RR540 The HSE Grain Dust Study · Health and Safety Executive The HSE Grain Dust Study workers'...

Health and Safety Executive

The HSE Grain Dust Study workers'exposure to grain dust contaminants,immunological and clinical response

Prepared by the Health and Safety Laboratory for the Health and Safety Executive 2007

RR540 Research Report



J R M Swan, D Blainey & B Crook Health and Safety Laboratory Harpur Hill Buxton Derbyshire SK17 9JN

Inhalation of airborne microorganisms and their associated contaminants in the workplace can cause a range of immunological and respiratory symptoms. The mechanisms through which these respiratory effects are caused are not all fully understood. The evaluation of worker exposure is essential for establishing causal relationships between occupational disease and one or several specific microorganisms or their associated contaminants.

This study investigated the role of microorganisms and their associated contaminants in the development of immunological and clinical response in workers exposed to grain dust. The objectives were:

■ To assess the exposure of grain workers in the UK to inhalable grain dust, the microbial contaminants in grain dust, including identification of the predominant microorganisms involved, and to endotoxin (bacterial cell wall toxins).

■ To measure the prevalence of immunological response to grain dust associated allergens in UK grain workers.

■ To examine the long term clinical and immunological effects of workplace exposure to grain dust and its contaminants in terms of its effect on respiratory health. This was done by establishing a cohort of 321 workers exposed to grain dust (farmers at 27 farms and dock workers at 2 docks in South East England) and maintaining as many as possible in the cohort for repeated immunoassay and clinical assessment over two study phases, Phase 1 from 1990 to 1993 and Phase 2 from 1997 to 2003.

This report and the work it describes were funded by the Health and Safety Executive (HSE). Its contents, including any opinions and/or conclusions expressed, are those of the authors alone and do not necessarily reflect HSE policy.

HSE Books

© Crown copyright 2007

First published 2007

All rights reserved. No part of this publication may bereproduced, stored in a retrieval system, or transmitted inany form or by any means (electronic, mechanical,photocopying, recording or otherwise) without the priorwritten permission of the copyright owner.

Applications for reproduction should be made in writing to:Licensing Division, Her Majesty’s Stationery Office,St Clements House, 216 Colegate, Norwich NR3 1BQor by email to hmsolicensing@cabinetoffice.x.gsi.gov.uk

ii

CONTENTS

1 INTRODUCTION......................................................................................... 11.1 Organic dust, grain dust and respiratory disease .................................... 11.2 Grain Dust ............................................................................................... 11.3 The role of atopy in occupational allergy ................................................. 21.4 Prevalence of allergic respiratory symptoms in UK workers .................... 21.5 The role of endotoxin in respiratory disease ............................................ 31.6 The UK maximum exposure limit (MEL) for grain dust ............................ 51.7 The Health and Safety Executive grain dust survey: aims....................... 5

2 MICROBIOLOGICAL EXPOSURE STUDY................................................ 72.1 Background ............................................................................................. 72.2 Study sites, Materials and Methods ......................................................... 72.3 Results summary and Discussion.......................................................... 112.4 Summary of microbiological exposure study ......................................... 232.5 Conclusions ........................................................................................... 30

3 IMMUNOLOGY STUDY ............................................................................ 313.1 Aim ........................................................................................................ 313.2 Study Population ................................................................................... 313.3 Immunoglobulin G immunoassay ......................................................... 323.4 Immunoglobulin E immunoassay .......................................................... 333.5 Results................................................................................................... 373.6 Discussion; immunology results ............................................................ 45

4 CLINICAL STUDY .................................................................................... 514.1 Introduction and Summary of Phase 1................................................... 514.2 Subjects................................................................................................. 514.3 Methods................................................................................................. 534.4 Results................................................................................................... 554.5 Summary of findings – Clinical study .................................................... 70

5 GENERAL DISCUSSION ......................................................................... 72

6 QUESTIONS ADDRESSED BY THE STUDY .......................................... 746.1 Has the MEL reduced sensitisation? ..................................................... 746.2 Did anyone develop new IgG or IgE during the study?.......................... 746.3 Should the MEL be reduced to reduce sensitisation?............................ 746.4 Does new sensitisation result in new symptoms?.................................. 75

7 OVERALL SUMMARY.............................................................................. 76

8 MAIN RECOMMENDATIONS................................................................... 78

9 ANNEX 1................................................................................................... 80

10 REFERENCES ...................................................................................... 90

iii

iv

EXECUTIVE SUMMARY

Objectives

Inhalation of airborne micro-organisms and their associated contaminants in the workplace can cause a range of immunological and respiratory symptoms. The mechanisms through which these respiratory effects are caused are not all fully understood. The evaluation of worker exposure is essential for establishing causal relationships between occupational disease and one or several specific micro-organisms or their associated contaminants.

This study investigated the role of micro-organisms and their associated contaminants in the development of immunological and clinical response in workers exposed to grain dust. The objectives were:

1. To assess the exposure of grain workers in the UK to inhalable grain dust, the microbial contaminants in grain dust, including identification of the predominant micro-organisms involved, and to endotoxin (bacterial cell wall toxins).

2. To measure the prevalence of immunological response to grain dust associated allergens in UK grain workers.

3. To examine the long term clinical and immunological effects of workplace exposure to grain dust and its contaminants in terms of its effect on respiratory health.

This was done by establishing a cohort of 321 workers exposed to grain dust (farmers at 27 farms and dock workers at 2 docks in South East England) and maintaining as many as possible in the cohort for repeated immunoassay and clinical assessment over two study phases, Phase 1 from 1990 to 1993 and Phase 2 from 1997 to 2003.

Main Findings

UK grain workers were frequently found to be exposed to more than 1 million bacteria and fungi per m3 air. Airborne bacteria exceeded 106 per m3 air and fungi exceeded 105 per m3 air at all the work places sampled. Levels of airborne endotoxin of over 10,000 EU/m3 were recorded at all but one workplace visited and personal exposures reached over 600 EU/m3 at every workplace. The Maximum Exposure Limit (MEL) for grain dust is an average of 10 mg/m3 of

The Maximum Exposure Limit (MEL) for grain dust is an average of 10 mg/m3 of total respirable dust in the air over an 8 hour period, and the maximum dust level should never exceed 30 mg/m3 measured over a 10 minute period. The MEL was introduced between Phases 1 and 2 of the study. Throughout both phases of the study, very high levels of dust were recorded with levels equal to, or greater than, 30 mg/m3 recorded in all work situations at 17 of the 19 workplaces visited.

The predominant micro-organisms present differed between freshly harvested grain and stored grain and grain for human and animal consumption, but not between different types of grain. The microbial species identified were used to select relevant isolates for use in immunoassays.

In the immunoassays, large numbers of sera were expected to show positive immunoglobulin G (IgG) response as the prevalent micro-organisms are also common environmental allergens to

v

which everyone is exposed to some extent. Between 13 and 39% of dockers, 25 to 41% of farmers and 24 to 35 % of the controls tested positive to one or more of the work-related allergens throughout the study. IgE sensitisation to work related allergens among the grain worker cohort showed evidence of a decrease since the introduction of the grain dust MEL, i.e., there was an overall trend towards a decrease in IgE positive responses across the study from Phase 1 to Phase 2. However, the long term trend towards improvement may not be shown clearly in this study as the majority of the participants were occupationally exposed to grain dust for many years before the MEL was introduced.

Although some workers who remained in the study throughout developed new IgG response or positive IgE response during the study, more workers lost their positive status than gained it. No dockworkers developed a sustained new positive status. There was a decrease in the number of non atopic workers who tested IgE positive to the work-related allergens, and none of the non-atopic dockworkers were IgE positive in Phase 2 of study. However, 11% of non atopic

However, 11% of non atopic farmers were still testing positive to one or more of the work related allergens at the end of the study. This suggests that it is possible to reduce exposure levels to below those required to trigger an IgE response in non-atopic workers, but that farm workers continue to be exposed to levels of grain dust and its contaminants that could trigger or maintain sensitisation.

In the two phases of the clinical study, grain workers were found to have a high prevalence of respiratory symptoms and sensitisation to environmental allergens including microorganisms, the grain itself and storage mites.

There was a close relationship between some respiratory symptoms (especially ‘wheeze in a dusty place’) and the presence of specific IgE to environmental allergens. There was no association between specific IgG and symptoms or lung function. Not all respiratory symptoms could be explained by IgE mediated immunological sensitisation. The longitudinal exposure of grain workers to high doses of endotoxin may be contributing to the development of work related ill health.

Lung function was found to be inversely related to the duration of exposure to grain dust in the first phase of the study, but the cause(s) of impaired lung function could not be identified from this phase. In the second phase of the study this relationship was not identified, and over the course of the study, both in those who took part in both phases, or those who just participated throughout Phase 2, there was no overall decline in lung function.

Main Recommendations

Although fungal allergens are recognised components of grain dust, this study has shown that not all grain workers’ impaired lung function and respiratory symptoms could be linked with immunological response. The study also highlighted the exposure of grain workers to large concentrations of endotoxin, which may contribute to respiratory symptoms but not immunological response. Guidance/information on endotoxin exposure in the workplace is recommended. The longitudinal health effects of high endotoxin exposure, and the adjuvant effects of endotoxin inhalation with other contaminants, require further investigation.

In many instances in grain handling, the use of respiratory protective equipment (RPE) is the only control option. This study highlighted that exposure to large numbers of micro-organisms and their toxins may occur in apparently ‘clean’ areas, where currently RPE is not worn. Improved cleanliness is recommended in rest areas, offices and vehicle cabs. Vehicle cabs may

vi

not be offering enough protection and the use of. RPE inside the cabs should be considered. The levels of IgE response in non-atopic farmers indicates that farmers continue to be exposed to high levels of grain dust work-related allergens. Observations made during the microbiological sampling indicated that better protection factor RPE, worn more frequently during specific high dust exposure tasks, would help reduce the levels of exposure, particularly during harvesting in areas/situations where it is not practicable to reduce dust exposure in other ways. It would be important to ensure that such RPE offered suitable levels of protection against endotoxin exposure.

vii

viii

1 INTRODUCTION

1.1 ORGANIC DUST, GRAIN DUST AND RESPIRATORY DISEASE

Ramazzini published work on the subject of occupational illness caused by grain dust over two centuries ago. This paper ‘Diseases of sifters and measurers of grain’ was published in 1713. He noted that the dust was made up ‘not only of the dust picked up from the threshing floor, but also another less innocent sort which is shed from the grain itself which is kept for long’. He described some of the symptoms - ‘the throat, lungs and eyes are keenly aware of serious damage’, ‘intense itching over the whole body of the sort sometimes observed in nettle rash’, ‘almost all who make a living by sifting or measuring grain are short of breath, cachetic and rarely reach old age’ (Ramazzini, 1713; cited in Becklake, 1980).

There is no clear understanding of the variety of mechanisms by which the problems associated with grain dust occur, though some have been identified. Organic dust from the grain itself can cause respiratory irritant effects and an immunological response. In addition to this, other biological materials present including fungi, actinomycetes, bacteria and their endotoxins, pollen, insects and arachnid mites and their debris are present in harvested and stored grain and may have effects on the respiratory system. The evaluation of worker exposure to microorganisms in the workplace in general, and of grain workers in particular, is essential for establishing causal relationships between occupational disease and one or several specific microorganisms or their associated contaminants. The number of known allergens is continually rising and the list of microorganisms that can cause occupational illness is increasing.

1.2 GRAIN DUST

Grain dust is the dust produced during harvesting, drying, handling, storage or processing of wheat, oats, barley, maize or rye. It is a complex mixture of fragments of grain, inorganic soil particles and associated organic contaminants. These contaminants may include plant cell debris, insect parts and mites as well as viable and non-viable microorganisms (vegetative cells and spores of fungi, actinomycetes and bacteria, and their components such as endotoxins and mycotoxins). This complicated mixture makes the effects of the grain dust on the lungs difficult to define, (Burrell, 1995; Lacey, 1990; Rylander, 1994). When grain is handled, clouds of this complex dust mixture are dispersed in the air. Inhalation of these dusts can lead to decreased lung function and the development of immunological respiratory symptoms which may include allergic rhinitis and asthma, chronic bronchitis, granulomatous pneumonitis (extrinsic allergic alveolitis, hypersensitivity pneumonitis), toxic pneumonitis (organic dust toxic syndrome/grain fever) and decline in lung function. The mechanisms by which these occur are not yet well understood (Becklake, 1980; Dimich-Ward et al., 1995; Fonn et al., 1993a; Fonn et al., 1993b; Hurst and Dosman, 1990; Mcduffie et al., 1991; Rylander, 1994).

The role of microbial exposure, the immunological effects and the relative importance of these grain dust contaminants compared to grain itself have not been established by previous research. Many of the microorganisms found in grain dust both during harvesting and after storage are known respiratory sensitisers e.g. Cladosporium, Alternaria, Aspergillus spp., Penicillium spp. which are well known as allergens (Darke et al., 1976; Dutkiewicz et al., 1985; Dutkiewicz et al., 1989; Lacey 1995; Marx et al., 1993) while Enterobacter agglomerans may also be a source of endotoxin (Dutkiewicz, 1976). It

1

is hard to define the precise effects of grain dust on the lungs because of the diversity of worker exposure and the range and diversity of symptoms involving different pathogenic mechanisms.

1.3 THE ROLE OF ATOPY IN OCCUPATIONAL ALLERGY

About 20% of the population are atopic. Atopic people have a hereditary predisposition to produce IgE antibodies and respond to tiny amounts of exposure to common environmental allergens with the persistent production of specific IgE antibodies. To determine atopic status, a subject is tested with a series of allergens. A positive test is an indicator of predisposition to further sensitisation to other allergens. Atopy is the single strongest risk factor for the development of asthma. Atopy can increase the risk of developing asthma by 10-20 fold when compared with those who are non-atopic (Holgate, 1997).

Other factors that affect the development of allergic disease are those of infection and toxicity. Infection can stimulate the immune response to adjuvant action or inflammation. Toxic factors such as tobacco smoke, air pollution and industrial exposures may enhance absorption of antigens as a result of pulmonary inhalation and inflammation. Alternatively, the toxic factors may act as adjuvants, it has been suggested that endotoxin is involved as an adjuvant.

1.4 PREVALENCE OF ALLERGIC RESPIRATORY SYMPTOMS IN UK WORKERS

The overall frequency with which occupational asthma occurs in the UK is unknown but it has been estimated that 2-6% of all the cases of adult asthma may be due to workplace exposures (HSE, 1998). In 1995 a survey of self reported work-related ill health was carried out. Information on self-reported occupational health data was collected from a representative national sample of adults in England and Wales. From these data it was estimated that about 138,000 people believe they are affected by work-related asthma. 62,000 of these people felt that their asthma was caused by, rather than made worse by, their work (HSE, 1998).

The ‘Surveillance of work-related and occupational respiratory disease in the UK’ - SWORD 99 - is a surveillance scheme of chest physicians reporting cases to produce a national picture of occupational respiratory disease (Meyer, 2001). In 1999 there were 1168 cases of occupational asthma, 43 of allergic alveolitis (13 cases of which were farmers lung) and 129 bronchitis reported (HSE,2001). There were 9 deaths from Farmers Lung in 1999 (HSE, 2001). Ross (1995) reported that follow-up studies showed that most patients with occupational asthma failed to recover and half left their employer. Patients who were exposed continually for a year or more after diagnosis had a worse outcome than those removed rapidly from exposure. In 1989 SWORD reported that 42% of cases of occupational asthma and 7% of allergic alveolitis were due to flour or grain dust. In 1999 8% of cases of asthma caused by sensitisation were attributed to flour and grain. In 2000 an estimated 797 new cases of occupational asthma was seen for the first time by occupational and chest physicians reporting to SWORD/OPRA schemes (HSE, 2001). These figures probably underestimate the problem as not all cases of work-related respiratory symptoms are seen by a doctor, and an occupational aetiology may not be considered in all cases. It is currently estimated by the Health and Safety Executive (HSE) that 1000 - 1500 new cases of occupational asthma occur every year in the UK (HSE, 1998).

As well as the very high personal cost of respiratory disease, these figures represent an increasing financial cost to industry not only in loss of manpower, but also compensation. In 1996, 410 people with occupational asthma were assessed as being disabled in order to qualify for benefit purposes (HSE, 1998).

2

1.5 THE ROLE OF ENDOTOXIN IN RESPIRATORY DISEASE

Endotoxin is found in the outer layer of the cell walls of all Gram-negative bacteria and some blue green algae. Occupational exposure to airborne endotoxin causes short-term illness, and may contribute to long-term worker ill-health. However at present there are no occupational exposure limits in place for endotoxin. A great number of people may be exposed to airborne endotoxin because of the wide range of workplace sources. The general population is exposed to low levels of endotoxin. Gram-negative bacteria are present in the oral cavities and intestinal tracts of humans and animals; they also live on the surfaces of animals and plants. Endotoxin has also been found in house dust. Endotoxin is present in occupational settings, mainly as a component of organic dusts such as those of vegetable origin contaminated with Gram-negative bacteria, grain dust, cotton dust, and dusts containing animal faeces in swine confinement buildings, poultry houses and sewage sludge. Air conditioning units, contaminated process waters or water-mix oil emulsions can also contain endotoxin.

The short term effects of endotoxin inhalation include a febrile response with leucocytosis and flu-like symptoms which resolve within 18 hours after initial intense exposure e.g. organic dust toxic syndrome (ODTS) or inhalation fever. No prior sensitisation is needed, antibodies do not develop and respiratory symptoms may or may not occur (Chan-Yeung et. al. 1992). Chronic effects occur following repeated occupational exposure to endotoxin e.g. Chronic bronchitis and chronic obstructive pulmonary disease (Clapp et al.1994; Olenchock et al. 1990). Long term decline in lung function has also been measured (Schwartz 1996).

Exposure to endotoxin in occupational environments and the resultant health consequences have been widely investigated. Several studies have investigated endotoxin exposure associated with grain dust and the subsequent health effects (Jagielo et al. 1996; Clapp et al. 1993 and 1994; Schwartz, 1996; Schwartz et al. 1995a; Smid et al. 1992 a and b; Blaski et al. 1996). These studies were reviewed in Swan and Crook (1999). These studies highlight the difficulties involved in determining which health effects resulting from the inhalation of organic dust mixtures are caused by the endotoxin component of the dust. The wide range of epidemiological studies in the review suggest that chronic exposure to endotoxin present in occupational organic dusts may cause chronic inflammation leading to chronic bronchitis and reduced lung function, however, more data is required on the longitudinal health effects of endotoxin exposure.

The most common route of occupational exposure to endotoxin is through inhalation of airborne endotoxin. The amount of airborne endotoxin in different occupational environments varies widely Some typical examples are summarised in Table 1.1 and exposure to endotoxin in grain dust is in Table 1.2. There have been few UK studies of occupational endotoxin exposure.

Although such investigation is outside the remit of this study, there is increasing recognition of the potential role of endotoxin as a synergistic component to increase respiratory ill health effects. For example, animal studies have demonstrated the potential for the combination of diesel exhaust particles and endotoxin (lipopolysaccharide) to aggravate endotoxin induced lung injury, as demonstrated for example by pro-inflammatory cytokines (Yanagisawa et al, 2003; Arimoto et al, 2005; Inoue et al, 2006). Other animal studies have shown the potential for exposure to a commonly used organothiophosphate pesticide to alter inflammatory response to endotoxin (Singh and Jiang, 2003), and a synergistic combination of the pesticide rotenone and endotoxin exposure has been hypothesised as being an acceleratory factor in neural degeneration (Gao et al, 2003).

3

Table 1.1. Summary of occupational exposure to airborne endotoxin.

Industry/activity airborne endotoxin concentration

(ng/m3) Agriculture Grain handling 0 - 54,900 Animal feed 0.1 - 1,870 Swine houses 31 - 75,000 Poultry farm 3.8 - 1,500 Vegetable processing 1 - 4,600 Industry Cotton industry 4 - 2,000 Biotechnology 0.07 - 162.85 Machining MWF 0.007 - 2,690 Fibreglass wash water 0.4 - 27,800 Waste processing sewage processing 0.08 - 310 garbage handling 0.02 - 131 composting 0.008 - 5,930 Offices Offices 0.018 - 1,200

Table 1.2 Endotoxin levels measured in grain dust Industry/activity mean airborne

endotoxin concentration

(ng/m3)

endotoxin level in bulk sample

(ng/mg)

Reference

Grain handling 54,900 2x105 - 5x105 Dutkiewitz et al., 86 Grain handling 237.2# Blaski et al., 96 Grain handling 2 - 23.7 # Beard et al., 96 Grain handling* 3 - 2,247 1 - 199 in dust Simpson et al., 98a grain elevator 0 - 0.74p 22.5 - 187.5 DeLucca et al., 84 Grain drying 214 Liesivuori et al., 94 drier emptying 16,100 Grain mill 105 Liesivuori et al., 94 silo emptying 159 - 8,850 Olenchock et al., 90b wheat dust 268.7 # Olenchock et al., 89 oat dust 283.3 #

# nanograms calculated from Endotoxin Units by dividing by 10 p personal sampling * UK study

4

1.6 THE UK MAXIMUM EXPOSURE LIMIT (MEL) FOR GRAIN DUST

The HSE guidance on the new maximum exposure limit (MEL) for grain dust came into effect on the 1 January 1992 (HSE Agricultural Information Sheet no. 3). It applies to all dust, including contaminants, arising from harvesting, drying, handling, storage or processing of barley, wheat, oats, maize or rye. The MEL is an average of 10 mg/m3 of total respirable dust in the air over an 8 hour period. This is a maximum and not a target. Dust levels must be reduced as far below the MEL as reasonably practicable. Higher concentrations can be permitted if exposure times are shorter, but the maximum dust level should never exceed 30 mg/m3 measured over a 10 minute period.

The MEL was set because of increasing evidence that exposure to high levels of grain dust causes ill health. HSE recommends that, when taking on new workers who may be exposed to grain dust, employers should check whether they have any pre existing chest disorders such as asthma. Employees should be encouraged to report any respiratory problems. The MEL did not take into account the fact that grain dust is a biologically active dust because, at the time, there had been no comprehensive study of the effects of the individual components of grain dust on the health of the grain workers. Consequently, there was a need to investigate these effects, and this study described in this report forms part of this investigation.

1.7 THE HEALTH AND SAFETY EXECUTIVE GRAIN DUST SURVEY: AIMS

The Health and Safety Executive (HSE) Grain Dust Survey has been an on-going longitudinal collaborative survey of worker exposure to grain dust in the UK. The survey involved assessments of the exposure to grain dust and its contaminants and clinical assessments of workers' respiratory health for workers handling grain in a range of occupational settings at 2 docks, 27 farms and, at the start, animal feed mills in the South East of England. The study population of 321 workers potentially exposed to grain dust was selected and recruited by a study team led by Dr David Blainey of the Medical Academic Unit, Broomfield Hospital, Chelmsford, who also carried out the clinical assessments. HSL provided dust and microbial exposure measurement and immunoassays of exposed workers’ sera.

The purpose of this study was to assess the long term clinical and immunological effects of workplace exposure to grain dust and its contaminants in terms of its effect on respiratory health, the prevalence of respiratory symptoms and airflow limitation in grain workers, and to examine the role of immunological and cellular markers in predicting the link between exposure to grain dust and the development of occupationally acquired disease of the respiratory tract.

A detailed examination was made of the exposure of a cohort of grain workers to microorganisms in dust in order to relate this to the incidence of immunological response and health effects. Airborne microorganisms were studied, both quantitatively and qualitatively, while grain was being handled on farms during harvest and after harvest; when stored grain was being moved and milled for feed on farms, and also during bulk handling of grain that was being imported or exported at dockside terminals. The findings from the microbiological study were then used in the subsequent studies to

5

measure the prevalence of immunological response to grain dust associated allergens in UK grain workers. In the clinical study, a computer administered respiratory health questionnaire was completed by volunteer workers at the farms and grain handling docks. Lung function tests were completed to measure respiratory airway calibre and serum samples were collected for use in the immunological studies.

The study was divided into 2 phases. Phase 1 took place from 1990 to 1993 and Phase 2 from 1997 to 2003, as far as possible using the same worker cohort.

This final report describes the outcome of the study. As there were three main parts to the study, the report will deal in separate sections with the assessments of workers’ microbiological exposure, with immunoassays and with clinical assessments.

6

2 MICROBIOLOGICAL EXPOSURE STUDY

2.1 BACKGROUND

The aim of the microbiological survey was to assess the longitudinal exposure of grain workers in the UK to total inhalable grain dust and to quantify and qualify their exposure to the microbial contaminants in the grain dust, including identification of the predominant microorganisms involved, in the second phase of the study measurement of endotoxin was included. The findings from the microbiological study were then used in the subsequent studies to measure the prevalence of immunological response to grain dust associated allergens in UK grain workers.

The microbiological exposure study was conducted on the two docks and a subset of the total of 27 farms in the South East of England which were included in the HSE Grain Dust Study. The farms were selected on the basis of the days on which they were carrying out relevant work. The study was carried out in two phases. In Phase 1, from 1990 to 1993, nine farms (F1-F9) and two dockside grain terminals (A and B) were included. In Phase 2, from 1997 to 2002, eight farms (F10-F17) and dock A were included.

2.2 STUDY SITES, MATERIALS AND METHODS

2.2.1 Farms

Air samples were taken from farms F1 to F5 and F10 to F17 during the harvest, and from farms F6 to F9 while grain was being handled after storage. At each farm, one to three workers were involved with grain handling. The activities taking place for each farm at the time of sampling are summarised below:

F1, 2 & 10. Barley was harvested, transferred to a tractor-drawn trailer and emptied into a barn or silo.

F3-5 & 11-17. Wheat was harvested, transferred to a tractor-drawn trailer and emptied into a barn or silo.

F6. Old wheat was loaded by tractor into lorries in one shed, and new grain was unloaded by tractor in a second shed.

F7. Oats, then barley, were milled and bagged in a barn, and the grain was shovelled into the barn manually and by tractor.

F8. Stored wheat was shovelled by tractor from a barn to a shed where it was milled.

F9. Stored wheat was sucked up from a barn floor to a storage bin. Men shovelled the wheat to the nozzle (this did not appear to be a very

dusty process).

