Medical Mycology January 2011, 49, 49–55

Effects of season and weather on blastomycosis in dogs:

Northern Wisconsin, USA
DENNIS J. BAUMGARDNER∗, DANIEL P. PARETSKY†, ZACHARY J. BAESEMAN‡ & ANDREA SCHREIBER§
∗Department of Family Medicine, Aurora UW Medical Group, University of Wisconsin School of Medicine and Public Health,

Downloaded from https://academic.oup.com/mmy/article/49/1/49/1390643 by Afyon Kocatepe University user on 08 November 2023

Center for Urban Population Health, Milwaukee, †Eagle River Animal Hospital, Eagle River, Wisconsin ‡University of Wisconsin
School of Medicine and Public Health, Madison, and §University of Wisconsin School of Medicine and Public Health, Center for
Urban Population Health, Milwaukee, Wisconsin, USA

Dogs and humans acquire Blastomyces dermatitidis infections from the same incom-
pletely defined habitat. Studies of blastomycosis cases have not consistently demon-
strated seasonality or significant antecedent climate effects. To determine the distribution
of disease by season, we studied over 18 years 219 dogs with blastomycosis from a sin-
gle veterinary practice in Northern Wisconsin. The 202 Vilas County dog addresses were
compared to 200 random-number selected addresses from the practice registry. Street
addresses were geocoded and mapped using ArcGIS, including ratio of cases/random
addresses to construct a control chart. Stepwise and linear regression was used to model
case counts by season and by 6 month warm (April–September) and cold periods, using
lagged local weather data. The geographic distribution of cases was found to be similar
regardless of season and time period, and no season exceeded control chart limits. Seasonal
distribution of cases was; winter (n ⫽ 53, 24%), spring (39, 18%), summer (79, 36%),
fall (48, 22%), p ⫽ 0.002. When cases were considered by 6-month warm/cold periods,
67% of variation is explained by the total precipitation which occurred two periods prior,
and lower average temperature, but higher maximum temperature one period prior
(p ⫽ 0.000). Weather parameters along with fixed and variable environmental factors likely
determine the occurrence of B. dermatitidis, perhaps as part of a ‘grow and tolerate
change’ model.
Keywords Blastomyces dermatitidis, blastomycosis, fungal ecology, dog, mycoses,
weather

Introduction drought and soil disturbances have been associated with

case clusters [5,9–11]. B. dermatitidis is poorly competi-
Blastomyces dermatitidis, the etiologic fungus of the sys-
tive in the environment, however the ability of this fungus
temic and cutaneous mycosis, blastomycosis, is endemic
to survive potentially harsh conditions and rapid environ-
to portions of North America, India and Africa [1,2]. Dogs
mental change may enable it to temporarily ‘bloom.’ This
and humans acquire the fungus from similar environmental
attribute may be a key to the propagation and ultimate
exposures [3–5], and thus far, no small mammal reservoir
infection of man and animal by this fungus [8].
has been identified [6]. The ecologic niche of this organism
The growth and dispersal of other dimorphic systemic
remains incompletely defined [7,8]. B. dermatitidis toler-
fungi have been associated with weather and other geo-
ates a wide range of temperatures, and periods of relative
climatic changes [12]. Regression-based climate modeling
has had variable success predicting future coccidioidomy-
cosis case incidence, but climate appears to account for
Received 4 February 2010; Received in final revised form 6 April 2010; much of the variability in southern Arizona. Increased ante-
Accepted 22 April 2010 cedent precipitation, temperatures and drought, followed
Correspondence: Dennis J. Baumgardner, Aurora Sinai Medical Center
– Outpatient Health Center, 1020 N. 12th Street, Suite #4180, Milwau-
by wind or excavation dust dispersal, leads to Coccidioides
kee, WI 53233, USA. Tel: ⫹1 414 219 5191; fax: ⫹1 414 219 3116; E-mail: infection (the ‘grow and blow’ hypothesis) [13]. Furthermore,
dennis.baumgardner@fammed.wisc.edu regarding Paracoccidioides brasiliensis, a recent study
© 2011 ISHAM DOI: 10.3109/13693786.2010.488658
 50 Baumgardner et al..

