South Wales, Australia 4 Centre for Infectious Diseases & Microbiology, Westmead Hospital, Westmead, New South Wales, Australia 5 Sydney Emerging Infections Biosecurity Institute, University of Sydney, New South Wales, Australia *Author for correspondence: Tel.: +61 2 9845 6255 n Fax: +61 2 9893 8659 n [email protected] 1
Cryptococcus gattii is an important primary and opportunistic pathogen, predominantly causing meningoencephalitis and pulmonary disease with substantial mortality. Initially considered geographically restricted to immunecompetent, highly exposed individuals in the tropics, an apparent epidemic in North America has led to new perspectives on its ecology, epidemiology and clinical associations, which are distinct from its sibling species Cryptococcus neoformans. The role of C. gattii molecular genotypes/subtypes in different settings is under investigation. Diagnostic and treatment strategies are similar to those for C. neoformans in immunocompetent hosts, although data indicate that more prolonged induction, as well as total duration of therapy, is required. Exclusion of CNS involvement is mandator y. Brain cr yptococcomas are characteristic of C. gattii infection, and raised intracranial pressure is common, for which surgery is often required. Immune reconstitution syndrome may occur. Ongoing C. gattii research and greater awareness and availability of specific diagnostic tests are required to improve patient outcomes.
Cryptococcal infections are responsible for approximately 1 million cases of meningo encephalitis annually, principally in HIVinfected individuals, with approximately 625,000 deaths [1]. The genus Cryptococcus con tains at least 70 distinct species [2]. However, two, Cryptococcus neoformans and Cryptococcus gattii, account for the majority of human dis ease. C. neoformans causes the most cases of cryptococcosis worldwide but C. gattii was, until recently, considered a geographically restricted rara avis of tropical and subtropical regions, and to be a variant of C. neoformans. It is now a sepa rate, distinct species, based on genetic and epide miological characteristics [3,4]. Its incursion into new environmental niches and ability to cause infection, including case clusters, in temper ate regions over the past 10–15 years in British Columbia, Canada and the Pacific Northwest (USA) [5–7] and Europe [8] have changed our appreciation of this pathogen. Although tradi tionally associated with a propensity to cause 10.2217/FMB.13.123 © 2013 Future Medicine Ltd
disease in immune-competent hosts, immuno compromised patients are also at risk of infec tion [7,9]. Evolution of disease associations has provided the impetus for contemporary clinical research into C. gattii. This review summarizes the recent ecology, clinical and molecular epidemiology, clinical fea tures and diagnostic and therapeutic aspects of C. gattii infections, highlighting novel develop ments and problematic management aspects rel evant to caring for patients with C. gattii crypto coccosis. Detailed discussion of the organism’s biology, and pathogenesis aspects, including the link thereof with ecological niches, are beyond the scope of this review. History
Cryptococcus was first reported in 1894 as a yeast in peach juice [10], but it was not until 1970 that Gatti and Eeckels described the organism as an unusual variant of C. neoformans causing men ingitis in a 7-year-old boy from the Congo [11]. Future Microbiol. (2013) 8(12), 1613–1631
Keywords antifungal drugs n brain and lung cryptococcomas n Cryptococcus gattii n epidemiology n immune reconstitution syndrome n raised intracranial pressure n
part of
ISSN 1746-0913
1613
Review
McMullan, Sorrell & Chen
Its original consideration as a variety of C. neoformans (C. neoformans var. gattii) was in part based on analysis of 628 clinical and 97 labora tory isolates, which demonstrated low prevalence of C. gattii in northern Europe, northern USA, Canada and Japan, compared with Australia, South America, southeast Asia and Africa [12]. Early reports of C. gattii infections stemmed mainly from Australia and Papua New Guinea (PNG) [13–18]. The emergence of C. gattii out breaks in North America in 1999 prompted reevaluation of these epidemiologic assumptions and investigation into C. gattii genetic diversity and recognition of molecular types (see ‘North America’ section). Ecology
Since the discovery of Eucalyptus camaldulensis and other Eucalyptus species by Ellis and Pfeiffer as putative ecological sources of C. gattii [13], the fungus has been recovered from >50 species of tree worldwide, including almond, avocado and cacti plants, soil, pine needles, vegetables and fruits (summarized in [19,20]). C. gattii has also been recovered from the environment of the coastal Douglas Fir and Western Hemlock on Vancouver Island (Canada) [21]. In Australia, seasonal flowering of eucalyptus has been linked with airborne dispersal of C. gattii spores [13], although Kidd et al. found no rela tionship with the flowering of any of the native trees of British Columbia [22]. Plants themselves are not susceptible to C. gattii infection, but are hypothesized to be an environmental reservoir. Environmental samples in the vicinity of human occupational and recreational activities further suggest a human role in C. gattii dispersal [23]. Taxonomy & serotypes
