Published ahead of print on April 30, 2003, doi:10.1164/rccm.200207-730OC
© 2003 American Thoracic Society
Fungal Levels in the Home and Lower Respiratory Tract Illnesses in the First Year of LifeDepartment of Environmental Health, Harvard University School of Public Health; Biostatistics Research Center, Division of Clinical Care Research, Department of Medicine, Tufts-New England Medical Center; Department of Biostatistics, Harvard University School of Public Health; and Channing Laboratory, Department of Medicine, Brigham and Women's Hospital and the Harvard Medical School, Boston, Massachusetts Correspondence and requests for reprints should be addressed to Diane R. Gold, M.D., M.P.H., Channing Laboratory, 181 Longwood Avenue, Boston, MA 02115. E-mail: diane.gold{at}channing.harvard.edu
The association between home dampness and lower respiratory symptoms in children has been well documented. Whether fungal exposures contribute to this association is uncertain. In a prospective birth cohort of 499 children of parents with asthma/allergies, we examined in-home fungal concentrations as predictors of lower respiratory illnesses (LRI) (croup, pneumonia, bronchitis, and bronchiolitis) in the first year. In multivariate analyses, we found a significant increased relative risk (RR) between LRI and high levels (more than the 90th percentile) of airborne Penicillium (RR = 1.73, 95% confidence interval [CI], 1.23, 2.43), dust-borne Cladosporium (RR = 1.52; CI, 1.02, 2.25), Zygomycetes (RR = 1.96; CI, 1.35, 2.83), and Alternaria (RR = 1.51; CI, 1.00, 2.28), after controlling for sex, presence of water damage or visible mold/mildew, born in winter, breastfeeding, and being exposed to other children through siblings. In a multivariate analysis, the RR of LRI was elevated in households with any fungal level at more than the 90th percentile (RR = 1.86; CI, 1.21, 2.88). Exposure to high fungal levels increased the risk of LRI in infancy, even for infants with nonwheezing LRI. Actual mechanisms remain unknown, but fungi and their components (glucans, mycotoxins, and proteins) may increase the risk of LRI by acting as irritants or through increasing susceptibility to infection.
Key Words: fungi • respiratory tract infections • infants • public health There is abundant literature discussing the relationship between home dampness and lower respiratory symptoms in early life (1–5). However, it is unknown whether this association is partly related to exposure to fungi, which thrive in damp conditions. Several studies have shown that the presence of visible mold growth is related to bronchitis, phlegm production, and chest illnesses (6–8). These studies relied on self-reported assessment of mold growth. There have been few attempts to quantify the relationship between fungal concentrations and the development of lower respiratory tract illnesses (LRIs) in the first year of life when children are in the process of developing immunocompetence. Therefore, we have explored whether exposure to fungi and fungal products, and not just dampness, is associated with LRIs early in life. Determining risk factors for LRIs in infants and very young children is important because both wheezing and nonwheezing LRIs in early years have been shown to be associated with significant morbidity later in life (9–12).
Study Protocol Participants were part of a metropolitan Boston prospective birth cohort designed to examine relationships between indoor allergens and the development of allergic sensitization and asthma. The screening and recruitment of families have been described elsewhere (13). In brief, with the approval of the Human Studies Committee at Brigham and Women's Hospital (Boston, MA), we enrolled 505 infants, or index children, from 499 families (six sets of twins) with a history of asthma or allergy in at least one parent. Consent was obtained to extract data (e.g., maternal age and child's birth weight) from the labor, delivery, and newborn nursery records. Of the 1,405 families initially screened, 906 were excluded from the study before the first home visit. Reasons for exclusion were as follows: reluctance to participate in a longitudinal study (51% of those who refused), plans to move within 1 year (39%), early loss to follow-up (9%), and other (1%). Mothers gave consent for screening at the time of the first home visit when the index child was 2 or 3 months of age. During this first home visit, a trained technician visited the child's home to obtain household and socioecomonic characteristics and to conduct air and dust sampling. Every 2 months, beginning when the child was 2 months of age, a follow-up telephone questionnaire was administered to the child's primary caregiver. Information was recorded on respiratory symptoms and illnesses experienced by the index child since the previous telephone interview, daycare attendance, and selected home characteristics. During the home visit, indoor air and dust samples were collected from the bedroom where the child usually slept. Indoor air samples were collected from each home using a Burkard culture plate sampler operated at 45 L/minute (calculated cutpoint = 2 µm) that collected particles onto Dichloran glycerol (DG18) agar in 90-mm petri dishes. Sequential duplicate air samples were collected in the bedroom 1–1.5 m above the area of the floor demarcated for dust collection. After air samples were collected, 2 m2 of the floor surrounding the newborn's bed was vacuumed for 5 minutes using a Eureka Mighty-Mite canister vacuum cleaner (The Eureka Co., Bloomington, IN) modified to collect dust in a 19- x 90-cm cellulose extraction thimble (Whatman International, Ltd., Herefordshire, UK). In cases in which both a smooth floor and a rug were present, 2.5 minutes were devoted to sampling the rug, and 2.5 minutes were spent vacuuming the smooth floor surface (14, 15). After at least 10 days of incubation, fungal colonies were identified using standard mycologic criteria. The number of colonies recovered on the air sample plates was adjusted for multiple impactions (16). The calculated concentrations of dust-borne fungi were the number of cfu per gram of dust and those of airborne fungi were cfu/m3 of air. The number of fungal colonies was determined only if sufficient dust was also available for allergen measurement. As a result, there are data on fungal levels from the child's bedroom floor for 419 homes. Analyses dealing only with dust-borne fungal levels are limited to these 419 children.
Definition of Fungal Predictors
Definition of Other Predictors
Definition of Outcome Variable A child was considered to have a wheezing LRI if any self-report of wheezing occurred before or concurrent with the doctor-diagnosed LRI. If no report of wheezing occurred, the LRI was classified as non-wheezing.
Statistical Analysis
Cohort Characteristics as Predictors of LRI Of the remaining 499 children included in the analysis, 367 (74%) had no doctor-diagnosed LRI. One hundred seven (21%) had a report of LRI in one questionnaire. Twenty (4%) reported LRI in two questionnaires, and five (1%) reported LRI in three or more questionnaires. We first identified potential independent predictors of LRI or potential confounders of the relationship between fungi and LRI. Tables 1–3 summarize familial and perinatal factors, parental and socioeconomic factors, and household and behavioral factors. In univariate analyses, males had an increased risk of LRI, whereas being born in the winter was inversely associated with LRI. Although we previously found low birth weight to be predictive of increased risk of wheezing (13), it was not associated with LRI. Breastfeeding was inversely associated with LRI. Factors such as maternal smoking during pregnancy, which we previously found to predict wheeze (18), and low income did not predict higher risk of LRI. Living in a single- or two-family home was marginally associated with an increased univariate risk of LRI. Exposure to other children, by the presence of siblings, was associated with an increased risk of LRI, as was water damage in the home or the presence of visible mold or mildew.
Description of Fungal Data Table 4 illustrates the distributions of the fungal data considered in the analysis. For each taxon, we report the number of houses from which the taxon was cultured. In addition, the 10th and 90th concentration percentiles are given, as well as the mean and median amount cultured for each taxon. Penicillium was the most common taxon sampled from the air, but Cladosporium was the taxon that had the highest levels recovered from the air. From the dust samples, Aspergillus was most commonly recovered, but yeasts were found in the highest levels.
Having high levels of one fungus did not necessarily imply that the household would have high levels of another fungus (Table 5) . For example, there were 44 homes with high levels of dust-borne Alternaria, but only 14 of those homes also had high levels of dust-borne Cladosporium.
Fungal Exposure as Predictor of LRI Table 6 illustrates the relationship between LRI and the specific airborne and dust-borne fungal concentrations in the house. In adjusted analyses, the fungi that remained significantly associated with LRI, even after controlling for other factors associated with doctor-diagnosed LRIs, were airborne Penicillium (RR = 1.73; 95% confidence interval [CI], 1.23, 2.43), dust-borne Cladosporium (RR = 1.52; CI, 1.02, 2.25), dust-borne Zygomycetes (RR = 1.96; CI, 1.35, 2.83), and dust-borne Alternaria (RR = 1.51; CI, 1.00, 2.28). The factors adjusted for included male sex, born in winter, breastfeeding, having siblings, and presence of water damage or visible mold/mildew in the home. The results for Alternaria and Zygomycetes were stronger for LRI with wheezing than for LRI without wheezing, but the results for Penicillium were stronger for nonwheezing LRIs.
In Table 7 , we see the effect of being in an environment where high fungal levels exist on the prevalence of at least one LRI in the first year. There were 324 homes that had high levels of at least one type of fungi. Children in environments with more than the 90th percentile for one or more taxa had an 86% increased risk of developing an LRI in the first year (RR = 1.86; CI, 1.21, 2.88). In addition, the effect of high household fungal levels was larger and stronger when focusing on nonwheezing LRIs compared with wheezing LRI (RR = 3.88; CI, 1.43, 10.52 versus RR = 1.58; CI, 0.95, 2.64).
