This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Kuhn, D. M. Articles by Ghannoum, M. A. Search for Related Content PubMed PubMed Citation Articles by Kuhn, D. M. Articles by Ghannoum, M. A.

Indoor Mold, Toxigenic Fungi, and Stachybotrys chartarum: Infectious Disease Perspective

Division of Infectious Diseases, Department of Medicine,¹ Center for Medical Mycology, Department of Dermatology,² University Hospitals of Cleveland, and Case Western Reserve University, Cleveland, Ohio 44106³

SUMMARY

INTRODUCTION

INDOOR AIR AND BUILDING-RELATED ILLNESS

FUNGI IN THE INDOOR ENVIRONMENT

Fungal Organisms in Damp Buildings

Technical Problems in Determining Fungal Exposure

Difficulties in measuring fungal organisms.

Difficulties in measuring mycotoxins.

Problems with Clinical Studies

ASSOCIATION OF STACHYBOTRYS SPECIES WITH ''SICK BUILDINGS''

FUNGI AND FUNGAL TOXINS

Mycotoxins

Stachybotrys Species as Pathogens

Associations with Human Disease

Mycotoxins from Stachybotrys

SPECIFIC TOXICITIES

Pulmonary

Rationale for concerns regarding Stachybotrys.

Organic toxic dust syndrome, hypersensitivity pneumonitis, and allergic lung disease.

Cleveland infant idiopathic pulmonary hemorrhage outbreak.

Potential mechanisms of Stachybotrys-induced lung injury.

Neurologic

Hematologic and Immunologic

General concerns.

Aflatoxin.

Trichothecenes.

Malignancy

General concerns.

Aflatoxin.

Other mycotoxins.

Hepatic

Endocrine

Renal

Pregnancy

Other Toxicities

CONCLUSIONS

ACKNOWLEDGMENTS

REFERENCES

Top Next References

INTRODUCTION

Top Previous Next References

 
Damp buildings often have a moldy smell or obvious mold growth, and some molds are known human pathogens. This has caused concern regarding potential health effects of moldy indoor environments. As a result, there have been many studies of moisture- and mold-damaged buildings. More recently, there have been a growing number of articles in the media and of lawsuits claiming severe illness as a result of indoor mold exposure, particularly to Stachybotrys chartarum. However, while many authors report a clear relationship between fungal contaminated indoor environments and illness, close examination of the literature reveals a much more confusing picture.

In this review, we discuss indoor environmental mold exposure and mycotoxicosis, with an emphasis on S. chartarum and its toxins (due to the breadth of the topic, we will not discuss better understood areas such as invasive disease caused by Aspergillus). We also discuss specific organ effects, focusing on illnesses purportedly caused by indoor mold. These illnesses include pulmonary, immunologic, neurologic, and oncologic disorders. We discuss the Cleveland infant idiopathic pulmonary hemorrhage (IPH) reports in some detail, since they provided much of the fuel for current concerns about Stachybotrys exposure. As we will see, while there is cause for concern about the potential effects of indoor mold exposure, particularly to Stachybotrys species, there is no well-substantiated evidence linking the presence of this fungus to health concerns elaborated in the scientific and lay press.

As patients and society at large become increasingly concerned that illnesses may be due to the home or work environment, an understanding of mycotoxins by microbiologists and clinicians (especially infectious-disease subspecialists) is of growing importance. Such knowledge is critical to the diagnosis of potential fungus-related disease and is necessary to assuage fears instilled by extensive media coverage (34; J. MacFarlane, 1997, Beware the mold Stachybotrys, http://www.cnn.com/HEALTH/9711/05/deadly.mold; J. McKenzie, 2001, Hidden menace: insurers worry about toxic mold claims, http://more.abcnews.go.com/sections/wnt/dailynews?toxicmold_010626.html; E. Moriarty, 2000, Invisible killers, OBS News, New York, N.Y.; N. Morris, 2001, Moldy Schools: are your kids getting sick at school? http://more.abcnews.go.com/sections/wnt/WorldNewsTonight/wnt010418_moldyschools_feature.html). Finally, such knowledge may be important in the wake of recent terrorist events in the United States. Some toxins, particularly aflatoxins and trichothecenes, have the potential to be used as weapons. There is evidence that several countries are currently involved in mycotoxin weapon research (30, 239, 465). The latter point is beyond the scope of this article.

INDOOR AIR AND BUILDING-RELATED ILLNESS

Top Previous Next References

 
It has long been postulated that exposure to damp, moldy home and workplace environments has detrimental health effects. At the beginning of the 18th century, Ramazzini, considered "the father of occupational medicine," described an illness of workers inhaling ‘foul and mischievous powder' from handling crops (116, 302). More recently, Platt et al. (333), found that occupants of wet, moldy buildings had an increase in subjective complaints. Brunekreef et al. (47) found a similar pattern in >6,000 children in six states in the United States and reported home dampness was a strong predictor of respiratory and other illness in this age group. The list of putative symptoms generally consists of upper respiratory complaints, including headache, eye irritation, epistaxis, nasal and sinus congestion, cough, "cold and flu" symptoms, as well as generalized gastrointestinal complaints (240). Taskinen et al. (409, 410) reported an increased prevalence of asthma in moisture-affected schools, although there were no objective measurements of respiratory disease. A number of studies have reported a relationship between similar symptoms and damp housing or workplace environments, although the proposed etiologies have varied (117, 118, 164, 416, 419, 444).