C. neoformans was considered a homogeneous single species until distinguishable serotypes based on capsular antigens were discovered [24]. These serotypes were traditionally character ized using rabbit antisera but monoclonal anti bodies against capsular antigens and molecu lar techniques are also available [25]. None of these methods are used in clinical practice (see ‘Microbiological diagnosis’ section for meth ods that are used). On the basis of serotypes, C. neoformans and C. gattii were divided into three varieties of a single species: C. neoformans var. neoformans and C. neoformans var. grubii (serotypes D and A, respectively) and C. neoformans var. gattii (serotypes B and C). Following further morphological and genetic studies [24], C. gattii was separated from C. neoformans, and 1614
Future Microbiol. (2013) 8(12)
the specific epithet or species name ‘Cryptococcus gattii’ agreed upon in 2002 [4]. However, this two-species paradigm is somewhat controversial in light of recent reports of intraspecies genetic diversity (see ‘C. gattii molecular types’ section). Nonetheless, the considerable genetic evolution ary distinction (time divergence of ~34 million years) between the two species justifies their separate species status [26]. C. gattii molecular types
There are at least four molecular types or geno types of C. gattii, VGI–VGIV, with a time divergence between molecular types of up to 12.5 million years [27,28]. Each genotype contains subtypes, as evidenced by a number of typing techniques, including multilocus sequence typ ing (MLST) analysis [29,30]. MLST, based on comparisons of sequences of multiple gene frag ments, provides reproducible and discriminatory strain differentiation [29], and is the current pre ferred genotyping method, since it is discrimi natory and standardized, allowing objective comparison of results. Under the auspices of the International Society of Human and Animal Mycology, a consensus set of seven genetic loci was selected as standards for MLST of C. gattii: six Cryptococcus housekeeping genes, CAP59, GPD1, LAC1, PLB1, SOD1 and URA5, and the ribosomal IGS1 gene [29]. The role of these genes in cryptococcal biology or pathogenesis, and the gene products they encode for, are summarized in Table 1. Details of primer design, amplification conditions and sequence analyses are given by Meyer et al. [31]. Other genotyping methods that have been proven to be discriminatory and have good clinical utility include PCR fingerprint ing, randomly amplified polymorphic DNA, PCR restriction fragment length polymorphism and amplified fragment length polymorphism (AFLP) [30]. In one large study, AFLP typing was used to investigate the C. gattii outbreak in Vancouver Island, where the vast majority of environmental and clinical isolates were AFLP6 (VGII) [21]. Genotype nomenclature has now been standardized [20]. Readers are referred to comprehensive reviews for further discussion of these other typing methods [30,31]. The fact that MLST has utility for geno typing and strain tracking of C. gattii is illus trated by worldwide studies characterizing the distribution and prevalence of genotypes with geographic region (summarized in [31]). A good example is the ability of the technique to iden tify the British Columbia case clusters as being predominantly caused by isolates of molecular future science group
Cryptococcus gattii infections
Review
Table 1. Recommended consensus multilocus sequence-typing loci† for molecular typing of Cryptococcus gattii isolates. Cryptococcal gene locus
Gene product
Role of gene product in cryptococcal virulence
CAP59
Capsular-associated protein
Protection against phagocytosis by host cells
GPD1
Glyceraldehyde-3-phosphate dehydrogenase
None
LAC1
Laccase
Melanin production, egress of cryptococci from lung
PLB1
Phospholipase
Tissue invasion
SOD1
Cu, Zn superoxide dismutase
Antioxidant function
URA5
Orotidine monophosphate pyrophosphorylase
None
IGS1
rRNA intergenic spacer
None
Consensus loci [29] are for typing for all cryptococci within the Cryptococcus neoformans–Cryptococcus gattii species complex. †
type VGII, subtypes VGIIa and VGIIb, while in neighboring Pacific Northwest, subtypes VGIIa and VGIIc were the most common [19,21,32,33]. The reasons for geographic differences in geno type distribution are unknown, but may relate to preferred ecologic niches for different genotypes and differences in growth conditions and viru lence characteristics. Although VGII genotypes exhibit different susceptibilities to azole anti fungals (see ‘Antifungal susceptibilities’ section) [34], the clinical relevance of different genotypes remains uncertain. Authoritative texts detail the global molecular epidemiology [20,31]. Molecular epidemiology