7

2.2.2 Docks

Samples were taken at Dock A while wheat, barley and maize were being handled. Workers were involved in a range of activities including loading and unloading grain from ships and lorries, moving stored grain, maintenance and cleaning, and office work. Samples were taken as background controls in the offices and rest area. Grain entered and left the dock by lorry or boat. It was transported between these and the storage silos indoors on open conveyor belts. These operated in the building and also in an overhead enclosed bridge which led out to the ships. Grain was piped from the conveyor belt into the lorries. Lorry loading was controlled from upper and lower loading galleries and workers here, as well as next to the lorries, were exposed to the dust generated when the grain reached the open lorry. Lorries delivering grain tipped it into a hatch at the side of the loading bay. This was a dusty process but did not require workers to stand nearby. Grain was unloaded from boats, either by suction or scoop, and loaded onto the ship down a chute. The offices were on the 5th floor of the building, and at the terminal entrance with the rest area adjacent.

Samples were taken in the following grain handling areas:

1. Lorry loading and unloading and scales room.

2. Grain movement in terminal: by basement conveyors in silos and ‘office’ in basement by upper conveyors in the bridge between silos and ship.

3. Dockside loading and unloading of ships.

4. Control room offices and rest area.

Dock B was much smaller than dock A and was not involved in Phase 2. Sampling in dock B, while wheat was loaded onto a ship from lorries and in a shed while wheat was being moved by a tractor, was only undertaken in the second year of the study during Phase 1. Grain was stored in large sheds and carried to the ship by lorries which tipped it onto the quay side from where it was sucked into the hold.

2.2.3 Bioaerosol sampling.

There are many different types of sampler available on the market that can be used to sample airborne micro-organisms (bioaerosols); each one has different advantages and limitations. In order to enumerate viable airborne particles, a sampler needs to be efficient in physically collecting particles of the required size range while minimising sampling stresses so that biological activity is not impaired. In practice, the best approach is to use more than one type of sampler to provide complementary results.

Three different bioaerosol samplers were used during Phase 1, to enable maximum recovery of the different species of microorganisms present to be achieved and to obtain information on particle size distribution (Crook et al., 1988). During Phase 2 filtration samplers, the method of sampling that performed best during Phase 1 and the preferred method for endotoxin sampling, were used. Static samplers were placed in areas where workers were likely to be exposed to grain dust. These included the inside of vehicle cabs, in work areas and in offices to provide a background control. Personal sampling was carried out extensively in Phase 2 in combination with static sampling.

8

The samplers used in Phase 1 were as follows:

• Andersen samplers, to separate airborne particles into six size fractions (more than 8.2 microns, 5.0-10.4, 3-6, 2-3.5, 1-2 and less than 1.0 microns) impacted directly onto the surface of agar media in petri dishes, operated at 25L min-1 for exact times between 30 seconds and 5 minutes depending on conditions (Andersen 2000 Inc., Atlanta, GA, USA) (Andersen, 1958).

• Aerosol monitors: Filter samplers loaded with polycarbonate membranes (37 mm diameter, 0.8 mm pore size) in disposable plastic cassettes (Nucleopore; Sterilin, Bibby, Stone) (Palmgren et al., 1986) and connected to battery operated portable vacuum pumps sampling air at 2L min-1 for up to 4 hours, (after sampling, the exact time was recorded). Dust deposits were washed from the surface of each filter with 5 ml of fluid to form an aqueous suspension, from which a ten fold stepwise dilution series was prepared, 0.1ml aliquots of appropriate dilution's were spread onto the surface of agar plates.

• Midget liquid impingers (SKC; Poole) charged with 10 ml of collection fluid (1/4 strength Ringer solution with 1% inositol (Oxoid)), into which airborne particles are suspended as the air is drawn through the sampler at 1L min-1 for up to 4 hours. The cell suspension was diluted in a tenfold stepwise series and 0.1ml aliquots of the appropriate dilutions were spread onto the surface of agar plates as above (May and Harper, 1957).

The use of the Andersen sampler was limited to open, less dusty areas because it overloaded quickly in the highly contaminated conditions. Since the pumps were not intrinsically safe they were not used in some enclosed areas because of a potential dust explosion hazard. Aerosol monitors and midget impingers were used with intrinsically safe vacuum pumps and could be used in highly contaminated areas without overloading because the samples could be diluted before plating, which also meant that they could be left to run for longer periods of time.

Quantitative results of airborne microorganisms were calculated from the long period sampling with midget impingers and aerosol monitors, while qualitative results from these were supplemented by data from short term sampling with Andersen samplers which also provided information on particle size distribution. An aerosol monitor and midget impinger sample were taken together in each sample site.

The samplers used in Phase 2 were as follows:

• IOM filtration samplers. This type of sampler superseded the use of the aerosol monitors on the basis of better collection efficiency and of consistency with HSE approved gravimetric dust sampling methods. Aerosol monitors are approved by NIOSH and EPA in the USA, but it is possible that electrostatic charges building up in the plastic holder could affect inlet efficiency (Crook, 1995a). The IOM sampler therefore has been optimised by HSL for bioaerosol sampling and analysis of bio-inhalable airborne dusts.

IOM filtration samplers were used with polytetrafluorethylene micro porous membrane filters (25 mm diameter, 0.5 µm - 0.8 µm pore size). The filters were weighed before and after sampling for total dust measurements. Samples were then processed as described for aerosol monitors. Endotoxin free buffer was used to wash the filters and an aliquot of the aqueous suspension removed and assayed for endotoxin content.

All the samplers and pumps were calibrated for correct airflow level with a rotameter (SKC) before use.

9

2.2.4 Microbial culturing

Combinations of five types of agar media and four different incubation temperatures were used. Bacteria were grown at 25oC and 37oC on nutrient agar (with the addition of chloramphenicol to prevent fungal growth); fungi at 25oC and 40oC on malt agar (with the addition of penicillin and streptomycin to prevent bacterial growth), and at 25oC on Dichloran glycerol (DG) 18 agar to reveal xerotolerant genera; thermophilic bacteria and actinomycetes were grown at 55oC on tryptone soy casein agar (Lacey and Dutkiewicz, 1976) and R8 agar (Amner et al., 1989). Plates were incubated for at least 7 days, and colonies counted at regular intervals until no more colonies emerged.

2.2.5 Quantitative microbiological analysis

Quantitative results for airborne microorganisms were calculated from the long-period sampling with midget impingers, aerosol monitors and IOM filtration samplers. Qualitative results from these were supplemented where possible by data from short-term sampling with Andersen samplers which also provided information on particle size distribution. After all the colonies had emerged, the number of colony-forming units (cfu) on each type of agar media, at each incubation temperature used, was calculated for every sample and converted to cfu/m3 air.

2.2.6 Microbiological Identification.

Representative colonies of the prevalent taxa from each site were isolated into pure culture and identified.

Fungi - Fungi were identified by direct observation of colonies growing on isolation plates and by microscopy. The total numbers in each taxon were recorded where possible.

Bacteria - Representative colonies of the bacteria most commonly occurring at each site were selected and isolated into pure culture. Bacteria were identified using colony morphology, Gram staining, cell shape and biochemical tests kits. The latter included the Biolog 96 well plate identification system (Atlas Bioscan Ltd, Hayward California) and API 20 and 50 well identification strips (BioMerieux (UK) Limited, Basingstoke), and the results were analysed by the proprietary computer software to compare results obtained with those from type species.

2.2.7 Endotoxin Analysis

Aliquots of the IOM filter sample suspension were centrifuged at 1000 g for 10 minutes to remove particles and dilutions of the supernatant were prepared for analysis. Samples were analysed using the Kinetic-QCL automated system (Bio-Whittaker Inc., Walkersville, Maryland, USA). Results are derived as Endotoxin Units (EU; 10 EU equates to 1 ng endotoxin as is quoted for some analytical methods) and calculated as EU /m3 air sampled.

10

2.3 RESULTS SUMMARY AND DISCUSSION

The microbiological results from both phases of the survey are summarised in Table 2.1. The microbiological data from Phase 1 was reported in detail in an HSL report number (IR/L/IM/93/01) and in Swan and Crook (1998). The raw data for Phase 2 is included in Annex 1. Table 2.2 lists the predominant fungi isolated and Table 2.3 the predominant bacteria.

11

Table 2.1. Summary of aerobiological contamination levels measured during Phase 1 and 2 at farms and docks.

Bacteria cfu/m3 Fungi cfu/m3 Thermophilic Bacteria and

Dust mg/m3

(Highest personal Endotoxin EU/m3

(Highest personal actinomycetes

cfu/m3 monitor) monitor)

Farms Phase 1 5.8 x 104 - 1.05 x 109 1.8 x 103 - 1.3 x 107 0 - 2.3 x 103 0.2 - 488 -

(102.7) Phase 2 1.09 x 103 -1.6 x 108 400 - 2.9 x 107 0 - 5.4 x 103 0.08 - 69.53

(69.53) <5 - 4.17 x 104

(1.85 x 104) Docks

Phase 1 8.1 x 103 - 1.4 x 1011 2.5 x 104 - 6.5 x 109 0 - 3.9 x 104 0.06 - 313.5 -(313.5)

Phase 2 6.6 x 103 - 3.1 x 108 193 - 8.1 x 106 0 - 1.37 x 104 1.56 - 54.29 (137.27)

20.6 - 7.7 x 106

(4.22 x 103)

12

Table 2.2. Fungi consistently isolated from airborne grain dust.

Species Dock A Dock B Farms Phase 1 Phase 2 Phase 1 Phase 1 Phase 2

Harvest Stored Harvest

Alternaria spp. + + * * + *

Aspergillus spp * * +

Cladosporium spp. * * * * *

Eurotium spp. * + +

Penicillium spp. * * * *

Yeasts + + + * +

* predominant (104 -106/m3 in most samples)

+ present (103-104/m3 in most samples)

13

Table 2.3. Bacterial taxa consistently isolated from airborne grain dust.

(not ranked in order of prevalence)

Microbacterium .

Coccal rods:

Gram negative spp. Gram positive spp. Rods: Agrobacterium Enterobacter agglomerans Pseudomonas corrugata Pseudomonas diminuta Pseudomonas fluorescens Pseudomonas glycosyles Pseudomonas/Xanthamonas maltophilia Pseudomonas marginalis Pseudomonas testosteroni Xanthomonas oryzae�

Rods: Bacillus licheniformis Bacillus subtilis

sppRhodococcus fascians

Curtobacterium spp.

Cocci: CDC Group A Micrococcus spp. Staphylococcus cohinii Staphylococcus epidermidis Staphylococcus xylosus

Samples were collected using midget liquid impingers aerosol monitors and IOM filtration heads sampling air at 2 l/min for up to 4 hours and Andersen samplers sampling at 28.3 l/min for between 0.5 - 5 min. Bacteria were grown at 25oC and 37oC on nutrient agar. Representative colonies of the prevalent taxa from each site were identified using BIOLOG and API test kits and PCR in 2002.

14

Figure 2.2 Airborne contaminants associated with grain handling on farms phase 2

1.00E+10

1.00E+8

1.00E+6

1.00E+4

1.00E+2

1.00E+0

1.00E-2

Bacteria (cfu) Fungi (cfu) Thermophilic bacteria and fungi (cfu) Endotoxin (EU) Dust (mg)

Airb

orne

con

tam

inan

ts lo

g 10

uni

ts/c

ubic

met

er

a b c d e f g h i j k l m n o p q r s

F10 a – d F11 e – k F14 l – n F17 o – r F13 s

Please note: no Figure 2.1 in this report

Wheat harvest.

Farm 10. a. Inside tractor; b. Inside combine; c. Outside combine; d. Next to grain discharge. Farm 11. e. Tractor driver personal monitor; f. Inside tractor; g. Outside tractor; h. Combine driver personal monitor; i. Inside combine; j. Outside combine; k. Control away from harvest. Farm 14. l. Tractor driver personal monitor; m. Inside tractor; n. Outside tractor. Farm 17 . o. Tractor driver/grain store worker personal monitor; p. Inside tractor; q. Outside tractor; r. In grain storeFarm 13. s. Control away from harvest.

15

Figure 2.5. Airborne contaminants associated with grain handling at dock A phase 2

1.00E+9

1.00E+7

1.00E+5

1.00E+3

1.00E+1

1.00E-1

Airb

orne

con

tam

inan

ts lo

g 10

uni

ts/c

ubic

met

er a

ir

Bacteria (cfu) Fungi (cfu) Thermophilic bacteria and actinomycetes (cfu) Endotoxin (EU) Dust (mg)

a b c d e f g h i j k l m n o p q r

Please note – no figures 2.3 or 2.4 in this report Dock A. a.Ship hull; b. Lorry loading gallery; c. Scales room; d. basement; e. Works office; f. Personal monitors (highest results); g. On boat.; h. Basement conveyor; I. Basement office; j. 7th floor cupola; k. Control room; l. Lorry loading; m. Rest area; n. Personal monitor loading ship; o. Static on the ship; p. Personal monitor basement and scales room; q. Basement office; r. Rest area.

16

Table 2.4. FARMS: Predominant fungi found in individual samples from farms 1-15 cfu/m3 x 105 (% of total)

Sample Total number Penicillium Cladosporium Alternaria

Phase 1 BARLEY HARVEST inside combine 1.40 0.02 (1.5) 1.00 (73.1) 0.21 (15.0) in field downwind 2.70 0.00 0.87 (32.2) 1.10 (41.0) WHEAT HARVEST outside combine 92.00 0.00 22.00 (26.0) 49.00 (53.0) inside combine 0.33 0.00 0.16 (48.5) 0.15 (45.4) in grain store 3.70 0.93 (25.1) 0.69 (18.6) 1.80 (5.0) by dresser 8.30 3.30 (40.0) 3.30 (40.0) 0.83 (10.0) STORED GRAIN moving old wheat 3.60 3.63 (99.5) 0.01 (0.25) 0.0 (0.25) milling 2.70 2.20 (82.0) 0.06 (2.2) 0.40 (14.8) Phase 2 BARLEY HARVEST outside combine 66.00 0.00 22.00(33.9) 28.00 (42.4) inside combine 0.68 0.00 0.20 (29.4) 0.31 (45.9) Inside tractor 1.40 0.00 0.54 (38.6) 0.72 (51.4) in grain store 19.50 0.00 15.60 (80.0) 3.90 (20.0)

Numbers are given in colony forming units per m3 air sampled. Samples were collected using midget liquid impingers aerosol monitors and IOM filtration heads sampling air at 2l/min for up to 4 hours. Fungi were grown at 25oC and 40oC on malt agar and at 25oC on DG18 agar.

17

Table 2.5. DOCKS: Predominant microorganisms found in individual samples from docks A and B.

cfu/m3 x 105 (% of total) Sample Total number Penicillium Aspergillus Cladosporium Alternaria

Phase 1 lorry unloading grain 320.00 0.00 6.90 (2) 83.00 (26) 97.00 (30) lorry being loaded imported grain

0.33 0.02 (6) 0.27 (82) 0.02 (6) 0.005 (2)

basement 19.00 2.60 (14) 9.70 (52) 2.60 (14) 0.00 cupola 1.70 0.0 (4) 1.00 (60) 0.00 0.07 (4) conveyor to ship 0.67 0.10 (15) 0.36 (54) 0.19 (28) 0.00 loading ship 7.60 0.24 (3) 0.00 4.00 (52) 1.30 (18) unloading ship 4.70 0.79 (17) 3.50 (74) 0.04 (1) 0.00 Phase 2 WHEAT FOR ANIMAL CONSUMPTION personal monitor 0.27 0.23 (85.0) 0.004 (1.7) 0.03 (10.0) 0.00 lorry unloading 0.09 0.02 (23.5) 0.005 (58.9) 0.08 (82.4) 0.00 basement 3.10 2.90 (92.5) 0.24 (7.7) 0.00 0.00 basement office 0.34 0.29 (85.5) 0.004 (1.3) 0.004 (1.3) 0.00 loading ship 1.50 1.30 (86.2) 0.10 (6.4) 0.00 0.00 WHEAT FOR HUMAN CONSUMPTION Lorry loading 0.06 0.01 (9.1) 0.01 (20.0) 0.006 (9.1) 0.006 (9.1) basement 1.10 0.00 0.39 (35.5) 0.47 (42.8) 0.16 (14.3) loading ship 10.00 0.00 4.00 (40.0) 4.0 0 (40.0) 2.00 (20.0)

Numbers are given in colony forming units per m3 air sampled. Samples were collected using midget liquid impingers aerosol monitors and IOM filtration heads sampling air at 2l/min for up to 4 hours. Fungi were grown at 25oC and 40oC on malt agar and at 25oC on DG18 agar.

18

2.3.1 Worker exposure to total inhalable dust

An examination of worker exposure to total grain dust in Phase 1 of the study was carried out by the HSL Field Scientific Support Unit and reported previously (Bagon, 1993). Dust measurements in Phase 2 were carried out by HSL Microbiology Section. In all years, except harvest 2002, dust levels equal to, or greater than, 30 mg/m3 were recorded in all work situations. Figures 2.2 and 2.5 summarise the dust levels recorded during Phase 2 of the study alongside the microbial and endotoxin levels measured in each dust sample.

Docks - During Phase 1 the results indicated a wide spread of personal exposures up to 313 mg/m3

(achieved during loading of grain on a ship). Dock workers were exposed to total inhalable dust levels exceeding 10 mg/m3 for over 40% of samples analysed during Phase 1 and 23% during Phase 2. The dock work during Phase 2 dust sampling was not continuous. However, dust levels of greater than 30 mg/m3 were still recorded during loading of grain on a ship, where levels reached 275.71 mg/m3, and levels in the basement work area in 2000 were over twice the maximum exposure limit. Maximum personal exposure was 6.14 mg/m3 working in the head house and basement and 137.27 mg/m3 whilst loading the ship in 2002.

Farms- During Phase 1 28% of samples from farms were over 10 mg/m3, most of which occurred during milling which produced 54% of results between 5-30 mg/m3. The highest personal exposure recorded was 102.7 mg/m3 (recorded on the driver of a combine harvester with no cab) and 41.4 mg/m3 was the next highest recorded, on a drier operator. During Phase 2 dust levels reached 69.5 mg/m3, 17.5% of samples were over 10 mg/m3, 14.2% were over 30 mg/m3, and the highest personal exposure was 69.5 mg/m3 recorded on a tractor driver.

2.3.2 Microbiological results

Figures 2.2 and 2.5 summarise the total yields of culturable fungi, bacteria and actinomycetes at each sampling site and include dust and endotoxin levels for Phase 2 samples. The predominant species of microorganism found during the survey are listed in Tables 2.2 and 2.3. The study design allowed for enumeration of identified fungi in individual samples, but only for an overall estimate of predominant bacteria. Concentrations of the predominant airborne fungi on farms are presented in Table 2.4 and at docks in Table 2.5. In the majority of samples taken during this survey the concentrations of airborne bacteria outnumbered the fungi, sometimes by several orders of magnitude, as described in more detail below.

Farms- (Figure 2.2 and Tables 2.1, 2.4)

F1, 2 and 10 during harvesting of barley: Airborne fungal spore concentrations ranged from 8.3 x 104

to 4.5 x 105 colony forming units per cubic metre (cfu/m3) during Phase 1 and 6.8 x 104 to 8.5 x 106

cfu/m3 in Phase 2. Concentrations of Alternaria and Cladosporium spp. inside the cab of a lorry collecting the harvested barley reached 2.6 x 105 cfu/m3 and 1.6 x 105 cfu/m3 respectively. Concentrations of airborne bacteria ranged from 1.2 x 105 to 1.3 x 107 cfu/m3 during Phase 1 and 2.4 x 104 - 1.6 x 108 cfu/m3 during Phase 2. Airborne actinomycetes were present in small numbers, ranging from none detected to 5.4 x 103 cfu/m3. Barley harvest dust contained around 29 - 73 % Cladosporium and 15 - 51 % Alternaria spp. viable spores.

F3-F5 and F11-F17 during harvesting of wheat: Concentrations of airborne fungal spores ranged from 1.8 x 103 to 1.3 x 107 cfu/m3 during Phase 1 and 400 to 2.9 x 107 cfu/m3 during Phase 2. Predominant fungi included Cladosporium (between 19 and 48%) and Alternaria spp. (between 5 and 53%),

19

3

3Alternaria spp. concentrations reached 4.9 x 106 cfu/m (53% of total fungi present) outside the combine harvester cabs and 1.5 x 104 cfu/m3 (45.4% of total fungi) inside, and Cladosporium spp 2.2 x 106 cfu/m3 (26% of total fungi) outside and 1.6 x 104 (48.5% of total) inside. Concentrations of airborne bacteria ranged from 5.8 x 104 to 1.0 x 109 cfu/m3 during Phase 1 and 1.1 x 103 to 1.6 x 108

cfu/m3. Airborne actinomycetes were present in small numbers ranging from none detected to 4.3 x 103 cfu/m3. Combine harvester and tractor drivers’ personal exposure levels to fungal spores ranged from 6.9 x 103 to 3.5 x 105 cfu/m3 and bacteria levels from 5.2 x 104 to 3.0 x 106 cfu/m .

F6-F9 during handling of stored grain: Conditions in grain stores varied. On some farms, particularly small ones, settled dust coated everything. Others were cleaner. Airborne fungal spore concentrations ranged from 6.8 x 103 to 1.1 x 106 cfu/m3. Predominant fungi during the milling of oats and barley included Penicillium spp. at levels of 2.2x105 cfu/m3 (82% of total fungi) and Cladosporium spp. at 5.9x103 cfu/m3 (2.2% of total fungi). The concentration of Penicillium spp. inside a tractor cab moving old wheat was 3.63x105 cfu/m3 (99.5% total fungi). Airborne bacterial concentration ranged from 1.3 x 104 to 2.1 x 107 cfu/m3. Bacteria consistently isolated from the grain dust included a range of Gram negative Pseudomonas and Enterobacter spp., Gram-positive CDC Group A, Rhodococus spp., Micrococcus spp. and Staphylococcus spp.. Airborne thermophilic bacteria and actinomycetes were present in small numbers, from none detected to 9.3 x 103 cfu/m3.

Docks (Figure 2.5 and Tables 2.1, 2.5 )

Lorry delivery of grain generated a great deal of dust but workers were not required to stand close by. However, workers in the loading galleries next to the lorries were exposed to dust when the grain reached the open lorry. Grain was transported round the docks on open conveyor belts that moved at speed, shaking the grain and generating clouds of dust. The locations in which these operated included the basement, 7th floor and the enclosed bridge leading out to the ship. The basement contained a cabin like office in which the workers can sit while monitoring the area. Dust was also generated where the grain dropped onto the conveyor belts. Unloading boats by scoop caused little dust, and unloading by suction even less. However, loading the ship using a chute caused dense clouds of dust, in the hold and dockside area; this was by far the dustiest procedure. Office and rest areas were monitored as background controls, however, even these areas were visibly dusty.

Concentrations of airborne fungal spore ranged from 2.5 x 104 to 6.5 x 109 cfu/m3, (office controls 3.1 to 5.3 x 104 cfu/m3) during Phase 1 and 262 to 8.1 x 106 cfu/m3 (control 193) during Phase 2. Penicillium spp. were isolated from all sites, numbers reached 2.6x105 cfu/m3 next to a conveyor carrying barley in the basement, and 1.3 x105 cfu/m3 (86%) on the ship deck, next to the hold, during wheat (for animal consumption) loading. Aspergillus spp. were also predominant at all sites, including A. fumigatus at concentrations of up to 2.7x104 cfu/m3 (82%) during loading of wheat into lorries, and 4.4x104 cfu/m3 on the ship deck next to the hold during the loading of animal feed wheat. Cladosporium spp. concentrations reached 4.0x105 cfu/m3 on the ship deck during wheat loading in Phase 1 and Phase 2 and 4.7 x104 cfu/m3 next to the basement conveyor carrying wheat, 2.6x105

cfu/m3 for barley. Airborne bacterial concentrations ranged from 8.1 x 103 to 1.4 x 1011 cfu/m3 (office controls 2.2 x 103 cfu/m3 to 1.2 x 106 cfu/m3) during Phase 1 and 6.6 x 103 to 3.1 x 108 cfu/m3 (1.1 x 103 cfu/m3 control) during Phase 2. Airborne thermophilic bacteria and actinomycetes were present in very small numbers ranging from none detected to 3.9 x 104 cfu/m3. In Phase 1 the dust from wheat grown for animal feed contained more fungi, bacteria and actinomycetes than the wheat grown for human consumption (the difference was not as marked in Phase 2). During handling of wheat, concentrations of airborne fungal spores reached 6.7 x 104 cfu/m3 with wheat for humans and 1.6 x 106 cfu/m3 with feed wheat; bacteria reached 1.6 x 105 cfu/m3 and 2.3 x 106 cfu/m3 respectively, and thermophilic bacteria and actinomycetes reached 8.0 x 102 cfu/m3 and 3.9x103 cfu/m3 respectively. In Phase 1 all samples of dust from the feed wheat contained A. candidus, but few or no colonies of this fungus were grown from other samples. Near to the conveyor leading from the silo to the boat,

20

concentrations of A. candidus reached 1.6 x 104 cfu/m3 . Aspergillus spp. formed 50% of the colonies grown including 1.6 x 104 cfu/m3 of A. fumigatus. Cladosporium spp. reached 1.9 x 104 cfu/m3. By contrast, in Phase 2 by far the most predominant isolate from all samples of dust from the animal feed wheat were penicillia spp., over 85% in most samples. Levels of penicillia spores reached 2.9 x 105

cfu/m3 in the basement and personal exposure levels reached 2.3 x 104 cfu/m3. Aspergillus levels were very low (1 - 8% in most samples). Larger numbers of thermophilic microorganisms were associated with the animal feed wheat than wheat for human food. In the wheat for human food, although concentrations of thermophilic bacteria and actinomycetes reached a maximum of 8.0 x 103 cfu/m3, none was detected in many of the samples, whereas they were present in concentrations exceeding 5 x 103 cfu/m3 in all samples taken during the handling of feed wheat.

Bacteria consistently isolated from the grain dust included Gram-positive spore-forming Bacillus spp. and cocci (Curtobacterium spp, Micrococcus spp. and Staphylococcus spp.), and a range of Gram negative Pseudomonas/Xanthomonas and Enterobacter spp.

The control room and rest/mess room were ‘clean’ areas where no personal protective equipment was worn. However, high levels of bacteria (1.5 x 105 cfu/m3 control rm., 4.1 x 103 cfu/m3 mess rm.) and fungi (1.6 x 104/ 580 cfu/m3) were recorded.

