- Email: [email protected]
Long term effect and allergic sensitization in newly employed workers in laboratory animal facilities
Respiratory Medicine 109 (2015) 1164e1173
Contents lists available at ScienceDirect
Respiratory Medicine journal homepage: www.elsevier.com/locate/rmed
Long term effect and allergic sensitization in newly employed workers in laboratory animal facilities Lena Palmberg, Britt-Marie Sundblad*, Anders Lindberg, Maciej Kupczyk 1, Karin Sahlander, Kjell Larsson Lung and Airway Research, Institute of Environmental Medicine, Karolinska Institutet, Stockholm, Sweden
a r t i c l e i n f o
a b s t r a c t
Article history: Received 25 March 2015 Received in revised form 4 June 2015 Accepted 12 June 2015 Available online 15 June 2015
Objective: The aim of this study was to identify targets predicting allergic sensitization to laboratory animals using shift in skin prick test to laboratory animals as primary outcome variable. Methods: In a prospective longitudinal study, personnel who were employed to work with laboratory animals at a medical university were investigated before and 6, 12 and 24 month after the start of employment. Lung function, bronchial challenges, exhaled NO and nasal lavage were performed and blood samples were drawn at all visits. Results: Seventy subjects attended all four visits and 13 (19%) became sensitized to laboratory animals during the two years of follow up. Lung function (VC and FEV1) deteriorated and blood levels of eosinophils and IL-2 producing lymphocytes increased after 24 months. An increased risk of developing laboratory animal allergy was significantly associated with female sex, atopy, symptoms associated with exposure to laboratory animals, low proportion of blood CD4þ cells, specific IgE to rat and mouse and high total IgE when starting to work with laboratory animals. Conclusions: A sensitization rate of 19% in 2 years, were demonstrated in laboratory animal workers. Atopy, increased total and specific IgE levels (rat and mouse) were the strongest predictors for laboratory animal sensitization. The progressive lung function impairment over time, observed in the whole study population may indicate that exposure in animal facilities induces harmful effects, irrespective to allergic sensitization. © 2015 Elsevier Ltd. All rights reserved.
Keywords: Laboratory animal allergy (LAA) Exposure Sensitization Risk factors
1. Introduction Occupational asthma represents 5e15 % of all adult asthma and is today the most common work related airways disease in the western world [1,2]. The prevalence of allergy has increased during the last decades and allergy against laboratory animals (LA) is a frequently observed work-related health problem among pharmaceutical research workers [3]. In cross-sectional studies up to nearly two thirds of the personnel who are exposed to LA report airway symptoms while working with animals [4,5] and approximately one of four is sensitized against LA. Of those who experience symptoms approximately 40% have skin symptoms including
* Corresponding author. Lung and Airway Research, Institute of Environmental Medicine, Karolinska Institutet, PO Box 287, S-171 77 Stockholm, Sweden. E-mail address: [email protected] (B.-M. Sundblad). 1 Present address: Department of Internal Medicine, Asthma and Allergy, Medical University of Lodz, Poland. http://dx.doi.org/10.1016/j.rmed.2015.06.007 0954-6111/© 2015 Elsevier Ltd. All rights reserved.
urticaria [6], and up to 80% have allergic rhino-conjunctivitis with nasal congestion, rhinorrhea, sneezing, and itchy, watery eyes [7]. Cough, shortness of breath and, wheeze, i.e., symptoms indicating asthma, have been reported in up to 22% of individuals exposed to LA. The onset of symptoms after commencing work with LA varies. In workers without previous rat exposure and a range of less than 30 dayse1369 days from the time of employment to the onset of symptoms was reported. The mean duration of employment before symptom onset was approximately one year for chest and skin symptoms and slightly more than 200 days for nose and eye symptoms [5]. Working in laboratory animal facilities not only involves exposure to laboratory animal allergens but also to non-allergenic compounds such as microbial products like endotoxins and peptidoglycans. Endotoxin (lipopolysaccharide, LPS) is recognised as a pro-inflammatory stimulus and exposure to endotoxin correlates with lung function impairment in farmers [8]. Effects of exposure to
L. Palmberg et al. / Respiratory Medicine 109 (2015) 1164e1173
non-allergenic agents in laboratory animal facilities have, however, not been much studied. It is most likely that exposure in laboratory animal facilities constitutes a pro-inflammatory stimulus which may influence lung function and bronchial responsiveness as have been demonstrated in other environments such as exposure in pig barns [9]. In farmers and agriculture industry workers occupational endotoxin exposure seems to reduce the risk of atopic sensitization whereas there is an association between exposure and bronchial hyper-responsiveness [10]. Individuals who applied for a job involving work with LA, at Karolinska Institutet, were longitudinally, prospectively, followed during the two first years of employment. The study was based on the assumption that 20e25% of the exposed employees would develop laboratory animal allergy (LAA) and the primary endpoint was a shift of skin prick test to LA. The main aims were to identify phenotypes and biological markers predicting development of allergy and symptoms and to explore possible effects of exposure to non-allergenic stimuli in laboratory animal facilities.
2. Material and methods 2.1. Study design Persons who applied for a job at Karolinska Institutet were eligible during a period of 3 years (2003e2006) and were included if they had the intention to work with LA and to remain as an employee at the work place for at least 2 years. Subjects were excluded if they had been working with LA during the last 6 months prior to employment and more than 2 years in total. Those who had a history of exposure to LA and positive skin prick test for LA (rat, mouse, rabbit, guinea pig and hamster) were excluded. Five subjects who at baseline had positive SPT to LA without any occupational exposure previous or during the study to these animals were included in the study to find out possible switch in SPT to other LA which they were exposed to during the time of employment. At the start of the study all participants were informed about the health risk associated with laboratory animal work and the importance of taking protective measures. The participants attended the laboratory twice, in the morning, at 4 occasions, prior to employment and after 6, 12 and 24 months of employment. At all visits a questionnaire, skin prick test, lung function test, exhaled nitric oxide (NO) measurement and measurement of bronchial responsiveness to a direct (methacholine) and an indirect (dry air hyperpnoea) stimulus were performed, the bronchial challenges on two separate days. Blood samples and nasal lavage were also collected. Asthma-medication (glucocorticosteroids and b2-agonists) were discontinued 24 h and anti-histamines 5 days prior to each visit. The study protocol was approved by the Karolinska Institute Ethics Committee (Protocol number 02-287, Stockholm, Sweden), and all individuals gave their informed consent to participate.
2.2. Exposure measurements Exposure in 11 animal facilities were assessed by personal air sampling described elsewere [11]. Allergen measurements of rat n1 and mouse m1 were made using unamplified sandwich ELISA according to previously described methods [11]. Endotoxin concentrations were analysed using a kinetic technique version of Limulus amebocyte lysate assay (Limulus Amebocyte lysate, Endosafe® Endochrome-K™ U.S. Lisence No. 1197, Coatech AB, Kungsbacka, Sweden), with E. coli 0111:B4 as standard. The detection limit was 0.1 EU/mL.
1165
2.3. Skin prick test (SPT) A simple skin prick test was performed using a panel of common allergens and allergens from commonly used laboratory animals (birch, timothy, mugwort, cat, dog, guinea pig, rabbit, hamster, Determatophagoides farinae, Cladosporium, Alternaria, (ALK-Abello A/S, Denmark) and rat, mouse, horse, latex, Dermatophagoides pteronyssinus, and Lepidoglyphus destructor, (Alergopharma, Germany)). Histamine dihydrochloride (10 mg/ml) served as positive and the diluent as negative control. A wheel of 3 mm in diameter was considered as positive reaction. 2.4. Questionnaire Two extensive questionnaires were used. The first questionnaire contained questions on heredity, smoking habits, usage of safety equipment, current and previous exposure to animals, airways disease, allergy, medication and eyes, nose, lower respiratory tract and skin symptoms. The second questionnaire contained detailed questions on current and previous work with laboratory animals and symptoms in connection with laboratory animal work. The questionnaires were sent to the participants who responded to the questions in advance and were then individually interviewed at each visit and exposure protection advices were given and followed up. 2.5. Lung function and bronchial responsiveness Forced expiratory volume in 1 s (FEV1) and vital capacity (VC) were measured using a wedge-spirometer (Vitalograph®, UK) according to the American Thoracic Society recommendations [12]. Local reference values were used [13,14]. Direct bronchial responsiveness was assessed by a methacholine challenge [15]. The subjects inhaled a nebulized methacholine solution of increasing concentrations, starting at 0.5 mg/mL followed by doubling doses until FEV1 decreased by 20% or the maximal concentration (32 mg/mL) was reached. On a separate day indirect bronchial responsiveness was assessed by hyperpnoea of dry air containing 5% carbon dioxide of ambient temperature breathed through a low-resistance, one-way valve in the sitting position (Aiolos Asthma Test®, Karlstad, Sweden). The target ventilation (35 FEV1 0.75 L) was maintained for 6 min and FVC and FEV1 were measured before and immediately, 1, 2, 5, 10, and 20 min after finishing the hyperpnoea. The highest, of two FEV1 measurements was registered. A fall in FEV1 of >10% of the pre-challenge value was regarded as a positive test [16]. 2.6. Measurement of exhaled nitric oxide Nitric oxide in exhaled air was analysed with chemiluminescence after reaction with ozone (NIOX®, Aerocrine, Sweden) according to current recommendations [17]. The subjects rinsed their mouth with water and then with sodium bicarbonate (10%) prior to the measurement. All the subjects were asked to refrain from eating green vegetables on the day of the visits. 2.7. Nasal lavage Nasal lavage was performed according to Bascom et al. [18] with minor modifications [19]. Briefly, the head was tilted back and then one nostril was flushed with 5 ml of 0.9 mg/mL saline. After 10 s the head was bent forward and the fluid was expelled into a plastic cup. The procedure was repeated in the other nostril and the lavage samples were pooled. After centrifugation, the supernatant was frozen (70 C) until further analyses.
