Since the low point of 1977 the incidence of infection attributed to Salmonella typhimurium in England and Wales has shown a steady increase with a total of 4956 cases reported in 1982 (OPCS, 1985). Although the number of general outbreaks and family outbreaks increased by 24 % and 54 % respectively between 1981 and 1982, the greatest increase (55%) was in the number of sporadic cases reported (PHLS, 1984). The cause of the recent increase in human infection is not known but may be associated with bovine infection and an explanation for the rise in sporadic cases could be that they comprise unrecognized outbreaks due to common food products distributed over wide geographical areas for long periods of time because of deep freezing (Galbraith, 1985). This hypothesis has prompted the Communicable Disease Surveillance Centre to mount a case-control study comparing sporadic cases of S. typhimurium infection with an unaffected control group in several parts of the country to find the vehicles of infection. Such an approach requires precise strain identification and detailed epidemiological information.

An enzyme-linked immunosorbent assay (ELISA) for the rapid diagnosis of antibiotic-associated colitis (AAC) is presented. Commercially available antisera to Clostridium difficile toxins contain antibodies to other antigens found in non-toxigenic C. difficile and other bacteria. Removal of these unwanted antibodies by absorption increased the specificity of ELISA for detection of C. difficile toxins. Specimens tested included 40 faecal extracts positive for cytotoxicity from cases of AAC, 30 diarrhoeic and 30 well-formed stools negative for cytotoxicity and 50 culture filtrates of toxigenic and non-toxigenic C. difficile and other clostridial species. Use of absorbed sera reduced false-positive reactions observed with faecal specimens frqm 23 to 8 %. About 90 % of specimens that were positive by the tissue culture cytotoxicity test were positive by ELISA using the absorbed sera. The relative merits of ELISA and other methods for the rapid diagnosis of AAC are discussed.

The isolation rate for Clostridium difficile in diarrhoeal stools was investigated in patients from general practice and community health centres over a 14-month period. C. difficile or its cytotoxin was detected in specimens from 89 (4·7 %) of 1882 patients studied and accounted for 30·3 % of all enteropathogenic micro-organisms isolated. Overall C. difficile was second only to Giandia lamblia in frequency. Recovery rates in the different groups of patients surveyed varied from 3·6 to 27·5 %. The relationship between stool culture results and stool cytotoxin assay also varied considerably between groups of patients studied. Coincident infections with a variety of enteropathogenic bacteria and intestinal parasites were diagnosed in 14 of the 89 patients. It was concluded that laboratories servicing this type of practice should be aware that C. difficile may be a cause of diarrhoea. An adequate clinical history should facilitate proper processing of the specimen.

A microbial survey of Sri Lankan desiccated coconut has been made on material purchased in supermarkets in Sheffield or on material obtained directly from the processing company. The total viable count (TVC) was reduced by spoilage and pasteurization from 104/g to 103/g. Most samples contained low levels of coagulase-positive Stephylococcus aureus suggesting that this commodity had been handled during production. One focus of contamination with Aspergillus flavus was found for each 8·34 g of desiccated coconut (mean contamination). The number of bacteria and moulds in spoiled coconut was significantly lower than that in coconut obtained from the processor or purchased from retail outlets. It is suggested that the accumulation of free fatty acids, aliphatic methyl ketones and secondary alcohols produced during fungal spoilage has had a bactericidal and fungicidal effect. The use of microbial specifications for foods is questioned in situations where there is evidence of microbial spoilage having taken place.

Controlled parallel experiments were performed on the Vienna test model for the evaluation of procedures for hygienic hand-disinfection in three laboratories (Vienna, Mainz, Birmingham). The degerming activity of four procedures, each taking 1 min, was assessed repeatedly and compared with that of a standard disinfection procedure (ST) using isopropanol 60 % (v/v). The mean log reductions (mean log RF) for each procedure were as follows: n−propanol 50% (v/v) 4·85 and 5·14 in Vienna (V) and Mainz (M) respectively, ethanol 70 % (v/v) + chlorhexidinegluconate 0·5% (w/v), 4·01 (V), 3·76 (M) and 4·00 in Birmingham (B). Washing procedures were less effective, mean log RF 's of 3·19 (V), 3·49 (M) and 3·04 (B) were obtained with povidone-iodine soap, and 2·91 (V), 3·37 (M) and 3·27 (B) with a liquid phenolic soap. Analysis of variance on the data from Vienna and Mainz revealed significant differences of means not only between procedures (‘preparations’) but also on repeat testing. To compensate for the influence of variables such as test subjects, laboratory and day, the Vienna test model provides a method of standardization by testing a ST in parallel with the test procedure (P).

Standardization of the results was obtained by pair-wise substraction, log . Analysis of variance on the resulting values demonstrated that comparability of the results between laboratories and on repeat testing was achieved. The relative variation of the measurements within the laboratories ranged from 0·9 to 4·2%. As assessed by power-analysis, a disinfection procedure will be detected as significantly (P= 0·1) inferior to the standard processes in 95 of 100 experiments if it produces a mean log RF that is at least 0·55–0·65 log units smaller than that of the standard.

Serogroup-specific antigen was extracted from a number of Legionella pneumophila strains and compared with phenol-water extracted lipopolysaccharide on the basis of gel filtration, chemical analysis, SDS-PAGE and reaction with serogroup-specific antibody in immunoblots. Serogroup specificity is apparently borne by the O side-chains of the lipopolysaccharide, which, although smooth in type, partitions in the phenol phase. For four serogroup 1 strains tested, there was no qualitative correlation between O side-chain length and pulmonary virulence for guinea-pigs.

The expression of toxigenicity by Vibrio cholerae, before and after exposure to various conditions of salinity, pH and cation composition and concentration, has been measured. Exposure to these conditions did not select for hyper- or hypo-toxigenic strains. This suggests that toxigenic V. cholerae O 1 are unlikely to lose their toxigenicity when exposed to environmental stress and that V. cholerae toxin production is not a response to the stresses included in this study. These results are consistent with an aquatic reservoir for toxigenic V. cholerae O 1.

Quantitative bacterial counts, concentrations of bile acids and bacterial enzyme profiles were measured in faecal samples from three tribal groups with distinctive dietary patterns in a rural area of northern Nigeria. Samples were obtained from Maguzawa (with a predominantly cereal diet), Hausa (cereal with regular meat consumption) and Fulani (cereal with frequent consumption of cows' milk). Numbers of bacteroides, clostridia and concentrations of bile acids were lower in the Maguzawa than the other dietary groups but these differences were not statistically significant. Bacterial enzyme profiles in each group were similar. The results are discussed with respect to possible influences of diet on the human intestinal microflora.

An outbreak of staphylococcal food poisoning in Europe caused by contaminated lasagne was detected and monitored by both national and international surveillance systems. The common source was a pasta-producing factory in Italy and high levels of Staphylococcus aureus were detected in packets of dried lasagne distributed in Luxembourg, the UK, France and Italy. Forty-seven cases were reported in the UK. Outbreaks of staphylococcal food poisoning attributed to mishandling during the food processing stage are uncommon and pasta as the food vehicle is rare. Prompt recognition of the outbreak and rapid identification of the food vehicle enabled most of the consignment to be withdrawn from the market.

Natural water sources used as drinking-water supplies by rural settlements in Sierra Leone were examined monthly over a one-year period to detect any seasonal variations in bacterial quality. The 37 °C colony count, levels of selected faecal indicator bacteria and the incidence of Salmonella spp. were monitored. A seasonality was demonstrated for all the variables, counts generally increasing with the progression of the dry season, culminating in peaks at the transition from dry to wet season. Some complications with respect to the interpretation of counts of faecal indicator bacteria from raw tropical waters are noted.

Duplicate specimens were taken with cotton-wool swabs from the upper tarsal conjunctiva of 63 patients living in Gambian villages in which trachoma is endemic and from 34 infants with ophthalmia neonatorum (ON) attending an outpatient clinic in The Gambia. The detection of Chlamydia trachomatis by direct immunofluorescence (IF) using a conjugated monoclonal antibody to its principal outer membrane protein was compared with isolation in cycloheximide-treated McCoy cells. For trachoma, the sensitivity and specificity of the immunofluorescent technique were 62% and 100% respectively if ten elementary bodies (EBs) was taken as the minimum requirement for positivity by IF. If all cases with one or more EB were considered positive, the sensitivity was 81% and the specificity 85%. For ON the sensitivity and specificity were 100% and 95% respectively, regardless of which criterion was used. In view of its simplicity and easy applicability to field conditions it seems likely that direct IF using monoclonal antibodies may be a useful technique for the detection of C. trachomatis in the conjunctival epithelium of patients with trachoma.

Two hundred and fifty clinical strains of Candida albicans and six isolates from a cross-infection outbreak were studied for their ability to assimilate 19 carbohydrates in the API-20C system (API Laboratory Products Ltd., Basingstoke, UK). The assimilation profiles were stable on repeat testing; at intervals in the 72 h duration of the test; and when incubated at different temperatures.

Although not a complete system of biotyping, the API-20C system shows high typability, and fair reproducibility and discrimination, having a limited role in indicating which isolates should be typed by more elaborate methods.

An outbreak of cryptosporidiosis in a circumscribed semi-rural population is described; some cases were infected concurrently with campylobacter. The results of a detailed case-control study and environmental surveillance are discussed.

A two-year prospective survey of patients with diarrhoea at a children's hospital detected 65 cases of cryptosporidiosis; 56 were index cases representing 1·4% of patients examined. There was a marked seasonal variation with fewer index cases in July–September (0·4–0·6%) than February–April (3·5%). All four adult cases were contacts of infected children and 21 (32%) of the patients were 12 months old or younger. Cryptosporidium was the fourth commonest diagnosed cause of gastroenteritis, the second most common non-viral cause, was responsible for 7·2% of all cases, and 14·7% of non-viral cases where a cause was identified. As such it should be sought routinely in cases of gastroenteritis in children.

A serological survey carried out on 1379 school-children from Skye and Mallaig in the N.W. of Scotland failed to show evidence of subclinical cases of hydatidosis. There appeared to be a relatively low level of exposure to the parasite, as judged by ELISA and IHA tests. Thirty-one children reacted positively in the ELISA; only one child had a transient positive result in the IHA test. Surveillance of these children will be continued for the present but the duration required is uncertain. The local population appeared to be well aware of the mode of spread of the parasite and the methods for its control.

The epidemiology of hydatid disease in man in England and Wales, based on the data collected between 1981 and 1983 at the Hydatid Reference Laboratory of the Public Health Laboratory Service, is presented. The incidence of hydatid disease was 42 cases per annum, with 2 cases per million population occurring in Wales and 0·2 per million in England. The highest prevalence was in London amongst the immigrant population. The incidence in the indigenous population was 15·5 cases per annum, with 5·6 cases per million occurring in South Powys, parts of South Wales and Herefordshire. The population most at risk is the rural, farming community in that area, but the West Midlands is also an apparent focus of the disease.