Ps Burge

Diagnosis of occupational asthma from time point differences in serial PEF measurements

Background: The diagnosis of occupational asthma requires objective confirmation. Analysis of serial measurements of peak expiratory flow (PEF) is usually the most convenient first step in the diagnostic process. A new method of analysis originally developed to detect late asthmatic reactions following specific inhalation testing is described. This was applied to serial PEF measurements made over many days in the workplace to supplement existing methods of PEF analysis. Methods: 236 records from workers with independently diagnosed occupational asthma and 320 records from controls with asthma were available. The pooled standard deviation for rest day measurements was obtained from an analysis of variance by time. Work day PEF measurements were meaned into matching 2-hourly time segments. Time points with mean work day PEF statistically lower (at the Bonferroni adjusted 5% level) than the rest days were counted after adjusting for the number of contributing measurements. Results: A minimum of four time point comparisons were needed. Records with ⩾2 time points significantly lower on work days had a sensitivity of 67% and a specificity of 99% for the diagnosis of occupational asthma against independent diagnoses. Reducing the requirements to ⩾1 non-waking time point difference increased sensitivity to 77% and reduced specificity to 93%. The analysis was only applicable to 43% of available records, mainly due to differences in waking times on work and rest days. Conclusion: Time point analysis complements other validated methods of PEF analysis for the diagnosis of occupational asthma. It requires shorter records than are required for the Oasys score and can identify smaller changes than other methods, but is dependent on low rest day PEF variance.

P224 How do the timings of reactions during specific inhalation challenge relate to real world exposures in occupational asthma

Timings of asthmatic reactions following specific inhalation challenge (SIC) have been defined as immediate, late, dual and prolonged immediate. How they translate into usual workplaces exposure is unknown. We postulated that those with an immediate component would start to react within one hour of regular occupational exposure and start to recover within one hour of leaving work, whereas those with a late reaction would have delayed starting and recovery. Those with dual or prolonged immediate reactions would show early deterioration and delayed recovery. We have compared the timings of reactions in 48 consecutive workers who had positive SIC and had kept serial PEF records at home and with real-world work exposures. These were analysed by the ABC plot from the Oasys analytical program which combines all measurements done on different work days into 2-hourly blocks and produces plots similar to that seen with SIC, with days away from work as the control exposure. Four experts independently scored the ABC plots to identify workplace deterioration starting at the first timepoint after the start of work, or ≥2 hours later (delayed deterioration), and starting to recover at the first timepoint after leaving work, or ≥2 hours later (delayed recovery). Records with disagreements were resolved in a joint meeting when all records available for an individual worker were compared. The relationship between SIC and real-world exposures is shown in the table. The relationship between laboratory and workplace reactions was only modest, complete concordance in 44%. Exposures may vary from day to day at work, or that the first reading at work was made before significant exposure has occurred; workers are instructed to make the last reading before work immediate before entering the workplace which should mitigate this. For those with immediate reactions alone during SIC, more showed deterioration in the first workplace reading than showed early recovery after leaving work. Abstract P224 Table 1 Comparison of timing of asthmatic reactions in the challenge chamber and in real world exposures at work Laboratory challenge reaction Start of deterioration Start of recovery Immediate Delayed Immediate Delayed Immediate (n=19) 13 6 9 10 Late (n=10) 4 6 4 6 Dual or prolonged immediate (n=20) 10 10 8 12

P156 Maintaining patients with idiopathic pulmonary fibrosis (ipf) on antifibrotic therapy; the nurses’ challenge