1166
L. Palmberg et al. / Respiratory Medicine 109 (2015) 1164e1173
2.8. Flow cytometry The blood samples were collected in EDTA vacutainer tubes (BD Bioscience) and a flow cytometric analysis was performed within 2 h using FACS Calibur™ (BD Bioscience). Fifty mL of blood and 20 mL MultiTEST CD45PerCP/CD3FITC/CD4APC/CD8PE (BD Bioscience) were added to TruCOUNT™ tubes and then incubated in darkness for 15 min. The red blood cells were then lysed with 1 mL FACS Lysing Solution™ (BD Bioscience). An absolute count of lymphocytes, monocytes, neutrophils, eosinophils, basophils, CD3þ, CD4þ and CD3þCD8þ subsets of T cells were determined by analysis in Multiset (BD Bioscience) using the TruCOUNT™ beads as an internal reference population. Interleukin-2, IL-4, IL-13 and IFN-g were measured in lymphocytes with intracellular cytokine staining. Heparinized peripheral blood in RPMI 1640 containing 2 mM glutamine (1:1 dilution) was stimulated with phorbol 12- myristate 13-acetate (PMA, 25 ng/mL) and ionomycin (1 mM, SigmaeAldrich) in presence of GolgiPlug (Brefeldin A, BD Bioscience) for 4 h at 37 C. Stimulated blood was stained with anti-CD4 APC (BD Bioscience). The samples were then lysed and incubated with Cytofix/Cytoperm™ (BD Bioscience Pharmingen). The permeabilized cells were washed with Perm/ Wash and resuspended in CellFIX™ (BD Bioscience). Unstimulated blood was stained using the same procedure as negative controls. Analysis was performed using a FACSCalibur™ (BD Bioscience) and CELLQuest™. The results were presented as proportion of cytokineproducing T-helper (CD4þ) cells [20]. 2.9. Analyses of CXCL-8 and ST2 The CXCL-8 (IL-8) concentration was measured in nasal lavage fluid using an in-house ELISA method. Commercially available antibody pairs (R&D system, Europe, Abingdon, UK) were used as previously described [21]. The detection range was 40e3200 pg/ mL. For serum analyses of sST2 (soluble Suppression of Tumorgenicity 2) DuoSet ELISA kit (R&D system, Europe, Abingdon, UK) was used. The detection range was 15.6e2000 pg/mL. For duplicate samples, an intra-assay coefficient of <10% was accepted for CXCL-8 and sST2. 2.10. Total and specific IgE Total Immunoglobulin E (tIgE), Phadiatop® (specific IgE of a mix of the most common allergens in Sweden) and specific IgE to rat (e87) and mouse (e88) allergens were measured in serum using ImmunoCAP100 (Phadia, Sweden). A specific IgE level above 0.35 kUA/L was considered positive. 2.11. Statistical methods Data are presented as mean and range (age and height), mean and 95% confidence intervals (lung function, exposure), or as median (25th 75th percentiles). Within- and between group comparisons of lung function data were assessed by ANOVA with Students paired or unpaired t-test when appropriate. Within group-comparisons for other outcome variables were assessed by Friedmans repeated measurements followed by Wilcoxon signed rank test when appropriate or directly with Wilcoxon signed rank test for parameters only measured twice. Between groups comparisons were assessed by Mann Whitney U test. Symptoms and skin prick test were analysed by means of Chi-square test. A p-value 0.05 was considered significant.
To identify factors, associated with allergic sensitization to laboratory animals, those who shifted from negative to positive skin prick test to laboratory animals during the study were compared to those who remained negative in skin prick test throughout the study. Odds ratios (OR) and risk ratios (RR) were assessed using univariate analysis of the following parameters: atopy (defined as at least 1 positive skin prick test to any allergen at baseline visit), positive skin prick test to cat and dog, female sex, age, smoking at study entry, symptoms (symptoms related to laboratory animal contact) at study entry, lung function, bronchial responsiveness, biomarkers (IL-2/CD4þ, IL-4/CD4þ, IL-13/CD4þ, IFN-g/CD4þ, CD3, CD4þ and CD8þ, cells and CD4/CD8 ratio, neutrophils, monocytes, lymphocytes, basophils and eosinophils in peripheral blood, IL-8 levels in nasal lavage fluid, total IgE, Phad, e87 and e88 and exhaled NO). For numerical variables median, 25 percentile and 75 percentile as possible cut-off values were evaluated. Chi-square test (or Fisher's exact test wherever appropriate) was applied. A multivariate logistic regression model including all potential predictors with univariate p-value less than 0.1 was built and adjusted for atopy, sex, age and smoking status. SPSS version 17.0 software (SPSS Inc, Chicago, Ill) was used for uni- and multivariate analyses and otherwise using Stat View (version 5.01; SAS Institute).
3. Results Out of 249 eligible subjects 119 fulfilled the inclusion criteria and 130 were excluded because they had worked for more than 2 years and the last 6 month with LA or, had positive SPT to mouse/rat that they have been occupationally exposed to or had planned to remain at the work place for less than 2 years. Of the included subjects, 17 declined participation, 102 attended visit 1 and 70 subjects attended all four visits. The reasons for loss of follow up (n ¼ 32) were termination of employment, pregnancy and unwillingness to continue. For analyses subjects were divided into subgroups based on two approaches (Fig. 1). The sensitized group (n ¼ 13, 18.6%), those who shifted from negative to positive skin prick test to laboratory animals during the study. Subjects (n ¼ 2) who had a positive SPT to LA at baseline without being occupationally exposed to this animal during the study and who became sensitized to another animal during the study were included in this group. The non-sensitized group (n ¼ 57, 81.4%), those who did not shift in skin prick test during the study. Subjects (n ¼ 3) who had a positive SPT to LA at baseline without being occupationally exposed to this animal during the study and who did not become sensitized to any other animal were included in this group. The atopic group (n ¼ 26, 37.1%), positive skin prick test to any allergen at study entry. The non-atopic group (n ¼ 44, 62.9%), no positive skin prick test at study entry.
3.1. Exposure The mean time of work with laboratory animals was 33 (25e42) hours/months with no difference between the groups (p ¼ 0.55; Table 1). In total, 187 air samples for allergen and 133 for endotoxin measurement were collected. Average endotoxin level was <1.2 mg/m3 and allergen levels of rat n1 and mouse m1 were 5.0 (0.01e115) ng/m3 and 73.9 (0.01e1168) ng/m3, respectively, in the animal facilities.
L. Palmberg et al. / Respiratory Medicine 109 (2015) 1164e1173
1167
Fig. 1. Sensitized and non-sensitized participating subjects in the groups with and without atopy at study entry. Shift in skin prick test to laboratory animal indicates change from negative to positive reaction to laboratory animals.
Table 1 Lung function and bronchial responsiveness in sensitized/non-sensitized and atopic/non-atopic subgroups.
M/F Smokers, n (%) Age, year, mean (range) Height, cm, mean (range) VC, L, mean (95%CI) FEV1, L, mean (95%CI) FEV1/VC, mean (95%CI) PD20, mg, median (25the75th percentile) Eucapnic hyperventilation, %FEV1 decrease, mean (95%CI) Laboratory animal work, hours/months, mean (95%CI)
All n ¼ 70
Sensitized n ¼ 13
Non-sensitized n ¼ 57
Atopic n ¼ 26
Non-atopic n ¼ 44
21/49 10 (14.3) 28 (20e53) 171 (155e195) 4.45 (4.21e4.68) 3.78 (3.60e3.97) 0.85 (0.84e0.88) 1.17 (0.36e2.71) 3.29 (2.68e3.90) 33 (25e42)
2/11 2 (15.4) 28 (21e53) 168 (158e188) 4.15 (4.25e4.78) 3.51 (4.25e4.78) 0.85 (0.81e0.88) 0.54 (0.40e1.18) 3.24 (2.17e4.30) 41 (11e71)
19/38 8 (14.0) 27 (20e43) 171 (155e195) 4.52 (4.25e4.78) 3.84 (3.63e4.05) 0.85 (0.84e0.87) 1.30 (0.36e3.18) 3.30 (2.58e4.01) 32 (23e41)
9/17 4 (15.4) 29 (20e53) 172 (159e195) 4.48 (4.09e4.87) 3.74 (3.44e4.03) 0.84 (0.81e0.87) 0.54 (0.19e1.91) 2.78 (1.90e3.66) 30 (16e44)
12/32 6 (13.6) 27 (20e39) 169 (155e195) 4.43 (4.13e4.74) 3.80 (3.55e4.06) 0.86 (0.84e0.88) 1.30 (0.43e10.2) 3.59 (2.77e4.42) 36 (24e47)
3.2. Skin prick test, Phadiatop® and total IgE Twenty-six subjects had positive skin prick test to any allergen at study entry, 13 shifted from negative to positive skin prick test to LA during the study. Two of the subjects showed conversion of two LA. The SPT reactions were not persistent, but shifted both in size and over time.
Phadiatop® and total IgE decreased during the two years study in the whole group (p 0.0003), in the non-sensitized (p 0.003), non-atopic (p < 0.05) and atopic (p 0.03) groups. In the sensitized group Phadioatop® (p ¼ 0.22) and total IgE (p ¼ 0.60) did not change during the study period. After two years of exposure the decrease in Phadiatop® in the atopic group exceeded that of the non-atopic group (p < 0.0001) (Fig. 2).
1168
L. Palmberg et al. / Respiratory Medicine 109 (2015) 1164e1173
Fig. 2. Blood eosinophils and basophils before and after 6, 12 and 24 months and total IgE and Phadiatop® before and 24 months of employment in sensitized and non-sensitized (A), atopic and non-atopic subjects (B). Median, 25th and 75th percentiles. x indicates p < 0.05, xxp < 0.01 and xxxp < 0.001 within group. ** indicates p < 0.01 and ***p < 0.001 between groups.
