Immunotherapy with cat- and dog-dander extracts

Immunotherapy with cat- and dog-dander extracts

Immunotherapy with cat- and dog-dander extracts Ill. Allergen-specific 1-year double-blind immunoglobulin placebo study responses in a Henning Lsw...

895KB Sizes 0 Downloads 13 Views

Immunotherapy with cat- and dog-dander extracts Ill. Allergen-specific 1-year double-blind

immunoglobulin placebo study


in a

Henning Lswenstein, Ph.D., Dr.Sc.,* Viggo Graff-Lonnevig, M.D.,** Gunilla Hedlin, M.D.,*** Henrik Heilborn, M.D.,**** Gunnar Lilja, M.D.,** K. Norrlind, M.D.,**** K-O Pegelow, M.D.,*** and Bo Sundin, M.D.*** Copenhagen,


and Stockholm,


An investigation was made of the specific antibody response to individual antigens in 40 patients taking part in a double-blind placebo study of immunotherapy with cat- or dog-dander extracts. Antigen-specific IgE levels were measured by means of CRIE, and the results were expressed as scores. The patients demonstrated IgE speci$cities toward 1 to 5 antigens. Cat-dander antigens Nos. 4 (cat Ag I) and 7, and dog antigens Nos. 6 and 13, produced the highest scores, but high IgE binding was also found for dog albumin. After 1 year of treatment, the IgE responses of the two treatment groups (allergen and placebo) were statistically indistinguishable from those before the start of treatment. There was a tendency toward a reduction in score for two of the dander antigens (cat Ag I and dog Ag No. 13) and no new IgE specificities appeared. Antigen-specific IgG levels were measured by means of CIE with patient serum incorporated in an intermediate gel, and the results were expressed as pluslminus precipitins. Only two patients had precipitating antibodies before the start of treatment (one against cat albumin and one against cat Ag I). During the course of treatment, the production of antigen-spec$c IgGs was observed in 18122 allergen- and II18 placebo-treated patients. For the cat allergentreated group, the speci$cities were directed against cat antigens Nos. 2, 3 (cat albumin), 4 (cat Ag I), and 7, and for the dog allergen-treated group, against dog antigens Nos. 2 (dog albumin), 13, and 20. Before and 1 year after the start of treatment, the patients’ homes were subjected to dust analyses for content of cat and dog antigens. These revealed that there was no statistically signijcant difference between the allergen-treated and placebo-treated groups, either before the start of treatment or after I year of treatment. The range of cat Ag I and of dog albumin found was 100 to 100,000 nglgm of dust. (J ALLERGYCLIN IMMUNOL 77:497-50.5, 1986.)

Extracts of cat and dog hair and dander are known to be complex antigenic mixtures.’ Patients’ IgE responsestoward the individual antigenspresentin these materials have been examined in a number of laboratory investigations213 by meansof CRIE. The patients’ antigen-specificIgG responsescanbe measured by IgG CRIE4 or by the technique of CIE with intermediate gel.’ Basedon a l-year double-blind study involving 40 From the *Protein Laboratory, University of Copenhagen, Copenhagen, Denmark, and ****Dandetyd, ***Huddinge, and **Saschka Hospitals, Stockholm, Sweden. Received for publication March 21, 1985. Accepted for publication Aug. 1, 1985. Reprint requests:Henning Lowenstein, the ProteinLaboratory,University of Copenhagen,Sigurdsgade34, DK-2200 Copenhagen, Denmark.

Abbreviations used CIE: Crossedimmunoelectrophoresis

CRIE: Crossedradioimmunoelectrophoresis QIE:

Quantitative immunoelectrophoresis

cases(accompanyingarticle, page 478), it was demonstrated that specific immunotherapy (hyposensitization) with partially purified cat- or dog-danderextracts had a statistically significantly greaterbeneficial effect on the patients’ bronchial sensitivity than treatment with placebo extracts. Similarly, in vivo and in vitro allergen-specificresponsesthroughout the entire treatment course revealed statistically significant differencesbetweenthe allergen- and placebo-treated patients with respectto total allergen-specific IgE re497

498 [email protected] et al.

