Seventy-three cultures of Clostridium difficile isolated both during, and in the period immediately following, an outbreak of infection in a group of three hospitals, were characterized by one-dimensional sodium dodecyl sulphate–polyacrylamide gel electrophoresis (SDS–PAGE) of whole-cell proteins. Each protein pattern was characterized by the presence of one or two dense bands which were highly reproducible. The protein patterns were used as the basis for a numerical analysis which divided the strains into five phenons (electrophoretic or EP types). The majority, 60 of the 73 cultures, belonged to a single phenon which included strains from both patients and the environment. We conclude that high-resolution SDS–PAGE of proteins provides an effective method for typing C. difficile and therefore for tracing the possible spread of epidemic strains in hospitals and other institutions, thereby allowing a better understanding of the epidemiology of the organism.

The incidence of Clostridium difficile-associated diarrhoea (CDAD) was investigated retrospectively at a 690-bed teaching hospital for the period 1983–92. Our aims were to determine: (i) the distribution by age and sex of patients with CDAD, (ii) the possibility of a seasonal trend and, (iii) the influence of infection control procedures, contamination of the hospital environment and the use of third-generation cephalosporins. The laboratory diagnosis of CDAD was based on demonstration of the organism by stool culture and/or detection of specific cytotoxin in stool filtrates. C. difficile was detected in 917 patients who were being investigated for diarrhoeal illness. Yearly isolations varied from a low of 49 in 1983 to a high of 120 in 1990 (Chi square for linear trend 128·8; P < 0·005). Most patients were elderly, with 63% aged 60 years or more; the majority (59%) were female. The relationship between culture of C. difficile and detection of cytotoxin in faecal extracts was also examined. Sixty percent of a sample of 132 isolates from patients in whom faecal cytotoxin was not detected produced cytotoxin in vitro, suggesting that culture is a more sensitive indicator of infection with C. difficile than cytotoxin detection. When the total number of faecal specimens received in the laboratory was used as a denominator there was an increase in the number of incident cases of CDAD between 1983 and 1990, apart from 1986. When occupied bed days was used as the denominator a similar trend was observed with a peak in 1990. These increases correlated with an increase in the use of third-generation cephalosporins at SCGH between 1983 and 1989 (Pearson's correlation coefficient, 0·90). The introduction of Body Substance Isolation in 1989, in conjunction with other infection control procedures, appears to have halted the rise, despite a continuing use of broad-spectrum cephalosporins. In order to reduce the number of cases of CDAD, either a reduction in levels of environmental contamination or a reduction in the use of third-generation cephalosporins is required. If this can be achieved the economic consequences, in terms of an opportunity cost, will be considerable.

A family outbreak of Salmonella enteritidis PT4 infection is described in which home-made ice cream was identified as the vehicle of infection. The ice cream contained approximately 105S. enteritidis PT4 organisms per gm and was probably contaminated by an infected shell egg containing between 105−108 organisms. The continued relevance of the Chief Medical Officer's warning on the use of raw shell eggs is highlighted.

Home-made ice cream using the same recipe as ice cream that had been incriminated as the cause of the family outbreak of S. enteritidis PT4 infection was used to study the growth of the organism that might have occurred in the 3–4 h it took to prepare the product. When the inoculum was in the stationary phase, as it would be from shell or other cross contamination, there was a lag phase of 3 h before growth occurred at room temperature. Even when actively multiplying organisms were introduced, as may be found in an infected egg, there was less than 3 log10 increase in the salmonella count in 4 h at room temperature. It was, therefore, given the high S. enteritidis count, unlikely that the ice cream was cross-contaminated.

By contrast, raspberry sorbet at pH 3·73 proved to be lethal to a large inoculum of S. enteritidis and may be a relatively safe raw egg containing product.

