ABSTRACr Employees at six sewage treatment plants and three drinking water plants were interviewed for the presence of specific medical symptoms. Serum immunoglobulin concentrations, white blood cell counts and fibrinogen degradation product concentrations (FDP) in urine were determined as were the number and species of airborne Gram negative rods in order to characterise exposure to aerosols of sewage water. The highest number of bacteria was found in areas where the sewage water was agitated. A significantly higher proportion of employees at sewage treatment plants reported skin disorders, diarrhoea, and other gastrointestinal symptoms than the control group. No significant differences were found between the groups for white blood cell count or serum immunoglobulin concentrations, except that IgM concentrations were slightly higher in the sewage workers. Some workers had serum transaminase concentrations in excess of normal; some of these returned to normal after the summer holiday. Among non-smokers a higher proportion of sewage treatment workers had increased amounts of FDP in urine. It is conceivable that the symptoms observed were caused by toxins from Gram negative bacteria.The occupational health risks of sewage workers have been investigated by several authors, and reviews have been presented by Anders' and Clark.2The emphasis in most investigations has been on the risk of contracting various types of infectious diseases.A previous study on sewage workers reported that when sewage sludge was heat-dried, employees experienced attacks of fever, chills, and eye inflammation after exposure to high concentrations of the dust.3 The concentrations of serum immunoglobulins and fibrogen degradation products (FDP) in urine were increased in exposed workers. The symptoms were interpreted as being caused by the endotoxin in the dust, originating from Gram negative bacteria in the sewage sludge.This present paper reports a study undertaken to determine whether specific symptoms were also present among sewage workers in plants where sludge was treated by conventional methods. Workers in drinking water plants served as a reference population. The presence of medical symptoms was studied in interviews and serum immunoglobulin concentrations, white blood cell counts, and FDP concentrations ip the urine were measured. The exposure to sewage water aerosols was characterised by measur-

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.

An outbreak of 18 pneumonia cases caused by Legionella pneumophila serogroup 1 occurred at a Swedish university hospital 1996 to 1999. Eight clinical isolates obtained by culture from the respiratory tract were compared to 20 environmental isolates from the hospital and to 21 epidemiologically unrelated isolates in Sweden, mostly from patients, by using pulsed-field gel electrophoresis (PFGE), amplified fragment length polymorphism analysis (AFLP), and monoclonal antibody (MAb) typing. All patients and most environmental isolates from the outbreak hospital belonged to the same genotypic cluster in both PFGE and AFLP. This genotype was distinctly different from other strains, including a cluster from a second hospital in a different part of the country. The MAb subtype of the outbreak clone was Knoxville except for three isolates that were Oxford. A variation in the MAb reactivity pattern was also found in a second genotypic cluster. These changes in the MAb reactivity pattern were due to the absence or presence of the lag-1 gene coding for an Oacetyltransferase that is responsible for expression of the lipopolysaccharide epitope recognized by MAb 3/1 of the Dresden Panel. In all MAb 3/1-positive strains, the lag-1 gene was present on a genetic element that was bordered by a direct repeat that showed a high degree of sequence homology. Due to this homology, the lag-1 gene region seemed to be an unstable element in the chromosome. MAb patterns are thus a valuable adjunct to genotyping methods in defining subgroups inside a genotypic cluster of L. pneumophila sg 1.

Bacterial contamination of various parts of the cotton plant and of cotton from different mills was investigated. The predominant bacterial species were Gram-negative rods mainly of the Enterobacter genus. When guinea pigs inhaled strains of these bacteria cultivated from cotton, a strong leucocyte mobilising capacity was found for Pseudomonas and Enterobacter but not for Agrobacterium or Bacillus species. The aetiology of the development of pulmonary symptoms after inhalation of bacteria-containing dusts and subsequent production of endotoxins is discussed.The presence of airborne Gram-negative bacteria in cotton mills was first reported by a commission investigating working conditions in cotton mills in Great Britain (Home Office, 1932 (Rylander and Snella, 1976). The leucocyte mobilising reaction occurs in several different animal species (Walker et al., 1975) and in man (Merchant et al., 1975). A close correlation was found between the number of airborne Gram-negative bacteria and the prevalence of byssinosis in an epidemiological study in different cotton mills (Cinkotai et al., 1977).In view of the apparent importance of the bacterial contamination of cotton in the development of pulmonary symptoms, the bacteria on different cotton plants and bale cotton were studied. The number of viable bacteria and the species were determined. The reactions after acute exposure to bacteria typically found on cotton were studied in animal experiments. Materials and methods BACTERIAL CULTURE OF COTTONSamples of cotton plants, bale cotton and cotton from blending machines, card rooms and waste from cards were collected from different cotton mills. One gram of the sample was suspended in 10 ml 0 9 % NaCl and shaken vigorously. The liquid was squeezed out of the sample, then put back and shaken again. Serial dilutions of the extracts were plated on nutrient agar. blood agar or Drigalski agar selective for Gram-negative rods and incubated at 37°C overnight. The total number of colony-forming units (cfu) was counted. Pilot experiments showed that incubation at 200C and 300C gave the same number of cfu.All samples were prepared in duplicate and the results were expressed as the mean number of cfu/g 204

The endotoxin concentration in air was measured in farms where 11 farmers had experienced febrile reactions or allergic alveolitis and in a random sample of farms with 17 symptomless farmers. Samples were obtained during normal dairy farming in eight reference farms (background samples) and in all farms during the handling of material which had probably caused symptoms or disease or, in reference farms, maximal spore exposure (worst case samples). In addition, parallel samplers were used in reference farms, one with a cyclone (5 ,um cutoff) and one without, to measure the dust and endotoxin concentrations in the respirable fraction and total dust. The endotoxin worst case values varied from < 0 01 to > 50 yg/m3 in symptom farms (median 6-4 pg/m3, geometric mean 2 2 ,g/m3) and from < 0 01 to > 50 ,g/m3 in reference farms (median 42 Mg/m3, geometric mean 29 ,ug/m3). This difference was not statistically significant. The background values in reference farms were 1*3 (median) and 04 (geometric mean) pg/m3. The differences between samples with and without cyclone and between background and worst case samples were statistically significant (p < 0-02). About 75% of the activity was found in the non-respirable fraction. No correlation was found between exposure to endotoxin and symptoms in farmers. There were weak, but statistically significant, correlations between endotoxin concentrations and total spore count or dust concentrations. The surprisingly high endotoxin values in the respirable fraction of air from environments which apparently did not cause symptoms raises the concern that the Limulus amebocyte assay might be sensitive to other components in the dust rather than endotoxin.