Atopy in Omani patients with asthma 5 Atopy in Omani patients with asthma *Maha Al-Amri1, Omar A Al-Rawas2, Bazdawi MS Al-Riyami2, Elizabeth R Richens3 ABSTRAC T. Objectives: To determine the range of serum IgE in healthy subjects and in asthmatic patients in Oman and to assess the degree of atopy in the asthmatic patients. Method: Serum IgE and in vivo (the skin prick test) and in vitro (the ImmunoCAP test) aller- gen-specific IgE levels were measured in 44 patients with asthma. Control groups were 19 healthy subjects and 27 asymptomatic allergic subjects. Results: The normal range for serum IgE in the Omani population was established at ≥ 101 IU/ml. The geometric mean (and 95% confidence interval) for asthmatic patients was 468 IU/ml (323–676). Positive results for allergen-specific IgE, defined as responses to ≥ 1 allergen mix in the ImmunoCAP and to ≥ 3 allergens in the skin prick test, occurred in 26/35 (74%) and in 34/44(77%) asthmatic patients respectively. Six out of 38 patients with serum IgE ≥ 101 IU/ml and 2/6 with levels <101 IU/ml gave negative and positive results respectively in the skin prick test. Overall, the degree of reactivity in the skin prick test correlated with the level of total serum IgE (r= 0.54, p<0.001). A similar correlation could not be established with ImmunoCAP reactivity, but sIgE levels ≥ 101 IU/ml were supported by a high frequency of positive ImmunoCAP responses for the majority of allergen mixes. Conclusions: Total serum IgE levels should be routinely monitored in asthmatic subjects as this may give an indication of atopy where skin prick testing is not indicated. Since in a minority of patients serum IgE levels and skin prick results do not predict in the same direction, all laboratory data should be interpreted in context of clinical history. Keywords: atopy, asthma, Oman 1Department of ENT, Al-Nahda Hospital, Ministry of Health, PO Box 393, Muscat 113, Sultanate of Oman. 2Department of Medicine, and 3Department of Microbiology and Immunology, College of Medicine and Health Sciences, Sultan Qaboos University, PO Box 35 Al-Khod, Muscat 123, Sultanate of Oman. *To whom correspondence should be addressed. The clinical expression of asthma is characterised by chronic eosinophilic inflammation, airway hyper- responsiveness to a variety of specific and non-specific stimuli and reversible airway obstruction.3,4 Whilst the pathogenesis of this inflammation has been shown to be increasingly complex, the interaction of immunoglob- ulin-E (IgE) with high-affinity mast cell receptors and the consequent release of inflammatory factors, are still widely accepted as the initiating/triggering events.5 Investigations of the role of the humoral immune system in the aetiology and presentation of asthma normally include the measurement both of total serum IgE (sIgE) reactivity and levels of allergen-specific IgE (asIgE). BRONCHIAL ASTHMA IS ONE OF THE MOST COM-mon chronic disorders, both of children and of adults, worldwide. Despite regional differ- ences in its prevalence, asthma presents considerable socio-economic burden to all societies. In Oman, stud- ies are under way to determine the epidemiology and severity of asthma in children, through participation in the International Study of Asthma and Allergies in Childhood (ISAAC).1 As measured by the occurrence of wheezing, there is a prevalence of 9% in 3–4-year-olds.2 If these figures project through to the entire population, they suggest a high burden of disease from asthma in Oman. SQU JOURNAL FOR SCIENTIFIC RESEARCH: MEDICAL SCIENCES 2002, VOL. 4, NO. –2, 5–23 ©SULTAN QABOOS UNIVERSIT Y �Richens atopy in Asthma ������ ��������� ����� ���� ������)������ ���( ����� �������� � ������� ����� � ������ ��� � ������� ��� �������� :�����:��� ������� �������� ������ ����� ��� �����IgE������� � ������ ������ ��� ���� ����� ���� � ������ �������� ���� ����� ��� ������� �����)������ ��� (����� ���� ��.�������:�� ����� ���� ��IgE���� � ����� �� �� ������� ����IgE������ �������� ������� ������ ������� �������"����� ��� "�� ������ � ������ ���“ImmunoCAP ”��� �� ��������������� ������� ������ �� � ����� ���� �������� ����� ����������� ����� ���� ��� ������ �� ������ ���� �����.�������:� �������� ������� ��� ���IgE�� �������� ��� >�������� ����/������� �������� �� )������ %����� ��� ( ��� �������� ���������� ����/��)���-��� .