2.3.3 Microorganisms and Particle size distribution

The particle size distribution data obtained from the Andersen samples showed a similar pattern for all samples taken at the farms and docks (Figure 2.6). More microorganisms (62% of actinomycetes, 41% of bacteria and 34% of fungi) were deposited in the first stage of the sampler than on other stages. This indicated that particles were larger than 8.2 µm aerodynamic diameter. About 12% of total particles were deposited on each of the five other stages, with aerodynamic size ranges of 5.0-10.4, 3.0-6.0, 2.0-3.5, 1.0-2.0 and up to 1.0 µm respectively. This is consistent with other reports that bacteria and actinomycetes aggregate more than fungi (Lacey and Dutkiewicz 1994).

21

Figure 2.6 Particle size data from Andersen samples.

70

60 Fungi Bacteria

Mic

roor

gani

sms (

cfu)

50 Actinomycetes

40

30

20

10

0

1 2 3 4 5 6 Stages

Andersen samplers sampled at 28.3 l/min for between 0.5 - 5 min. Particles are separated according to their aerodynamic diameter, into six size fractions, before being deposited onto agar plates. Stage 1, more than 8.2 µm; stage 2, 5.0-10.4µm; stage 3, 3-6µm; stage 4, 2-3.5µm; stage 5, 1-2µm; stage 6, less than 1.0µm .

2.3.4 Endotoxin Results

Levels of airborne endotoxin were high at the farms and the dock during all grain handling activities, they are summarised in Table 2.6. Levels of over 10,000 EU/m3 were recorded at all the workplaces visited and personal exposures reached over 600 EU/m3 at every workplace.

The highest exposure areas at the dock were on the ship hold during loading when levels reached 7.7 x 106 EU/m3 and in the basement area where levels reached 1.8 x 106 EU/m3. Personal exposure levels to endotoxin were extremely high ranging from 13.9 to 4.2 x 103 EU/m3, work in the head house and basement resulted in personal exposures of 1.2 x 103 - 4.2 x 103 EU/m3. Personal exposure to endotoxin during ship loading reached 1.9 x 105 EU/m3. Of particular concern are the endotoxin levels in the control room and rest/mess room which are supposedly ‘clean’ areas where no personal protective equipment was worn, however, high levels of endotoxin up to1,585/ 765EU/m3 were recorded.

During harvest and grain handling on the farms airborne endotoxin levels reached 4.2 x 104 EU/m3 by a grain store hopper. Levels inside vehicle cabs ranged from <5 to 8.25 x 103 EU/m3. The personal exposure levels ranged from <5 to 1.85 x 104 EU/m3 recorded on a tractor driver during the harvest.

22

Table 2.6. Summary of Phase 2 endotoxin results

Site EU/m3 x 103

Personal exposure Static sample Farms Inside combine 0.139 - 8.25

n=5, x = 1.87 <0.005 - 2.18

n=5 x=0.6165 Outside combine 0.193 - 19.5

n=6, x = 15.2 Inside tractor 0.298 - 185.00

n=8 x=3.72 <0.005 - 6.74

n=6 x=1.30 Outside tractor 0.039 - 17.2

n=6 x=3.9 Grain store 1.17 0.0004 - 41.7

n=6 x=8.9 Farm Office <0.005 - 2.16

n=2 Control 0.5 - 5

n=2 Docks On Ship 0.0591 - 190

n=3 x=1111.00 74- 7700 n=5 x=2800

Lorry loading controll area 1.6 - 1.8 n=2

Scales room 0.0206 - 210 n=3 x=70

Basement/Head house 1.20 - 4.20 n=5 x=2.70

8.9- 1800 n=3 x=610

Basement office 2.1- 3.0 n=2

Office/control room 0.36 0.001 - 3.4 n=3 x=1.8

Mess/rest room 0.0245 - 0.770 Cleaner (general areas) 0.0139- 0.3191

n=2

n=number of samples, x = sample average

(The Dutch recommended health-based occupational exposure limit for airborne endotoxin is 50 EU/m3 based on personal inhalable dust exposure measured as an eight-hour time-weighted average see Section 2.4.7)

2.4 SUMMARY OF MICROBIOLOGICAL EXPOSURE STUDY

Workers handling or working in the vicinity of grain being moved at the dockside or on farms were exposed to airborne dusts containing concentrations of microorganisms that frequently exceeded 1 x 106 m-3 of air. In all years dust levels equal to, or greater than, 30 mg/m3 were recorded in all work situations except on farm 17 during the 2002 harvest. Environmental levels of airborne endotoxin were high at the farms and the dock during all grain handling activities, levels of over 10,000 EU/m3

were recorded at 17 of the 19 workplaces visited and personal exposures reached over 600 EU/m3 at these workplace.

23

2.4.1 Dust levels

Peak exposure levels to grain dust remained high throughout the study, and in Phase 2 were still frequently over the MEL introduced between Phases 1 and 2. In particular, levels were high in the dock basement area and at farms during combining and in their stores. However, overall there was a reduction of about 40% both at Dock A and at the farms in the number of samples in which grain dust exposure was over 10mg/m3(from 40% of samples to 23% at the dock and from 28% to 17.5% at the farms). Personal dust exposure also decreased at the Dock between Phases 1 and 2, with the maximum personal exposure levels measured falling from 313 mg/m3 to 137 mg/m3, however dock work during Phase 2 dust sampling was not continuous.

2.4.2 Microbiological results

The Phase 2 findings confirmed and expanded on the Phase 1 findings. Exposure levels to bacteria and fungi remained very high throughout the study. The aerobiological sampling provided “snapshots” of exposure levels, however, there was an overall trend at the dock towards a decrease in the maximum levels measured of total bacteria and fungi from Phase 1 to Phase 2, nevertheless the maximum levels sampled per cubic meter of air were still extremely high over 8 million for fungal spores and 300 million for bacteria. This reduction was not seen at the Farms.

The overall picture obtained was that quantitatively the microbial populations did not differ greatly (Table 2.1) from site to site, year to year or from grain to grain. Individual samples varied and levels of airborne dust peaked during various procedures such as the loading of the ship, however, the overall levels of airborne microorganisms were consistently high.

Qualitatively, populations of fungi, bacteria and actinomycetes differed little in dust from different grains. The largest qualitative differences found were between freshly harvested grain and stored grains and between grain for human and animal consumption. The predominant microorganisms did not change from year to year. There were some slight differences in the infrequently isolated fungi. Overall, numbers of airborne fungi at the docks and farms were similar but, as might be expected, species differed between the two areas. During harvest, the microorganisms in the dust are mostly saprophytic "field fungi" that colonise the grain during growth, such as Cladosporium spp., Alternaria spp., Verticillium spp., and bacteria such as Enterobacter agglomerans (Pantoea agglomerans, Erwinia herbicola) and Pseudomonas spp. (Dutkiewicz, 1997; Lacey, 1980).

Bacterial numbers were highest in the dust generated during handling of freshly harvested grain. The field fungi, Alternaria spp. and Cladosporium spp., were the predominant fungi in the dust generated during harvesting. No Aspergillus flavus were isolated, and numbers of Aspergillus fumigatus were very low in fresh grain. There were some slight differences in the infrequently isolated fungi, during Phase 1 more Verticillium were isolated and during Phase 2 more Paecilomyces. Yeast levels also varied year by year and were higher in Phase 1 with very few Aureobasidium isolated in Phase 2.

Barley generated the largest concentrations of airborne microorganisms, bacteria and fungi, both reaching 3.0 x 108 cfu/m3 next to conveyors carrying barley at dock A .

Once harvested and stored, grain becomes colonised by a different range of microorganisms, depending on storage conditions - especially water content, oxygen content and temperature. As a result, the constituents of grain dust generated during harvesting are different from those in dust generated when stored grain is handled. If the grain is stored dry (12-13% water content) microorganisms present at harvest may survive but do not proliferate. If the water content of the grain is greater, some spores of ‘storage fungi’ naturally present may germinate and grow, including Aspergillus spp., Eurotium spp., and Penicillium spp. (Lacey, 1980). These fungi can grow and displace field fungi in drier grains. To prevent fungal growth, a water content in grain of less than

24

13% is required (Lacey, 1988; Lacey, 1995). In badly stored, damp grain the increased metabolic activity among microorganisms can lead to spontaneous heating in the stored grain, which, with enough water, can reach 65-70oC and the development of a succession of different species which are increasingly thermotolerant or thermophilic, including allergenic fungi and actinomycetes (Darke et al. 1976; Lacey and Dutkiewicz, 1994). Aspergillus, Penicillium and Eurotium spp. were the predominant fungi in the dust at the docks.

Dust clouds created during the handling of animal feed wheat contained many more thermophilic fungi and bacteria (particularly A. candidus, A. flavus, Penicillium and Bacillus spp.) than the grains for human consumption.

The lack of thermophilic actinomycetes throughout this study indicates that the grain handled was stored fairly well, although the presence of high levels of A. candidus and Penicillia in animal feed grain indicated that this was less well stored than grain for human consumption. Saccharopolyspora (Faenia) rectivirgula and Saccharomonospora viridis, both previously associated with farmer's lung disease, were present in small numbers only at dock A in year 2 of Phase 1, and dock B.

2.4.3 Comparisons with other studies

Previous UK studies of the microbial content of dust associated with stored grain have mainly investigated mouldy or moist grain, and were carried out some time ago. Non-UK studies have not gone into detail when investigating microbial populations, and some only include settled grain dust. The study reported here investigated the microbial content of airborne dust generated from grain that was harvested and stored using modern techniques; consequently the grain was in good condition.

There have been many studies on the health effects of grain dust and its microbial content on workers. The majority of studies on the health effects of grain dust exposure concentrate on the respiratory health of the exposed workers (Carvalheiro et al., 1995; Dimich-Ward et al., 1995; doPico 1980; Pollock et al., 1980; Senthilselvan et al., 1996; Von Essen et al., 1990). Some studies include measurements of grain dust exposure as well as respiratory health (Corey et al., 1982; Fonn et al., 1993b; Heederick et al., 1994; Huy et al., 1991; Massin et al., 1995; Tielemans et al., 1994). A few studies added immunological assessments of the workers including atopic status and sensitivity to fungal antigens (Chan-Yeung et al., 1992; doPico et al., 1982; Lin et al., 1994; Zuskin et al., 1994). Few studies have included the exposure of workers to micro-organisms during grain handling at farms and docks in the UK or the immunological effect these might have on the workers. Most notable was Darke et al. (1976) who studied respiratory disease in 76 workers handling grain during harvesting at 16 farms in the UK between 1970 and 1973; they did not look at dust from stored grain. The predominant micro-organisms and their relative numbers found during harvesting were similar to those found in the study reported here. Differences between years and crops were quantitative rather than qualitative. They also found few actinomycetes and actinomycetes and bacteria accounted for fewer than 10% of the total spores in the dust. By contrast we found larger numbers of bacteria than fungi in most of our harvest samples. Darke et al. (1976) used a cascade impactor to obtain total spore counts and a six-stage Andersen sampler to obtain identifications and quantification of viable fungi, bacteria and actinomycetes. A personal Casella sampler was worn by the combine harvester drivers to obtain personal data of exposure levels of total spores. A respiratory symptoms questionnaire was carried out as well as lung function tests, skin prick tests using antigens from fungi isolated from the combine harvester dust, and precipitin tests against the organisms implicated in farmer's lung. A quarter of the participants complained of respiratory distress while working on combine harvesters or near grain elevators or dryers. Symptoms were cough, wheezing and breathlessness; sometimes these were so severe as to prevent work. Drivers were exposed to up to 20 million spores per cubic meter of air; Cladosporium spp . were predominant with Alternaria and Verticillium/Paecilomyces also abundant. The relative frequency of different colony types could not be determined accurately

25

because the plates were overloaded. Antigen extracts from these species produced positive skin-prick reactions, precipitin reactions and rapid decreases in FEV1 when inhaled by affected workers. Actinomycetes and bacteria accounted for less than 10% of total spores; there were very few actinomycetes.

In a South African study, Fonn et al. (1993a) used Casella personal filter samplers, and Andersen samplers to investigate levels of viable micro-organisms in grain dust. Gram positive cocci were the most predominant contaminant found (3.0 x 106 - 3.5 x 106 cfu/m3 air); they did not report their microbial findings in detail. 0.01-958 x 106 airborne fungal spores were found at Canadian grain elevators (Lacey, 1980) with Ustilago spores predominating; Aspergillus, Mucor and Cladosporium were present in most samples, and Alternaria in large numbers at some sites. Results in the study reported here were similar in concentration but differed in that Penicillium spp. predominated and Mucor was not frequently isolated. In three studies in southern USA, Palmgren et al. (1983), DeLucca et al. (1984) and DeLucca and Palmgren (1987) investigated the viable bacterial population in settled grain dust. They found that the predominant bacterium was E. agglomerans (up to 87% of all isolates), followed by Pseudomonas spp. (up to 28% of all isolates). Klebsiella spp., Citrobacter spp. and Serratia spp. were also identified. One study included quantification of fungi; Aspergillus spp. and Penicillium spp. predominated. Dutkiewicz (1978a) investigated the viable microbial population of airborne grain dust in Polish grain stores and mills using a slit sampler that impacts particles directly onto agar plates. He found bacterial levels of 2.26 x 104 - 1.3 x 106 cfu/m3 air. E. agglomerans predominated but Staphylococcus epidermidis, Streptomyces spp., Acinetobacter spp., Bacillus cereus and Pseudomonas spp. were also present in high numbers.

Several USA studies have looked at exposure to grain dust. Todd and Buchan (2002) and Buchan (2002) found that 58 % of total dust and 33% of respirable dust samples exceeded the eight hour TWA at the corn storage facilities. 85% of the samples contained over 500EU/m3 air with a maximum of 1.7x106 EU/m3 air. Viet et. al. (2001) took measurements during a Colorado wheat harvest, all samples were personal samples, total dust levels ranged from 0.09-15.33mg/m3 air, endotoxin levels were 4.4-744.4EU/m3 air.

2.4.4 ‘Normal’ levels of airborne microorganisms

To put the exposure levels recorded above into context, some ‘normal’ levels of airborne microorganisms are included here. Sampling representative background levels during the study was problematical as grain dust can travel long distances. This was shown in the fairly high background samples taken in the ‘clean’ dock A offices where concentrations of airborne fungal spores ranged from 580 to 5.3 x 104 cfu/m3 and airborne bacterial concentrations ranged from 1.0 x 103 cfu/m3 to 1.2 x 106 cfu/m3. This dust was observed to be carried in on workers clothes and blown in on the wind through the open doors. Natural atmospheric viable microbial conditions in a typical suburban area were reported to be 0 - 7.2 x 103 (mean 273) cfu/m3 mesophilic fungi, 0 - 193 (mean 2.1) cfu/m3

thermophilic fungi, 0 - 71 (mean 1) cfu/m3 A. fumigatus, 42 - 1.6 x 103 (mean 79) cfu/m3 bacteria. The highest concentrations occurred during summer and autumn (Jones and Cookson, 1983). Crook and Lacey (1988) reported concentrations of viable airborne micro-organisms outdoors to be: 500 cfu/m3

total bacteria, 10 cfu/m3 Gram-negative bacteria, 1,200 cfu/m3 total mesophilic fungi, 300 cfu/m3

thermophilic fungi and 60 cfu/m3 thermophilic bacteria and actinomycetes. Ambient levels of viable airborne bacteria in an agricultural area were reported by Bovallius et al. (1978), 2 - 3.4 x 103 (mean 99) cfu/m3, and in a city 100 - 4.0 x 103 cfu/m3 (mean 850). These levels are about 10 - 107 fold lower than those recorded during the handling of grain dust. Workers handling, or working in the vicinity of, grain being moved at the dockside or on farms were exposed to airborne dusts containing concentrations of micro-organisms that frequently exceeded 106 per m3 of air. Airborne bacteria and fungi exceeded 104 per m3 air in all areas sampled.

26

2.4.5 Exposure to airborne microorganisms inside air-conditioned cabs

Darke et al. (1976) recommended that all combine harvesters should have cabs to protect workers from dust. All tractors and combines in this study had cabs. Although all air entering the combine cabs was filtered and air conditioned, and concentrations of airborne micro-organisms inside cabs on the combine harvesters were decreased by 10 to 100 fold, numbers inside were still high - with up to 7.0x105 cfu/m3 fungi and 2.74 x 103 to 4.2x106 cfu/m3 bacteria measured during this study. These results are important because farm workers rely on the cab filtration to provide protection and do not wear personal protective equipment inside the cabs. The high levels of airborne micro-organisms found inside the air-conditioned cabs during Phase 1 of the grain dust study resulted in a study of combine harvester and tractor filtration systems. This was a collaborative study carried out by the Microbiology and Filtration Sections at HSL and reported by Thorpe et al. (1997), the details of which are summarised below.

In this study the performance of air-conditioning systems in agricultural vehicles was examined; the efficiency of the filters was measured, along with air leakage past the filters. Measurements of total airborne dust and viable micro-organisms inside and outside the cabs were also made during the harvest of 1994 at 3 farms on 4 combine harvesters and 4 tractors.

The numbers of bacteria outside the cabs ranged from 4.5 x 105 to 5.5 x 108 cfu/m3 inside, and from 1.5 x104 to 4.1 x 106 cfu/m3. The numbers of fungi outside the cabs ranged from 2.3 x 105 to 8.7 x 107

cfu/m3 inside, and from 1.8 x104 to 3.0 x 106 cfu/m3. Grain dust levels ranged from 1.56 - 103.7 mg/m3 outside and 0.09 - 2.36 mg/m3 inside. The ratio of bacteria outside to inside the cabs ranged from 26 to 1120, for fungi from 15 to 156 for grain dust from 6.4 to 216. The protection factors varied widely between filters of different materials and even between filters of the same material. Predominant micro-organisms in all samples were Cladosporium spp. (5.5 - 13 µm long x 4-6 µm wide) and Alternaria spp. (50µm long x 3-6 µm wide).

The protection factors for bacteria were higher than those for fungi. This was unexpected because bacteria are smaller than fungi and, therefore, more likely to penetrate the filter or leak past the seals. Possible explanations for this were that the bacteria aggregated together in clumps, or were attached to dust particles. Another explanation was that the high levels of micro-organisms inside the cabs were a result of redispersion of settled dust in the cab, or dust carried in on the driver or entering the cab when the door was open. As the bacteria are more susceptible to dehydration, there would be fewer viable bacteria inside the cabs.

The conclusion of this study was that leakage of dust into the cab through the ventilation inlet is not a problem. The high levels of micro-organisms in the cab are due to re-dispersion of settled dust, dust brought into the cab on the body of the driver, or dust entering the cab when the door is opened. A protection factor of 100 is a realistic aim at present. Anything higher would be difficult to achieve and negated every time the door was opened. It was interesting to note that the old filters often performed better than the new ones, as the layer of dust clogging the old filters added to the filtration protection.

2.4.6 The potential importance of bacteria and endotoxin in respiratory sensitisation

In addition to the potential role of fungi in respiratory allergy, this study has highlighted the potential importance of bacteria and endotoxin in grain dust. Total concentrations of bacteria were higher than fungi, and gram-negative bacteria contributed to this total. Other studies have also found high levels of Gram-negative bacteria, particularly Enterobacter agglomerans (Dutkiewicz, 1976; 1986; 1987) which has been shown to be a cause of occupational allergy in farmers (Dutkiewicz, 1978b; Dutkiewicz et al., 1985). Dutkiewicz (1987) investigated bacteria in the indoor farming environment

27

and found that the most common were staphylococci and other cocci, spore forming bacilli, corynebacteria and Gram-negative rods - similar results to those reported here. They concluded that Enterobacter agglomerans was the greatest hazard. This was one of the predominant bacterial taxa found in this study.

Gram-negative bacteria are hazardous due to their endotoxin content. E agglomerans has a potent endotoxin (Dutkiewicz, 1986). There is a substantial body of evidence that endotoxin has a major role in occupational respiratory disease amongst grain workers (Clapp et al., 1994; Schwartz et al., 1994, 1995, 1996; Smid et al., 1994) and other workers exposed to Gram-negative bacteria (Douwes et al.,1997; Michel et al., 1995, 1997; Sandström et al., 1994).

2.4.7 Occupational exposure limits for Endotoxin.

At present there are no occupational exposure limits (OEL) for endotoxin in place in any country. Several of the studies reviewed in thesis have recommended ‘no effect’ levels based on their results. The International Journal of Occupational and Environmental Health Criteria Document on ‘Endotoxins in the Environment’ written by the Committee on Organic Dusts International Commission on Occupational Health contains a paper by Rylander (1997) which suggests some ‘no effect’ levels for environmental endotoxin. An occupational exposure standard based on ‘no effect’ levels is being considered in the Netherlands. The Dutch recommended health-based occupational exposure limit for airborne endotoxin is 50 EU/m3 (approximately 4.5 ng/m3), based on personal inhalable dust exposure measured as an eight-hour time-weighted average (Heederik and Douwes, 1997).

Rylander (1997) recommends guidelines for ‘no effect’ levels for environmental endotoxin based on the values for persons with a history of atopy or asthma:

Disease ng/m3

Toxic pneumonitis 200 Airways inflammation 10

Systemic effects 100

These figures were based on the results of the studies reviewed in the Criteria Document on Endotoxins in the Environment (The Committee on Organic Dusts International Commission on Occupational Health, 1997). Except for the studies by Michel (1997) and Michel et al. (1995, 1992) the exposure experiments used to calculate these figures were conducted on healthy volunteers or workers. Michel has shown that asthmatics and atopics have a more acute reaction to LPS and an increased risk of developing pulmonary effects after endotoxin exposure. A ‘no effect’ level would therefore be expected to be lower for such subjects. Studies involving already exposed workers may have been affected by the ‘healthy worker’ effect and tolerance may have obscured the dose-effect relationship and raised the ‘no effect’ levels. There is evidence that endotoxin can have an adjuvant effect on allergic reactions (Michel et al., 1991), this has important implications for the study of the effects of complex organic dusts such as grain dust which contain known allergens as well as very high levels of endotoxin.

In the study reported here levels of airborne endotoxin of over 10,000 EU/m3 were recorded at 17/19 of the workplaces visited and personal exposures reached over 600 EU/m3 at these workplaces. 77.5% of farm and 91.2% of dock LPS samples were over 50EU/m3, 67.5% of farm and 82.3% of dock LPS samples were over 200EU/m3. Of particular concern are the endotoxin levels in the control room and rest/mess room which are supposedly ‘clean’ areas where no personal protective equipment was worn,

28

however, high levels of endotoxin up to1,585 in the control room and 765EU/m3 in the rest/mess room were recorded.

Several USA studies have looked at exposure to endotoxin in grain dust. Todd and Buchan (2002) and Buchan (2002) found that 85% of the samples contained over 500EU/m3 air with a maximum of 1.7x106 EU/m3at a corn storage facility (no personal sampling was carried out). Viet et. al. (2001) took measurements during a Colorado wheat harvest, all samples were personal samples, endotoxin levels were 4.4-744.4EU/m3 air.

There are very little data available on the longitudinal health effects of high endotoxin exposure, and the adjuvant effects of endotoxin inhalation with other contaminants. Von Essen (1997) reviewed endotoxin exposure in grain dust and airways obstruction, he concluded that endotoxin is believed to be responsible for some or all of the respiratory effects of grain dust. Baur et. al. (2003) published two case reports concluding that grain dust induces asthmatic reactions and ODTS which are not of allergic origin and they suspected the high concentrations of endotoxins associated with grain dust.

The high levels of endotoxin exposure recorded during this study including the high personal exposure levels, indicate that the repeated dose toxicity of the endotoxin component of organic dust may make an important contribution to the development of chronic lung disease. Although no quantitative data was collected it was apparent that many of the grain workers taking part in this study were not protected by PRPE whilst exposed to high levels of endotoxin. At the farms PRPE was not worn during the harvest except in the grain stores yet the farmers were exposed to endotoxin levels well over proposed safety limits whilst in the fields. At the docks endotoxin levels in the mess room and offices were very high and again no PRPE was worn in these areas.

2.4.8 Other Microbial Toxins

Mycotoxins may contribute to occupational lung disease in farmers. Aspergillus spp. and Penicillium spp. produce mycotoxins and both were present in the dust generated during the handling of stored grain during this study. Airborne mycotoxin levels during grain handling are low (Lacey and Dutkiewicz, 1994; Selim et al., 1998 ) but, in one study, ten out of fifteen grain dust samples contained mycotoxin (Palmgren et al., 1983). Todd and Buchan (2002) found mycotoxins present at all sites during a study of Colorado corn storage. Their possible role in causing respiratory symptoms is not fully understood and the presence of mycotoxins in grain dust has not been studied in the UK.

Bacterial toxins other than endotoxin, such as Pseudomonas exotoxin A, could also play a role in respiratory disease.

(1-3)β-D-glucan is a polyglucose compound in the cell walls of fungi, some bacteria and plants. It is a potent inflammatory agent and may also be a respiratory immunomodulatory agent. Exposure to (1-3)β-D-glucans has been associated with an increased prevalence of atopy, and a decrease in FEV1 (Thorn and Rylander, 1998; Fogelmark and Rylander, 1994).

29

2.5 CONCLUSIONS

This study has provided previously unreported information on the wide range and large numbers of dust, endotoxin, fungi and bacteria to which workers handling grain in the UK are exposed. It has characterised the exposure of workers during normal work activity, highlighted the very high levels of exposure to endotoxin, and has provided the basis for the study of the immunological, toxicological and clinical effects of occupational exposure to grain dust contaminants.

30

3 IMMUNOLOGY STUDY

3.1 AIM

To determine which specific biological components of grain dust isolated during the microbiological survey are eliciting an immunological response in UK grain workers. The details of methods used during Phase 1 of the study are described in ‘The Respiratory Effects of Grain Dust’ (Bainey et. al., 1995).

3.2 STUDY POPULATION

The study population was made up of volunteers from two different docks, a variety of different farms, and animal feed mills in Essex and East Anglia. 304 Male farmers, dock workers and mill workers were studied during Phase 1. Phase 2 did not include mill workers. Not all the participants of the immunology study gave a blood sample every year and in Phase 1 many of the blood samples were not large enough for every test to be carried out on each sample.

