Emergence of Cryptococcus gattii—Pacific Northwest, 2004–2010

  • July 23, 2010 / 59(28);865–868


Cryptococcus gattii has been increasing in the Pacific Northwest, where it has been seen in organ transplant recipients and normal hosts, in addition to other populations.

Cryptococcus is a genus of fungi, of which two species, Cryptococcus neoformans and Cryptococcus gattii, cause nearly all human and animal cryptococcal infections. Whereas C. neoformans primarily affects persons infected with human immunodeficiency virus (HIV) worldwide, C. gattii primarily affects HIV-uninfected persons in tropical and subtropical regions (1). In December 2004, a case of human C. gattii infection was reported in Oregon, associated with an outbreak on Vancouver Island and in mainland British Columbia, Canada (2). A second C. gattii case was reported in Oregon in 2005, and 12 more cases were reported in 2006 and 2007. In 2008, in response to the emergence of C. gattii in the United States, CDC, state and local public health authorities, and the British Columbia Centre for Disease Control (BCCDC) formed the Cryptococcus gattii Public Health Working Group (1). States began collecting epidemiologic information on patients and sending isolates to CDC. By July 2010, a total of 60 human cases had been reported to CDC from four states (California, Idaho, Oregon, and Washington) in the Pacific Northwest. Among 52 patients for whom travel history was known, 46 (88%) said they had not traveled to British Columbia or any other C. gattii–endemic areas, suggesting they acquired the infection locally. Among 45 patients with known outcomes, nine (20%) died because of C. gattii infection, and six (13%) died with C. gattii infection. Physicians should consider C. gattii as a possible etiology of a cryptococcal infection among persons living in or traveling to the Pacific Northwest or traveling to other C. gattii–endemic areas.

Multilocus sequence typing subcategorizes C. gattii into four genotypes: VGI, VGII, VGIII, and VGIV. Further genetic analysis divides the VGII genotype into three subtypes: VGIIa, VGIIb, and VGIIc (3). Although VGII is the genotype most commonly associated with the outbreak in the United States and British Columbia, it is uncommon in other C. gattii–endemic parts of the world, where VGI is isolated most frequently (3).

During 1999, C. gattii began appearing in animals and humans on Vancouver Island and, beginning in 2004, among mainland British Columbia residents who had no exposure to Vancouver Island (2,4). By the end of 2007, a total of 218 human cases had been reported to BCCDC (5). Studies from British Columbia and elsewhere showed a median incubation period of 6–7 months, with a range of 2–13 months (1). The median age of patients in British Columbia was 59 years, with age-specific incidence highest among persons aged 70–79 years (5). Only 38% of patients had an identifiable immunosuppressive condition (5). (Reported case-fatality rates either from or with C. gat-tii infection was 9% (5). Studies on Vancouver Island found C. gattii spores in the environment, often in association with trees and soil (6).

The two human infections reported from Oregon in 2004 and 2005 were from C. gattii subtypes VGIIa and VGIIc (3). The VGIIc subtype had not been found previously anywhere in the world; the VGIIa isolate was genetically distinct from the British Columbia VGIIa isolates (4). Neither patient had traveled to Vancouver Island or any other known C. gattii–endemic area. In early 2006, a resident of Orcas Island, Washington, developed C. gattii VGIIa infection with a strain indistinguishable from the British Columbia VGIIa strain (7).

In October 2009, the C. gattii Public Health Working Group formalized a surveillance system for C. gattii and housed it at CDC. The system includes standardized human and veterinary case report forms and isolate submission protocols. Standardized case report forms include questions about patient demographics, health history, and illness onset and course, and are completed by state or local health departments through interviews with patients or their family members. For the surveillance system, a case is defined as an illness occurring on or after January 1, 2004, in a U.S. resident or animal with a culture-confirmed isolate of C. gattii. Case report forms are completed after a patient isolate is confirmed as C. gattii at CDC. Chart reviews and patient or family member interviews for collection of patient information are carried out by state and local health departments to gather data on symptoms and illness course. For patients whose illness occurred before the surveillance system was initiated, isolates were sent to CDC or BCCDC at the time of illness; chart reviews and patient or family member interviews carried out at the time of illness were used to complete the case report forms.

