farmers

above definition are in this paper grouped together to a patient group called AST/RHIN.

Allergens. House dust mite, cat, dog, horse and pollen allergen preparations were biologically standardised glycerinated Soluprick extracts (Allergologisk Laboratorium, Copenhagen, Denmark) with an activity of 5 HEP. Allergens of moulds, cow dander, soy bean, wheat and barley flour used were 1:20 w/v extracts from the same company. Skin test extracts with storage mites were not available.

The chironomid larvae extract used for SPT and NPT was produced at the Allergy Lab., Sahlgrenska hospital, Gothenburg, Sweden. It was produced in the following way: Deep frozen suspensions of larvae from red feather mosquitoes (Chironomus species), were purchased as food for aquarium fishes in a zoo shop. After thawing the product was disintegrated by freeze pressing. As the chironomid fish food contained quite a lot of fluid, no buffer solution was added. The particulate residues were removed by centrifugation and filtration using 0.65 um and finally 0,22 um Millipore filters. The solution was diluted 10 times with 50% glycerol and 0,5% phenol in Jensen’s buffer (1.45 g KH2PO4, 7.6 g Na2HPO4, 4,8 g NaCl in 1 litre of aqua destilata) before being used as a prick test reagent. Extracts used for the provocation test were diluted 10 times with Jensen’s buffer without glycerol. The chironomid extract produced from fish food is in this paper called CHIR.

The CHIR extract was controlled for non-specific histamine liberating properties by testing with SPT in 20 normal individuals without known atopic diseases. No positive reactions were found in these individuals.

Skin tests. SPTs were performed on the volar side of the forearm. The wheal reactions were read after 15 minutes and recorded in accordance with the recommendations of the Standardisation Committee of the Northern Society of Allergology (2). Thus a wheal reaction of the same area as that of a histamine reference (histamine HC1 1 g/l) was recorded as three plus (3+). A reaction of half the size of the histamine reference was recorded as 2+. The mean (+/- 1 stand. dev.) diameters of the wheals elicited by the histamine reference was 3.9 +/- 0.9 mm.

SPT results >/2+ were regarded as positive. Patients with positive skin test with one or more of the allergens used were regarded as atopics.

RAST. IgE antibodies were determined by RAST using Phadebas RAST (Pharmacia Diagnostics, Uppsala) according to the recommendations by the manufacturer. The following allergens were used in the RAST: CHIR (coupled to bromocyanide activated filter paper discs at Allergy lab., Sahlgrenska hospital, Gothenburg. For incubation of 100 discs 6 ml of the CHIR extract, earlier described, was used after dilution with 6 ml of 0.1 mol/L NaHC03.), Nimitti midge (NM, Cladotanytarsus lewisi, Pharmacia disc i72), Aedes communis (AC, Pharmacia disc i71) and one storage mite (Lepidoglyphus destructor, LD, Pharmacia disc d71). Furthermore, RAST was performed by Dr X Baur, Munich (4) with his chironomid haemoglobin (Chb) and whole body Chironomus thummi (CT) allergen preparations on 34 sera selected because of a positive SPT with CHIR.

The RAST results were expressed as PRU/ml and Phadebas RAST classes 0-4, using the Phadebas RAST Reference System (Pharmacia Diagnostics, Uppsala, Sweden).

Statistics. The statistical significances of correlations were tested with Chi square test and with Pitman’s non-parametric permutation test (5). In order to avoid the problems with mass significances only p<0.01 was regarded as significant, since several hundreds of tests were performed.

Since there were possibly reciprocal influences between some factors related to AST/RHIN (farm area, age, use of chemicals) a non-parametric partial correlations test, which could eliminate such influences, was used for analyses of correlations between factors (19). For the selected variables a logistic model was used and the linear combination of the variables used in

Table 2. Test results with Dermatophagoides pteronyssinus (DP), Lepidoglyphus destructor (LD), CHIR, Nimitti midge (NM), Chironomus thummi (CT) and Chironomid haemoglobin (CHb) in 9 patients having positive RAST with CHIR.

