Blastomycosis

Blastomycosis is considered endemic to the south-central, southeastern, and midwestern US states, particularly those bordering the Ohio and Mississippi Rivers and in parts of the United States and Canada surrounding the Great Lakes and the Saint Lawrence River (2,5). Hyperendemic foci exist in north-central Wisconsin and western Ontario. Areas of blastomycosis endemicity are based primarily on reports of symptomatic disease and was first described in the late 1930s (5). Although the epidemiology of blastomycosis is not as well understood as that of the other mycoses endemic to North America, the incidence seems to be increasing in some states, including Illinois, Indiana, and Wisconsin (2).

Cases of canine blastomycosis are often recognized as sentinels for human disease, presumably because dogs’ outdoor exposures are similar to, but potentially more extensive than, those of their human counterparts (6). A study in Illinois estimated blastomycosis incidence among dogs to be >8 times that among humans, and during 2001–2007, the annual incidence among dogs increased 50-fold (7). In addition to these apparent increases, blastomycosis has also been documented far outside the known disease-endemic areas. One report describes 2 cases probably acquired in the Pacific Northwest: 1 case of ocular blastomycosis in a golden retriever that had never left Washington state and 1 case of disseminated disease in a Rottweiler mix with no history of travel outside of British Columbia (Table 1) (8).

Similarly, several cases of blastomycosis in humans have also occurred far outside the traditionally defined disease-endemic area. A report from 1951 describes disseminated blastomycosis in an agricultural worker in Oregon (9). More recently, an immunosuppressed 6-year-old girl was thought to have contracted primary cutaneous blastomycosis in north-central Florida (10). Two unrelated cases of osseous blastomycosis of the knee without apparent pulmonary involvement were probably acquired in Nebraska (11): the first case was in a previously healthy man from western Nebraska who sustained a knee injury involving farm equipment ≈1 year before diagnosis, and the second case was in a man from central Nebraska with a 1-month history of knee pain but no acute knee injury (11). These cases highlight the difficulties associated with diagnosing blastomycosis in patients with no exposures to disease-endemic areas and whose disease manifestations are unusual (11). Similarly, 2 other cases in previously healthy adults occurred in Colorado (12); each patient had pulmonary disease and had been extensively exposed to soil while excavating prairie dog burrows on the eastern slope of the Rocky Mountains (12). De Groote et al. suggest that higher-than-normal rainfall in that region could have contributed to conditions favorable for Blastomyces growth and sporulation (12).

Moisture is believed to be an influential factor in the growth and dispersal of Blastomyces, although the precise ecology of the organism is not well understood, partly because of the difficulties associated with its recovery from the environment (2). Previously, blastomycosis was believed to be caused by 1 species, B. dermatitidis, but phylogenetic analysis indicates that B. dermatitidis is probably 2 species, B. dermatitidis and B. gilchristii (30). Furthermore, B. gilchristii is hypothesized to inhabit a specific ecologic niche in areas of hyperendemicity, whereas B. dermatitidis may be adapted to a wider range of environmental conditions and distribution throughout North America may be scattered (30). Additional research into the genetic, geographic, and clinical differences between these 2 species may contribute to a better understanding of blastomycosis epidemiology, both inside and outside the traditionally defined areas of endemicity.

Coccidioidomycosis

Coccidioidomycosis is caused by Coccidioides spp. and is endemic to the southwestern United States and parts of Mexico and Central and South America. An estimated 60% of Coccidioides infections are asymptomatic (3,31). The remaining 40% of infections most commonly cause an influenza-like illness also known as Valley fever, which is often self-limiting but can result in serious illness, particularly because some cases progress to severe pulmonary or disseminated disease. Prior Coccidioides infection usually provides immunity against reinfection, which can be assessed by use of a skin test antigen.

The first complete description of the geographic distribution of coccidioidomycosis was accomplished through large-scale evaluations of coccidioidin sensitivity prevalence. During the mid-1940s to early 1950s, coccidioidin skin tests were performed on ≈110,000 lifetime residents of a single county, most of whom were white men and women 17–21 years of age (32). The study identified Arizona, California, Nevada, New Mexico, Utah, and Texas as Coccicioides-endemic states; the highest rates of skin test positivity (50%–70%) were in California’s southern San Joaquin Valley and Arizona’s Sonoran Desert (32). Point-source coccidioidomycosis outbreaks near the borders of the areas identified by Edwards and Palmer further contributed to what is known about the geographic distribution of Coccidioides spp. (32). For example, in an outbreak that occurred in 1970, at least 61 archeology students were affected after excavating Native American ruins near Chico, Butte County, California, ≈70 miles north of the recognized disease-endemic area at the time (13). In 1972, at least 17 persons were infected during a similar excavation near Red Bluff, Tehama County, California, 20 miles north of the previous outbreak near Chico (14). In 2001, another outbreak occurred among 10 workers at an archeological site in Dinosaur National Monument in northeastern Utah, ≈200 miles north of the previously defined disease-endemic area (15). These outbreaks support the idea that foci of Coccidioides exist outside of the traditional areas of endemicity, yet they are not always represented on maps depicting these areas.

