Phase II of the International Study of Asthma and Allergies in Childhood (ISAAC II): rationale and methods

Questionnaires

Standardised core questionnaires were developed on demographics, wheezing, allergic rhinitis and atopic eczema. These include detailed questions on the occurrence and severity of atopic symptoms, identical to those used for parents of children aged 6–7 yrs in ISAAC I 6. Supplementary questionnaires dealt with additional respiratory symptoms, as well as the clinical management and medical treatment of asthma and wheezing, hay fever, rhinitis and flexural dermatitis. The risk factor questionnaire enquires about current and previous (first year of life) living conditions, and collects information about factors such as birthweight, gestational age, place of birth, socioeconomic status, breast feeding, family size, family history of atopic diseases, infections, vaccinations, crowding, pets, environmental tobacco smoke, cooking fuels, damp housing, type of windows, type of bedding, area of residence, physical activity and nutritional factors. All questionnaires were for completion by parents.

Examination for flexural dermatitis

Participating children were examined for visible flexural dermatitis according to a photographic protocol 17 and detailed instructions on its use in the field. The examination involved inspection of the skin: 1) around the eyes, 2) around the sides and front of the neck, 3) in front of the elbows, 4) behind the knees, and 5) in front of the ankles. The presence or absence of signs of visible flexural dermatitis was recorded for each of the five areas. A training package of high-quality clinical photographs and instructions for fieldworkers was developed, which also included a test set of pictures for central evaluation of fieldworker quality 18.

Skin-prick testing for atopy

Extracts of six common aeroallergens (Dermatophagoides pteronyssinus, D. farinae, cat hair, Alternaria tenuis, mixed tree pollen and mixed grass pollen) and control solutions produced by ALK (Hørsholm, Denmark) were used. Cockroach and other allergens of localrelevance could be added. Briefly, a drop of each allergen extract and the positive (10 mg·mL⁻¹ histamine) and negative (diluent) control was placed on to the skin of the volar side of the left forearm and pierced vertically using 1-mm ALK lancets. After 15 min, the outer contour of the weal reaction was outlined using a fine felt-tip pen, and the result expressed asthe mean of the lengths of the longest diameter and theperpendicular line through its centre. Detailed recommendations for training and monitoring of fieldworkers were provided in order to assess reproducibility and consistency over time.

Bronchial challenge with hypertonic saline

A minimum of two baseline spirograms was recorded, and the higher of two reproducible (within 5%) measurements of forced expiratory volume in one second (FEV₁) was recorded as baseline FEV₁. Bronchial reactivity was assessed by changes in FEV₁ after inhalation of nebulised saline using an ultrasound nebuliser 19. The children inhaled hyperosmolar (4.5%) saline for periods of increasing duration: 0.5, 1, 2, 4, and 8 min. FEV₁ was measured 1 min after each inhalation period and the next challenge period followed after 3 min. If the FEV₁ fell by 10–15% from the baseline value, the exposure time was repeated. If, after two repetitions, the FEV₁ remained 10–15% below the baseline value, the duration of the inhalation period was doubled again according to the protocol. Either way, bronchial challenge was stopped if either the FEV₁ had fallen by ≥15% from baseline or the total inhalation period of 15.5 min had been reached. The saline canister and tubing were weighed before the first and after the last inhalation period in order to measure the aerosol dose delivered. Participating children were asked to withhold bronchodilator medication before the challenge. Regular use of inhaled steroids was recorded, but not withheld. In children with a baseline FEV₁ of <75% of the predicted value, no bronchial challenge was performed and an inhaled bronchodilator was administered.

Serum samples for analyses of immunoglobulin E

Blood samples were collected by venipuncture into plain sample tubes and allowed to clot at room temperature. The serum was separated by centrifugation (2,500–3,000 revolutions per minute for 15 min) and pipetted off in 1-mL aliquots into plastic tubes with screw corks. Samples were frozen, stored at ≤−20°C and transported under dry ice.

Serum levels of total immunoglobulin (Ig)E and IgE directed against a panel of common inhalant allergens (Phadiatop^TM) were quantified using the Pharmacia CAP System^TM (Pharmacia Diagnostics, Uppsala, Sweden). The analyses were performed in one central certified (SWEDAC) laboratory (Dept of Clinical Immunology, Karolinska University Hospital, Stockholm, Sweden). In three centres, analyses had already been performed using the same methods but in another laboratory. The comparability of results obtained in different laboratories was tested in 400 samples from two centres 20, which were analysed in the central laboratory and another laboratory (two laboratories in the case of total IgE). There was good agreement between results: for total IgE levels, the concordance correlation coefficient was >0.99; and for specific IgE classes, the weighted κ was 0.88.