Blood was collected by nursing staff from the Broomfield Hospital research team (see Clinical study section of this report). A sample of venous blood for immunological analysis was taken from each participant using the vacutainer systemTM (Becton Dickinson) and the serum separated and stored at minus 20oC until analysed.

Table 3.1 Workers who took part in the Immunological survey

Site Number of workers in immunological survey

Phase 1 Phase 2

1998 1999 2001/02

Docks 128 27 23

17

Farms 116 135 105

63

Mills 60 - -

-

Controls - 30 21

20

Total 304 192 151

100

The control population used in Phase 1 were workers from an electronics factory who were not occupationally exposed to microorganisms or grain dust. No further information on the control population is available. The control population used in Phase 2 were dock workers and hospital porters who were unexposed to grain dust.

Eleven dockworkers and eleven farmers took part in both the Phase 1 immunology study and the final year of the study (2001/2). These workers are referred to here as the ‘stayers’.

31

3.3 IMMUNOGLOBULIN G IMMUNOASSAY

3.3.1 Why IgG assay was performed

IgG is the most common immunoglobulin found in normal human serum. There are four subclasses of human IgG; in this study total IgG was measured. IgG was measured as a marker of worker exposure to the components of the grain dust. IgG is part of the normal immune response to antigen, when an individual first encounters an antigen the cells of the immune system may produce an immune reaction in response to the antigen. In a typical immune response to antigen the primary response consists of IgM and IgG antibodies, the secondary response consists predominantly of IgG. The presence of raised levels of specific IgG in serum indicate immunological response to exposure to a particular antigen. Raised levels of IgG are not necessarily associated with clinical symptoms. IgG is associated with hypersensitivity type III reactions when immune complexes of antibody/antigen can be formed for example in the lungs following repeated exposure to large amounts of inhaled antigen. IgG antibodies against specific antigens in organic dusts are a characteristic immunological feature of extrinsic allergic alveolitis/granulomatous pneumonitis (farmer's lung).

The Enzyme-Linked-Immunosorbent-Assay (ELISA) was used to measure levels of antigen-specific IgG (against the fungi, and grain) in workers' serum. The ELISA method used in Phase 2 was slightly different from that used in Phase 1 as the reagents used during Phase 1 were no longer on the market. The levels of antigen-specific IgG in the serum from the grain workers was compared with the levels in the control population.

3.3.2 IgG ELISA method

The ELISA assay was performed with the antigen passively adsorbed onto the base of 96-well flat-bottomed polystyrene plates. The amount of antigen bound to the well is critical and has to be optimised. Too much antigen can enhance non-specific binding; too little antigen can leave only a small capacity for specific binding resulting in too low an assay sensitivity. Serum is then added to the well and antigen-specific antibody binds to the antigen. The plate is washed and then anti-human-IgG antibody added. This binds to the IgG bound to the antigen. The Phase 2 method used peroxidase conjugated anti-human-IgG antibody (for the Phase 1 method the plate was washed again and peroxidase conjugated antibody specific for the anti-human-IgG antibody was added). The plate is washed and O-phenylenediamine dihydrochloride (OPD) is added. The peroxidase enzyme catalyses a reaction with the OPD producing a yellow coloured soluble end-product. The enzyme reaction is stopped by the addition of acid. The optical density can then be read.

The ELISAs were controlled by including a positive Aspergillus fumigatus test and high test sera for each specific antigen taken from the initial control test plate and repeated on each subsequent plate to enable plate to plate variation to be taken into account. The extract of A. fumigatus and a positive anti-A. fumigatus serum were purchased from the National Institute for Biological Standards and Control (NIBSC, UK). Micro-titre plates (Nunc-Immuno plates, Gibco BRL) were incubated overnight at 4oC with 0.1 - 1 µg/ml test antigen and 0.5 µg/ml A. fumigatus control antigen (Table 4.3). The plates were then washed 3 times and diluted 1/500 worker's serum and control serum were then added and the plate incubated for 2 hours at room temperature. The plate was washed again and 50µl of a 1/1000 dilution of peroxidase labelled monoclonal anti-human IgG (mouse)(Sigma) in 1% Tween 20 in PBS was added. The plate was incubated for 1 hour at room temperature. The plate was again incubated for 2 hours and washed. 50µl of 7.4mM o-phenylenediamine (OPD) and 5.9mM hydrogen peroxide in 0.01M citrate phosphate buffer (pH6) were added. The plate was incubated for 15 minutes, then 2M

32

sulphuric acid was added to stop the reaction. The optical density of each well was read at 492 nm. Workers with serum levels of specific IgG that gave an optical density over 2 standard deviations above the mean optical density of the non-occupationally exposed control population were taken to be positive.

3.3.3 Optimisation of dilutions of allergen extract and sera

Optimum dilutions for the extracts (Table 3.2) and sera were found by carrying out checkerboard ELISAs using samples from each group of extracts in a range of dilutions and a range of dilutions of known positive sera. All allergen extracts were diluted in carbonate bicarbonate buffer pH 10. All sera were diluted 1/500 in 1% Tween 20 in PBS.

Table 3.2 Antigen concentrations used

Antigen Antigen concentration µg/ml

NIBSC A. fumigatus 0.5 Test Fungi except 1.0 Eurotium spp. Eurotium spp. 0.5 Wheat 0.1 Maize 1

3.4 IMMUNOGLOBULIN E IMMUNOASSAY

3.4.1 IgE and allergy; why IgE assay was performed

Five classes of antibody are recognised in humans, immunoglobulin (Ig) A, D, E, G and M. IgE is present in the blood stream in much smaller amounts than other antibodies (50-300ng/ml in normal individuals (Sutton and Gould, 1993). Its main role is to defend the body against parasites, but in industrialised western countries IgE is more commonly known as the antibody involved in immediate allergic responses (type I hypersensitivity reactions). IgE is also involved in type IV delayed-type hypersensitivity reactions.

Diseases in which IgE-mediated allergy may be involved are allergic migraine headache, anaphylaxis and angioedema, asthma, conjunctivitis, dermatitis (contact and atopic), eczema, extrinsic allergic alveolitis, certain gastrointestinal disorders, hypersensitivity pneumonitis, rhinitis, sinusitis and uriticaria (Holgate and Church, 1993).

Allergy is caused by over production of IgE in response to common environmental and occupational antigens. It is not known why some common substances e.g. house-dust mites, pollen and fungi, stimulate IgE synthesis in some people and not in others. There are several

33

suggested explanations including variation in mucosal permeability, T cell cytotoxin imbalances and genetic predisposition.

Type 1 immediate hypersensitivity reactions - There are two stages to type 1 immediate hypersensitivity reactions: initial sensitising exposure to allergen, followed by a repeat exposure that leads to the release of allergic mediators which cause the symptoms of allergic disease. Inhaled, injected or ingested allergen is presented to the immune system which recognises it as foreign and produces IgE. The allergen-specific IgE binds to the high affinity receptors on mast cells and basophils and sensitises the cells. Allergic mediators are not released until re-exposure to the allergen. When the allergen is next inhaled, injected or ingested it binds to the antigen-specific IgE bound to the surface of immune cells This crosslinking of the IgE receptor complex by antigen activates the cell, releasing histamine and other mediators which cause type 1 immediate hypersensitivity and contribute to the later response and regulation of IgE synthesis (Holgate and Church, 1993).

Type IV delayed type hypersensitivity - Antigen-sensitised T cells release lymphokines upon subsequent contact with the allergen. The lymphokines induce inflammatory reactions and attract macrophages which release more mediators of inflammation. Symptoms occur about 8-24 hours after re-exposure to the allergen.

3.4.2 Detection of IgE by RAST

Specific IgE against the individual contaminants of grain dust was measured as an indication of potential development of immunological symptoms in response to grain dust leading to possible clinical symptoms. Quantitative measurement of allergen-specific IgE in serum requires a sensitive test to detect the minute quantities found in subjects. The Radio-Allergo-Sorbent-Test (RAST) is currently the standard laboratory technique, the system at HSL is UKAS Accredited. Using the RAST technique, specific IgE for a large number of antigens can be tested for, using one serum sample from the subject.

3.4.3 Bulk culturing of allergens and extract preparation

Fungi – Fungi (isolated as described in the Microbiology section of this report) were grown in bulk culture using Czapek Dox liquid medium (Oxoid), supplemented with 1% maltose, 0.5% dextrose, 0.25% CaCl2, 1/4 strength minimal essential medium (MEM) amino acids, 1/4 strength MEM nonessential amino acids and 1/4 strength MEM vitamins. This is a synthetic and protein-free medium specially developed for use in the production of fungal antigens. 250ml of medium in a conical flask was inoculated with pure culture, stoppered with a foam bung loosely covered with foil, and incubated on a shaker at 25oC for 8 weeks with 1 final week without shaking to allow a sporing mycelial mat to grow.

Extract preparation was carried out at 4oC to minimise autodigestion. Cultures were frozen then thawed, placed in an ice bath and homogenised followed by sonication. The suspensions were then incubated overnight to allow the release of cell contents and protein molecules into solution. The extracts were then centrifuged and the supernatant dialysed against 20mM ammonium bicarbonate buffer for 24 hours followed by freeze drying over night and reconstitution in a small amount (5-10ml) of PBS. The extract was then dialysed against PBS to remove any residual ammonium bicarbonate and before having the protein content assayed. Extracts were stored at -20oC.

34

Actinomycetes - For Phase 2 Actinomycete extracts that had been quality controlled using positive non human sera became available on the market and were purchased from Microgen Bioproducts Ltd.

Bacteria - Lawns of bacteria were grown on nutrient agar (Oxoid) plates. The bacteria were harvested by scraping from the surface of the agar and re-suspending in PBS. The bacteria were then frozen and thawed before sonication in an ice-bath. The extracts were incubated overnight to allow the release of cell contents and protein molecules into solution, then centrifuged and the supernatant dialysed against PBS for 24 hours. The protein content was then assayed. The extracts were stored at -20oC.

Storage mites - 70% pure storage mite cultures (Acarus siro, Tyrophagus spp., Lepidoglyphus destructor) were purchased from CSL,York. Extracts were made by rotating the mites overnight in 20% w/v 0.02M ammonium bicarbonate. The suspension was centrifuged and the supernatant dialysed against ammonium bicarbonate buffer overnight. This was then freeze-dried and reconstituted in PBS before a final dialysis against PBS and protein assay. The extracts were stored at -20oC.

Grains - 20% w/v solutions made of whole wheat flour and homogenised maize in PBS were rotated overnight. Insoluble material was removed by centrifugation and the supernatant dialysed against PBS overnight before protein assay. The extracts were stored at -20oC.

3.4.4 RAST assay method

Cellulose disks are used to provide a large surface area for antigen to bind to; the disks are first activated using cyanogen bromide; a reactive imidocarbamate group is introduced to the hydroxyl groups of the cellulose in the paper enabling it to bind covalently to the amine groups of the antigen proteins (Axen and Ernbach, 1971). The disc coated with antigen is then incubated with the test serum. Antigen-specific antibodies in the serum bind to the disc. Excess serum is washed off and the disk incubated with radio-labelled anti-human IgE which binds to IgE bound to the disk. Excess labelled anti-human IgE is washed off. The amount of radioactivity, measured by a gamma counter, is proportional to the quantity of specific IgE in the patient's serum.

Allergen disc preparation - Paper discs (Schleicher and Schull no. 589\3) were activated with cyanogen bromide using the method of Ceska et al. (1972). The disks were washed and dried overnight. The dry disks were stored frozen at -20oC.

Mixed allergen discs (Table 3.3) were prepared using 1mg of mixed protein in 15ml of 0.1M sodium bicarbonate buffer per 100mg of activated discs. Atopy discs were coated with grass pollen, cat fur and house dust mite (Swann Technology, Royston, Herts). Discs were rotated with the extract at room temperature overnight. The discs were then washed and stored in PBS with 0.1% Tween 20 at -20oC.

35

Table 3.3 Antigens tested

Antigens Atopy grass pollen, cat fur, house dust mite Fungi

Alternaria spp, Aspergillus flavus, Aspergillus fumigatus, Aureobasidium, Cladosporium spp, Eurotium spp, Paecilomyces spp, Penicillium spp, Scopulariopsis, Verticillium spp.

Bacteria Pseudomonas spp, Agrobacteria spp, Staphylococcus spp, Pseudomonas agglomerans, Micrococus spp, Rhodococus spp.

Storage Mites Acarus siro, Lepidoglyphus destructor, Tyrophagus spp. Actinomycetes Micropolyspora faeni, Saccharomonospora viridis,

Streptomyces spp., Thermoactinomyces thalpophilus, Thermoactinomyces vulgaris, Thermomonospora curvata

Grain Wheat, Maize

IgE measurement - Each sample was tested in duplicate. 200µl of a 1:4 dilution of serum in PBS was added to an allergen disc and incubated overnight at room temperature. The serum was then removed and the discs were washed. I125-labelled rabbit anti-human IgE antibody (Kabi Pharmacia) was diluted 1:1 with PBS, and 100ml were added to each disc and incubated overnight. The excess I125-labelled antibody was removed and the discs were again washed. I125 bound to the discs was measured using a gamma counter (Canberra-Packard). The washing was initially carried out by machine, after the machine irrecoverably broke down, during processing of the 1998 sera, washing was carried out by hand. Quality controls were included in every assay, high quality control sera was cat positive serum purchased from the National Institute for Biological Standards and Control (NIBSC, UK), low quality control serum was prepared by the HSL Health Effects Section (SOP 1.4). The high and low sera were tested with atopy discs, the low sera was also tested with the grain dust allergen test discs.

The results were expressed as RAST percent binding, which was defined as the amount of radioactivity bound to the disc as a percentage of the total radioactivity added. The IgE results are presented using a ‘cut off’ of 1% binding of radiolabelled anti-human IgE above which a result is deemed to be positive. This enables comparison of the results across the two Phases of the study.

36

3.5 RESULTS

Sera were assayed for specific IgG and IgE against the allergen extracts from the microbiological study. These allergen extracts are referred to here as the work-related allergens. The overall results are summarised in Tables 3.4. Table 3.5 summarises the overall results from IgE assay in terms of percentage positive sera from workers. Percentages given were calculated from the number of sera tested for each disc/extract, not from the total number of workers in the study, as not all workers gave blood and in some cases, during Phase 1 only, there was not enough blood to carry out the full range of assays. In Table 3.6 IgE results are compared between atopic and non-atopic workers. Table 3.7 summarises the results of IgG assay for sera tested against extracts of grain material and of fungi. Table 3.8 summarises the results of IgG assay for sera tested against extracts of bacteria. In Table 3.9 the numbers of positive and negative sera in Phase 1 of the study are compared with Phase 2 for the ‘stayers’, examining the numbers that went from remained positive, remained negative or went from positive to negative or negative to positive.

37

Table 3.4 Immunoassay results summary

Workers’ location

1990 1998 1999/00 2001/02

% atopic % +ve IgE to one or more

% +ve IgG to one or more

% atopic % +ve IgE to one or

% +ve IgG to one or

% atopic % +ve IgE to one or

% +ve IgG to one or

% +ve IgE to one or

% +ve IgG to one or

extracts* extracts* more more more more more more extracts* extracts* extracts* extracts* extracts* extracts*

Docks 19.8 18.2 12.7 33.3 7.4 39.3 43.5 21.7 21.7 17.6 17.6

Farms 28.9 28.8 30.2 23.3 21.2 40.7 25.7 22.9 41.9 20.6 25.4

Mills 9.6 4.5 16.9 - - - - - - - -

Controls - - - 33.3 6.7 26.6 14.3 9.52 23.8 0 35

* excluding bacterial and atopy discs

38

Table 3.5 Summary of RAST IgE results

Allergen disc % positive volunteers

Docks Farms Controls

1990

Atopy 19.8 26 15

Fungi 6.6 15.6 15

Grain 2.4 6.8 5

Bacteria 0 1 0

Actinomycetes 5.8 3.2 10

Storage Mites 9.5 18.4 15

% positive to one or more discs* 18.2 28.8 -

1998

Atopy 33.3 23.3 33.3

Fungi 0 3.6 3.3

Grain 3.7 5.1 3.3

Bacteria 0 0 0

Actinomycetes 0 0 0

Storage Mites 3.7 15.3 3.3

% positive to one or more discs* 7.4 21.2 6.7

1999

Atopy 43.5 25.7 14.3

Fungi 8.7 4.8 4.8

Grain 13 9.5 4.8

Bacteria 0 0 0

Actinomycetes 0 0 0

Storage Mites 8.7 16.2 0

% positive to one or more discs* 21.7 22.9 9.5

2001/2

Fungi 5.9 4.8 0

Grain 11.8 4.8 0

Bacteria 0 0 0

Actinomycetes 0 0 0

Storage Mites 0 14.3 0

% positive to one or more discs* 17.6 20.6 0

*excluding bacterial extracts and atopy

39

Table 3.6 Summary of RAST IgE results divided into subjects who are or are not also atopic.

% positive workers Allergen disc Docks Farms Controls

atopic not atopic atopic not atopic atopic not atopic

1990

Fungi 1.7 5.2 3.1 8.2 - -

Grain 1.7 0.9 5 2 - -

Bacteria 0 0 0 1.1 - -

Actinomycetes 3.5 2.7 0 1.2 - -

Storage Mites 6.9 3.5 13 2.6 - -

% positive to one or more discs*

9.5# 10.3# 16# 13# - -

1998

Fungi 0 0 1.5 2.2 3.3 0

Grain 3.7 0 4.4 0.7 3.3 0

Bacteria 0 0 0 0 0 0

Actinomycetes 0 0 0 0 0 0

Storage Mites 3.7 0 10.2 5.1 3.3 0


7.4 0 13.1 8 6.7 0

1999

Fungi 8.7 0 1.9 2.9 4.8 0

Grain 13 0 6.7 1.9 4.8 0



Storage Mites 8.7 0 9.5 6.7 0 0


21.7 0 13.3 9.5 9.5 0

2001/2

Fungi 5.9 0 1.6 3.2 0 0

Grain 11.8 0 3.2 1.6 0 0



Storage Mites 0 0 4.8 9.5 0 0


17.6 0 9.5 11.1 0 0

*excluding bacterial extracts and atopy # not all tested for atopy

40

Table 3.7 Summary of ELISA IgG results

Source of Allergen % positive volunteers

1990 1998 1999/2000 2001/2002

Docks Farms Mills Docks Farms Controls Docks Farms Controls Docks Farms Controls

Wheat 8.7 17 1.7 3.6 7.4 6.6 4.3 9.5 9.5 5.9 7.9 10

Maize 5.6 3.5 0 0 0 3.3 4.3 1.9 0 5.9 4.8 10

Alternaria 0 4.1 5.1 0 12.6 6.6 4.3 16.2 4.8 5.9 4.8 5

A. candidus 0.8 6.6 5.5 14.3 9.6 9.9 0 0.95 9.5 0 4.8 0

A. fumigatus 0.8 0 1.7 10.7 16.3 3.3 8.7 21.9 4.8 5.9 6.3 5

A. flavus 2.3 1.6 0 0 0.74 3.3 0 0.95 0 0 4.8 0

Cladosporium 2.3 4.3 9.1 0 1.48 3.3 4.3 6.7 0 0 7.9 5

Eurotium 0.8 7.6 1.8 0 0 0 0 0 0 0 3.2 5

Penicillia 0 1.7 3.6 10.7 2.96 3.3 0 0.95 0 0 3.2 15

% positive to one or more fungi

4.8 15.7 13.5 39.3 37.8 23.3 13 36.2 14.3 17.6 25.4 30

NOTE: DUE TO FORMATTING PROBLEMS THERE IS NO PAGE 41 IN THIS REPORT

42

Table 3.8 Summary of Phase 1 ELISA IgG results – bacteria

Source of Allergen % positive volunteers Docks Farms Mills

Ps. diminuta 36 4.3 36.7 Ps. maltophilia 18.9 0 73.3 Ps. testosteroni 0 0 1.6 Ps. glycinii 3.3 0 5.4 Ps.corrugata 0.8 0 0 Ps. fluorescens 0.8 0.9 3.2 Staph. cohini 1.6 1.7 1.6 Staph. epidermidis 5.5 7.8 3.2 Staph. xyloses 18.9 8.6 10 Sphingobacter 14.2 10.3 6.7 Curtobacterium 0 0 0 E. agglomerans 3.9 0.9 10 % positive to 1 or more Pseudomonas

43.3 5.2 86.7

% positive to 1 or more Streptomycete

21.3 14.6 13.3

% positive to 1 or more bacterial extract

50.4 25.9 88.3

43

Table 3.9 ‘Stayers’ population immunology results summary

Worker location

Assay 1990 2001/2 + - + Æ + + Æ - -Æ - -Æ +

Dockers IgG 3 8 0 3 6 2 IgE 4 7 2 2 7 0

Farmers IgG 0 11 0 0 10 1 IgE 4 7 3 1 7 0

+ positive

- negative

+ Æ + positive in 1990 and still positive in 2001/2

+ Æ - positive in 1990 and negative in 2001/2

-Æ - negative in 1990 and still negative in 2001/2

-Æ + negative in 1990 and positive in 2001/2

44

3.6 DISCUSSION; IMMUNOLOGY RESULTS

3.6.1 Background to determining the immunological response to grain dust and its contaminants

Antigen production - Although there are several methods available for the production of microbial antigens, there is not, as yet, a generally accepted, standardised method for the growth and production of a particular antigen. During antigen production there are many variables. Fungi are very complex. They are able to grow under different and continuously changing nutritional and environmental conditions. Each species is made up of many strains. Different isolates of the same species can produce allergen extracts with different properties and characteristics. Each micro-organism has different optimum growth conditions. The amount and types of specific antigens produced can vary with culture and extraction methods and from batch to batch. Growth factors, such as temperature, time, oxygen availability, size of inoculum, and composition of the media including pH, carbon and nitrogen sources, are important in determining allergen content. Standardisation and quality control are not used between commercial producers or individual laboratories. Commercial extracts can be very variable in quality. Aas et al. (1980) investigated several different Cladosporium herbarum extracts and found that they possessed widely differing properties, including percentage of protein content and antigen presence, resulting in differing RAST results as well as differences in precipitates in crossed immunoelectrophoresis with antiserum. Extracts produced from the same strains of an organism, grown under identical culture conditions and for the same length of time, can vary in potency. Yunginger et al. (1976) studied 12 commercially available Alternaria extracts using the RAST method. They found that there was no relationship between the manufacturer's designation of concentration and actual allergenic potency. Wardrop et al. (1977) studied 36 Scottish Dairy farmers and found that 18% of their population were positive to commercial Faenia rectivirgula extract but 29% were positive when local strains of Faenia rectivirgula were used.

For the above reasons the microbial isolates used in antigen extract production for use in this study were those obtained from the environments in which the study population were exposed rather than commercial preparations. The antigen extracts were prepared from the micro-organisms obtained during the microbiological survey. The methods used in extract production were those previously optimised and validated in studies at HSL (unpublished data).

The fungi were grown in bulk culture using modified Czapek Dox liquid medium. This is a synthetic and protein-free medium specially developed for use in the production of fungal antigens. The actinomycetes for Phase 1 were also grown in a specially developed synthetic medium (Kurup and Fink, 1977), for Phase 2 commercially produced quality controlled extracts became available and these were purchased from Microgen Bioproducts Ltd. as very few actinomycetes were isolated from the aerobiological study. During dialysis the majority of any components of the media left in the antigen solution are removed. This avoids contamination of the final product with components of the growth medium. Fungal and actinomycete cultures were left to stand, enabling the growth of a mycelial mat on the surface of the culture to increase spore production. It is important to have as wide a range of antigens as possible in the extract, including spore antigens, as spores are inhaled in bioaerosols as well as mycelia.

Wheat and corn/maize were used for the grain extracts because it is well documented that wheat, rye and barley allergens cross-react (Sandiford et al., 1995; Sanches-Mange et al., 1992; HSL unpublished data).

45

The Control population - The non-occupationally exposed control population were included here to compare results with occupationally exposed workers and calculate ‘cut offs’ above which a worker is deemed positive. This control population was required because the antigens used in this study are all commonly occurring antigens that many people come into contact with on a daily basis. The influence of non-occupational exposure can be compensated for, as far as possible, by comparing the occupationally exposed population with a non-occupationally exposed population, assuming that both populations are similarly exposed outside the workplace (Eduard, 1997). Ideally the non-occupationally exposed population should be matched with the study population as much as possible for socio-economic factors, age, sex, smoking etc. The non-occupational exposures for the two populations can never be guaranteed to be exactly the same, however, if statistically significant, well matched study populations are used, the increased levels of specific IgG and IgE in the occupationally exposed population can be said to be due to occupational exposure to the specific antigens. The controls used in Phase 1 of the study were workers in an electronics factory recruited for a separate study and very little serum was available. This has been remedied in Phase 2 of the study with a matched control population from two sources. Firstly dock workers who were not occupationally exposed to grain dust were recruited and secondly hospital porters were recruited. VacutainersTM

(Beckton Dickinson) were introduced to facilitate the collection of larger volumes of serum from all these workers.

Other studies have also compared occupationally unexposed populations with occupationally exposed populations e.g. Godnic-Cvar et al.(1999) while investigating the immunological status of brewery workers. Again, common environmental allergens of barley and fungi were used. Aspergillus and Penicillium spp. are both common in the general environment but Avila and Lacey (1974) found higher IgG antibody levels to these antigens in their exposed population than in the non-occupationally exposed control population. Park et al. (1998) found non-occupationally exposed subjects with specific IgG to corn dust.

Observations and trends- Not all the volunteers took part in the immunology study and of those who did not all gave blood every year. The numbers involved are too low to be statistically analysed however many observations and trends can be noted looking at the overall populations. 11 dockworkers and 11 farmers took part in both the Phase 1 immunology study and the final year of the study (2001/2) these workers are referred to here as the ‘stayers’.

3.6.2 IgG Results

The workers involved in this study were shown in the microbiological study to be exposed to a large number of airborne micro-organisms. The development of specific IgG antibodies in response to the grain dust associated antigens is to be expected in this population.

All of the populations involved had a very high prevalence of positive IgG to the extracts tested, including the non-occupationally exposed control population- in fact in the final year (2001/2) a higher percentage of the control population tested positive than the occupationally exposed populations. The increase in responses to the allergens could be linked to increases in exposures in general for example due to the weather conditions in those years. This reflects the fact that these are all commonly found environmental antigens to which all populations are exposed on a daily basis. This was found to be particularly the case for the bacteria in Phase 1 and so they were not included in Phase 2 of the study.