sponses (as measured by skin prick tests and RAST) and IgG responses (as measured by various IgG RASTs) (accompanying article, page 488). The present study was concerned with the patients’ antigen-specific IS; responses (measured by CIE with an intermediate gel incorporating patient serum), antigen-specific IgE responses (measured by means of CRIE), and with analyses for cat or dog antigens in the dust from the patients’ homes. The purpose of this investigation was to compare the antigen-specific IgG and IgE levels before and during treatment and to examine their potential as parameters for immunologic. posttherapeutic control of hyposensitization, and to examine whether any new IgE or IgG specificities appeared during the course of treatment. Furthermore. it was essential for the evaluation of this immunotherapy study to establish the immunologic specificities of the patient groups and to know whether any changes had occurred in the levels of the specific antigens that the patients were exposed to in their homes. MATERfAt. Antigens


The batches of standardized. partially purified cat- and dog-allergen extracts (Soluprick SQ. ALK, Copenhagen, Denmark) described previously, page 488, were used. The identification and quantification of known cat allergens’ ’ and dog ailergens’. ’ were made by QIE relative to internal standards (DAS-76). The following cat antigens were identified: cat Ag Nos. 2, 3 (equal to cat albumin), 4 (equal to

cat Ag 1). 5. 6. 7. 8. Il. 17. and 18; and dog Ag Nos. 2 (dog albumin). 3. 6, 13, X. and 20. The ratio of cat Ag 1 per cat albumin was determined to be 0.092 (w/w) by rocket immunoclectrophoresis with use of cat 0580 as a standard’, the amounts ofcat albumin being 0. I2 mg/mg of dry weight. ‘Phc content of dog albumin in the dog-allergen extract was determined to be 0.20 mgimg of dry weight by means of rocket immunoelectrophoresis by use of purified dog albumin as standard.’ Total number of antigens in the catallergen extract was 15, and 16 in the dog-allergen extract. Only antigens eliciting rise to IgE and/or IgG responses in the patients by the methods used will be discussed.

Antibodies Antibodies against hair and dander allergens of cat and dog were raised by immunizing each of three rabbits with the DAS cat- or dog-hair and dander extract,’ respectively. The immunization and subsequentisolation of the immunoglobulins was performed according to the method of Harhoc and Ingild.” During this isolation. the immunoglobulins were concentrated fivefold.

QIE QIE analyses, namely, CIE with an intermediate gel, and CRIE were performed as described recently’ and as specified in the Figures.



CIE CIE was performedon 1% agarosegel containing 0.073 mol/L of Tris, 0.025 m&L of barbital, 0.0006 mol/L of calcium lactate, and 0.0003 mol/L of sodium azide (pH 8.6, 25” C) at 15” C by use of 10 V/cm for 30 minutes in the first dimension and 2 V/cm for 15 hours in the second dimension. Anodes were to the right for the first dimension, and at the top for the second dimension. An intermediate gel containing no antigen or antibodies was placed between the first dimension gel and the anodic antibody-containing gel in order to improve resolution in the second dimension. For individual antigens the sensitivity of the methods is about IO to 100 rig/ml with a coefficient of variation 1?I3% to 8%.’



Precipitating antibodies against the individual antigens were determined by the CIE with intermediate gel technique.’ Conventional CIE was performed with 200 ~1 of patient serum incorporated in the intermediate gel. A pool of serum from 1000 normal individuals served as control. Before disclosing the code revealing whether treatment had been administered with allergen or with placebo, the CIEs were evaluated independently by two persons for the presence of spurs and “inward feet,” which indicate the presence of precipitating antibodies in the patient sera that crossreact with the corresponding, heterologous rabbit antibodies.5 A 100% agreement between the two evaluations was obtained.

CRlE CRIE autoradiographswere performed from day 0 to day 3, from day 3 to day 4. and from day 4 to day I 1, corrcsponding to 3, I, and 7 days, respectively, or for 3 days only. The degree of radiostaining of individual precipitates was evaluated by visual comparison of CRIE autoradiographs (140 nCi ?-labeled anti-IgE) with autoradiographs of mancini rings with twofold serial dilution of “‘I-antiIgE.” For each of the nine mancini rings, a score number Nx was calculated from the equation

where D is the dilution factor for “‘I-anti-IgE. For each CRIE autoradiograph, the score Nx for the individual prccipitates was evaluated independently by at least two pcrsons. The scores were corrected for nonspecific radiostaining by use of a pool of sera from normal individuals. Examples of CRIE autoradiographs for individual patients. allergoprints, are presented in Fig. 4.

Dust analyses Dust analyses were performed by counter-current immunoclectrophoresis according to the method of Schwartz et al.Y at the Diagnostic Laboratory, ALK, Copenhagen, Denmark. Dust samples (I gm), collected before the start of and I year after hyposensitization, were extracted, the extracts were freeze-dried, and twofold dilutions were subjected to clectrophoresis against purified species-specific


77 3


rabbit anticat and antidog antibodies. The results are expressedas titers corresponding to the dilutions. The correspondingabsoluteamountsof cat Ag 1” and dog albumin* were also calculated.


with cat and dog dander


Statistics All comparisons within the same group of patients at different times were made with the Wilcoxon matchedpaired signed-rank test. Comparisons between groups of patients at the same time were made by Mann-Whitney statistics. All stochasticprobabilities CO.025 were considered significant.




The 40 patients participating in this study have been described in a preceding article, page 478. The antigenspecific tests reported here were carried out for all patients independentof whether or not the patients had been diagnosedas cat or dog allergic. Thus, the following subgroups of patients will be discussed: cat allergen-treated, catallergic placebo-treated, and cat-allergic, untreated, the last group being identical with the dog allergen- plus dog-allergic placebo-treatedgroup. The samesubgroupsare applicable to the dog-allergic patients, i.e., dog allergentreated, dog-allergic placebo-treated,and dog-allergic untreatedgroups of patients. Matching

of patients

by means of CRIE

In the run-in period, serafrom all the 45 selectedpatients were submittedto both cat and dog CRIE. The allergoprints for those patients who consideredcat to be most relevant (accompanyingarticle page478) were compared,and those pairs of patients who gave a maximum similarity in CRIE responsewere selected.The sameprocedurewas performed for the patients who considered dog to be most relevant. Most often several solutions were possible, and the pairs could therefore also be selectedso that they had the same sex, approximately the same age, clinical history, and threshholdin bronchial challenge. Finally, the two treatment alternatives, allergen and placebotreatment,were allocated randomly within the pairs. RESULTS Dust analyses Dust samples were collected in all patients’ homes, as far as possible, and analyzed for content of both cat and dog antigens ( 151/ 160 tests resulted). The dust analyses revealed that 10 of 40 and seven of 40 homes were devoid of detectable cat and dog antigens in 1982 and 1983, respectively. The distribution of cat and dog antigens in the homes of the various groups of patients (allergen-treated, cat-allergic; placebotreated, cat allergic; allergen-treated, dog-allergic; and placebo-treated, dog allergic) was, however, quite similar (Fig. l), and no statistically significant dif-

*[email protected]: Unpublished.












e, 0

rn 2


32 128 512



32 128 512 Titer

n Dog Ag‘s , 0 Cat Ag‘ s FIG. 1. The distribution of cat- and dog-antigen levels in dust samples from the patients’ homes. All homes were tested for both cat and dog antigens. A and P refer to homes of allergen- and placebo-treated patients, respectively; 82 and 83 refer to before and after 1 year of treatment, respectively.

ferences were found between the allergen- and placebo-treated groups (Mann-Whitney) nor between antigen levels detected in 1982 and 1983 for the same homes (Wilcoxon). The level of cat antigenic dust corresponds to 100 to 100,000 ng of cat Ag 1 per gram of dust and of dog antigenic material to 100 to 100,000 ng of dog albumin per gram. Precipitating


An example of CIE with intermediate gel-containing patient sera drawn before and after allergen treatment with cat-allergen extract is presented in Fig. 2. The results are summarized in Fig. 3 in which only those antigens eliciting rise to the production of precipitating antibodies and present in the treatment extracts are included. Precipitating antibodies were detected only in the allergen-treated group with the exception of two patients in the cat-allergic group who had precipitating antibodies against cat albumin and cat Ag 1, respectively, before the start of the hyposensitization. The specificities of the precipitating antibodies detected were directed against cat Ag Nos. 2, 3 (cat albumin), 4 (cat Ag l), and 7, and dog Ag Nos. 2 (dog albumin), 13, and 20.




et al.


ffi. 2. Determination of precipitating antibodies against individual cat antigens by means of CIE with patient serum incorporated in the intermediate gel. The quantity of cat-dander extract applied was 30 &g and 5 to 7 pi/cm of anticat-dander antibodies were incorporated in the antibody-containing gel. Patient serum (200 ~1) from A (before start of treatment) and from 6 (after 1 year of treatment) was incorporated in the intermediate gels. Experimental conditions, otherwise, are as described in the text. The presence of precipitating antibodies against Ag 4 (cat Ag 1) is indicated by the spur (marked by an arrow). The same spur was not present in A.

labeled with 3 in the CIE in Fig. 4, I) and dog albumin

No of Ptt


Clef?,labeled with 2 in the CIE in Fig. 4. I); however. Dog 82

Cat 82

10 i

Dog 83


2 -4-




L 13


G Placebo

FtG. 3. Distribution of patients’ precipitating antibodies against individual allergens in cat or dog extracts. The /efi part of the figure includes data for the cat allergenor placebo-treated patients and the right part data for the dog allergenor placebo-treated patients.

CRIE Examples of individual patients’ allergoprints in CRIE are presented in Fig. 3. All patients were tested simultaneously in both cat and dog CRIE, and it is clear from Fig. 4, A to H. that there is often a simultaneous uptake of IgE in the cat albumin (rig/u,

the intensity of the radiostaining of the albumins differs significantly, e.g., Fig. 4. B and H. Most of the patients reacted to allergens from both of the allergen sources, but some patients (Fig. 4, A and C) reacted with more IgE specificities to dog-dander allergens, whereas other cases reacted mainly to cat-dander allergens and with higher intensity, especially cat Ag 4 (cat Ag I) and 7 (Fig. 4. D, F. and G). There were also examples in which the patients’ IgE responses appeared to be directed mainly against the serum components (Fig. 4, E and H). The CRIE scores for all the selected patients before the start of treatment are illustrated in Tables I and II. These data (plus data for an additional five patients who, caused by external factors not related to this study, were excluded from the clinical study) (accompanying article, page 478) were used for allocation of the patients into two matched groups. The qualitative similarities between the allergoprints for the matched patients were compared with those of a random matching excluding the use of CRIE. This demonstrated that the weighted mean for paired patients had a difference in specificity of one allergen only. whereas the random distribution would have demonstrated a weighted mean difference of two allergens. It was also found that all the cat allergen-treated patients had high IgE responses toward antigen 4 (equal to cat


77 3


with cat and dog dander


FIG. 4, A-l. Simultaneous cat- and dog-allergen CRIE performed with sera of eight different catand/or dog-allergic patients. The dog antigens were applied to the left, the cat antigens to the right, and were run against their respective rabbit antibodies in the CRIE procedure. The corresponding Coomassie blue-stained CIE reference pattern is illustrated in I, in which the most important allergens are also marked. In CIE the amounts of cat and dog extract applied were 30 and 30 pg, and the amounts of anticat- or antidog-dander antibodies 5.4 and 4.3 r&cm, respectively. Experimental conditions were otherwise as described.6 The autoradiography was carried out for 3 days.

Ag 1) and that all the dog allergen-treated patients had high levels of IgE against the dog-danderantigen No. 13. The averageIgE CRIE scorestoward the individual antigens per number of responding patients are presentedin Fig. 5. This Figure includes data for all the 45 patients primarily selectedfor this study. The antigens most often eliciting rise to IgE binding were cat antigens Nos. 4 (equal to cat Ag l), 3 (equal to cat albumin), and 7, and dog antigens Nos. 2 (equal to dog albumin) and 13. The highest averageIgE CRIE scoreswere, however, obtained for the cat-danderantigens Nos. 4 (equal to cat Ag l), 5, and 7, and the dog antigens Nos. 6, 13, and X. The results of the treatment for the various patient groups expressedas mean CRIE scoresfor the individual allergens are presented in Figs. 6 and 7. In Fig. 6 the data for the cat-allergic group are listed,

andthe statistical analysiswith the Mann-Whitney test revealedthat there were no statistical differenceswith respectto the scoresfor any of the allergensbetween the cat allergen-treated, placebo-treated,or untreated groups with the exception of a difference in the level of IgE directed against cat Ag No. 4 (cat Ag 1) observed when the allergen-treated and untreated patients were compared. A comparison of the mean CRIE score for any of the antigens obtained before treatment and after 1 year of treatment demonstrated a clear, although it was not statistically significant, tendencytoward a decreasein the level of IgE against cat Ag No. 4 (cat Ag 1) in the allergen-treatedpatients. Similarly, in Fig. 7, the mean scoresfor the dogallergic patients were comparedfor the different treatment groups. Statistically significantly different IgE levels were found for Ag 6 and Ag 13 when the allergen-treatedand untreated groups were compared,

[email protected] L0wenstein et al.