During January 1, 2004–July 1, 2010, a total of 60 human cases of C. gattii infection were reported to CDC (Figure): 43 from Oregon, 15 from Washington, one from California, and one from Idaho. Approximately half (54%) of the patients were male (Table); patients ranged in age from 15 to 95 years, with the highest proportion of patients (45%) aged 50–69 years (Table). Among 47 patients for whom such information was known, 38 (81%) had an underlying condition that might have predisposed them to infection, including three patients with HIV infections. Of all patient isolates, 50% were subtype VGIIa, 32% were VGIIc, 10% were VGIIb, 5% were VGI, and 3% were VGIII. The most common clinical finding was pneumonia, occurring among 57% of patients.

Figure Figure.

Cases of Cryptococcus gattii infection* (n = 51) with known illness onset date,† by quarter—California, Idaho, Oregon, and Washington, 2004–2010.
Source: Cryptococcus gattii Public Health Working Group.
*Defined as illness occurring on or after January 1, 2004, in a U.S. resident with a culture-confirmed isolate of C. gattii.
†Includes estimated date for one patient each in 2007, 2008, and 2010, and two patients in 2009.
Alternate Text: The figure above shows cases of Cryptococcus gattii infection (n = 51) with known illness onset date, by quarter, in California, Idaho, Oregon, and Washington. During January 1, 2004–July 1, 2010, 60 human cases of C. gattii infection were reported to CDC: 43 from Oregon, 15 from Washington, one from California, and one from Idaho.

Table Table.  Microbiologic, demographic, and clinical characteristics of patients with Cryptococcus gattii infection*—California, Idaho, Oregon, and Washington, 2004–2010
  1. Abbreviation: HIV = human immunodeficiency virus.

  2. Source:Cryptococcus gattii Public Health Working Group.

  3. * Defined as illness occurring on or after January 1, 2004, in a U.S. resident with a culture-confirmed isolate of C. gattii.

  4. †Categories might not add to 100% because of rounding.

  5. § Frequencies are reported among patients for whom specific information was available.

  6. ¶ Includes single or bilateral kidney (three patients), bilateral lung (two), liver (one), and heart (one).

  7. ** Among patients without HIV or solid organ transplant, includes the following: recent history of oral steroid use (21 patients), non-HIV immunosup-pressive condition (16), lung disease (15), cancer (11), kidney disease (10), cardiovascular disease (nine), diabetes (eight), liver disease (five), rheumatic condition (five), and connective tissue disorder (three).

Sex (n = 52)
Age (yrs) (n = 51)
Race (n = 27)
 Weight loss16(40)
 Loss of appetite13(33)
 Chest pain12(28)
 Muscle pain11(26)
 Neck stiffness8(21)
 Night sweats7(17)
Clinical findings§
 Lung cryptococcomas13(34)
 Brain cryptococcomas3(25)
Predisposing conditions (n = 47)
 Had predisposing condition38(81)
 HIV infection3(8)
 Solid organ transplant¶8(21)
 No predisposing condition9(19)
Hospitalization status (n = 44)
Outcome (n = 45)
 Died from C. gattii9(20)
 Died with C. gattii, from another cause6(13)
 Survived, no response to treatment10(22)
 Survived, partial response to treatment5(11)
 Survived, recovered fully3(7)
 Survived, status unknown12(27)

Among the 45 patients for whom outcome was known, nine (20%) died because of C. gattii infection and six (13%) died with C. gattii infection; two of the nine who died from C. gattii infection had no predisposing condition. In addition to human cases, 52 veterinary cases (among cats, dogs, ferrets, sheep, camelids, elk, horses, goats, and porpoises) were reported to CDC from California, Hawaii, Oregon, and Washington.