Dust samples for analyses of aeroallergens and endotoxins

Instructions for dust collection are in the ISAAC II manual and in the more detailed protocol 21, including photographic instructions, developed by Utrecht University (Utrecht, The Netherlands), which has been in use since 2001. Vacuum cleaners with a ≥800-W engine were used in combination with ALK filter holders and 70-mm paper filters (ALK or Schleicher&Schuell, Dassel, Germany). Dust was sampled separately from two sites within homes: the mattress or sleeping place of the index child participating in the study, and the floor in the living room. The beds were vacuumed on the sheets or, if there were no sheets, directly on the mattress or sleeping surface. The original instructions were to vacuum soft surfaces for 2 min·m⁻², covering an area of ≥2 m² (≥4 m² if possible), and hard surfaces for 1 min·m⁻², covering an area of ≥4 m² (ISAAC II manual). This was subsequently made more specific as follows: 1) mattress, whole area for 2 min; 2) carpeted floors, 2 m² for 4 min; 3) smooth floor with ≥4-m² rug, 2-m² rug for 4 min; and 4) smooth floor with no rug or smaller rug(s), 4-m² smooth floor for 4 min (Utrecht University protocol). After collection of dust, the filter box with the filter was removed, the lid was attached and then the box was stored in a plastic bag at −20°C until analysis (ISAAC II manual). Alternatively, after dust collection, the filter with dust was transferred to a preweighed plastic tube directly after the vacuum cleaning, facilitating transport and analysis (Utrecht University protocol).

Laboratory analyses of the dust of European centres took place at Utrecht University, at the Institute for Risk Assessment Sciences. Dust of non-European centres was analysed at the laboratory of the Wellington Asthma Research Group (Wellington School of Medicine and Health Sciences, Wellington, New Zealand). After weighing, floor dust was extracted using Tween-20 in water 22 and analysed for endotoxin, using a kinetic chromogenic Limulus amoebocyte lysate (LAL) test, using only one batch of LAL reagents (LAL lot no. 1L676S and lipopolysaccharide standard lot no. 2L0090; BioWhittaker, Verviers, Belgium) for all ISAAC analyses. Endotoxin concentrations were expressed as units of endotoxin per gram of dust (1 endotoxin unit=0.09 ng). Mattress dust was extracted using phosphate-buffered saline containing 0.05% Tween-20, and analysed for the allergens D. pteronyssinus antigen 1 (Der p 1), D. farinae antigen 1 (Der f 1), dog antigen 1 (Can f 1) and cat antigen 1 (Fel d 1), using the enzyme immunoassay reagents and methods of Indoor Biotechnologies (Cardiff, UK), essentially as described previously 23. The lower limits of detection are 8 ng·mL⁻¹ for Der p 1, 6 ng·mL⁻¹ for Der f 1, 20 ng·mL⁻¹ for Can f 1 and 0.4 ng·mL⁻¹ for Fel d 1 for five-fold diluted samples. Samples with undetectable amounts of allergen or endotoxin were assigned a value two-thirds that of the lowest observed amount per gram of dust for the specific component determined.

DNA preparation for genotyping

The alternative sources of DNA from study participants included frozen storage of whole blood or buffy coats and storage of whole blood as dried spots. Samples originating from populations with prevalence rates of category III infectious agents (such as hepatitis B and HIV) of >1% required extraction in a containment level 3 laboratory.

Whole blood and buffy coat samples were received frozen in tubes containing ethylenediamine tetra-acetic acid, with typical volumes ranging 4–10 mL. Extractions were performed using the Promega Wizard genomic DNA purification kit (Promega UK, Southampton, UK), using the standard protocol adjusted for sample volume. Extraction of blood spots was carried out using the Qiagen Midi Kit (Qiagen, Crawley, UK). Blood spots were cut from filter paper and further fragmented using autoclaved scissors. Cell lysis and protein precipitation was carried out using volumes adjusted for sample volume.

Following extraction, all samples were quantified using a Molecular Probes Pico Green double-stranded DNA Quantification Kit (Invitrogen, Paisley, UK). Blood spots yielded 0.0–48.0, buffy coats 0.0–1,013.0 and whole blood 0.0–578.0 ng·µL⁻¹ DNA. The corresponding total amount of DNA from each source was blood spots 0.9–2.1, buffy coats 3.5–69.5 and whole blood 47.0–69.0 µg. Primer extension preamplification was used to increase the quantity of DNA by a factor of 40. This involves using random primer sequences to amplify template DNA, resulting in whole genome amplification 24.

The aim of the genetic module is to genotype the samples with regard to all published genetic polymorphisms associated with childhood asthma and allergies. Several single nucleotide polymorphisms will be typed from each gene in order to create haplotypes that may be compared across populations. Single nucleotide polymorphisms will also be chosen according to their heterozygosity level, with minor allele frequencies of ≥20%. Genotyping will be carried out using the Sequenom System (Sequenom, Inc., San Diego, CA, USA), which is based on an assay of primer extension with detection by matrix-assisted laser desorption time-of-flight mass spectrometry (MALDI-TOF). Genotyping is performed at the University of Oxford (Oxford, UK).