Minced beef samples inoculated with Escherichia coli O157 were cultured in buffered peptone water supplemented with vancomycin, cefsulodin and cefixime (BPW-VCC) and subcultured to cefixime tellurite sorbitol MacConkey (CT-SMAC) agar both directly and after immunomagnetic separation (IMS) of the organism with magnetic beads coated with an antibody against E. coli O157 (Dynabeads anti-E. coli 0157, Dynal, Oslo). E. coli O157 was recovered from initial inocula of 200 organisms/g by direct subculture and 2 organisms/g by IMS. Twelve strains of E. coli 0157 of different combinations of phage type, H antigen and toxin genotype were all recovered from initial inocula of two organisms/g by IMS. Nonspecific binding of other organisms to the magnetic beads could be reduced by washing of the beads in PBS with Tween-20 0·002–0.005% E. coli O157 was not bound by magnetic coated with an unrelated antibody.

During investigation of a dairy herd that was possibly linked to a small outbreak of infection with E. coli O157, the organism was isolated from 2 of 279 forestream milk samples from individual cattle; both isolates were made only by the IMS technique. IMS is rapid, technically simple, and a specific method for isolation of E. coli O157 and will be useful in epidemiological studies.

Utilizing the Office of Population Censuses and Surveys (OPCS) Omnibus Survey, it was possible to measure the frequency with which a stool culture was obtained following episodes of diarrhoea in adults. Interviewing over 8000 adults, over a 4-month period between October 1992 and January 1993, 633 persons (7·9%) reported one episode of diarrhoea in the previous month, and 5·4 % of these individuals with diarrhoea reported that a stool had been requested for examination. No significant regional differences were observed with the sample size available. The estimate of the rate of diarrhoea in adults was just under one episode per person per year.

Faecal samples collected from three populations of healthy adult volunteers (290 pigfarmers, 316 abattoir workers, 160 (sub)urban residents) living in the south of The Netherlands were analysed for the prevalence and degree of antibiotic resistance of Escherichia coli.

Significant differences in prevalence of resistance to amoxicillin, neomycin, oxytetracycline, sulfamethoxazole and trimethoprim were observed. The pig-farmers showed the highest percentages of resistance and the (sub)urban residents the lowest. In contrast no significant differences in high degrees of resistance were observed, except for neomycin.

Although both pigfarmers and abattoir workers have regular contact with pigs differences in prevalences of resistance were observed. However, because abattoir workers with intensive and less intensive pig(carcass) contact did not show significant differences, this is probably not the only important source of resistant E. coli in pigfarmers.

The high antibiotic use by pigfarmers (5%) and abattoir workers (8%) than by (sub)urban residents (0%) did not result in significantly different resistance percentages.

The genomes of 34 recent United Kingdom isolates of Neisseria meningitidis serogroup C were examined by restriction enzyme digestion and by conventional and pulsed-field gel electrophoresis (PFGE). Strains were assigned to groups on the basis of the Dice similarity coefficient; strains with values of >92% were considered to be clonally related. Twelve clones were identified by PFGE. Strains of two clonal groups predominated. Restriction endonuclease analyses using the ‘high frequency cleavage’ endonuclease Stu I and conventional electrophoresis gave 11 groups; in general it had lower resolving power than PFGE.

Clonal dissemination of two different MRSA strains, both clumping factor negative, has been observed in Germany for more than a year. Both strains possess the mec-A determinant and each exhibits a characteristic genomic DNA fragment pattern. One strain has spread in the north, the other in the south-west of Germany.

Intensive care units are mainly affected by MRSA-infections and probably play a special role in further intra- and inter-hospital spread.

An outbreak of food intoxication involving over 265 cases in western United States occurred in October 1991. Staphylococcus intermedius was implicated as the aetiologic agent. Representative outbreak isolates (five clinical and ten from foods) produced type A enterotoxin. DNA fragments generated by four restriction endonucleases and analysed by pulsed-field gel electrophoresis (PFGE) provided definitive evidence that all isolates from nine different counties in California and Nevada were derived from a single strain. The PFGE pattern of these outbreak isolates was distinct from those of a heterogeneous collection of seven S. intermedius strains of veterinary origin and five unrelated S. aureus laboratory strains. The data show a significant PFGE pattern heterogeneity not only among members of different Staphylococcus species but also within members of the same species and even the same enterotoxin type. The results indicate that PFGE is a valuable strain-specific discriminator for the epidemiological characterization of S. intermedius. To our knowledge, this represents the first documented foodborne outbreak caused by S. intermedius. These findings suggest that the presence of S. intermedius and other species such as S. hyicus in food should be reason for concern.

