The online literature databases PubMed and Google Scholar were searched for English-language articles about fungal infections related to natural disasters that were published as of April 2013. Search terms included combinations of “disaster,” “natural disaster,” “tornado,” “hurricane,” “earthquake,” “tsunami,” and “flood” with “fungal infection,” “fungal disease,” “fungus,” or “mold.” References cited in relevant articles were also reviewed, and the authors’ personal records were searched for conference abstracts.

Of the 1.5 million species of fungi on Earth, ≈300 are known human pathogens (20). Pathogenic fungi exist in a broad range of natural habitats but are believed to be more common in subtropical and tropical areas of the world, probably because of environmental restrictions on their growth or propagation (20). Known geographic habitats of some pathogenic fungi (for example, Blastomyces, Coccidioides, and Histoplasma) are well defined, but others (such as Aspergillus and other molds) are thought to be ubiquitous. The abundance and distribution (and therefore, potential to cause disease) of environmental fungi probably depend on climatic or environmental factors such as ambient temperature and moisture (20). Examples include coccidioidomycosis incidence in Arizona, which has been shown to correlate with hot, dry conditions; blastomycosis clusters observed in association with rainfall after periods of decreased precipitation; and incidence of Penicillum marneffei penicilliosis, an opportunistic infection which is endemic to Southeast Asia and increases in incidence during rainy months (21,22). Aspergillosis and other invasive mold infections have also been postulated to correlate with seasons or weather patterns in some geographic areas (23).

Although seasonal changes and weather effects probably play a role in the growth and distribution of many pathogenic fungi, environmental disruption is a key factor in the dispersal of these organisms and their resulting potential for causing infection. Both small-scale earth-disrupting activities, such as excavation or construction, and events affecting larger areas, such as earthquakes (16), tsunamis (5–7,10–15,24,25), and tornadoes (4), have been linked to the occurrence of fungal infections. Natural disasters can cause large-scale disruption of fungal habitats, which can lead to clusters of respiratory, cutaneous, or other forms of fungal disease.

Studies of clinical specimens collected from persons injured during disasters highlight the diversity of potential fungal pathogens in the natural environment. For example, after an 8.0-magnitude earthquake in Wenchuan, China, during 2008, 19 strains of fungi were identified in wound, sputum, and blood cultures from 123 injured persons (26), and fungi accounted for 7.6% of clinical isolates obtained from 42 patients with crush syndrome (27). Similarly, after a 1970 tornado in Texas, United States, fungi were identified in 8 (6.5%) of 124 wound isolates from 24 hospitalized patients (Fusarium, unspecified yeast, Rhodotorula, Aspergillus, Hormodendrum [now Cladosporium], and Cephalosporium), and in 4 (10.5%) of 38 wound isolates from 23 ambulatory patients (28). In each of these reports, multiple organisms were isolated from many of the patients, and the fungi recovered may not necessarily have been agents of infection.

The risk for wound infections after a natural disaster is high when wounds are contaminated with water, soil, or debris (32). In addition, damage to the local health care infrastructure can compromise the ability to properly irrigate contaminated wounds with sterile solution or promptly treat injured persons with topical or systemic antimicrobial drugs (32). These factors can result in severe, often polymicrobial, infections of otherwise relatively minor injuries (32). Although most documented disaster-associated soft tissue infections are bacterial (typically gram-negative pathogens such as Aeromonas, Escherichia coli, and Klebsiella) (33), fungal wound infections can also occur, and they could be under-recognized because they can be clinically similar to bacterial infections, particularly during the early stages of infection.

Mucormycosis, caused by fungi that belong to the order Mucorales, is perhaps the most recognized example of post-disaster fungal soft tissue infection. Necrotizing fasciitis can result, and case-fatality rates of ≈30% are frequently described, although early diagnosis and treatment has been shown to lead to better outcomes (34). The first documented instance of disaster-associated mucormycosis occurred after the 1985 volcanic eruption in Armero, Colombia, which caused an estimated 23,000 deaths and ≈4,500 injuries (17). According to a report of 38 patients with necrotizing lesions who were hospitalized after the volcano, 8 patients had infections caused by the mucormycete Rhizopus arrhizus (oryzae) (17).

Similarly, a cluster of mucormycosis cases caused by Apophysomyces trapeziformis occurred among 13 persons who were severely injured in the May 22, 2011, tornado in Joplin, Missouri, United States (4). Penetrating trauma and an increased number of wounds were shown to be independent risk factors for mucormycosis. Whole-genome sequencing of A. trapeziformis isolates from case-patients’ wounds (Figure) showed 4 nonidentical but closely related strains of A. trapeziformis. This finding, considered with case-patients’ receipt of medical care at different hospitals, suggested that the infections were acquired from the natural environment as a result of exposure to organic matter and water, which are likely reservoirs for mucormycetes (4).

In addition to the 2 clusters described, several isolated cases of post-disaster soft tissue mucormycosis have been reported, notably among persons injured during the 2004 Indian Ocean tsunami (13–15). These reports illustrate some of the clinical challenges associated with soft tissue mucormycosis, caused by organisms that may initially appear indistinguishable from other types of wound infections but require aggressive treatment with intravenous antifungal medication and surgical debridement (13,14).

Other agents of fungal soft tissue infections in survivors injured during the 2004 Indian Ocean tsunami include Fusarium (which later caused systemic infection) in 1 tourist (15) and Cladophialophora bantiana in 2 other tourists (10). Subcutaneous C. bantiana infection has also been associated with a tornado-associated injury in which the patient was inoculated by a contaminated wood splinter but did not have symptoms until ≈16 years later (35). Soft tissue fungal infections have also been documented among persons who were not directly injured during a disaster but who sustained minor trauma while performing post-disaster tasks: in Texas after Hurricane Ike in 2008, chromoblastomycosis was diagnosed in 3 patients, all of whom had histories of cancer and all of whom described clearing brush and fallen trees near their homes after the storm (8).

