|
|
||||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Clinic for Pneumology (Drs. Miedinger, Chhajed, Tamm, Stolz, and Leuppi) and Hospital Pharmacy (Dr. Surber), University Hospital, Basel, Switzerland.
Correspondence to: David Miedinger, MD, Department of Internal Medicine, University Hospital, Petersgraben 4, 4031 Basel, Switzerland; e-mail: miedingerd{at}uhbs.ch
Abstract
Background: Subjects with asthma do not meet medical requirements for professions such as firefighting.
Objective: To prospectively determine the diagnostic value of respiratory symptoms and various tests used in the assessment of asthma in a cohort of firefighters.
Methods: A questionnaire, spirometry, direct and indirect airway challenge tests, exhaled nitric oxide, and skin-prick tests were administered prospectively to 101 of 107 firefighters employed in Basel, Switzerland. Asthma was defined as the combination of respiratory symptoms with airway hyperresponsiveness.
Results: Six of 101 firefighters (6%) had physician-diagnosed asthma, which could be confirmed in 4 firefighters. In contrast, asthma was diagnosed in 14% (14 of 101 firefighters). Wheezing was the most sensitive symptom for the diagnosis of asthma (sensitivity, 78%; specificity, 93%). Other respiratory symptoms showed a higher specificity than wheezing but a markedly lower sensitivity. Bronchial airway challenge with mannitol was the most sensitive (92%) and specific (97%) diagnostic test for asthma. Using a cutoff point of 47 parts per billion, nitric oxide had a similar specificity (96%) but lower sensitivity (42%) compared to the direct (methacholine) and indirect (mannitol) airway challenge tests.
Conclusion: Asthma was considerably underdiagnosed in firefighters. The combination of a structured symptom questionnaire with a bronchial challenge test allows to identify patients with asthma and should routinely be used in the assessment of active firefighters and may be of help when evaluating candidates for this profession.
Key Words: asthma test exhaled nitric oxide firefighters mannitol methacholine skin prick test
Asthma is a chronic inflammatory disorder of the airways that is associated with inflammation, airway obstruction, and airway hyperresponsiveness (AHR). The diagnosis of asthma is based on clinical symptoms such as wheezing, breathlessness, chest tightness, and cough.12 Reversible airway obstruction on spirometry3 or variability in daily peak expiratory flow4 are also accepted methods for the diagnosis of asthma. Epidemiologists often use wheezing within the last 12 months combined with positive response to a bronchial challenge test (BCT) to define current asthma.567 However a considerable number of patients with asthma may not have wheezing within the last 12 months.2 Measurement of fractional exhaled nitric oxide (FENO) can also be used to identify subjects with atopic asthma.89
Subjects with asthma do not meet medical requirements for professions such as firefighting.101112 An exacerbation of asthma during the execution of duty may put the subject as well as coworkers in potential danger. The definition of self-reported physician-diagnosed asthma used in epidemiologic studies213 relies on a positive answer to the questions, "Do you have asthma?" and "Has this been confirmed by a doctor?" This has the potential to underestimate the diagnosis of asthma. Guidelines for diagnosing asthma in firefighters do not incorporate different available tests for routine assessment unless specifically requested by the examining physician.101112 It is known that there may be underreporting of symptoms at job recruitment to avoid exclusion from the job.1415
In professions such as firefighting, it is important to have a highly sensitive objective tool to diagnose asthma. This study prospectively determined the diagnostic value of different respiratory symptoms, spirometry, direct and indirect BCT, and FENO in the assessment of asthma in a cohort of firefighters.
Materials and Methods
Study Subjects
Basel is an industrial city in the northern part of Switzerland. The local municipal fire department employs 107 full-time firefighters. We excluded the records of the only female firefighter from the analysis. Firefighters were involved in all types of work when responding to an incident, so they were all exposed to similar conditions during work. Our study was approved by the local ethics committee. All subjects gave written informed consent.
Questionnaire
All participants answered a self-administered questionnaire with items of the Swiss Study on Air Pollution and Lung Diseases in Adults (SAPALDIA) I questionnaire,13 an extended version of the questionnaire used in the European Community Respiratory Health Survey (ECRHS).16 The questionnaire contained items about respiratory symptoms, living conditions, active and passive smoking, and occupational and leisure exposure to air pollutants.
Lung Function
Spirometry was performed using American Thoracic Society criteria.17 A spirometer (EasyOne; ndd Medizintechnic; Zurich, Switzerland) was used to measure FVC and FEV1. The higher of two values for FEV1 repeatable to within 100 mL were recorded, and the percentage of predicted values18 was calculated.