L. Palmberg et al. / Respiratory Medicine 109 (2015) 1164e1173
3.3. Baseline factors associated with development of laboratory animal allergy In the univariate model high bronchial responsiveness to methacholine (median slope >17.17%/mg methacholine) was associated with an increased risk of skin prick test shift to laboratory animals (p ¼ 0.034). Possible protective factors included low bronchial responsiveness to methacholine indicated by PD20FEV1 >2.71 mg (75th percentile; p ¼ 0.028) and slope <6.66%/mg (25th percentile) methacholine (p ¼ 0.028) and Phadiatop® <0.05, (25% percentile; p ¼ 0.03), e87 < 0.003 (25th percentile; p ¼ 0.029), and e88 ¼ 0 (25th percentile; p ¼ 0.0034) at study entry. Major risk factors for a skin prick test shift to laboratory animals included positive skin prick tests to cat and dog, serological indicators of atopy and low proportion of circulating CD3þ cells (<1.28 cells/mL) at study entry (Table 2). When adjusted for possible confounding factors in multivariate logistic regression models, female sex, atopy (e87 > 0.01, e88 > 0.004, total IgE > 15.5 kUA/L), low proportion of circulating CD3 cells (<0.92 cells/mL) and symptoms remained significantly associated with an increased risk of developing laboratory animal allergy (Table 2).
1169
3.6. Exhaled NO Exhaled NO differed between atopic and non-atopic at all visits (Fig. 5). In the sensitized group exhaled NO level was higher after 6 (p < 0.01) and 12 (p < 0.05) months of employment than in the nonsensitized group but no difference was found at study entry and after 24 months of employment. 3.7. Blood
The prevalence of work-related symptoms did not change during the study within the four subgroups, (Fig. 3). At 12 months the sensitized group had more symptoms from the eyes (p ¼ 0.04) and nose (p ¼ 0.03) than had the non-sensitized group. Symptoms did not differ significantly between the atopic and non-atopic groups.
The number of eosinophils increased in the whole group and in the non-sensitized, and the non-atopic groups (p < 0.0001 for all) and basophils in blood increased over time in the atopic group (p ¼ 0.03), (Fig. 2). Intracellular expression of IL-2 increased in the whole group (p ¼ 0.0003), in the non-sensitized group (p ¼ 0.0006) and the nonatopic group (p ¼ 0.0018) during the 2 years of follow up (Fig. 6). Interferon- g did not change over time in any subgroup but was higher in the atopic than in the non-atopic group after 6 (p ¼ 0.03) and 24 months (p ¼ 0.04) of employment (Fig. 6). The intracellular expression of IL-4 and IL-13 was enhanced in atopic group compared with the non-atopic group at study entry (p ¼ 0.001 and p ¼ 0.008, respectively) and after 6 months (p ¼ 0.002 and p ¼ 0.007, respectively). Enhanced expression was also found for IL-4 at 24 months (p ¼ 0.01) and IL-13 at 12 months (p ¼ 0.02) of employment (Fig. 6). Soluble ST2, involved in pro-inflammatory responses and playing a role in allergic inflammation, did not change over time and did not differ between the groups.
3.5. Lung function and BHR
3.8. Nasal lavage
Lung function (VC and FEV1) deteriorated over time in the whole group (n ¼ 70) and in the non-sensitized group (Fig. 4). This result was calculated using local Swedish reference values (13, 14). When using GLI2012 the deterioration lung function was even more enhanced (data not shown). Lung function change did not differ over time between the sensitized and non-sensitized and between the atopic and non-atopic groups. Bronchial responsiveness to methacholine (PD20) differed between the atopic, 0.54 mg (0.19e1.91), median (25the75th perc.) and non-atopic group 1.30 mg (0.43e3.18) at study entry (p ¼ 0.05) and increased significantly during the study in the atopic group (p ¼ 0.02; PD20 after 24 month was 0.38 mg (0.12e1.00)). Bronchial response to hyperpnoea of dry air (% FEV1 decrease) did not change over time in the whole group or in any of the subgroups.
The levels of the inflammatory mediator CXCL-8 in nasal lavage were not altered over time and did not differ between the subgroups.
3.4. Symptoms
4. Discussion One out of five persons who commenced working in laboratory animal facilities became sensitized against LA during the first two years of employment. The incidence of allergic sensitisation is similar to what previously has been reported following exposure to LA [6,22e24]. Of those who became sensitized against LA, more than three out of four were atopic at study entry, clearly showing atopy as the major risk factor. Conversion in SPT to one or more LA was the primary outcome in this study. Eight of the 13 sensitized subjects were sensitized at
Table 2 Adjusted odds ratios in multivariate logistic regression models for risk factors at study entry associated with the development of laboratory animal allergy (in decreasing order). Factor (at study entry)
Adjusted OR
95% CI
P-value
e88 > 0.004, kUA/La Female sexb e87 > 0.01, kUA/La Total IgE > 15.5, kUA/La Atopy, any positive skin prick testc CD3 < 0.92, cells/mLa Symptoms in connection with laboratory animal work, yes/noa e87 < 0.003, kUA/La Phad < 0.05, kUA/La
19.31 9.86 9.05 8.12 7.45 5.40 5.26 0.135 0.076
2.04e182.98 1.08e89.8 1.89e43.3 1.36e48.32 1.71e32.42 1.08e27.09 1.00e27.41 0.03e0.635 0.006e1.045
0.010 0.042 0.006 0.021 0.007 0.040 0.049 0.011 0.054
e87 ¼ specific IgE to rat and e88 specific IgE to mouse. a Adjusted for atopy, sex, age and smoking. b Adjusted for atopy, age and smoking. c Adjusted for sex, age and smoking.
1170
L. Palmberg et al. / Respiratory Medicine 109 (2015) 1164e1173
Fig. 3. Work-related symptoms from airways, skin, eyes and nose before and after 6, 12 and 24 months of employment in the sensitized, non-sensitized, atopic and non-atopic groups. At 12 months the sensitized group had more symptoms from the eyes (p ¼ 0.04) and nose (p ¼ 0.03) than had the non-sensitized group.
the last follow-up at 24 month. Thus, it was impossible to show if the conversion was persistent in these subjects and in some other subjects the positive SPT to LA varied over time. Two explanations to a temporarily positive SPT at one or two follow-up occasions may be related to changed exposure-response and ongoing educational effort with increased prevention. In subjects who became sensitized, symptoms increased following exposure whereas symptoms did not change over time in non-sensitized, atopics or non-atopics indicating a causal relationship between allergic sensitization and the development of clinical symptoms. In the sensitized group symptoms, predominantly from the eyes and nose indicating allergic rhinoconjunctivitis, occurred within one year of employment whereas no such symptoms were experienced in the non-sensitized group. It was thus demonstrated that symptoms from the eyes and nose were the most common and the first symptoms to appear after commencement of exposure which is consistent with previous findings [5,7,25]. Exposure-response relationship with regard to prevalence of sensitization has been demonstrated in laboratory animal workers [26] and is more pronounced in atopic workers. In our study individual (personal) exposure levels to allergen and endotoxin were not assessed, but average exposure in animal facilities as an estimate of exposure levels are presented. Longitudinal data collected over a 12 year period, showed an elevated incidence rate of laboratory animal allergy (LAA) for workers with more than 6 h per week exposure to LA [3]. The average exposure time in our study exceeded this limit in all groups but with a large individual variation. Lung function deteriorates with increasing age in healthy individuals [13,14]. In the present study VC and FEV1, decreased over time whereas the FEV1/VC-ratio remained unchanged throughout the study. After correction for age there was an additional VC and FEV1 impairment added to the age-related lung function deterioration, following exposure in laboratory animal facilities. The decrease of FEV1 over time was calculated using local reference values and when re-calculated using GLI2012 reference lung
function deterioration over time was even more pronounced. The decrease in VC and FEV1 was not associated with a FEV1/VC-ratio decrease indicating a development of restrictive rather than obstructive ventilation impairment. The decrease in VC and FEV1 was close to 1% of predicted value during the two years of follow up, which is a doubling of the deterioration rate which is expected by ageing. The findings thus indicates that exposure in laboratory animal facilities accelerates lung function impairment over time. The consequence of this is not clear and we are not aware of any prolonged prospective studies of laboratory animal workers showing the consequences of prolonged continuous exposure. Pacheco et al. found, that laboratory animal workers exposed to the highest cumulative endotoxin exposure demonstrate a significant CD14 gene-environment interaction associated with lung function [27]. Atopic workers were particularly affected by endotoxin exposure but, in our study we saw no significant difference in lung function decrease between atopic and non-atopic subjects. Atopic subjects had enhanced bronchial responsiveness to methacholine compared with non-atopic at study entry and increased bronchial responsiveness was identified as a risk factor for allergic sensitisation. We also showed that low bronchial responsiveness to methacholine was associated with a lower risk of developing LAA. These findings are in line with previous results in a 2 years follow-up cohort study [22] where atopy in addition with family history of allergy, bronchial methacholine threshold, IgE level and working with rats were associated with an increased risk of becoming sensitized to rat and mice. We observed increased bronchial responsiveness during follow up only in those who were atopic whereas bronchial responsiveness remained unaltered in subjects who were non-atopic at study entry. This finding indicates an on-going progressive process once the allergic sensitisation has been established and the exposure is maintained. It is noteworthy that only bronchial responsiveness to a direct stimulus is associated with allergic sensitisation whereas bronchial response to an indirect stimulus (dry air hyperpnoea) was insensitive in this respect. Intracellular lymphocyte IL-2 expression and the number of eosinophils in peripheral blood increased with time in the whole
L. Palmberg et al. / Respiratory Medicine 109 (2015) 1164e1173
1171
Fig. 4. Lung function (VC and FEV1) in percent of predicted values before and after 6, 12 and 24 months of employment in sensitized and non-sensitized, and atopic and non-atopic groups. Mean and 95% CI. x indicates p < 0.05 within group.* indicates p < 0.05 between groups.