Dockers: Although throughout the study some dockers did develop new IgG positive test results and the overall response to the work related allergens actually increased peaking at 39.3% in 1998, the general trend amongst the ‘stayers’ was a decrease in IgG positive tests across Phase 1 and Phase 2.

46

Farmers: Amongst the farmers population IgG response peaked at 41.9% in 1999/2000, there was an overall decrease in IgG positive responses to 25.4% by 2001/2. In the overall farmers population the numbers who increased their positive status were very similar to the numbers who decreased their positive status. One of the eleven farmer ‘stayers’ developed an IgG positive status across Phase 1 and Phase 2.

Controls: In the control population the overall response to the work related allergens also increased peaking at 35% in 2001/2. During Phase 2, three controls developed new IgG to the work related allergens but four lost their positive status.

The prevalence of positive IgG to Aspergillus fumigatus was higher in the occupationally exposed population. Farmers had a slightly higher prevalence of positive IgG to grain, ‘field’ fungi and ‘storage’fungi including A. fumigatus.

The control population had a slightly higher prevalence of positive IgG to grain and in 2001/2 the control population had the highest prevalence of IgG positives to Penicillium spp., again reflecting the common environmental presence of these allergens.

In this study population no relationship was identified between a positive IgG to any extract and symptoms (see section 4; Clinical study for more details).

As discussed below, many studies, have shown that development of IgG antibodies is not an indicator of disease development.

In Phase 1 large numbers of subjects were positive to IgG for the bacterial extracts, this reflects the fact that levels of these bacteria in the grain dust to which the workers were exposed were very high. However, the levels of bacterial specific IgG in the control population were very close to that in the study population. This reflects the fact that these are commonly found environmental bacteria to which the general population are exposed in daily life. A larger control population would be required to increase the accuracy of the ‘cut offs’ and enhance the significance of the results unfortunately despite additional recruitment amongst hospital porters this was not possible.

3.6.3 Comparisons of the IgG results from this study with other studies

In early studies, IgG was detected non-quantitatively using the double diffusion method of precipitating antibody detection. More recently ELISA-based methods have been used. These are more sensitive and give better quantification. Few studies have investigated IgG levels in workers exposed to grain dust; most of these investigated the presence of IgG specific to one particular fungus or actinomycete, using double immunodiffusion (DD).

Using DD, Darke et al. (1976) found that 71% of 34 farm workers were IgG positive to various fungi. They found that 64% of workers with symptoms, and 74% of those without symptoms, showing positive precipitins to one or more fungal antigens. There were no precipitins to A. fumigatus or actinomycetes but precipitins to Verticillium, Aphanocladium and Paecilomyces spp. were found. Wardrop et al. (1977) studied 36 Scottish dairy farmers and found 42% were positive to various fungal and actinomycete antigens, mostly Penicillium spp. and Streptomyces spp. There was no correlation between disease and seropositivity. Malmberg et al. (1985) found that 16% of 80 Swedish farmers were positive to various fungal and actinomycete antigens.

Some studies have used ELISAs to measure specific IgG in farmers. Schønheyder et al. (1985) investigated serum levels of IgG specific to A. fumigatus in 182 Danish farmers and 105 controls. They found that farmers with high antibody activity tended to have fewer respiratory symptoms than

47

farmers with lower antibody activity. Hjort et al. (1986) investigated serum levels of IgG specific to A. umbrosis in 181 Danish farmers, 137 farmers' spouses and 104 male controls. Using ELISAs higher levels of IgG antibodies were found in farmers than in their spouses and non-farming controls. No correlation could be demonstrated between antibody levels and respiratory symptoms or lung function parameters. Katila et al. (1986) revisited a population of dairy farmers and investigated levels of IgG to M.faeni, T. vulgaris, A. umbrosis and A. fumigatus they found that changes in antibody titres were of more diagnostic use than the antibody level itself. Increased titres occurred in farmers with continuous exposure and were associated with the appearance of symptoms in previously symptom-free individuals. Park et al. (1999) investigated specific anti-graindust extract IgG levels in 43 grain workers and 27 non-exposed controls and found that there was no association between the prevalence of specific IgG antibodies and the presence of respiratory symptoms. They also concluded that production of IgG to grain dust allergens is a response to grain dust exposure and not related to development of respiratory symptoms. It is not possible to compare directly the results reported here with the above studies of specific IgG in grain workers as different methods and allergens were used in each study.

The immunological studies of farm workers reported in the literature have concentrated on antibodies specific for fungi and actinomycetes. Dutkiewicz et al. (1985) investigated sensitisation to Enterobacter agglomerans in two grain workers both of whom were positive for precipitins.

Eduard (1997) estimates that IgG antibodies can be detected in response to respiratory exposure to 105

viable spores and 104 non-viable spores. The results from the airborne microorganism survey reported here showed that the workers involved in this study were frequently exposed to more than 105 viable spores.

3.6.4 IgE Results

Exposure to grain dust and the antigens associated with it varies throughout the year, with farmers in particular, handling both freshly harvested grain and stored grain. This exposure to the full range of grain dust contaminants was reflected in the development of antibodies to the antigens studied in all three of the worker populations.

Dockers: There was an overall trend towards a decrease in positive IgE status among the dockworkers. No dockworkers developed sustained positive levels of IgE during the study, 2 who were negative in Phase 1 tested positive midway through Phase 2 but returned to their negative status in their final samples. Two atopics lost their positive status (one to fungi and actinomycetes and one to storage mites and actinomycetes, this worker also tested positive to fungi in 1999).

Farmers: There was an overall trend towards a decrease in positive IgE status among the farmers. Five farmers developed sustained positive levels of IgE during the study (3 to storage mites, 1 to grain and 1 to fungi), five who were negative in Phase 1 tested positive midway through Phase 2 but returned to their negative status in their final samples. Seven lost their positive status (2 to storage mites, 3 to grain and 2 to fungi).

Eleven of the 2001/2 farmers were ‘stayers’. Two of these eleven tested in 1990 were atopic, one atopic was IgE positive for storage mites in 1990 and not in 2001/2 one atopic tested positive for grain throughout the study.

Atopic workers - All the controls with positive IgE responses to the work related allergens were also atopic. All the dockworkers in Phase 2 with positive IgE responses to the work related allergens were also atopic.

48

Non atopic farmers tested positive to fungi, grain and storage mites, a higher percentage of nonatopics than atopic farmers tested positive to storage mites.

In Phase 1 atopic dock workers and farmers were more likely to have positive levels of IgE to one or more grain dust allergen disc, atopy is not a prerequisite, non-atopic farm workers did also have positive levels of IgE to grain dust allergens. In Phase 1 there was no significant difference in IgE levels between the control and test populations, again reflecting the fact that these are common environmental antigens. However, the presence of positive IgE to grain dust-associated allergens indicates that allergic mechanisms could be the cause of symptoms in some of the workers. In Phase 1 a higher percentage of non-atopics tested positive to work related allergens than did in Phase 2. In Phase 2 there were no non-atopic dockworkers who also had positive levels of work related IgE, however, up to 11.1% of non-atopic farmers had positive levels of work related IgE.

IgE positive status to grain was most prevalent in the dockworker population followed by farmers and then controls. IgE positive status to storage mites was most prevalent by far in the farm worker population.

3.6.5 Comparisons of the IgE results from this study with other studies.

It is not possible to compare directly the results reported here with other studies of specific IgE in grain workers as different methods and allergens were used in each study. However, several studies have found that production of specific IgE to occupational allergens does not explain all respiratory disease in farmers exposed to organic dusts. The studies carried out by Spiewak et al. (2001), Darke et al. (1976), Lewis et al. (1986) and Park et al. (1998) had some areas of commonality with the study reported here and are discussed below.

In a previous study, Blainey et al. (1989), found that 25% of grain store workers had positive skin prick tests to storage mite and 10% to grain. In RAST tests 22% of grain workers were positive to storage mite and 10% to grain. The results reported here are lower 3.7 - 18.4% of workers were positive to storage mite and 2.4 - 13% to grain. In Phase 1 a significant association was found between work-related symptoms and specific storage mite IgE, but not between work related symptoms and positive skin tests to grain allergen. Spiewak et al. (2001) carried out skin prick tests in 73 Polish arable farmers and found very similar results to those reported here, 5.6% were positive to grain dust, 19.1% to 1 or more allergens, up to 17.8% to individual storage mite extracts and 4.1% to Aspergillus fumigatus they also found positive tests to bacteria and actinomycetes, 4.1% to Pantoea agglomerans and 4.1% to Saccharomonospora rectivirgula, there were no IgE positives to bacteria or actinomycetes in Phase 2 of the study reported here. Spiewak et al. found that the skin prick results did not correlate well with the work related symptoms. Darke et al. (1976) carried out skin prick tests with various fungal allergens in 34 British farm workers. 35.9% were IgE positive to various fungi, most commonly Verticillium, Aphanocladium and Paecilomyces spp. These were higher levels than reported here but skin prick tests are not directly comparable to RAST S. Lewis et al. (1986) tested farmers with RAST assays and found 3.45% positive to barley dust extract and 5.14% positive to wheat dust extract. This was more comparable to the results reported here. Alvarez et al. (1996) found that 5 out of 6 farmers had positive IgE to wheat extract allergens. Park et al. (1998) studied 43 Korean animal feed workers. He also found, using skin prick tests and an ELISA to measure specific IgE to grain dust, that specific IgE to grain dust was detected in 40% of symptomatic workers and only 11% of asymptomatic workers. This indicates that although immunological response to the grain dust allergens could explain some of the symptoms it does not explain all of them. Like the study reported here, the study by Park et al. (1998) also found that atopic workers were more likely to develop specific IgE to grain dust. Zuskin et al. (1994) also used skin prick tests in their investigation of a range of workers exposed to organic dusts including grain dust. They also found that although

49

exposure to organic dust aerosols may be associated with frequent immunological reactions, the development of specific IgE antibodies does not predict objective respiratory impairment.

Cross-reactivity of allergens - Cross-reactivity between species is a complicating factor in immunological assay; for example Alternaria cross reacts with Ulocladium (Cole and Sampson 1984). Storage mites have been shown to cross-react with house dust mites (Griffin et al., 1989). In the study reported here the majority of workers who were positive to storage mites were also atopic.

Blood Collection - It is preferable, in a survey such as this, to time the collection of blood to during, or as soon as possible after, the activity of handling the grain. This was the case with the dock workers involved with the study who have an almost all-year exposure to grain dust. However, the collection of blood from the farmers took place during their quieter season outside the busy harvest period. IgG and IgE serum antibody levels increase during exposure to high levels of microorganisms but then decrease as the antibodies are removed slowly from the circulation. IgG antibodies have a half-life of 2-4 weeks; IgE of 2-3 days unbound and 21 days when bound to its receptors on mast cells and basophils. Therefore it is best to collect the serum soon after the harvest. However, incidental exposure to small amounts of the microorganisms concerned during other activities can boost the levels of circulating antibody in already sensitised individuals. This study may therefore underestimate IgG and IgE levels in some of the farmers. This underestimation is reduced by the sensitivity of the assays used in detecting tiny amounts of antibody still in the serum.

50

4 CLINICAL STUDY

4.1 INTRODUCTION AND SUMMARY OF PHASE 1

In both Phase 1 and Phase 2 of this study the occurrence of respiratory symptoms and airflow limitation in this cohort of workers was assessed through from a respiratory questionnaire, spirometric tests and analysis of peak flow records.

Phase 1 began in 1990, which established a cohort of workers exposed to grain dust for a significant part of their working lives. Many factors affect exposure levels including the tasks undertaken by the worker, environmental factors such as weather conditions and control measures in place, and the type and quality of stored grain. Groups with different types of grain exposure were therefore included in the 1990-93 study. Farmers have intense exposure during cereal harvesting, and again when the grain is being handled for transportation or for milling animal feed. Dockworkers have a more continuous, though sometimes equally intense, exposure.

The results of this survey of grainworkers from three separate sites were related to the results of environmental surveys of the dust levels and biological constituents of the dust. The main findings in the 321 grain workers (86% of the available workforce) were that FEV1 was reduced by exposure to grain dust, and a high proportion had symptoms (Blainey et al, 1995). Specific IgE to the principal environmental contaminants of grain storage mites, field fungi (e.g. Alternaria) and storage fungi (e.g. Aspergillus spp.) was found in 16.5% of grainworkers; 41% of subjects with positive "work-related" IgE had work-related respiratory symptoms (WRS). The significant relationship between WRS and immunological responses to storage mites and other allergens contaminating grain dust could form part of the explanation for grain-induced lung symptoms, but only a proportion of the symptoms and lung function changes could be explained by these results. The results demonstrated wide variations between individuals, and it was not clear whether some subjects may be predisposed to respiratory effects or whether factors other than individual predisposition and variation in exposure levels accounted for this. Follow up of these workers over two successive years was then undertaken to see if individual predisposing factors could account for symptoms or reduced airway calibre found in some of these subjects.

Twenty one subjects (13%) developed new work-related respiratory symptoms during Phase 1 of the study. These subjects were significantly more likely to be atopic than those with no WRS (46.7% vs. 10.8%) and to have specific IgE to storage mites and to grain. It appeared that atopic subjects may be more susceptible to the effects of grain dust. This did not appear to be a function of exposure levels.

The Phase 1 study showed no evidence of significant change in FEV1 in the study participants over the 3 year measurement period. In those who remained in the Phase 1 study, atopic workers were more likely to develop new work-related symptoms during the study period, but there was no relationship with overall dust exposure levels. The MEL was introduced during the period of the study and came into force after exposure assessments had been made.

4.2 SUBJECTS

4.2.1 Phase 1 (1990/1993)

Volunteers, selected from lists of names and addresses supplied by the National Farmers Union (NFU) for the East Anglian area, were invited to take part from two different dock sites, from a variety of different farms and from animal feed mills in Essex and East Anglia.

51

4.2.2 Phase 2 (1997/2002)

Farms which had taken part in the 1990-93 survey were approached to see if they would take part, and one of the two docks was invited to continue with the study. Because the measured dust exposure and prevalence of symptoms at the mills had been so low in Phase 1, they were not included in Phase 2. The second dock was no longer functioning. 151 out of a potential 320 farm workers, and 33 out of a potential 51 dockworkers agreed to take part.

In the farms, all subjects were likely to undertake any job which involved dust exposure; in the docks, all staff with grain dust exposure were included in the analysis. Adjustments for social class were not made.

The control population, all of whom undertook skilled manual work, was initially recruited from workers in the docks who had no grain dust exposure. However after the first year of Phase 2 it was established that some of these subjects had in fact had significant grain dust exposure in the past. Approaches were made to several employers of skilled manual labour in the region with workforces in excess of 100 employees, but unfortunately none of these employers were prepared to give permission for their workforce to be approached. Estate workers from a NHS Trust in Essex joined this cohort in 2001. None of the control population finally selected had a history of working in the grain industry.

The study was approved by Mid Essex, Southend, Basildon and West Essex LREC’s. Informed consent was obtained from all subjects in accordance with the Local Research Ethical Committee guidelines. Subjects for Phase 2 of the study consisted of dock workers from one dock and farmers.

The break down of the participants taking part in the survey are outlined in table 4.1

Table 4.1 Participants in the Survey

Dockers Mill workers

Farmers Controls Total Remained from previous year (1998-2002)

1990 130 66 125 321

1991 87 37 96 220

1992 78 30 95 203

1998 33 151 30 214

1999 22 120 21 163 147

2000 26 103 13 142 142

2001 16 92 19 127 115

2002 18 81 15 114 112

Altogether, 98 subjects left the study between 1998 and 2002. The reasons that were given, where these could be established, are set out in the table below.

52

Table 4.2. Reasons for leaving the study

Reason for leaving

Number of Farmers

Number of Dockers

Retired 3 1 Illness 2 2 Died 1 1 Moved 3 Left industry

7 5

Redundancy 1 Unable to contact

18 5

Self withdrawal

43 6

4.3 METHODS

4.3.1 Respiratory health questionnaire

A validated self-administered respiratory questionnaire (Venables 1993) was used to assess the presence of cough, wheeze, shortness of breath (SOB) and work-related symptoms (WRS). Additional questions to relate symptoms to workplace exposure, and smoking and occupational history were added to the questionnaire.

The questionnaire was administered under the supervision of one of four trained respiratory nurses. Participant subjects were required to complete the questionnaire themselves.

Definitions of symptoms as obtained from the questionnaire are outlined in Table 4.3

A separate occupational survey was completed by the respiratory nurse to identify occupational history and exposure.

Exposure duration was defined as the number of years employed in the grain industry.

53

Table 4.3 Definition of symptoms

Cough Short of breath Work related symptoms

Cough in the morning in winter Cough during the day or night in winter Cough for more than 3 months per year Cough on exertion

Woken by wheeze Wheeze on waking Wheeze on exertion Wheeze in a dusty place Chest sounds wheezy

Woken by difficulty

in breathing on waking Chest tight

Any of the previous

weekends or

holidays of a week

Wheeze

in breathing Difficulty

felt breathing become difficult

symptoms plus improvement at

improvement on

or more

Subjects were not asked to record if their symptoms deteriorated over the duration of a shift; this could only have been done for the dockers where defined shifts were worked.

Symptoms derived from questionnaires do not necessarily provide a measure of the severity of a health effect, so in order to measure this, the impact of respiratory symptoms on health, the St George’s Respiratory Questionnaire was used (Jones et al, 1992). The SGRQ measures several aspects of the health effect. For the purpose of this study we have used the ‘Activity’, ‘Impact’, and ‘Total’ scores derived from the SGRQ as a measure of the health impact of grain dust exposure. The SGRQ was administered as well as the conventional questionnaire in 2001 and the results from that year have been included in this analysis.

4.3.2 Spirometry

Spirometry measurements were recorded in the standing position using a Vitalograph bellows spirometer with a 12 second carriage. Measurements were recorded in accordance with the American Thoracic Society guidelines (1987). The subjects performed at least three FVC curves with a variation of less than 5%. The best FEV1 value was selected and used for analysis. The highest FVC from a valid curve was selected as the FVC. The results of Spirometry were expressed in % of predicted value for FEV1 and FVC using the reference values of the European Coal and Steel Community (1993 ).

The respiratory nurses undertaking spirometric measures had all received a standardised training. There was personnel change in 1999 and 2001.

Two identical spirometers were used in this study. They were regularly calibrated using a calibrating syringe.

54

4.3.3 Peak flow recording

Subjects were asked to record their Peak Expiratory Flow Rate (PEFR) 4 times a day every day for 4 weeks, using a mini Wright peak flow meter. They were also asked to record if they had worked with grain, time of waking, starting and finishing work, and time of going to bed. Daily variability in PEFR was defined as (maximum PEFR - minimum PEFR)/(mean PEFR).

4.4 RESULTS

4.4.1 Respiratory Health Questionnaire

The age, occupational histories and grain exposed employment duration were not significantly different between employment groups (see Table 4.4).

Table 4.4 Age and duration of exposure

Mean age Mean years exposure

% smokers Male Curr ent

Ex non

Dockers 48.18 16.9 21.4 42.8 35.7 Farmers 46.07 25 18.4 19.7 61.7 Controls 47.5 - 8.1 43.4 48.6

There were no significant differences in age between dockers, farmers and controls, or in duration of exposure.

Although there were fewer current smokers among the controls, the overall consumption as measured by pack years was not different between the occupational groups as Figure 4.1 below demonstrates.

55

Figure 4.1 Smoking in pack years in relation to occupation

0

5

10

15

20

25

pack years

ls

Ex

ll

Dockers Farmers Contro

smoking pack years

Current overa

The prevalence of symptoms was similar for each year. The percentage prevalence of symptoms for

each symptoms as defined above and for each year is shown across the whole group surveyed in

Figure 4.2. This graph represents the point prevalence in the whole group included in the survey in

each year. The prevalence of symptoms in the subjects who remained in the study in all years is shown

later.

60.00% %

prevalence of symptom

s50.00%

40.00%

30.00%

20.00%

10.00%

0.00%

Cough Wheeze SOB WRS

1990 1991 1992 1998 1999 2000 2001 2002

year of study

Figure 4.2 Prevalence of symptoms across the whole study group

56

There were no significant differences in the prevalence of each individual symptom between each year, and no significant trend over the ten years of the study period. In particular there was no significant difference between the prevalence of symptoms in Phase 1 (1990 – 93, before and encompassing the introduction of the MEL) and Phase 2 (1998-2002).

Wheeze was the commonest symptom recorded, followed by SOB, then cough, then WRS.

The breakdown of symptoms in relation to atopy is shown in Table 4.5. Although there are clear differences between the different groups, the numbers are too small for there to be any significant differences by contingency table analysis. This is also shown graphically in Figure 4.3. The table and graph express the relationship between atopy and symptoms in different ways. In the table, the proportion of the total group who have the symptom and are atropic or non-atopic is shown. In the graph, the proportion of subjects with each symptom who are atopic or non atopic is shown. Thus for example, the table shows that 7.4% of atopic dockers reported wheeze, and 11.1% of non atopic dockers. The graph shows that of the dockers reporting wheeze, 40% were atopic.

Table 4.5 Prevalence of symptoms in occupational groups in relation to Atopy

Occupational

group

Atopic or non

atopic

Percentage

with Cough

Percentage

with Wheeze

Percentage

with SOB

Percentage

with WRS

Controls Atopic 5.71 5.71 2.86 8.57

Non Atopic 20.0 20.0 2.86 8.57

Dockers Atopic 11.11 7.40 0 11.1

Non Atopic 29.6 11.1 3.70 25.93

Farmers Atopic 5.84 7.30 6.57 5.84

Non Atopic 26.2 16.8 4.34 9.50

It was observed that for the cough and wheeze, the proportion of non atopics with symptoms was higher, but the differences were not significant at the 5% level.

57

prevalence of atopy in symptomatic subjects in each occupational group

60

50 40

% a

topi

c Controls 30 Dockers

20 Farmers

10

0 Cough Wheeze WRS Wheeze

dust

Figure 4.3. Prevalence of atopy in symptomatic subjects in each occupational group

There were 63 subjects who only remained in Phase 2 of the study for 1 or 2 years. Of these 11 were controls, 4 were dockers and 48 were farmers. The reasons for leaving where these could be established are given in table 3.2. Analysis of the prevalence of symptoms in those who did not remain in the study compared with those who did showed no significant differences for any symptom, including work related symptoms and symptoms of chest tightness or wheeze on exposure to grain dust.

4.4.2 The Health Impact of Respiratory symptoms - SGRQ scores

A total of 97 subjects completed the SGRQ questionnaire. Scores for activity, impact and the total score are given below for the symptoms derived as described above in Table 4.6

Table 4.6: Health Impact of symptoms

Whole group

Cough +ve* Dyspnoea +ve*

WRS +ve* +ve*

97 15 6 25 13 0 1.63 10.99 1.63 11.47

Activity score

5.96 11.21 12.17 12.17 12.17

6.24 12.15 14.9 9.83 17.79

Wheeze

Number Impact score

Total score *Symptoms derived from Venables Questionnaire

The median Total SGRQ score was 4.97 in non-smokers, 5.93 in current smokers and 8.09 in ex smokers. The median Total score in subjects with abnormal reduced airway calibre (FEV1<80%

58

predicted) was 7.89, compared to 5.17 in subjects with FEV1>80% predicted. These differences are not significant. However there was a significant relationship between the SGRQ score and FEV1% predicted (correlation coefficient –0.283, p=0.014) but not for FVC %.

This relationship would be expected as a low FEV1 is a marker for chronic respiratory disease (Barnes P, 2004). However as shown below the absolute prevalence of low FEV1 is low in this study.

These scores have been compared with published scores in other respiratory diseases below in Figure 4.4.

0 10 20 30 40 50 60

l

SGRQ scores

Tota Impact Activity SGRQ score type

SGRQ scores in relation to other respiratory diseases

Grain WRS Mild asthma COPD

Figure 4.4. St Georges Questionnaire responses in subjects with Grain WRS in this study

compared with subjects from other studies with mild asthma, and with COPD

The scores relating to Mild asthma and COPD are derived from the use of this questionnaire in defined populations (Jones, 2001). The comparator populations are clearly not matched for health, but the use of this measuring tool can provide an indication of the global impact on health of these occupational symptoms.

This graph shows that the impact of work related symptoms in this group is low by comparison with those recorded in other respiratory diseases. This group of grain workers as a whole did not report poor health status as a result of their symptoms. Even in those with work-related symptoms the overall impact was low. It would appear that the occupational symptoms recorded by these grain workers do not represent a major adverse health impact from grain dust.

4.4.3 Immunological results in Phase 2; relation to symptoms

The prevalence of sensitisation to common inhalant allergens (atopy), storage mites, environmental fungi, and grain dust are outlined in Table 4.7. The relationship between individual symptoms and

59

RAST tests for the principal positive specific IgE results are given in the table below. These results were similar for all years of Phase 2 of the study.

Table 4.7: relationship between positive IgE for Atopy, Mites and grain and individual

symptoms: Odds ratios and 95% confidence intervals (significant results in bold)

Question Atopy Mites Grain Cough in morning in winter?

1.037 0.996-1.078

1.015 0.979-1.050

0.99 0.95-1.03

Cough in day or 1.074 .895 0.171 night – winter? 0.83-1.39 0.36-2.23 0.116-0.254 Cough > 3 months per year?

0.506 0.28-0.914

0.857 0.403-1.822

2.32 1.07-5.022

Cough on exertion?

1.09 0.65-1.84

0.69 0.25-1.93

0.15 0.1-0.23

Wheeze on exertion?

1.92 1.03-3.54

1.27 0.64-2.52

0.77 0.39-1.52

Sleep broken by wheeze?

5.62 1.91-12.6

1.58 1.38-1.8

0.45 0.14-1.43

Wheeze on 1.04 1.02 0.99 waking? 0.99-1.09 0.98-1.06 0.95-1.03 Wheezing in a 1.34 2.87 3.1 dusty place? 0.77-2.33 1.29-6.36 1.24-7.75 Chest sounds 1.095 0.958 0.944 wheezy? 0.625-1.917 0.461-1.991 0.454-1.961 DIB on waking 1.04 1.02 0.99 up? 0.99-1.09 0.98-1.06 0.95-1.03 Chest feels tight 0.831 0.98 1.21 / DIB? 0.474-1.458 0.471-2.039 0.579-2.52

The strong relationship (Odds ratio and 95% confidence limits < >1) noted between a positive specific IgE to occupational allergens and the symptoms ‘wheezing in a dusty place’ strongly supports the role of these allergens in causing symptoms.