The causal relationship between damp housing and illness is unclear. Establishing such a relationship is complicated since there are a variety of pollutants in the indoor environment (143) including volatile organic compounds such as toluene, benzene, alkenes, aromatic hydrocarbons, esters, alcohols, aldehydes, and ketones (208, 267, 277, 310, 332, 357, 450); radon (256); combustion gases, sulfur dioxide, nitrogen dioxide, carbon dioxide, ozone (16, 20, 357); and the essentially ubiquitous formaldehyde (205, 357). Other items (copy paper) and activities (photocopying and video terminal exposure) have been linked to symptoms (178). Other studies have suggested that shade, organic debris, landscaping quality, central electrostatic systems, ventilation rates, temperature, noise levels, dust control compliance, and patient gender may be important (23, 42, 77, 215, 264, 298, 308, 344, 437, 440), as well as the presence of tobacco smoke (81, 123, 276). Psychosocial issues may be playing a role in building-related complaints. Several studies have reported that the quality of the work environment, stress, and somaticization may all be significant (31, 42, 89, 232, 276, 307, 308, 413).

The indoor environment also contains a wide range of microorganisms including bacteria (e.g., Legionella and other gram-negative species) (85), mycobacteria (9, 415), and molds (161), as well as their products, including endotoxins and mycotoxins. There may often be a much higher bacterial load than fungal load (161, 416).

Most fungi are metabolically active over a broad temperature range (203); however, high moisture and relative humidity are required for optimal growth (69). The lowest relative humidity supporting mold growth is approximately 75%, although the requirements of Stachybotrys are much higher, around 93% at 25°C (142). Increasing temperature and nutritional status of the substrate can lead to lower moisture requirements. Surfaces that are soiled or have susceptible paint or paper do not need to be as damp for mold to develop. While promoting mold growth, moisture itself may be critical in "sick-building syndrome" (SBS) illnesses, since humidity affects mite and ozone levels, as well as off-gassing, salt, and acid formation (26). The links between moisture damage, any of these related cofactors, and building-related illnesses are not clear (24, 27, 50, 83, 260). For example, dust mites are notorious allergic agents and produce many of the upper airway symptoms ascribed to mold exposure or SBS; moreover, they are almost always found in association with mold species (90, 262, 359), confounding moisture- and mold-related findings. Gram-negative bacteria, endotoxin, and mycobacteria are found in water-damaged buildings in association with mold (9, 85). To our knowledge, only one paper has actually reported a lack of association between symptom prevalence and endotoxin, dust mites, or other nonfungal agents (85). In moldy office buildings there is an association between microbial contamination and repeated flooding or stagnant pools of water (280). Some geographic locales are obviously more likely to be affected than others. For example, 12% of English building stock suffer serious dampness; extrapolation suggested that there were 2.5 million affected dwellings in the United Kingdom but that 60% of these were from condensation rather than overt flooding (362; Anonymous, Bldg. Res. Estab. Semin. Proc., 1981). Readers interested in an in-depth review of these issues are referred to the recent comprehensive report by the Institute of Medicine (175).

FUNGI IN THE INDOOR ENVIRONMENT

Top Previous Next References

 

A final problem in measuring fungal organisms in the indoor environment relates to selective sampling. As noted above, Stachybotrys species rarely exist in isolation. They are often present in settings which select for a host of other fungal species and their potential mycotoxins, as well as bacteria, mycobacteria, arthropods, and man-made organic chemicals. However, most studies cited below have used methods that preferentially select for Stachybotrys species and mycotoxins. Of more immediate scientific and medicolegal concern, many studies of purportedly affected housing are surveying only for Stachybotrys species while ignoring other organisms (our unpublished experience).

Difficulties in measuring mycotoxins. As discussed in detail below, similar problems exist regarding the detection and significance of indoor environmental mycotoxins. Many purported fungal volatiles are in fact common and are not unequivocally fungal in origin (267). While some true mycotoxins have been detected in indoor air, this has usually been in the context of heavy industrial contamination (240, 295). Although it is occasionally possible to collect mycotoxins by using air filters followed by extraction (318, 388), they are usually isolated from inert dust or building materials (9, 79, 111, 267, 294). This may misrepresent exposure, since the compounds are not volatile. In the case of Stachybotrys, toxin-bearing spores are produced in a slimy mass with high moisture content (Fig. 1), becoming airborne only when dry and disturbed or when attached to other particles such as dust (161). Serologic testing of potentially exposed individuals is not useful, since specific immunoglobulin levels do not correlate with exposure (419).

View larger version (297K): [in this window] [in a new window]   FIG. 1. Morphology of S. chartarum. (A) Representative section of damaged wallboard. Note the areas of black discoloration. (B) Pure culture of S. chartarum, initially obtained from contaminated wallboard (40x objective). (C) Higher-power view of the same culture as in panel B (100x oil objective). (D) Scanning electron micrograph of conidia at the tip of a conidiaphore. (E) Scanning electron micrograph of mature conidia. Slime has been removed by scanning electron microscope processing. Panels D and E are reprinted from Nelson, http://www.apsnet.org/online/feature/stachybotrys/ with permission.

Such findings may explain the confusing results of earlier studies. For example, some authors have claimed links between childhood asthma and damp, moldy housing (401). While retrospective questionnaires reported more wheezing, cough, and chest cold symptoms in children from affected houses, the degree of bronchospasm was not different between groups. Thus, despite the claim that there was a causal association between moldy houses and wheezing, there was no supporting objective evidence. Some studies which claim that moisture and mold were associated with respiratory infections, cough, and wheezing (again with no objective measures) also fail to show differences in asthma prevalence between case and control schools (409, 410). Other authors report that despite claims of symptoms being more prevalent in case groups (reporting exposure to fungi, pets, mold odor, and dampness), actual asthma prevalence was no different (177).

ASSOCIATION OF STACHYBOTRYS SPECIES WITH "SICK BUILDINGS"

Top Previous Next References

 
Because of concerns of mold-induced building-related illness and the particular characteristics of Stachybotrys species, there has been growing concern about the health of occupants of Stachybotrys-"damaged" buildings (9, 79, 91, 133, 150, 157, 184, 188, 197, 241, 318, 423, 424). Many authors have reported ill effects in relation to Stachybotrys, although it is critical to note these reports are often associations rather than proof of causation. Hodgeson et al. (164) reported building-related illness in Florida; this was described as symptoms consisting of mucosal irritation, fatigue, headache, and chest tightness that occurred within weeks of moving into the affected building. The symptoms were purportedly caused by S. chartarum and A. versicolor, although a number of other species were seen. The authors identified mycotoxins including satratoxins G and H (see below) in moldy ceiling tiles, although the significance of these findings is unclear. While they concluded the symptom outbreak was likely a result of inhalation of fungal toxins, there was in fact no clear evidence (e.g., laboratory parameters) to support the claim. Tuomi et al. (420) examined Finnish buildings with water damage and identified a host of fungal organisms and mycotoxins (satratoxins G and H, T-2 toxin, and the aflatoxin precursor sterigmonisin) in bulk samples, although the relationship between the organisms and toxins was unclear, as explored below. Examining buildings with building-related illness complaints, Johanning et al. (197) isolated satratoxin H and spirocyclic lactones from water damaged material. The authors implied these mycotoxins were the cause of respiratory and immune problems, although, as we discuss below, the claims are questionable. Other authors have reported anecdotal cases of illness in which S. chartarum and mycotoxins have been isolated from building materials, but again there are few objective measurements of illness or clear etiologic links to the fungus (79). While authors claim the health effects are similar to past cases of stachybotryotoxicosis, such effects are often vague, poorly described, and clearly not the same as the serious illnesses of equine stachybotryotoxicosis and alimentary toxic aleukia described below.

One of the best studies of building-related illness showed minimal relation to Stachybotrys. Miller et al. (267) examined 50 Canadian homes in which the occupants had complaints of respiratory or allergic symptoms for which there was no explanation, although at the time of the study, occupants of only 6 houses had "building-related illnesses." Looking at air exchange rate, moisture levels, and analyzing air and dust for fungus and fungal products in 37 of the homes, they found S. chartarum in only one house; analysis of the 6 "sick" houses did not indicate fungus-related disease. During parts of the year when windows are open, indoor fungi are comparable to outdoor species (Cladosporium, Alternaria, and Aureobasidium) (391). However, in this study, outdoor air spores were negligible and Penicillium and other soil fungi were most important. Toxigenic fungi included P. viridicatum, Trichoderma viride, P. decumbens, and A. versicolor. House dust usually contained "appreciable" amounts of filamentous fungi and yeast, and so it was expected that spores could be found in air, depending on the activity in the room.

Recently, there has been a great concern regarding exposure of school children in "contaminated schools," sometimes resulting in building closures (367, 409, 414; Norris, ABC News online article). In fact schools may have lower mean viable mold spore counts than the students' homes (105). In one 22-month study of 48 schools in which there were concerns regarding indoor air quality and health (rhinitis and congestion which improved when the students were away from school), fewer than 50% of affected schools had fungal CFUs higher than outdoor air (72). In 11 schools where complaint areas had samples with the same organisms as outdoors, Stachybotrys was found, but only on surface swabs and not air specimens. The researchers did not look for other etiologies, nor were there objective measures of illness. Taskinen et al. (409, 410) also reported an increase in asthma in moisture- and mold-affected schools but presented no objective measurements of asthma and very limited immune data, including surprisingly low incidences of positive skin prick tests. Other authors have presented similar findings, reporting that "exposed" children had a higher prevalence of respiratory symptoms and infection, doctor visits, and antibiotic use, and got better post renovation (367). However despite claiming "[exposure]...increased the indoor air problems of the schools and affected the respiratory health of the children," the study was neither controlled nor blinded, and presented no physical diagnosis or objective measures.

Other evidence suggests that Stachybotrys exposure is not responsible for these building-related episodes. Sudakin (404) examined water-damaged buildings in the Pacific Northwest, due to occupants' neurobehavioral and upper respiratory health complaints (there were no objective pulmonary data) and found S. chartarum in only 1 of 19 cellulose agar cultures from building materials; the fungus was not detected in any of the above samples. While employees felt better after being relocated, there was no evidence that Stachybotrys was a causative agent. Even when large amounts of fungus are detected, analysis often fails to show direct links between symptomatic residents and fungal growth (41). In studies reporting that exposure to home dampness and mold may be a risk factor for respiratory disease, other factors such as smoking may be more contributory (84). In buildings with moisture problems where mycotoxins have been identified, a variety of species are identified, and links between a particular organism and toxin often cannot be established (420).

Despite these problems and an almost complete lack of objective evidence to support guidelines, broad recommendations have been made concerning indoor mold exposure, acceptable air contamination limits, and remediation goals. The sources range from individual authors (267, 292) to the American Academy of Pediatrics (7a) to government agencies (15). Nikodemusz et al. (292) declared that microbial monitoring of air is important even though the organisms the author found were not pathogens. While Miller et al. (267) admit that their "data seriously call into question any attempt to set arbitrary standards for fungal CFU values," they proposed that some fungi should be considered unacceptable, e.g., pathogens and certain toxigenic species such as S. chartarum, even though complete elimination would be untenable. The same authors stated that it is reasonable to assume there is a problem if a single species predominates with >50 CFU m^-3; that <150 CFU m^-3 is acceptable if there is a mix of benign species; and that there is no problem when up to 300 CFU of Cladosporium or other common phylloplane fungi m^-3 is isolated. Notably there is no source material to support these assertions. The American Association of Pediatrics produced guidelines in the wake of the Cleveland IPH story (7a), again without substantial evidence. More moderate recommendations (while recognizing that the presence of fungi does not necessarily imply illness) would appear reasonable (240). These could include maintaining heating, ventilation, and air conditioning (HVAC) systems, controlling humidity, inspecting and repairing water damage and other sources of contamination, regularly cleaning the home environment with dust removal, cleaning carpets, removing visible mold growth, and formulating guidelines to standardize the levels of fungal and bacterial contamination.

There have been a number of media reports on the abandonment or destruction of buildings contaminated with S. chartarum (34; McFarlane, CNN online article; Moriarty, CBS News program). It is unlikely that such extreme measures are warranted. Methods are discussed further below, but it is important to note that individuals get better with remediation efforts (6, 72, 79, 191, 321, 367), although perhaps not always (164). Simple methods, including removing damaged material and spraying affected areas with bleach, are generally effective in controlling contamination and result in "clean" air samples (430). In some cases, temperature and humidity control may be adequate (142).

FUNGI AND FUNGAL TOXINS

Top Previous Next References

 

Mycotoxins are diverse secondary metabolites produced by fungi growing on a variety of foodstuffs consumed by both animals and humans (Table 2) (76). Clinical toxicological syndromes caused by ingestion of large amounts of mycotoxins have been well characterized in animals and range from acute mortality to slow growth and reduced reproductive efficiency. The effects on humans are much less well characterized (Table 3) (76, 329). Outbreaks of various types of animal mycotoxicosis have occurred worldwide in livestock, including sweet clover poisoning, moldy- corn toxicosis, cornstalk disease, bovine hyperkeratosis, and poultry hemorrhagic syndrome (76, 133).

A variety of factors affect toxin occurrence (157). Many toxins are secondary metabolites, produced under suboptimal growth conditions (48) or in the presence of limited nutrients (161). (For reviews of toxin synthesis, see references 46 and 406.) Temperature, relative humidity, moisture, and growth rate all affect fungal mass as well as toxin synthesis. Aflatoxin production by Aspergillus is dependent on concentrations of O₂, CO₂, zinc, and copper, as well as physical location (A. fumigatus and A. flavus grow in trench silos, while upright silos favor Fusarium species) (402). Ochratoxin production relates to air exhaustion (69), patulin production relates to limiting nitrogen, ergot production relates to phosphate limitation (48), and A. parasiticus toxin production relates to temperature (35). These considerations are critical, since the recovery of toxigenic species from any environment does not substantiate the presence of a mycotoxin (mycotoxin production is not a necessary result of fungal growth) (48). Indeed, the conditions necessary for mycotoxin production are usually very different from those required for growth; for example, Fusarium tricintum produces a significant amount of T-2 toxin at 15°C but little at higher temperatures (69).

The most notorious and best described of the mycotoxins are the aflatoxins. In the early 1960s, an outbreak of turkey X disease in England, in which over 100,000 fowl died, was later traced to contaminated peanuts from Brazil (454). Aflatoxins were subsequently identified as the toxic agent. While made primarily by Aspergillus species, these toxins are also produced by Penicillium and Fusarium species (406). A. flavus makes aflatoxin B (AFB), while A. parasiticus produces both AFB and AFG. AFM₁ and AFM₂ are oxidative metabolic products made after ingestion and appear in milk, urine, and feces. The aflatoxins are toxic, immunosuppressive, mutogenic, teratogenic, and carcinogenic, and their main target is the liver. Most have been classified as type 1 carcinogens (172). AFB₁ is probably the most potent liver carcinogen for a variety of species, including humans (102). Aflatoxin-related disease can occur in outbreaks, causing acute, often fatal, liver injury. The compounds have been best studied in veterinary practice, where they show the most potent effects. Toxicity is species, age, and route dependent; for example, farm animals ingest large quantities in feed (402). Species variability may relate to the ability to form epoxide derivatives in liver microsomes and endoplasmic reticulum.

The case of aflatoxin also illustrates the problems of elucidating clinically relevant levels of mycotoxins. Determining actual exposure levels is exceedingly difficult, even in known contaminated foodstuffs (334). While aflatoxin contaminates many imported goods (from almonds to melon seeds), there is a large variation in toxin distribution. Data validity is suspect when looking at small quantities, since aflatoxin is normally found in only very limited portions of a food lot and levels in such samples can range from 0 to >400,000 ng/g. For many mycotoxins, it becomes a matter of how hard one looks, and as more sensitive methods are developed, more toxins are found. Currently, at least 29 mycotoxins have been identified in commercially available foods or feeds (57), and in rare cases of high feed contamination they have been found in meat, milk, and eggs.

While in 1938 this animal disease was associated with Stachybotrys species, it was nearly a decade before the etiologic organism was identified in contaminated grain as S. alternans var. jateli (133, 429). In 1837, Corda first defined the Stachybotrys genus, from a strain growing on domestic wallpaper in Prague (75). S. alternans and S. atra are obsolete species names, and the organism has been renamed S. chartarum (www.doctorfungus.org/imageban/synonyms/stachybotrys.htm). S. chartarum is dematiaceous and is a member of the Fungi Imperfecti. Interested readers should refer to the article by Jong and Davis (200) as well as the recent on-line review by Nelson (B. D. Nelson, 2001, Stachybotrys chartarum: the toxic indoor mold, APSnet, http://www.apsnet.org/online/feature/stachybotrys).

In the 1940s Drobotko et al. (103, 104) reported finding the fungus on straw and fed it to horses, which developed the characteristic equine illness. At the same time, scientists recognized that the disease was toxin mediated and that exposure to isolated toxin could produce symptoms (133). Subsequently, Drobotko coined the term "stachybotryotoxicosis." By the late 1940s, similar outbreaks of livestock disease had been reported in the rest of the USSR and Eastern Europe, as well as among nonequine species (134, 387, 434, 455). More recently, animal disease has been seen from Europe to South Africa (150, 212, 224, 225, 373; P. Le Bars and J. Le Bars, Proc. 5th Int. Working Conf. Stored Product Prot., 1990). In South Africa, a case of sheep disease was seen in the 1990s after animals consumed heavily contaminated grain cubes (374). The affected animals had fever, listlessness, oral lesions, pancytopenia, hemorrhage, opportunistic infections, and a significant mortality rate.

Members of the Stachybotrys genus exist worldwide, from Finland to the South Pacific Islands (133, 434), and were identified in the United States in the 1940s (442). In general, the organism is found in soil and strata rich in cellulose (hay, straw, grain, hemp, plant debris, dead roots, wood pulp, cotton, fabrics, paper, book bindery glue, plant fiber-processing plants, etc.) (133). It has also appeared in cigarette tobacco (1, 115). The organism can survive a wide temperature range, only dying at temperatures of >60°C. The fungus can survive over winter, spores stay viable for years to decades, and conidia retain viability despite passage through the gastrointestinal tract (but are killed in composting and degradation of manure). These factors, combined with their geographic range, suggest that Stachybotrys species are essentially ubiquitous. While disinfectants kill conidia and mycelia, cell walls are quite stable.

Despite the association of Stachybotrys with animal and human disease, early researchers were unable to fulfill Koch's postulates with the fungus. There was no evidence that the fungus itself was a pathogen, and scientists could not transmit infection or disease by injection of tissue from affected animals (104). At most, injection of the fungus caused a local response but no systemic invasion (429). It was only with the identification and application of Stachybotrys toxins that the nature of the disease process was understood.

View larger version (21K): [in this window] [in a new window]   FIG. 2. Structures of some representative mycotoxins.

The most famous purported case of mass human trichothecene toxicity was in fact due to Fusarium. The illness was initially dubbed "septic sore throat" and subsequently called alimentary toxic aleukia (ATA) (7, 193-195, 252, 253, 274, 369, 380, 408, 429). It occurred in Russia in the early 20th century, most notably prior to and during World War II, and was characterized by several stages. Initially, there was oral mucosal ulceration and gastroenteritis. Subsequently, there was pancytopenia accompanied by fatigue, vertigo, and hypotension. The illness had a substantial mortality rate, at least in part due to opportunistic bacterial infections developing in the later stages of the disease (53, 69, 370). One-third of family members who ate contaminated grain became ill, and one-third of those died; this was responsible for thousands of deaths. Where nutrition was good, morbidity and mortality were much lower (7, 274). As in the case of the equine stachybotryotoxicosis, Koch's postulates could not be fulfilled during research into the disease's etiology. The illness was subsequently attributed to trichothecene mycotoxins in overwintered grain infected with F. sporotrichioides or F. poae (194). The grain had been left in the fields due to the severe conditions prior to and during the war.

The precise toxic cause of the illness was never fully identified. A number of mycotoxins were obtained from laboratory cultures of F. sporotrichioides and F. poae: trichothecenes (T-2, HT-2, and neosolaniol), sterols (sporofusarin, sporofusariogenin, poaefusarin, and poaefusariogenin), and fatty acids (29, 195, 273, 382, 452, 453). The relevance of these mycotoxins to the observed human illness is still not clear. T-2 toxin was commonly produced and may be responsible (69), and T-2 toxin produces an ATA-like illness in cats (233, 234); however, many of the other compounds also produced toxic effects in cats (195). ATA has not been seen since the last Soviet reports in the late 1940s, and so there have been no recent episodes to study by modern analytical techniques. (In 1970, Saito and Tatsuno [353] described four human cases of Akakabi-byo, or red mold disease, which had some similarity to ATA. Analysis of fungus-contaminated grain revealed F. graminanum and mycotoxins including nivalenol, fusarenone-X, and diacetoxyscirpenol.) The reasons for the disappearance of ATA are probably multifactorial, including improved grain handling and nutrition. Several important points are worth mentioning. First, while the clinical picture of ATA is similar to that of equine stachybotryotoxicosis, the former disease is almost certainly caused by Fusarium rather than Stachybotrys species. Second, the exact toxicologic mechanism has not been determined. Third, the population affected by ATA also suffered severe nutritional deficiency (252, 253), which can produce similar symptoms.

While the effects of large amounts of orally ingested mycotoxins on animals have been well described, systemic effects of exposure to inhaled mycotoxins have been much less characterized. As noted above, workers exposed to aerosolized mold or mycotoxins during the equine stachybotryotoxicosis outbreak had a very different constellation of symptoms from animals ingesting affected grain. Only a few mycotoxins have been conclusively shown in aerosols, including aflatoxin, some trichothecenes (157), zearalenone (69), and secalonic acid D (112). This is because purified mycotoxins are not volatile (370), due to their high molecular weights (71). Therefore, inhalation exposure probably occurs through inhalation of airborne particulates containing mycotoxins, such as dust and fungal components (314, 443; H. B. Schiefer, Proc. 5th Int. Conf. Indoor Air Qual. Climate, 1990). Moreover, since mycotoxins have been postulated to normally be confined in spores, it is doubtful that they frequently reach the lower airways due to size limitations. The depth of particle penetration is inversely proportional to size; the upper airways trap particles of 10 to 60 µm, while particles 2 to 4 µm in diameter can reach the alveoli (141). Mold spore size depends on the organism; e.g., Alternaria spores are more than 7 µm in diameter, while thermophillic actinomycetes are less than 1 µm (141). This point will become important below, when discussing models of stachybotryotoxicosis.

SPECIFIC TOXICITIES

Top Previous Next References

 

Possible links between mold and respiratory disease have been recognized for more than a century; literature describing connections between indoor air and pulmonary disease is cited above. Initial study of the problem came about due to work-related problems in several areas, including farming and industry (73, 376). Early Soviet literature described a toxicosis associated with inhalation of dust heavily contaminated with spores of a variety of fungi (e.g., A. fumigatus, A. niger, Dendrodochium toxicum, and Stachybotrys), which was subsequently dubbed respiratory mycotoxicosis or pneumomycotoxicosis (360, 361). This illness was considered an occupational disease in many professions, including those involving work in binder twine factories, cottonseed oil processing plants, grain elevators, and mills (133, 360). However, occupational lung disease occurs in many industries, generally in the absence of an infectious agent (73), so that it is often unclear if fungi are the responsible agents or epiphenomenona.

With the dissemination of the concept of toxigenic indoor mold, scientific reports (see below) and legal claims (13, 14, 17; C. Kingdollar, 2001, Pollution litigation review—August 2001, http://www.facworld.com/FACworld.nsf/doc/pollitrev0801) of mold-induced respiratory complaints have become commonplace. For example, Johanning et al. (197) reported that exposure to toxigenic S. chartarum and other atypical fungi was associated with disorders of the respiratory system, although pulmonary disease was never documented beyond subjective complaints. However, it must be noted that many different conditions can produce essentially the same respiratory symptoms (129). These range from the benign, such as congestion and cough from rhinitis, to reactive airways disease to more serious syndromes including alveolitis, bronchiectasis, and pulmonary fibrosis. Within the rubric of subjective complaints subsequently described as "asthma," there may be a variety of actual syndromes ranging from asthma to allergic bronchopulmonary aspergillosis, hypersensitivity pneumonitis, emphysema, pulmonary fibrosis, and pulmonary hemosiderosis (73, 128). It is also critical to distinguish conditions which are readily reversible from those which produce permanent damage: most mold-related respiratory diseases in fact appear reversible (72). This is especially important given legal claims of permanent lung injury and "lung scarring" (Anonymous, 2000, New details from mold investigation: news comes too late for former employee, ChannelCincinnati, http://www.channelcincinnati.com/cin/news/investigations/stories/investigations-20000126-221100.html; Kingdollar, Pollution litigation review—August 2001 online article; M. McConnell, 2001, Awareness of mold problem increasing among homeowners, hvac industry, http://www.snipsmag.com/CDA/ArticleInformation/features/BNP_Features_Item/0,3374,20041,00.html). Although these conditions can be diagnosed by methods ranging from radiology to pulmonary function tests (PFTs) to biopsy, most studies have not done so, relying only on subjective reports instead. A number of papers have made claims regarding asthma and interstitial and emphysematous lung disease with no data beyond subjective questionnaires, which cannot diagnose many of these conditions (164). As noted above, objective studies often reveal poor correlation between complaints (on retrospective questionnaires) and actual pathology. Thus, a main concern lies in determining if there is actually disease beyond mild upper airway inflammatory responses and in whether these symptoms are due to fungus as opposed to other contaminants (73, 319).

Organic toxic dust syndrome, hypersensitivity pneumonitis, and allergic lung disease. Exposure to massive amounts of fungus can cause a significant, but transitory, acute lung injury. Anecdotal reports have documented disease, generally in farmers, who inhaled large quantities of organisms (116, 286, 324). All describe an acute organic dust toxic syndrome or silo unloaders syndrome (also called atypical farmer's lung or pulmonary mycotoxicosis). This illness differs from hypersensitivity pneumonitis (HP) in that it is transient, occurs in naive patients, needs intense exposure, neutrophils and not lymphocytes are found on brochoalveolar lavage (BAL), and fungal precipitin testing is negative (377). The episodes described occurred shortly after inhalation of unusually massive amounts of fungus and resulted in cough, respiratory distress (sometimes requiring ventilatory support), fever, fatigue, alveolar and interstitial infiltrates on chest X-ray, and leukocytosis. These cases are relatively unusual since lung biopsy alveolar specimens showed fungus, including A. niger (286), Fusarium, Penicillium (324), and in one case five different organisms (116). In addition to fungal organisms, histopathologic testing showed acute and organizing diffuse alveolar damage or bronchopneumonia (324). Serologic studies for antigens capable of causing hypersensitivity pneumonitis were negative (286). Most importantly, all patients recovered, with no residual chest X-ray abnormalities or residual deficits (324). While one author claimed that these cases were due to fungal toxins (116), there was nothing to suggest that these responses were more than an acute pneumonitis or near-drowning-type response. To our knowledge, these cases are the only ones where fungi have been demonstrated in lung tissue (aside from the quite different diseases of frank pulmonary fungal infection with organisms such as Aspergillus), with one exception noted below. This is an important point, since as discussed below, most animal models of Stachybotrys pulmonary toxicity rely on direct inoculation of organisms into the airway, in concentrations high enough that they can later be seen in lung tissue.

Allergic pulmonary effects of mold have been well described, ranging from upper airway inflammation (rhinitis with coincident conjunctivitis) to asthma, allergic pulmonary aspergillosis (the latter usually affecting patients with intrinsic lung disease and bronchiectasis), and HP (73, 301, 345, 376). While an in-depth discussion of this topic is beyond the scope of this review, some findings are worth mentioning. Occupational lung disease alone consists of a number of subtypes, including industrial bronchitis, occupational asthma (in which a very small amount of offending agent can cause bronchospasm after airways sensitization, e.g., in animal handlers and crab processors), byssinosis (Monday morning fever), and grain dust-induced bronchitis (73). There is substantial evidence that the last two illnesses are caused by endotoxin exposure (58, 156, 235).

In several cases, HP has been documented after exposure to indoor mold (usually Penicillium species), generally in the setting of faulty ventilation systems (2, 37, 126, 356). Summer-type HP in Japan is probably caused by T. cutaneum (10, 355, 385, 393, 458), although supporting studies are limited by lack of objective data as well as cocontamination (11 case homes had 3,536 strains of fungi [459]). While some forms of HP appear to be due to fungi, the disease can be caused by dozens of agents, ranging from bird proteins to thermophilic actinomycetes (376). Additionally, the differential diagnosis of the condition includes organic dust toxic syndrome, humidifier fever, secondary (from chemotherapy, radiotherapy, inhaled toxins, and pneumoconiosis) and idiopathic pulmonary fibrosis, granulomatous disease (e.g., sarcoid), and congestive heart failure (376). HP is reversible if the patient is removed from exposure to the offending agent, although in some cases prolonged (years) exposure may lead to pulmonary fibrosis and pulmonary hypertension.

Despite claims that building contamination increases asthma symptoms, there is a profound lack of objective data. Notably, one of the few studies to employ objective measures of lung function (PFTs with methacholine challenge) failed to show lower respiratory effects (284). Another paper reported decreased diffusion capacity in individuals with respiratory symptoms who worked a problem building, but this is of unclear significance, given there were no changes in other PFTs and it was not certain if the symptoms were related to the building or preexisting conditions (164). A small Finnish study reported new cases of asthma in occupants of a water-damaged building, confirmed by Sporobolomyces salmonicolor inhalation provocation tests (381) but simultaneously claimed the illnesses were not immunoglobulin E (IgE) mediated. Finally, an experiment which allowed workers to have individual control of their ventilation systems actually led to higher concentrations of airborne dust and fungi while producing fewer symptoms (263).

In some cases it appears there is a correlation between SBS and upper airway allergic symptoms, although the etiology is unclear and is probably varied (176; H. M. Ammann, 2001, Is indoor mold contamination a threat to human health? http://www.doh.wa.gov/ehp/oehas/mold.html). Objective studies have found nasal mucosal hyperreactivity (305), and alterations in tear film stability (285). Jaakkola and Jaakkola (178) found that symptoms were associated with the use of photocopy paper.

Researchers have sometimes found an association between building-related symptoms and mold-specific IgE, although it occurred in a minority of patients. Surprisingly, there was not an association with self-reported hay fever or asthma (222), even though allergic asthma is IgE mediated (345). These authors apparently did not check for IgE to nonmold allergens. Other authors have failed to show links between atopy and symptoms. Higher mold IgE levels have been found in individuals exposed to water damaged structures (242, 368). Several groups have reported links between Alternaria allergen sensitivity (measured by specific IgE), but the relationship was not as strong as that between routine indoor allergens and asthma (325). Of a group of Chinese schoolteachers exposed to moldy sugarcane, 42% developed allergic alveolitis. After the outbreak, patients were found to have elevated IgE levels, and Penicillium and Mucor species were identified on the sugarcane (155). Other studies failed to show associations between IgE and disease (164). Nasal lavage of individuals from contaminated buildings show increased tumor necrosis factor alpha, interleukin-6 (IL-6), and nitric oxide levels in relation to periods of documented exposure (163). However, the study did not document which fungus was responsible or if the results were controlled for confounding factors. Other work found increased concentrations of eosinophils, eosinophil cationic protein, and myeloperoxidase in the nasal lavage fluid of office workers in damp buildings compared to controls (435, 436). Damp buildings had larger amounts of molds and bacteria, as well as evidence of volatiles from degraded polyvinyl chloride floor coatings. Polyvinyl chloride degradation products have been linked to symptoms by using objective physical measures (444). Thus, while evidence suggests a larger amount of allergic markers in individuals exposed to water-damaged structures, the etiology of these changes is far from certain (242). Furthermore, potential fungal antigens are many, and not well identified. While airborne 1-3-ß-D-glucan is widely quoted to cause airways inflammation, there is limited evidence to support the assertion (132, 351, 412).

