Independent effects of intestinal parasite infection and domestic allergen exposure on risk of wheeze in Ethiopia: a nested case-control study

BACKGROUND

Why asthma is rare in rural subsistence societies is not clear. We tested the hypotheses that the risk of asthma is reduced by intestinal parasites or hepatitis A infection, and increased by exposure to dust-mite allergen or organophosphorus insecticides in urban and rural areas of Jimma, Ethiopia.

METHODS

From 12876 individuals who took part in a study of asthma and atopy in urban and rural Jimma in 1996, we identified all who reported wheeze in the previous 12 months, and a random subsample of controls. In 1999, we assessed parasites in faecal samples, Der p 1 levels in bedding, hepatitis A antibodies, serum cholinesterase (a marker of organophosphorus exposure), total and specific serum IgE, and skin sensitisation to Dermatophagoides pteronyssinus in 205 cases and 399 controls aged over 16 years. The effects of parasitosis, Der p 1 level, hepatitis A seropositivity, and cholinesterase concentration on risk of wheeze, and the role of IgE and skin sensitisation in these associations, were analysed by multiple logistic regression.

FINDINGS

The risk of wheeze was independently reduced by hookworm infection by an odds ratio of 0.48 (95% CI 0.24-0.93, p=0.03), increased in relation to Der p 1 level (odds ratio per quartile 1.26 [1.00-1.59], p=0.05), and was unrelated to hepatitis A seropositivity or cholinesterase concentration. In the urban population, D pteronyssinus skin sensitisation was more strongly related to wheeze (9.45 [5.03-17.75]) than in the rural areas (1.95 [0.58-6.61], p for interaction=0.017), where D pteronyssinus sensitisation was common, but unrelated to wheeze in the presence of high-intensity parasite infection.

INTERPRETATION

High degrees of parasite infection might prevent asthma symptoms in atopic individuals.

Decreased prevalence of sensitization to cats with high exposure to cat allergen

We investigated the relationship between current exposure to cat allergen and sensitization to cats. A questionnaire was administered and skin prick testing and home visits for collection of dust samples (Fel d 1; ELISA) were performed in 2502 adults (mean age, 31.8 years; age range, 18-58 years; 1251 women). The results for Fel d 1 in relation to sensitization to cats were analyzed for 10 deciles of cat allergen exposure (cut points [microg/g]: 0.05, 0.34, 0.48, 0.72, 1.13, 1.92, 7.2, 44, 151). The prevalence of sensitization to cat was significantly decreased in the lowest and the highest exposure groups. In the multivariate regression analysis (age, sex, socioeconomic status, and current smoking being adjusted for), the risk of sensitization to cats was significantly increased with medium exposure to Fel d 1 (3rd centile, OR 2.3, 95% CI 1.2-4.4, P =.01; 4th centile, OR 2.1, 95% CI 1.1-4.0, P =.03; 5th centile, OR 2.2, 95% CI 1.2-4.3, P =.04, 6th centile, OR 2.5, 95% CI 1.3-4.9, P =.005). These results indicate that the prevalence of sensitization to cat is decreased in the lowest and highest cat allergen exposure groups.

The role of indoor allergen exposure in the development of sensitization and asthma

The role of indoor allergen exposure in the development of sensitization and asthma remains a subject of controversy. From a number of cross-sectional and longitudinal studies we can conclude that there is a very close association between allergen exposure and the sensitization of an individual. The dose-response relationships seem to differ between allergens; house dust mite and cockroach allergens appear to have a positive linear relationship, whereas cat allergens appear to act quite differently, with maximum sensitization developing at moderate exposure levels. Very low levels of cat allergen exposure are likely to induce no response and very high levels are likely to develop a form of tolerance, with a modified T helper cell type 2 response and the production of IgG4 antibodies and but not IgE. The relationship between indoor allergen exposure and asthma is, however, less clear. The proposed mechanism for the development of disease is that allergen exposure causes sensitization, and continued exposure leads to airway responsiveness and inflammation. As yet, the evidence for allergen exposure being a primary cause of asthma remains weak, and the results of ongoing prospective, randomized allergen avoidance trials are awaited to clarify this issue.

Repeat tuberculin testing in BCG-vaccinated subjects in the United Kingdom. The booster effect varies with the time of reading

The booster effect varies between populations, but has not been studied in the UK. The aim of this study was to investigate the effect of repeat tuberculin tests at 1 wk in BCG-vaccinated healthy subjects (all hospital employees) in the UK; we have assessed whether a booster effect is present 48 and 72 h after injection. Twenty-six subjects received two tuberculin tests (both 10 units) administered by the Mantoux technique-Tests 1 and 2. At Test 2 there was a significant increase in induration at 48 h (mean, 7.8 mm; p < 0.001), no difference at 72 h (mean, 0.2 mm; p = 0.93), and a reduction at 96 h (mean, -4.2 mm; p = 0.02). There were more positive results (> 15 mm induration) at Test 2 compared with Test 1 at 48 h (19 vs. 9, respectively; p = 0.002), but similar numbers at 72 h (11 vs. 10, respectively). These results show that repeat tuberculin testing in this BCG-vaccinated population caused a booster effect that varied with the time of measurement and was maximal at 48 h.

Risk factors for asthma in urban Ghana

BACKGROUND

Asthma is increasing in prevalence and severity in Africa. Previous studies have suggested that the prevalence of atopy in West Africa was low.

OBJECTIVE

We sought to investigate the risk factors for asthma in Ghanaian school children.

METHODS

Fifty children (age range, 9-16 years) with a physician diagnosis of asthma and asthma symptoms within the previous 12 months and 50 age- and sex-matched healthy control subjects were randomly selected and evaluated by means of questionnaire, skin testing, total and specific IgE measurements, and allergen level measurements from bed dust samples (mite, cat, dog, and cockroach).

RESULTS

Asthmatic children were exposed to higher levels of mite allergens than were control children (geometric mean, 19 microg/g [95% CI, 13.6-26.5] vs. 11.2 microg/g [7.4-15.7]; P <.05). Cat and dog allergen levels were low. There was a marked dissociation between skin test responses and the presence of specific IgE to cat and dog (CAP method). However, 84% of subjects with positive cat dander-specific IgE levels in cat CAP tests and negative skin test responses did not have Fel d 1-specific IgE (chimeric ELISA). In the univariate analysis significant associations with the patient group were found for sensitization to mite (odds ratio [OR], 9.3; 95% CI, 3.7-23.4) and cockroach (OR, 3.9; 95% CI, 1.3-11.6), inner-city residence (OR, 3.5; 95% CI, 1.4-8.9), asthma in family member (OR, 3.5; 95% CI, 1.4-9.0), low (<5) position in sibship (OR, 3.6; 95% CI, 1.2-11), presence of smoker in home (OR, 3.7; 95% CI, 1.2-11.9), small household size (OR, 0.57; 95% CI, 0.35-0.93), and use of electricity as domestic fuel (OR, 0.34; 95% CI, 0.12-0.97). In the multivariate analysis sensitization to mites remained the strongest risk factor associated with the asthmatic group (OR, 10.4; 95% CI, 3.5-30.9). The other significant associations were inner-city residence (OR, 4.8; 95% CI, 1.5-5.2), sensitization to cockroach (OR, 4.9; 95% CI, 1.3-18.6), and position in sibship of less than 5 (OR, 6.3; 95% CI, 1.3-29.4).

CONCLUSION

Sensitization to dust mite and cockroach allergens, inner-city residence, and low position in sibship were independent risk factors for asthma in Ghanaian children.

Pet allergen levels in homes in Ghana and the United Kingdom

The frequency of cat and/or dog ownership in Ghana is comparable to that in the United Kingdom (approximately 50%). However, in Ghanaian communities pets are predominantly kept outdoors. Levels of pet allergens (Fel d 1 and Can f 1) in 100 Ghanaian homes (49 without pets) were compared with levels in 410 homes in the United Kingdom (258 without pets). Homes with pets in the United Kingdom contained much higher allergen levels than homes with pets in Ghana (for Fel d 1: mean difference, 275-fold; 95% CI, 129-fold to 594-fold; P <.0001; for Can f 1: mean difference, 75-fold; 95% CI, 33-fold to 169-fold; P <.0001). Homes without cats in the United Kingdom contained significantly higher levels of Fel d 1 than homes with cats in Ghana (mean difference, 3.7-fold; 95% CI, 2.0-fold to 7.2-fold; P <.0001). In the United Kingdom, homes with dogs contained 75-fold (95% CI, 47-fold to 139-fold) more Can f 1 than homes without dogs, whereas in Ghana, homes with dogs contained 3.1-fold (95% CI, 1.5-fold to 6.1-fold; P =.003) more Can f 1 than homes without dogs. In the United Kingdom, homes with cats contained 77-fold more Fel d 1 (95% CI, 46-fold to 129-fold; P <.0001) than homes without cats; there was no difference in cat allergen levels between homes with cats and homes without cats in Ghana. In conclusion, levels of pet allergens in Ghanaian homes with pets were (1) between 75-fold (dog) and 275-fold (cat) lower than levels in homes with pets in the United Kingdom and (2) lower than or comparable to levels in homes without pets in the United Kingdom.

Lymphoproliferative responses in cord blood and at one year: no evidence for the effect of in utero exposure to dust mite allergens

BACKGROUND

Maternal allergen exposure beyond the 22nd week of pregnancy may be important in foetal T cell priming. Allergen-specific cord blood mononuclear cell (CBMC) immunoproliferative responses without corresponding bacterial antigen responses (tetanus toxoid), have been suggested as evidence of in utero sensitization.

OBJECTIVES

To investigate the relationship between lymphoproliferative responses at birth and at 1 year with maternal and 1-year infants house dust mite allergen exposure.

METHODS

Home visits and dust sampling were performed by the 20th week of pregnancy, immediately after birth, and then at 1 years of age. Der p 1 was assayed using a two-site immunometric ELISA. CBMC immunoproliferative responses (AIM V serum-free medium; 1 x 105 cells/well) were measured for 225 neonates (171 had a high risk of atopy (HR)--both parents skin test positive; 59 had a low risk of atopy (LR) - both parents skin test negative, no history of atopy) by 3H-Thymidine (1microCi/well) incorporation after stimulation in primary culture with phytohaemagglutinin (PHA) (1 microg/mL), house dust mite [HDM] extract (30 microg/mL), immunopurified Der p 1 (30 microg/mL), Tetanus toxoid (TT) (aluma free, 30 Lf/mL) or vehicle. Blood was collected from 144 infants at the age of 1 years and stimulated proliferative responses were assessed using the same procedure.

RESULTS

PHA-stimulated lymphoproliferative response was significantly lower in HR compared to LR neonates (mean difference 38%, 95% CI 15%-54%; P = 0.003); significantly lower proportion of positive CBMC responses to HDM occurred in LR than in HR neonates (30.4% vs. 46.6%; P = 0.034). There was no relationship between Der p 1 levels in maternal bed and CBMC immunoproliferative responses, despite the 21 000-fold range of maternal Der p 1 exposure. No significant differences in magnitude, or in proportion of positive responses to any stimulant were observed between the neonates at low, medium or high tertile of allergen exposure. Immunoproliferative responses at birth were not predictive of 1-year PBMC responses. There was no relationship between maternal allergen exposure in pregnancy and 1-year PBMC proliferative responses. However, the proportion of positive proliferative responses at 1 years significantly increased with increasing infant Der p 1 exposure at 1 years.

CONCLUSION

These results indicate that the magnitude of immunoproliferative responses are unrelated to maternal mite allergen exposure and cannot be used as evidence for in utero sensitization to inhalant allergens. The immunoproliferative responses at 1 year seem to shift away from the genetically influenced responses at birth towards responses to specific stimulants which correlate with environmental exposure to those specific stimulants. These data support the concept of sensitization to inhalant allergens occurring in early life, but not in utero.

Effect of environmental manipulation in pregnancy and early life on respiratory symptoms and atopy during first year of life: a randomised trial

BACKGROUND

Asthma places huge demands on health-care services, and its prevalence is increasing. Reduction of exposure to environmental allergens could offer a realistic chance for primary prevention. Our aim was to ascertain whether or not living in a low-allergen environment reduces the risk of asthma and atopic diseases in infants.

METHODS

We assigned infants to four risk groups according to parental atopic status. We enrolled 291 high-risk couples (both parents atopic, no pets) into a prospective, prenatally randomised, cohort study, and allocated them to environmental manipulation, in which measures to reduce prenatal and postnatal allergen exposure were undertaken (active HRA) (n=145) or no intervention (control HRC) (n=146). Two further prospective groups were studied: 161 high-risk infants with pets in the home (HRP group) and 168 low-risk infants, whose parents were both non-atopic (LR group). The main outcome measures were signs and symptoms of atopic disease at 1 year of age.

FINDINGS

103 families dropped out or were lost to follow up. At age 1 year we followed-up 133 HRA, 118 HRC, 140 HRP, and 126 LR infants. Children in the HRA group were less likely to have respiratory symptoms during the first year of life than those in the HRC group. The most pronounced differences were in the relative risks for severe wheeze with shortness of breath (relative risk 0.44 [95% CI 0.20-1.00]), prescribed medication for the treatment of wheezy attacks (0.58 [0.36-0.95]), and wheezing after vigorous playing, crying, or exertion (0.18 [0.04-0.79]). Probability of respiratory symptoms in HRC and HRP infants was similar, whereas it was much lower in the LR than in the HRC group. Cat ownership was significantly associated with sensitisation to cats (24.6 [3.04-199.05]; p=0.003).

INTERPRETATION

Environmental manipulation reduces some respiratory symptoms in the first year of life in high-risk infants. Further follow up is needed, however, to ascertain whether living in a low-allergen environment reduces allergy and asthma in later life.

B-cell isotype control in atopy and asthma assessed with cDNA array technology

B-cell isotype switching and the production of IgE is regulated by a variety of gene products through different mechanisms. A better understanding of these processes has the potential to identify markers of disease and new therapeutic targets. The aim of the study was to investigate human B-cell isotype control and IgE production in atopy and asthma with cDNA array technology. Eighteen atopic asthmatic, eight atopic nonasthmatic, and fourteen healthy control subjects were included. Peripheral blood mononuclear cells were separated by gradient centrifugation, mRNA was purified, and the reverse-transcribed probes were hybridized to cDNA membranes. Group differences were assessed with the Mann-Whitney U-test. Twenty-three of seventy-eight tested IgE-related genes had significantly altered expression in atopy and asthma compared with that in the healthy subjects. The differentially expressed genes include surface molecules involved in T- and B-cell interaction and activation, cytokines, intracellular signaling products, and transcription factors. In conclusion, both atopic nonasthmatic and atopic asthmatic individuals had activated proinflammatory pathways, a minimal requirement for B-cell isotype switching, and a clear net pro-IgE cytokine climate.

Exposure and sensitization in infants and children

The complex relationship between allergen exposure, atopic sensitization and asthma in individuals and in populations has recently been a subject of controversy. A number of studies have demonstrated that allergen exposure in sensitized asthmatic individuals increases the severity of disease. A simple dose-response relationship between mite allergen exposure and specific sensitization in infants and children has been confirmed. However, the concept that there is a direct relationship between allergen exposure and the initiation of asthma has been challenged. The relationship between allergen exposure and subsequent disease development is complex, and is confounded by a number of important factors. Populations and individuals are exposed to a mixture of several allergens, irritants and pollutants, and we know very little about the impact of these mixtures and their possible synergistic effect. The dose-response relationship between allergen exposure and allergic disease may not be linear, and may be different for different allergens. For example, a protective effect of cat ownership on sensitization and allergic disease has been reported, raising the question of whether the dose-response relationship between exposure and sensitization may be different for cat compared with mite allergen. It has been suggested that many children who are exposed to a high level of cat allergen make a modified T helper type 2 response, characterized by the presence of IgG4 antibody to cat proteins without becoming allergic (i.e. no IgE response), which could be regarded as a form of tolerance. This could explain a decreased risk of asthma in children living in homes with cats, without invoking a concept of a shift in the balance of T helper types 1 and 2 responses. Cat allergen is ubiquitous, and passive exposure (e.g. home without cats and public places) may induce specific IgE responses in non-cat owners, whereas those exposed to very high levels of cat allergen may initially mount an IgE response, which may be replaced by a modified T helper type 2 response (tolerance). It is likely that the population susceptibility to allergic sensitization and also end-organ responsiveness has altered, and allergen exposure may still be important in initiating disease in an increasingly susceptible population, although the pattern may differ for different allergens.

NAC Manchester Asthma and Allergy Study (NACMAAS): risk factors for asthma and allergic disorders in adults

Asthma and atopic disorders are the most common chronic diseases in the developed countries. Knowledge of the risk factors for these disorders may facilitate the development of preventive strategies aimed at reducing prevalence rates. To investigate the risk factors for asthma and allergic diseases in a large number of adults who are the parents of children in the National Asthma Campaign Manchester Asthma and Allergy Study. All pregnant women and their partners attending "Booking" antenatal clinics were invited to take part in the study. Questionnaire data were collected including the history of asthma and other atopic diseases, pet ownership and smoking habits, and skin prick tests were performed. The prevalence of atopy and the risk factors for asthma and allergic disorders were investigated in all subjects who completed the questionnaire and underwent skin testing. Statistical analysis was carried out using logistic regression. Initially, risk factors were assessed by univariate analysis to see how each potential explanatory variable affected the probability of having allergic disease. Variables were then tested in a forward stepwise multivariate analysis. In 5687 adult subjects there was a very high (48.2%) prevalence of atopy, and 9.7% of subjects had a diagnosis of asthma. In a multivariate regression analysis sensitization to dust mite, cat, dog and mixed grasses were all independently associated with asthma. The odds ratios for current asthma increased with the increasing number of positive skin tests (any two allergens - OR 4.3, 95% CI 3.3-5.5; any three allergens - OR 7.0 95% CI 5.3-9.3; all four allergens - OR 10.4, 95% CI 7.7-14; P < 0.00001). Dog ownership (OR 1.31, 95% CI 1.10-1.57; P = 0.003) and current smoking (OR 1.36, 95% CI 1.15-1.62; P = 0.0004) were significantly and directly associated with "asthma ever". Thirteen per cent of participants reported a history of eczema. In the multivariate analysis the strongest independent associate of eczema was sensitization to dog (OR 1.37, 95% CI 1.14-1.63, P < 0.0001). Apart from dog, the strength of the association between sensitization to common allergens and eczema appeared to be much lower than in the case of asthma. The prevalence of hay fever was high (20.6%), and in the multivariate analysis the association between sensitization to pollen and hay fever was extremely strong (OR 13.6, 95% CI 11.3-16.3; P < 0.0001). The results of the current study emphasize the importance of sensitization to indoor allergens in asthma. However, evidence of a possible direct role of allergen exposure in asthma causation remains unclear.

Patient concerns in their first year with type 2 diabetes: patient and practice nurse views

One year after diagnosis, 250 patients with type 2 diabetes identified concerns related to having diabetes, in response to an open question and checklist. Their practice nurses independently reported what they believed were the patients' concerns. Nurses identified patients' main concerns in only 20% of cases. Patients' most frequent main concerns were 'fear of getting worse', 'following dietary advice' and 'damage caused by diabetes'. Nurses most frequently cited 'following dietary advice', 'illness or pain unrelated to diabetes' and 'overweight'. BMI over 25 was related to patient concerns about 'overweight' but not 'following dietary advice' nor 'taking exercise', mentioned also by people of lower weight. Only 19/106 patients with raised HbA1c levels reported concerns about high blood glucose levels. Nurse education in patient-centred care should build on the findings that patients focus more on current and future burden of symptoms and treatment than on blood glucose levels.

Apoptosis signals in atopy and asthma measured with cDNA arrays

A variety of studies have stressed the importance of the control of inflammatory cell longevity and the balance of pro-survival and pro-apoptotic signalling. Recently, asthma was found to be associated with reduced apoptosis of inflammatory cells in lung tissue. The aim of the study was to investigate the systemic activation of apoptosis pathways using cDNA array technology in atopy and asthma. Eighteen atopic asthmatics (AA), eight atopic non-asthmatic (AN) and 14 healthy control subjects (C) were included in the study. Peripheral blood mononuclear cells were separated with gradient centrifugation, mRNA purified and the reverse-transcribed probes hybridized to cDNA arrays. The signals were compared by standardizing to the 100 most expressed genes and group differences assessed with the Mann-Whitney U-test. We found a concerted up-regulation of several pro-survival cytokines and growth factors in AN and AA. FAS and FASL were not differentially expressed, but FAST kinase was over-expressed in AN and AA. The tumour necrosis factor pathway was activated in AN and AA with increased cytokine and receptor levels and increased TRAF2, an intracellular signalling product. There were indications of a down-regulated p53 system. In contrast, the Bcl-2 family of genes showed a net pro-apoptotic profile in AN and AA. The group of caspases showed a constant gene expression pattern in all groups. In conclusion, significant differences in the expression of apoptosis-related genes were found in peripheral blood of atopic individuals with and without asthma. cDNA array technology proved to be useful and may be complementary to DNA-based studies in order to analyse interactive and multidimensional pathways as shown here for apoptosis.

Allergen avoidance

Allergen exposure is a risk factor for sensitisation, prevalence and severity of allergic disease. Effective allergen avoidance strategies have been devised. The cost-benefit in public health terms should be established by prospective controlled studies that are under way around the world.

GP selection of postgraduate education courses has implications for colleagues: messages for course providers and for those writing practice professional development plans

BACKGROUND

The Department of Health's review of continuing professional development in general practice advocates setting team and individual goals.

AIM

To explore how general practitioners (GPs) share learning experiences with colleagues, focusing on how GPs choose courses as one factor influencing sharing.

METHOD

Interviews were conducted with 21 GPs using grounded theory methodology. The responses were coded by six researchers from psychology, education, and general practice.

RESULTS

Much sharing with colleagues took place, though not always immediately following a course. GP explanations revealed four reasons for course selection that influenced the degree of sharing: 1. Attendance to meet group needs encouraged rapid sharing and could involve course attendance with colleagues. 2. Attendance to enhance 'special interests' could either encourage or inhibit sharing. 3. Attendance in pursuit of 'personal interests' peripheral to general practice did not result in sharing within the primary care team. 4. Attendance to meet personal learning needs did not involve sharing when needs were not currently shared with colleagues.

CONCLUSION

Course selection and subsequent sharing have implications for course providers and those writing personal and practice professional development plans.

Comparison of pharmacokinetics and systemic effects of inhaled fluticasone propionate in patients with asthma and healthy volunteers: a randomised crossover study

BACKGROUND

Inhaled corticosteroids are currently the cornerstone of asthma treatment. Some studies of high-dose fluticasone propionate in patients with no or mild asthma have, however, suggested substantial systemic absorption. We investigated the pharmacokinetics of fluticasone propionate in patients with asthma receiving appropriate doses for severity.

METHODS

We did a double-blind, randomised, crossover study in 11 patients with asthma and 13 matched healthy controls (age 20-65 years; asthma patients forced expiratory volume in 1 s <75% and stable on high-dose inhaled corticosteroids). Patients received one 1000 microg intravenous dose or 1000 microg daily for 7 days inhaled (via spacer device) fluticasone propionate. In the 12 h after dosing, we monitored plasma fluticasone propionate and cortisol concentrations by mass spectrometry and competitive immunoassay with use of direct chemiluminescence. Analysis was by intention to treat.

FINDINGS

After inhalation, geometric mean values were significantly lower in the asthma group than in controls for fluticasone propionate plasma area under curve (1082 [95% CI 850-1451] vs 2815 pg mL(-1) h(-1) [2262-3949], -62% difference [45-72]; p<0.001), maximum concentrations (117 [91-159] vs 383 pg/mL [302-546], -68% [-50 to -81]; p<0.001), and systemic bioavailability (10.1 [7.9-14.0] vs 21.4% [15.4-32.2], -54% [-27 to -70]; p=0.001). Intravenous-dose clearance, volume of distribution at steady state, plasma half-life, and mean residence time, were similar in the two groups. Less suppression of plasma cortisol concentrations was seen in the asthma group than in controls 4-12 h after inhaled fluticasone propionate.

INTERPRETATION

Systemic availability of fluticasone propionate is substantially less in patients with moderate to severe asthma than in healthy controls. Inhaled corticosteroids that are absorbed through the lungs need to be assessed in patients who are receiving doses appropriate for disease severity, and not in normal volunteers.

Synthetic pillows contain higher levels of cat and dog allergen than feather pillows

The use of non-feather pillows has increased over the last few decades. Recently, we found significantly higher levels of dust mite allergens in synthetic pillows than in feather ones. This study investigated the levels of pet allergens in feather and synthetic pillows. Dust samples were collected from 14 pairs of pillows (consisting of one synthetic fibre-filled and one feather-filled). Each pair of pillows had been on the same bed for at least 2 years. The pillows were vacuumed for 1 min on each side through a 355-microm diameter mesh onto a 5-microm vinyl filter, producing a sample of fine dust. Samples were extracted, and cat (Fel d 1) and dog (Can f 1) allergens determined using monoclonal antibody-based enzyme-linked immunosorbent assay (ELISA). Results were expressed both as total amount of allergen recovered and concentration of allergen per unit weight (ng/g). Total levels of pet allergens were significantly higher in the synthetic pillows (Fel d 1: 6.7-fold difference [95% CI 1.5-29.7], p=0.01; Can f 1: 8-fold difference [95% CI 1.6-39.5], p=0.01). Similarly, when the results were expressed as ng/g, synthetic pillows contained significantly more allergens than feather pillows (Fel d 1: 3.7-fold difference [95% CI 1.3-10.1], p=0.01); Can f 1: 4.4-fold difference [95% CI 1.5-13.2], p=0.01). We have therefore demonstrated that synthetic pillows contain significantly more pet allergens than feather pillows, supporting the view that tightly woven encasements surrounding feather pillows act as a barrier for allergens.