Reported by:E DeBess, DVM, PR Cieslak, MD, Oregon Public Health Div. N Marsden-Haug, MPH, M Goldoft, MD, R Wohrle, DVM, C Free, E Dykstra, PhD, Washington State Dept of Health. RJ Nett, MD, Div of Vector Borne Infectious Diseases; T Chiller, MD, SR Lockhart, PhD, J Harris, PhD, Div of Foodborne, Bacterial, and Mycotic Diseases, National Center for Zoonotic, Vector-Borne, and Enteric Diseases, CDC.

Editorial Note

C. gattii is an emerging infection in the United States. C. gattii appears to differ from its sibling species, C. neoformans, both in its clinical aspects (e.g., less responsive to antifungal drugs and more likely to cause tumor-like lesions called cryptococcomas) and its ecologic niche (2,8). In addition, whereas the primary risk factor for C. neoformans cryptococcosis is severe immunosuppression (e.g., from HIV infection), risk factors for C. gattii infection in the United States appear to include both immunocompromise and exposure to specific regions of environmental fungal colonization (2,8). Many cases of C. gattii infection are likely not recognized because distinguishing between C. gattii and C. neoformans requires plating on differential media not routinely available in clinical microbiology laboratories; therefore, many cryptococcal infections are never speciated. In addition, cryptococcal infections generally are not notifiable diseases in the United States, although C. gattii is now reportable in one state, Washington, as a rare disease of public health importance.

Until 1999, most human C. gattii infections were reported from Australia and other tropical and subtropical regions, including parts of Africa, Asia, the Mediterranean, South America, and southern California (8). Fungal spores are known to colonize the nasal cavity and spread to other body sites, causing meningitis, pneumonia, and the development of lung, brain, or muscle cryptococcomas (8). The infection is not known to be transmitted among or within animal species. Although C. gattii had been isolated rarely from environmental sources and patients in the United States before 2004 (2), U.S. outbreaks had not been reported.

Because C. gattii typically has been regarded as tropical or subtropical in geographic distribution, its emergence in a temperate climate suggests that the pathogen might have adapted to a new climatic niche, or that climatic warming might have created an environment in which minimum threshold conditions for C. gattii spore survival and propagation are attained consistently (1,2). Alternatively, the environmental conditions supportive of C. gattii might be broader than previously suspected, or earlier propagation might have been inhibited by low concentrations of pathogen in the environment. In addition, infections might have occurred in the Pacific Northwest before the recognized increase in human cases, but too rarely to attract attention. However, retrospective speciation of 49 cryptococcal isolates from the Pacific Northwest obtained from 1997 through 2003 (7) and 31 isolates from Vancouver Island obtained from 1987 through 1998 (9) revealed exclusively C. neoformans, suggesting that the recent increase in reports of C. gattii represent actual emergence of the species in the region and not just an increase in disease awareness and reporting.

Additional systematic surveillance is needed to track C. gattii infection, along with increased awareness of the infection among public health practitioners, physicians, and veterinarians. In 2010, for the first time, surveillance data for C. gattii were reported at the Council of State and Territorial Epidemiologists meeting. The C. gattii Public Health Working Group is continuing disease surveillance and planning to conduct speciation of banked isolates of Cryptococcus. Improved surveillance should enable better assessment of the incidence of the disease and also its clinical manifestation and course.

Physicians should consider C. gattii as a possible etiology of infection when treating patients (particularly those who are HIV negative) who have signs and symptoms of cryptococcal infection, and should ask patients about recent travel to the Pacific Northwest, British Columbia, or other C. gattii–endemic areas. Physicians, particularly in the Pacific Northwest, should report suspected C. gattii infections and submit clinical isolates to their state health departments when requested.


The findings in this report are based, in part, on contributions by the Cryptococcus gattii Public Health Working Group, state laboratories, and the British Columbia Centre for Disease Control and Prevention.