Pii: s0091-6749(99)70181-8

House dust mite avoidance for children
with asthma in homes of low-income
families
Gail G. Shapiro, MD,a Timothy G. Wighton, PhD,a Tamara Chinn, RN,a Jon
Zuckerman, BA,a A. Heather Eliassen, BA,a Joseph F. Picciano, BA,a and Thomas
A. E. Platts-Mills, MD, PhDb Seattle, Wash, and Charlottesville, Va
Background: Home exposure to high levels of house dust mite
allergen has been shown to aggravate airways reactivity and
Objective: The purpose of this study was to determine whether
specific house dust mite control measures could reduce expo-
sure levels and asthma severity.
ways inflammation and clinical disease.5-9 Secondary Methods: This double-blinded, randomized trial compared
variables included assessment of symptoms, quality of asthma progression over 1 year in children whose homes
life, spirometry, urgent care, and hospital visits, as well received standard environmental control intervention with
as prednisone courses for acute exacerbations.
those whose homes received aggressive intervention for dust
mite elimination. The primary end point was doubling in PD20
methacholine.
Patient selection and clinical measures
Results: Symptom scores and quality-of-life scores were simi-
lar for the standard and aggressive intervention groups. PD20
This study represents a double-blinded, randomized clinical methacholine doubling occurred in 9 members of the aggres-
trial; only the home dust collector was unblinded, and he did not sive intervention group vs 4 control patients (P < .05). Dust
communicate home information to others. Children aged 6 to 16 mite levels decreased in the aggressive intervention homes
years with asthma were recruited from clinics that serve lower compared with the standard intervention homes (P < .05).
socioeconomic neighborhoods. The children had been diagnosed by Conclusion: Aggressive dust mite intervention decreased dust
a physician as having asthma and satisfied the definition of having mite levels and improved bronchial hyperresponsiveness. (J
mild-to-moderate persistent disease according to criteria set forth in Allergy Clin Immunol 1999;103:1069-74.)
the National Heart, Lung, and Blood Institute guidelines.10 Thesecriteria included 1 or more of the following: cough or wheeze more Key words: Asthma, dust mite, bronchial hyperresponsiveness,
than once a week, nocturnal symptoms, documentation of peak expiratory flow rate (PEFR) at 60% to 80% of predicted level, andcurrent asthma symptoms causing some impairment to normal There is growing concern in the medical and lay com- activity, including school performance. In addition, patients had to munities that asthma should be treated more aggressive- meet all of the following inclusion criteria: (1) at least 1 urgent care ly from both an environmental and pharmacotherapeutic visit caused by asthma in the previous 6 months; (2) history of point of view. Continuous exposure to high levels of required albuterol at least 5 times per month; (3) positive metha- environmental allergens, in particular house dust mites, choline challenge response at a concentration of 10 mg/mL or less; has been shown to aggravate airways reactivity and may and (4) positive skin prick test response to Dermatophagoides be a critical factor in the development and perpetuation pteryonyssinus, Dermatophagoides farinae, or both (ie, wheal diameter at least 3 mm greater than that produced by the saline con- The goal of this study was to determine whether spe- trol). Patients were excluded if their families were already carryingout environmental control measures in the home as reported by the cific house dust mite control measures could reduce field staff on the first home visit.
exposure levels to this allergen and, subsequently, reduce Parents and older children consented (and younger children the severity of asthma in children. The primary end point assented) to be randomized to either a standard or aggressive envi- was change in airways reactivity to methacholine ronmental control program. The trial was approved by the Investi- because this response can be roughly correlated with air- gational Review Board of Children’s Hospital and Regional Med-ical Center, Seattle, Washington.
At the initial visit, all patients provided a history and underwent physical examination, quality-of-life assessment, and skin prick From aA.S.T.H.M.A., Inc, Seattle; and bthe Division of Allergy, University of tests to dust mites, cat, cockroach, and regional pollens. Spirometry Virginia Medical Center, Charlottesville.
was performed, with short-acting β-agonists having been withheld Supported by National Institute of Allergy and Infectious Diseases grant #1 for at least 4 hours. Patients were instructed on peak flow tech- UO1 AI34578-01, AI/ES-34607, and AI-20565.
niques and charting of peak flow and symptoms, which was to be Received for publication Oct 29, 1998; revised Mar 2, 1999; accepted for done morning and night. Patients received cromolyn, nedocromil, or triamcinolone metered-dose inhalers at the discretion of the investi- Reprint requests: Gail G. Shapiro, MD, A.S.T.H.M.A., Inc, 4540 Sand Point gator, with the inhaled steroid (ie, triamcinolone) being reserved for those with more severe disease, as determined by history of 1/1/98647
albuterol use and pulmonary function.