In late 1992, three cases of smear-positive tuberculosis were diagnosed among secondary school students in Lodi, Italy. The three attended different schools but travelled on the same bus. Schoolmates, other bus riders, family members, and friends underwent tine testing and X-rays. Of the 3188 students tested, 277 (8·7 %) were reactors. Independent risk factors for tine reactivity among students included living in the same town (odds ratio [OR] = 4·8; 95% confidence intervals [CI] = 3·8–6·0); having classroom contact (OR = 4·4; 95% CI = 3·4–5·7); or riding the same bus (OR = 5·4; 95% CI = 4·3–6·7) as a smear-positive case. Twenty-four cases of pulmonary tuberculosis were identified. The index case was a student whose father had had cavitary tuberculosis. Despite being tine test positive in 1989, he was not given prophylaxis and was lost to follow-up. This large outbreak emphasizes the need for identification and prompt chemoprophylaxis of reactors, especially in vulnerable adolescent populations.

Giardia lamblia is a common and increasing cause of gastrointestinal illness in the UK. We report a case-control study that examined risk factors for giardiasis. Patients with giardiasis were identified from reports to the Consultants in Communicable Disease in1 Avon and Somerset, and age-sex matched controls were obtained from their general practitioners' lists. Details of travel history, water consumption and recreational water use were collected by postal questionnaire. Over the period July 1992 to May 1993, 74 cases and 108 matched controls were obtained. The data were analysed using conditional logistic regression. Swimming appeared to be an independent risk factor for giardiasis (odds ratio 2·4, 95% CI 1·0 to 6·1, P = 0·050). Travel (P = 0·001), particularly to developing countries, and type of travel (P = 0·004) - that is, camping, caravanning or staying in holiday chalets - were also observed to be significant risk factors. Other recreational water use and drinking potentially contaminated water were found to be not statistically significant after adjustment for other factors.

Influenza B virus reinfection in Japanese children was studied epidemi-ologically during 1979–91 and virologically during 1985–91. During this investigation, there were four epidemics caused by influenza B viruses, each of which accompanied antigenic drift. Between the epidemics in 1987/88 and 1989/90, the viruses changed drastically, both genetically and antigenically. The minimum rate of reinfection with influenza B virus during the whole period was 3–25% depending on the influenza seasons. The antigens of primary and reinfection strains of influenza B virus isolated from 18 children during 1985–90, which covered three epidemic periods, were studied by haemagglutination inhibition tests. The results showed that the viruses isolated in the 1984/85 and 1987/88 influenza seasons, which belonged to the same lineage, were antigenically close, and reinfection occurred with these viruses. The results of amino-acid analysis of the HA1 polypeptide of these viruses corresponded with those of antigenic analysis. There were no specific amino-acid changes shared by the primary infection and reinfection influenza B viruses; the patients were infected with the viruses epidemic at that time.

Sera from an age-stratified sample of 7196 individuals, submitted for diagnostic purposes to four public health laboratories in England in 1986/7, were tested for hepatitis A antibody. The serological profiles, which showed marked regional differences, were consistent with declining incidence in the past. The decline in the incidence of hepatitis A has resulted in an increase in susceptibility in adults. This has three main consequences: an increase in the average age of infection may be leading to an increase in morbidity; normal immunoglobulin may become less protective against hepatitis A; the risk of transmission through blood products contaminated by viraemic blood donors may rise.