Fig. 5. Exhaled NO before and after 6, 12 and 24 months of employment in sensitized and non-sensitized, atopic and non-atopic groups. Median, 25th and 75th percentiles. * indicates p < 0.05 and **p < 0.01 between groups.
study population. Enhanced expression of IL-2 indicates proliferation of T-lymphocytes of T helper 1 (Th1) type and activation of Bcells, while increased number of eosinophil's is a characteristic feature of allergic conditions. Significant changes over time in this respect were observed only in the non-sensitized group. The finding may reflect a small but important difference in levels of the measured parameters at study entry. The exposure seems to induce activation of T-cells and eosinophils in non-sensitized subjects whereas the exposure, in sensitised subjects, is not strong enough
to affect cells which are already activated. Krop et al. showed IL-4 response ex vivo in subjects sensitized to rat [28] but when measuring intracellular IL-4 in our study we saw only response in atopic subjects. Sensitized subjects had higher total IgE, Phadiatop®, and specific IgE (e87 and e88) than had non-sensitised subjects. As a consequence of this the risk of allergic sensitisation to laboratory animals was increased with increased levels of total IgE, Phadiatop®, and specific IgE (e87 and e88) at study entry. These results indicate that
1172
L. Palmberg et al. / Respiratory Medicine 109 (2015) 1164e1173
Fig. 6. Intracellular Th1 (IL-2, IFNg) and Th2 (IL-4, IL-13) cytokines before and after 6, 12 and 24 months of employment in sensitized and non-sensitized (A), atopic and non-atopic subjects (B). Median, 25th and 75th percentiles. xx indicates p < 0.01 and xxxp < 0.001 within group. * indicates p < 0.05 and **p < 0.01 between groups.
the subjects sensitized to both primary and secondary LAA were atopic, a finding that is supported by others [24]. In a10-year LAA prevention program it was also stated that atopy was a stronger risk factor for develop LAA than a family history of allergy or asthma [29]. Atopic subjects had higher levels of exhaled NO than non-atopic subjects at study entry and during follow-up. In addition, exhaled NO increased more in the sensitized group than, in the nonsensitized group during the first year. This is in line with Adisesh
et al. who, in a cross-sectional study, found that exhaled NO is increased in the beginning of the process of developing laboratory animal allergy, predominantly in those who develop asthma [30]. In the present study the levels of exhaled NO increased more in the sensitised group, as could be anticipated, but the levels of exhaled NO at study entry did not constitute a predictor for development of LAA. We are aware of the fact that the absence of an unexposed control group is a weakness of our study. The subjects are, however,
L. Palmberg et al. / Respiratory Medicine 109 (2015) 1164e1173
followed repeatedly during two years thereby constituting their own controls. 5. Conclusions This study established that one of five newly employed subjects became sensitised to laboratory animals within two years from the start of employment in the laboratory. The main risk factor for laboratory animal sensitization was atopy. Furthermore, we found progressive lung function impairment and an increased level of circulating eosinophils and IL-2 producing lymphocytes in the whole study population indicating systemic effects caused by exposure in laboratory animal facilities irrespective of allergic sensitization. Competing interests None of the authors have declared any conflict of interest. Authors'contribution LP, KL conception and design, analysis and interpretation and drafting the manuscript for important intellectual content. BMS, KS recruitment of patients, analysis and interpretation and drafting the manuscript for important intellectual content. AL, MK analysis and interpretation and drafting the manuscript for important intellectual content. All authors read and approved the final manuscript. Acknowledgement The authors would like to acknowledge Inger Ericsson, Kicki Olsson, Gun-Britt Åkerman, Kristin Blidberg, Karin Strandberg, Ida ele and Anne Renstro €m for excellent technical assistance. von Sche The study was supported from Swedish Council for Working Life and Social Research, Swedish Asthma and Allergy Association, Konsul Th Berg's and Karolinska Institutet. References [1] S.K. Meredith, V.M. Taylor, J.C. McDonald, Occupational respiratory disease in the United Kingdom 1989: a report to the British Thoracic Society and the Society of Occupational Medicine by the SWORD project group, Br. J. Ind. Med. 48 (5) (1991) 292e298. [2] J. Tan, J.A. Bernstein, Occupational asthma: an overview, Curr. Allergy Asthma Rep. 14 (5) (2014) 431. [3] L. Elliott, D. Heederik, S. Marshall, D. Peden, D. Loomis, Incidence of allergy and allergy symptoms among workers exposed to laboratory animals, Occup. Environ. Med. 62 (11) (2005) 766e771. [4] D.H. Bryant, L.M. Boscato, P.N. Mboloi, M.C. Stuart, Allergy to laboratory animals among animal handlers, Med. J. Aust. 163 (8) (1995) 415e418. [5] P. Cullinan, D. Lowson, M.J. Nieuwenhuijsen, S. Gordon, R.D. Tee, K.M. Venables, et al., Work related symptoms, sensitisation, and estimated exposure in workers not previously exposed to laboratory rats, Occup. Environ. Med. 51 (9) (1994) 589e592. [6] K. Aoyama, A. Ueda, F. Manda, T. Matsushita, T. Ueda, C. Yamauchi, Allergy to laboratory animals: an epidemiological study, Br. J. Ind. Med. 49 (1) (1992) 41e47.
1173
[7] R.K. Bush, R.A. Wood, P.A. Eggleston, Laboratory animal allergy, J. Allergy Clin. Immunol. 102 (1) (1998) 99e112. [8] D. Heederik, R. Brouwer, K. Biersteker, J.S. Boleij, Relationship of airborne endotoxin and bacteria levels in pig farms with the lung function and respiratory symptoms of farmers, Int. Arch. Occup. Environ. Health 62 (8) (1991) 595e601. [9] B.M. Sundblad, I. von Scheele, L. Palmberg, M. Olsson, K. Larsson, Repeated exposure to organic material alters inflammatory and physiological airway responses, Eur. Respir. J. 34 (1) (2009) 80e88. [10] L.A. Smit, D. Heederik, G. Doekes, J.W. Lammers, I.M. Wouters, Occupational endotoxin exposure reduces the risk of atopic sensitization but increases the risk of bronchial hyperresponsiveness, Int. Arch. Allergy Immunol. 152 (2) (2010) 151e158. [11] A. Renstrom, G. Bjoring, A.U. Hoglund, Evaluation of individually ventilated cage systems for laboratory rodents: occupational health aspects, Lab. Anim. 35 (1) (2001) 42e50. [12] ATS, Standardization of spirometry, Am. J. Respir. Crit. Care Med. 152 (1995) 1107e1136. € m, P. Malmberg, K. Agarwal, Reference values for lung function [13] H. Hedenstro tests in females: regression equations with smoking variables, Bull. Eur. Physiopathol. Respir. 21 (1985) 551e557. € m, P. Malmberg, H.V. Fridriksson, Reference values for pulmo[14] H. Hedenstro nary function tests in men: regression equations which include tobacco smoking variables, Upsala J. Med. Sci. 91 (1986) 299e310. [15] P. Malmberg, K. Larsson, S. Thunberg, Increased lung deposition and biological effect of methacholine by use of drying device for bronchial provocation tests, Eur. Respir. J. 4 (1991) 890e898. [16] G. Argyros, J. Roach, K. Hurwitz, A. Eliasson, Y. Phillips, Eucapnic volontary hyperventilation as a bronchoprovocation technique, Chest 109 (1996) 1520e1524. [17] Society AT, ATS/ERS recommendations for standardized procedures for the online and offline measurement of exhaled lower respiratory nitric oxide and nasal nitric oxide, Am. J. Respir. Crit. Care Med. 171 (8) (2005) 912e930. [18] R. Bascom, U. Pipkorn, L. Lichtenstein, R. Naclerio, The influx of inflammatory cells into nasal washings during the late respose to antigen challenge, Am. Rev. Respir. 138 (1988) 406e412. [19] B.-M. Larsson, L. Palmberg, P.O. Malmberg, K. Larsson, Effect of exposure to swine dust on levels of IL-8 in airway lavage fluid, Thorax 52 (1997) 638e642. [20] K. Sahlander, K. Larsson, L. Palmberg. Increased serum levels of soluble ST2 in birch pollen atopics and individuals working in laboratory animal facilities. J. Occup. Environ. Med.;52(2):214e218. [21] K. Larsson, G. Tornling, D. Gavhed, C. Muller-Suur, L. Palmberg, Inhalation of cold air increases the number of inflammatory cells in the lungs in healthy subjects, Eur. Respir. J. 12 (4) (1998) 825e830. [22] E.J. Krop, D.J. Heederik, R. Lutter, G. de Meer, R.C. Aalberse, H.M. Jansen, et al., Associations between pre-employment immunologic and airway mucosal factors and the development of occupational allergy, J. Allergy Clin. Immunol. 123 (3) (2009) 694e700, 700 e1e3. [23] Lutsky II, I. Neuman, Laboratory animal dander allergy: I. An occupational disease, Ann. Allergy 35 (4) (1975) 201e205. [24] A. Hollander, G. Doekes, D. Heederik, Cat and dog allergy and total IgE as risk factors of laboratory animal allergy, J. Allergy Clin. Immunol. 98 (3) (1996) 545e554. [25] R.K. Bush, G.M. Stave, Laboratory animal allergy: an update, ILAR J. 44 (1) (2003) 28e51. [26] A. Hollander, D. Heederik, G. Doekes, Respiratory allergy to rats: exposureresponse relationships in laboratory animal workers, Am. J. Respir. Crit. Care Med. 155 (2) (1997) 562e567. [27] K.A. Pacheco, C.S. Rose, L.J. Silveira, M.V. Van Dyke, K. Goelz, K. MacPhail, et al., Gene-environment interactions influence airways function in laboratory animal workers, J. Allergy Clin. Immunol. 126 (2) (2010) 232e240. [28] E.J. Krop, M.A. van de Pol, R. Lutter, D.J. Heederik, R.C. Aalberse, J.S. van der Zee, Dynamics in cytokine responses during the development of occupational sensitization to rats, Allergy 65 (10) (2010) 1227e1233. [29] L.E. Goodno, G.M. Stave, Primary and secondary allergies to laboratory animals, J. Occup. Environ. Med. 44 (12) (2002) 1143e1152. [30] L.A.K.S. Adisesh, D.H. Yates, D.C. Snashell, A.J. Newman-Taylor, P.J. Barnes, Exhaled and nasal nitric oxide is increased in laboratory animal allergy, Clin. Exp. Allergy 28 (1997) 876e880.