Pirfenidone and nintedanib are the first licensed drugs for IPF. Both reduce the rate of disease progression but have significant intolerability issues limiting long-term use. This study aims to identify areas where changes in practice might improve outcomes. We compared baseline characteristics of disease severity with the reasons for stopping treatment and the proportion stopping treatment within the first 4 weeks, for all patients from our centre up to December 2016, including those receiving treatment on a named patient basis. Data was available for 150/153 patients started on pirfenidone and 56/57 on nintedanib. Table 1 compares patients who were dispensed treatment but never took it, patients who stopped after the first prescription, and those continuing treatment. 8 patients consented to treatment but never took the drug; they were older and had less advanced disease. 39 patients stopped treatment after the first prescription; they had slightly more advanced disease, and had a longer interval between drug delivery and the first nurse (60.7 SD 57 days vs. 45.4 SD 30; (p=0.032), and more often had a shared care arrangement, but did not have a lower BMI. Photosensitivity for pirfenidone and diarrhoea for nintedanib were usually managed by treatment and were uncommon reasons for stopping antifibrotic therapy, which was more often related to upper GI intolerance and disease progression. Abstract P156 Table 1 Comparison of patients dispensed pirfenidone or nintedanib for IPF who never took the dispensed medicine, who stopped after the first prescription, or who continued treatment after the first prescription. *p<0.05, **p<0.01, ***p<0.001 Prescription dispensed Once and never taken Once and not repeated continued Number 8 39 159 FVC% predicted (SD) 89.5 (27) 69.3 (16) 74.8 (14) ** Composite physiology index 36.4 (16) 56.0 () 48.9 (10) *** Shared care% 38 20 ** Age 78.0 (6) 72.5 (7) 70.4 (9) * BMI 30.0 (2) 28.5 (6) 29.2 (5) Conclusion Patients stopping antifibrotic treatment early cannot be identified from baseline data but can be reduced by intensive nursing support. Prescribing centre based ILD-CNS’s should be responsible for early treatment tolerability and aim to see patients monthly until established on treatment. Uncertain delays between prescription and drug delivery make this more difficult.

S106 Distribution of occupational and non-occupational causes in hypersensitivity pneumonitis diagnosed by an interstitial lung disease expert panel

Introduction The aim of this study was to describe the demographic, clinical, exposure history and investigatory Results of hypersensitivity pneumonitis (HP) cases diagnosed by an expert regional NHS interstitial lung disease (ILD) board, to determine the extent and distribution of occupational aetiology. Methods All HP cases identified from a database over a 13 year period were included. The demographic, clinical, exposure history and investigatory Results were reviewed from clinical notes. Cases were categorised in to three groups: occupational, non-occupational and no known cause. Hypothesis testing at the 95% confidence level was used to identify significant differences between the 3 groups. Results A total of 127 cases (13 occupational; 34 non-occupational; 80 no known cause) of HP were identified. Men were more likely to experience occupational HP (p=0.029) compared to the other groups. Occupational HP cases were younger (p=0.002), more likely to experience weight loss (p=0.004), to have systemic symptoms (p=0.007) and a recurrence of symptoms (p<0.001). Occupational HP were due to metal working fluids (MWF), birds, mould/fungi, Farmer’s lung, and cleaning and treatment sprays. Non-occupational HP were due to birds, mould/fungi in the home and medication. Percentage lymphocyte count in broncho-alveolar lavage (BAL) was significantly raised in occupational HP (p=0.001). Where no causative agent was identified, there was a greater absence of exposure history (13.2%–32.5%). Conclusion Occupations where there is exposure to birds or those working in trades where industrial processes and the use of chemical compounds predominate, are at risk of occupational HP. Birds remain an important cause of non-occupational HP. Clinical features such as weight loss, systemic symptoms and recurrent symptoms should raise a suspicion of an occupational cause of HP. BAL remains an important investigatory tool in HP, especially in occupational HP. An exposure history, especially an occupational history, is mandatory when assessing suspected cases of HP.

Histologist's original opinion compared with multidisciplinary team in determining diagnosis in interstitial lung disease

Guidelines recommend that multidisciplinary interstitial lung disease meeting (ILD MDT) decisions become the gold standard for diagnosis, replacing the histologist from this position, and identify this as requiring supportive evidence. We have compared diagnoses from lung biopsy material made by expert histologists with the subsequent consensus opinion from a properly constituted ILD MDT in 71 consecutive patients referred to a regional thoracic unit. MDT changed the original histological diagnoses in 30% (95% CI 19.3% to 41.6%) and strengthened the diagnoses from probable to confident in a further 17% (95% CI 9.1% to 27.7%). The assessment of hypersensitivity pneumonitis, non-necrotising granulomas and organising pneumonia accounted for the majority of the changes.