postulates that soil water storage and El Niño activity in county is forested, most commonly with mature upland
the prior 3 years, and absolute air humidity the year before mixed forest (a mixture of needleleaf, especially Pinus
case incidence explains 49% of the yearly variance in an strobus, Pinus resinosa, and Abies balsamea, and broad-
area of Brazil [14]. Penicillium marneffei, an opportunistic leaf, especially Betula, Populus and Acer saccharum trees
dimorphic fungus of Southeast Asia, is associated with soil taller than 9.1 m) with the balance composed largely of
exposure during rainy seasons and with burrowing bamboo waterways, including approximately 3,710 km of com-
rats [15]. bined lake, river and stream shorelines. Cases of blasto-
The precise relationship of climatic factors, if any, to mycosis in dogs from this area have been associated with

Downloaded from https://academic.oup.com/mmy/article/49/1/49/1390643 by Afyon Kocatepe University user on 08 November 2023

B. dermatitidis has not been elucidated. Seasonality might close proximity to waterways, recent excavation, and
suggest certain environmental factors important to the sandy soils [18,28].
habitat of Blastomyces or factors that vary by season. Pre-
viously, no consistent seasonality has been associated with
blastomycosis [16–22]; however, predominant winter/fall
Dog cases and distribution
onset of localized pulmonary cases (which suggested
warmer weather exposure to the investigators due to incu- Cases included all dogs diagnosed with blastomycosis
bation period) has recently been reported from a portion of between 1 August 1990 and 31 July 2008 from a single
Canada typified by very cold winters [23]. We recently veterinary practice in Eagle River, which is located in
studied a comprehensive series of 151 human pulmonary southeast Vilas County. Excluded were dogs not residing
cases over 25 years from Vilas County, Wisconsin, USA in the Northern Wisconsin service area of this practice (in
[24, and additional unpublished data]. No seasonal trends and adjacent to Vilas County) to avoid bias or excessive
(by symptom onset) or relationships to local weather tem- distance from the local weather station by visitors, or dogs
perature, precipitation, dew point or wind data were brought from more remote areas due to reputation of the
noted. practice regarding this disease. A total of 219 street address
Dogs are considered harbingers of human blastomyco- locations of blastomycosis cases were collected with 202
sis [25] due to their involvement in mixed human and dog of these dogs residing in Vilas County. The epidemiologic
outbreaks [4,5,26], same-household cases over time [27], features and waterway and soil associations of earlier por-
and similar geographic distributions of human and dog tions of this case registry have been previously reported
cases in studied areas [17,18,28]. In highly endemic Northern [18,28]. All cases were diagnosed clinically by veterinar-
Wisconsin, USA, per-capita dog cases of blastomycosis ians experienced in the identification of blastomycosis in
occur at a rate 14-fold higher than that of humans [18]. this highly endemic area. Laboratory confirmation included;
This higher frequency of cases in a defined area may make (a) direct microscopy of throat/tracheal/bronchial washes
dogs more suitable for the study of associated climatic fea- or swabs in 50% of dogs, (b) direct microscopy of integu-
tures. We report the seasonal and geo-climatic features of ment, ocular, prostate or skeletal abscesses, lesions or drain-
a series of 219 dogs with blastomycosis over an 18-year age in 26%, (c) direct microscopy of internal organs at
period from a single veterinarian practice in Northern necropsy in 5%, (d) serology in 6%, (e) cytology in 2%
Wisconsin. and (f) urine antigen in 2% of the animals. Finally, 9% of
dogs were diagnosed only by typical clinical and radio-
logic findings. Date of diagnosis was used to assign cases
Methods to month or season, consistent with other large studies of
dogs [22,29].
Setting
Two hundred street addresses were obtained from the
A physical description of Vilas County, Wisconsin, USA 2001 total practice computerized database (a study mid-
has previously been reported [17]. In brief, it is located at point year) to determine the general distribution of practice
latitude 46° North and longitude 89–90° West, is 2,246 dogs in Vilas County, compared to blastomycosis case
square km in area, and has a population of 20,000 persons. addresses. The 2,020 county mailing addresses were num-
The average elevation is approximately 518 m. The aver- bered sequentially and a random number table [30] was used
age high temperature in Eagle River (the practice location) to select control households. No attempt was made to exclude
in July (warmest month) is 25.6°C and the average low is addresses associated with human or dog blastomycosis during
12.8°C. These average high and low temperatures for any time period. This portion of the analysis was restricted
January (coldest month) are –6.1°C and –17.8°C, respec- to the 202 Vilas County case dogs due to the existence of the
tively. The average annual precipitation is 75.2 cm. Snow 2001 database from a prior study [28] and to avoid map
covers the ground approximately 120 days per year, typi- density error due to small numbers of cases outside the
cally melting in April or May. Approximately 80% of the county.
© 2011 ISHAM, Medical Mycology, 49, 49–55
 Blastomycosis and weather 51

