Published ahead of print on January 24, 2008, doi:10.1164/rccm.200707-991OC
© 2008 American Thoracic Society doi: 10.1164/rccm.200707-991OC
Long-Term Outcomes in a Prospective Cohort of Apprentices Exposed to High-Molecular-Weight Agents1 Axe de Recherche en Santé Respiratoire, Sacré-Coeur Hospital, Montreal, Canada; 2 The Department of Epidemiology, Biostatistics, and Occupational Health, McGill University, Montreal, Canada; and 3 Institute for Risk Assessment Sciences, Environmental Epidemiology Division, Utrecht University, Utrecht, The Netherlands Correspondence and requests for reprints should be addressed to Denyse Gautrin, Ph.D., Axe de Recherche en Santé Respiratoire, Sacré-Coeur Hospital, 5400 Gouin Blvd West, Montreal, Canada H4J 1C5. E-mail: d.gautrin{at}umontreal.ca
Rationale: We conducted a long-term (8-yr) follow-up of 408 apprentices entering programs involving exposure to high-molecular-weight allergens. Objectives: The objectives were to assess the frequency of new and persisting sensitization, symptoms, and bronchial hyperresponsiveness in relation with job history after ending apprenticeship and to examine characteristics significantly associated with the incidence and remission of these occupational outcomes. Methods: A respiratory symptom questionnaire, skin prick tests with work-related allergens (laboratory animals, flour, and latex), spirometry, and methacholine challenge were administered. The association between incidence or remission of these outcomes and individual characteristics at baseline and end of apprenticeship was examined. Measurements and Main Results: In subjects who at any time during follow-up held a job related to their training (78%), the incidence of sensitization, rhinoconjunctival and chest symptoms, and bronchial hyperresponsiveness at follow-up was 1.3, 1.7, 0.7, and 2.0 per 100 person-years, respectively. The remission of these outcomes acquired during apprenticeship was 18.5, 9.6, 9.6, and 12.4 per 100 person-years, respectively, in subjects no longer in a job related to training. Several clinical, immunological, and functional characteristics at baseline and acquired during apprenticeship were found to be significantly associated with the incidence and remission of the outcomes. Conclusions: The incidence of sensitization, symptoms, and bronchial hyperresponsiveness was lower while at work than during the apprenticeship period. A high proportion of subjects in a job not related to training experienced remission of symptoms acquired during apprenticeship.
Key Words: occupational sensitization • bronchial responsiveness • work-related chest symptoms • work-related rhinoconjunctivitis symptoms • asthma in the workplace
IgE-mediated sensitization and development of rhinoconjunctival and respiratory symptoms represent a significant risk in workers exposed to various high-molecular-weight agents, such as flour in bakers and pastry-makers, latex in health professionals, and laboratory animals in animal health workers (1). Although population- and sentinel-based studies, medicolegal statistics, and cross-sectional studies in high-risk workplaces have been used to assess frequency, few prospective cohorts have been proposed to estimate the risk to health associated with exposure at work to high- and low-molecular-weight agents. Some population-based prospective studies examined the risk of occupations to cause nasal or chest symptoms. Riu and coworkers showed that the risk of developing rhinitis was higher in adolescents enrolled in high-risk occupations (odds ratio [OR] 1.4; 95% confidence interval [CI], 1.0–2.1), especially those exposed to low-molecular-weight agents (2). In the European Community Respiratory Health Society (ECRHS)-II study performed in almost 7,000 participants from 13 countries, a significant asthma risk (OR, 1.6; 95% CI, 1–2.3) was documented in subjects enrolled in jobs with exposure to substances known to cause occupational asthma, including variants with and without a latency period (3). Students entering an apprenticeship program represent a suitable model for examining incidence and risk factors. De Zotti and coworkers have shown that the cumulative incidence of skin sensitization to relevant antigens in 125 trainee bakers was 10.1% after 30 months (4). Walusiak examined 287 apprentice bakers and found that the respective cumulative incidence of sensitization to a program-related antigen, occupational allergic rhinitis, and occupational asthma (OA) was 12.4%, 12.5%, and 8.7% over a 2-year period (5). Since 1992, we have performed a number of apprentice-based cohort studies. In nearly 800 apprentices exposed to high-molecular-weight allergens, the incidence of sensitization to a program-related specific allergen was 7.9 per 100 person-years (PY) in animal health, 4.2 per 100 PY in pastry-making, and 2.5 per 100 PY in dental hygiene (6). The incidence of probable occupational rhinoconjunctivitis (nasal symptoms and specific immunological reactivity) and OA (specific immunological reactivity and bronchial hyperresponsiveness) was 5.7 and 2.7 per 100 PY, respectively, in apprentices exposed to various animals over the 10- to 44-month training period (7). The incidence of probable OA was 3% after 15 months in 286 welding students (8). The use of the prospective model in apprentices not only provides useful information on incidence and risk factors but also offers the possibility of reexamining the apprentices after they enter the workforce. A proportion of these students find a job related to their training, whereas others work in different fields. We conducted a long-term follow-up study of apprentices who trained in a field with exposure to high-molecular-weight agents to assess the incidence and remission of sensitization, symptoms, and bronchial hyperresponsiveness in relation with job history after ending apprenticeship and to examine characteristics that were significantly associated with the incidence and remission of these occupational allergy outcomes.