P123 What causes occupational asthma in cleaners

Domestic cleaners show a consistently raised risk of developing asthma in population-based epidemiological studies. The main risks appear to be the use of bleach and sprayed cleaning agents. These risks are now found in other occupations, such as healthcare due to the requirement for disinfection of clinical areas. Since 2000, there has been 26 cleaners and healthcare workers with occupational asthma (OA) reported to Shield (West Midlands registry for OA) thought to be due to cleaning agents. The majority of these were using chlorine-releasing substances containing dichloro-isocyanurate for disinfection purposes. We have carried out 8 specific inhalation challenge tests to chlorine-releasing tablets mixed with cold water but none of these have elicited an asthmatic reaction despite the workers previously showing work-related changes on serial peak flows when exposed. In a separate worker, a positive reaction was seen when challenged to a washing up liquid containing triclosan (another chlorine based substance). Due to the negative nature of the majority of tests, we considered the possibility that the chlorine itself may not be the provoking factor, but that chlorine-releasing formulations have the potential to react with amines from urine or bodily fluids when they are being used to disinfect clinical areas. We report a 48 year-old health-care assistant for the elderly exposed to Haztabs (a chlorine-based cleaner) at work who had clear evidence of occupational asthma from serial peak expiratory flow records (Oasys score 3.44, ABC score 40 L/min/hr, 5 positive timepoints), but negative tests to Haztab exposure when mixed with cold water. Adding urine (5%) to the Haztab mixture resulted in a dual immediate and late asthmatic reaction (Figure 1). She showed mild airway hyper-reactivity pre challenge which was not repeated post challenge due to a continued low FEV1 of 39% predicted. Conclusions The SIC results suggested that chloramines were the cause of her asthma, similar to the agents causing occupational asthma in workers exposed to indoor swimming pool vapours/mists. Abstract P123 Figure 1 Specific inhalation challenge testing showed no asthmatic response to hartabs or urine alone. The chlorine urine mixture provoked a dual asthmatic reaction and sustained drop in FEV1 of up to 34% from base line. The grey box denotes the exposure.

P11 Pirfenidone treatment is only available in the UK for a minority of patients with usual interstitial pneumonitis

Introduction Usual Interstitial Pneumonitis (UIP) may be caused by asbestos exposure (asbestosis), collagen vascular diseases (CVD) or agents causing hypersensitivity pneumonitis (HP), or may be idiopathic (IPF). Referrals to our ILD MDT have increased since the limited availability of pirfenidone which is only prescribable for IPF with a FVC 50–80% predicted. We have reviewed presentation in 2014 to identify patients suitable for Pirfenidone prescription. Methods Our hospital provides a regional service for Pirfenidone prescription. All patients with a MDT diagnosis of definite UIP were included. A standard proforma requested information on exposures, CVD and antibodies relevant to CVD and HP. If there were no relevant exposures and CVD was excluded, a diagnosis of IPF was made. Asbestosis, HP and CVD associated UIP were diagnosed when the relevant information supported this, unspecified UIP was diagnosed in the absence of specific information. Results 202/546 referrals in 2014 were judged to have definite UIP after consideration by a fully constituted ILD MDT including histopathologists, radiologists, clinicians and CNS’. After exclusion of 22 with asbestosis, 11 with CVD associated UIP, 17 with CPFE and 17 with HP associated UIP, 51 with IPF and 84 with unspecified UIP remained for consideration of Pirfenidone treatment. Only 64/135 suitable patients had a FVC 50–80% predicted (Figure 1).Abstract P11 Figure 1 Conclusions The decisions of an ILD MDT are limited by the completeness of investigation. We found causes for 50/200 patients with UIP. Pirfenidone was not prescribable for 53% of otherwise suitable patients. If the FVC limit was raised to 90% 32% would still be excluded, including 2 who died of their disease within 12 months. The FVC is often preserved even in terminal IPF.

P131 Understanding Health Beliefs And Behaviour In Workers With Occupational Asthma

Introduction Long delays from symptom onset to the diagnosis of occupational asthma have been reported in the UK, Europe and Canada and workers are often reluctant to seek medical help or workplace solutions for their symptoms.1,2 Reducing this delay could improve workers’ quality of life, and reduce the societal cost of occupational asthma. This study aimed to explore reasons behind such delays. Methods A purposive sample of 20 individuals diagnosed with, or under investigation for, occupational asthma (median age=52; 70% male; 80% white British) undertook a single semi-structured interview. Interviews were transcribed verbatim and thematic analysis was undertaken in order to explore health beliefs and identify barriers to diagnosis. Results Four themes were identified: (1) workers’ understanding of symptoms, (2) working relationships, (3) workers’ course of action and (4) workers’ negotiation with healthcare professionals. Understanding of symptoms varied between individuals, from a lack of insight into the onset, pattern and nature of symptoms, through to misunderstanding of what they represented, or ignorance of the existence of asthma as a disease entity. Workers described reluctance to discuss health issues with managers and peers, through fear of job loss and a perceived lack of ability to find a solution. Conclusion The evolution of workers’ understanding depended upon how actively they looked to define symptoms or seek a solution. Proactive workers were motivated to seek authoritative help and negotiate inadequate healthcare encounters with GPs. In summary there appear to be a number of key influences motivating a worker to seek an explanation for their symptoms or a definitive solution, which are represented in the model in Figure 1. Understanding workers’ health beliefs will enable policy makers and clinicians to develop better workplace interventions for identifying occupational asthma. References Fishwick D et al. Prim Care Respir J 2007;16: 304–10 Poonai N et al. Can J Public Health 2005;96:230–3 Abstract P131 Figure 1 The major influences on workers’ health-seeking behaviour