High fungal levels and the presence of mold, mildew, or water damage were independent predictors of LRI in the first year of life. High fungal levels were highly statistically significant, whereas the variable representing the presence of mold, mildew, or water damage exhibited a strong trend (p = 0.06). The two variables, however, were virtually uncorrelated (R = 0.08). Of the houses with water damage or visible mold or mildew, 75.4% had high fungal levels; of the houses without water damage or visible mold or mildew, 67.9% had high fungal levels.
Fungal Exposure and Related Health Effects Fungal exposure, even to one type of fungus, is complex. Exposure often includes allergens, irritants, toxins, and sometimes even potentially infectious organisms (19). People are routinely exposed to more than 200 different species of fungi. Exposure occurs universally and is impossible to avoid completely. Often there are no adverse effects from these exposures, but at times, exposure to fungi can directly and indirectly influence an individual's health. Recently, there has been a significant focus on health effects associated with fungal exposure. Several studies have shown that the presence of visible mold growth is related to bronchitis, phlegm production, and chest illnesses (6–8). These studies relied on self-reported assessment of mold growth. Relationships between exposure to fungi, such as Alternaria, Aspergillus, Penicillium, and Trichoderma, and asthma exacerbation and allergic rhinitis in children (20, 21) have been suggested. In addition, the mycotoxins produced by the fungus Stachybotrys chartarum have been shown to produce bronchiolar inflammation and hemorrhage in laboratory mice (22). Stachybotrys was recovered very rarely in our homes, both because it is rare in homes without water damage and because the culture media used (DG18) does not support its growth well.
Household Fungal Levels as a Predictor of LRI The results for Alternaria and Zygomycetes were stronger for LRI with wheezing than for LRI without wheezing, but the results for Penicillium and the composite were stronger for nonwheezing LRIs. Although children who have repeated wheeze in early infancy have a greater risk of allergic asthma in later childhood than those without wheeze, for many children, early wheeze in infancy may represent nonallergic inflammation in small airways. Sensitivity to inhaled allergens, including mold, as measured by skin testing or by radioallergosorbent test, is uncommon in infancy. Although children who repeatedly wheeze with LRI in infancy are more likely to develop allergic asthma, many wheezing LRIs are related to nonallergic inflammatory responses in small airways. In infancy, nonwheezing LRIs are even less likely than wheezy LRIs to be allergy related. The strong association of higher household fungal levels with nonwheezing LRIs suggests that for many children the mechanism leading to this association in infancy is likely to be nonallergic.
Exposure to Fungal Components Mycotoxins are by-products of fungal metabolic processes. They accumulate in fungal spores, mycelia, and growth substrates (28), and inhalation exposure occurs in particulate form. It is well established that the ingestion of mycotoxins can cause deleterious health effects. However, much less is known about the effect of inhalation. Many mycotoxins have been shown to interfere with macrophage functioning, which is an important part of the defense system of the lung (29, 30). It may be the case that mycotoxins in the low concentrations likely to be present indoors can have some irritant effect in infants, such as inflammation, that would increase the likelihood of contracting a symptomatic LRI in the first year of life. To digest foods, fungi excrete enzymes into the environment to break down complex carbon compounds. These enzymes are some of the major fungal allergens (19). It may be that these enzymes, which can elicit an allergic response in susceptible individuals with fully developed immune systems, can cause a physiologic response in infants that will make them more susceptible to contracting a LRI. Alternaria alternata extracts have been shown to induce cell shrinking and cell desquamation in adults (31).
Potential Study Limitations
Conclusion Finally, our study suggests that exposure to high fungal levels can increase the risk of nonwheezing as well as wheezing LRI during infancy. The mechanisms for these associations may be, in great part, nonallergic.
Supported by grant AI/EHS35786 from the National Institutes of Health and the Environmental Science and Engineering Program through unrestricted departmental funds and through the Yamaguchi Endowment (P.C.S.). Received in original form July 22, 2002; accepted in final form April 26, 2003
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
METHODS
0.05) fungal taxa and significant independent predictors of LRI. Relative risks (RRs) were computed from PROC GENMOD using the log link and exponentiating the parameter estimates. Multivariate models were independently validated using the lasso approach for model shrinkage described by Tibshirani (
3)-ß-D-glucan, induces inflammation. In addition, it appears that (1 