Study Design The study population was previously described on entering the training programs (9) and at the end of the apprenticeship (6, 7, 10–13). Figure 1 shows the schedule of visits in the current follow-up study.
Subjects The 629 subjects who completed their training and remained in our study until then were eligible for the follow-up study; among them, 181 could not be traced. We contacted 448 subjects, of whom 408 agreed to participate. We have complete questionnaire data for 384 subjects (61.0%). Skin prick tests results were available for 368 subjects (58.5%), and bronchial responsiveness data were available for 281 subjects (44.6%). The 24 remaining subjects cancelled their appointment with the research personnel on more than one occasion or were unavailable within the time frame set for the follow-up.
Methods The follow-up assessment included (1) answering the face-to-face administered standardized International Union Against Tuberculosis and Lung Disease (IUATLD)-ECRHS respiratory questionnaire (14, 15), which contains questions on work-related symptoms and work history after ending apprenticeship; (2) skin prick tests to work-specific allergens (e.g., rat urinary proteins [Allergon AB; Pharmacia, Angelholm, Sweden], mouse dander, rabbit dander, latex, mixed flours, and wheat and bran flour [Omega, Montreal, PQ, Canada]) that were used in studies performed in the same subjects (6); and (3) spirometry and bronchial responsiveness to methacholine (16). These tests were performed at Sacré-Coeur Hospital or at a participating training center; for those unable to come to these sites, a home visit was done that included the questionnaire and skin prick tests. For 16 subjects, the questionnaire was administered by telephone. The outcomes studied were specific skin sensitization, work-related rhinoconjunctivitis and chest symptoms, and bronchial hyperresponsiveness. Skin sensitization to a program-specific allergen, symptoms of work-related rhinoconjunctivitis, and chest symptoms were considered "new" if they were absent through the apprenticeship and present any time after ending apprenticeship. Probable OA was defined as the occurrence of new skin sensitization to a program-specific allergen and significant increase in bronchial responsiveness at follow-up compared with that measured at entry into apprenticeship (3.2-fold or greater change in the provocative concentration of methacholine causing a 20% fall in FEV1–PC20, which corresponds to the 95% CI for within-subject variability of the test (17) as assessed by histamine in our laboratory and assuming that PC20 histamine is highly comparable with PC20 methacholine) (18), or a change at follow-up to a PC20 of 16 mg/ml or less in subjects with an initial PC20 of greater than 32 mg/ml. For each outcome, we defined a profile to describe the time course from end of apprenticeship to follow-up visit after ending apprenticeship. The profiles were negative-negative, negative-positive (new event), positive-negative (remission), and positive-positive (persistence). The events of interest were, for each outcome, incident cases during follow-up (negative-positive) and remission (positive-negative).
A work history was obtained on job(s) held after ending apprenticeship, with start and finish dates, number of hours worked per week, type of industry, main activity of the workplace, and job title. A three-point score was developed that reflected the relatedness of jobs held with training (score 0: not related; score 1: main activity of the workplace related; score 2: job title related). A score was attributed to each participant after a consensus agreement of two authors (DG and ES) blindly as to the outcome status. The duration of work was computed and totaled for all jobs considered related to training (score The protocol was approved by the ethics committee of Sacré-Coeur Hospital and each participant signed a consent form.
Statistical Analysis Questionnaire items considered relevant for our outcomes were atopic diathesis, family history of asthma, atopic status (skin test reactivity to common allergens), bronchial responsiveness at baseline, and work-related symptoms during apprenticeship; this represented 36 parameters. The associations between these factors and the events of interest in subjects with a history of work related to training were estimated by ORs and 95% CIs adjusting for confounders (sex, age, and smoking) through multiple logistic regression analysis. We did weighted regression analyses using duration of follow-up (time between end of apprenticeship and follow-up visit) as weights to account for the increased risk of developing the outcomes associated with a longer time of follow-up. All analyses were performed with R (R Development Core team 2007, Vienna, Austria). A P value less than 0.05 was considered statistically significant.