Geographic Information System analysis and weather data Results

Vilas County street addresses were geocoded with ArcGIS The individual/moving range control chart of the number
software (ESRI, Redlands, CA) using imported maps (GDT of cases of blastomycosis per consecutive season, based on
Dynamap, Lebanon, NH). Maps were generated on multi- date of diagnosis, is illustrated in Fig. 1. Case occurrences
ple scales appropriate for analysis, and the distance to near- remain within the upper control limits, without change in
est waterway was calculated for each case. Case distribution moving range over this time period, suggesting ‘normal’
maps were made for the separate time periods 1990–1995, variation (i.e., no apparent outbreaks).
1996–2002 and 2003–2008. Maps of the ratio of cases to Figure 2 illustrates the monthly distribution of cases, by

Downloaded from https://academic.oup.com/mmy/article/49/1/49/1390643 by Afyon Kocatepe University user on 08 November 2023

controls were generated for each combined season (National date of diagnosis, December–November. The distribution of
Oceanic and Atmospheric Administration definitions) of case numbers, by season, is significant overall (p ⫽ 0.002)
winter (December–February), spring (March–May), sum- and is as follows; winter (n ⫽ 53, 24%), spring (39, 18%),
mer (June–August) and fall (September–November). summer (79, 36%), fall (48, 22%), that is 140/219 (64%, p ⬍
Monthly weather data, including; (a) maximum and mini- 0.001) of cases occurred from May–October. The seasonal
mum temperatures, (b) mean maximum, minimum and aver- distribution of cases, if assigned month was adjusted one
age temperatures, and average dew point, (c) total precipitation, month previously as in Light et al. [23] (n ⫽ 41, 38, 83, 57,
and (d) average wind speed and maximum wind gust were respectively, winter–fall), did not differ significantly from the
obtained from a nearby local weather station (KEGV, Eagle distribution based on diagnosis date (p ⫽ 0.49).
River, WI). Missing data (41/222 recordings of monthly tem- The geographic distribution of the 202 Vilas County cases,
perature data, and 81/222 wind data points) were calculated compared to controls, was similar from season to season
by obtaining the equivalent reading from another Vilas County (Fig. 3). The geographic distribution of these cases by time
weather station (KARV, Arbor Vitae, WI), located 27 km west, periods 1990–1995, 1996–2002 and 2003–2008 did not
and adjusting the reading with the difference between such differ (data not shown).
readings from the two stations during three mid-point years Although the median distance of dog case residence to
(1999–2001). nearest waterway was not significantly different (p ⫽ 0.09)
over the four seasons (winter, 155 m; spring, 204 m; sum-
mer, 137 m; fall, 183 m), summer-diagnosed cases were
Statistical analysis closer to the nearest waterway than spring cases (p ⫽ 0.02).
MINITAB statistical software (State College, PA) was uti- Weather parameters varied significantly in this locale dur-
lized for data analysis. Categorical data was analyzed using ing the study period. The maximum temperature was 37°C,
chi-square tests. Mood’s median test was used to compare the minimum ⫺42°C. The monthly mean maximum tem-