Contents lists available at ScienceDirect
Respiratory Medicine journal homepage: www.elsevier.com/locate/rmed
Long term effect and allergic sensitization in newly employed workers in laboratory animal facilities Lena Palmberg, Britt-Marie Sundblad*, Anders Lindberg, Maciej Kupczyk 1, Karin Sahlander, Kjell Larsson Lung and Airway Research, Institute of Environmental Medicine, Karolinska Institutet, Stockholm, Sweden
a r t i c l e i n f o
a b s t r a c t
Article history: Received 25 March 2015 Received in revised form 4 June 2015 Accepted 12 June 2015 Available online 15 June 2015
Objective: The aim of this study was to identify targets predicting allergic sensitization to laboratory animals using shift in skin prick test to laboratory animals as primary outcome variable. Methods: In a prospective longitudinal study, personnel who were employed to work with laboratory animals at a medical university were investigated before and 6, 12 and 24 month after the start of employment. Lung function, bronchial challenges, exhaled NO and nasal lavage were performed and blood samples were drawn at all visits. Results: Seventy subjects attended all four visits and 13 (19%) became sensitized to laboratory animals during the two years of follow up. Lung function (VC and FEV1) deteriorated and blood levels of eosinophils and IL-2 producing lymphocytes increased after 24 months. An increased risk of developing laboratory animal allergy was significantly associated with female sex, atopy, symptoms associated with exposure to laboratory animals, low proportion of blood CD4þ cells, specific IgE to rat and mouse and high total IgE when starting to work with laboratory animals. Conclusions: A sensitization rate of 19% in 2 years, were demonstrated in laboratory animal workers. Atopy, increased total and specific IgE levels (rat and mouse) were the strongest predictors for laboratory animal sensitization. The progressive lung function impairment over time, observed in the whole study population may indicate that exposure in animal facilities induces harmful effects, irrespective to allergic sensitization. © 2015 Elsevier Ltd. All rights reserved.
Keywords: Laboratory animal allergy (LAA) Exposure Sensitization Risk factors
1. Introduction Occupational asthma represents 5e15 % of all adult asthma and is today the most common work related airways disease in the western world [1,2]. The prevalence of allergy has increased during the last decades and allergy against laboratory animals (LA) is a frequently observed work-related health problem among pharmaceutical research workers [3]. In cross-sectional studies up to nearly two thirds of the personnel who are exposed to LA report airway symptoms while working with animals [4,5] and approximately one of four is sensitized against LA. Of those who experience symptoms approximately 40% have skin symptoms including
* Corresponding author. Lung and Airway Research, Institute of Environmental Medicine, Karolinska Institutet, PO Box 287, S-171 77 Stockholm, Sweden. E-mail address: [email protected] (B.-M. Sundblad). 1 Present address: Department of Internal Medicine, Asthma and Allergy, Medical University of Lodz, Poland. http://dx.doi.org/10.1016/j.rmed.2015.06.007 0954-6111/© 2015 Elsevier Ltd. All rights reserved.
urticaria [6], and up to 80% have allergic rhino-conjunctivitis with nasal congestion, rhinorrhea, sneezing, and itchy, watery eyes [7]. Cough, shortness of breath and, wheeze, i.e., symptoms indicating asthma, have been reported in up to 22% of individuals exposed to LA. The onset of symptoms after commencing work with LA varies. In workers without previous rat exposure and a range of less than 30 dayse1369 days from the time of employment to the onset of symptoms was reported. The mean duration of employment before symptom onset was approximately one year for chest and skin symptoms and slightly more than 200 days for nose and eye symptoms [5]. Working in laboratory animal facilities not only involves exposure to laboratory animal allergens but also to non-allergenic compounds such as microbial products like endotoxins and peptidoglycans. Endotoxin (lipopolysaccharide, LPS) is recognised as a pro-inflammatory stimulus and exposure to endotoxin correlates with lung function impairment in farmers [8]. Effects of exposure to
L. Palmberg et al. / Respiratory Medicine 109 (2015) 1164e1173
non-allergenic agents in laboratory animal facilities have, however, not been much studied. It is most likely that exposure in laboratory animal facilities constitutes a pro-inflammatory stimulus which may influence lung function and bronchial responsiveness as have been demonstrated in other environments such as exposure in pig barns [9]. In farmers and agriculture industry workers occupational endotoxin exposure seems to reduce the risk of atopic sensitization whereas there is an association between exposure and bronchial hyper-responsiveness [10]. Individuals who applied for a job involving work with LA, at Karolinska Institutet, were longitudinally, prospectively, followed during the two first years of employment. The study was based on the assumption that 20e25% of the exposed employees would develop laboratory animal allergy (LAA) and the primary endpoint was a shift of skin prick test to LA. The main aims were to identify phenotypes and biological markers predicting development of allergy and symptoms and to explore possible effects of exposure to non-allergenic stimuli in laboratory animal facilities.
2. Material and methods 2.1. Study design Persons who applied for a job at Karolinska Institutet were eligible during a period of 3 years (2003e2006) and were included if they had the intention to work with LA and to remain as an employee at the work place for at least 2 years. Subjects were excluded if they had been working with LA during the last 6 months prior to employment and more than 2 years in total. Those who had a history of exposure to LA and positive skin prick test for LA (rat, mouse, rabbit, guinea pig and hamster) were excluded. Five subjects who at baseline had positive SPT to LA without any occupational exposure previous or during the study to these animals were included in the study to find out possible switch in SPT to other LA which they were exposed to during the time of employment. At the start of the study all participants were informed about the health risk associated with laboratory animal work and the importance of taking protective measures. The participants attended the laboratory twice, in the morning, at 4 occasions, prior to employment and after 6, 12 and 24 months of employment. At all visits a questionnaire, skin prick test, lung function test, exhaled nitric oxide (NO) measurement and measurement of bronchial responsiveness to a direct (methacholine) and an indirect (dry air hyperpnoea) stimulus were performed, the bronchial challenges on two separate days. Blood samples and nasal lavage were also collected. Asthma-medication (glucocorticosteroids and b2-agonists) were discontinued 24 h and anti-histamines 5 days prior to each visit. The study protocol was approved by the Karolinska Institute Ethics Committee (Protocol number 02-287, Stockholm, Sweden), and all individuals gave their informed consent to participate.
2.2. Exposure measurements Exposure in 11 animal facilities were assessed by personal air sampling described elsewere [11]. Allergen measurements of rat n1 and mouse m1 were made using unamplified sandwich ELISA according to previously described methods [11]. Endotoxin concentrations were analysed using a kinetic technique version of Limulus amebocyte lysate assay (Limulus Amebocyte lysate, Endosafe® Endochrome-K™ U.S. Lisence No. 1197, Coatech AB, Kungsbacka, Sweden), with E. coli 0111:B4 as standard. The detection limit was 0.1 EU/mL.
1165
2.3. Skin prick test (SPT) A simple skin prick test was performed using a panel of common allergens and allergens from commonly used laboratory animals (birch, timothy, mugwort, cat, dog, guinea pig, rabbit, hamster, Determatophagoides farinae, Cladosporium, Alternaria, (ALK-Abello A/S, Denmark) and rat, mouse, horse, latex, Dermatophagoides pteronyssinus, and Lepidoglyphus destructor, (Alergopharma, Germany)). Histamine dihydrochloride (10 mg/ml) served as positive and the diluent as negative control. A wheel of 3 mm in diameter was considered as positive reaction. 2.4. Questionnaire Two extensive questionnaires were used. The first questionnaire contained questions on heredity, smoking habits, usage of safety equipment, current and previous exposure to animals, airways disease, allergy, medication and eyes, nose, lower respiratory tract and skin symptoms. The second questionnaire contained detailed questions on current and previous work with laboratory animals and symptoms in connection with laboratory animal work. The questionnaires were sent to the participants who responded to the questions in advance and were then individually interviewed at each visit and exposure protection advices were given and followed up. 2.5. Lung function and bronchial responsiveness Forced expiratory volume in 1 s (FEV1) and vital capacity (VC) were measured using a wedge-spirometer (Vitalograph®, UK) according to the American Thoracic Society recommendations [12]. Local reference values were used [13,14]. Direct bronchial responsiveness was assessed by a methacholine challenge [15]. The subjects inhaled a nebulized methacholine solution of increasing concentrations, starting at 0.5 mg/mL followed by doubling doses until FEV1 decreased by 20% or the maximal concentration (32 mg/mL) was reached. On a separate day indirect bronchial responsiveness was assessed by hyperpnoea of dry air containing 5% carbon dioxide of ambient temperature breathed through a low-resistance, one-way valve in the sitting position (Aiolos Asthma Test®, Karlstad, Sweden). The target ventilation (35 FEV1 0.75 L) was maintained for 6 min and FVC and FEV1 were measured before and immediately, 1, 2, 5, 10, and 20 min after finishing the hyperpnoea. The highest, of two FEV1 measurements was registered. A fall in FEV1 of >10% of the pre-challenge value was regarded as a positive test [16]. 2.6. Measurement of exhaled nitric oxide Nitric oxide in exhaled air was analysed with chemiluminescence after reaction with ozone (NIOX®, Aerocrine, Sweden) according to current recommendations [17]. The subjects rinsed their mouth with water and then with sodium bicarbonate (10%) prior to the measurement. All the subjects were asked to refrain from eating green vegetables on the day of the visits. 2.7. Nasal lavage Nasal lavage was performed according to Bascom et al. [18] with minor modifications [19]. Briefly, the head was tilted back and then one nostril was flushed with 5 ml of 0.9 mg/mL saline. After 10 s the head was bent forward and the fluid was expelled into a plastic cup. The procedure was repeated in the other nostril and the lavage samples were pooled. After centrifugation, the supernatant was frozen (70 C) until further analyses.