Methacholine BCT
Methacholine testing was done according to the protocol of the SAPALDIA I,13 which is identical with the protocol of ECRHS "method two short protocol."1619 The maximum cumulative dose of methacholine administered in this protocol is 2 mg. Subjects who reported a respiratory tract infection in the previous 3 weeks were rescheduled. All those with a prediluent FEV1/FVC ratio
0.6 were invited to undergo methacholine challenge. Methacholine was delivered at room temperature using a dosimeter (Mefar MB3; Mefar; Bovezzo, Italy) set to deliver the aerosol over a period of 1 s. All solutions of methacholine to test the firefighters were prepared by the hospital pharmacy at University Hospital, Basel, Switzerland.
Mannitol BCT
Challenge with mannitol was performed using a protocol described by Anderson et al.20 Mannitol is administered as a dry powder in capsule form inhaled from a dry powder inhaler device. All those with a prediluent FEV1/FVC ratio
0.6 were invited to undergo mannitol challenge. An empty capsule was used as a control at baseline. The FEV1 value measured after the empty capsule was used to calculate the percentage decrease in FEV1 in response to the mannitol challenge. If the FEV1 fell by 10% based on the FEV1, the dose producing this fall was repeated. The challenge was stopped if the FEV1 fell by
15%, or when the maximum dose had been administered.
FENO Measurement
FENO was measured according to ATS guidelines21 by using a nitric oxide analyzer (NIOX; Aerocrine AB; Solna, Sweden). The subject exhaled to residual volume, took a deep breath over 2 to 3 s through the mouthpiece to total lung capacity, and exhaled immediately through the mouthpiece over 10 s against an oral pressure from 5 to 20 cm H2O, maintaining a flow from 0.045 to 0.055 L/s. This was achieved by computed biofeedback software.
Allergic Sensitization
Sensitization to allergens was measured by reactions to skin-prick tests on the forearm according to the protocol of the SAPALDIA.13 Nine different allergens or allergen mixtures were tested: mixture of six grass, mixture of three trees, the molds Alternaria alternate and Cladosporium herbarum, cat and dog epithelium, and the house dust mites Dermatophagoides pteronyssinus and Dermatophagoides farinae (ALK-Abello; Round Rock, TX). Histamine was used as the positive control and a saline/glycerol solution as the negative control. Atopy was defined as a positive response to at least one of the allergens tested.
Asthma Definition
We defined asthma as respiratory symptoms and a history of wheezing not restricted to the last 12 months combined with any measurement of AHR (provocative dose of methacholine causing a 20% fall in FEV1 [PD20] and/or provocative dose of mannitol causing a 15% fall in FEV1 [PD15]).
Statistical Analysis
Continuous variables are expressed as mean ± SD or as geometric means with interquartile range, and categorical variables are expressed as relative frequencies and percentages. A receiver operating characteristic (ROC) curve was plotted that allowed a graphical representation of sensitivity and specificity in order to view the best cutoff level for diagnosis of asthma. All tests were performed using software (SPSS Version 12; SPSS; Chicago, IL; and Excel; Microsoft; Redmond, WA).
Results
One hundred one of 106 male firefighters (95%) gave informed consent for this study. They were employed for a mean of 16 years (range, 2 to 36 years). Thirty-three of the 101 firefighters (33%) were current smokers (mean, 23 pack-years). Baseline characteristics can be seen in Table 1 .
|
|
Two firefighters were classified as not having asthma. One firefighter was completely symptom free, and results of all administered tests to detect asthma were normal. The other firefighter complained of wheezing in the last 12 months and occasional nocturnal chest tightness. He was receiving asthma medication, had normal spirometry results, and had no AHR.
Fifteen firefighters (15%) had a positive test response to methacholine, denoted by a drop > 20% in FEV1 from baseline up to a cumulative dose of 2 mg of methacholine. Fourteen firefighters (14%) had positive test results to mannitol, denoted by a drop > 15% in FEV1 from baseline up to a cumulative dose of 635 mg of mannitol. In two firefighters, the mannitol BCT could not be accomplished due to cough before the total test dose of 635 mg of mannitol could be administered in one case, and due to technical problems in one case.