distance to waterway data. To determine if the variation in perature ranged from 27°C to ⫺13°C, the mean minimum
the number of cases each consecutive season remained within from ⫺23°C to 16°C. Monthly total precipitation varied from
limits of anticipated variation, the time series was analyzed zero to 31.6 cm, and maximum wind gusts ranged from
by means of a data control chart. Values for number of cases 37–129 km per hour.
by season were plotted sequentially in the individual value When case numbers and monthly weather data were
portion of the chart and the mean obtained. A second (mov- analyzed by season (3 month blocks), the maximum tem-
ing range) chart was constructed of the differences between perature in the prior season was the single variable most
the successive seasonal values. The average of these values predictive of the resultant case numbers, and was inversely
(the average moving range) was multiplied by the constant proportional (p ⫽ 0.002; T score –3.28; R-squared, adjusted
3.27 to establish the upper control limit of the moving range ⫽ 12%). The best-fit model for seasonal data explains 32%
chart, in the usual fashion [31]. Multiplying the average mov- of the variation (p ⫽ 0.001) and involves four significant
ing range by the constant 2.66, and adding the product to the variables including an inverse relationship with maximum
mean value of the number of cases per season determined temperature in the two seasons prior to index season and
the upper control limit for the individual value chart. direct relationships with mean maximum temperature in
Consecutive cases by season, and also by 6 month ‘warm’ the two prior seasons, as well as mean average temperature
(April–September) and ‘cold’ (October–March) periods, were and total precipitation for the four prior seasons.
taken as outcome variables and subjected to stepwise regres- When warm vs. cold weather time periods (6 month
sion using weather parameters lagged by one to four time blocks) were the unit of analysis, total precipitation two
periods (seasonal data) or by one or two 6 month (warm vs. periods prior was the single variable most predictive of the
cold) periods as predictor variables (alpha ⫽ 0.15 to enter and resultant case numbers in a direct relationship (p ⫽ 0.003;
remove). Best-fit regression models were then constructed T score 3.31; R-squared, adjusted ⫽ 29%). Extending
and equations generated. case numbers through March 2009 (end of 2008 ‘cold
© 2011 ISHAM, Medical Mycology, 49, 49–55
 52 Baumgardner et al..

Downloaded from https://academic.oup.com/mmy/article/49/1/49/1390643 by Afyon Kocatepe University user on 08 November 2023

Fig. 1 Data control chart of blastomycosis cases in dogs, Northern Wisconsin, USA, by season of diagnosis, Years 1990–2008. Numbers of cases for
each season were plotted sequentially in the individual value (upper) chart and the mean obtained. The moving range (lower) chart was constructed of
the differences between the successive seasonal values, and the mean of these differences obtained (see text). Control limits were calculated using a
–
constant [31]. UCL ⫽ upper control limit; LCL ⫽ lower control limit; R⫽ average moving range