1070 Shapiro et al
TABLE I. Baseline characteristics of the study population
TABLE II. Doubling PD20 methacholine
Standard
Aggressive
es), change in FEV1 levels (classified as mild, moderate, or severe), and changes in household dust mite concentration (classified as low, moderate, or high). In each of these analyses, P values of less than .05 were considered indicative of statistically significant group dif- Four weeks later, patients returned for methacholine challenge, and diaries were reviewed. A home visit occurred soon after this at Patient sample
which patients were randomized to the standard or aggressive envi-ronmental intervention arms. Patients were again seen in the clinic Forty-four patients were randomized, and 36 complet- after 6 and 12 months for history, physical examination, quality-of- ed the evaluation. Five patients (3 in the standard and 2 life assessment, spirometry, and methacholine challenge. Peak flow in the aggressive intervention groups) failed to complete and symptom diaries were collected, and families were questioned the trial because of noncompliance with medications and regarding their children’s asthma severity in this interval compared visits. Three (2 in the standard and 1 in the aggressive with baseline values. Sequential pulmonary function and metha- intervention groups) failed to complete the study because choline challenges were done at approximately the same time of of worsening asthma: 1 patient withdrew on the advice of day and at least 30 days after steroid bursts.
his primary care doctor, and 2 were withdrawn when Home intervention and analysis
FEV1 values were too low to allow methacholine chal- Standard intervention included a general discussion of the need to dust and vacuum the house weekly and avoid clutter in the bed- Mean age and racial distribution were similar between room. The home visit to patients in this group involved vacuuming groups. Racial minorities are overrepresented in this of pillow and mattress, application of tannic acid placebo to the sample compared with the composition of the greater child’s bedroom carpet every 4 months, and phone call reminders to Seattle area, where the minority population is approxi- families to spray tannic acid placebo on carpets at 2, 6, and 10 mately 13% (communication from Chamber of Com- months. Aggressive intervention consisted of the application of dust mite–impermeable covers (donated by Allergy Control Products, Medication profiles of the groups were similar, with 9 Inc, Ridgefield, Conn) to mattress, box spring, and pillow; laundry and 11 of the standard and aggressive intervention service delivery of a clean blanket and 4 sets of bed linens every patients, respectively, taking cromolyn; 5 standard and 7 month; and tannic acid application to the bedroom and living roomcarpet every 2 months. Families were instructed to dust and vacuum aggressive intervention patients taking triamcinolone; weekly and to avoid clutter. A vacuum cleaner was provided if a and 3 standard and 1 aggressive intervention patient family in either group lacked one, which occurred in only one using albuterol only in spite of anti-inflammatory recom- instance. With each delivery of bedding, compliance with environ- mendations. This changed little during the trial.
mental intervention (eg, special covers on beds) was confirmed.
Dust samples were collected from all homes at baseline and 4, 8, Diaries and symptoms
and 12 months after randomization. The collection procedure con- Attempts to collect daily symptom and peak flow sisted of 2 minutes of vacuuming of the following areas: 2 m2 of the diaries were futile. Although patients were encouraged to upper surface of the child’s mattress (sheets removed but special mail these, there were no reminder calls built into the encasings in place), 1 m2 of carpet near and underneath the child’s protocol. Symptom scores and quality-of-life outcomes bed, a major piece of upholstered furniture in the family area, 1 m2of carpet underneath a major piece of upholstered furniture in the were assessed at baseline and at 6- and 12-month visits.
family area, and 1 m2 of the kitchen floor. The technique for prepa- Parents rated their child’s asthma severity as mild, mod- ration, storage, and analysis of the samples has been described.11 erate, or severe. There were no significant changes frombaseline to 12 months for each group nor were there any Statistical analysis
differences between the groups. Quality of life was mea- Logistic regression was used to detect a significant effect of the sured on a 14-point scale, with 0 representing no symp- aggressive household intervention compared with standard treat- toms and 14 representing many symptoms. These scores ment. A doubling in PD20 dosage units, determined by using base- were similar for the groups and did not change during the line and 12-month follow-up results, was compared between groups year. Exacerbations measured in terms of hospitaliza- after adjusting for baseline concentrations of dust mite in the home tions, emergency department visits, and steroid bursts and baseline levels of bronchial hyperresponsiveness (in PD20 were similar for the 2 groups. Although there were more dosage units). The chi-squared statistic was used to detect signifi-cant differences between groups for categorical measures, including steroid bursts for the standard than for the aggressive the incidence of patient exacerbation during the course of the study intervention group (11 vs 5), this did not reach statistical (emergency department visits, hospitalizations, and steroid cours- Shapiro et al 1071
TABLE III. Medication use of patients with PD20 doubling
Patient no.