P133 Longitudinal Decline In Fev1 In Occupational Asthma Due To Irritants Is Not Altered By Removal From Exposure

Introduction Airborne irritants and allergens in the workplace can induce and trigger occupational asthma (OA). OA due to sensitisation is associated with an accelerated rate of decline in FEV1 (100 mls/yr) compared with healthy controls (25 mls/yr)1, which improves on removal from exposure. We sought to describe the rate of FEV1 decline in patients with irritant-induced OA before and after their removal from exposure. Methods Cases of irritant-induced OA reported between 1991 and 2011 were identified from the SHIELD database (a voluntary reporting scheme for OA) and their demographic characteristics and serial FEV1 measurements were gathered. Generalised estimating equations with an exchangeable correlation structure were used to calculate an average rate of FEV1 decline for all patients before and after removal from exposure. Results A total of 526 FEV1 readings (179 prior to removal, 347 post removal) were gathered from 52 patients. 30 patients had FEV1 data both before and after removal; 5 patients had FEV1 data only before removal, and 17 patients had FEV1 data only after removal; demographics were not significantly different between the groups. FEV1 decline prior to removal of the irritant was 44 ml/year (95% CI: 32–58) and FEV1 decline after removal was 49 ml/year (95% CI: 36–62). There was no significant difference between the intercepts of the two lines, implying no improvement in FEV1 after removal from exposure. Conclusion In this cohort, irritant-induced OA was associated with an accelerated decline in FEV1, which persisted after removal from the irritant. These results might be attributed to differences in the underlying pathology of sensitisation and irritant-induced OA, differences in patient behaviour, or differences in treatments offered to the two groups. Reference Anees W et al. FEV1 decline in occupational asthma. Thorax 2006;61:751–5

P205 21 years of SHIELD: decreasing incidence of Occupational Asthma in the West Midlands, UK?

Introduction Notifications of occupational asthma (OA) to the West Midlands SHIELD reporting scheme have declined between 1991 and 2011. This may be due in part to reporter fatigue or restrictions on reporting, under-recognition, or true reduction in incidence of OA due to workplace control measures. We aimed to describe trends in reports of OA to the SHIELD database over a 21-year period and investigate reasons for any changes. Methods Descriptive statistics were performed on demographic and annual notification count data (for total notifications and individual causative agents). Count data were scaled to give a count per million workers using West Midlands’ mid-year population estimates. A number of statistical analyses were undertaken to calculate trends in annual notifications: (i) non-parametric Kendall tau-b correlation (reference method), (ii) a negative binomial regression model, calculating incident rate ratios (IRRs), including a step-change analysis, and (iii) a logistic regression model calculating annual reporting odds ratios for 13 common causative agents. Results 1637 cases of OA were notified between 1991 and 2011. Most cases were notified from the Birmingham Heartlands specialist clinic and notifications made elsewhere in the West Midlands fell from 16 per million workers in 1995 to zero in 2004, with very few cases (4 per million workers) after that. A significant non-linear decrease in annual total notifications was observed over the study period (IRR = 1.056; 95% CI = 1.012–1.102; p = 0.012), as was the case for most causative agents studied. However cleaning products showed a 6% increase in notifications year-on-year (IRR = 1.056; 95% CI = 1.012–1.102; p = 0.012; Figure 1). Although the incidence of isocyanate-related OA decreased significantly after 2005, the proportion of annual notifications due to isocyanates relative to other agents increased year on year (tau = 0.43; p = 0.007). Conclusions Falling incidence of OA between 1991 and 2011 is explained in part by under-recognition by healthcare professionals, and by reporter fatigue, though for some causative agents this may be attributed in part to better workplace controls. Isocyanates remain the most common cause of OA but reports have fallen recently along with colophony, latex and glutaraldehyde, which have almost disappeared. Conversely cleaning product related OA is increasingly recognised. Abstract P205 Figure 1. Annual notifications of occupational asthma due to cleaning products (cases per million workers) recorded 1991–2011 in West Midlands UK, with prediction line based on negative binomial regression analysis.