Cleveland infant idiopathic pulmonary hemorrhage outbreak. In the wake of an outbreak of infant pulmonary hemorrhage (60, 65), Montana et al. (278) described a cluster of 10 infants from the Cleveland area who were diagnosed with IPH after presenting with severe respiratory disturbances requiring intensive care treatment. Of the affected infants, 50% had symptom recurrence after returning home, although these events occurred anywhere from days to many months afterward. Epidemiologic investigation using retrospective questionnaires and home examinations concluded that home water damage prior to the clinical event was the main risk factor (OR = 16). Other factors appeared to include "any relative who coughed blood" (OR = 33), birth weight (normal birth weight OR = 0.12), breastfeeding (OR = 0.16), and home smoking (OR = 7.9); the last two factors became nonsignificant after stratification in a model which included home water damage. Affected infants were found to have significant differences from controls in red blood cell indices, hemolysis, and serum cholinesterase levels. There were several problems with the study. First, case and control infants were significantly different with regard to sex, race, birth weight, breastfeeding, smoking, and the presence of electric fans; nonsignificant differences existed for gestational and maternal age. The definition of close relatives with pulmonary hemorrhage (PH) was erroneous, since there are many causes of hemoptysis in adults other than true PH, especially in smokers (129). Moreover, there was no evidence of increased illness in other family members, despite seemingly common exposure to fungi. Finally, a prior paper had described a cluster of Greek infants with IPH, possibly caused by pesticides in grain stored near children's sleeping areas (56). While the Cleveland study did not find an effect of pesticides, trace amounts of household pesticides were present in case and control homes, and the home of the one child who died had the largest amount of airborne solvents and pesticide residues. The case children in Cleveland had significantly decreased levels of serum cholinesterase compared to controls, though still within the normal range.

The same group examined whether fungal exposure was a risk factor for IPH (123). The hypothesis arose from previous work on the issue of pulmonary disease from wet building and fungal exposure, and because hemolysis on case infants' blood smears might implicate stachybotryotoxicosis (since Stachybotrys toxins may produce hemorrhagic disease and hemolysis in animals) (133, 373, 407). Retrospective questionnaires and home surveys appeared to rule out pesticides, personal care products, or maternal cocaine use as etiologies. In general, case houses had more fungi, but the OR was near 1 for all species but Stachybotrys, which was 9.83 for IPH (risk of disease given a 10-fold increase in fungi). While Aspergillus, Cladosporium, and Penicillium species were abundant in case infant homes, matched analysis failed to demonstrate differences between case and control homes. The authors concluded that infants with IPH were more likely than controls to live in homes with toxigenic S. chartarum and other fungi, although toxigenicity was not in fact demonstrated. As later noted by the Centers for Disease Control and Prevention CDC (61-64), surface samples of mold cannot reliably predict airborne contamination or individual exposure (323), since release of spores is variable and dependent on environmental factors including temperature, relative humidity, light, and air movement (228). Moreover, isolation of organisms may not reflect toxin levels in homes, since toxin production is also variable and depends on many factors, as noted above (157).

In 1999, the same group published an overview of the Cleveland investigations, with 37 case infants (including 12 fatalities) reported from 1993 to 1997; they also mentioned 138 cases in the United States during the preceding 5 years (122). The clinical profile of the larger group of cases was heterogeneous, including 4 with stresses "that do not normally produce respiratory failure" (anesthesia, hypernatremia, water intoxication, and febrile seizure); 4 cases were associated with upper airway obstruction and 1 was associated with probable asphyxiation; 11% had developmental delay or failure to thrive; 22% had seizures; 19% had another infection (including Pneumocystis carinii pneumonia); and there were cases of hemolysis with hemoglobinuria. All 22 patients available for follow-up bronchoscopy had ongoing hemosiderosis (most >6 months), and 39% of infants needed additional therapy for reactive airway disease. The latter finding does not reflect previous reports of this disease. Importantly, gram-negative bacteria and endotoxin levels were not examined. While mold isolates produced mycotoxins, including satratoxins G and H, there was no difference between case and control-homes in this ability (189). Other fungi capable of producing mycotoxins may have confounded the picture, including two strains of Memnoniella echinata (since reclassified as Stachybotrys echinata) (21), which made trichodermol, trichodermin, and griseofulvins (189, 190), as well as Cladosporium species (279). Details regarding over 100 cases reported nationally are not available (405).

Subsequently, the CDC published a retraction of its support of the papers' conclusions (61-64), due to apparent shortcomings of the aforementioned studies. The CDC stated that, because of insufficient evidence, an association between S. chartarum and infant IPH was not proven. They also reported that "while [it is] advisable to remediate homes of mold for a variety of reasons, [it is] not solely because of the purported Stachybotrys/Acute IPH association." Primary criticisms