period’), the best-fit model explained 67% of the variation mapped by season. This suggests the presence, in these
(p ⬍ 0.001 for the model and the three variables) and has areas, of some relatively fixed environmental factors that
the following equation, i.e., number of new cases in 6 are important in the ecology of Blastomyces dermatitidis.
month period ⫽ 0.967 (mean maximum temperature one These presumably include two previously identified asso-
period prior) – 0.907 (mean average temperature one period ciations with Northern Wisconsin blastomycosis case
prior) ⫹ 0.554 (total precipitation two periods prior) – addresses, i.e., sandy soils and relatively low elevation
13.7. Figure 4 illustrates the predicted vs. actual cases waterways [28,29]. This study also reveals interesting
by 6 month blocks during this time period. associations with weather parameters prior to blastomyco-
sis diagnosis.
This investigation has several limitations. It is a retrospec-
Discussion tive study of moderate sample size in a single geographic
In this highly endemic area, the home-site geographic dis- location that may not be representative of other populations.
tribution and new case occurrences of blastomycosis in We were also limited to the weather parameters recorded
dogs have remained consistent over 18 years. The geo- by the local weather station. Detailed histories of the hab-
graphic distribution of these cases is also similar when its of these dogs were not obtained except as reported for
the first 3 years of the study [18]. It was not possible to
ascertain the precise onset dates of symptoms in these dogs
and the incubation period of Blastomyces dermatitidis
(sometimes regarded as 5–12 weeks) is variable and not
known with certainty [32]. There is no evidence to date
that time from exposure to diagnosis in dogs should vary
by season.
The seasonal distribution of our cases (summer/fall
predominant) differs somewhat from that of the recent
human series of Light et al. [23] (fall/winter predominant),
even when cases were adjusted one month previous to
onset. In addition, it differed from the uniform seasonal
distribution of our own human case series from the same
Fig. 2 Aggregate monthly distribution of dog blastomycosis cases, by area as the present report [24]. It is unclear whether our
month of diagnosis, December–November, years 1990–2008. results reflect a difference in geographic location in the
© 2011 ISHAM, Medical Mycology, 49, 49–55
 Blastomycosis and weather 53

Downloaded from https://academic.oup.com/mmy/article/49/1/49/1390643 by Afyon Kocatepe University user on 08 November 2023

Fig. 3 Distribution of Vilas County, Northern Wisconsin, USA, cases of blastomycosis in dogs, by season. Shading represents the ratio of cases to
controls by quintiles of cases (i.e., the lowest quintile [lightest shading] represents the geographic distribution of the 20% of cases with the lowest density
of cases, compared to controls, by map location). Q1 ⫽ first (lowest) quintile, Q2 ⫽ second quintile, etc.

former case, or differences in outdoor exposures and hab- case numbers. While this best-fit model only directly cor-
its, such as sniffing and digging by dogs, compared to relates with case occurrences, one might suggest that it
humans, in both cases. The timing of the upswing of dog indirectly correlates with growth and dispersal of this
cases in our study would correlate with spring thaw and environmentally acquired fungus. If this is the case, the
presumed increase in outdoor recreational open waterway latter two parameters of the model suggest positive effects
and other exposures. Indeed, summer-diagnosed cases of more extreme temperature rises about a relatively cool
resided closer to waterways. The later spring thaw in time period, consistent with the hypothesis that Blasto-
Canada may partially explain the delayed case peak of myces dermatitidis is favored by such an environmental
Light et al. [23], compared to our study. The May–October change [8].
predominance of cases in our study is similar to that of a This hypothesis fits with previous findings related to
previous Wisconsin study [29] using similar date of diag- blastomycosis and the etiologic fungus in this area
nosis monthly assignment, without the second peak in including sandy soils [28] and close proximity to water-
November and December that they observed. The major ways [18]. Sandy soils are more prone to daily tempera-
peak in a Louisiana, USA dog series [22] occurred in ture extremes due to changes in air temperature and
August–October, perhaps partly reflecting climate differ- sunlight. They also dry out quickly following rainfall and
ences from Wisconsin. are prone to wet-dry cycles. These daily temperature and
When warm vs cold weather 6-month time periods moisture changes would be more extreme where the for-
were the unit of analysis, increased moisture on average est canopy is open due to waterways. Thus, weather-
of 1 year prior to diagnosis, and lower average tempera- induced effects on soil would likely be most pronounced
ture, but higher maximum temperature on average of in habitats previously associated with B. dermatitidis in
6 months prior to diagnosis were associated with increased this region.

