The relation between the health of workers and the environment in swine confinement buildings was investigated in a study of 57 workers on 30 swine farms in southern Sweden and 55 matched controls. Swine workers reported significantly higher frequencies of respiratory symptoms, more frequent colds and absence due to chest illness, and a history of pneumonia. The increased frequency of symptoms of respiratory disease was related to the number of years and per cent of the day spent working with swine. Symptoms were also associated with respirable dust, total dust, endotoxin in total dust, and number of microbes in the air of the work environment.

Summary Three cases of humidifier fever were detected in an office environment. Flavobacteria were found in the contaminated water in a humidifier. After an experimental exposure, the three persons with previous symptoms suffered from fever and slight respiratory symptoms. A leucocytosis and an increase in the number of segmented white blood cells were found the day following the exposure. General immuno‐globulins as well as antibodies to Flavobacterium and endotoxin were slightly elevated in the exposed group. The possibility that endotoxins may be the causative agent by means of an indirect complement activation is discussed.

The presence of humidifier disease in a printing factory was investigated with particular emphasis on airborne endotoxins. The water in the humidifier was contaminated with Pseudomonas. The amount of airborne endotoxin when the humidifier was operating was 0-13-0-39 jug/m^ Twenty of the fifty workers investigated reported typical symptoms of fever, chills, and chest tightness when the humidifiers were operating. Symptoms were more frequent among non-smokers.The estimated inhaled dose of endotoxin was found to be sufficient to cause the observed symptoms. The determination of airborne endotoxins in future episodes of humidifier disease is recommended.

ABSTRACr Changes in respiratory function (FEV,) and polymorphonuclear neutrophils (PMN) on nasal epithelium were studied in 68 students and 39 cotton mill workers in an experimental cardroom. The exposure was characterised by the vertical elutriator dust and endotoxin levels. A dose related decrease was found for FEV, which was more pronounced in smoking workers. The thresholds for no FEV, reaction were 0-58 mg/i3 dust and 0*17 ,g/m3 endotoxin for students and 0-43 mg/m3 and 0.08 ug/m3 for smoking workers. (FEVy) and the number of Gram-negative bacteria in the processed cotton. 5 The effect has been related to bacterial lipopolysacchrides (endotoxin) present in the outer cell wall of Gram-negative bacteria.6 In some studies a larger reaction has been found among smokers compared with non-smokers at the same exposure levels.2To explore further the reactions among previously unexposed subjects and cotton mill workers after exposure to cotton dust studies were undertaken in an experimental cardroom where the dust level could be varied. Workers from cotton mills and students not previously exposed to cotton dust were used as test subjects. FEV, was measured before and after a four hour working session which for the cotton mill workers took place on a Monday. The number of polymorphonuclear leukocytes (PMN) in the nasal secretion was determined. Information on subjective symptoms such as chest tightness and fever was obtained from questionnaires.Received 10 January 1983. Accepted 11 July 1983 Material and methodsThe study was performed in an experimental cardroom measuring 6-4 x 5.3 x 2.8 m. A carding machine without a protective cover was situated in the middle of the room and cotton laps were processed. The dust levels in the room could be varied by adjusting the ventilation.In the experiments United States middle grade bale cotton was used. For some of the experiments washed cotton was prepared. The washing procedure consisted of flushing water over a web of cotton on a rayon manufacturing line whereafter the cotton was dried. The maximum temperature during the process was 66°C.The exposure in the cardroom was determined using three vertical elutriators7 placed on different sides of the carding machine. The filters collecting respirable dust were weighed on a semimicro balance (Sartorius 2474) using an alpha-radiating source to prevent static electricity. The mean value from two or three filters was calculated and expressed as mg/i3.After weighing, the elutriator filters were shaken in 10 ml of pyrogene free water, which was serially diluted. The amount of airborne endotoxin was determined using the Limulus amoebocyte lysate (LAL) technique.8 Then 0.1 ml of Limulus lysate (Cape Cod Associates) was added to 0 1 ml portions of dilutions of dust filter water extracts. The last dilution giving a stable clot was read as the Escherichia coli endotoxin equivalent concentration. Dilutions were also prepared with commercial E coli endotoxin (E coli 026-B26 Difco) to assess the 340

