Go to full issue: June 2003: Volume 31, Number 4 (PDF)
Blastomycosis is a systemic mycosis caused by Blastomyces dermatitidis, a dimorphic fungus that exists as a mold in the environment and as a pathogenic yeast form in the body. Blastomycosis is endemic in the central and southeastern United States, particularly in the Mississippi and Ohio River valleys and the Great Lakes states. The reservoir is rich, moist soil; transmission occurs through inhalation of aerosolized conidia from contaminated soil. The median incubation period, based on limited outbreak information, is 45 days (range, 21 to 106 days). Most infections are asymptomatic or self-limiting. In clinical cases, acute pulmonary symptoms ranging from mild to fulminant are the most common manifestation; however, the infection may disseminate to the skin, bones, genitourinary system, and central nervous system (CNS). The case fatality rate is approximately five percent nationwide.
Epidemiology of Blastomycosis in Minnesota, 1999-2002
From 1999, when systematic surveillance for blastomycosis was instituted in Minnesota, through 2002, 140 laboratory-confirmed cases of blastomycosis were reported to the Minnesota Department of Health (MDH). A confirmed case was defined as a Minnesota resident with B. dermatitidis cultured or visualized from tissue or body fluid. The 140 reported cases include 18 cases that occurred during an outbreak of blastomycosis in Mountain Iron, Minnesota in 1999. The median annual number of cases from 1999 to 2002 was 30.5 (range, 28 to 49 cases) (Figure 1). The median annual incidence of blastomycosis statewide over the period was 0.6 cases per 100,000 population. St. Louis County residents accounted for 42 cases reported to MDH from 1999 to 2002, followed by Itasca (15), Hennepin (14), Ramsey (11), Cass (seven), Anoka (five), and Beltrami (five) Counties (Figure 2). The median annual incidence of blastomycosis in St. Louis County residents was 2.7 cases per 100,000 population (range, 1.5  to 14.0  per 100,000). The probable county of exposure to B. dermatitidis was identified for 89 cases; St. Louis, Itasca, Cass, and Beltrami Counties accounted for 66% of cases (Figure 3).
The median age of blastomycosis cases was 44 years (range, 4 to 83 years) (Figure 4). Overall, 67% of cases were male. However, only five (28%) of the 1999 outbreak-associated cases were male. If outbreak cases are excluded, the gender difference among sporadic cases from 1999 to 2002 is more remarkable; 88 (72%) of 122 sporadic cases were male. The distribution of race among the cases was as follows: white, 85%; Native American/Alaskan Native, 8%; Asian/ Pacific Islander, 4%; black, 3%, and other 1%. Seventy-four percent of cases were diagnosed during June through December, with a peak in September (Figure 5).
Of 115 cases with reported symptom information, 92 (80%) reported cough, including 17 (15%) with hemoptysis; 85 (74%) reported fatigue; 81 (70%) reported fever, and 29 (25%) reported skin sores. Sixty-eight (65%) of 104 cases reported weight loss; 61 (59%) reported night sweats and chills; 50 (48%) reported headache, and 50 (48%) reported chest pain. Eighty-eight (67%) of 131 cases were hospitalized for a median of 7 days (range, 1 to 56 days). There were 11 fatal cases from 1999 to 2002, resulting in a case fatality rate of 8%. Cases with an underlying chronic illness were more likely to be fatal (five of 22 [23%] vs. one of 82 [1%]; relative risk, 18.9; 95% confidence interval, 2.3- 153.3; p=0.001). The 22 cases with chronic illness included nine with diabetes, three with leukemia, and ten with other neoplastic, degenerative, or inflammatory disorders.
Seventy-nine (66%) of 119 blastomycosis cases with reported clinical information were limited to pulmonary infection, but in 30 (25%) cases the infection had disseminated to skin or soft tissues (68%), bones or joints (16%), the CNS (5%), the eye (3%), or other tissues (8%). Ten (8%) cases were categorized as extra-pulmonary only; most of these were localized soft tissue infections following a traumatic injury such as a puncture wound. The diagnosis was confirmed in 112 (90%) of 125 cases by culture of bronchoalveolar lavage fluid (42%), sputum (32%), wound exudate (16%,) or lung aspirate (9%). In 54 (43%) cases, the organism was identified in a smear of sputum (41%), bronchoalveolar lavage fluid (36%), lung aspirate (12%), or wound exudate (12%). Sixteen (13%) cases were confirmed by histopathology of lymph nodes, lung nodules, or trans-bronchial biopsies. Serology was positive in nine cases.