© 2011 ISHAM, Medical Mycology, 49, 49–55

 54 Baumgardner et al..

important for growth, and for natural conidia dispersal

[38]. In vitro, it has been shown to assimilate and/or toler-
ate a variety of substrates [39] and is unusually tolerant of
ammonia [40]. Soil ammonia concentrations increase as
moisture decreases, particularly in sandy soils with inter-
mittent nitrogenous deposits such as animal waste (nitrify-
ing organisms often do not operate at water potentials at
which ammonifiers such as Penicillium are still active

Downloaded from https://academic.oup.com/mmy/article/49/1/49/1390643 by Afyon Kocatepe University user on 08 November 2023

[34]). Were B. dermatitidis to propagate (perhaps follow-
ing repetitive wet-dry cycles) in a microenvironment con-
taining such substances, at a time of subsequent extremes
of temperature (and perhaps other undiscovered factors)
this fungus may temporarily ‘bloom’ due to its ability to
survive such change better than its nearby competitors. The
next cycle of this ‘grow and tolerate change’ model could
Fig. 4 Graph of actual vs. predicted cases, using best-fit regression then ensue. While our study did not associate wind with
model for cases of dog blastomycosis in Northern Wisconsin, October
blastomycosis cases, the terminal event prior to infection
1990–March 2009. Case numbers are for consecutive six-month warm
(April–September) and cold (October–March) time periods. Circles (0) by these dogs may well have been local soil disturbing
and solid lines ⫽ predicted case numbers. Plus signs (⫹) and dotted lines events such as digging and sniffing, excavation [5,18], or
(- - - -) ⫽ actual case numbers. local wind bursts not captured by this data.

A predominance of blastomycosis cases following snow

melt in this region is not inexplicable. Tundra soil micro- Acknowledgements
bial biomass in Colorado, USA has been shown to reach
This study was supported, in part, by a donation to the St
its (predominantly fungal) annual peak under snow [33].
Luke’s Foundation by Mr and Mrs Charles Goldsworthy,
Finally, although identification of a predictive model based
Eagle River, WI. The authors thank Karen Blanchard for
on preceding weather parameters may have no practical
manuscript preparation.
utility for blastomycosis from a public health perspective,
the association with climatic factors is most intriguing.
Certainly, case occurrences may not accurately reflect Declaration of interest: None.
population size of Blastomyces in the environment. None-
theless, our results suggest that weather ultimately effects
References
growth and sporulation of this fungus. Soil, its microor-
ganisms, flora and fauna, and their biochemical processes 1 Chapman SW. Blastomyces dermatitidis. In: Mandell GL, Bennett JE,
Dolin R, eds. Principles and Practice of Infectious Diseases, 6th edn.
and interactions are complex [34]. The occurrence of
Philadelphia: Elsevier, 2005: 3026–3040.
B. dermatitidis then may be a non-linear dynamical system 2 Al-Doory Y, DiSalvo AF (eds). Blastomycosis. New York: Plenum,
as described for other populations and infectious diseases 1992.
[35]. Such dynamics and cross-scale interactions, includ- 3 Furcolow ML, Busey JF, Menges RW, et al. Prevalence and incidence
ing climate and disturbances, have been proposed in other studies of human and canine blastomycosis. Am J Epidemiol 1970;
92: 121–131.
soil-related environmental systems [36]. Recently, Barrozo
4 Armstrong CW, Jenkins SR, Kaufman L, et al. Common-source out-
et al. [14] suggested that preceding (large-scale) El Niño break of blastomycosis in hunters and their dogs. J Infect Dis 1987;
activity is associated with yearly variance of paracoccid- 155: 568–570.
ioidomycosis cases. Perhaps large scale disturbances, 5 Baumgardner DJ, Burdick JS. An outbreak of human and canine
long-term trends (such as sustained drought [10]), varia- blastomycosis. Rev Infect Dis 1991; 13: 898–905.
6 Baumgardner DJ, Summerbell R, Krajden S, et al. Attempted isola-
tions in weather, and micro-ecological factors act dynami-
tion of Blastomyces dermatitidis from native shrews in northern
cally across multiple scales to effect the growth, dispersal Wisconsin, USA. Med Mycol 2005; 43: 413–416.
and ultimate infection by this environmentally-acquired 7 DiSalvo AF. The ecology of B. dermatitidis. In: Al-Doory Y, DiSalvo AF,
fungus. eds. Blastomycosis. New York: Plenum, 1992: 43–73.
A ‘grow and blow’ hypothesis has been proposed for 8 Restrepo A, Baumgardner DJ, Bagagli E, et al. Clues to the presence
of pathogenic fungi in certain environments. Med Mycol 2000;
Coccidioides wherein the fungus grows under moister con-
38(Suppl. 1): 67–77.
ditions, followed by spore formation under drier condi- 9 Kitchen MS, Reiber CD, Eastin GB. An urban epidemic of
tions, followed by spore dispersal by blowing dusts [37]. North American blastomycosis. Am Rev Respir Dis 1977; 115:
For B. dermatitidis, periods of increased moisture may be 1063–1066.