A study was performed to assess medical symptoms and pulmonary function among persons working in 30 swine confinement buildings on 28 farms. Data were taken from 57 persons working on these farms and on 55 matched controls.Health data collected included an interview for occupational history, health history, and disease symptoms. Pulmonary function testing was done before and after a daily work period [Ferris, 1978; Gardner, 19791, and a blood sample was collected to test for antibodies to agents in the work environment pelin et al, 19851. The questionnaire used for the interview consisted of two parts. The first part asked for specific symptoms (primarily respiratory symptoms) that the workers attributed directly to their work in swine stables. These questions were asked both before and after the work shift. The first part of the questionnaire was compiled from two separate questionnaires that had been used extensively in previous farm health studies [Donham and Gustafsson, 1982; Donham et al, 1984 a,b, Donham et al, 1986; Kolmodin-Hedman et al, manuscript in preparation]. The second part consisted of the complete American Thoracic Society (ATS) Epidemiology Standardization Project questionnaire [Ferris, 19781, which had been translated into Swedish.The questionnaire had also been used extensively in a similar study in Iowa. The second part of the questionnaire was also used to interview controls, who were farmers or farm workers from the same geographical area, who worked in any type of farming except swine or poultry buildings. These controls were matched to the swine farm workers by age (+5 years), sex, and smoking history (f5 cigarette pack years). Figure 1 summarizes some of the significant findings from the first part of the questionnaire interview. Bronchitis symptoms were very commonly reported, with 60% reporting cough and 40% reporting cough with phlegm. Symptoms of bronchoconstriction were reported by 28% of the workers. Eye, nose, and throat irritation

The prevalence of byssinosis and of chronic bronchitis was studied in a questionnaire investigation among workers in bale opening areas, carding rooms, and spinning rooms in five Swedish cotton mills. Airborne dust and Gram-negative bacteria were measured. Nineteen per cent of the interviewed workers reported symptoms of light byssinosis (grade 1/2). The prevalence of symptoms was not related to the duration of employment, and cases of byssinosis were found among people who had worked in cotton mills for only a few years. A significantly higher proportion of male than female workers reported symptoms. No difference in the extent of byssinosis was found between smokers and non-smokers, but the prevalence was significantly higher among those workers who had ceased smoking. The prevalence of byssinosis was related to the number of airborne viable Gram-negative bacteria as well as to the dust level in the different mills.

Cadmium, mercury, and lead concentrations were determined in deep-frozen kidney cortex biopsies taken from 36 living, healthy Swedish kidney donors (18 males and 18 females), who were 30-71 (mean 53) years of age. Information about occupation, smoking, the presence of dental amalgam, and fish consumption could be obtained for 27 of the donors. The samples (median dry weight 0.74 mg) were analyzed using inductively coupled plasma mass spectrometry, and the results were transformed to wet-weight concentrations. The median kidney Cd was 17 micrograms/g (95% confidence interval, 14-23 micrograms/g), which was similar in males and females. In 10 active smokers, the median kidney Cd was 24 micrograms/g, and in 12 who never smoked, it was 17 micrograms/g. The median kidney Hg was 0.29 micrograms/g, with higher levels in females (median 0.54 micrograms/g) than in males (median 0.16 micrograms/g). Subjects with amalgam fillings had higher kidney Hg (median 0.47 micrograms/g, n = 20) than those without dental amalgam (median 0.15 micrograms;g/g, n = 6), but kidney Hg was below the detection limit in some samples. Nearly half of the samples had kidney Pb below the detection limit. The median kidney Pb was estimated as 0. 14 micrograms/g. This is the first study of heavy metals in kidney cortex of living, healthy subjects, and the results are relatively similar to those of a few previous autopsy studies, indicating that results from autopsy cases are not seriously biased in relation to kidney metal concentrations in the general population. Cd concentrations in those who never smoked were relatively high, indicating considerable Cd intake from the diet in Sweden. The effect of dental amalgam on kidney Hg was as expected, although the reason for the difference in Hg levels between males and females is unclear.

Cadmium, mercury, and lead concentrations were detennined in deep-frozen kidney cortex biopsies taken from 36 living, healthy Swedish kidney donors (18 males and 18 females), who were 30-71 (mean 53) years of age. Information about occupation, smoking, the presence of dental amalgam, and fish consumption could be obtained for 27 ofthe donors. The samples (median dry weight 0.74 mg) were analyzed using inductively coupled plasma mass spectrometry, and the results were transformed to wet-weight concentrations. The median kidney Cd was 17 pdgg (95% confidence interal, 14-23 pg/g), which was similar in males and females. In 10 active smokers, the median kidney Cd was 24 p/g, and in 12 who never smoked, it was 17 pg/g. The median kidney Hg was 0.29 pglg, with higher levels in females (median 0.54 plgg) than in males (median 0.16 pg/g). Subjects with amalgam fillings had higher kidney Hg (median 0.47 pg/g, n = 20) than those without dental amalgam (median 0.15 pgig, n= 6), but kidney Hg was below the detection limit in some samples. Nearly half of the samples had kidney Pb below the detection limit. The median kidney Pb was estimated as 0.14 pg/g. This is the first study of heavy metals in kidney cortex of living, healthy subjects, and the results are relatively similar to those of a few previous