Statistical analysis by contingency table analysis of the relationship between symptoms and the prevalence of positive IgG responses to aerobiological materials encountered in the workplace showed only one significant relationship. The prevalence of specific IgG to any of the occupational agents was lower in subjects who reported that their symptoms were improved at weekends or holidays (i.e. work-related symptoms). Only 6.1% of subjects who reported that their symptoms improved at weekends, for example, had a positive IgG response, and 92.7% of subjects with work related symptoms improving on holidays of a week or more had negative IgG responses. No other symptoms had any relationship with IgG by contingency table or logistic regression analysis.

60

4.4.4 Relationship between immunological findings and smoking, duration of exposure and age

Logistic regression was used to examine the relationship between immunological findings and age, smoking and duration of exposure. No relationship was observed for atopy, or grain or fungi specific IgE, but a positive IgE to storage mites was correlated with both smoking pack years and duration of exposure. Subjects with a positive RAST to mites were less likely to be smokers, and had a shorter duration of exposure (p<0.05).

4.4.5 Spirometry results from Phase 2

Results are summarised in Table 4.8.

Table 4.8 Spirometric results at all 5 measurement points in Phase 2 of the study.

Year 1998 1999 2000 2001 2002

Number of subjects 204 162 141 126 112

FEV1 absolute (Litres) 3.866 3.822 3.767 3.766 3.868

FEV1 % predicted 100.83 101.02 99.73 100.5 104.15

FVC absolute (Litres) 4.961 4.93 4.89 4.94 5.16

FVC % predicted 104.8 105.4 104.7 106.85 111.87

FEV1, and FVC both expressed as a percentage predicted values, is illustrated in Figure 4.5, with standard errors as illustrated. There was no significant difference between each year. Although both FEV1 and FVC (Figure 4.6) showed a decline in mean value between year 0 and year 2, and an improvement in year 4, this difference was not statistically significant.

61

FEV1 % 1998 - 2002 Means & Intervals

97.5

100.0

102.5

105.0

107.5

110.0

FEV% 98 %FEV1 99 %FEV 00 %FEV 01 %FEV02 Std ErrMean1.96*Std Err

Figure 4.5. FEV1 % predicted Phase 2 (1998 –2002)

Means & Intervals

102.5

105.0

107.5

110.0

112.5

115.0

117.5

FVC % 1998 -2002

FVC %99 %FVC 99 %FVC 00 %FVC 01 % FVC 02 Std ErrMean1.96*Std Err

Figure 4.6. FVC % predicted Phase 2 (1998 – 2002)

The FEV1 and FVC absolute in those 96 subjects who had measurements in all years of the Phase 2 study is shown in Figure 4.7. As in the group as a whole there was a noticeable decline in 2000 but there was no statistical significance to this trend.

62

1998 1999 2000 2002

FEV1 5

FVC

1

3.6 3.65 3.7

3.75 3.8

3.85 3.9

3.95

2001

Year of study

4.75 4.8 4.85 4.9 4.95

5.05 5.1 5.15 5.2

FEV

FVC

Figure 4.7 Absolute FEV1 and FVC in subjects who remained in Phase 2

4.4.6 Abnormal lung function

A significant reduction in airway calibre is often defined (arbitrarily) as 80% of predicted values. The prevalence of abnormal lung function is shown in Table 4.9

Table 4.9 The prevalence of abnormal lung function

Year Dockers Farmers Controls

1998 <1% 5.41% 0

1999 <1% 6.17% 0

2000 <1% 4.25% 0

2001 0 4.76% <1%

2002 <1% 5.36% 0

The prevalence of abnormal lung function in atopic subjects is shown in Table 4.10. Overall, atopics made up 20% of the study population.

63

Table 4.10 The prevalence of abnormal lung function in atopic subjects

Year Atopic Non-Atopic

1998 6.86% 5.71%

1999 6.79% 6.19%

2000 4.96% 3.13%

2001 5.55% 6.33%

2002 6.25% 6.85%

4.4.7 Spirometry and symptoms -1998

Table 4.11 shows the relationship between FEV1 and FVC, as percent predicted, and symptoms, in the first year of Phase 2 of the study. Similar results were obtained for all other years.

Table 4.11 relationship between symptoms, FEV1 and FVC

Subjects with Cough

Subjects with

Wheeze

Subjects with SOB

Subjects with WRS

FEV1 as % predicted

96.61% 98.25% 98.94% 96.15%

FVC% as % predicted

102.21% 99.26% 104.14% 98.98%

4.4.8 Change in lung function over the course of Phase 2

The change in lung function over the course of Phase 2 has been analysed as below. All subjects who had at least 3 separate points of acceptable spirometric recordings were included in this analysis.

There were 147 subjects who fulfilled these criteria. The overall annual decline in FEV1 over the 4year period of this study was 26.02 mls/year. There was no significant difference between farmers and dockers, (∆FEV1 20.05 ml/yr) and controls (∆FEV1 66.27 ml/yr), and no significant relationship between the level of decline in FEV1 and the prevalence of symptoms (WRS, cough, wheeze, SOB). Neither was there any significant relationship between the results of immunological tests (a positive to atopy, or a positive work-related IgE) and the level of decline in lung function.

A comparison of the mean annual decline in FEV1 over the course of Phase 2 of the study is shown in Figure 4.8 below in relation to the number of years spent in the study.

64

It is apparent from this that the smallest decline in FEV1 is experienced by those subjects who spent the greatest time in the study, and that those in Phase 2 who only had 2 or 3 years of recording had a greater decline in FEV1. The difference between the extremes is statistically significant. Those who completed only 2 years of the study �FEV1 was 47.08 ml/yr, but �FEV1 was 14.33 ml/yr in those who completed all 4 years. This difference was statistically significant.

Mean annual change in FEV1 (mls)Means & Intervals

-100

-50

0

50

100

DFEV1 2 yr DFEV1 3yr DFEV1 4yr Std ErrMean1.96*Std Err

Figure 4.8. Mean annual change in FEV1 (ml)

It is to be expected that the measurement variation over 3 measurements will be higher than over 5 as the standard error bars on the above graph confirm. It therefore remains possible that those with the shortest duration of exposure to grain dust had more variability in the lung function measurement than those with longer duration of exposure.

The relationship between selected symptoms as measured at the outset of Phase 2 of the study, and levels of annual decline in FEV1 (∆FEV1) is shown below in Figure 4.9. There is no suggestion that the presence of symptoms at the outset of the study is associated with a more rapid decline in FEV1, as the prevalence of symptoms was the same regardless of the change in FEV1.

65

0

10

20

30

40

50

)

<30 >30 >50 >100 change in FEV1( mls/yr

prevalence of symptoms in relation to change in FEV1

WRS Cough Wheeze

Figure 4.9. Prevalence of symptoms in relation to change in FEV1

Similar findings were evident in the analysis of change in FEV1 over the much shorter period of Phase 1 of the study. Although only 3 recordings points were available in this study, a comparison could be made of the change in FEV1 in those who had spirometry at all three points and were included in both Phase 1 and Phase 2, and those who had 3 recordings but only remained in Phase 1. Of 163 subjects who fulfilled these criteria; 49 were ‘stayers’ and 114 non stayers. The change in FEV1 in ‘stayers’ was +12.25 ml/yr, but in non stayers �FEV1 as –15.26 ml/yr.

These results suggest that workers who do not remain in the industry are likely to have a higher rate of decline in FEV1, and may be part of the explanation for turnover in the industry. However as this was not related to the presence of atopy or specific IgE to work related allergens, it may be the nonspecific effect of irritant dust exposure that is responsible – perhaps mediated through endotoxin.

4.4.9 Spirometry, duration of exposure and smoking

In Phase 2 of the study, there was no relationship between the duration of exposure to grain dust and FEV1 or FVC as expressed as percentage of the predicted value (Correlation coefficient –0.13, p>0.05). Smoking showed the expected relationship (on stepwise multiple regression) with FEV1 as percentage predicted, but there was no relationship with FVC % predicted.

Neither smoking in pack years or duration of exposure independent of age was found to have any significant influence on �FEV1. However �FEV1was highly correlated with absolute and percentage predicted FEV1(correlation coefficient –0.38, p<0.01)

66

4.4.10 Peak flow data

A total of 95 farmers completed at least 4 weeks PEFR recordings. There were 601 days (20.24%) recorded working with grain, 2368 days (79.75%) away from grain. Mean daily variability (MDV) with grain was 8.82%, variability away from grain was 7.74%. 13(14%) reported WRS and 14(15%) were current smokers. Analysis demonstrated that both current smoking and work related respiratory symptoms (WRS) influenced MDV with highest MDV in subjects with WRS when working with grain (10.79%). However current smoking appeared to reduce MDV when working with grain (5.47%) see Table 4.9.

More detailed analysis of daily peak flow recordings (PFR) in 77 of these grainworkers identified different patterns of responses to grain dust. Records for 2357 days were analyzed; 30% of days included grain dust exposure and 70% did not. To allow for diurnal variation, records were normalized for a mean baseline taken as the first PFR within 1 hour of waking and expressed as change from baseline for each day and each individual.

PFR were compared on days with and without grain dust exposure. The expected diurnal PFR rise occurred later on grain- exposed days. PFR at 3-10 hours rose by up to +3.5% irrespective of grain dust exposure but by 10 hours was –0.2% from baseline on grain exposed days compared to +2.64% without grain exposure (p=0.006). In subjects with WRS, PFR was –9.35% from baseline at 12 hours on grain exposed days, vs +2% in subjects with no symptoms.

An analysis of the 1998 data using the OASYS system for detecting occupational asthma was undertaken by Burge and co-workers (Anees et al, 2002). 61% of the workers in the first year of the Phase 2 study returned records which were analysed by OASYS, but only 70% of these had adequate data for OASYS to interpret. 9 subjects had records fulfilling the criteria for work-related deterioration in PEF, but only 1 had work related symptoms suggestive of occupational asthma. There were 7 subjects with questionnaire-based symptoms but a negative OASYS PEF analysis. There was no relationship between positive immunology and an OASYS score of > 2.5.

This demonstrated that there was little overlap between occupational asthma as identified by the OASYS system, and symptoms or immunological sensitisation in these subjects. It was acknowledged in this study, however, that the discontinuous nature of exposure in the workers participating in this study may have reduced the ability of OASYS –2 to detect occupational asthma. Of importance to the interpretation of effects in the respiratory tract in this study, however, is the demonstration that variable airflow limitation with a work-related effect is found independently of ‘work-related’ symptoms, or sensitisation. This suggests that a mechanism other than occupational asthma due to sensitisation to occupational allergens may be responsible for variation in airflow calibre in some subjects.

It is possible that the variability in airway calibre attributable to grain exposure and associated with symptoms, which are work related but not necessarily symptoms of asthma, is due to endotoxin exposure. One way to determine if this is the case would be to examine patterns of respiratory symptoms for a ‘Monday morning’ effect, i.e., a short term worsening of symptoms on re-exposure followed by progressive improvement with continued exposure. This has been demonstrated in workers exposed to high levels of endotoxin in the cotton industry (Fishwick et al, 2002). The grain study data have not at this stage been examined to determine if this effect may have occurred, but further interrogation of the data is planned and may be the subject of later publications.

67

4.4.11 Subjects common to phase 1 and phase 2 (‘stayers’)

Introduction - 58 subjects who had taken part in Phase 1 of this study were included in Phase 2. This represented 22.8% of the 254 subjects from farms and docks who had taken part in the first year of the Phase 1 study, though the major difference in subject numbers took place between years 1 and 2 of Phase 1. As the original report showed, there was a dramatic reduction in the numbers of dockworkers between years 1 and 2 of the Phase 1 study.

The lack of continuity is explained by major restructuring of both arable farming and the docks between 1990 and 1998. Agriculture continues to decline as an employer, and the docks experienced a major restructuring between 1990 and 1993, with a >50% reduction in the workforce.

Analysis of the results from 1990 of those who remained may be useful if the measurements made in 1990 can be used to predict the likelihood of leaving the industry. 24 were dockworkers 35 were agricultural workers

Symptoms - The percent prevalence of symptoms in those who stayed compared to those who left is shown in Table 4.12. The recording of symptoms in this table was made at entry to the study in 1990.

Table 4.12 Prevalence of symptoms

Cough Wheeze Breathlessness Work-related symptoms

Stayers 22% 34% 10% 18% Non stayers 38.6% 43.96% 8.7% 22.7%

Cough was significantly more common in non stayers, but the prevalence of the other symptoms was not significantly different between the two groups.

Lung Function - No significant differences in mean age, FEV1 (absolute or % predicted) FVC, duration of employment or smoking history were found (Table 4.13). However, abnormal lung function (FEV1<80%) prevalence, measured at entry into the study in 1990 was 6.12% in stayers but 18.09% in non stayers. This difference was significant at the 5% level by contingency table analysis.

Table 4.13 Comparison between subjects who stayed in the study and those who left.

Age FEV1 absolute FVC absolute Employment duration

FEV1 % predicted

Mean Stayers 41.65 3.81 5.12 17.91 yrs 95.31% Mean non

stayers 43.94 3.56 4.78 18.52 yrs 90.55%

Immunological results - 21.4% of the leavers were atopic compared with 33.3% of the stayers The presence of any occupational allergen was higher in the group that stayed (21.2%) than in the non stayers (13.38%), but these differences were not statistically significant by contingency table analysis. There were no significant associations between immunological sensitisation and remaining in, or leaving, the study.

68

The change in FEV1 over 12 yrs in the subjects who remained was a mean of -0.07 L and change in FVC of - 0.08 L over the 12 years. This average of 6 mls/year is considerably less than expected, as a decline of 20ml/yr was identified in a healthy control population by Chan-Yeung as well as in healthy populations (Chan-Yeung et al, 1981,1992; Barnes P, 2004). The small negative direction of change indicates a healthy worker effect). In longitudinal studies of Canadian grain workers (Chan-Yeung, 1981, Huy et al, 1981) grain workers had a higher rate of decline in FEV1 than control workers after 23 years but there was no difference after 12 years of follow up. As in the present study, those who left between survey points in that study were more likely to have symptoms. However other studies (Pahwa et al, 1994) have demonstrated that in some grain workers, the annual loss of FEV1 can be related to duration of exposure to grain dust. Even this study, however, acknowledged that the loss of FEV1 levelled off during the later years in the grain industry, supporting the hypothesis that those with the greatest decline in FEV1 are most likely to leave the industry.

Eleven dockers and eleven farmers who remained in the study had a decline in FEV1 of >250 ml over 12 years (‘rapid decline’). In other studies (e.g. Enarson et al, 1985) where a rapid decline in FEV1 in some grain handlers has been noted (in this study >100 ml/year), rapid decline was associated with higher dust exposures. It was not possible in this study to attribute individual workers to a high or lopw dust exposure category because of the variability of exposure. We have examined whether these subjects had any characteristics which might have predicted a more rapid decline in FEV1.

The prevalence of atopy in the subjects with rapid decline in lung function was higher than in those without. 6/22 (27%) in those with had rapid decline compared to 5/36 (13.9%) in subjects without. This difference (X2 = 1.59) is not statistically significant although the numbers are small. A positive test to storage mites is also not significant (a prevalence of a positive test of 27.2% in subjects with a rapid decline compared to 17.4% in subjects with <250 mls decline in FEV1). A positive specific IgE to any occupational allergen was the same in both groups – 26.7% positive in rapid decline but 18.7% positive in slow decline (no significant difference).

Work-Related Symptoms or chest symptoms with grain dust exposure at the start of the study do not identify those with rapid decline. 35.2% with rapid decline reported WRS in 1990 compared to 29.4% in those with <250ml decline in FEV1 over 12 years.

There are therefore no features other than cough which identify those less likely to remain in the study. Rapid decline in FEV1 does not necessarily result in leaving the industry, and the prevalence of sensitisation to environmental agents was no different in those who stayed compared to those who remained.

4.4.12 Conclusions – Stayers study

1. A 'survivor' population of 58 grain workers has been identified who have been studied for more than 8 years. Although directly comparable measurements could not be made in those who left the study, those who remained appeared to have a lower mean annual change in FEV1 than subjects who left the study after 2 years or less.

2. The principal reason for large changes in the study population appeared to be industrial restructuring.

3. Cough and impaired lung function at the outset appeared to be associated with a higher probability of leaving the grain industry.

4. Age, employment duration, atopic status and the presence of positive responses to occupational allergens were not predictors of leaving the study.

69

4.5

5. These findings indicate a healthy worker effect with a lower decline in FEV1 than would be expected in the general population. Subjects who did not remain had more variable airway caliber, worse lung function, and were more likely to report cough.

6. No single factor or combination of factors could reliably predict leaving or remaining in the study.

SUMMARY OF FINDINGS – CLINICAL STUDY

• Symptom prevalence from the questionnaire was not significantly different between Phase 1 and Phase 2 of the study, despite the considerable turnover of subjects involved in the study and the introduction of the MEL in 1992. (Figure 4.2)

• Symptoms of cough, but no other symptoms, were associated with smoking; no relationship was found between any symptom and duration of exposure.

• The overall health impact of symptoms as defined by the St George’s Respiratory Questionnaire, was comparable with those of mild asthma, and much less severe than those associated with COPD in previously published studies of respiratory health impact in these defined populations. This may be an example of the healthy worker effect.

• Some individual symptoms, particularly symptoms arising in the workplace such as wheeze in a dusty place, were strongly associated with atopy and positive specific IgE to mites and grain. Fungi specific IgE were also associated with symptoms but the prevalence of positive responses to fungi was lower.

• In Phase 1 but not in Phase 2, symptoms described by participants as ‘work-related’ were significantly associated with the presence of specifically work-related specific IgE.

• Subjects with specific IgE to storage mites had a shorter duration of exposure and had smoked less. This relationship was not seen for other occupational allergens.

• FEV1 and FVC were unchanged over the period of the study in the group as a whole, and there was no significant trend towards change (decline) in FEV1 or FVC either in the group analysed as a whole, and in those who remained in the study for more than 8 years.

• In Phase 2, unlike Phase 1, there was no significant relationship between the duration of exposure to grain dust and FEV1. This may reflect the effects of prolonged previous exposure in subjects studied in 1990.

• There was no significant relationship between symptoms and lung function, whether symptoms were measured by the Venables questionnaire or the SGRQ.

• Baseline symptoms in 1990 or in 1998 were not a useful predictor of remaining in the study or leaving, although cough was more common in those who did not remain between 1990 and 1998.

• The �FEV1 was greatest in those who spent least time in the study but was not significantly related to atopy, positive work-related IgE or symptoms at entry into the study.

70

• Peak flow rates in subjects with grain dust exposure showed a decline from baseline compared with flow rates on non exposed days, especially marked in subjects with work-related symptoms.

• A total of 129 Peak flow patterns analysed by OASYS, which is specifically designed to look for patterns of occupational asthma, showed that a work-related pattern could be demonstrated in some records but only 1 of these had symptoms suggestive of work-related asthma. There was a poor correlation with positive immunology (positive specific IgE to work related allergens).

71

5 GENERAL DISCUSSION

There was a clearcut relationship in both Phase one and Phase two of the study between symptoms and sensitisation to work-related allergens that are likely to cause specific IgE responses. This was most clearly identified with ‘work-related symptoms’ in general in Phase 1 of the study, and individual symptoms related to grain dust exposure in Phase 2. These findings provide support for the hypothesis that in some individuals exposed to grain dust, symptoms may be caused by occupational sensitisation to work-related allergens. This correlated poorly with exposure measures such as duration of exposure, and with any variability in lung function that was identified over the longer period of study either from Phase 1 to Phase 2 or within Phase 2. Those who remained in the study did not become progressively sensitised to occupational allergens, did not demonstrate a deterioration of their lung function and did not develop a higher prevalence of symptoms. Those who left the study could not be clearly identified from baseline symptoms or sensitisation to specific IgE, although if they remained for sufficient time to have more than 2 recordings of FEV1 leavers were found to have a more rapid decline in FEV1 unrelated to dust exposure.

These findings strongly suggest that in this study there was a healthy worker effect. The factors causing individual subjects to leave the industry have not been unequivocally identified in this study, at least partly due to the massive restructuring that has taken place in both docks and farms between 1990 and 2002. However it is possible to infer some features of the characteristics of those who left by comparing the whole group with those who stayed. Cough was more common in those who left. Analysis of change in FEV1 in particular suggests that there is an inverse correlation between �FEV1 and length of time in the study. This would not be expected if symptoms were principally a consequence of occupational sensitisation through IgE mediated mechanism. If that were the case, greater variability in airway calibre and possibly an overall decline in FEV1 during the course of the study would be expected. Although greater variability in airway calibre was found in those who left the study before its conclusion, there were actually a slightly higher prevalence of atopy and immunological sensitisation to work-related allergens in those who remained. Occupational asthma as such is unlikely to be a major cause of leaving the industry or of long term respiratory damage.

The difficulty identifying any specific work-related asthma patterns using the OASYS system is more likely to be related to variation in dust exposure than an absence of occupational asthma in this workforce. The close association between the ‘bronchial reactivity’ symptoms, particularly those which are work or dust-related, and immunological sensitisation to work-related allergens, suggests that at least in some subjects occupational asthma is a cause of symptoms. The St George’s respiratory Questionnaire results do suggest that the level of morbidity is low where occupational asthma is causing symptoms.

The lower prevalence of symptoms, better lung function and lower change in FEV1 (�FEV1), in the subjects who remained in the study are all likely to be the consequence of a healthy worker effect (Chan-Yeung et al, 1981,1992).

Symptomatic subjects who do not have occupational asthma may well be experiencing the effects of endotoxin exposure. It is known, and this study confirms, that agricultural workers are heavily exposed to endotoxin. Endotoxin can cause acute changes in airway calibre and airways inflammation (Schwartz et al, 1994,1995, Clapp, 1993). However this study was not designed to demonstrate cross shift changes in FEV1 typical of those seen in byssinosis (McKerrow et al, 1958: Schwartz et al, 1995). In order to demonstrate that effect in these subjects, the analysis of the peak flow records of these subjects may demonstrate a typical ‘endotoxin effect’ but this analysis has not yet been done. Preliminary analysis of the patterns of peak flow charts suggested a greater 8-10 hour decline on grain exposed days in symptomatic subjects than on non grain-exposed days but these charts have not been specifically analysed for the ‘Monday morning effect’ seen in byssinosis.

72

Other studies have suggested that there is a relationship between �FEV1 and duration of exposure. For example, in the many Vancouver grain dust studies, (Chan-Yeung et al, 1981,1992), it was possible to demonstrate a relationship between �FEV1 and exposure levels. In that study, controls had an annual decline in FEV1 of approximately 20 mls/year, workers with average grain dust exposure of < 4mg/mls had a mean decline of 10 mls/yr, and those with mean dust exposure of > 9 mg/ml had a mean decline in FEV1 of 35 ml/year. Our subjects who remained in Phase 2 for just 2 or 3 years had a mean annual decline of approximately 50 mls/year, but in those who remained for 4 years it was just 14 mls/yr, and even less in the ‘stayers’ who remained in for 8-12 years. It was not possible in this study to relate change in FEV1 directly to dusty exposure, as exposure was highly variable between and within individuals. There was more sustained and continuous exposure in dockworkers than in farmers, but despite this there was no significant difference in �FEV1 between the two groups.

Phase 1 of the study suggested that there may be an increasing prevalence of work-related symptoms in association with the development of immunological sensitisation to environmental allergens, but this has not been confirmed by the longer study which shows a remarkable degree of symptoms and specific IgE stability. Another important difference between Phase 1 and Phase 2 are that the relationship between FEV1 and duration of exposure to grain dust which was identified there was not seen in the subjects included in Phase 2. It may well be that those who participated in the first year of the Phase 1 study had had previously higher or different exposures resulting in a degree of chronic impairment in lung function, but this was not seen in Phase 2.

It is also possible that the low �FEV1 seen in those who remained in the study could have resulted from the reduction in the MEL. However this seems unlikely, since the point prevalence of symptoms was identical in every year of the study between 1990 and 2002, and that the prevalence of reduced lung function in the groups as a whole was no different in 1990 compared with 2002. It remains possible, however, that the higher prevalence of abnormal lung function in subjects who left the study soon after 1990 may reflect previously much higher levels of dust exposure.

The present study was not designed to identify the acute effects of dust exposure such as the organic dust toxic syndrome or acute hypersensitivity pneumonitis. The former is generally associated with very high levels of exposure to mouldy organic dusts and thought to be due in part to endotoxin effects (Schenker et al, 1998). The acute symptoms of this syndrome are unlikely to have been detected by the methods used in this study, but it is known that the prevalence of symptoms due to mouldy grain ius much lower in the East of England which is relatively free of rainfall than elsewhere in the UK or Europe (Blainey, 1989). Hypersensitivity pneumonitis will also produce both acute and long-term symptoms. Hoiwever these do not usually result in short term reductions in airway caliber or asthma-like symptoms, and long term chnges in lung function of those affected by hypersensitivity pneumonitis are a reduction in lung volumes not FEV1. This was not seen in the present study (Schenker et al, 1998).

73

6 QUESTIONS ADDRESSED BY THE STUDY

6.1 HAS THE MEL REDUCED SENSITISATION?

The overall trend in the study is a decrease in IgE positive status. Between 1990 and 2001/2 IgE positive dock workers decreased by 0.6% and farmers by 8.2%. The clearest reduction in sensitisation was amongst the non-atopic dockworker population, the percentage of those who tested positive to the occupational allergens was reduced from 10.3% in Phase 1 to 0% in Phase 2.

The dockworker population stayed very steady, 11/17 2001/2 workers also took part in 1990 and therefore the long-term trend towards improvement may not be very marked amongst this population as these workers were occupationally exposed to grain dust for years before the MEL was implemented. The use of better dust control and PPE in the docks in recent years may not yet be showing in the immunological status of these workers. However the general trend towards lower levels of sensitisation amongst even these workers, and the decrease from 10.3% to 0 in the numbers of non-atopics who were positive to the work related allergens, indicates that new recruits may be benefiting from the implementation of the MEL.