Baseline medication
PD20 (dosage units)
12-month medication
PD20 (dosage units)
prn, As required; qd, daily; bid, twice daily; tid, three times daily; qid, four times daily.
Pulmonary function and methacholine
The same was true for Can f 1 (dog antigen). Table IV reactivity
lists the relative amounts of pet antigen in homes. For catantigen, the number of patients with a positive skin prick Pulmonary function was similar between groups at the test response and the number of patients with a cat in the beginning and end of the trial. Mean FEV1 (L/sec) was home in addition to having a positive skin test response 1.93 ± 0.97 for the standard intervention group and 1.85 ± are noted. Skin testing for dog antigen was not performed.
0.56 for the aggressive intervention group at baseline, with Cockroach antigen concentration (Bla g 2) was mea- little change during the year. Describing asthma on the sured as being low (<1 µg/g dust), moderate (1 to <8 g/g dust), or high (≥8 µg/g dust). Only 3 samples con- dicted value), moderate (60% to 79% of predicted value), tained moderate levels; the rest were low.
or severe (59% or less of predicted value), the groups had Correlation of PD
a similar composition, which was constant for the year.
20 doubling, dust mite con-
Bronchial hyperresponsiveness, measured as PD centration, and cat antigen concentration
showed initial similarity between the groups. The 12- There was no direct correlation between changes in month challenge showed a more pronounced increase in PD20 and dust mite level (Fig 2). No threshold determi- PD20 for the aggressive intervention group than the stan- nation appeared to change this. The 13 patients (4 in the dard intervention group. There was a doubling of PD20 standard and 9 in the aggressive intervention groups) for 4 members of the standard intervention group and 9 who experienced a doubling of PD20 had a mean members of the aggressive intervention group (P = .049).
decrease in house dust mite concentration of 5.4 µg/g, Changes in PD20 are seen in Tables II and III. Changes in and those who did not had a mean decrease of 2.4 µg/g.
daily asthma therapy, including controlling for inhaled The presence of cats did not have a predictable effect steroid use, did not account for this difference.
on dust mite environmental control outcome. The 7patients with positive skin test responses to cat antigen Antigen content of dust samples
and a cat in the home had a mean increase in PD20 of 10.9 Antigen content of samples showed abundant quanti- dosage units, and those without cat had a mean increase ties of dust mite, cat, and dog allergen. The quantities of of 7.8 dosage units, showing that the presence of a cat did cockroach allergen were very low. Dust mite (pooled Der not prohibit improvement in PD20. Removing the 7 p 1 and Der f 1) concentration was categorized as low (<2 patients with positive skin test responses to cat antigen µg/g dust), moderate (2 to <10 µg/g dust), or high (≥10 and a cat in the home from analysis reduces the power µg/g dust). A change in designation to a lower category substantially. The relationship between doubling PD20 occurred in 16.7% of the standard intervention homes and dust mite intervention becomes a trend (P = .17). The and 50% of the aggressive intervention homes (P = .03; removal of patients with a positive skin test response to Fig 1). From randomization to 12 months later, the stan- cat and cat antigen in the home without a cat reduces the dard intervention group had a mean 33% increase in dust sample too drastically for meaningful analysis.
mite levels, whereas the treatment group had a mean19.6% decrease in dust level (P = .2).
DISCUSSION
Levels of Fel d 1 were moderate (<8 µg/g dust) or high (≥8 µg/g dust) in almost all homes at each visit regardless This trial began with the hypothesis that aggressive of cleaning regimen or the presence of a cat in the home.
environmental intervention directed at house dust mite 1072 Shapiro et al
FIG 1. Changes in dust mite concentration levels by group (low, <2 µg/g; moderate, 2 to 10 µg/g; and high, ≥10
µg/g). The number of patients with household dust mite concentrations in the low, moderate, and high rangeare depicted at the baseline and 12-month collection periods. The change to a lower concentration range wassignificant for the aggressive compared with the standard intervention groups (P < .03).
TABLE IV. Cat/dog antigen levels
Moderate
Standard
Aggressive
Standard
Aggressive
Standard
Aggressive
All values are from the first home visit.
SPT, Skin prick test.
elimination could have a positive effect on asthma in dust courses; aggressive intervention group, 5 courses), this mite–sensitive children from homes of lower-income did not reach statistical significance.
families. The change in the primary outcome variable, The concentrations of cat and dog antigens in these doubling of PD20, confirmed this premise. Along with homes were impressive. The number of homes with this change, a decrease in dust mite antigen in home sam- moderate and high levels of pet antigen and no resident pet is in keeping with the findings of others.13 Cleaning It is disappointing that home symptom diaries and practices directed at dust mite mitigation did not signifi- peak flows were not available. Had we built in more cantly decrease these levels. Some of our subjects had patient visits or phone contacts, we might have been positive skin test responses to cat antigen and had sub- more successful with this. However, other studies on stantial concentrations of antigen in the home, a situation asthma among low-income patients have had similar dif- that could have obscured more dramatic benefits of dust ficulties in obtaining peak flow or symptom diaries.12 mite environmental control. We cannot prove, however, Parental reporting of overall asthma severity and quality that cat antigen decreased the benefit of dust mite inter- of life at 6-month visits failed to show intergroup differ- ences. Although there were differences in the number of The literature to support dust mite intervention for prednisone courses (standard intervention group, 11 asthma control is extensive.14-20 Nonintervention studies Shapiro et al 1073
FIG 2. Changes in mean PD20 before to 12 months after initiation of dust mite intervention. Mean PD20 is
expressed for the standard and aggressive intervention groups at baseline and 12 months later. The standard
group showed a mean increase in PD20 of 24.2%, whereas the aggressive intervention group showed a mean
increase of 63.3%.
show correlations between antigen levels and clinical or removed carpeting and upholstery with those living in markers of disease severity.4 Custovic et al1 described a standard bedrooms. Patients with aggressive environmen- correlation between mite exposure, methacholine reac- tal control had significant improvement in the tivity, and decreased FEV1 in adults. Chan-Yeung et al2 FEV1/forced vital capacity ratio, PEFR, and bronchial noted that dust mite–sensitive asthmatic children showed hyperresponsiveness (histamine), but not FEV1, com- a positive relationship of mean daily symptom score to total mite allergen level and a negative relationship of Our trial represents the first demonstration of the pos- daily mean PEFR to dust mite level. Sensitization to dust itive effect of dust mite reduction measures in a low- mites has been associated with acute childhood asthma income US population. It confirms previous observations in more affluent communities of a correlation between Australian investigators have reported that changes in decreased dust mite concentrations and improvement in allergen concentration in bedding were significantly cor- bronchial hyperresponsiveness.21 In concert with other related with bronchial hyperresponsiveness to histamine trials, bronchial hyperresponsiveness appears to be more and to symptom scores.16 They cited small within-patient sensitive to change with environmental intervention than variations in PD20 explained by antigen levels and failed, FEV1, PEFR, or symptoms.16,17 The contribution of as we did, to see significant changes in peak flow or other antigens in the home and the complexity of dealing FEV1. European investigators have documented the with them in American homes, particularly pet antigens, value of impermeable mattress covers combined with use of filters in improving bronchial hyperresponsiveness.17Reduction in mite antigen correlated with a significant REFERENCES
1. Custovic A, Taggart SCO, Francis HC, Chapman MD, Woodcock A.
Exposure to house dust mite allergens and the clinical activity of asthma.
were unchanged, as with our evaluation.
J Allergy Clin Immunol 1996;98:64-72.