The association between respiratory symptoms and asthma is shown on Table 3 . The association of different tests for the diagnosis of asthma is shown on Table 4 . A sensitivity analysis for the BCT with methacholine and FENO measurement in diagnosing alternative definitions of asthma can be found on Table 5 . The proportion of positive challenge test result or diagnosis of asthma tended to be higher in subjects complaining respiratory symptoms in the workplace compared to those without respiratory symptoms: positive methacholine BCT result (17% vs 15%, p > 0.05), positive mannitol BCT result (21% vs 12%, p > 0.05), and asthma (17% vs 12%, p > 0.05). The ROC curve for the FENO measurements is shown on Figure 1 . The skin-prick test result was positive in 12 of 14 patients with asthma.
|
|
|
|
Our study shows that asthma is considerably underdiagnosed in firefighters, and the best approach for a diagnosis is the combination of symptoms with a positive test result to mannitol challenge. Swiss firefighters undergo a medical screening program including spirometry prior to employment and also regular medical examination when in service to confirm the status "fit for duty."10 Physician-diagnosed asthma is often used to assign the diagnosis of asthma to a subject in an epidemiologic study setting13 or in a screening questionnaire at conscription.10 This might not reflect the final diagnosis of asthma as has been shown in studies from the British and American armies.1415 In the British army, Sinclair and coworkers15 found exercise-induced asthma in 29% recruits who had a history of probable asthma during childhood but had no symptoms and treatment during the last 4 years. Nish and Schweitz14 examined 192 recruits who failed a preenlistment exercise training test and had denied having asthma since the age of 12 years. Fifty-nine percent of these recruits had asthma diagnosed subsequently based on further investigations.14 In our study, only 6% of firefighters reported a diagnosis of asthma; in two of them, the diagnosis could not be confirmed. However, 10 firefighters had asthma diagnosed based on symptoms and positive BCT results. In one case, it could be argued that the diagnosis was COPD rather than asthma or a combination of both diseases. The mean age of the firefighters as well as duration in the service were similar to literature reports.222324 The diagnosis of asthma is important from the treatment perspective and also may affect work performance. It has been shown that firefighters who where fighting a fire had an increase in AHR when measured within 24 h after exposure.25 Furthermore, legal aspects have to be considered when firefighters with a diagnosis of asthma are exposed to risks.
Symptom perception leading to the diagnosis of asthma may be limited in asthmatics26 as well as physicians.2728 There is an ongoing debate about how to define the "gold standard" for the diagnosis of asthma. Some authors2 have used the definition of doctor-diagnosed asthma, although this definition has been criticized because there had been a wide variation reported in the criteria for making a diagnosis.29 Studies designed to identify asthmatics often yield more information (eg, symptoms, lung function measurements, and results of bronchial hyperresponsiveness) than general practitioners have when they assign the diagnosis of asthma to a subject.29 Once the diagnosis is made, it has to be recalled by the subject when asked in a questionnaire, and it is possible that subjects underreport a diagnosis of asthma because they are afraid that the diagnosis would declare them unfit for service and they would probably have to change their profession. Measuring AHR with BCT has been shown to be reliable and is not influenced by symptom perception and reporting.30 AHR is a fundamental characteristic of asthma.129 The findings of our study support the use of regular screening with BCTs in such a profession.
In clinical practice, the diagnosis of asthma often relies on a positive history of wheeze and other symptoms consistent with asthma such as cough and dyspnea. However, in a study by Baumann and coworkers,31 wheeze, cough, and dyspnea only occurred simultaneously in 36% of asthmatics. Burrows and coworkers32 identified wheezing as the most frequent finding in a longitudinal study with patients with a recent diagnosis of asthma. Wheezing in the last 12 months was the most efficient symptom for the diagnosis of asthma (Table 3).
