Symptoms of mothers and infants related to total volatile organic compounds in household products
VOLATILE ORGANIC COMPOUNDS (VOCs) are a diverse group of relatively low-molecular-weight compounds with lower- and upper-limit boiling points of 50-100 [degrees]C and 240-260 [degrees]C, respectively. These solvents are all liquids at room temperature and are strongly lipophilic. VOC concentrations in indoor air are dependent on their sources (i.e., from combustion processes or from synthetic organic compounds used in buildings, furnishings, and household products where they may be released in low concentrations as active ingredients or contaminants). Common household products that contain VOCs include air fresheners, deodorizers, insect sprays, cleaners, polishes (1) (e.g., furniture polish, floor polish), and personal care products such as hairspray. Decreased ventilation in energy-efficient homes may increase VOC concentrations, whereas sink effects may contribute to their reduction. (2) Simulation of indoor insecticide application has shown indoor levels to be within the range expected to produce health complaints.
Health symptoms attributed to VOCs include a panoply of responses (e.g., unwanted sensory effects such as unpleasant smells and sensory irritation, mucosal irritation resulting from toxic effects that interfere with the normal functioning of organs or tissues, and genotoxic effects). Researchers have investigated these health symptoms in different samples and populations; objectives have varied and methods have differed. (4) The effects of VOCs may not be immediate and acute, but may occur following cumulative doses. Odor sensation may lead to secondary outcomes (e.g., vomiting, altered respiratory response). Symptoms of indoor exposure to VOCs are often similar to those reported by individuals with Sick Building Syndrome. (5) In studies of the latter, investigators typically have found VOC levels to be below threshold limit values established for occupational settings. (6-8) Occupational standards–set at levels 10-1,000 times lower than the expected no-effect levels in humans–may be inappropriate for household environments, in which the exposure period may be more protracted than in the work setting. The home environment may contain an unknown mixture of chemicals that can interact synergistically at various air temperatures and humidity levels to exacerbate health effects from exposure to a single compound. (9-11) Health effects may result from levels lower than would be expected from exposure to a single compound; concentration gradients, rather than total concentrations, may potentially affect health adversely.
Non-occupational exposure limits are relevant for vulnerable infants, the elderly (> 65 yr of age) and ill individuals who spend most of their time indoors. These populations may spend their greatest number of indoor hours in the bedroom, where there is little air exchange. Pregnant women and their infants [less than or equal to] 6 mo of age spend 77% and 81%, respectively, of their time in their home. (12) The pregnant woman and her fetus may be highly vulnerable in that a rise in plasma lipids during pregnancy may enhance the solubility of inhaled lipophilic substances, such as organic solvents. (13-16)
Method
The Avon Longitudinal Study of Parents and Children (ALSPAC) is a population-based study of children born to women who resided in Avon (United Kingdom) during their pregnancy and who had an expected delivery date between April 1, 1991, and December 31, 1992. (17) There were 14,541 pregnant women enrolled in this study, and a cohort of 13,971 of their children was still being followed at age 12 mo. The goal of the ALSPAC is to evaluate environmental, genetic, and social factors that can influence the health of infants and their mothers. Information was collected from mothers through self-report questionnaires at different times during their pregnancy, as well as after the infant’s birth, to ascertain family and household characteristics, parental occupations, and other socioeconomic factors. The purpose of this study within the ALSPAC was (a) to determine indoor levels of VOCs relative to the use of specific household products and (b) to identify households in which total VOC (TVOC) levels were high. Investigation of the entire cohort of children and their parents further identified common health effects at different points of data collection.
The Department of the Environment (U.K.) commissioned Building Research Establishment Ltd (BRE [Watford, U.K.]) to monitor the levels of VOCs within 170 homes randomly selected from the total ALSPAC cohort. Monitoring began in November 1990, and approximately 10 new homes/mo were enrolled during the study period. (11) The home of each pregnant woman was monitored from approximately 6 mo gestation to 6 mo following birth. (12) The 1st birth in the cohort was in April 1991. VOCs were monitored with TENAX[TM] tubes (TENAX Corp. [United Kingdom]), which were exposed for more than 1 mo for each of the 12 mo of sampling in the main bedroom and living room of each home in the study. The returned sampling tubes were stored in 3plastic bags at room temperature and were analyzed within 4 wk of collection of the tubes. Sampling tubes exposed for fewer than 21 days or for more than 42 days were rejected. For each batch of 10 tubes, 3 were retained at room temperature as blanks–without removal of their caps. (11,18) Following thermal desorption, chemical analysis of the sampler tubes was achieved with gas chromatography, which measured VOCs and specific compounds, including those commonly found in indoor air (i.e., benzene, toluene, m-xylene, o-xylene, decane, 1,2,4-trimethylbenzene, and undecane). We used toluene levels to calculate the TVOC concentration (i.e., a summation of individual compounds [mainly [C.sub.6]-[C.sub.17] compounds]). (19)
We asked subjects to complete a questionnaire that had questions about the frequency of use of 9 common household products that contain high proportions of VOCs. A total of 13,164 women completed the 1st questionnaire when they were 8 wk pregnant. Of these women, 10,976 completed a 2nd questionnaire 8 mo after birth, and 10,119 completed a 3rd questionnaire when their child was 21 mo of age. We assumed that information about household product use during early pregnancy reflected routine use of these products–rather than later uses which might include cleaning that occurred because the infant was now a member of the household (e.g., use of products to ensure special cleanliness in the infant’s environment). The types of household products examined were window cleaners, carpet cleaners, dry-cleaning fluids, turpentine or white spirit, paint stripper, house paints or varnishes, pesticides, other aerosols or sprays, and air fresheners. The categories of use were (a) never or less than once per week, (b) once per week, and (c) daily on most days.