( � �������� ������� ����� �� ���IgE���� �� ������� ������ ������� ������� ������� >��� ������ �� ������� ������ ������ ����“Immuno CAP ”� >������� �� ������� ������ ������ "����� ��� "�� ����� ��/��)�� ( %�����/��)�� (%������� ��� ����� ��� �� .�� �������� ������ ����� ����� �������� › IgE ����� ����/� �� ��������� ����� ����� ����IgE<�������� ���� /������� � ����� ����� ����� ��)������� ��� (������ �� "����� ��� "��� ���� .��� ����� �������"����� ��� "� ����� ������� �� �����IgE�� ������� ��� � ����� ��IgE����� �� >�������� ����/�� ������� ������� �������� ������� ������ �������� ����.������:� ����� �������IgE����� ���� �� ������ ���� ����� ��� ����� �� � ������� ����� ��� ��������� ������ ���)������ ��� (������ ������� ��� ��� �� ��� �"����� ��� "�� ����� �IgE����� �� ������ ����� ������� �� ������� ��� ��� ����� ��� ���� � ������ �� ��� �� ����� �� ���� �. 6 asIgE has been measured by in vivo (skin prick test: SPT) and in vitro (ImmunoCAP test) techniques. In the present study, we have investigated the fol- lowing features of atopy in asthmatic patients in Oman: () the relationship of total serum sIgE levels to reac- tivity in the skin prick test (SPT) and to levels of asIgE, (2) the range of allergen reactivity in the Omani patients and (3) the extent of concordance between the SPT and the asIgE tests. P A T I E N T S A N D M E T H O D S SUBJEC TS Forty four patients with asthma [7, male; 27, female; 33.7 y±5.4 (mean±SD age)] seen routinely in the asthma clinic in Sultan Qaboos University Hospital, Muscat, were studied. The history of allergic asthma was estab- lished by full clinical history and examination. The standard Bencard questionnaire was applied to all the patients included in the study, and skin tests were performed using the panel of Bencard skin test aller- gens. In addition, sIgE and in vitro asIgE levels were also measured. Control groups were: () 27 subjects [3, male; 4, female; 33.0 y±4.] with previous history of allergy, but currently asymptomatic, the Asymptomatic Allergic Subjects (AAS); (2) 9 [, male; 8, female; 29.3y±8.5] Normal Control Subjects (NCS). M E T H O D S . SERUM IGE This was measured using a micro particle enzyme immunoassay6 in the IMX system (Abbot Laboratories, IL 60064, USA). 2. ALLERGEN-SPECIFIC IGE IN VIV O TEST: THE SKIN PRICK TEST (SPT) This was measured using the Bencard system, using a panel of 25 allergens. Positive reactions in this test were assessed as wheal-and-flare reactions greater than 3 mm in patients who also gave positive reactions on provoca- tion with histamine (positive control). 3. ALLERGEN-SPECIFIC IGE IN VITRO TEST: THE IMMUNOCAP TEST AsIgE was measured using the automated ImmunoCAP system (Pharmacia and Upjohn Diagnostics, Uppsala, Sweden). Undiluted test sera and standards were incu- bated with allergens covalently coupled to a CNBr- activated cellulose derivative (the ImmunoCAP) for 30 minutes. β-galactosidase conjugated anti-IgE mono- clonal antibodies against a range of allergens were added. The mixture was incubated for a further 50 minutes dur- ing which time a complex was formed. Unbound enzyme anti-IgE was washed away and the bound complex was incubated with a developing agent, 4-methylumbellif- A L - A M R I E T A L 24 10 100 1000 Healthy (n=19) AAS (n=8) Asthma (n=44) sI g E g eo m et ri c m ea n (IU /m l) Bars represent geometric mean values sIgE, with 95% confidence limits Figure 1. Ranges of sIgE levels in normal control subjects, asymptomatic allergic subjects and patients with asthma. Bars represent geometric mean values sIgE, with 95% confidence limits 7 lar frequency of response to the ImmunoCAP allergens Euroglypus maynei and Hollister-Stier Laboratories dust preparation [Table a]. By contrast, the SPT allergen house dust, (73% response) showed a similar frequency of response to each ImmunoCAP individual and mixed preparations of the Dermataphagoides species allergens, and also to the Euroglypus maynei and the Hollister- Stier Laboratories dust preparations. SKIN PRICK TESTS . Frequency of response The frequencies of positive responses to SPT allergens are shown in Table 2. The highest frequency was to house dust, where 73% of subjects responded, followed by cat fur (43%), house dust mite and lobster (each at 38%), mixed threshing (36%) and maize (34%). There was a less than 20% response rate for the majority of allergens. When SPT allergens were examined on an individ- ual basis, there was no significant difference in the fre- quency of responses for the majority of allergens between those patients with serum IgE levels ≥ 0 and <0 IU/ ml respectively. The one exception to this was the aller- gen house dust mite, to which there was an increased response at sIgE levels ≥ 0 (χ2= 4.37, p<0.05). 