A=asthma, R=rhinitis, AR=asthma and rhinitis, nd=not determined

Patient		Diagnosis	Skin test results (no of plusses			RAST results (PRU/ml)						NPT with CHIR
Age	Sex		CHIR	DP	CHIR		NM	CT	Chb	AC	LD
37	M	AR	3	0	1.8		0.5	0.47	0	2.1	6.6		nd
28	M	AR	2	2	0.5		0.5	0.15	0	0	0		nd
45	M	A	3	0	0.4		0	0	0	0	0		0
55	M	AR	3	0	0.6		0	0.63	0	0	2.4		1
32	F	AR	2	0	0.5		0	0.2	0	0	0.35		0
32	M	A	3	2	12.3		1.6	3.35	0	1.3	15.0		nd
63	M	AR	3	0	1.5		0	0.03	0	0	1.6		0
36	M	R	4	3	0.6		0	nd	0	0	0.9		nd
62	M	A	3	0	1.5		0	nd	0	0	1.6		0

the logistic model, was determined. The probability of AST/RHIN as a function of the linear combination was determined using isotone regression.

Results

Prevalence of asthma and rhinitis. Of the 212 farmers who were examined at the clinic because their answers indicated allergic airways or skin diseased, 136 farmers received a diagnosis of AST/RHIN (29 asthma, 59 rhinitis and 48 asthma + rhinitis). Most of them had a non-atopic asthma or rhinitis (see test results!). Furthermore, of the farmers, who did not come to an investigation, 14 were regarded as having AST/RHIN (5 asthma, 9 rhinitis), because of their answers to the questionnaires. The distribution of asthma, seasonal rhinitis and perennial rhinitis is shown in Figure 1.

Definite prevalence figures cannot be calculated from our study, firstly because only 76% of the farmers answered the questionnaires and secondly because some diagnoses are doubtful, based only on the patients’ history. If farmers who did not answer the questionnaires are AST/RHIN is 150/592 = 25,3%.(asthma 5,7%, rhinitis 11,5%, asthma and rhinitis 8,1%).

No grading of disease severity was done during the first phase of the study. For this reason, a second questionnaire was later sent to those farmers given a diagnosis of asthma. It was found that only 16 farmers needed daily medication for their asthma and only 28, i.e. 3,6% of all farmers, had used drugs for asthma during the last month. Sixteen farmers, i.e. 2% of all the farmers, regarded their asthma being of such a severity that it influenced their working capacity.

Of the wives and farm-workers examined at the Clinic, 75 received a diagnosis of AST/RHIN. A significant difference regarding the prevalence of asthma was noted between farmers having large farms (area of cultivation > 50 hectares) ant those with small farms. (> 25 hectares). In the former group 25% of the farmers had asthma whereas only 12% of those in the later group had asthma (p< 0.01).

Use of chemical means of control. The farmers with AST/RHIN used chemical means of control more often than healthy farmers (p<0.001) (Figure 2). The farmers with AST/RHIN furthermore had used chemicals for a longer time than other farmers. Also the number of days per year handling chemicals was significantly correlated to AST/RHIN (p<0.01) The multivariate analyses (see Statistics) showed that the correlation between AST/RHIN and handling of chemicals could not be explained by other factors such as age or area of cultivation.

There was no significant correlation between number of years with chemicals and the results of skin prick tests or RAST.

Environmental factors eliciting symptoms. Thirty eight per cent of patients with AST/RHIN reported daily symptoms. Various environmental factors with relation to the working at the farm were reported to elicit symptoms by more than 80 per cent of the patients with AST/RHIN. Grain handling was the most often named work related factor (Figure 3).

Patients reporting symptoms with relation to work most often got their symptoms during the work. In a few patients, however, symptoms appeared only several hours after work.

SPT results. Seventy-four patients (55 farmers and 19 farm workers) had one or more positive SPT, i.e. they could be regarded as atopics. This indicates that 9% (55/592) of the farmers in Laholm are atopic and that (35% 74/210) of the rural patients with AST/RHIN are atopic.

The frequency of positive SPT results with the different allergens in subjects with AST/RHIN is shown in Figure 4. CHIR gave the highest number of positive test results (33 patients, i.e. 16% of the total AST/RHIN group); 45% of atopic subjects

Table 3. Comparison between SPT results in patients with asthma and/or rhinitis tested at the Allergy Clinic 1977-1989 (according to ref. 7) and in rural patients with asthma and/or rhinitis (according to the present study).

Allergen	Prevalence (%) of positive test results.
	Atopic patients with AST/RHIN		All patients with AST/RHIN
	Clinic patients n=4224	Rural patients n=74	Clinic patients n=7307	Rural patients n=210
Chironomids 1)	12	45	6	16
Mugwort	16	25	9	9
D. pteronyssinus	12	25	7	9
Cat	23	11	13	4
Dog	13	11	7	4
Timothy	22	11	13	4
Birch	24	11	14	4
Cladosporium	1	3	<1	1
Horse	9	1	5	<1

1) SPT with chironomids performed only 1983-1989

with AST/RHIN), while animal dander and pollen allergens showed positive test results in rather few cases. Also some (5%) of the subjects who were investigated because of other symptoms but without asthma or rhinitis had positive SPT with CHIR.

Of the 33 individuals with a >/2+ SPT result against CHIR, 8 had a 2+, 22 had a 3+ and 2 patients had a 4+ reaction. Sixteen patients had an isolated positive reaction with CHIR while 17 had positive SPT to common inhalant allergens as well. The number of patients with positive SPT to one or more of the common allergens was 56.

RAST results. The results of RAST to CHIR, NM, AC and LD are shown in Table 1. Positive RAST results were found more often with LD than with the insects (CHIR + NM + AC). RAST to CHb was negative in all sera tested. In 71 patients without AST/RHIN no positive RAST results were obtained with any of the allergens used.

Some details of the 9 patients with positive RAST to CHIR are given in Table 2. All but two of them had positive SPT against one or more common inhalant allergens besides CHIR.

Of the 20 patients with positive RAST to LD, 8 had positive SPT against CHIR as well as common allergens, 2 positive SPT to only CHIR, 7 positive SPT to common allergens but negative to CHIR and 3 negative SPT to all the allergens.

Relationship between test results and symptom eliciting factors. Of patients with work-related asthmatic symptoms 30% were atopic and of those with work-related rhinitis 36% were atopic. Twelve per cent of patients with work-related asthma or rhinitis had positive RAST with the storage mite LD. No significant correlations were found between the patients’ reports of factors at work inducing symptoms (see Figure 1) and SPT results with CHIR or RAST results with the storage mite.

Relationship between SPT and RAST results and between test results with different allergens.

Of the 34 patients with positive SPT with CHIR only 9 had positive CHIR RAST and 3 positive NM RAST Simultaneous sensitisation against CHIR, DP and LD was common (Table 2). SPT with CHIR was positive in 10 of 19 patients who were RAST positive with LD, and RAST with LD was positive in 10 of the 34 patients who had positive SPT with CHIR. RAST with LD was positive in 11 of 18 patients with positive SPT with DP. Significant correlations were found between RAST results with CHIR, AC and LD (p<0.001). A positive correlations also existed between SPT results with DP and RAST results with LD (p<0.001). (SPT was not performed with LD).

Discussion

The study has shown that asthmatic symptoms and rhinitis are frequently occurring in farmers. The estimated prevalence figure for asthma and/or rhinitis, 25%, is roughly the same as that found for the general population in our region (9). Higher figures were reported by van Hage-Hamsten et al (23) who found symptoms from airways or eyes in 40% of farmers on the island of Gotland. Iversen et al (15) found asthma in 7,7%, rhinitis in 9,4% and chronic bronchitis in 23,6% of Danish farmers.

A relationship was found to exist between handling of chemical means of control and the presence of AST/RHIN. As far as we know such a correlation has not earlier been shown. One could speculate that these chemicals, like diesel exhaust and other air pollutants, can induce airways diseases in susceptible individuals. In the above mentioned Danish study pig farming was found to be a risk factor for asthma (15).

In our study SPT with 16 different allergens was performed on 210 rural patients reporting symptoms of asthma and/or rhinitis (AST/RHIN). Seventy-four (35%) of this patient group showed positive SPT with one or more allergens, of which CHIR gave the highest number of positive reactions. Such allergens are not usually included in the concept ”common inhalant allergens”. Only 56 (27%) of the patients with AST/RHIN had positive SPT with the common allergens, such as pollens, mites, animals or moulds and thus fit into the usual definition of atopics. Of common allergens D. pteronyssinus (DP), mugwort and cow showed most positive reactions. It should be noted that test extracts with storage mites were not available for our study.

The prevalence of positive SPT (27% - 35%) in our patients with respiratory symptoms, i.e. the prevalence of atopics, does not significantly differ from the figures of other European studies (12,24). The observed allergy pattern with relatively high figures for sensitisation against mites and low figures for pollen is also in agreement with other studies in farmers (6, 12, 16, 24).

Our patients with AST/RHIN showed a much lower frequency of positive SPT to such common inhalant allergens as birch pollen, grass pollen and animal danders than urban patients with asthma or rhinitis, tested at our outpatient Allergy Clinic (7). (Table 3). Also in other parts of the world a similar difference in sensitisation rates has been reported between rural and urban populations. (11, 20, 22).

The reason for the finding that urban patients are more often sensitised against common inhalant allergens than rural patients is not clear. One could speculate that large amounts of allergens such as those in rural environmental might induce tolerance whereas the lower doses typical of the urban milieu could sensitise people. It is also possible that urban people are exposed to larger quantities of unspecific air pollutants, acting as adjuvant factors for sensitisation, which could explain a higher rate of sensitisation in urban areas.

Some allergens seem to be more important in farmers than in urban patients. A high sensitisation rate has bee reported with storage mites, which are found in the byres and hay and grain stores (6, 21, 23). Nine per cent of our farmers with AST/RHIN, corresponding to 26% of our atopic farmers with AST/RHIN, had a positive RAST to a storage mite, LD. The figures are in agreement with those of other studies from Sweden (23), Finland (21), Denmark (16), Scotland (6) and Iceland (12).

Allergy against the mosquito Aedes aegypti has been reported in the Baikal-Amir Main Line (1). Inhalant allergy against mosquitoes belonging to the family Chironomidae has earlier been reported as a common cause of bronchial asthma in Sudan (10), Japan (13) and in USA (17), and recently two European studies reported on chironomid allergy (8, 20). We found a high percentage of positive SPT to CHIR in our farmers, higher than in our urban patients (7). The reason is not obvious.

The positive SPT with CHIR could mean a true and clinically relevant allergy against chironomids. It could also mean, however, a sensitisation without clinical importance. The positive SPT could even be biologically falsely positive, i.e. irrelevant non-allergenic contents of the extracts could induce wheal reactions in a proportion of the subjects. It should be noted that our CHIR was not a standardised allergen extract and the allergen concentration might be performed employing improved extracts, in order to evaluate the real relevance of chironomid allergy.

The CHIR allergen used by us probably consisted of larvae of Chironomus thummi thummi (CT) which by Baur et al (4) was reported to elicit hypersensitivity among workers engaged in producing food for pet fishes. They found haemoglobin (CHb) contained in the larva to be an important allergen. None of our patients had a positive RAST with CHb. If they have a clinically relevant allergy to some chironomid allergens, it is most likely not directed against the haemoglobin allergen but against other allergens.

In order to prove a connection between farm work and sensitisation to chironomids, the relevant allergens should be possible to find in the farming environment. In Japan chironomid allergen has been found in house dust (12). We have not performed such studies among the patients presented in this study.

We found a high degree of correlation among test results with CHIR, DP and LD. It would depend on simultaneous sensitisation to several allergens in the patients’ environment, or depend on true immunological cross reactivity. It has been suggested that there exists a pan-allergy to insects in subjects who have been sensitised to one or a few insects and allergenic similarities may also exist with some non-insect members of the phylum Arthropods (3).

Acknowledgements

We thank Dr X Baur for kindly performing RAST analyses with allergens not available to us. This study was supported by the SLO Foundation, Asthma & Allergy Association, Gustaf V:s Foundation, the Council of Environmental Medicine in Halland and the Swedish Medical research Council, grant no 16x - 105.

References

1. Ado AD, Chervinskaya TA, Bondareva GP, Shustova VI. Allergy to mosquito. Immunologia 1988: No 6: 47-51.

2. Aas K, Belin L. Standardization of diagnostic work in allergy. Acta Allergol 1972; 27: 439-468.

3. Baldo B A, Panzani R C. Detection of IgE antibodies to a wide range of insect species in subjects with suspected inhalant allergies to insects. Int Archs Allergy appl Immun 1988; 85: 278-287.

4. Baur X, Dewair M, Fruhmann M, Aschauer H, Pfletschiner J, Baunitzer G. Hypersensitivity to chironomids (non-biting midges): localization of the antigenic determinants within certain polypeptide sequences of hemoglobins (erythrocruorins) of Chironomus thummi thummi (Diptera). J Allergy Clin Immunol 1982;69:66-76.

5. Cuthbert OD, Jeffrey IG, McNeill HB, Wood J, Topping MD. Barn allergy among Scottish farmers. Clin Allergy 1984;14:197-206.

6. Eriksson NE. Does the sensitization pattern in atopics change? (In Swedish) (Abstr). Hygiea, 98;No 5:99, 1989, Stockholm, Sweden.

7. Eriksson NE. Prevalence of allergy and obstructive lung disease among adults in Halland. (In Swedish) (Abstr). Hygiea, 98;No 5:99, 1989, Stockholm, Sweden.

8. Eriksson NE, Jonsson P, Rydén B. Hypersensitivity to larvae of chironomids (non biting midges). Cross sensitization with crustaceans. Allergy 1989;44:303-313.