Patients with asthma may have a normal lung function, and a BCT is needed to identify patients with asthma.3334 In direct tests (eg, methacholine or histamine), the agonist acts directly on specific receptors on the bronchial smooth muscle, causing a contraction. The indirect tests (eg, exercise eucapnic voluntary hyperpnea, hypertonic saline solution, adenosine 5'-monophosphate, or mannitol) act via osmotic stimuli on inflammatory cells that then release mediators like histamine, leukotrienes, and prostaglandins that act on smooth-muscle cells.35 In asthmatics, responsiveness to an indirect BCT is better reflected with indexes of airway inflammation such as sputum eosinophils or FENO than a direct BCT with methacholine.36 In the current study, mannitol BCT had a higher efficiency and Youden index compared to methacholine BCT. Indirect BCTs are more specific for the diagnosis of asthma373839 although less sensitive to identify AHR in a laboratory population.40 Using direct BCT in a young and healthy population often produces false-positive responses in subjects without symptoms of asthma.740414243 Exercise-induced bronchoconstriction can be missed with direct BCTs such as histamine or methacholine4445 but may be better diagnosed using indirect BCT.464748 In our firefighters with a diagnosis of asthma, three patients had negative responses to methacholine BCT but showed a positive mannitol BCT response. We have used the protocol previously used in a large population study, the ECRHS.19 There are other protocols for methacholine challenge testing, and in some protocols the total test dose is higher.4749 It is possible that the three asthmatic firefighters with negative test results in our study would have had a fall of 20% in FEV1 when challenged with a higher total dose, but this would have yielded a higher rate of firefighters with a positive test results without respiratory symptoms. Additionally, there seem to be more positive BCT responses especially to mannitol in the group of firefighters complaining of respiratory symptoms at the workplace. Thirteen of 14 firefighters had a positive BCT response to mannitol, and only 11 of 14 subjects had a positive BCT response to methacholine. We suggest that in the presence of respiratory symptoms and normal spirometry findings, a BCT should be performed. It appears that a combination of BCT to mannitol and symptoms is superior to BCT to methacholine and symptoms to identify subjects with asthma, although due to the small numbers of asthmatics in our study group we cannot postulate BCT with mannitol to be the new "gold standard" for the diagnosis of asthma in firefighters.
When working with a self-containing breathing apparatus or when scuba diving, the workload is increased leading to an increase in minute ventilation. To avoid corrosion of the metal parts of the breathing apparatus, air is dehumified before it is compressed and filled in the bottle. These apparatus use compressed air that cools down when used, is decompressed, and is released from the respirator (Boyles law). As dry cold air is a typical stimulus that is used in indirect BCTs, there is a possibility that bronchoconstriction can occur in subjects with a positive indirect BCT result even if that subject is otherwise not complaining of respiratory symptoms that are hallmarks of asthma.5051
In the past years, the measurement of nitric oxide has been proposed as a noninvasive test that can reliably detect subjects with asthma.89 In our population, FENO showed a similar specificity as the direct and indirect BCT but a relatively low sensitivity when using a cutoff value of 47 parts per billion (ppb). A strong association has been reported between elevated FENO and nasal nitric oxide and skin-prick test scores, total IgE, and blood eosinophilia in mild asthmatics.5253 In a study54 investigating military conscripts, exercise-induced bronchoconstriction and elevated FENO were only associated in atopic individuals. In our study, only 2 of 14 patients with asthma had a negative skin-prick test result. However, the skin-prick test was obviously not specific for the diagnosis of asthma.
Although our study population was representative, including 95% of all firefighters of the municipal firefighting department, it can be argued that the group size is relatively small. Further studies are needed to confirm our findings in firefighters in other countries and in firefighters mainly engaged in hazardous material or wild land fire operations, and when firefighters undergo the different preemployment screening and regular check procedures.
We did not analyze data on work performance and medical leaves in the firefighters; therefore, we cannot determine the impact of asthma or a positive BCT response on these variables. It can be argued that although having asthma, these asthmatic firefighters are still in the active workforce and are therefore not suffering from a clinically relevant asthma. Further studies are needed to evaluate the influence of asthma on the work performance of firefighters. No doubt, findings will be based not only on host responses but on the types and intensity of fire and smoke exposures. Fire departments are not allowed to perform challenge tests in asymptomatic candidates who have a negative respiratory history, due to the discriminatory impact of false-positive test results. However, all candidates should undergo challenge testing if symptoms or history suggest asthma, even when baseline spirometry findings are normal.
In summary, asthma was considerably underdiagnosed in firefighters. The combination of a structured symptom questionnaire with the mannitol BCT allows to identify patients with asthma. Whether subjects without symptoms but with a positive mannitol challenge response have to be excluded from firefighting or should be treated with inhaled steroids has to be determined.
Footnotes
Abbreviations: AHR = airway hyperresponsiveness; BCT = bronchial challenge test; ECRHS = European Community Respiratory Health Survey; FENO = fractional exhaled nitric oxide; PD15 = provocative dose of mannitol causing a 15% fall in FEV1; PD20 = provocative dose of methacholine causing a 20% fall in FEV1; ppb = parts per billion; ROC = receiver operating characteristic; SAPALDIA = Swiss Study on Air Pollution and Lung Diseases in Adults
The study was supported by a grant from Boehringer Ingelheim (Switzerland) GmbH, Basel.
The authors have no conflicts of interest to disclose.
Received for publication September 13, 2006. Accepted for publication February 22, 2007.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |