Ontario: Expert Panel Tackles The Health Risks of Toxigenic Mold and Fungus Indoors

Over the years, health departments have investigated community concerns about the health effects of poor indoor air quality. Many factors contribute to poor indoor air quality which may result in a variety of symptoms. The role of the health department is to investigate and determine which factors contribute to the symptoms experienced and to initiate controls. In the case of mold contamination of the indoor environment, assessment of health risk is often difficult. Determining health risk to individuals in contrast to risk to a community presents a public health challenge.

In order to assist health departments, the Environmental Health and Toxicology Unit (EHTU) of the Disease Control Service, Public Health Branch, convened a panel of experts in respiratory medicine, allergy, mycology, epidemiology and community medicine, to review the issue of mold contamination indoors. What follows is the Expert Panel Report which includes a scientific summary, points of consensus, response to specific questions, and recommendations.

Expert Panel Report
Scientific Summary

The health risks of molds in indoor environments are associated with the following potential pathogenic mechanisms

1. allergic (IgE-mediated asthma) and other immunemediated mechanisms (hypersensitivity pneumonitis);

2. toxic and irritative mechanisms (eg. mycotoxins, glucans, volatiles (pulmonary hemosiderosis, non-specific respiratory and flu-like symptoms);

3. infectious mechanisms (eg. illnesses in immunocompromised individuals from normally non-pathogenic organisms (asperigillosis, zygomycosis)

4. carcinogenic mechanisms (eg. aflatoxins).

The evidence for the clinical effects of exposure to indoor air molds is clear for conditions such as asthma, hypersensitivity pneumonitis, fungal infections, organic dust toxic syndrome and aflatoxin-related cancers.

However, the association between mold and non-specific symptomatology or more precisely, the direct chemical effects of mycotoxins and non-specific clinical syndromes (including respiratory and other symptoms), is less well understood.

• Characterization of human exposure to molds in residential settings is difficult with current methodologies. Difficulties relate to the counting of spores, multiple species of molds that may be growing in indoor environments, and determination of the fraction of viable spores. In future, in vitro techniques based on relevant biomarkers may provide better exposure information.

• Methods for measuring the production of mycotoxins in home settings are inexact and studies often rely on proxy measures such as occupant-reported mold or water damage.

• The symptoms and reported health effects recorded in these studies are difficult to validate objectively.

At this time it is not possible to definitely conclude that molds are responsible for the wide variety of reported symptoms.

Molds are ubiquitous in indoor and outdoor environments. Because of the limitations in quantifying risk, the burden of mold-related illness in the population is difficult to determine. For illnesses such as asthma for which there is good evidence of a mold-related etiology, the population burden is judged to be small. However, there are situations in which significant quantities of mold may develop resulting in significant health risks.

Such situations are often easily remediable. Remediation of a known amplifier (an indoor substrate where fungal growth is occurring) rests on removal of the amplifier and moisture control. Fungicides have limited if any value in these situations. Remedies for situations in which mold has accumulated over time are less straightforward. The current recommendations of HEPA (high efficiency particulate arresting) filter vacuuming and common sense cleaning are usually effective provided moisture and amplifier control are also undertaken. The CMCH booklet, "Clean-up procedures for mold in houses", provides the details of these clean-up procedures.

Consensus points

• Epidemiological studies have demonstrated a clear association between the presence of molds in homes and a number of respiratory and other symptoms.

• Many of the symptoms that have been associated with molds are nonspecific, and have also been associated with a number of other factors.

• Clinical evidence indicates that both asthma and hypersensitivity syndromes can be caused by exposure to mold in susceptible individuals.

• In vitro studies on fungal mycotoxins and glucans suggest a biological basis for human respiratory effects.

• Although molds can cause clear health effects under certain circumstances, the burden of illness in the general population remains to be clarified.

• The nature of the dose-response relationship for symptoms associated with molds is not clear. The possible existence of a threshold dose below which health effects would not occur needs to be considered in assessing population health risks.

Panel responses to specific questions

1. Are the health impacts of indoor air molds sufficient to be called a health hazard under the Health Protection and Promotion Act?

Under the HPPA, a health hazard means "(i) a condition of a premises, (ii) a substance, thing, plant or animal other than man, or (iii) a solid, liquid, gas or combination of any of them, that has or is likely to have an adverse effect on the health of any person".

Epidemiological and clinical evidence clearly indicates that molds contribute to respiratory and other symptoms. However, such effects tend to occur in susceptible individuals, or when exposures are high. Consequently, the Panel concluded that molds can pose a health hazard under certain conditions within the context of the HPPA.

2. What conditions must be present to reasonably anticipate that there is a health hazard?

With the possible exception of infectious or allergic diseases, it is generally difficult to attribute adverse health effects observed in individuals to specific environmental exposures, including exposure to indoor air molds. However, the Panel was able to identify factors that increased the risk of an adverse health effect occurring following exposure to indoor air molds. Adverse health effects have been observed with increased frequency in susceptible individuals subjected to high levels of exposure for example infants who developed pulmonary hemosiderosis. High inoculum potential also increases risk. The more noxious (high potential for mycotoxin production) species of mold pose the greatest risks.

3. What factors should be considered in host response?

A number of factors can affect individual response to exposure to indoor air molds. Infants, asthmatics, individuals with pre-existing respiratory conditions, and individuals with compromised immune systems are likely to be at increased risk.

4. What conditions can a Medical Officer of Health include in order to mitigate the health hazard?

Indoor air molds can be controlled in a number of ways. In general, indoor air molds can be reduced by eliminating conditions favourable to growth. Specifically, reducing water availability (including leaks and condensation) to the organisms will generally reduce growth. Disinfection or removal of affected materials, or materials in which molds may proliferate, are effective in eliminating molds. The Panel noted that because of the uncertainty associated with the risks of molds, mitigation actions need to be developed in the context of each situation.

Recommendations

• Because of the observation of pulmonary hemosiderosis among children exposed to high levels of mold in Cleveland, it may be worthwhile to investigate the occurrence of this condition in the Toronto area. If an excess of pulmonary hemosiderosis were observed, the role of Stachybotrysin the etiology of this condition should be investigated.

• The panel noted that there remain a number of uncertainties concerning the relationship between molds and health effects and that there continue to be ongoing investigations worldwide. The panel recommends that this issue be reviewed again after an appropriate interval.

As recommended by the Expert Panel, the EHTU will examine the prevalence of pulmonary hemosiderosis in infants in Ontario and pursue appropriate follow up studies. The Public Health Branch thanks the Expert Panel for their participation.

Panel Members

Dr. Daniel Krewski - Chair
Bureau of Chemical Hazards
Environmental Health Directorate
Health Canada

Dr. Susan Tarlo
Allergy
University of Toronto and Toronto Hospital

Dr. Robert Sales
Respiratory Medicine
University of Ottawa and Ottawa General Hospital

Dr. Vivek Goel
Medical Epidemiology
University of Toronto

Dr. Richard Summerbell
Mycologist
Ministry of Health

Dr. Ross Upshur
Community Medicine
University of Toronto

CONTACT

Patricia Powell, MEd, CPHI(C)
Environmental Health and Toxicology Unit
Disease Control Service
Public Health Branch
Telephone 416 327-7427
Facsimile 416 327-7439

This article appeared in the Public Health & Epidemiology Report Ontario (PHERO) Newsletter, Volume 8, Number 2, February 28, 1997. PHERO is published monthly by the Public Health Branch, Ministry of Health, North York. Telephone 416 327-7623.

PULMONARY HEMOSIDEROSIS IN ONTARIO INFANTS, 1991-1995

Abstract

Pulmonary hemosiderosis is a rare condition of infancy. In 1994, a cluster of cases of this disorder occurred in Cleveland. This cluster was investigated by the local health department and the Centers for Disease Control in Atlanta. The investigation demonstrated a strong association of the cases with water damage (flooding). The toxigenic mold, Stachybotrys atra, was found in all of the infants homes, though its concentration was not statistically associated with the cases. Other mold species were also identified as were low concentrations of several household pesticides.

There is a chain of evidence in these cases which points to the potential for mold as a contributory etiologic factor for pulmonary hemosiderosis. We began to investigate its occurrence in Ontario in order to determine if there is any evidence of clustering of this rather rare condition here. This report is the result of a preliminary examination of hospital separations in Ontario, focusing on infants under one year of age and a diagnosis of pulmonary hemosiderosis. The project was designed to be completed in a timely and cost effective manner.

Ontario hospital separations from 1991-1995 were screened for infants admitted to hospitals with the ICD-9 codes of 516.1 (idiopathic pulmonary hemosiderosis) and/or 275.0 (disease of iron metabolism). A potential case was an infant who had no known secondary cause of pulmonary hemorrhage/hemosiderosis after review of other diagnoses in the separation date file.

The results demonstrate that an unusual number of cases (10) occurred in the 1991 fiscal year while none occurred in the other years reviewed in this study. We believe that the Ontario hospital separations data base should be more carefully examined using the information gained in the initial survey. A geographic analysis of cases with an extension of the search for other ICD codes for pulmonary disease in infants may support a more intensive research project examining the environmental causes of pulmonary hemorrhage/hemosiderosis in Ontario infants. Should this association be confirmed, information about the prevention of exposure of children to moldy homes will require more aggressive dissemination by public health and other agencies.

Introduction

In December, 1994 the CDC reported a cluster of eight cases of idiopathic pulmonary hemosiderosis diagnosed at the Rainbow Babies and Children s Hospital (RBCH) in Cleveland from January 1993-November 1994. During the preceding ten years only 3 cases of this disorder had been diagnosed at the same hospital.(1)

A case control study was performed using 10 case-infants and 30 age-matched controls from the same area in Cleveland(2,3). The study found a decreased incidence of breastfeeding (odds ratio [OR]=0.2; 95% confidence interval [CI]=0-1.2) and an increased incidence of smoking for the case-infants (OR=7.9; CI=0.9-70.6) which did not achieve a level of statistical significance. However, all ten of the case-infants had resided in homes where severe water damage had occurred in the 6 months prior to admission versus 7(23%) of the 30 controls (OR=16.3; CI=2.6-infinity). Case infants were also more likely to have had close relatives with pulmonary hemorrhage (OR=33.14; CI= 5.1-infinity). Lastly, 50% of the infants had recurrent pulmonary hemorrhage after returning to their homes.

Visual inspection of the children s homes, fungal quantification and fungal identification were then performed. There was a higher concentration of fungi in the case infants homes (OR=1.6; CI=1.0-30.8). Among the molds in the homes was Stachybotrys atra. It is a candidate causative organism for infant pulmonary hemorrhage because its toxins have been implicated in hemorrhagic disorders in animals. The study suggests that environmental risk factors may have contributed to pulmonary hemorrhage in the case-infants.

All of the ten infants presented with hemoptysis associated with pallor and an abrupt cessation of crying in the absence of fever. Extensive investigation failed to find a specific etiology for the hemoptysis and the diagnosis of idiopathic pulmonary hemosiderosis was therefore assigned.

TABLE 1

Causes of Hemoptysis in Children
Infection

Bacterial
Lung abscesses
Necrotizing pneumonia
Tuberculosis
Bronchiectasis (cystic fibrosis immune
deficiency, immotile cilia)

Fungal
Actinomycosis
Aspergillosis
Coccidioidomycosis
Histoplasmosis

Parasitic
Echinococcosis (hytadid disease)
Paragonimiasis
Strongyloidiasis
Foreign Bodies

Congenital Defects

Cardiovascular
Congenital heart defects
Absent pulmonary artery
Ateriovenous malformation
Hemangiomatous malformation
Telangiectasia (Osler-Weber-Rendu
Syndrome)

Other
Pulmonary sequestration
Bronchogenic cyst
Intrathoracic enteric cyst

Vasculitis
Periarteritis nodosa
Adapted from Rosenstein(4)

Autoimmune Disorders
Wegener's granulomatosis
Pulmonary hemosiderosis
Milk Allergy
Goodpasture's syndrome
Collagen vascular disease

Trauma
Compression or Crush Injury
latrogenic
Postsurgical
Diagnostic lung puncture
Transbroncial biopsy

Inhalation of Toxins

Endobronchial Metastases

Primary Lung Tumors
Benign (hamartoma, neurogenic tumors)
Malignant (bronchial adenoma,
bronchogenic carcinoma, pulmonary
blastoma)

Endometriosis

Drug Induced
Propylthiouracil

Pulmonary Embolism
Hemoglobinopathy With Pulmonary Infarct
Factitious

Hemoptysis, defined as the expectoration of blood or blood tinged sputum, is unusual in the pediatric age group. Massive hemoptysis, as occurred in the case-infants in Cleveland is even rarer. The most common cause of hemoptysis in children is infection, followed by aspiration of a foreign body (not necessarily for infants). Other important causes include cardiac malformations, vascular malformations, and autoimmune disorders(4). Pulmonary hemosiderosis is an unusual cause of infant pulmonary hemorrhage and is estimated to account for less than 5%(2) of infant cases. A detailed list of the causes of hemoptysis in children is given in Table 1.

Pulmonary hemosiderosis is a nonspecific finding that usually indicates previous hemorrhage or aspiration of blood. Essentially, blood iron phagocytosed by alveolar macrophages is stored within them as hemosiderin. The causes of pulmonary hemosiderosis are usually divided into secondary cardiac or systemic causes (listed in Table 1 for hemoptysis) and primary disorders intrinsic to the lung(5). In its primary form there is no demonstrable cause for pulmonary hemorrhage. Often primary hemosiderosis is divided into four variants (1) an isolated form; (2) a form associated with allergy to cow s milk; (3) a form associated with myocarditis or pancreatitis; and (4) a form associated with progressive glomerulonephritis (Goodpasture s syndrome)(6). Some others only classify the first form of the disease as idiopathic pulmonary hemosiderosis (IPH)(5) whereas others group all four into IPH(6). Either way, IPH is the most common type of pulmonary hemosiderosis. Generally, infants who were previously healthy and then have chronic recurring episodes of pulmonary hemorrhage of unknown cause are diagnosed as having IPH(3). Pulmonary hemosiderosis can present in one of two ways (1) fulminant onset with recurrent acute hemoptysis (as in the Cleveland cluster); or (2) insidious onset with anemia, pallor, weakness, and lethargy(6).

Cassimos et al. described 30 cases of IPH in Greece over a 20 year period(7). The cases occurred in household clusters where children were sleeping in rooms near stored grain. The authors suggested that the illnesses may have been associated with chronic pesticide exposure. More recently, the clusters in Cleveland suggest another environmental cause of IPH; exposure of susceptible infants to molds, specifically, Stachybotrys atra. A search of the Ontario hospital separation data base was done to explore in a timely and cost-effective manner if there are unrecognized clusters of IPH that may be due to an environmental exposure.

Methods

The Ontario hospital separations data base was searched for children under five years of age at the time of hospitalization with diagnosis of idiopathic pulmonary hemosiderosis (ICD-9 516.1) and/or disease of iron metabolism (ICD-9 275.0). The fiscal years 1991-1995 were searched.

These individual records were then examined to see if the cause of pulmonary hemosiderosis was attributable to a known cause of infant pulmonary hemosiderosis according to the exclusion criteria used by the CDC(8). Known causes of pulmonary hemorrhage/hemosiderosis in infants include bacterial pneumonia; septicemia; tuberculosis; bronchiectasis; pulmonary malignancy; Gaucher s disease; Wegener s granulomatosis; cardiac anomalies; and vascular anomalies. Separations of pulmonary hemosiderosis not meeting the exclusion criteria were counted as being potential cases.

Results

A total of 23 separations were found that had at least one of the two ICD-9 codes used in the search. There were no potential cases for the years 1992-1995. On average there were 2.25 separations found by the search ranging from 1 to 4 for any given year between 1992 and 1995. In 1991 there were 10 potential cases from a total of 14 separations identified in the search. Significantly, none of these potential cases reported pulmonary hemorrhage or hemoptysis. (Table 2).

Discussion

This data review was carried out following reports that an outbreak of infant pulmonary hemorrhage/hemosiderosis in Cleveland, OH was associated with water damage and high levels of the mold in the children s homes; Stachybotrys atra was one of the molds identified in the homes of affected children. Ontario has much of its population in cities and towns which, like Cleveland, borders one of the Great Lakes and has a similar climate. Therefore, it is quite possible the same environmental conditions are present in homes in Ontario where infants are found and causing pulmonary hemorrhage/hemosiderosis in a number of susceptible individuals.

This study was undertaken to explore that possibility. Screening the provincial hospital separation data for idiopathic pulmonary hemosiderosis can quickly and inexpensively determine if there was potentially a cluster of these cases. If that is the case, follow up with a geographic analysis and an examination of risk factors in each child can potentially discover if the case could be attributed to water damage of homes and/or Stacybotrys exposure. Given the small number of potential cases and the specificity of the diagnosis, the approach taken was deemed the most appropriate first step.

Table 2

Hospital Separations by Fiscal Year Children Under 5 Years with ICD-9 Codes 516.1/275.0

Fiscal Year Separations with Number with Number of ICD-9 code 516.1 or 275.0 Pulmonary Hemosiderosis Potential Cases

1991 14 10 10
1992 2 1 0
1993 2 0 0
1994 1 0 0
1995 4 1 0

Results of search of Ontario hospital's separation data. Children 5 years of age or younger who were diagnosed with ICD-9 codes 516.1 (idiopathic pulmonary hemorrhage) and/or 275.0 (Disease of iron metabolism) were screened from the data base. Potential cases were those children with pulmonary hemorrhage/hemosiderosis who had no known cause for this disorder based on the separation data.

The results suggest a time-limited cluster of cases of IPH 1991. Ten potential cases appear in this year versus none in the following four years. None of these potential cases is reported to have pulmonary hemorrhage or hemoptysis. It is not known if this is because this symptom is truly absent (insidious IPH) or not captured by the database (fulminant IPH). If these ten cases did not have hemoptysis, then they would be quite different from the Cleveland case-infants presentation. Does this mean a new form of the same phenomenon, a new phenomenon, or merely coincidence?

As a follow up to this study, we will examine the hospital separation data base further by extending the number of years searched (1988-1997), by including several other ICD codes which may capture the diagnosis of IPH, and by limiting the search to infants under 12 months of age. A broader search of the database extending the years and the separation diagnostic codes is needed to clarify the preliminary results obtained here. The outcome of this second search may warrant a more detailed examination of the situation (such as a chart review) and epidemiologic study (such as a case-control study) which could provide more definitive evidence for this rare disease and novel association.

Source

Howard Shapiro, MD
Community Medicine Resident
University of Toronto

Contact

Lesbia F. Smith, MD
Senior Medical Consultant
Environmental Health and Toxicology Unit
Population Health Service
Public Health Branch
smith99@mail.moh.gov.on.ca

References

1. CDC. Acute pulmonary hemorrhage/hemosiderosis among infants- Cleveland, January 1993-November 1994; MMWR 1994; 43(2) 881-883.

2. CDC. Update pulmonary hemorrhage/hemosiderosis among infants-Cleveland, Ohio, 1993-1996. MMWR 1997; 46(2) 33-35.

3. Montana E, Etzel RA, Allan T, Horgan TE, Dearborn, DG. Cleveland, Ohio, 1993-1996. MMWR 1997; 46(2) 33-35.

4. Montana E, Etzel RA, Allan T, Horgan TE, Dearborn DG. Environmental risk factors associated with pediatric idiopathic pulmonary hemorrhage and hemosiderosis in a Cleveland community Pediatrics 1997; 99(1) e5. World Wide Web site http//pediatrics.org/cgi/content/99/1/e5

5. Rosenstein BR. Hemoptysis. In Hilman, B.C. ed. Pediatric Respiratory Disease Diagnosis and Treatment. Philadelphia, PA WB Saunders Co; 1993 533-543.

6. Cutz E. Idiopathic pulmonary hemosiderosis and related disorders in infancy and childhood. Perspect. Pediatr. Pathol. 1987; 11 47-81.

7. Levy J. and Wilmott, R. Pulmonary Hemosiderosis. In Hilman, B.C. ed. Pediatric Respiratory Disease Diagnosis and Treatment. Philadelphia, PA WB Saunders Co; 1993 543-549.

8. Cassimos CD, Chryssanthopoulos C, Panagiotidou C. Epidemiologic observations in idiopathic pulmonary hemosiderosis. J. Pediatr. 1983 102(5) 698-702.

9. CDC. CDC study of acute pulmonary hemorrhage/hemosiderosis in infants. World Wide Web site http//p e d s c c m . w u s t l . e d u / R E S E A R C H / Pulmonary_hemorrhage.html

This article appeared in the Public Health & Epidemiology Report Ontario (PHERO) Newsletter, Volume 9, Number 3, March 27, 1998. PHERO is published monthly by the Public Health Branch, Ministry of Health, North York. Telephone 416 327-7623.

Population Health Service Comment

Mold growth indoors continues to be an issue of growing concern because of its potential relationship to adverse health effects. In the past year, several instances of large-scale mold contamination in buildings have come to the attention of public health departments. These have always occurred with flooding inside the building or breaching of walls from the outside. In the aftermaths of recent weather related disasters (i.e. flooding, ice storm), there are likely to be many instances of premises developing mold problems. Stachybotrys atra has received quite a bit of attention in the media. However, as is pointed out in the article above, it may not be the cause of any particular health problem if it is found in a premise. We highly recommend that staff dealing with this issue review the articles referenced in this report, in particular, those by Montana and Etzel, which clearly describe the relationships found between dampness, mold and infant illness. In the light of these reports and the media attention which the mold issue has received, public health staff can anticipate increased community concern and demand for information.

Three publications provided by the Public Health Branch to public health departments will be useful in anticipating problems and responding to concerns The Expert Panel on the Health Risks of Toxigenic Molds (PHERO 1997), Indoor Air Quality Resource Document, and the CMHC booklet, Clean-up Procedures for Mold in Homes. Public Health documents can be forwarded to you electronically by contacting the branch directly at the e-mail addresses listed with the contacts.

Sources and Contact

Patricia Powell, MEd, CIPHI
Senior Inspection Consultant
Environmental Health and Toxicology Unit
Population Health Service
Public Health Branch
powell99@mail1.moh.gov.on.ca

This article appeared in the Public Health & Epidemiology Report Ontario (PHERO) Newsletter, Volume 9, Number 3, March 27, 1998. PHERO is published monthly by the Public Health Branch, Ministry of Health, North York. Telephone 416 327-7623.