These more recent studies reinforce other trials from 2. Chan-Yeung M, Manfreda J, Dimich-Ward H, Lam J, Ferguson A, War- the last decade. Murray and Ferguson19 studied asthmatic ren P, et al. Mite and cat allergen levels in homes and severity of asthma.
children in dust-free bedrooms that included impermeable Am J Respir Crit Care Med 1995;152:1805-11.
covers for pillows, mattress, and box spring; no carpets; 3. Sporik R, Holgate ST, Platts-Mills TAE, Cogswell JJ. Exposure to house- dust mite allergen (Der p 1) and the development of asthma in childhood.
and limited toys. Children living in these special rooms A prospective study. N Engl J Med 1990;323:502-7.
compared with children living in standard bedrooms 4. Nelson RP, DiNicolo R, Fernandez-Caldas E, Seleznick MJ, Lockey RF, showed fewer days of wheezing, fewer days requiring Good RA. Allergen-specific IgE levels and mite allergen exposure in medication, and fewer days with abnormal peak flow children with acute asthma first seen in an emergency department and innonasthmatic control subjects. J Allergy Clin Immunol 1996;98:258-63.
rates. Walshaw and Evans20 evaluated adult asthmatic 5. Burrows B, Sears MR, Flannery EM, Herbison GP, Holdaway MD, Silva subjects and compared those living in bedrooms with spe- PA. Relation of the course of bronchial responsiveness from age 9 to age cial bed covers, weekly laundered bedding, and cleaned 15 to allergy. Am J Respir Crit Care Med 1995;152:1302-8.
1074 Shapiro et al
6. Sparrow D, O’Connor G, Colton T, Barry CL, Weiss ST. The relationship 14. Platts-Mills TAE. Dust mite allergens and asthma—a worldwide prob- of non-specific bronchial responsiveness to the occurrence of respiratory lem. J Allergy Clin Immunol 1989;83:416-27.
symptoms and decreased levels of pulmonary function. The Normative 15. Platts-Mills TAE, Woodfolk JA, Chapman MD, Heymann PW. Changing Aging Study. Am Rev Respir Dis 1987;135:1255-60.
concepts of allergic disease: the attempt to keep up with real changes in 7. Burney PGJ, Britton JR, Chinn S, Tattersfield AE, Papacosta AO, Kelson lifestyle. J Allergy Clin Immunol 1996;98:S297-306.
MC, et al. Descriptive epidemiology of bronchial reactivity in an adult 16. Marks GB, Tovey ER, Green W, Shearer M, Salome CM, Woolcock AJ.
population: results from a community study. Thorax 1987;42:38-44.
The effect of changes in house dust mite allergen exposure on the sever- 8. Rijcken B, Schouten JP, Weiss ST, Speizer FE, van der Lende R. The ity of asthma. Clin Exp Allergy 1995;25:114-8.
association of airways responsiveness to respiratory symptom prevalence 17. van der Heide S, Kauffman HF, Dubois AE, de Monchy JGR. Allergen and to pulmonary function in a random population sample. Bull Eur reduction measures in houses of allergic asthmatic patients: effects of air- cleaners and allergen-impermeable mattress covers. Eur Respir J 9. Clifford RD, Howell JB, Radford M, Holgate ST. Associations between respiratory symptoms, bronchial response to methacholine, and atopy in 18. Korsgaard J. Mite asthma and residency. A case-control study on the two age groups of schoolchildren. Arch Dis Child 1989;64:1133-9.
impact of exposure to house-dust mites in dwellings. Am Rev Respir Dis 10. Guidelines for the diagnosis and management of asthma. Bethesda (MD): National Institutes of Health; National Heart, Lung, and Blood Institute; 19. Murray AB, Ferguson AC. Dust-free bedrooms in the treatment of asth- April 1997. NIH publication no. 97-4051.
matic children with house dust or house dust mite allergy: a controlled 11. Childhood Asthma Management Program Research Group. The Child- hood Asthma Management Program (CAMP): design, rationale, and 20. Walshaw MJ, Evans CC. Allergen avoidance in house dust mite sensitive methods. Control Clin Trials 1999;20:91-120.
adult asthma. Q J Med 1986;58:199-215.
12. Rosenstreich DL, Eggleston P, Kattan M, Baker D, Slavin RG, Gergen P, 21. Ehnert B, Lau-Schadendorf S, Weber A, Buettner P, Schou C, Wahn U.
et al. The role of cockroach allergy and exposure to cockroach allergen in Reducing domestic exposure to dust mite allergen reduces bronchial causing morbidity among inner-city children with asthma. N Engl J Med hyperreactivity in sensitive children with asthma. J Allergy Clin Immunol 13. Bollinger ME, Eggleston PA, Flanagan E, Wood RA. Cat antigen in homes with and without cats may induce allergic symptoms. J AllergyClin Immunol 1996;97:907-14.

Source: http://centerforhealthyhousing.org/Portals/0/Contents/Article0715.pdf