© 2011 ISHAM, Medical Mycology, 49, 49–55

 Blastomycosis and weather 55

10 Proctor ME, Klein BS, Jones JM, Davis JP. Cluster of pulmonary 24 Baumgardner DJ, Baeseman ZJ, Schreiber A. Geographic distribution
blastomycosis in a rural community: evidence for multiple high-risk of human blastomycosis by season in northern Wisconsin. Wis Med J
environmental foci following a sustained period of diminished pre- 2008; 107: 339–340.
cipitation. Mycopathologia 2002; 153: 113–120. 25 Sarosi GA, Eckman MR, Davies SF, Laskey WK. Canine blastomycosis
11 Blondin N, Baumgardner DJ, Moore GE, Glickman LT. Blastomycosis in as a harbinger of human disease. Ann Intern Med 1979; 91: 733–735.
indoor cats: suburban Chicago, Illinois, USA. Mycopathologia 2007; 26 MacDonald PDM, Langley RL, Gerkin SR, Torok MR, MacCormack
163: 59–66. JN. Human and canine pulmonary blastomycosis, North Carolina,
12 Baumgardner DJ. Commentary: environmental determinants of dimor- 2001–2002. Emerg Infect Dis 2006; 12: 1242–1244.
phic systemic mycoses – the macro and the micro. Int J Epidemiol 27 Baumgardner DJ, Paretsky DP. Blastomycosis: more evidence for
2009; 38: 1649–1650. exposure near one’s domicile. Wis Med J 2001; 100: 43–45.

Downloaded from https://academic.oup.com/mmy/article/49/1/49/1390643 by Afyon Kocatepe University user on 08 November 2023

13 Comrie AC, Glueck MF. Assessment of climate-coccidioidomycosis 28 Baumgardner DJ, Steber D, Glazier R, et al. Geographic information
model: model sensitivity for assessing climatologic effects on the system analysis of blastomycosis in northern Wisconsin, USA: water-
risk of acquiring coccidioidomycosis. Ann NY Acad Sci 2007; 1111: ways and soil. Med Mycol 2005; 43: 117–125.
83–95. 29 Archer JR, Trainer DO, Schell RF. Epidemiologic study of
14 Barrozo LV, Mendes RP, Marques SA, et al. Climate and acute/sub- canine blastomycosis in Wisconsin. J Am Vet Med Assoc 1987; 190:
acute paracoccidioidomycosis in a hyper-endemic area in Brazil. Int 1292–1295.
J Epidemiol 2009; 38: 1642–1649. 30 Brown BW Jr., Hollander M. Statistics: A Biomedical Introduction.
15 Vanittanakom N, Cooper CR Jr., Fisher MC, Sirisanthana T. New York: John Wiley & Sons, 1977: 415.
Penicillium marneffei infection and recent advances in the epide- 31 Wheeler DJ. Understanding Variation: The Key to Managing Chaos.
miology and molecular biology aspects. Clin Microbiol Rev 2006; 19: Knoxville, TN: SPC Press, 1993.
95–110. 32 Legendre AM. Blastomycosis in animals. In: Al-Doory Y, DiSalvo AF,
16 DiSalvo AF. The epidemiology of blastomycosis. In: Al-Doory Y, eds. Blastomycosis. New York: Plenum, 1992: 249–264.
DiSalvo AF, eds. Blastomycosis. New York: Plenum, 1992: 75–104. 33 Schadt CW, Martin AP, Lipson DA, Schmidt SK. Seasonal dynamics
17 Baumgardner DJ, Buggy BP, Mattson BJ, Burdick JS, Ludwig D. of previously unknown fungal lineages in tundra soils. Science 2003;
Epidemiology of blastomycosis in a region of high endemicity in 301: 1359–1361.
north central Wisconsin. Clin Infect Dis 1992; 15: 629–635. 34 Paul EA (ed). Soil Microbiology, Ecology and Biochemistry, 3rd edn.
18 Baumgardner DJ, Paretsky DP, Yopp AC. The epidemiology of blas- Oxford: Elsevier, 2007.
tomycosis in dogs: north central Wisconsin, USA. J Med Vet Mycol 35 Hastings A, Hom CL, Ellner S, Turchin P, Godfray HCJ. Chaos in
1995; 33: 171–176. ecology: is mother nature a strange attractor? Annu Rev Ecol Syst
19 Chapman SW, Lin AC, Hendricks KA, et al. Endemic blastomycosis 1993; 24: 1–33.
in Mississippi: epidemiological and clinic studies. Semin Respir Infect 36 Peters DPC, Havstad KM. Nonlinear dynamics in arid and semi-ar-
1997; 12: 219–228. id systems: interactions among drivers and processes across scales.
20 Vasquez JE, Meha JB, Agrawal R, Sarubbi FA. Blastomycosis in J Arid Environments 2006; 65: 196–206.
northeast Tennessee. Chest 1998; 114: 436–443. 37 Comrie AC, Glueck MF. Assessment of climate-coccidioidomycosis
21 Dwight PJ, Naus M, Sarsfield P, Limerick B. An outbreak of human model: model sensitivity for assessing climatologic effects on the risk of
blastomycosis: the epidemiology of blastomycosis in the Kenora acquiring coccidioidomycosis. Ann NY Acad Sci 2007; 1111: 83–95.
catchment region of Ontario, Canada. Can Commun Dis Rep 2000; 38 McDonough ES, Wisnewski TR, Penn LA, Chan DM, McNamara WJ.
26: 82–91. Preliminary studies on conidial liberation of Blastomyces dermatitidis
22 Arceneaux KA, Taboada J, Hosgood G. Blastomycosis in dogs: 115 and Histoplasma capsulatum. Sabouraudia 1976; 14: 199–204.
cases (1980–1995). J Am Vet Med Assoc 1998; 213: 658–664. 39 Baumgardner DJ, Laundre B. Studies on the molecular ecology of
23 Light RB, Kralt D, Embil JM, et al. Seasonal variations in the clinical Blastomyces dermatitidis. Mycopathologia 2000; 152: 51–58.
presentation of pulmonary and extrapulmonary blastomycosis. Med 40 Baumgardner DJ. Microecology of Blastomyces dermatitidis: the
Mycol 2008; 46: 835–841. ammonia hypothesis. Med Mycol 2009; 47: 745–752.

This paper was first published online on Early Online on 24 May

2010.

© 2011 ISHAM, Medical Mycology, 49, 49–55