2. Association between sIgE level and positivity in SPT The overall positive response in the SPT was defined as reactivity to at least 3/25 of the SPT allergens in the panel, and 34/44 patients (77%) were positive and 0/44 patients (23%) were negative in this test. Of the 38 patients with sIgE ≥0 IU/ml, 32 (84%) gave a positive response in the SPT, compared with 2/6 (33%) patients with sIgE <0 IU/ml—χ2=7.64, p<0.0. Of the 0 patients with an overall negative response in the SPT, 4 had sIgE <0 IU/ ml and 6 had sIgE ≥ 0 IU/ml. Four patients gave zero responses in the SPT; their sIgE levels were 467, 276, 224 and 57 IU/ml [Figure 2]. There was an overall positive correlation between sIgE level and the number of aller- gens reactive in the SPT (r=0.54, p<0.00) [Figure 2]. ALLERGEN SPECIFIC IGE . Frequency of response For these tests, data is available for both the AAS group and the asthmatic patients. The frequency of responses to ImmunoCAP allergen mixes is shown in Table 3. With regard to airborne allergens, 74% of asthmatic patients responded to the ImmunoCAP allergen, House dust mix, the same proportion as responded to house AT O P Y I N O M A N I PAT I E N T S W I T H A S T H M A eryl-β-d-galactoside. After 0 minutes, the reaction was stopped and the fluorescence detected in a fluorimeter. The developed fluorescence was directly proportional to the concentration of allergen specific IgE in the serum.7 For individual allergens, standardisation permits calcu- lation of IgE levels in allergen specific units. The system is calibrated over six ranges: <0.35 kUa/l (not atopic) to >00 kUa/l (strongly positive). For allergen mixes, only a positive/negative result is given. STATISTICAL ANALYSIS The frequency of responses to SPT and ImmunoCAP allergens was compared by χ2 analysis. Correlations between sIgE levels and allergen-specific IgE data were made using Pearson’s product-moment correlation coef- ficient and regression analysis. The significance of the differences in log sIgE levels between subject groups was compared by t-test. Differences in quantitative reactions to ImmunoCAP allergens between different groups of subjects were made using the Mann-Whitney U test. Ranges of sIgE levels are shown as the 95% confidence interval (CI) of the geometric mean. R E S U L T S TOTAL SERUM IGE LEVELS Total sIgE levels for the three groups are illustrated in Figure . The geometric mean and 95% CI for the asth- matic patients were 468 IU/ml and 323–676. This was significantly higher than for either the AAS—30 IU/ml (3–68) (t=5.73, p<0.00)—or the NCS—68 IU/ml (4– 0) (t=5.79, p<0.00). There was no significant difference between these latter two groups. In this study therefore, the upper limit of the normal range has been set at <0 IU/ml. While 38 asthmatic patients had sIgE levels above this limit—632 IU/ml (45–887)—there were 6 patients in whom the sIgE levels were within the normal range— 67 IU/ml (54–83). COMPARISON OF ALLERGENS USED IN THE IN VIVO AND IN THE IN VITRO TESTS FOR ATOPY A similar frequency of positive responses was shown for the majority of allergen pairs analysed in the in vivo and in vitro systems (p value of χ2>0.05) [Table a]. The only exception to this was the SPT allergen house dust mite, to which there was a 38% response. This was significantly lower than the response frequency of greater than 70% for each of the three ImmunoCAP Dermataphagoides species allergens and the ImmunoCAP House dust mix (p values of χ2<0.05) [Table b], but it showed a simi- 8 dust in the SPT. However, when the response to indi- vidual ImmunoCAP allergens was analysed, it became apparent that the highest proportion (73%), of subjects responded to Dermatophagoides pteronyssinus and only 44% responded to the ImmunoCAP house dust prepara- tion, (Hollister-Stier Laboratories) i.e the reverse of the situation seen with SPT allergens [Table 2]. The response was greater than 50% for other airborne allergens. The numbers of asthmatic patients tested with the food aller- gen mixes was very low, so the results can only be inter- preted with extreme caution. For all airborne allergens, and for nut mix, but not for cereal or for fish mixes, there was a significantly higher proportion of positive responses in asthmatic patients compared with the AAS group [Table 3]. 