T&4M.Et. Cat antigens ARtlganS

4 5 I iI 12 19 20 21 31 36

37 39 41 44 4x < >rn SEM

Placebo-treated patients No. 2 8 14 15 ii 22. 76 2x 34 12 47 40 $0 -:: .:'j)i SEM





4 0 4 2 2 0 0 8 0 8 0 0 0 I 0 1.9 0.73

3 0

3 3 3 0 0 7 0 7 3 3 0 0 1 2.1 0.62

32 4 32 32 64 8 8 32 2 64 32 16 4 x 32 24.7 5.2

4 4 0 4 0 2 0 8 2 2 0 2 0 2.2 0.66

3 I 3 3 0 3 0 7 7 3 0 7 I 2.9 0.70

32 2 32 32 8 32 I 32 32 32 8 32 4 21.s 3.9

and for Ag 2, 6, and X when the placebo-treatedwith untreatedgroups were compared.When the IgE binding in CRIE for the patients before and after 1 year of treatment was compared, a statistically significant reduction in the level of IgE with specificities directed againstdog-danderAg 13 was observedfor the group of allergen-treatedpatients. During the treatment period, a number of new IgE specificities were observed for both the cat allergicand dog allergic-treated groups. However, thesenew IgE specificities also appearedin the placebo-treated and untreatedgroups in amountsstatistically indistinguishable from those in the untreated group. A comparison between the summed CRIE score before and after 1 year of treatment for the allergen-





2 0 0 2 2 0 0 0 0 0 0 4 0 1 0 0.73 0.32

8 0 16 32 8 0 8 16 0 8 16 4 0 8 0 8.3 2.3

0 0 0 0 0 0

4 8 0 0 4 0 0 0 8 1.6 0.76

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

49 4 55 71 79 8 20 71 2 87 55 27 4 18 41

0 0 2 2 0 0 0 4 0 2 0 2 0 0.92 0.37

8 0 4 16 0 8 0 32 16 4 4 64 0 12.0 5.0

4 0 0 0 0 8 0 0 0 0 0 0 0 0.92 0.66

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

51 7 41 57 8 53 1 83 57 43 12 107 5

treated, placebo-treated,and untreated groups of patients was also made. It appearsthat this rather crude parameteris not changedby hyposensitization. Only for certain of the individual antigens were significant reductions in the IgE level in the allergen-treatedrelative to the placebo-treatedgroups observed (Figs. 6 and 7). DlSCUSSfON In two preceding articles in this issue of the JOURNAL, pages478 and 488, the results of a double-blind placebostudy of hyposensitizationwith extractsof cat or dog dander were presented.The main results were a reduced bronchial sensitivity combined with an increasein the allergen-specific IgG level for the aller-


77 3


with cat and dog dander


TABLE II. Dog antigens Antigens Allergen-treated patients No. 1 18 23 29 30 33 45 < >*t SEM









3 3 7 7 3 3 3 4.14 0.74

8 2 2 32 32 0 0 10.9 5.6

8 2 16 16 16 64 16 19.7 7.7

0 0 2 8 0 0 4 2.0 1.2

0 0 0 4 0 0 4 1.1 0.7

0 0 0 0 0 0 0 0 0

0 0 0 0 0 0 0 0 0

19 I 27 67 51 67 27

16 64 4 0 4 17.6 11.9

8 32 4 0 0 8.8 6.0

32 0 8 2 0 8.4 6.1

0 0 0 0 4 0.8 0.8

0 0 0 0 0 0 0

0 0 0 0 0 0 0

87 159 31 9 11

Placebo-treated patientsNo. 9 10 17 24 38 < >*t SEM

31 63 15 7 3 23.8 10.9

gen-treated patients relative to the placebo-treated groups. These results might have been influenced by changesin the levels of sensitizing material within the patients’ homes; dust samples from the latter were thereforeanalyzed, and no statistically significant differenceswere found for the dust-antigenlevels, either betweenthe various patient groups before the start of the study or within a given patient group before the start of the study as compared with the level after 1 year of treatment. We realize that thesedust analyses do not provide an accuratemeasureof the amount of dust reaching the patients’ lungs that will, of course, dependon the level of physical activity in the patients’ homes. However, it appearshighly probable that the averagelevels of physical activity for the various patient groups are statistically identical becauseof the very equal distribution in age and sex among the patient groups and becauseall patients were instructed to minimize any contact with cats or dogs before the start of this study (accompanyingarticle, page 478). The presenceof precipitating antibodiesagainstthe individual antigens was measured by a simple gel technique. The levels of precipitating antibodieswere rather low as comparedwith, for example, the level of precipitating antibodiesagainstcow’s milk antigens found in most normal individuals. I’ However, there was a statistically extremely significant differentiation between treated and nontreatedpatients, providing a clear indication that the injected antigens act immunologically in the patients. However, we do not know

whether this additional precipitation of the cat or dog antigens is due only to the patients’ IgG and IgG antibodies; it might also be influenced (inhibited or enhanced depending on their relative amounts) by the nonprecipitating antigen-specific IgEs and IgGs through their competition for the various epitopes.The responsestoward the individual antigens did not reflect the patients’ allergoprints, possibly caused by technical difficulties involved in uniform evaluation of the individual precipitates. It is, however, crucial that no precipitating antibodies were found having specificities against antigenstoward which the patient did not have IgE. Statistically significant changesin the levels of IgE directed against the various identified allergens were detectedonly in a few cases.Comparedto other studies, the general changes were rather small but we might well expect a further decreasein antigen-specific IgE levels after a longer treatmentperiod, as was the casein anotherhyposensitizationstudy.*However, the largestchangeswere found for the danderproteins toward which the highest numberof patientshavetheir IgE directed. Although the extracts used contain a high percentageof dry weight of cat- and dog-serum albumin, respectively, the IgE level was not changed very much, possibly becauseof a rather low significanceof theseproteins as allergens.‘* From theseresults the conclusion can be drawn that a high content of somedanderproteins (especially Cat Ag 1 and Dog Ag 13) must be included in allergenic extracts. Pa-




et al. c Dog CRIE score +,

CRIE ecore / + ve Pt





S Serum

















Antigen numbers are listed below the various


S indicates that the antigen is present in serum; +ve reactions lists the number of patients having specific IgE directed against the above antigen.

[ 25

CRIE scoreapt






IL- 4ll ua3







11 17

L 2




FIG. 7. Mean CRIE scores for the patients in the dog allergen-treated (A), placebo-treated (P). or untreated (U) patient groups before (82) and after (83) 1 year of treatment. Identification numbers for the various antigens are listed at the bottom.






FIG. 5. The average IgE CRIE scores toward the individual cat and dog antigens per patient responding to the antigen in question. The score system is explained in the text.

c Cat


Ag *

FIG. 6. Mean CRIE scores for the patients in the cat allergen-treated fA). placebo-treated (P), or untreated (lJ) patient groups before (82) and after (83) 1 year of treatment. Identification numbers for the various antigens are listed at the bottom.

tients generally demonstrated greater reactions to these proteins than to the serum components. The patients’ immunologic changes as a result of the treatment were also very significant with respect to antibodies with specificities directed against the dander component\. The preliminary CRIE analyses were also used for

allocation of the patients in the allergen- and placebotreatment groups. This increased the similarities between the two study groups, especially for the numerically superior cat-allergic groups, but only to a minor degree for the smaller dog-allergic groups. The CRIE analyses were performed for all 40 patients by use of both cat- and dog-allergen extracts. This revealed that the dog allergen-treated groups generally responded more strongly to cat antigens than the cat allergen-treated groups responded to the dog antigens, which might explain the clinically better effect obtained for the cat allergen-treated than for the dog allergen-treated patients. No new allergies arising as a result of the specific hyposensitization were detected by means of the CRIE method. This is consistent with the results of a recent study of pollen hyposensitization treatment,” in which no new allergies had appeared after 3 years. Further investigation of the long-term effect of hyposensitization with cat- and dog-dander extracts is in progress. We thank Kirsten Bathe Billesbelle for the skillful technical performance of the immunochemical analyses and Henrik lpsen for the statistical analyses.


77 3


REFERENCES 1. Lowenstein H: Allergene von katze, hund, rind und pferd. Allergologie 4:265, 1981 2. Lowenstein H, Lind P, Weeke B: Identification and clinical significance of allergenic molecules of cat origin. A part of the DAS 76 study. Allergy 40:430, 1985 3. Blands J, Lowenstein H, Weeke B: Characterization of extract of dog hair and dandruff from six different dog breeds by quantitative immunoelectrophoresis. Identification of allergens by crossed radioimmunoelectrophoresis (CRIE). Acta Allergol 321147, 1977 4. Sondergaard I, Djurup R, Weeke B: A modified crossed radioimmunoelectrophoretic method for determination of allergen-specific IgG antibodies (IgG-CRIE). Elimination of nonspecific binding by use of F(ab’)2-fragments of rabbit antibodies and I-labeled protein A. Allergy 39:135, 1984 5. Axelsen NH: Intermediate gel in crossed and in fused rocket immunoelectrophoresis. Stand J Immunol S(suppl 1):71, 1973 6. Harboe N, Ingild A: Immunization isolation of immunoglobulins, estimation of antibody titer. Stand J Immunol 2(suppl)l:71, 1973

with cat and dog dander

7. Lowenstein H: Quantitative immunoelectrophoretic methods as a tool for the analysis and isolation of allergens. Prog Allergy 25:1, 1978 8. Bsterballe 0, Ipsen H, Weeke B, Lowenstein H: Specific IgE response toward allergenic molecules during perennial hyposensitization: a three-year prospective double-blind study. J ALLERGYCLIN IMMLJNOL71:40, 1983 9. Schwartz B, Gravesen S, Petersen BN, Weeke ER: Laboratory investigations for allergens in the home. Ugeskr Laeger 141:882, 1979 10. Lowenstein H, Gravesen S, Larsen L, Lind P, Schwartz B: Indoor allergens (house dust mites, pets, etc.). J Allergy Clin Immunol (in press) Il. Lowenstein H, Krasilnikoff PA, Bjerrum OJ, Gudman-Hoyer E: Occurrence of specific precipitins against bovine whey pro teins in serum from children with gastrointestinal disorders. Int Arch Allergy 55:514, 1977 12. Ohman JL, Jr, Marsh DG, Goldman M: Antibody responses following immunotherapy with cat pelt extract. J ALLERGY CLIN IMMUNOL 69:320, 1982

The importance of a pause between the inhalation of two puffs of terbutaline from a pressurized aerosol with a tube spacer Ssren Pedersen,

M.D. Aalborg, Denmark

In a double-blind crossover study, 16 children with asthma were treated with two puffs of terbutaline (0.25 mg per puff)lplacebo from a pressurized aerosol with a tube spacer. The puffs of terbutaline were taken either immediately after each other (TT) or with a pause of 3 minutes (3TT) or 10 minutes (IOTT) between the two puffs. All active treatments compared with placebo resulted in a significant bronchodilation, both under normal day-to-day conditions and during acute attacks of bronchoconstriction (p < 0.01). Under basic conditions there was no statistically significant difference in increase in FEV, after Ti”, 3TT, and IOTT treatments (22%, 24%, and 25%, respectively). During attacks of acute wheeze, however, a pause between the two puffs of terbutaline improved bronchodilation [email protected] from 49% (TI’), to 48% (3TT), and 78% (1OTT) (p < 0.01). There was no statistically signt$cant dt#erence between 3TT and 1OTT treatments. It is concluded that pauses between the doses of inhaled terbutaline is likely to improve bronchodilation during episodes of wheeze or poor control of symptoms, whereas there appears to be no need to recommend pauses between the puffs in the routine dayCLINIMMJNOL77:505-9, to-day management of children with moderate asthma. (J ALLERGY 1986.)

When patients with asthmaaretreatedwith an aerosol delivering a P,-agonist, it is often recommended that a doseof two puffs be administeredwith a pause From the Allergy Unit, Department of Pediatrics, Aalborg Hospital, Aalborg, Denmark. Received for publication April 29, 1985. Accepted for publication Sept. 3, 1985. Reprint requests: S&en Pedersen, M.D., Department of Pediatrics, Aalborg Hospital, DK-9100, Aalborg, Denmark.

of a few minutes duration between the puffs instead of taking the two puffs rapidly after each other. Dividing the doseis believedto improve bronchodilation and to increasethe possibility of a more optimal distribution in the lungs of the inhaled drug. On the same assumption a pause between the inhalation of a p2agonist and other therapeutic agentsis recommended. Since many patients find pauses between the inhalations most inconvenient, it is important that the effect andnecessityof sucha pauseis studiedin detail. 505