1166
L. Palmberg et al. / Respiratory Medicine 109 (2015) 1164e1173
2.8. Flow cytometry The blood samples were collected in EDTA vacutainer tubes (BD Bioscience) and a flow cytometric analysis was performed within 2 h using FACS Calibur™ (BD Bioscience). Fifty mL of blood and 20 mL MultiTEST CD45PerCP/CD3FITC/CD4APC/CD8PE (BD Bioscience) were added to TruCOUNT™ tubes and then incubated in darkness for 15 min. The red blood cells were then lysed with 1 mL FACS Lysing Solution™ (BD Bioscience). An absolute count of lymphocytes, monocytes, neutrophils, eosinophils, basophils, CD3þ, CD4þ and CD3þCD8þ subsets of T cells were determined by analysis in Multiset (BD Bioscience) using the TruCOUNT™ beads as an internal reference population. Interleukin-2, IL-4, IL-13 and IFN-g were measured in lymphocytes with intracellular cytokine staining. Heparinized peripheral blood in RPMI 1640 containing 2 mM glutamine (1:1 dilution) was stimulated with phorbol 12- myristate 13-acetate (PMA, 25 ng/mL) and ionomycin (1 mM, SigmaeAldrich) in presence of GolgiPlug (Brefeldin A, BD Bioscience) for 4 h at 37 C. Stimulated blood was stained with anti-CD4 APC (BD Bioscience). The samples were then lysed and incubated with Cytofix/Cytoperm™ (BD Bioscience Pharmingen). The permeabilized cells were washed with Perm/ Wash and resuspended in CellFIX™ (BD Bioscience). Unstimulated blood was stained using the same procedure as negative controls. Analysis was performed using a FACSCalibur™ (BD Bioscience) and CELLQuest™. The results were presented as proportion of cytokineproducing T-helper (CD4þ) cells [20]. 2.9. Analyses of CXCL-8 and ST2 The CXCL-8 (IL-8) concentration was measured in nasal lavage fluid using an in-house ELISA method. Commercially available antibody pairs (R&D system, Europe, Abingdon, UK) were used as previously described [21]. The detection range was 40e3200 pg/ mL. For serum analyses of sST2 (soluble Suppression of Tumorgenicity 2) DuoSet ELISA kit (R&D system, Europe, Abingdon, UK) was used. The detection range was 15.6e2000 pg/mL. For duplicate samples, an intra-assay coefficient of <10% was accepted for CXCL-8 and sST2. 2.10. Total and specific IgE Total Immunoglobulin E (tIgE), Phadiatop® (specific IgE of a mix of the most common allergens in Sweden) and specific IgE to rat (e87) and mouse (e88) allergens were measured in serum using ImmunoCAP100 (Phadia, Sweden). A specific IgE level above 0.35 kUA/L was considered positive. 2.11. Statistical methods Data are presented as mean and range (age and height), mean and 95% confidence intervals (lung function, exposure), or as median (25th 75th percentiles). Within- and between group comparisons of lung function data were assessed by ANOVA with Students paired or unpaired t-test when appropriate. Within group-comparisons for other outcome variables were assessed by Friedmans repeated measurements followed by Wilcoxon signed rank test when appropriate or directly with Wilcoxon signed rank test for parameters only measured twice. Between groups comparisons were assessed by Mann Whitney U test. Symptoms and skin prick test were analysed by means of Chi-square test. A p-value 0.05 was considered significant.
To identify factors, associated with allergic sensitization to laboratory animals, those who shifted from negative to positive skin prick test to laboratory animals during the study were compared to those who remained negative in skin prick test throughout the study. Odds ratios (OR) and risk ratios (RR) were assessed using univariate analysis of the following parameters: atopy (defined as at least 1 positive skin prick test to any allergen at baseline visit), positive skin prick test to cat and dog, female sex, age, smoking at study entry, symptoms (symptoms related to laboratory animal contact) at study entry, lung function, bronchial responsiveness, biomarkers (IL-2/CD4þ, IL-4/CD4þ, IL-13/CD4þ, IFN-g/CD4þ, CD3, CD4þ and CD8þ, cells and CD4/CD8 ratio, neutrophils, monocytes, lymphocytes, basophils and eosinophils in peripheral blood, IL-8 levels in nasal lavage fluid, total IgE, Phad, e87 and e88 and exhaled NO). For numerical variables median, 25 percentile and 75 percentile as possible cut-off values were evaluated. Chi-square test (or Fisher's exact test wherever appropriate) was applied. A multivariate logistic regression model including all potential predictors with univariate p-value less than 0.1 was built and adjusted for atopy, sex, age and smoking status. SPSS version 17.0 software (SPSS Inc, Chicago, Ill) was used for uni- and multivariate analyses and otherwise using Stat View (version 5.01; SAS Institute).
3. Results Out of 249 eligible subjects 119 fulfilled the inclusion criteria and 130 were excluded because they had worked for more than 2 years and the last 6 month with LA or, had positive SPT to mouse/rat that they have been occupationally exposed to or had planned to remain at the work place for less than 2 years. Of the included subjects, 17 declined participation, 102 attended visit 1 and 70 subjects attended all four visits. The reasons for loss of follow up (n ¼ 32) were termination of employment, pregnancy and unwillingness to continue. For analyses subjects were divided into subgroups based on two approaches (Fig. 1). The sensitized group (n ¼ 13, 18.6%), those who shifted from negative to positive skin prick test to laboratory animals during the study. Subjects (n ¼ 2) who had a positive SPT to LA at baseline without being occupationally exposed to this animal during the study and who became sensitized to another animal during the study were included in this group. The non-sensitized group (n ¼ 57, 81.4%), those who did not shift in skin prick test during the study. Subjects (n ¼ 3) who had a positive SPT to LA at baseline without being occupationally exposed to this animal during the study and who did not become sensitized to any other animal were included in this group. The atopic group (n ¼ 26, 37.1%), positive skin prick test to any allergen at study entry. The non-atopic group (n ¼ 44, 62.9%), no positive skin prick test at study entry.
3.1. Exposure The mean time of work with laboratory animals was 33 (25e42) hours/months with no difference between the groups (p ¼ 0.55; Table 1). In total, 187 air samples for allergen and 133 for endotoxin measurement were collected. Average endotoxin level was <1.2 mg/m3 and allergen levels of rat n1 and mouse m1 were 5.0 (0.01e115) ng/m3 and 73.9 (0.01e1168) ng/m3, respectively, in the animal facilities.
L. Palmberg et al. / Respiratory Medicine 109 (2015) 1164e1173
1167
Fig. 1. Sensitized and non-sensitized participating subjects in the groups with and without atopy at study entry. Shift in skin prick test to laboratory animal indicates change from negative to positive reaction to laboratory animals.
Table 1 Lung function and bronchial responsiveness in sensitized/non-sensitized and atopic/non-atopic subgroups.
M/F Smokers, n (%) Age, year, mean (range) Height, cm, mean (range) VC, L, mean (95%CI) FEV1, L, mean (95%CI) FEV1/VC, mean (95%CI) PD20, mg, median (25the75th percentile) Eucapnic hyperventilation, %FEV1 decrease, mean (95%CI) Laboratory animal work, hours/months, mean (95%CI)
All n ¼ 70
Sensitized n ¼ 13
Non-sensitized n ¼ 57
Atopic n ¼ 26
Non-atopic n ¼ 44
21/49 10 (14.3) 28 (20e53) 171 (155e195) 4.45 (4.21e4.68) 3.78 (3.60e3.97) 0.85 (0.84e0.88) 1.17 (0.36e2.71) 3.29 (2.68e3.90) 33 (25e42)
2/11 2 (15.4) 28 (21e53) 168 (158e188) 4.15 (4.25e4.78) 3.51 (4.25e4.78) 0.85 (0.81e0.88) 0.54 (0.40e1.18) 3.24 (2.17e4.30) 41 (11e71)
19/38 8 (14.0) 27 (20e43) 171 (155e195) 4.52 (4.25e4.78) 3.84 (3.63e4.05) 0.85 (0.84e0.87) 1.30 (0.36e3.18) 3.30 (2.58e4.01) 32 (23e41)
9/17 4 (15.4) 29 (20e53) 172 (159e195) 4.48 (4.09e4.87) 3.74 (3.44e4.03) 0.84 (0.81e0.87) 0.54 (0.19e1.91) 2.78 (1.90e3.66) 30 (16e44)
12/32 6 (13.6) 27 (20e39) 169 (155e195) 4.43 (4.13e4.74) 3.80 (3.55e4.06) 0.86 (0.84e0.88) 1.30 (0.43e10.2) 3.59 (2.77e4.42) 36 (24e47)
3.2. Skin prick test, Phadiatop® and total IgE Twenty-six subjects had positive skin prick test to any allergen at study entry, 13 shifted from negative to positive skin prick test to LA during the study. Two of the subjects showed conversion of two LA. The SPT reactions were not persistent, but shifted both in size and over time.
Phadiatop® and total IgE decreased during the two years study in the whole group (p 0.0003), in the non-sensitized (p 0.003), non-atopic (p < 0.05) and atopic (p 0.03) groups. In the sensitized group Phadioatop® (p ¼ 0.22) and total IgE (p ¼ 0.60) did not change during the study period. After two years of exposure the decrease in Phadiatop® in the atopic group exceeded that of the non-atopic group (p < 0.0001) (Fig. 2).
1168
L. Palmberg et al. / Respiratory Medicine 109 (2015) 1164e1173
Fig. 2. Blood eosinophils and basophils before and after 6, 12 and 24 months and total IgE and Phadiatop® before and 24 months of employment in sensitized and non-sensitized (A), atopic and non-atopic subjects (B). Median, 25th and 75th percentiles. x indicates p < 0.05, xxp < 0.01 and xxxp < 0.001 within group. ** indicates p < 0.01 and ***p < 0.001 between groups.