2. Association between sIgE level and positivity to ImmunoCAP mixes Due to the limited numbers there was no discrimina- tion in the proportions of those patients with high and those with low sIgE levels reacting to allergen mixes in the ImmunoCAP test [Table 4]. However, when those patients with sIgE ≥ 0 IU/ml were considered sepa- rately, it is apparent that positive results to ImmunoCAP allergen mixes occurred in the following proportions of patients: Food 3/6 (8%) House dust 6/22 (72%) Moulds 5/8 (63%) Tree pollen 6/0 (60%) Grass 8/2 (38%) Thus, apart from grass allergen mix, levels of sIgE ≥ 0 IU/ml are supported by reactivity to ImmunoCAP aller- gen mixes in a diagnosis of atopy. COINCIDENCE OF RE AC TIVIT Y TO SPT AND TO IMMUNOC AP ALLERGENS IN INDIVIDUAL PATIENTS Data was available from 3 asthmatic patients for both tests. Positivity was again defined as reactivity to ≥ 3 SPT allergens and to ≥ ImmunoCAP allergen. Twenty three (74%), 2 (67%) and 9 (58%) patients were posi- tive in the SPT, in the ImmunoCAP and in both tests respectively. These data, together with the mean sIgE levels for each group, are shown in Table 5. This illus- trates very clearly that SPT positivity is associated with high levels of sIgE, while negativity in the SPT is asso- ciated with low levels of sIgE. A significantly higher proportion of the patients positive in SPT had a posi- tive response in the ImmunoCAP compared with those negative in the SPT (χ2=9.04, p<0.0). That is, there is an overall tendency for the SPT and the ImmunoCAP to predict in the same direction. The levels of sIgE were similar in ImmunoCAP positive and negative subjects, both in SPT positive (t=0.28, df=2, p>0.05) and nega- tive (t=0.23, df=6, p>0.05) patients. A L - A M R I E T A L 25 Figure 2 0 0.5 1 1.5 2 2.5 3 3.5 4 0 2 4 6 8 10 12 number of positive skin prick tests lo g s lg E r=0.54, p<0.001 Horizontal line represents upper limit of normal range for IgE Horizontal line represents demarcation of overall negative and positive results in SPT Figure 2. Correlation between sIgE levels and degree of positivity in the skin prick test. The horizontal line represents upper limit of normal range for IgE. Vertical line represents demarcation of overall negative and positive results in SPT. 9 D I S C U S S I O N Total sIgE levels and allergen-specific tests, both in vivo and in vitro, are commonly used to diagnose atopic dis- eases. At the present time, there is no established data on either sIgE levels or the immune status of asthmatic patients in this country. Identification of triggering aller- gens in the Gulf region is important for the informed clinical management of allergic diseases. This is particu- larly so where desensitisation therapy is considered. It is equally important to establish the clinical efficiency of in vivo and in vitro diagnostic tests for allergic diseases. The present study compares methodologies used in the diagnosis of asthma in adult Omani patients attending the Asthma Clinic at Sultan Qaboos University Hospital in Oman; we report here the results of asIgE tests in the context of total sIgE levels in the study population. The mean sIgE level in healthy Omani sub- jects was 68 IU/ml with a range of 4 to 0 IU/ml. These levels are comparable to those of other populations.8,9 The asIgE data in this study has been analysed in the context of a normal range for sIgE of <0 IU/ml. The mean level for the AAS group, 30 IU/ml with a range 3 to 68 IU/ml, was lower than that of the healthy Omani subjects; only one of the AAS had an sIgE level above the normal range. It is perhaps of relevance here that the entire AAS group were expatriates. Hence the dif- ferences found may illustrate a real difference between the two control groups, since varia- tions in sIgE may occur due to environmen- tal factors such as diet and parasite burdens