Infant symptoms and health problems since birth (i.e., diarrhea, vomiting, cough or cold, high temperatures, earache, wheezing, breathlessness, rashes, and hospital admissions) were gleaned from the mother via a structured questionnaire at 8 mo postpartum. Respondents were also asked to report their own symptoms (e.g., anxiety, headache/migraine headache, cough or cold, influenza, wheezing, eczema, urinary tract infection, nausea, vomiting, diarrhea). Depression (during the week prior to completing the questionnaire) was also assessed with the Edinburgh Postnatal Depression Score (EPDS)–a 10-item scale for which scores of 13 or more are strongly predictive of clinical depression. (20) Women who scored a standard depression screening measurement (21) greater than 9 were defined as depressed for the purposes of this study.
Statistical analyses. Mean TVOC levels were calculated on the basis of the monthly values from the living rooms and main bedrooms of the homes monitored in the BRE indoor air study (N = 170). Households with less than 5 TVOC readings for the year were excluded from the analysis. TVOC levels were dichotomized into 2 percentiles: < 75th percentile and [greater than or equal to] 75th percentile. Use of each of the 9 household products during early pregnancy was dichotomized to < 1/wk and [greater than or equal to] 1/wk. We used Pearson’s chi-square and Fisher’s Exact test (crosstabs) to evaluate the relationships between VOC levels in the homes and product use during early pregnancy. We then used products that were statistically significantly associated with higher TVOC levels in the analysis of the entire cohort to determine if use of these products was associated with reporting of symptoms for infants or mothers.
For the total cohort, we applied logistic-regression analysis to obtain adjusted odds ratios (ORs) for each symptom with use of a specific product for different frequencies of use, to determine if the odds of experiencing a symptom increased as use of the product increased. Adjustments were made for education, mother’s age, housing tenure, number of children in the home, number of smokers in the home, paid job subsequent to birth of the child, dampness or condensation in the home, mold in the home, type of winter heating fuel, and month the questionnaire was completed. The first 6 variables controlled for socioeconomic status; the latter 4 controlled for seasonal ventilation differences that might have influenced the build-up of VOCs.
In this study, we found that the use of air fresheners and aerosols was significantly associated with certain symptoms in mothers and their infants. Both products were also associated with increased levels of VOCs in the home. We examined symptoms associated with the use of either or both products, rather than with individual VOC exposures, thereby reducing to some degree the problem of multiple comparisons. If use of a specific product was associated with the same symptom across time periods or across populations, the findings would lend credibility to the existence of a causal association between use of that product and the symptom. Causal effects were also likely to occur if similar associations were found across products (air fresheners and aerosols).
Our results showed a statistically significant association between product use and infant diarrhea and maternal headache; the association with maternal headache was found at 2 points in time and for each product (Table 4). Whereas infant symptoms were associated with the use of these products, the symptoms were not validated by physical findings (e.g., ear infection with reporting of infant earache) or by laboratory tests (analysis of blood or urine for breakdown products from use of aerosols or air fresheners). (22) Vomiting and diarrhea in the infant may be a response to unfamiliar chemicals, but these symptoms may also be associated with exposure to unknown factors. For maternal symptoms of headache or migraine headache, there is no better procedure than self-report, whereas depression was measured using a validated scale.
Measured levels of TVOCs in the homes of subjects were well below established no-effect levels for acute symptoms in humans. (23) Most of these levels were set for occupational settings and with adults in mind and may, therefore, not be applicable to our study populations. The health effects on infants exposed to low levels of VOCs for protracted time periods are difficult or impossible to estimate–even in an experimental setting. Women may be more receptive physiologically to the effects of solvents than men, given that women have a larger percentage of body fat. This fact may be particularly important after childbirth. There may also be an additive or synergistic effect of exposure to the complex mixture of unknown chemicals in the home environment. Air fresheners, for example, often contain limonene–a terpene added to many household products to improve fragrance. Limonene reportedly acts synergistically with ethyl alcohol on the skin permeability of mice. (24,25)
We have been unable to find reports of studies on the effects of VOCs in infants, although there are some studies of short-term effects in adults. For example, a population of healthy nonsmoking adults revealed significant exposure effects for odor intensity, eye and throat irritation, headache, and drowsiness after exposure to a mixture of VOCs at 15- to 30-min intervals. (26) Consumer products that provide fragrance have shown toxic properties following exposure to relatively high levels of polar VOCs, but the literature is unclear with respect to low-level chronic exposure to these substances. (27) Perhaps reactions to air fresheners and aerosols may constitute a specific syndrome. (28,29) For example, bronchial hyperresponsiveness has been associated with indoor limonene concentrations. (30).