L. Palmberg et al. / Respiratory Medicine 109 (2015) 1164e1173
3.3. Baseline factors associated with development of laboratory animal allergy In the univariate model high bronchial responsiveness to methacholine (median slope >17.17%/mg methacholine) was associated with an increased risk of skin prick test shift to laboratory animals (p ¼ 0.034). Possible protective factors included low bronchial responsiveness to methacholine indicated by PD20FEV1 >2.71 mg (75th percentile; p ¼ 0.028) and slope <6.66%/mg (25th percentile) methacholine (p ¼ 0.028) and Phadiatop® <0.05, (25% percentile; p ¼ 0.03), e87 < 0.003 (25th percentile; p ¼ 0.029), and e88 ¼ 0 (25th percentile; p ¼ 0.0034) at study entry. Major risk factors for a skin prick test shift to laboratory animals included positive skin prick tests to cat and dog, serological indicators of atopy and low proportion of circulating CD3þ cells (<1.28 cells/mL) at study entry (Table 2). When adjusted for possible confounding factors in multivariate logistic regression models, female sex, atopy (e87 > 0.01, e88 > 0.004, total IgE > 15.5 kUA/L), low proportion of circulating CD3 cells (<0.92 cells/mL) and symptoms remained significantly associated with an increased risk of developing laboratory animal allergy (Table 2).
1169
3.6. Exhaled NO Exhaled NO differed between atopic and non-atopic at all visits (Fig. 5). In the sensitized group exhaled NO level was higher after 6 (p < 0.01) and 12 (p < 0.05) months of employment than in the nonsensitized group but no difference was found at study entry and after 24 months of employment. 3.7. Blood
The prevalence of work-related symptoms did not change during the study within the four subgroups, (Fig. 3). At 12 months the sensitized group had more symptoms from the eyes (p ¼ 0.04) and nose (p ¼ 0.03) than had the non-sensitized group. Symptoms did not differ significantly between the atopic and non-atopic groups.
The number of eosinophils increased in the whole group and in the non-sensitized, and the non-atopic groups (p < 0.0001 for all) and basophils in blood increased over time in the atopic group (p ¼ 0.03), (Fig. 2). Intracellular expression of IL-2 increased in the whole group (p ¼ 0.0003), in the non-sensitized group (p ¼ 0.0006) and the nonatopic group (p ¼ 0.0018) during the 2 years of follow up (Fig. 6). Interferon- g did not change over time in any subgroup but was higher in the atopic than in the non-atopic group after 6 (p ¼ 0.03) and 24 months (p ¼ 0.04) of employment (Fig. 6). The intracellular expression of IL-4 and IL-13 was enhanced in atopic group compared with the non-atopic group at study entry (p ¼ 0.001 and p ¼ 0.008, respectively) and after 6 months (p ¼ 0.002 and p ¼ 0.007, respectively). Enhanced expression was also found for IL-4 at 24 months (p ¼ 0.01) and IL-13 at 12 months (p ¼ 0.02) of employment (Fig. 6). Soluble ST2, involved in pro-inflammatory responses and playing a role in allergic inflammation, did not change over time and did not differ between the groups.
3.5. Lung function and BHR
3.8. Nasal lavage
Lung function (VC and FEV1) deteriorated over time in the whole group (n ¼ 70) and in the non-sensitized group (Fig. 4). This result was calculated using local Swedish reference values (13, 14). When using GLI2012 the deterioration lung function was even more enhanced (data not shown). Lung function change did not differ over time between the sensitized and non-sensitized and between the atopic and non-atopic groups. Bronchial responsiveness to methacholine (PD20) differed between the atopic, 0.54 mg (0.19e1.91), median (25the75th perc.) and non-atopic group 1.30 mg (0.43e3.18) at study entry (p ¼ 0.05) and increased significantly during the study in the atopic group (p ¼ 0.02; PD20 after 24 month was 0.38 mg (0.12e1.00)). Bronchial response to hyperpnoea of dry air (% FEV1 decrease) did not change over time in the whole group or in any of the subgroups.
The levels of the inflammatory mediator CXCL-8 in nasal lavage were not altered over time and did not differ between the subgroups.
3.4. Symptoms
4. Discussion One out of five persons who commenced working in laboratory animal facilities became sensitized against LA during the first two years of employment. The incidence of allergic sensitisation is similar to what previously has been reported following exposure to LA [6,22e24]. Of those who became sensitized against LA, more than three out of four were atopic at study entry, clearly showing atopy as the major risk factor. Conversion in SPT to one or more LA was the primary outcome in this study. Eight of the 13 sensitized subjects were sensitized at
Table 2 Adjusted odds ratios in multivariate logistic regression models for risk factors at study entry associated with the development of laboratory animal allergy (in decreasing order). Factor (at study entry)
Adjusted OR
95% CI
P-value
e88 > 0.004, kUA/La Female sexb e87 > 0.01, kUA/La Total IgE > 15.5, kUA/La Atopy, any positive skin prick testc CD3 < 0.92, cells/mLa Symptoms in connection with laboratory animal work, yes/noa e87 < 0.003, kUA/La Phad < 0.05, kUA/La
19.31 9.86 9.05 8.12 7.45 5.40 5.26 0.135 0.076
2.04e182.98 1.08e89.8 1.89e43.3 1.36e48.32 1.71e32.42 1.08e27.09 1.00e27.41 0.03e0.635 0.006e1.045
0.010 0.042 0.006 0.021 0.007 0.040 0.049 0.011 0.054
e87 ¼ specific IgE to rat and e88 specific IgE to mouse. a Adjusted for atopy, sex, age and smoking. b Adjusted for atopy, age and smoking. c Adjusted for sex, age and smoking.
1170
L. Palmberg et al. / Respiratory Medicine 109 (2015) 1164e1173
Fig. 3. Work-related symptoms from airways, skin, eyes and nose before and after 6, 12 and 24 months of employment in the sensitized, non-sensitized, atopic and non-atopic groups. At 12 months the sensitized group had more symptoms from the eyes (p ¼ 0.04) and nose (p ¼ 0.03) than had the non-sensitized group.
the last follow-up at 24 month. Thus, it was impossible to show if the conversion was persistent in these subjects and in some other subjects the positive SPT to LA varied over time. Two explanations to a temporarily positive SPT at one or two follow-up occasions may be related to changed exposure-response and ongoing educational effort with increased prevention. In subjects who became sensitized, symptoms increased following exposure whereas symptoms did not change over time in non-sensitized, atopics or non-atopics indicating a causal relationship between allergic sensitization and the development of clinical symptoms. In the sensitized group symptoms, predominantly from the eyes and nose indicating allergic rhinoconjunctivitis, occurred within one year of employment whereas no such symptoms were experienced in the non-sensitized group. It was thus demonstrated that symptoms from the eyes and nose were the most common and the first symptoms to appear after commencement of exposure which is consistent with previous findings [5,7,25]. Exposure-response relationship with regard to prevalence of sensitization has been demonstrated in laboratory animal workers [26] and is more pronounced in atopic workers. In our study individual (personal) exposure levels to allergen and endotoxin were not assessed, but average exposure in animal facilities as an estimate of exposure levels are presented. Longitudinal data collected over a 12 year period, showed an elevated incidence rate of laboratory animal allergy (LAA) for workers with more than 6 h per week exposure to LA [3]. The average exposure time in our study exceeded this limit in all groups but with a large individual variation. Lung function deteriorates with increasing age in healthy individuals [13,14]. In the present study VC and FEV1, decreased over time whereas the FEV1/VC-ratio remained unchanged throughout the study. After correction for age there was an additional VC and FEV1 impairment added to the age-related lung function deterioration, following exposure in laboratory animal facilities. The decrease of FEV1 over time was calculated using local reference values and when re-calculated using GLI2012 reference lung
function deterioration over time was even more pronounced. The decrease in VC and FEV1 was not associated with a FEV1/VC-ratio decrease indicating a development of restrictive rather than obstructive ventilation impairment. The decrease in VC and FEV1 was close to 1% of predicted value during the two years of follow up, which is a doubling of the deterioration rate which is expected by ageing. The findings thus indicates that exposure in laboratory animal facilities accelerates lung function impairment over time. The consequence of this is not clear and we are not aware of any prolonged prospective studies of laboratory animal workers showing the consequences of prolonged continuous exposure. Pacheco et al. found, that laboratory animal workers exposed to the highest cumulative endotoxin exposure demonstrate a significant CD14 gene-environment interaction associated with lung function [27]. Atopic workers were particularly affected by endotoxin exposure but, in our study we saw no significant difference in lung function decrease between atopic and non-atopic subjects. Atopic subjects had enhanced bronchial responsiveness to methacholine compared with non-atopic at study entry and increased bronchial responsiveness was identified as a risk factor for allergic sensitisation. We also showed that low bronchial responsiveness to methacholine was associated with a lower risk of developing LAA. These findings are in line with previous results in a 2 years follow-up cohort study [22] where atopy in addition with family history of allergy, bronchial methacholine threshold, IgE level and working with rats were associated with an increased risk of becoming sensitized to rat and mice. We observed increased bronchial responsiveness during follow up only in those who were atopic whereas bronchial responsiveness remained unaltered in subjects who were non-atopic at study entry. This finding indicates an on-going progressive process once the allergic sensitisation has been established and the exposure is maintained. It is noteworthy that only bronchial responsiveness to a direct stimulus is associated with allergic sensitisation whereas bronchial response to an indirect stimulus (dry air hyperpnoea) was insensitive in this respect. Intracellular lymphocyte IL-2 expression and the number of eosinophils in peripheral blood increased with time in the whole
L. Palmberg et al. / Respiratory Medicine 109 (2015) 1164e1173
1171
Fig. 4. Lung function (VC and FEV1) in percent of predicted values before and after 6, 12 and 24 months of employment in sensitized and non-sensitized, and atopic and non-atopic groups. Mean and 95% CI. x indicates p < 0.05 within group.* indicates p < 0.05 between groups.