related to geographical location of origin. The level of sIgE in the healthy Omani population can only be validated by increasing the size of the cohort studied. Such studies are currently in progress. In agreement with previous findings,8 the geometric mean (469 IU/ml) and the range (323–676 IU/ml) of sIgE were increased in the Omani asthmatic patients compared to healthy control subjects. Since sIgE contrib- utes significantly to the outcome of skin tests, the clinical significance of these sIgE levels and the interpretation of the results from the allergen-specific tests can only be assessed in context of the normal range for this popula- tion. However, in one study, skin reactivity was Table I a. SPT and ImmunoCAP allergens showing similar frequencies of responses on testing in asthmatic patients ImmunoCAP allergens SPT allergens Gx1: grass (mix) grass, hay dust, straw, mixed threshings t23: cupressus trees t72: queen palm trees Tx7: Tree (mix) trees m3: Aspergillus A. niger, A. fumigata, hay dust, straw, mixed threshings m6: Alternata A. alternata, hay dust, straw, mixed threshings Mx2: Mould (mix) A. alternata, hay dust, mixed threshings d74: E. maynei house dust, house dust mite h2: Hollister-Stier labs house dust, house dust mite Dl: D. pteronyssinus house dust d2: D. farinae house dust d3: D. microcereus house dust Hx2: House dust mix house dust el: cat cat fur e3: horse horse hair e4: cow cow hair e81: sheep sheep wool e85: chick feathers fi: egg egg f2: milk milk, chocolate f4: wheat wheat f8: maize maize f80: lobster lobster Fx2: Fish mix cod Fx3: Cereal (mix) wheat, maize For each ImmunoCAP allergen shown above there was no significant difference in frequency of response (p values of χ2>0.05) by comparison with related SPT allergens. Table 1 b. SPT and ImmunoCAP allergens showing different frequencies of response in asthmatic patients ImmunoCAP allergens SPT allergen Dl: D. pteronyssinus 73% house dust mite 38% d2: D. farinae 73% d3: D.microcereus 77% Hx2: House dust mix* 72% There was a significantly higher frequency of response (p values of χ2 <0.05) to each of these ImmunoCAP allergens than to the SPT allergen “house dust mite.” *Hx2 contains dl, d2, h2 and i6 (Blatella germanica) allergens. AT O P Y I N O M A N I PAT I E N T S W I T H A S T H M A 20 noted in 35% of asthmatic patients in the presence of low levels of sIgE.10 Similarly, six patients in this study had sIgE levels <0 IU/ml, and two of them (33%) showed positivity in the SPT. Where there is the coincidence of a low level of sIgE and positivity in the SPT in an indi- vidual patient, difficulties may occur in the classification of the asthma. Positive SPT responses and allergen-specific IgE anti- bodies are both important markers of disorders in the upper respiratory tract. However, studies of their inter- active functioning are seriously hampered by differences in the commercial antigen extracts available for use in the two tests. At the outset of this study, we established comparability of response between the two antigen sys- tems used for a wide range of allergens. We employed the Pharmacia ImmunoCAP system for in vitro asIgE meas- urements and the Bencard system for in vivo SPT testing. Both systems are dependent on the quality of allergen extracts used in the tests. Quality control is difficult and in practice it can only be achieved by correlation with clinical history data by an experienced allergologist. Lack of standardisation between allergens of different manufacturers is also a major concern.11 The introduc- tion of recombinant allergens, whilst encouraging in this respect12 brings further problems. Due to the high spe- cificity of B- and T-cell responses, the number of indi- vidual patients responding to a recombinant allergen will be restricted compared to the number responding to a heterogeneous allergen prepared from natural sources, because such allergens contain multiple epitopes and different isoforms.12 Hence, the use of recombinant allergens may lead to underdiagnosis of allergic patients. Studies in patients from the Gulf region should ideally be performed using allergens prepared locally. These are not available at present. In their absence, it is difficult to draw definitive conclusions on the relative diagnos- tic efficiencies of assay modalities