During 2010–2011, three unrelated coccidioidomycosis cases were identified in south-central Washington, far north of the area of known Coccidioides endemicity (16). Whole-genome sequencing of 1 clinical isolate from the patient and soil isolates recovered from the patient’s location of exposure revealed that the isolates were identical, providing direct evidence that the infection was acquired in Washington (33).

Evidence of coccidioidomycosis far outside the areas of known endemicity has also been seen in fossil records. Coccidioides spherules were morphologically identified in 2 fossilized, 8,500-year-old bison mandibles recovered from a flood plain in central Nebraska, suggesting that bison had migrated from disease-endemic areas or that Coccidioides previously inhabited a different or broader geographic range (34).

Histoplasmosis

Histoplasmosis-endemic areas were also established by using nationwide skin testing to evaluate histoplasmin sensitivity among ≈70,000 white persons, 17–21 years of age, who were lifetime residents of a single county (35). The highest proportion of positive reactors (60%–90%) occurred in states bordering the Ohio and Mississippi River valleys; a zone of moderate prevalence (30%–60%) extended outward around the central area for up to 300 miles (35). Histoplasma spp. seems to be less geographically restricted than Blastomyces or Coccidioides and is also endemic to parts of central America and various other locations worldwide, such as Africa and Asia. Because Histoplasma grows well in soil containing bird or bat droppings, the organism probably exists in microfoci outside of these broadly defined regions (4).

Histoplasma can infect many animal species and has been found in domestic and wild animals far outside the traditional disease-endemic areas, including disseminated disease in a dog with no history of travel outside of western Idaho (8) and in a northern sea otter found in Alaska (17). Otters do not migrate, so the authors of that report hypothesized that the infection was acquired from windborne spores or spores carried on the wings, feet, or beaks of migratory birds (17). Spores carried in seawater may be another possible explanation (36).

Histoplasmosis has also occurred in cats in the putatively non–disease-endemic states of Colorado, New Mexico, and California (18,19). Multilocus sequence typing of cat tissue samples indicated that the infecting strains of H. capsulatum were closely related to but clustered separately from the North American-1 clade (1 of 2 clades common to North America), suggesting that the genetic differences represent either geographic variation among H. capsulatum or differences in the strains capable of infecting animals and humans (18). Histoplasmosis in other cats near Vacaville, California (G. Thompson, unpub. data), and in two 6-month-old raccoons rescued near San Francisco (20) provides further evidence that Histoplasma may be established in California.

Among humans, histoplasmosis potentially acquired in California was first described in 1949; in a series of 5 cases of histoplasmosis in children in California, 1 patient was a girl who had never traveled >100 miles from San Francisco and another was a girl who had lived only in California and Wyoming, which is also not a traditional histoplasmosis-endemic area (21). Notably, these histoplasmosis diagnoses were based on chest radiograph findings and positive results for histoplasmin skin tests, which can cross-react with Coccidioides. A more recent case occurred in the Central Valley of California, in an immunocompetent 87-year-old man who seroconverted during a febrile illness and in whom Histoplasma endocarditis later developed (22). Another case provides further evidence of histoplasmosis far west of the traditional disease-endemic area; in an immunocompetent woman from Arizona with no history of travel to histoplasmosis-endemic areas, an initial gastrointestinal infection was later followed by development of an intramedullary spinal cord abscess (23). Another 6 unrelated cases were reported from Montana (5 cases) and Idaho (1 case); of these 6 patients, 5 had immunocompromising conditions and 3 experienced substantial diagnostic delays, probably because of the low index of suspicion for histoplasmosis in an unusual location (24).

Histoplasmosis has also been observed both south and north of the known disease-endemic areas. In a series of 7 cases of disseminated histoplasmosis in HIV/AIDS patients from south Florida, 5 patients had no relevant travel history (25). In northern Florida, acute pulmonary histoplasmosis developed in a college student after he had explored a bat-infested cave (26). New York is repeatedly described as a non–histoplasmosis-endemic area and is typically not represented on maps depicting histoplasmosis-endemic areas, probably because the skin test surveys by Manos et al. estimated <10% positive reactors for the entire state and <2% positive reactors in certain counties (35). However, cases reported from New York date back several decades, including an outbreak beginning in 1978 at a prison in rural central New York; this outbreak was suspected to have been related to removal of accumulations of bird droppings and trees that served as bird roosting sites (27). Evidence also exists of histoplasmosis acquisition in urban areas of New York, such as in a previously healthy child from Staten Island with extensive exposure to birds (28) and in a series of 5 cases of disseminated histoplasmosis in HIV/AIDS patients from the south Bronx who had no exposures to birds or bats (29).