Fig. 5. Exhaled NO before and after 6, 12 and 24 months of employment in sensitized and non-sensitized, atopic and non-atopic groups. Median, 25th and 75th percentiles. * indicates p < 0.05 and **p < 0.01 between groups.
study population. Enhanced expression of IL-2 indicates proliferation of T-lymphocytes of T helper 1 (Th1) type and activation of Bcells, while increased number of eosinophil's is a characteristic feature of allergic conditions. Significant changes over time in this respect were observed only in the non-sensitized group. The finding may reflect a small but important difference in levels of the measured parameters at study entry. The exposure seems to induce activation of T-cells and eosinophils in non-sensitized subjects whereas the exposure, in sensitised subjects, is not strong enough
to affect cells which are already activated. Krop et al. showed IL-4 response ex vivo in subjects sensitized to rat [28] but when measuring intracellular IL-4 in our study we saw only response in atopic subjects. Sensitized subjects had higher total IgE, Phadiatop®, and specific IgE (e87 and e88) than had non-sensitised subjects. As a consequence of this the risk of allergic sensitisation to laboratory animals was increased with increased levels of total IgE, Phadiatop®, and specific IgE (e87 and e88) at study entry. These results indicate that
1172
L. Palmberg et al. / Respiratory Medicine 109 (2015) 1164e1173
Fig. 6. Intracellular Th1 (IL-2, IFNg) and Th2 (IL-4, IL-13) cytokines before and after 6, 12 and 24 months of employment in sensitized and non-sensitized (A), atopic and non-atopic subjects (B). Median, 25th and 75th percentiles. xx indicates p < 0.01 and xxxp < 0.001 within group. * indicates p < 0.05 and **p < 0.01 between groups.
the subjects sensitized to both primary and secondary LAA were atopic, a finding that is supported by others [24]. In a10-year LAA prevention program it was also stated that atopy was a stronger risk factor for develop LAA than a family history of allergy or asthma [29]. Atopic subjects had higher levels of exhaled NO than non-atopic subjects at study entry and during follow-up. In addition, exhaled NO increased more in the sensitized group than, in the nonsensitized group during the first year. This is in line with Adisesh
et al. who, in a cross-sectional study, found that exhaled NO is increased in the beginning of the process of developing laboratory animal allergy, predominantly in those who develop asthma [30]. In the present study the levels of exhaled NO increased more in the sensitised group, as could be anticipated, but the levels of exhaled NO at study entry did not constitute a predictor for development of LAA. We are aware of the fact that the absence of an unexposed control group is a weakness of our study. The subjects are, however,
L. Palmberg et al. / Respiratory Medicine 109 (2015) 1164e1173
followed repeatedly during two years thereby constituting their own controls. 5. Conclusions This study established that one of five newly employed subjects became sensitised to laboratory animals within two years from the start of employment in the laboratory. The main risk factor for laboratory animal sensitization was atopy. Furthermore, we found progressive lung function impairment and an increased level of circulating eosinophils and IL-2 producing lymphocytes in the whole study population indicating systemic effects caused by exposure in laboratory animal facilities irrespective of allergic sensitization. Competing interests None of the authors have declared any conflict of interest. Authors'contribution LP, KL conception and design, analysis and interpretation and drafting the manuscript for important intellectual content. BMS, KS recruitment of patients, analysis and interpretation and drafting the manuscript for important intellectual content. AL, MK analysis and interpretation and drafting the manuscript for important intellectual content. All authors read and approved the final manuscript. Acknowledgement The authors would like to acknowledge Inger Ericsson, Kicki Olsson, Gun-Britt Åkerman, Kristin Blidberg, Karin Strandberg, Ida ele and Anne Renstro €m for excellent technical assistance. von Sche The study was supported from Swedish Council for Working Life and Social Research, Swedish Asthma and Allergy Association, Konsul Th Berg's and Karolinska Institutet. References [1] S.K. Meredith, V.M. Taylor, J.C. McDonald, Occupational respiratory disease in the United Kingdom 1989: a report to the British Thoracic Society and the Society of Occupational Medicine by the SWORD project group, Br. J. Ind. Med. 48 (5) (1991) 292e298. [2] J. Tan, J.A. Bernstein, Occupational asthma: an overview, Curr. Allergy Asthma Rep. 14 (5) (2014) 431. [3] L. Elliott, D. Heederik, S. Marshall, D. Peden, D. Loomis, Incidence of allergy and allergy symptoms among workers exposed to laboratory animals, Occup. Environ. Med. 62 (11) (2005) 766e771. [4] D.H. Bryant, L.M. Boscato, P.N. Mboloi, M.C. Stuart, Allergy to laboratory animals among animal handlers, Med. J. Aust. 163 (8) (1995) 415e418. [5] P. Cullinan, D. Lowson, M.J. Nieuwenhuijsen, S. Gordon, R.D. Tee, K.M. Venables, et al., Work related symptoms, sensitisation, and estimated exposure in workers not previously exposed to laboratory rats, Occup. Environ. Med. 51 (9) (1994) 589e592. [6] K. Aoyama, A. Ueda, F. Manda, T. Matsushita, T. Ueda, C. Yamauchi, Allergy to laboratory animals: an epidemiological study, Br. J. Ind. Med. 49 (1) (1992) 41e47.
1173
[7] R.K. Bush, R.A. Wood, P.A. Eggleston, Laboratory animal allergy, J. Allergy Clin. Immunol. 102 (1) (1998) 99e112. [8] D. Heederik, R. Brouwer, K. Biersteker, J.S. Boleij, Relationship of airborne endotoxin and bacteria levels in pig farms with the lung function and respiratory symptoms of farmers, Int. Arch. Occup. Environ. Health 62 (8) (1991) 595e601. [9] B.M. Sundblad, I. von Scheele, L. Palmberg, M. Olsson, K. Larsson, Repeated exposure to organic material alters inflammatory and physiological airway responses, Eur. Respir. J. 34 (1) (2009) 80e88. [10] L.A. Smit, D. Heederik, G. Doekes, J.W. Lammers, I.M. Wouters, Occupational endotoxin exposure reduces the risk of atopic sensitization but increases the risk of bronchial hyperresponsiveness, Int. Arch. Allergy Immunol. 152 (2) (2010) 151e158. [11] A. Renstrom, G. Bjoring, A.U. Hoglund, Evaluation of individually ventilated cage systems for laboratory rodents: occupational health aspects, Lab. Anim. 35 (1) (2001) 42e50. [12] ATS, Standardization of spirometry, Am. J. Respir. Crit. Care Med. 152 (1995) 1107e1136. € m, P. Malmberg, K. Agarwal, Reference values for lung function [13] H. Hedenstro tests in females: regression equations with smoking variables, Bull. Eur. Physiopathol. Respir. 21 (1985) 551e557. € m, P. Malmberg, H.V. Fridriksson, Reference values for pulmo[14] H. Hedenstro nary function tests in men: regression equations which include tobacco smoking variables, Upsala J. Med. Sci. 91 (1986) 299e310. [15] P. Malmberg, K. Larsson, S. Thunberg, Increased lung deposition and biological effect of methacholine by use of drying device for bronchial provocation tests, Eur. Respir. J. 4 (1991) 890e898. [16] G. Argyros, J. Roach, K. Hurwitz, A. Eliasson, Y. Phillips, Eucapnic volontary hyperventilation as a bronchoprovocation technique, Chest 109 (1996) 1520e1524. [17] Society AT, ATS/ERS recommendations for standardized procedures for the online and offline measurement of exhaled lower respiratory nitric oxide and nasal nitric oxide, Am. J. Respir. Crit. Care Med. 171 (8) (2005) 912e930. [18] R. Bascom, U. Pipkorn, L. Lichtenstein, R. Naclerio, The influx of inflammatory cells into nasal washings during the late respose to antigen challenge, Am. Rev. Respir. 138 (1988) 406e412. [19] B.-M. Larsson, L. Palmberg, P.O. Malmberg, K. Larsson, Effect of exposure to swine dust on levels of IL-8 in airway lavage fluid, Thorax 52 (1997) 638e642. [20] K. Sahlander, K. Larsson, L. Palmberg. Increased serum levels of soluble ST2 in birch pollen atopics and individuals working in laboratory animal facilities. J. Occup. Environ. Med.;52(2):214e218. [21] K. Larsson, G. Tornling, D. Gavhed, C. Muller-Suur, L. Palmberg, Inhalation of cold air increases the number of inflammatory cells in the lungs in healthy subjects, Eur. Respir. J. 12 (4) (1998) 825e830. [22] E.J. Krop, D.J. Heederik, R. Lutter, G. de Meer, R.C. Aalberse, H.M. Jansen, et al., Associations between pre-employment immunologic and airway mucosal factors and the development of occupational allergy, J. Allergy Clin. Immunol. 123 (3) (2009) 694e700, 700 e1e3. [23] Lutsky II, I. Neuman, Laboratory animal dander allergy: I. An occupational disease, Ann. Allergy 35 (4) (1975) 201e205. [24] A. Hollander, G. Doekes, D. Heederik, Cat and dog allergy and total IgE as risk factors of laboratory animal allergy, J. Allergy Clin. Immunol. 98 (3) (1996) 545e554. [25] R.K. Bush, G.M. Stave, Laboratory animal allergy: an update, ILAR J. 44 (1) (2003) 28e51. [26] A. Hollander, D. Heederik, G. Doekes, Respiratory allergy to rats: exposureresponse relationships in laboratory animal workers, Am. J. Respir. Crit. Care Med. 155 (2) (1997) 562e567. [27] K.A. Pacheco, C.S. Rose, L.J. Silveira, M.V. Van Dyke, K. Goelz, K. MacPhail, et al., Gene-environment interactions influence airways function in laboratory animal workers, J. Allergy Clin. Immunol. 126 (2) (2010) 232e240. [28] E.J. Krop, M.A. van de Pol, R. Lutter, D.J. Heederik, R.C. Aalberse, J.S. van der Zee, Dynamics in cytokine responses during the development of occupational sensitization to rats, Allergy 65 (10) (2010) 1227e1233. [29] L.E. Goodno, G.M. Stave, Primary and secondary allergies to laboratory animals, J. Occup. Environ. Med. 44 (12) (2002) 1143e1152. [30] L.A.K.S. Adisesh, D.H. Yates, D.C. Snashell, A.J. Newman-Taylor, P.J. Barnes, Exhaled and nasal nitric oxide is increased in laboratory animal allergy, Clin. Exp. Allergy 28 (1997) 876e880.