The follow-up time after ending the apprenticeship was on average 7.56 ± 1.78 years (SD) (range, 4 –11.7 yr). Subjects who took part in the follow-up and others who did not were compared for immunological (i.e., skin reactivity to common allergens), clinical (i.e., asthma diagnosed by a physician, nasoconjunctival and chest symptoms) and functional (i.e., bronchial hyperresponsiveness) characteristics; no differences were found between the two groups except for smoking, which was more common at baseline in the participants (Table 1). Similarly, the proportions of participants who developed skin reactivity to a program-specific allergen or work-related nasoconjunctival and chest symptoms during apprenticeship were not different than among nonparticipants. A total of 315 subjects had a job related to their training, and 69 (18.0%) did not. Subjects who chose a job related to their training did not differ in terms of incidences of specific sensitization, reported work-related symptoms, or bronchial hyperresponsiveness during apprenticeship (Figures 2–5  ).
Figure 2 shows incidence and remission figures on specific skin sensitization according to the relatedness of the work history to the training program. In subjects who were in a job related to their training, 22 out of 225 developed a new specific sensitization (1.3 per 100 PY), in contrast to 9.7 and 7.3 per 100 PY during the training period, respectively, for subjects with a job not related or related to their training. Sixteen of the 22 (1.8 per 100 PY) newly sensitized subjects in a job related to their training had been in the animal health program, four (1.0 per 100 PY) had been in pastry-making, and two (0.7 per 100 PY) had been in dental hygiene. Of the 20 subjects who became sensitized during their training and had a job not related to their training, 16 (18.5 per 100 PY) lost sensitization by comparison with a lower occurrence of remission (i.e., 9.5 per 100 PY) among those who were still in the same field. Figures 3 and 4 give similar information regarding work-related rhinoconjunctival and chest symptoms. The incidence of rhinoconjunctival symptoms was much lower at the time of follow-up (1.7 per 100 PY) than during the training period (12.9 per 100 PY) in those in a job related to their training. Of the 26 incident cases after ending apprenticeship, 15 (57.7%) had been in the animal health program. The incidence of chest symptoms was lower in those in a job related to their training during the follow-up period (0.7 PY) than during the apprenticeship (1.7 PY). Of 16 incident cases, 10 trained in animal health. Figure 5 presents results related to the increase in bronchial responsiveness. The incidence of bronchial hyperresponsiveness was, respectively, 5.8 during apprenticeship and 2.0 per 100 PY during follow-up in subjects with a job related to their training. Of the 27 incident cases of bronchial hyperresponsiveness, 12 were from the animal health program, 8 were from pastry-making, and 7 were from dental hygiene. Almost all subjects (6/7) who acquired bronchial hyperresponsiveness during their training and were no longer in a job related to their apprenticeship lost this feature. There were 23 out of 277 (8.3%) new cases of probable OA during the apprenticeship and 6 out of 201 (3.0%) after ending apprenticeship in those with a job history related to training. Two subjects with probable OA on ending apprenticeship and having left the field were no longer affected; by comparison, probable OA persisted in one-third (7/21) of those who were still working in the same field. Table 2 presents clinical, immunological, and functional characteristics at baseline and acquired during apprenticeship and their association with new specific sensitization, work-related symptoms and increase of bronchial responsiveness in those with a work history related to training; the referents are subjects who did not develop the outcome. The adjusted odds ratios are presented only for characteristics significantly associated with the outcomes. In multivariate analyses, sensitization to mites and bronchial hyperreponsiveness remained significantly associated with new sensitization to a work-related allergen, physician-diagnosed asthma and bronchial hyperresponsiveness at start remained significantly associated with incidence of chest symptoms, and sensitization to pets at baseline and respiratory symptoms at the end of apprenticeship remained significantly associated with increase in bronchial hyperresponsiveness. Table 3 lists characteristics significantly and inversely associated with remission during follow-up compared with persistence of sensitization to a work-related allergen, work-related symptoms, and increases of bronchial responsiveness developed during training in subjects who had ever held a job related to training. Factors significantly and inversely associated with remission of outcomes were sensitization to pets at baseline in the case of sensitization to a work-related allergen, sensitization to pets at baseline for rhinoconjuctivitis symptoms, and baseline hyperresponsiveness for increase in bronchial hyperresponsiveness.