9. Gad El Rab MO, Kay AB. Widespread immunoglobin E-mediated hypersensitivity in the Sudan to the "green nimitti" midge, Cladotanytarsus lewisi (Diptera: Chironomidae). J. All. Clin. Immunol. 1988;66: 190-197.

10. Gergen PJ, Turkelhaub MD, Kover MG. The prevalence of allergic skin test reactivity to eight common aeroallergens in the US population: Results from the second national health and nutrition examination survey. J Allergy Clin Immunol 1987;80:669-79.

11. Gislason D, Gravesen S, Asmundsson T, Magnusson V. Immediate type allergy in two farming communities in Iceland. 1. Prevalence and main allergens. Laeknabladid 1988;74:303-8 (Summary in English).

12. Igarashi T, Murakami G, Adachi YG, Matsuno M, Saeki Y, Suzuki Y, Okada T, Kawai K, Kumagai A, Sasa M. Common occurence in Toyama of bronchial asthma induced by chironomid midges. Japan J Exp Med 1987;57:1-9.

13. Ishii A, Matsuoka H, Uchida JY et al. Chironomid midge and allergy around lake Kojima, Okayma. In International symposium on mite and midge allergy (president: T Miyamoto), Tokyo, 1988.

14. Iversen M, Dahl R, Korsgaard J, Hallas T, Jensen EJ. Respiratory symptoms in Danish farmers: an epidemiological study of risk factors. Thorax 1988;43:872-877.

15. Iversen M, Korsgaard J, Hallas T, Dahl R. Mite allergy and exposure to storage mites and house dust mites in farmers. Clin and Exp Allergy 1990;20:211-219.

16. Kagen SL, Yunginger JW, Johnson R. Lake fly allergy: incidence of chironomid sensitivity in an atopic population (Abstract). J Allergy Clin Immunol 1984;73(N:o 1 part 2).

17. Kino T, Chihara J, Fukuda K, Sasaki Y, Shogaki Y, Oshima S. Allergy to insects in Japan. 3. High frequency of IgE antibody responses to insects (moth, butterfly, caddis fly and chironomid) in patients with bronchial asthma and immunochemical quantitation of the insect-related airborne particles smaller than 10 um in diameter. J. All. Clin. Immunol. 1987;79:857-866.

18. Mantel N. Chi-square tests with one degree of freedom. Extensions of the Mantel-Haenszel procedure. J Am Stat Ass 1963;58:690.

19. McHugh SM, Credland PF, Tee RD, Cranston PS. Evidence of allergic hypersensitivity to chironomid midges in an English village community. Clin Allergy 1988;18:275-85.

20. Terho EO, Husman K, Vohlonen I, Rautalahti M, Tukianinen H. Allergy to storage mites or cow dander as a cause of rhinitis among finnish daily farmers. Allergy 1985;40:23-6.

21. Turner KJ, Dowse MB, Stewart GA, Alpers MP. Studies on bronchial hyperreactivity, allergic responsiveness and asthma in rural and urban children of the highlands of Papua New Guinea. J Allergy Clin Immunol 1986;77:558-66.

22. Van Hage-Hamsten M, Johansson SGO, Zetterström O. Predominance of mite allergy over allergy to pollens and animal danders in a farming population. Clin Allergy 1987;17:417-423

23. Van Hage-Hamsten M, Johansson SGO, Höglund S, Tull P, Zetterström O. Storage mite allergy is common in a farming population. Clin Allergy 1985;15:555-65.

Correspondence: Nils E Eriksson, M.D., Lung and Allergy Clinic, Department of Medicine, County hospital, S-30185 Halmstad, Sweden.

Of the 592 farmers who answered the questionnaire, 150 farmers had asthma and/or rhinitis. The figure shows the distribution of the diagnoses.

The use and previous use of chemical means of control among 442 healthy farmers and 150 farmers with asthma and/or rhinitis. The figures indicate percentages.

Number of patients getting asthmatic symptom or rhinitis from various environmental factors at the farm.

Distribution of positive skin prick tests with 16 different allergens in 210 rural patients with asthma and/or rhinitis. The figures indicate percentages.

Allergen	Number of positive RAST results	Percentages of
		The whole group (n=210)	Patients with positivie SOPT with any allergen (n=74)	Patients with postivie SPT with common inhalant allergens (n=58)
CHIR	9	4	12	16
Nimitti midge	3	1	3	5
Aedes communis	2	1	3	4
L. destructor	20	9	26	36