Only 11/163 of the 2001/2 farmers also took part in the 1990 immunology study so the ‘stayers’ population can not be observed in the same way as the dockers. However, there was an overall decrease in IgE positives to work related allergens in the farming population from 28.8% positive in 1990 to 20.6% positive in 2001/2. The number of non-atopic farmers who tested positive to the occupational allergens was reduced from 13% in Phase 1 to 11% at the end of Phase 2. Again the majority of farm workers who are responding immunologically to work-related allergens will have been exposed and sensitised before the MEL was implemented. Although our study population was not sustained, generally farm workers stay within the industry.

6.2 DID ANYONE DEVELOP NEW IGG OR IGE DURING THE STUDY?

Yes. Some participants did develop new IgG or IgE during the study, however, more lost their positive status than gained it. No dock workers developed a new sustained positive status, 2 tested positive midway through the study but had returned to their negative status by the time of their final blood sample. 5 farm workers developed a new sustained positive status, 5 tested positive midway through the study but had returned to their negative status by the time of their final blood donation and 7 lost their positive status by the time of their final blood donation.

6.3 SHOULD THE MEL BE REDUCED TO REDUCE SENSITISATION?

Throughout the study very high levels of dust were recorded with levels equal to, or greater than, 30 mg/m3 recorded in all work situations at 17 of the 19 workplaces visited. The MEL for grain dust is an average of 10 mg/m3 of total respirable dust in the air over an 8 hour period, the maximum dust level should never exceed 30 mg/m3 measured over a 10 minute period.

6.3.1 Should the MEL be reduced to reduce IgG levels amongst workers?

No, specific IgG is generally produced as a response to exposure to a particular allergen, even very small amounts of allergen can trigger an IgG response in some people. As discussed in section 3.6, many studies have shown that development of IgG antibodies is not an indicator of disease development. As shown in this study, the only relationship with symptoms and IgG was an inverse

74

relationship with symptoms improving at weekends or holidays, but the significance of this is uncertain.

6.3.2 Should the MEL be reduced to reduce IgE levels amongst workers?

The success of the controls put in place to reduce exposure at the docks may have contributed to reducing the number of positive tests in non-atopic dock workers to zero which shows that it is possible to reduce IgE levels amongst the workers and suggests that the farm worker exposure levels still need to be reduced. In the farming population exposure levels continue to be so high/frequent that even non-atopic workers, who are not predisposed to sensitisation, are becoming sensitised. The results in the dockworkers suggest that it should be possible to reduce this exposure.

Whilst ‘large’ exposures to antigen may be required to trigger initial sensitisation in the workers, in already sensitised persons only a tiny amount of antigen is enough to trigger a repeat response therefore it will take ‘a generation’ of workers before any reductions in IgE responses can be clearly measured.

It is very important to reduce exposure to the sensitising allergens as far as possible. Due to the presence of atopic and already sensitised workers in the workforce and the nature of the work involved, it may not be possible to reduce the levels of exposure to below those required to trigger an IgE response in some of the workforce without the correct use of efficient respiratory protective equipment (RPE).

It is possible that the nature of farmers exposure needs to be re-examined. The farming workplace is less controlled than the docks, it was observed during sampling visits that no RPE was worn during the grain harvest except in the stores, yet workers in the fields were exposed to extremely high levels of dust both inside and outside the air conditioned harvest vehicles. Farmers rely on the air conditioning in their cabs to provide protection, but this and previous studies (Thorpe et. al. 1997) have shown that they provide only limited protection. By increasing farmers’ awareness of their exposure, and by advising when to wear RPE it will be possible to decrease their potential exposure and therefore possibly their sensitisation. In some work situations, for example during the grain harvest, it may not be possible to implement a lower MEL therefore awareness of RPE and when to wear it may be more important than the level at which the MEL is set.

6.4 DOES NEW SENSITISATION RESULT IN NEW SYMPTOMS?

In the subjects who developed ‘new’ specific work-related IgE during the course of the study, there were no distinguishing features at the beginning of the study to identify them at higher risk, or to indicate adverse consequences. 4/5 reported some respiratory symptoms, but in only 1/5 were these described as work related. The rate of decline in FEV1 was not different in this group by comparison with others. The significance of the finding is therefore unclear.

Thus neither symptoms nor new sensitisation appear to be able to predict the development of adverse consequences for those who remain in the industry. In fact those who remain do not have more symptoms, and have less average decline in lung function than those who leave, so for these subjects it has not been possible to demonstrate any adverse consequences of remaining in this study.

It is very difficult to determine whether this is simply a healthy worker effect or the result of reductions in exposure as a consequence of the introduction of the MEL. The trend toward a lower prevalence of sensitisation from 1998 – 2002 does not help to discriminate as only a small proportion of those tested in 2002 had remained in the study from 1990.

75

7 OVERALL SUMMARY

UK grain workers are exposed to high levels of endotoxin, microorganisms and dust. The levels of endotoxin recorded here are greatly in excess of those shown to cause adverse health effects (the Dutch recommended a health-based occupational exposure limit for airborne endotoxin of 50 EU/m3

(DECOS, 1998)), which indicates that the longitudinal exposure to endotoxin may have an important contribution to the work related symptoms experienced by grain workers.

Exposure levels to grain dust were high, and still frequently over the MEL, in particular in the dock basement area and at farms during combining and in their stores. Personal dust exposure at the Dock appeared to decrease between Phases 1 and 2, however dock work during Phase 2 dust sampling was not continuous.

Exposure levels to airborne bacteria and fungi remained very high throughout the study. Actinomycete and A. fumigatus levels were low throughout indicating that the grain was well-kept and in good condition.

IgE sensitisation to work related allergens among grain workers has decreased since the introduction of the grain dust MEL. There was an overall trend towards a decrease in IgE positive responses across the study from Phase 1 to Phase 2. The long term trend towards improvement may not be shown clearly in this study as the majority of the participants were occupationally exposed to grain dust for many years before the MEL was introduced. There was a decrease in the number of non atopic workers who tested IgE positive to the work-related allergens, none of the non-atopic dockworkers were IgE positive in Phase 2 of study. However, 11% of non atopic farmers were still testing positive to one or more of the work related allergens at the end of the study. This indicates that it is possible to reduce exposure levels to below those required to trigger an IgE response in non-atopic workers and that farm workers continue to be exposed to unacceptably high levels of grain dust.

Symptoms and lung function in UK grain workers could be partly explained by sensitisation to environmental allergens. However, there are indications from this study that factors such as endotoxin exposure may play a part in the factors which lead individuals to leave the industry, although it is impossible to say from this study that it plays any part in the development of any long term decline in lung function. Those with the greatest short term decline in lung function remained in the study for the shortest period and those who remained had very low levels of recorded lung function change. Immunological sensitisation does not appear to predispose workers to a greater risk of symptoms, or to leaving the industry. The decline in FEV1 identified in Phase 1 by relating to duration of exposure did not appear to be explained by immunological sensitisation. However it is of importance to note that this relationship was not found in subjects taking part in Phase 2, and it is likely that Phase 1 included subjects who had spend longer in the industry when levels of dust exposure were higher than they are now. The introduction of the MEL in 1993 does not appear to have influenced the prevalence of symptoms or the degree of sensitisation, but may well have influenced the relationship between exposure and long term decline in FEV1.

Clearly some subjects do have both symptoms of occupational asthma and sensitisation to work related allergens. However in neither Phase 1 or Phase 2 could it be clearly demonstrated that this was the principle, or even a major, source of symptoms, or caused subjects to leave the industry. It is suggested that endotoxin may explain some of the work related symptoms. This suggestion may be strengthened by the detailed analysis of ‘endotoxin-like’ effects on airway calibre in these subjects.

It is clear from the analysis of those subjects who remain in the industry over a long period, however, that these very high levels of endotoxin exposure do not have a long term effect on lung function and do not appear to cause chronic lung damage. This may be a selection effect in that those who leave the industry may have been those the most susceptible to lung damage.

76

This study has highlighted the dangers of exposure to microorganisms and their toxins in apparently ‘clean’ areas, such as offices, rest areas and vehicle cabs, were personal respiratory protective equipment is not worn. Improved cleanliness is recommended in rest areas and offices. Vehicle cabs may not be offering enough protection and the use of personal respiratory protective equipment inside the cabs should be considered.

Guidance/information for both employers and workers on endotoxin exposure in the workplace is recommended.

It is not possible to conclude from this study that subjects who remain in this industry now will suffer any long term consequences of exposure. Neither an exposure-related decline in airway caliber nor an increasing prevalence of symptoms were apparent in those subjects who remained exposed to grain dust over long periods.

There are some indications from this study that those who are affected by grain dust may leave the industry early. The subjects who did not remain were more likely to have cough and variable lung function. It is possible that close monitoring of exposed subjects may be of value in detecting those likely to experience symptoms and changes in FEV1 and may identify subjects at risk.

77

8 MAIN RECOMMENDATIONS

1. The findings from the microbiological study have been used to conduct the subsequent studies of immunological status of UK grain workers.

2. The longitudinal health effects of high endotoxin exposure, and the adjuvant effects of endotoxin inhalation with other contaminants, require further investigation.

3. This study highlights the exposure of grain workers to extremely high levels of endotoxin. Guidance/information on endotoxin exposure in the workplace is recommended.

4. This study has highlighted the dangers of exposure to microorganisms and their toxins in apparently ‘clean’ areas were currently personal respiratory protective equipment is not worn. Improved cleanliness is recommended in rest areas, offices and vehicle cabs. Vehicle cabs may not be offering enough protection and the use of. personal respiratory protective equipment inside the cabs should be considered.

5. The levels of IgE in the non-atopic farmers indicates that farmers continue to be exposed to very high levels of grain dust work-related allergens. Observations made during the microbiological sampling indicate that higher quality and more frequently worn personal respiratory protective equipment would help reduce the levels of exposure, particularly during harvesting in areas/situations where it is not practicable to reduce dust exposure in other ways.

6. Personal respiratory protective equipment that offers suitable levels of protection from endotoxin exposure is required.

7. Longitudinal monitoring of symptoms and FEV1 to detect short term changes may help to identify subjects at risk.

DUE TO FORMATTING PROBLEMS THERE IS NO PAGE 79 IN THIS REPORT

78

9 ANNEX 1

Microbiological results tables for Phase 2 of the study.

Farm 10 cfu/m3

bacteria 25oC bacteria 37oC fungi malt 25oC fungi DG18 25oC fungi malt 40oC Thermophilic bacteria and Actinomycetes

In combine A 7.1 x 105 5.7 x 103 6.8 x 104 3.5 x 104 0 0

In combine B 4.9 x 105 1.05 x 105 4.1 x 105 8.3 x 104 555 0

outside combine A 8.8 x 107 8.5 x 105 6.4 x 106 6.6 x 106 350 210

outside combine B 1.6 x 108 6.4 x 105 8.5 x 106 9.9 x 106 1.6 x 104 628

in tractor 2.9 x 106 2.7 x 104 2.0 x 105 4.9 x 104 0 60

outside tractor 2.4 x 104 - - - 0 61

next to tipping 1.6 x 108 7.0 x 106 7.4 x 106 5.2 x 106 650 5.4 x 103

PM Personal monitor ND* Unable to obtain endotoxin reading due to sample interference with the assay.

80

Farm 11 CFU/m3

HSL Sample number

Sample site

number

Bacteria 25oC

Bacteria 37oC

Fungi 25oC Thermophilic bacteria and

Actinomycete

Endotoxin EU/m3

Dust mg/m3

Corresponding blood sample

99,004,833 1 Tractor 1, PM driver 7.72 x 105 5.00 x 104 2.00 x 105 0 655 2.6 F99139 collecting grain from combine

9,900,4834 2 Tractor 2, PM driver bailing cut wheat

1.98 x 105 3.13 x 104 4.17 x 104 313 298 2 F99140

9,900,4835 3 Combine 1, PM combine driver

9.07 x 104 6.85 x 103 7.55 x 103 46 363 1.8 F99154

9,900,4836 4 In tractor 1 with PM1 2.74 x 103 45 716 0 13 0.08

9,900,4837 5 In tractor 2 with PM2 8.88 x 104 - 7.21 x 103 104 36 0.6

9,900,4838 6 In combine 1 with PM3 3.94 x 103 1.25 x 103 1.66 x 103 0 857 0.13

9,900,4839 7 Outside tractor 1 1.04 x 106 5.00 x 104 1.71 x 105 249 884 3.6

9,900,4840 8 Outside tractor 2 6.46 x 105 5.21 x 103 1.31 x 105 0 445 2.5

9,900,4841 9 Outside combine 1 2.47 x 107 6.47 x 105 4.78 x 106 415 1.30x104 19.3

9,900,4842 10 Front of storage site grain hopper

1.41 x 107 1.21 x 106 6.86 x 105 358 4.17x104 43

9,900,4843 11 Control in field away from harvest

1.70 x 103 537 2.86 x 103 0 ND 0.2

9,900,4844 12 unexposed Control filter 0 0 0 0 ND 0

Farm 12 CFU/m3

HSL Sample number

Sample site

number

Bacteria 25oC

Bacteria 37oC

Fungi 25oC Thermophilic bacteria and

Actinomycete

Endotoxin EU/m3

Dust mg/m3


9,900,4845 13 Combine 2, PM driver 1.49 x 106 1.55 x 104 7.56 x 104 0 ND* 1 F99104

9,900,4846 14 PM grain store operator & pig pen worker

7.06 x 106 8.89 x 104 5.39 x 104 833 ND* 3 F99105

81

9,900,4847 15 In combine 2 4.59 x 105 6.25 x 104 6.26 x 104 69 ND* 38

9,900,4848 16 In grain store by hopper 3.99 x 105 2.33 x 105 2.63 x 103 4.34 x 103 ND* NR

9,900,4849 17 outside combine cab 2 7.37 x 107 5.23 x 105 3.34 x 106 1.14 x 103 ND* 17

9,900,4851 18 tractor 3, collecting grain from combine, PM driver

9.39 x 104 2.22 x 103 2.44 x 104 167 ND* 0.26

9,900,4852 19 outside Tractor 3 8.29 x 104 3.45 x 103 5.59 x 103 0 118 0.65

9,900,4853 20 Inside tractor 3 3.44 x 105 2.89 x 103 6.59 x 103 43 ND* 0.8

9,900,4854 21 In pig house 3.73 x 105 0 6.56 x 104 0 ND* 33

9,900,4855 22 Control away from harvest 46 0 0 0 ND* NR

Farm 13, CFU/m3

HSL Sample number

Sample Bacteria 25oC

Bacteria 37oC

Fungi 25oC Thermophillic bacteria and

Actinomycete

Endotoxin EU/m3

Dust mg/m3


9,900,4855 23 Tractor 4, PM driver, also worked in grain store

1.09 x 103 3.1 x 103 1.65 x 105 56 691 0.5 F99007

9,900,4856 24 PM Farmer's wife, a little grain store work

9.34 x 104 6.53 x 103 3.85 x 104 159 26 NR F99008

9,900,4857 25 inside tractor 4 1.62 x 104 1.51 x 103 2.52 0 58 1.05

9,900,4858 26 outside tractor 4 2.88 x 106 7.08 x 104 4.48 x 104 224 3.76x103 NR

9,900,4859 27 pm tractor driver 1.28 x 105 3.73 x 103 1.55 x 104 0 445 NR F99155

9,900,4860 28 Inside combine 3 2.51 x 104 171 3.02 x 103 0 27 40.3

9,900,4861 29 Outside combine 3 3.34 x 106 5.24 x 106 3.1 x 106 328 x 103 1.94x104 12.55

9,900,4862 30 Grain store 1.97 x 105 2.44 x 103 3.1 x 104 145 610 69.12

9,900,4863 31 Control away from harvest 290 261 464 0 0.5 NR

Farm 14 CFU/m3

82

HSL Sample number


Bacteria 37oC


Actinomycete

Endotoxin EU/m3

Dust mg/m3


9,900,4864 32 Tractor 5, PM driver 3.00 x 106 3.52 x 105 3.46 x 105 118 1.85x104 31.6 F99147

9,900,4865 33 Combine 4, PM driver 5.17 x 104 2.02 x 102 6.86 x 103 65 139 1.91 F99044

9,900,4866 34 Inside tractor 5 4.27 x 105 2.81 x 104 8.25 x 104 110 647 0.4

9,900,4867 35 Outside tractor 5 4.2 x 104 2.94 x 103 1.37 x 104 0 39 18

9,900,4868 36 Inside combine 4 8.09 x 104 6.01 x 103 7.54 x 103 0 18 0.19

9,900,4869 37 Outside combine 4 1.02 x 107 6.09 x 106 2.9 x 107 1,015 193 2.96

9,900,4870 38 In grain store 1.88 x 106 1.17 x 105 6.25 x 105 66.5 0.4 0.68

9,900,4871 39 Control away from work 4.72 x 103 399 2.23 x 105 66.5 5 41.58

Farm 15, CFU/m3

HSL Sample number


Bacteria 37oC


Actinomycete

Endotoxin EU/m3

Dust mg/m3


9,900,4872 40 PM Farmer, 1 hr combine driving, a little store work

7.84 x 104 1.89 x 103 2.62 x 104 0 390 33.01 F99034

9,900,4873 41 Combine 5, PM Driver 9.35 x 105 5.47 x 103 3.63 x 104 0 188 NR F99035

9,900,4874 42 tractor 6, PM driver 8.77 x 105 3.23 x 104 2.48 x 105 719 1.58x103 69.53 F99036

9,900,4875 43 Inside combine 5 6.02 x 104 6.37 x 103 7.25 x 103 0 0.62 0.87

9,900,4876 44 Outside combine 5 2.54 x 108 6.21 x 106 2.91 x 106 1.94 x 103 1.95x104 0.05

9,900,4877 45 Inside tractor 6 1.26 x 106 1.06 x 106 1.34 x 105 96 291 2.83

9,900,4878 46 Outside tractor 6 1.34 x 104 736 x 104 1.77 x 103 0 ND NR

9,900,4879 47 In grain store 1.42 x 104 2 x 105 3.96 x 105 169 2.45x103 NR

9,900,4880 48 Control away from harvest 1.05 x 103 101 2.4 x 103 0 ND 17.96

9,900,4881 49 Unexposed control 0 0 0 0 ND NR

83

PM Personal monitor ND* Unable to obtain endotoxin reading due to sample interference with the assay.

�Farm 16 CFU/m3

HSL Sample number

Sample site number

Bacteria 25oC Bacteria 37oC Fungi 25oC

Thermophilic bacteria and

Actinomycete

Endotoxin EU/m3

Dust mg/m3


20,205,191 1 Combine driver PM 1.22x106 4.93x103 2.06x105 809 8.25X103 2.71 200,205,192 2 + 12 Outside combine 5.88x107 1.82x105 1.69x107 221 3.58X104 15.94

+ 200205199

200,205,193 3 Inside combine 1.31x105 211 3.13x104 78 2.18X103 0.33 200,205,194 4 Tractor driver PM 2.90x106 4.92x103 3.33x105 ND 4.70X103 6.02 20,025,195 5 Outside tractor 1.94x107 6.53x104 2.53x106 67 1.72X104 4.67

200,205,196 6 Inside tractor 3.47x105 1.03x104 4.87x104 67 6.74X103 1.71 200,205,197 10 Grain store 1.27x106 6.09x103 1.69x105 ND 7.02X103 3.6 200,205,200 13 Grian store operator PM 2.23x105 65 4.46x104 ND 1.17X103 1.33 200,205,198 11 Control in office 2.23x104 471 1.40x104 ND 2.16X103 0.31

84

Farm 17 CFU/m3

HSL Sample number

Sample site number

Bacteria 25oC Bacteria 37oC Fungi 25oC

Thermophilic bacteria and

Actinomycete

Endotoxin EU/m3

Dust mg/m3


7 Office control 2.89x103 22 2.23x103 297 <5 0.19 200205279 200,205,280 8 Combine driver PM 9.31x105 1.49x103 2.45x105 1.12x103 <5 1.59 200,205,281 8 Combine inside, static 2.76x105 241 9.47x104 332 <5 0.23 200,205,282 12 + 19 Combine outside 1.48x108 4.24x104 1.10x107 1.25x103 3.55x103 26.73

+ 200205288 200,205,283 14 Farm manager (not

working on grain) 1.83x105 2.47x103 2.14x104 ND <5 1.07

200,205,284 15 Tractor outside 2.16x106 3.75x103 3.06x105 181 1.06x103 4.03 200,205,285 16 Tractor inside 3.99x105 661 5.10x104 268 <5 0.68 200,205,286 17 Tractor driver/grain store

worker PM 2.72x106 1.46x104 7.35x105 422 2.90x103 8.6

200,205,287 18 Grain store static 3.09x106 9.01x103 5.12x105 338 1.63x103 4.07

85

Dock visit 1998

cfu/m3

site bacteria 25oC bacteria 37oC fungi malt fungi DG18 fungi malt 40oC Thermophilic Endotoxin EU/m3

25oC 25oC bacteria and Actinomycetes

Ship Hull A Loading English wheat for human

8.7 x 107 3.2 x 107 1.0 x 106 2.05 x 105 0 3.7 x 103 7.7 x 106

Ship Hull B� consumption 3.1 x 108 1.7 x 108 8.1 x 106 8.0 x 105 0 3.8 x 103 5.58 x 106

Lorry Loading Control Gallery

loading wheat and soya

5.3 x 106 - 604 0 0 242 1.8 x 103

Scales Room 6.7 x 105 5.0 x 105 6.3 x 103 3.4 x 103 1.1 x 103 346 2.1 x 105

Basement 1.9 x 107 1.6 x 107 1.3 x 105 1.1 x 106 8.4 x 103 1.2 x 103 1.8 x 106

Office 2.3 x 104 2.1 x 103 193 129 64.5 0 3.4 x 103

July 2000 Dock A visit - loading ship with English wheat for animal consumption

Site cfu/m3

86

Sample number

Bacteria 25oC

Bacteria 37oC

Fungi Malt 25oC

Fungi DG18 25oC

Fungi Malt 40oC

Thermophilic Bacteria &

Actino 55oC

Endotoxin EU/m3

Dust mg/m3

Worker number

TD1 Personal Monitor - head house and basement

2.8 x 105 4.36 x 104 2.67 x 104 6.14 x 104 1.2 x 103 2.58 x 103 4,218.6 6.14 D005


1.26 x 105 6.37 x 104 3.24 x 104 5.15 x 104 735 1.96 x 103 3,917.6 5.26 D013


1.88 x 105 6.37 x 104 1.89 x 104 4.35 x 104 326.5 1.58 x 103 2,030.9 3.15 D006

TD4 Personal Monitor - tower 2 cabin

2.33 x 104 5.74 x 103 1.02 x 104 1.15 x 104 0 0 358.1 5.14

TD5 Personal Monitor - loading on ship

2.9 x 104 2.59 x 104 1.06 x 105 - 0 0 59.1 1.66 D011

TD6 Personal Monitor - workparty (cleaning)

1.66 x 103 0 - - 0 0 13.9 1.73 D033

TD7 Personal Monitor - workparty (cleaning)

1.3. x 104 1.35 x 104 2.69 x 103 6.16 x 103 115.5 0 319.1 4.73 D032

TD8 Personal Monitor - JS

2.7 x 104 3.48 x 103 2.61 x 103 1.30 x 104 - 0 88.5 2.96

TD 10 Ship Hold 6.34 x 106 4.66 x 105 1.75 x 105 1.27 x 105 8.03 x 103 1.33 x 104 73,897.2 54.29 TD 11 Ship Hold 3.7 x 106 2.36 x 105 1.53 x 105 1.57 x 105 2.65 x 103 9.06 x 103 122,108 49.76 TD 12 Basement by

conveyor 6.5 x 106 1.78 x 105 1.41 x 105 2.15 x 105 4.45 x 103 1.80 x 104 29,701.7 22.7

TD 13 Basement office 1.4 x 105 3.0 x 104 3.39 x 104 4.22 x 104 572 1.23 x 103 2,119 4.17 TD 14 Scales room 4.55 x 103 0 0 0 0 0 20.6 1.56 TD 15 7th floor cupola 2.4 x 103 - 5.28 x 103 5.76 x 103 0 0 59.2 1.48 TD 16 control room 1.01 x 105 1.44 x 104 7.63 x 103 1.60 x 104 864 1.92 x 103 1,880.6 1.92 TD 17 mess room 1.05 x 103 7.38 x 102 580 263 0 0 24.5 3.29 TD 18 lorry loading control 4.52 x 104 1.17 x 104 8.39 x 103 9.14 x 103 322 752 1,585.2 5.63 TD 21 unexposed control 0 0 0 0 0 0 <5EU/filter -

87

Dock A 2003

CFU/m3

Sample number

Bacteria 25oC Bacteria 37oC Fungi Malt 25oC

Fungi Malt 40oC

Thermophilic Bacteria and

Actinomycetes 55oC

Endotoxin EU/m3

Dust mg/m3

1 Worker loading on ship pm 2.0x107 2.38E+06 8.2x104 8 0 1.9X105 118 2 Static on ship 3.9x107 4.07E+06 1.0x106 3,806 0 4.5X105 275 3 Worker in basement and scales 4.1x105 2.77E+05 1.5x104 2,841 0 1.2X103 3

room pm 4 Worker loading on ship pm 2.2x107 1.79E+06 2.1x104 846 0 1.5X105 137 5 Static in mess room 4.1x103 5.76E+02 2.5x102 0 0 7.7X102 0.3 6 Static basement office 1.1x106 8.57E+04 1.8x104 1,748 0 3.0X103 4 7 Static basement conveyor area 4.3x105 2.92E+06 1.2x105 7,216 0 8.9X103 17 8 Static head house 1.2x106 1.29E+05 3.7x104 2,389 0 1.4X104 7 9 Static 7th floor control room 6.6x103 3.76E+02 2.6x102 0 0 1.0X101 0.3

10 Static 7th floor cupola floor 2.7x104 2.39E+03 8.8x102 0 0 1.6X102 0.3 11 Worker in basement and scales 4.5x105 1.35E+05 1.4x104 1,946 0 2.2X103 4

room pm 12 Static in scales room 2.4x105 2.31E+05 8.9x103 2,789 0 6.9X102 2

NOTE: DUE TO FORMATTING PROBLEMS THERE IS NO PAGE 89 IN THIS REPORT

88

10 REFERENCES

Aas K, Leegaard J, Aukrust L, Grimmer O. (1980) Immediate type hypersensitivity to common moulds. Allergy, 35, 443-451.