from studies using dif- ferent commercial allergens. Variations between them may reflect affinity and epitope recognition patterns within individual IgE responses rather than imprecision in the test. A further complication is that in vivo and in vitro tests monitor different aspects of asIgE, tissue- bound (i.e. stable over many months) in the SPT, ver- sus serum levels (with a half-life of only 2 days) in the ImmunoCAP. Despite these factors, we found no statistical differ- ence in response between the majority of comparable allergens in the ImmunoCAP and the SPT, though other studies suggest a higher overall frequency of response in the SPT.14,15 This is probably a reflection of the more per- sistent biological life of tissue bound IgE. The only major discrepancy in this study between comparable allergens was between the house dust and house dust mite aller- gens. In agreement with another reported study, the ImmunoCAP was more sensitive for house dust mite allergens.16 By contrast, the SPT was more sensitive for the house dust allergen. These data are suggestive of a crossover in specificities of these particular allergens A L - A M R I E T A L Table 2. Skin prick test (SPT ) response to allergens in asthmatic patients in relation to total sIgE level Allergen Positive response (n=44) Positive response in patients with sIgE <101IU/ml (n=6) ≥101IU/ml (n=38) Cat fur 19 (43%) 2 (33%) 17 (45%) Cowhair 4 (9%) 0 (0%) 4 (11%) Feathers 10 (23%) 0 (0%) 10 (26%) Horse hair 7 (16%) 1 (17%) 6 (16%) Human hair 6 (14%) 0 (0%) 6 (16%) Sheep wool 13 (29%) 0 (0%) 13 (34%) House dust 32 (73%) 3 (50%) 29 (76%) Hay dust 12 (27%) 1 (17%) 11 (29%) House dust mite* 17 (38%) 0 (0%) 17 (45%) Cotton 6 (14%) 1 (17%) 5 (13%) A.alternata 6 (14%) 0 (0%) 6 (16%) A.fumigata 5 (11%) 0 (0%) 5 (13%) A.niger 3 (7%) 0 (0%) 3 (8%) Grass 13 (30%) 1 (17%) 12 (32%) Mixed threshings 16 (36%) 1 (17%) 15 (39%) Trees 3 (7%) 0 (0%) 3 (8%) Straw 9 (20%) 1 (17%) 8 (21%) Cheese 4 (9%) 1 (17%) 3 (8%) Chocolate 1 (2%) 1 (17%) 0 (0%) Cod 0 (0%) Egg 2 (4%) 0 (0%) 2 (5%) Lobster 17 (38%) 3 (50%) 14 (37%) Maize 15 (34%) 3 (50%) 12 (32%) Milk 0 (0%) 0 (0%) Wheat 1 (2%) 1 (3%) Histamine 44 (100%) 6 (100%) 38 (100%) *Using a cut off point of 101 IU/ml, sIgE levels do not differentiate frequency of SPT response to individual allergens, except, in the case of “house dust mite”, which is only responsive in the SPT at sIgE levels ≥101 IU/ml (χ2=4.37, p<0.05). 2 from the two different systems. There is as yet no agree- ment as to whether the SPT and the ImmunoCAP tests are17 or are not18 complementary to each other. In this study, 34/44 (77%) of patients gave positive responses to the panel of allergens in the SPT, and the association of these positive responses with raised levels of sIgE9 was confirmed. Of the 0 (23%) patients negative in the SPT, 6 had sIgE levels ≥ 0 IU/ml and 4 had sIgE levels <0 IU/ml. Six out of 44 patients (4%) had sIgE levels <0 IU/ml, and 2 of these were positive in the SPT. Reactivity to individual allergens in the SPT was associ- ated with a large range of values of sIgE levels in different patients e.g., amongst those subjects with reactivity to 4 SPT allergens, the range of sIgE levels was 99–5,52 IU/ ml [Figure 2]. It has been suggested that where high lev- els of sIgE are found in atopic patients, they additionally reflect immune dysregulation. IgE itself may be involved in allergen uptake via the CD23 receptor (FceR).5 Such IgE-mediated allergen presentation may lead to a con- tinuous (over) activation of the immune system due to high levels of IgE. This could explain the deterioration in the clinical condition of children with asthma, which is characterised both by increased severity of symptoms and sensitivity to an increased number of (non related) allergens.5 Four patients in this study showed no reactivity to any SPT allergens. Three of them had sIgE levels above the normal range. This apparent anomaly may merely be a reflection of the lack of knowledge of the precise amount of the relevant allergen required to provoke an SPT reaction in each individual patient, that is, inadequate technical understanding of the fundamental aspects of the tests. It has been shown that with similar levels of allergen-specific IgE, the amount of allergen required for a positive SPT may