C. gattii comprises approximately 11% of C. neoformans species complex isolates globally and has a higher prevalence in Australia and PNG but also comprises significant disease burden in Asia, South America and, more recently, North America, Europe and Africa [8,20]. Understand ing the true global epidemiology is limited by the absence of systematic environmental sampling and lack of routine speciation of Cryptococcus isolates in many laboratories [33]. Oceania
In Australia and PNG, where infection is endemic, molecular type VGI is the most com mon C. gattii genotype; VGII genotype is much less common and its occurrence is geographically restricted [35,36]. The annual incidence of C. gattii in Austra lia is estimated to be 0.61/million people per year, with the majority of infections reported in healthy hosts, although, in a recent study, future science group
immunocompromised individuals comprised 25% of patients [9]. In PNG, the disease is reported in immunocompetent adults and chil dren, with an incidence of up to 42 cases/million people per year [19,37]. Molecular type VGIII strains have been reported in PNG and New Zealand [20]. South America
C. gattii is well-reported in Brazil, predomi nantly in HIV-uninfected patients and of the VGII molecular type [20]. In Colombia, isolates of genotypes VGI–VGIV have all been reported, while in Argentina, molecular type VGI pre dominates. Of note, infections in children due to C. gattii are particularly reported from the Brazilian eastern Amazon areas [38]. North America
Previously not considered endemic for C. gattii, during the emergence of infection in British Columbia, C. gattii cases occurred in humans as well as companion animals and wildlife on Van couver Island (Canada; reviewed in [19,21,39–41]). C. gattii was also identified in soil and environ mental samples [21,40]. The majority of isolates were of the molecular subtype VGIIa or VGIIb [21,40]. Infections in the Pacific Northwest fol lowed (>90 cases), predominantly caused by iso lates of subtypes VGIIa and VGIIc; the latter is a novel subtype not found outside the USA [33]. Epidemiological factors particular to the North American cases included a high proportion of respiratory infection in immunocompromised individuals (38–59%) [6,7]. The case clusters led to much speculation on the mechanism of www.futuremedicine.com
1615
Review
McMullan, Sorrell & Chen
introduction of this species. The high frequency of common genotypes suggested that this repre sented a clonal outbreak, with potential contrib uting factors, such as climate change (average: 0.8°C warming) [42] and anthropogenic activities [23]. Potential ‘parent’ genotypes of the North American molecular profiles have been found in Australia and Colombia but, as yet, with no clear lineage [43–45]. Europe
C. gattii is uncommonly reported in Europe, and mainly appears to have been acquired through travel to endemic regions, although apparently autochthonous cases are reported, and C. gattii has been isolated from Eucalyptus trees in Italy and Carob trees in Spain [8,46]. Asia
The epidemiology of C. gattii in Asia varies with locality. In Vietnam and China, infections are not uncommon in HIV-uninfected individuals (predominantly VGI genotype). In Thailand, C. gattii was reported in 12/18 (66%) isolates before the AIDS epidemic, but since then, has accounted for only six of 169 (4%) isolates [47]. In India, C. gattii molecular types VGI, VGII and VGIV are reported, although in one recent review, the majority were VGI. Contrary to find ings in Australia, Eucalyptus trees were not the predominant environmental niches [48]. VGI and VGII genotype isolates have been recovered in Malaysia, and VGII and VGIII isolates in Korea, although very uncommonly [20]. Africa
Although C. gattii infection in Africa is uncom mon, of note, isolates of genotype VGIV, rare elsewhere, are reported from Botswana, the Democratic Republic of Congo, Kenya, Malawi, Rwanda, Zimbabwe and South Africa in individuals with AIDS (reviewed in [19, 49–52]). Molecular type VGI strains have been noted in the Democratic Republic of Congo and those of type VGII in Senegal [43]. C. gattii environmen tal niches have included almond trees in Tunisia and Eucalyptus camaldulensis in Egypt [20]. Epidemiology & host risk groups
C. gattii is an important primary patho gen. In comparison with C. neoformans, data regarding clinical risk factors and manifes tations of C. gattii infections are relatively sparse, partly due to the fact that, until recently, many laboratories did not distinguish between C. gattii and C. neoformans [19]. In 1616
Future Microbiol. (2013) 8(12)
Australia, where C. gattii is endemic, the two species have been routinely distinguished using canavanine–glycine–bromothymol blue agar [53] since the early 1990s [Chen S, Pers. Comm.]. Until recently, C. gattii infections have been thought to primarily affect apparently healthy hosts [14,16–18]. However, in recent studies, only 62–72% of patients were previously healthy [9,41]. Furthermore, patient immunocompro mise was associated with increased mortality (29 vs 5%; p