Alvarez MJ, Tabar AI, Quirce S, Olaguibel JM, Lizaso MT, Echechepia, Rodriguez A and Garcia BE (1996) Diversity of allergens causing occupational asthma among cereal workers as demonstrated by exposure procedures. Clinical and Experimental Allergy 26 147-153

American Thoracic Society. (1997) Standardization of spirometry: 1997 update. Am Rev Respir Dis 1987; 136:1285-93

Amner W, Edwards C, McCarthy AJ (1989) Improved medium for the isolation of the farmers lung organism, Saccharomonospora viridis. Appl and Environ Microbiol 55, 2669-2674.

Andersen AA (1958) New sampler for the collection, sizing and enumeration of viable airborne particles. J Bacteriol 76,147-484.

Anees W, Huggins V, Blainey D, Pantin C, Robertson K, Burge S 2002 Evaluation of an expert system for the interpretation of serial peak flow measurements in the diagnosis of occupational asthma in a field trial HSE report, 450/2002 (ISBN 0 7176 2525 7)

Arimoto T, Kadiiska MB, Sato K, Corbett J, Mason RP. (2005). Synergistic production of lung free radicals by diesel exhaust particles and endotoxin. Am J Respir Crit Care Med. 15, 379-87.

Avila R; Lacey J (1974) The role of Penicillium frequentans in suberosis (respiratory disease in workers in the cork industry). Clin Allergy 4(2):109-17

Axen R, Ernbach S. (1971) Chemical fixation of enzymes to cyanogen halide activatedpolysaccharides. Eur. J. Biochem 18, 351-360.

Bagon, D. (1993) Grain dust results associated with Grain Handling. HSE Report. FOD, FCG.

Barnes P (2004) Small airways in COPD NEJM 350:2635-2637

Beard MD, Buchan RM, Reeves FB, Salasek M. (1996) Pulmonary Reactive Components of Winter Wheat Grain Dust at Colorado Farms and Elevators. Appl. Occup. Environ. Hyg 11, 1409-16.

Beckett WS, (2000) Occupational Respiratory Disease. N Engl J Med 342 406-413

Becklake, M.R. (1980) Grain dust and health: state of the art. In: Occcupational pulmonary disease:Focus on grain dust and health. Eds: Dosman JA, Cotton DJ Publ: Academic Press Inc., London.

Blainey JL, Blainey AD (1999) Peak Flow Variability in Agricultural Workers Exposed to Grain Dust Am J Resp & Critical Care Medicine 159 (3)pt 2 A299

Blainey AD (1990) Respiratory Disease in Grain Workers. Respiratory Medicine; 84:93-95

Blainey AD, Topping MD, Ollier S, Davies RJ (1989) Allergic respiratory disease in grain workers;The role of storage mites. J Allergy Clin Immunol, 84,296-303.

90

Blainey, A.D., Power, M., Gould, J.R.M., and Crook, B. (1995) The respiratory and immunological effects of grain dust in the UK Thorax 50(suppl 2) A71

Blaski CA, Clapp WD, Thorne PS, Quinn TJ, Watt JL, Frees KL, Yalga SJ, Schwart DA. (1996) The role of atopy in grain dust induced airway disease. Am. J. Respir. Crit. Care. Med. 154, 334-340

Bovallius A, Brucht B, Roffey R, Anas A. (1978) Three year investigation of the natural airborne bacterial flora at four localities in Sweden Appl. Environ. Microbiol 35, 847-852

Burrell R. (1995) Immunotoxic reactions in the agricultural environment. Ann. Agric. Environ. Med. 2, 11-20.

Carvalheiro MF, MarquesGomes MJ, Santos O, Duarte G, Henques J, Mendes B, Marques A, Avila R. (1994) Symptoms and Exposure to endotoxin among brewery employees. Am. J. Ind. Med. 25, 113-5.

Ceska M, Eriksonn R and Vargo JM. (1972) Radioimmunosorbent assay of allergens. J. Allergy Clin. Immunol. 49, 1-7.

Chan-Yeung M, DimichWard H, Enarson DA, Kennedy SM. (1992) Five cross-sectional studies of grain elevator workers. Am. J. Epidemiol 136, 1265-79.

Chan-Yeung M, Enarson DA, Kennedy SM. (1992) The impact of grain dust on respiratory health. Am. Rev. Respir. Dis. 145, 476-87.

Chan-Yeung M, Schulzer L, McLean E, Dorken F, Tan S, Lam D, Enarson D, Grzybowski S (1981) A follow up study of the Grain Elevator workers in the port of Vancouver Arch Env Health 36:75-81

Clapp WD, Thorne PS, Frees KL, Zhang X, Lux CR, Schwartz DA. (1993) The effects of Inhalation of Grain Dust Extract and Endotoxin on Upper and lower Airways. Chest 104, 825-30.

Clapp WD, Becker S, Quay J, Watt JL, Thorne PS, Frees KL, Zhang X, Koren HS, Lux CR, Schwartz DA. (1994) Grain Dust-induced Airflow Obstruction and Inflammation of the Lower Respiratory Tract. Am. J. Resp. Crit Care Med. 150, 611-7.

Cole G.T and Samson R.A (1984) The Conidia In : Mould Allergy Eds. Al- Doory Y and Domson J.F. Lea and Febiger, Philadelphia, USA

Committee on Organic Dusts International Commission on Occupational Health, (1997) Endotoxins in the environment: A Criteria Document. Int. J. Occ. Environ. Health 3,S1-S48.

Corey P, Hutcheon M, Broder I, Mintz S. (1982) Grain elevator workers show work-related pulmonary function changes and dose-effect relationships with dust exposure. B. J. Ind. Med. 39, 330-7.

Crook B. (1995) Inertial samplers: Biological perspectives. In Bioaerosols Handbook Ed. Cox CS amd Wathes CM. Pub. CRC Lewis London.

Crook B, Lacey J. (1988) Enumeration of airborne micro-organisms in work environments. Environ. Tech. Lett. 9, 515-20.

Darke CS, Knowelden J, Lacey J, Milford-Ward A. (1976) Respiratory disease of workers harvesting grain. Thorax 31, 294-302.

91

55

DECOS Duch expert committee on occupational standards. (1998) Health -based recommended occupational exposure limit for Endotoxins. Gezondheidsraad.

DeLucca AJ, Godshall MA, Palmgren MS (1984) Gram-negative bacterial endotoxins in grain elevator dusts. Am. Ind. Hyg. Assoc. J. 45(5),336-339.

DeLucca AJ, Palmgren MS (1987) Seasonal variation in aerobic bacterial populations and endotoxin concentrationsin grain dusts. Am. Ind. Hyg. Assoc. J. 48(2) 106-110

Dimich-Ward HD, Kennedy SM, Dittrick MA, DyBuncio A, Chan-Yeung M.(1995) Evaluation of the respiratory health of dock workers who load grain cargoes in British Colombia. Occup. Environ. Med. 52, 273-8.

DoPico GA (1980) Report on diseases. Am. J. Ind. Med. 1980,10,261-265 .

DoPico GA, Flaherty D, Bhansali P, Chavaje N (1982) Grain fever syndrome induced by inhalation of airborne grain dust. J. Allergy. Clin. Immunol 69, 435-43.

Dutkiewicz J(1978a) Exposure to dust-borne bacteria in agriculture. I. Environmental studies. Arch Environ Health 33,250-259.

Dutkiewicz J (1978b) Exposure to dust-borne bacteria in agriculture. II. Immunological studies. Arch Environ Health 33,260-270.

Dutkiewicz J. (1976) Studies on endotoxins of Erwinia herbicola and their biological activity. Zbl. Bak. I. Abt. Orig. 236, 487-508.

Dutkiewicz J. (1986) Microbial hazards in plants processing grain and herbs. Am. J. Indust. Med. 10, 300-2.

Dutkiewicz J(1987) Bacteria in the farming environment Eur. J. Resp. Dis. 71, 71-88

Dutkiewicz J (1997) Bacteria and fungi in organic dust as a potential health hazard. Ann Agric Environ Med 4, 6-11.

Dutkiewicz J, Kus L, Dutkiewicz E, Warren CPW (1985) Hypersensitivity pneumonitis in grain farmers due to sensitization to Erwinia herbicola Ann Allerg 54,65-68.

Dutkiewicz J, Olenchock SA, Sorenson WG, Gerencser VF, May JJ, Pratt DS, Robinson VA. (1989) Levels of bacteria, fungi, and endotoxin in bulk and aerosolised corn silage. Appl Environ Microbiol

Eduard W (1997) Exposure to non-infectious microorganisms and endotoxins in agriculture. Ann Agric Environ 4 179-186

Enarson D, Vedal S, Chan-Yeung M (1985) Rapid decline in FEV1 among grain handlers Am Rev Resp Dis 132:814-817

Fishwick D, Raza SN, Beckett P, Swan JR, Pickering CA, Fletcher AM, Niven RM, Francis H, Rawbone R, Curran AD. (2002). Monocyte CD14 response following endotoxin exposure in cotton spinners and office workers. American J. Industrial Med. 42 437-442.

92

350

Fogelmark and Rylander, (1994) Pulmonary inflammation induced by repeated inhalations of (1-3)β-D-glucans and endotoxin Int.J. Exp. Pathol. 74,85-90.

Fonn S, Groeneveld HT, deBeer M, Becklake MR. (1993A) An Environmental and Respiratory Health Survey of Workers in a Grain Mill in the Johannesburg Area, South Africa. American Journal of Industrial Medicine 24, 387-400.

Fonn S, Groeneveld HT, deBeer M, Becklake MD (1993B) Relationship of respiratory health status to grain dust in a Witwatersrand grain mill: Comparison of workers' exposure assessments with industrial hygiene survey findings. AM J Ind Med 24, 401-11.

Gao HM, Hong JS, Zhang W, Liu B. (2003). Synergistic dopaminergic neurotoxicity of the pesticide rotenone and inflammogen lipopolysaccharide: relevance to the etiology of Parkinson's disease. J Neurosci. 15, 1228-36.

Godnic-Cvar J, Zuskin E, Mustajbegovic J, Schachter EN, Kanceljak B, Maca J, Ilic Z and Ebling Z (1999) Respiratory and immunological findings in brewery workers. American Journal of Industrial Medicine 35 68-75

Griffin P, Ford AW, Alterman L, Thompson J, Parkinson C, Blainey AD, Davies RJ and Topping MD (1989) Allergenic and anitgenic relationship between three species of storage mite and the house dust mite, dermatophagoides pteronyssinus. Journal of Allergy and Clinical Immunology 84 108-117

Heederik D, Smid T, Houba R, Quanjer PH. (1994) Dust-related decline in lung function among animal feed workers. Am. J. Ind. Med. 25, 117-9.

Heederik D, Douwes J. (1997) Towards an occupational exposure limit for endotoxins? Ann. Agric. environ. Med. 4, 17-19.

Hjort C, Andersen P, Schonheyder H and Stenderup A (1986) Occurrence of Aspergillus unbrosus antibodies in Danish farmers. Allergy 41 104-109

Holgate ST (1997) The cellular and mediator basis of asthma in relation to natural history. The Lancet

Holgate ST, Church MK (1993) Allergy Pub: Gower Medical Publishing London

HSE Information sheet (1993) Grain dust - maximum exposure limit. Agricultural sheet no.3.

HSE (1993) Grain dust Guidance Note EH66

HSE Information sheet (1993) Health hazards from grain dust during the loading and unloading of ships. Docks Sheet No 2.

HSE (1993) Grain Dust in Maltings (maximum exposure limit) HSE Guidance Note EH62

HSE (1997) Health and Safety statistics 1996/97 HSE Books ISBN 0 7176 1461 1

HSE (1998) Medical aspects of occupational asthma Guidance Note MS25

Hurst TS, Dosman JA (1990) Characterisation of health effects of grain dust exposures. Am J Ind Med 17 27-32

93

Huy T, Schipper KD, Chan-Yeung M, Kennedy SM. (1991) Grain Dust and Lung Function. Am. Rev. Respir Dis. 144, 1314-21.

Inoue KI, Takano H, Yanagisawa R, Hirano S, Kobayashi T, Ichinose T, Yoshikawa T. (2006) Effects of organic chemicals derived from ambient particulate matter on lung inflammation related to lipopolysaccharide. Arch Toxicol. 2006 Apr 26; [Epub ahead of print]

Jagielo PJ, Thorne PS, Watt JL, Frees KL, Quinn BS, Schwartz DA. (1996a) Grain dust and endotoxin inhalation challenges produce similar inflammatory responses in normal subjects. Chest 110, 263-70.

Jones BL, Cookson JT. (1983). Natural atmospheric microbial conditions in a typical suburban area. Appl. Environ. Microbiol 45, 919-934.

Jones PW (2001) Health status measurement in chronic obstructive pulmonary disease Thorax 56: 880-887

Jones PW, Quirk FH, Baveystock CM, Littlejohns P. (1992) A self-complete measure of health status for chronic airflow limitation. The St. George's Respiratory Questionnaire. Am Rev Respir Dis. 145(6):1321-7.

Katila ML, Ojanen TH and Mantyjarvi RA (1986) Significance of IgG antibodies against environmental microbial antigens in a farming population. Clinical Allergy 16 459-467

Kurup VP and Fink JN (1977) Extracellular antigens of micropolyspora faeni grown in synthetic medium Infection and Immunity 15(2) 608-613

Lacey J (1988) Actinomycetes as biodeteriogens and pollutants of the environment. In: Goodfellow M, Williams ST: Actinomycetes In Biotechnology, Academic press, London 359-432.

Lacey J (1980) The microflora of grain dusts. In: Dosman JA, Cotton DJ :Occupational pulmonary disease: focus on grain dust and health 189-200.

Lacey J. (1990) Grain dust and health. Postharvest. News. and. Information. 1, 113-7.

Lacey J (1995) Airborne agents of occupational lung disease and their detection. Ann Agric Environ Med 2, 31-35.

Lacey J, Dutkiewicz J (1976) Methods for examining the microflora of mouldy hay. J Appl Bact 41, 13-1.7

Lacey J, Dutkiewicz J (1994) Bioaerosols and occupational lung disease. J. Aerosol Sci. 25, 13711404

Lewis DM, Romeo PA and Olenchock SA (1986) Prevalence of IgE Antibodies to Grain and Grain Dust in Grain Elevator Workers. Environmental Health Perspectives 66 149-153

Liesivuori J, Kotimaa M, Laitinen S, Louhelainen K, Ponni J, Sarantila R, Husman K. (1994) Airborne endotoxin concentrations in different work conditions. Am. J. Ind. Med. 25, 123-4.

Lin FJ, Kennedy SM, Chan-Yeung M. (1994) Effect of grain exposure and smoking on the longitudinal changes in immediate skin reactivity. Clinical and Experimental Allergy 24, 949-54.

94

Malmberg P, Rask-Andersen A, Palmgren U, Hoglund S, Komodin-Hedman B and Stalenheim G (1985) Exposure to microorganisms, febrile and airway-obstructive symptoms, immune status and lung function of Swedish farmers Scand J Work Environ Health 11 287-293

Marx JJ, Twiggs JT, Ault BJ, Merchant JA, Fernandez-Caldas E (1993) Inhaled aeroallergen and storage mite reactivity in a Wisconsin farmer nested case-control study. Am Rev Respir Dis 147, 3548.

Massin N, Bohadana AB, Wild P, Kolopp S, M.N., Toamain JP (1995) Airway responsiveness to methacholione, respiratory symptoms, and dust exposure levels in grain and flour mill workers in Eastern France. American Journal of Industrial Medicine 27, 859-69.

May KR, Harper GJ (1957) The efficiency of various liquid impinger samplers in bacterial aerosols. Brit J Ind Med 14, 287-297.

McDuffie HH, Pahwa P, Dosman JA. (1991) Respiratory health status of 3098 Canadian grain workers studied longitudinally. AM. J. Ind. Med. 20, 753-62.

Michel O. (1997) Human challenge studies with endotoxins. Int. J. Occ. Environ. Health. 3 S18-S26.

Michel O, Gianni R, Duchateau J, Vertongen F, Le Bon B, Sergysels R. (1991) Domestic endotoxin exposure and clinical severity of asthma. Clin. exp. allerg. 21, 441-448.

Michel O, Gianni R, Le Bon B, Content J, Duchateau J, Sergysels R. (1992) Inflammatory response to acute inhalation of endotoxin in asthmatic patients.

Am. Rev. Resp. Dis. 146, 352-357

Michel O, Duchateau J, Plat G, Cantinieaux B, Hotimsky A, Gerain J, Sergysels R. (1995) Blood inflammatory response to inhaled endotoxin in normal subjects. Clin. Exp. Allerg 25, 73-9.

Michel O, Gianni R, Le Bon B, Content J, Duchateau J, Sergysels R. (1997) Dose response relationship to inhaled endotoxin in normal subjects. Am. J. Resp. Crit. Care Med. 145, 1157-1164

Olenchock SA, Lewis DM, Mull JC. (1989) Effects of different extraction protocols on endotoxin analyses of airborne grain dusts. Scand. J. Work Environ. Health. 15, 430-5.

Olenchock SA, May JJ, Pratt DS, Piacitelli LA, Parker JE. (1990) Presence of endotoxins in different agricultural environments. Am. J. Ind. Med. 18, 279-84.

Pahwa, P, Senthilselvan S, McDuffie H, Dosman JA (1994) Longitudinal estimates of pulmonary function decline in grain workers Am J Respir Med Crit Care 150:656-662

Palmgren U, Strom G, Blomquist G, Palmgren P,(1986) Collection of airborne microorganisms on Nuclepore filters, estimation and analysis - CAMNEA method. J. Appl. Bacteriol. 61, 401-406.

Palmgren MS, Lee LS, DeLucca AJ, Ciegler A (1983) Preliminary study of microflora and mycotoxins in grain dust from New Orleans area grain elevators. Am Ind Hyg Assoc J 44 485-488

Park H-S and Jung K-S (1998) Enhanced neutrophil chemotactic activity after bronchial challenge in subjects with grain dust-induced asthma Annals Allergy, Asthma & Immunology 80 257-262

95

Park H-S, Suh C-H,Nahm D-H and Kim H-Y (1999) Presence of specific IgG antibody to grain dust does not go with respiratory symptoms. J Korean Med Sci 14 39-44

Park H-S, Nahm D-H, suh C-H, Kwon O-Y, Kim KS, Lee S-W and Chung H-K (1998) Occupational asthma and IgE sensitisation to grain dust J Korean Med Sci 13 275-80

Pollock R, Shaw A, Patel KR, Symington IS, Greenberg M (1980) Ventilatory function of granary workers in the west of Scotland: In: Dosman JA, Cotton DJ: Occupational Pulmonary Disease: Focus on grain Dust and health. 285-296.

Quanjar PH Tammeling GJ, Cotes JE, (1993) Standardized lung function testing . Eur Respir J 10:1683-93

Ramazzini B(1713) De morbis artificum diatriba. Translation by Wrght WC (1964) Pub: Hafner, New York.

Ross DJ, Sallie BA, McDonald JC. (1995) SWORD'94:surveillance of work-related and occupational respiratory disease in the UK. Occup. Med. 45, 175-8.

Rylander R. (1994) Organic Dusts and Lung Disease: The Role of Inflammation. Ann. Agric. Environ. Med. 1, 7-10.

Rylander R. (1997) Evaluations of the risks in endotoxin exposures. Int. J. Occ. Environ. Health. 3 S32-S36.

Rylander R. (1994) Organic Dusts and Lung Disease: The Role of Inflammation. Ann. Agric. Environ. Med. 1, 7-10.

Sanchez-Monge R, Gomez L, Barber D, Lopez-Otin C, Armentia A and Salcedo G (1992) Wheat and barley allergens associated with baker's asthma Biochem J 281 401-405

Sandiford CP, Tee RD, Newman-Taylor AJ. (1995) Identification of crossreacting wheat, rye, barley and soya flour allergens using sera from individuals with wheat-induced asthma. Clinical and Experimental allergy. 25, 340-9.

Sandström T, Bjermer L, Rylander R. (1994) Lipopolysaccharide (LPS) Inhalation in Healthy Subjects Causes Bronchoalveolar Neutrophilia, Lymphocytosis, and Fibronectin Increase. Am. J. Ind. Med. 25, 103-4.

Schenker M et al (1998) Respiratory Health hazards in Agriculture Am J Resp Crit Care Med 158:S1-S76

Schønheyder H, Møller-Hansen C, Andersen P and Stenderup A (1985) Serum antibodies to Aspergillus fumigatus in Danish farmers. Sabouraudia Journal of Medical and Veterinary Mycology 23 93-99

Schwartz DA. (1996) Grain dust, endotoxin, and airflow obstruction. Chest 109, 57S-63S

Schwartz DA, Thorne PS, Jagielo PJ, White GE, Bleuer SA, Frees KL (1994) Endotoxin responsiveness and grain dust-induced inflammation in the lower respiratory tract. The American Physiological Society 267, L609-17.

96

Schwartz DA, Thorne PS, Yaglan SJ, Burmeister LF, Olenchock SA, Watt JL, Quinn TJ. (1995a) The Role of Endotoxin in Grain Dust-induced Lung Disease. Am. J. Respir. Crit. Care. Med. 152, 603-8.

Schwartz DA, Donham KJ, Olenchock SA, Popendorf WJ, Van Fossen DS, Burmeister LF, Merchant JA. (1995b) Determinants of longitudinal changes in spirometric function among swine confinement operators and farmers. Am. J. Respir. Crit. Care. Med. 151, 47-53.

Selim MI; Juchems AM; Popendorf W (1998) Assessing airborne aflatoxin B1 during on-farm grain handling activities. Am Ind Hyg Assoc J 59(4):252-6

Senthilselvan A, Pahwa P, Wang P, McDuffie H, Dosman JA (1996) Persistent wheeze in grain elevator workers should not be ignored. Am J Respir Crit Care Med 153, 701-5.

Simpson JCG, Niven RMcl, Pickering CAC, Oldham LA, Fletcher AM, Francis HC. (1999) Comparative Personal Exposures to organic dusts and endotoxin. Ann. Occ. Hyg. 43; 107 –115.

Singh AK, Jiang Y. (2003). Lipopolysaccharide (LPS) induced activation of the immune system in control rats and rats chronically exposed to a low level of the organothiophosphate insecticide, acephate. Toxicol and Health. 19, 93-108.

Smid T, Heederik D, Mensink G, Houba R, Boleij JSM. (1992a) Exposure to dust, endotoxins, and fungi in the animal feed industry. Am. Ind. Hyg. Assoc. J. 53, 362-8.

Smid T, Heederik D, Houba R, Quanjer PH. (1992b) Dust and endotoxin-related respiratory effects in the animal feed industry. Am. Rev. Respir. Dis. 146, 1474-1479.

Sutton BJ, Gould HJ (1993) The human IgE network. Nature 366 421-28

Swan JRM, Crook B. (1998) Airborne microorganisms associated with grain handling. Annals of Agricultural and Environmental Medicine 5, 7-15.

Thorn P, Rylander R, (1998) Airways inflammation and glucan in a rowhouse area. Am. J. Resp. Crit. Care. Med. 157,1798-1803

Thorpe A, Gould JRM, Brown RC, Crook B (1997) Investigation of the performance of agricultural vechile cab filtration systems.J Agric Engin.Res 66 135-149

Tielemans E, Heederik D, van Pelt W. (1994) Changes in ventilatory function in grain processing and animal feed workers in relation to exposure to organic dust. Scand. J. Work. Environ. Health 20, 43543.

Venables KM, Farrer N, Sharp L, Graneek BJ Newman Taylor AJ. (1993) A Respiratory Symptoms Questionnaire for asthma epidemiology: Validity and Reproducibility. Thorax 48: 214-219.

Von Essen, S.G., Thompson AB, Robbins RA, Jones KK, Dobry CA, Rennard SI (1990) Lower respiratory tract inflammation in grain farmers. Am. J. Ind. Med. 17, 75-76.

Wardrop VE, Blyth W and Grant IWB (1977) Farmer's lung in a group of Scottish dairy farms British Journal of Industrial Medicine 34 186-195

Yanagisawa R, Takano H, Inoue K, Ichinose T, Sadakane K, Yoshino S, Yamaki K, Kumagai Y, Uchiyama K, Yoshikawa T, Morita M. (2003). Enhancement of acute lung injury related to bacterial endotoxin by components of diesel exhaust particles. Thorax. 58, 605-12.

97

Yunginger JW; Jones RT; Gleich GJ (1976) Studies on Alternaria allergens. II. Measurement of the relative potency of commercial Alternaria extracts by the direct RAST and by RAST inhibition. J Allergy Clin Immunol 58(3):405-13

Zuskin E, Schachter EN, Kanceljak B, Mustajbegovic J, Wiltek TJ. (1994) Immunological and respiratory reactions in workers exposed to organic dusts. Int. Arch. Occup. Environ. Health. 66, 31724.

Published by the Health and Safety Executive 03/07



Inhalation of airborne microorganisms and their associated contaminants in the workplace can cause a range of immunological and respiratory symptoms. The mechanisms through which these respiratory effects are caused are not all fully understood. The evaluation of worker exposure is essential for establishing causal relationships between occupational disease and one or several specific microorganisms or their associated contaminants.

This study investigated the role of microorganisms and their associated contaminants in the development of immunological and clinical response in workers exposed to grain dust. The objectives were:

■ To assess the exposure of grain workers in the UK to inhalable grain dust, the microbial contaminants in grain dust, including identification of the predominant microorganisms involved, and to endotoxin (bacterial cell wall toxins).

■ To measure the prevalence of immunological response to grain dust associated allergens in UK grain workers.

■ To examine the long term clinical and immunological effects of workplace exposure to grain dust and its contaminants in terms of its effect on respiratory health. This was done by establishing a cohort of 321 workers exposed to grain dust (farmers at 27 farms and dock workers at 2 docks in South East England) and maintaining as many as possible in the cohort for repeated immunoassay and clinical assessment over two study phases, Phase 1 from 1990 to 1993 and Phase 2 from 1997 to 2003.

This report and the work it describes were funded by the Health and Safety Executive (HSE). Its contents, including any opinions and/or conclusions expressed, are those of the authors alone and do not necessarily reflect HSE policy.

RR540

www.hse.gov.uk

Date post:	01-Sep-2018
Category:	Documents
Upload:	nguyenque
View:	221 times
Download:	0 times

RR540 The HSE Grain Dust Study · Health and Safety Executive The HSE Grain Dust Study workers'...

Documents