differ by as much as a factor of 00 between patients.19 Thus, there may be a threshold amount of allergen required to provoke a response for each individual; this will be elucidated only when puri- fied allergens become generally available. From this study, it is obvious that raised levels of serum IgE and positivity in the SPT both aid the diag- nosis of atopy, but the absence of these factors does not exclude it. Raised sIgE levels are thought to be predic- tive of the subsequent development of allergic disease in children,19,20 but in the adult, it does not appear that, alone, they invariably predict an allergic state. Genetic and environmental factors also play an important role in the production of clinical symptoms. Overall, the SPT is considered to have a better diag- nostic efficiency and sensitivity than in vitro methods,21,22 AT O P Y I N O M A N I PAT I E N T S W I T H A S T H M A Table 3. Proportions of asthmatic and AAS* subjects reacting to allergen mixes using the ImmunoCAP system Allergens Patient group AAS (%) Asthmatic (%) p value of χ2 Grass pollen mix 2/19 (11%) 9/24 (37%) <0.05 Tree pollen mix 3/20 (15%) 9/15 (60%) <0.05 Mould mix 2/19 (11%) 5/12 (42%) <0.05 House dust mix 9/27 (33%) 26/35 (74%) <0.05 Food (nut) mix 1/16 (6%) 3/3 (100%) <0.01 Food (fish) mix 3/18 (17%) 3/8 (38%) >0.05 Food (cereal) mix 3/17 (18%) 7/15 (47%) >0.05 *AAS: Asymptomatic allergic subjects Table 4. Numbers of asthmatic patients positive to ImmunoCAP mixes compared to their total serum IgE Allergen mix (n)* Response to sIgE concentration of <101 IU/ml ≥101 IU/ml χ2 # p nega- tive posi- tive nega- tive posi- tive Grass pollen (24) 2 1 13 8 0.23 >0.05 Tree pollen (12) 2 0 4 6 0.60 >0.05 Mould (12) 4 0 3 5 2.10 >0.05 House dust (25) 1 2 6 16 0.20 >0.05 Food mixes combined (18) 1 1 3 13 0.00 >0.05 *Number of cases tested #Non-parametric χ2 for two independent samples Table 5. Coincidence of reactivity in SPT and ImmunoCAP Test result n Geometric mean sIgE (IU/ml) and (95% CI) Positive SPT +ive immunocap –ive immunocap 23 19 4 776 (519–1161) 796 (512–1245) 684 (225–2080) Positive SPT +ive immunocap –ive immunocap 8 2 6 166 (67–302) 169 (30–1146) 156 (77–315) 22 and it has the advantage of providing immediate infor- mation for both patient and physician. The high degree of sensitivity offered by the SPT is important when a patient is assessed for potentially life threatening aller- gies, for example, from penicillin or stinging insects. It is worth emphasising, however, that in a minority of patients the expected association of sIgE level with SPT reactivity does not occur. As previously reported, some patients were found with sIgE levels in the normal range, but with multiple SPT reactivity,10 and the converse, where some patients had raised sIgE levels but little or no SPT reactivity. These findings are suggestive of false positive and false negative reactions in the SPT, and they emphasise the care that must be taken in the categorisa- tion of such patients. It is most important that a detailed clinical history is taken, as a third component of the diag- nostic procedure. The subjects in the AAS group had, as expected, low levels of sIgE and reduced ImmunoCAP reactivity compared with the asthmatic patients. C O N C L U S I O N We have described here a population of asthmatic patients in Oman who have a sIgE profile comparable to that found in similar studies.9 Estimation of sIgE may be helpful in the diagnosis of atopic asthma, but it has poor correlation with symptoms. In individual patients there is a good correlation between sIgE level and the degree of overall reactivity to the panel of allergens in the SPT, irrespective of ImmunoCAP reactivity. sIgE levels, in combination with the SPT, should therefore be used in the routine diagnosis of allergy in the majority of asthmatic patients who have no complications from skin conditions or medication or who are considered so sensitive by history that anaphylaxis is possible. In cases where these types of difficulties are found, use of the ImmunoCAP may be useful since its wide range of measurement permits a good discrimination of nega- tive cases from those with atopic disease. Due to non- specific binding, high levels of sIgE may interfere with both in vivo and in vitro assays of asIgE. In view of the occasional occurences of false positive and false negative reactions in the Skin Prick Test (SPT), its results should always be interpreted in the context of a detailed clini- cal history. R E F E R E N C E S . ISAAC Steering Committee. Worldwide variation in preva- lence of symptoms of asthma, allergic rhinoconjunctivitis and atopic eczema. Lancet, 998, 35, 225–232. 2. Al-Riyami MSB, Al-Rowas OAS, Al-Riyami AA, Jasim LG, Mohamed AJ. Prevalence of asthma symptoms in Omani schoolchildren. SQU J Sci Res: Med Sci, 200, 3, 2–27. 3. Kroegel C, Virchow J-C, Luttmann W, Walker C, Warner JA. Pulmonary immune cells in health and disease: the eosinophil leucocyte. Eur Resp J, 994, 7, 59–543. 4. Smith L, McFadden ER. Bronchial hyperreactivity revisited. Ann Allergy, Asthma and Immunol, 995, 74, 454–469. 5. Mudde GC, Bheekha R, Bruijnzeel-Koomen CAFM. Con- sequences of IgE/CD23 mediated allergen presentation in allergy. Immunol Today, 995, 6, 380–382. 6. Yman L, Roosdorp N, Schroder H, Andrae MI. Methods for the determination of IgE and allergen-specific antibodies. International Allergy Symposium, Excerpta Medica, 98, 67, 74–83. 7. Paganelli, R. Ansotegui J, Sastre J, Lang CE, Roovers MH, de Groot H, et al. Specific IgE antibodies in the diagnosis of atopic disease. Clinical evaluation of a new in vitro test system, UNICAP TM, in Six European allergy clinics. Clin Allergy, 988, 8, 58–587. 8. John AB, Lee HS, Lee Fy, Chug HH. Allergen skin test and total IgE in adults with rhinitis in Singapore. Asian Pac J Allergy Immunol, 996, 4, 9–2. 9. Janson DF, Rijeken B, Schouten JP, Kraan J, Weiss ST, Timens W, et al. The relationship of skin test positivity, high serum total IgE levels, and peripheral blood eosinophilia to sympto- matic and asymptomatic airway hyperresponsiveness. Am J Respir Crit Care Med, 999, 59, 924–93. 0. Hoshina K, Kawasaki A, Mizushima Y, Yano S. Positivities of skin test and IgE RAST to allergens in bronchial asthma with normal serum IgE levels. Arerugi, 99, 40, 6–20. . Dolen WK. Allergen extract standardization: reality. myth or dream. Ann Allergy Asthma Immunol, 996, 75, 8–82. 2. Crameri R, Lidholm J, Gronlund H, Stuber D, Blaser K, Menz G. Automated specific IgE assay with recombinant allergens: evaluation of the recombinant Aspergillus fumigatus allergen I in the CAP system. Clin Exp Allergy, 996, 26, 4–49. 3. Hayglass KT. Allergy: Who, why and what to do about it. Immunol Today, 995, 6, 505–507. 4. Ewan PW, Coote D. Evaluation of a capsulated hydrophilic carrier polymer (the ImmunoCAP) for measurement of spe- cific IgE antibodies. Allergy, 990, 45, 22–29. 5. Gleeson M, Cripps AW, Hensley MJ, Wlodarczyk JH, Henry RL, Clancy RL. A clinical evaluation in children of the Phar- macia ImmunoCAP system for inhalant allergens. Clin Exp Allergy, 996, 6, 692–702. 6. Plebani M, Bourghesan F, Faggian D. Clinical efficiency of in vitro and in vivo tests for allergic diseases. Ann Allergy Asthma Immunol, 995, 74, 23–28. 7. Droste JH Kerhof M de Monchy JG, Schouten JP, Rijken B. Association of skin test reactivity, specific IgE, total IgE and eosinophils with nasal symptoms in a community-based pop- ulation study. The Dutch ECRHS Group. J Allergy Clin Immu- nol, 996, 97, 922–932. 8. Witteman AM, Stapel SO, Perdok GJ, Sjamsoedin DH, Jansen HM, Aalberse RC, et al. The relationship between RAST and skin tests in patients with asthma or rhinitis: a quantitative study with purified major allergens. J Allergy Clin Immunol, A L - A M R I E T A L 23 996, 97, 6–25. 9. Kjellman NI. Predictive value of high IgE levels in children. Acta Paediatric Scand, 976, 65, 465–47. 20. Haahtela T, Suoniemi I, Jaakonmaki I, Bjorksten F. Relation- ship between sIgE and the occurrence of immediate skin test reactions and allergic disorders in young people. Allergy, 982, 37, 597–602. 2. Ownby DR. In vivo versus in vitro. Pediatr Clin North Am AT O P Y I N O M A N I PAT I E N T S W I T H A S T H M A 1988, 35, 995–009. 22. Tschopp JM, Sistek D, Schindler C, Leuenberger P, Perruchoud AP, Wuthrich B, et al. Current allergic asthma and rhinitis: diagnostic efficiency of three commonly used atopic mark- ers (IgE, skin prick tests, and Phadiatop). Results from 8329 randomized adults from the SAPALDIA Study. Swiss study on Air pollution and Lung Diseases in Adults. Allergy, 998, 53, 606–63.