Although respiratory and cutaneous infections are the most commonly described forms of fungal infection after natural disasters, other, more invasive fungal infections have also been observed. An outbreak of Aspergillus meningitis after the 2004 Indian Ocean tsunami was associated with the use of spinal anesthesia for cesarean section infant delivery for 6 previously healthy women in Sri Lanka (12). The first 5 case-patients were initially treated for bacterial meningitis, but the discovery of Aspergillus during the post-mortem examination of the index case-patient led to the use of amphotericin B and voriconazole in the surviving case-patients (12). Investigation of various medical supplies revealed that syringes from a central storage facility were contaminated with A. fumigatus, probably as a result of suboptimal storage conditions in a humid warehouse (12).

Disaster-induced water damage to structures can create moist environments that can promote indoor fungal growth, but the extent to which damp indoor spaces and mold growth affect human health remains somewhat ambiguous (36). A 2004 report by the Institute of Medicine found sufficient evidence of association between indoor mold exposure and upper respiratory tract symptoms, cough, and wheezing, and evidence of an association between indoor mold exposure and some noninfectious health conditions that included asthma symptoms in persons with asthma and hypersensitivity pneumonitis in some groups of people (36). Although the report found no association between indoor mold exposure and invasive infection in healthy persons, there was evidence to support a link between exposure to Aspergillus and aspergillosis in severely immunocompromised persons (36).

Few data clearly demonstrate that indoor mold exposures increase the risk for invasive infection in post-disaster settings. Flooding lasted for weeks after Hurricanes Katrina and Rita made landfall on the US Gulf Coast in August and September 2005, respectively, leading to visible mold growth in 46% of 112 inspected homes (37). Despite these high levels of indoor mold growth documented in some areas, 1 study showed no elevated risk for fungal infections among immunocompromised patients exposed to water-damaged buildings after Hurricane Katrina; 1 patient, 1.2% of the profoundly immunocompromised study population, had a mold infection (caused by a Cladosporium sp.), which resolved without antifungal treatment (9). Colonization in the absence of related clinical symptoms was observed in persons who returned to their water-damaged homes after Hurricanes Rita and Katrina: the mucormycete Syncephalastrum was detected in various clinical specimens from 8 persons whose self-reported exposures to mold ranged from none to heavy, but none had evidence of invasive infection (38).

After the 2011 Great East Japan Earthquake and subsequent tsunami, a medical relief team observed unexplained chronic cough among a group of previously healthy persons living in a temporary refuge (25). Fungal cultures of sputum samples from 6 persons yielded Aspergillus fumigatus, A. flavus, and basidiomycetous fungi; culture plates exposed inside the refuge showed a similar fungal profile, suggesting that the indoor environment may have played a role in the patients’ infections (25). Although the authors of that report state that the patients’ coughs resolved after treatment with the antifungal itraconazole, the extent to which an infectious process was responsible for the illnesses is unclear (25).

Climate change could be affecting the ecology of pathogenic fungi in ways that are not yet fully understood; even minor or gradual changes in temperature, moisture, and wind patterns might affect fungal growth, distribution, and dispersal (20). For example, warmer average global temperatures may allow the geographic range of fungi typically restricted to tropical and subtropical environments, such as Cryptococcus gattii, to expand into areas that are currently more temperate (20). Global warming has also been hypothesized to select for fungi with tolerance to warmer temperatures (20). The relative scarcity of fungal diseases among mammals has been hypothesized to be associated with the inability of many fungal species to survive at temperatures >37°C; however, warmer ambient temperatures may enable nonpathogenic fungi to acquire the ability to infect warm-blooded hosts (20).

Huppert and Sparks suggest that global climate change is contributing to greater frequency and severity of extreme weather events and that current patterns of population growth, urbanization, and human activity create conditions that render many communities increasingly vulnerable to these hazards (39). Coupled with an increased risk for natural disasters, a larger or more geographically widespread ecologic burden of pathogenic fungi could lead to greater numbers of disaster-associated fungal infections through any of several mechanisms: inhalation of spores dispersed as a result of geophysical disruption, traumatic implantation of fungi into wounds contaminated with organic matter, or infection associated with suboptimal medical care where the local health care system has been damaged or destroyed. Altogether, a combination of factors including genetic and biological aspects of host–pathogen interactions; changing features of the physical environment; and social, political, or economic influences could lead to the emergence of new fungal pathogens or increased numbers of infections by known pathogens (40).

Disasters are complex events that can result in a wide range of health effects, although infectious disease outbreaks as an immediate consequence of disasters are uncommon. Health care providers should be aware of the possibility for cases or clusters of community-acquired or health care–associated fungal infections among disaster survivors because these infections often appear clinically similar to bacterial infections and can be associated with serious illness and death. These infections can occur in persons who do not have the typical immunocompromising risk factors for fungal infection but who have experienced near-drowning, trauma, or other unusual exposure to the environment, such as a dust storm. A fungal infection should be considered early if a patient has a persistent or progressive infection that is not responding to initial antibacterial treatment, particularly because rapid diagnosis and administration of appropriate antifungal therapy can improve patient outcomes. Prompt restoration of disaster-affected aspects of the local health care infrastructure may help facilitate earlier diagnosis and treatment and possibly reduce the risk for infection associated with the use of contaminated medical equipment or substandard care. Strategies to reduce disaster-associated fungal infections should be considered within the broader context of comprehensive and sustainable risk reduction methods to prevent disaster-related injury and illness.