Eight (8%) of 102 blastomycosis cases from 1999 to 2002 had potential occupational exposures. Six cases were workers involved with excavation and outdoor construction, and two were forestry workers in endemic counties. Excavation has previously been identified as a risk factor for infection.1 From 1999 to 2002 in Minnesota, 36 (36%) sporadic blastomycosis cases reported excavation at or near their residence within 3 months of onset of illness. This was the most commonly reported potential risk factor, followed by woodcutting (29%), hiking (25%), fishing (18%), and owning or visiting a cabin (18%).
1999 Mountain Iron Outbreak
Prior to 1999, the only blastomycosis outbreak identified in Minnesota occurred in 1972 in Itasca County.2 In September 1999, a cluster of blastomycosis cases was reported from the town of Mountain Iron in St. Louis County. The ensuing investigation involved active surveillance for human and veterinary cases, interviewing and serological screening of town residents, a case-control study to identify risk factors for infection, soil cultures, and a meteorologic review.
A human case was defined as a Mountain Iron resident who had B. dermatitidis cultured or visualized from sputum or bronchial lavage fluid. Eighteen human cases were identified. All lived in a single neighborhood of approximately 200 households. The median age was 38 years (range, 7 to 70 years). Thirteen (72%) cases were female. Ten (56%) were hospitalized, for a range of 1 to 22 days. There were no fatalities. Cases were more likely to report other ill family members (odds ratio, 6.8; p<0.05) and lived closer to a recent new house excavation site than healthy neighborhood controls (p=0.05). Hunting, fishing, hiking on a neighborhood trail, gardening, or owning an ill dog were not associated with illness.
A canine case was defined as a dog from Mountain Iron from which B. dermatitidis was cultured or visualized from sputum, skin, or bronchial lavage fluid; a suspect case had either a chronic cough or non-resolving skin lesion. Nineteen confirmed and four suspect canine cases lived in the same neighborhood as human cases. In this outbreak, dogs were not useful as sentinels because their onsets of illness were generally concurrent with or after human case illness onsets (Figure 6).
Serologic testing of neighborhood volunteers by immunodiffusion and complement fixation did not identify additional cases. Only two of 11 culture-confirmed cases and none of the other 157 residents tested developed a detectable antibody response. These findings illustrated the inadequacy of available serologic tests for screening or diagnostic purposes. All of the human isolates of B. dermatitidis were the same genotype by random amplified polymorphic DNA polymerase chain reaction and DNA sequencing, but different from 18 historical or reference isolates. All soil samples tested from the neighborhood were negative for B. dermatitidis by selective culture and mouse assay. B. dermatitidis is rarely isolated from the environment.
In the months preceding the outbreak, Mountain Iron had above average precipitation (p=0.05), temperature (p<0.01), and dew point (p<0.01) than in the comparable timeframe in previous years. Weather conditions and recent disruption of soil at the excavation site probably contributed to this outbreak.
Diagnosis and Treatment A high index of suspicion is important for timely diagnosis of blastomycosis, because the presenting symptoms frequently mimic acute bacterial pneumonia. Radiographic findings are variable but may include single or multiple segmental or lobar infiltrates, single or multiple nodules, or larger masses. All currently available serodiagnostic tests lack sensitivity; a negative result is not helpful in ruling out blastomycosis. The easiest and quickest method of diagnosis is examination by light microscopy of sputum or wound exudate for large, broad-based budding yeast cells. Most patients with blastomycosis can be successfully treated with a 6-month course of itraconazole. Blastomycosis cases presenting with Acute Respiratory Distress Syndrome (ARDS) or CNS involvement require immediate and aggressive therapy with amphotericin B.3
There are no known practical measures for the prevention of blastomycosis. Minimizing morbidity and mortality from blastomycosis depends primarily on early recognition and appropriate treatment of the disease. Currently, MDH and the Minnesota Board of Animal Health are studying the epidemiology of canine blastomycosis cases in Minnesota to better define endemic areas in the state. Canine cases are more numerous than human cases, and in many cases, the probable location of exposure to the organism can be more easily identified (Figure 7). Veterinary cases of blastomycosis are also reportable; during 1997-2002, a median of 53 cases per year were reported (range, 36 to 98).
To report a human case of blastomycosis, or for further information on blastomycosis, call 651-201-5414 or 1-877-676-5414.
1. Baumgardner DJ, Burdick JS. An outbreak of human and canine blastomycosis. Rev Inf Dis 1991;13:898-905.
2. Tosh FE, Hammerman KJ, Weeks RJ, Sarosi GA. A common source epidemic of North American blastomycosis. Am Rev Respir Dis 1974;109:525-529.
3. Davies SF, Sarosi GA. Epidemiological and clinical features of pulmonary blastomycosis. Sem Resp Inf 1997;12:206-218.