The number of subjects with probable OA during or after apprenticeship was not sufficient to allow for similar analyses. Of the 55 incident cases of sensitization to a work-related allergen during the apprenticeship (Figure 2), 40 (72.7%) had also developed work-related rhinoconjunctivitis symptoms during the same period that persisted at work in 16 cases, and three others (3/55 or 5.5%) developed rhinoconjunctivitis symptoms at work. Nine of these 55 subjects (16.4%) had developed respiratory symptoms during training that persisted at work in three cases, and four others (7.3%) developed respiratory symptoms after ending apprenticeship. There were 109 incident cases of work-related rhinoconjunctivitis symptoms during the apprenticeship, accompanied in 14 of these cases (12.8%) by respiratory symptoms that persisted at work in four cases (Table 4). Seven subjects (7/94 or 7.4%) developed work-related respiratory symptoms after ending apprenticeship. Seventy-one of these 109 incident cases had assessment of bronchial responsiveness (Table 5). Twenty-one out of 71 (29.6%) developed hyperresponsiveness during apprenticeship, and 12 out of 50 (24.0%) developed hyperresponsiveness while at work.
This study is the first long-term follow-up of apprentices (workers at the time of the study) exposed to high-molecular-weight agents during their training and at risk of sensitization, symptoms, and bronchial hyperresponsiveness. Relevant information on the outcome of subjects who did or did not have a work history related to training is provided. We found that the incidence of sensitization in those with a job related to training was lower than that found at the time of apprenticeship (1.3 vs. 7.3/100 PY). The same applies for the incidence of rhinoconjunctivitis, which was also lower after ending apprenticeship; a decrease in the incidence of chest symptoms and bronchial hyperresponsiveness was also seen although less clearly (overlap of 95% CIs). In the case of remission, subjects who had developed specific sensitization or showed an increase in bronchial responsiveness during apprenticeship and no longer in the same field tended to lose these features more so than those in the same field, although 95% CIs overlapped. These findings have public health implications. Surveillance of OA should be prioritized in the first years after starting exposure because the incidence of its hallmarks decreases afterward. Finally, the subjects who trained in the animal health program had the highest incidence of specific sensitization at the follow-up assessment compared with other programs, which was also the case in the training period (6). The reason for this higher frequency also found by others in workers exposed to laboratory animals is not clear, but it might be that proteins derived from animals represent more potent allergens (19).
These changes in frequency (incidence and remission) of sensitization and disease are unlikely to be due to frequently incriminated host factors such as atopy and smoking because the frequency of these factors did not change. Exposure expressed by the number of hours spent in a job related to training did not go down because those with a work history compatible with training had worked, according to our definition, at least 40 hours/week during 1 year after ending apprenticeship, which was greater than during apprenticeship. However, in our study, exposure was assessed by an indirect index; that is, the number of hours spent per week during the apprenticeship or while at work. Because we did not assess directly the environmental concentrations of the antigens, we cannot exclude the possibility that these concentrations were higher during the apprenticeship. The same rationale applies for the acquisition of outcomes in a few subjects who were classified in the "job not related to apprenticeship" category. Participants with part-time work might have been exposed to the relevant allergen but wrongly classified as non-exposed. We suspect that the time-course of sensitization and disease slows down after the first few years of exposure; this may be related to host characteristics that were not examined in the current work. It is likely that the more susceptible cases are identified in the first few years after starting exposure (i.e., during apprenticeship) and that the less susceptible cases show up afterward (i.e., during work). Another possibility could be the development of specific IgG that confers protection against symptomatology and probable disease (20). Symptomatology that does not diminish in frequency over time can be influenced by personal factors, such as over- or underreporting that can lead to biased estimates. Also, incident bronchial hyperresponsiveness can be related to the development of nonoccupational respiratory symptoms. The lower incidence of specific sensitization, symptoms, bronchial hyperresponsiveness, and probable OA may also be partially related to the healthy worker effect. This is unlikely to have affected our results because the 88 subjects who did not have a job related to their apprenticeship did not have a higher incidence of specific sensitization, symptoms, and increase in bronchial responsiveness at the end of apprenticeship than those choosing a job in the same field (Figures 2–5 It is difficult to interpret the time course of features acquired during apprenticeship in subjects who worked in an environment related to their apprenticeship compared with those who did not. Frequently, subjects who developed sensitization, symptoms, and disease during the apprenticeship no longer demonstrated these acquired features at follow-up; this was somewhat more frequent with regard to specific sensitization and bronchial responsiveness if they worked in an environment not related to their apprenticeship. In subjects with objectively diagnosed OA, it has been shown that there is generally persistence of asthma (21) and of IgE-mediated sensitization to the causal agent (22) after cessation of exposure. Although there is a diminution of IgE-mediated sensitization after stopping exposure to an occupational agent in subjects with OA (23, 24), Lemière found that nearly 75% of 16 subjects with OA to high-molecular-weight agents had elevated levels of specific IgE 2 years or more after cessation of exposure (22). In contrast, the results of our study suggest that early markers of OA may be transient and more rapidly reversible when subjects are no longer in a work environment related to their training than OA. Remission of immunological reactivity in subjects who were still in the same type of working environment is difficult to explain. Misclassification of exposure may have occurred. For example, students may have been exposed predominantly to laboratory animals during their training and only to domestic pets while at work. Job title was the question that was asked and not the type of animals participants were exposed to. However, such an apparent reversal of immunological sensitization has also been noticed in subjects sensitized to ubiquitous allergens, and the hypothesized mechanism is related to development of specific IgG and IgG4 antibodies (20). It would therefore be relevant to quantify the levels of specific IgE, IgG, and IgG4 antibodies in subjects showing one of the three profiles, namely those who acquired sensitization, those who remain sensitized, and those who apparently lose sensitization. Factors associated with the development of sensitization, symptoms, and increased bronchial responsiveness up to 11 years after first exposure at beginning of apprenticeship (see Table 2) are of the same type as those previously found to be associated with these outcomes during the apprenticeship period (6, 10). Some factors pertain to the status of immunological sensitization to ubiquitous allergens, others to general nasal and respiratory symptomatology, and others to the status of bronchial responsiveness to methacholine. Although the latter has generally not been incriminated in the likelihood of developing OA (21), the level of bronchial responsiveness at the time of entry in apprenticeship was associated with incident sensitization to a work-related allergen, new work-related chest symptoms, and increased bronchial responsiveness. Several characteristics were found to be significantly associated with remission compared with persistence of sensitization, symptoms, and bronchial hyperresponsiveness (see Table 3). There is evidence that the absence of IgE-mediated sensitization to ubiquitous allergens and of bronchial hyperresponsiveness at the time of entry into the apprenticeship was associated with the remission of work-related characteristics in those who continued to be exposed to the relevant agent(s). On the other hand, work-related symptoms that developed during apprenticeship were significant determinants of persistence of the study outcomes after entering a workplace with similar exposures. Due to the small number of subjects who developed the outcomes, the power of the multivariate analyses does not allow us to draw firm conclusions on the most significant predictors. Our main objective was to identify all significant determinants. The incidence of nasoconjunctival symptoms (1.7 per 100 PY) was greater than that of respiratory symptoms (0.7 per 100 PY). Although much neglected, rhinitis and occupational rhinitis are more frequent than asthma and OA (25, 26). Students who acquired various features (sensitization, changes in bronchial responsiveness, and symptoms) during apprenticeship did not reorient their careers. The majority of participants in the follow-up chose a job related to their apprenticeship (315/384 [82.0%]) (Figure 3). Although all these students were given information on their health status after completing the tests performed during their apprenticeship, this did not influence their decision. Orbon and coworkers found that the choice of vocational and working careers of adolescents was minimally affected by asthma symptoms or asymptomatic bronchial hyperresponsiveness (27). In a prospective cohort study of nearly 5,000 participants, Radon and coworkers confirmed these findings and showed no significant associations between baseline asthma, allergic rhinitis, or atopic dermatitis and selection of jobs with asthma risk (28). There was a tendency for teenagers with allergic rhinitis to select high-risk jobs less frequently (28). Our long-term study was performed in subjects exposed to high-molecular-weight agents. It would be useful to conduct similar long-term studies with workers exposed to low-molecular-weight agents that are potential causal agents in approximately 60% of subjects referred for possible OA (29).
The authors thank Mrs. Katherine Lieber for editing this manuscript.
Supported by the Canadian Institutes of Health Research (grant MOP-53118), the Medical Research Council of Canada (grant MT-12256), and the Institut de Recherche Robert-Sauvé en Santé et en Sécurité du Travail (grant 099-164). This article has an online supplement, which is accessible from this issue's table of contents at www.atsjournals.org Originally Published in Press as DOI: 10.1164/rccm.200707-991OC on January 24, 2008 Conflict of Interest Statement: None of the authors has a financial relationship with a commercial entity that has an interest in the subject of this manuscript. Received in original form July 6, 2007; accepted in final form January 24, 2008
Related articles in AJRCCM:
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
AT A GLANCE COMMENTARY
1). Only workers who had worked at least 40 h/wk in such a job for the equivalent of 1 year or more including part-time work were considered as having a job related to training. 
16 mg/ml in subjects with initial PC20 >32 mg/ml).