Current average annual incidence in 5–14-year olds was estimated to vary between regions from 0·5–1·9%. This supports the view that, in the absence of a vaccination programme, hepatitis A will remain endemic unless there are further improvements in living conditions and standards of hygiene. A vaccine giving long-lasting protection could eliminate hepatitis A transmission with modest coverage at a young age. Targeting childhood vaccination on economically deprived areas or using vaccine to control outbreaks might be more effective policies.

To determine whether or not occupational exposure to sewage is associated with a higher seroprevalence of hepatitis A virus (HAV) infection, 600 sewage workers in Singapore were tested for total (IgG and IgM) antibody to HAV by enzyme immunoassay. Using logistic regression with stepwise procedure, the adjusted seroprevalence of sewage workers was 2·2 times higher than that of another non-occupationally exposed population group. Seroprevalence was significantly correlated with age and educational levels, the association being independent of the occupational association. The epidemiological data in the study show that sewage workers have an increased occupational risk of acquiring HAV infection and should be protected by active immunization.

A stratified and randomly-selected population sample was identified in 1990 in order to study the seroprevalence of nephropathia epidemica (haemorrhagic fever with renal syndrome) in Northern Sweden. Sera from 1538 subjects (750 men, 788 women), 25—64 years of age, were analysed for the presence of Puumala virus (PUV) specific-IgG by the indirect immunofluorescence antibody test. Specific IgG was detected in sera from 83 subjects (5·4%). Men and women had similar seroprevalence rates. The highest seroprevalences were found in subjects 55 years or older (8·0%) and among farmers and forestry workers (15·9%). The geographic distribution of seropositive individuals was uneven and there were significantly more seropositive persons in rural than in urban areas (P < 0·05).

The purpose of this research was to determine whether trappers in northern Alaska acquired immunity to rabies virus from non-bite exposures while trapping and skinning arctic foxes (Alopex lagopus). In coastal Alaska recurring epizootics presumably provide trappers ample opportunity for contact with rabid animals. Serum neutralization analyses of blood samples collected from 26 individuals were conducted. All but three had negative rabies neutralizing antibody levels (< 0·05 I.U./ml). Two of these had previously received rabies vaccine but one individual who had trapped for about 47 years with an estimated harvest of over 3000 foxes and who had never received pre- or post-exposure rabies vaccination had a rabies serum neutralizing antibody concentration of 2·30 I.U./ml. This represents the first report of an unvaccinated person acquiring rabies virus antibody with a titre above the 0·5 I.U./ml level considered acceptable by the World Health Organization.

An enzyme immunoassay employing recombinant Norwalk virus capsid protein was evaluated for the measurement of IgA responses. Tests on 23 volunteers and patients known to have been infected with Norwalk virus (NV) showed that 19 developed significant IgA responses, 2 had unchanging levels of IgA and 2 failed to respond. There was no evidence of IgA responses to NV following infection with Hawaii or Snow Mountain-like viruses.

Tests on sera from patients involved in outbreaks associated with eating contaminated shellfish suggest that some patients may have been infected with more than one strain of calicivirus. The use of the rNV EIA for measuring IgA and IgG responses in patients involved in a major outbreak of food poisoning affecting hospital staff indicated that the causative agent was probably NV.

An enzyme immunoassay (EIA) for astrovirus type 1 together with immune electronmicroscopy (IBM) was used to type a collection of 162 astroviruses obtained from 1981–93 from children with diarrhoea. The EIA was found to be specific for astrovirus type 1. Astrovirus types 2–4 were typed by IEM.

Astrovirus type 1 was the prevalent serotype 107/125 (86%), followed by type 3 (8%), type 4 (6%) and type 2 (1 %). Six samples containing astrovirus could not be typed or detected by EIA because they were coated with coproantibodies; 11 others were not identified. Virus particles could no longer be detected in 15/162 (9%) samples following storage for ≥ 2 years.

Selected samples containing astrovirus types 1–4 were passaged in CaCO2 cells and their identity confirmed by one or both assays. One sample was shown to have remained viable for 10 years when stored as an aqueous suspension at −20 °C.

Two patients with severe combined immune deficiency disease (SCID) were shown to be excreting astrovirus type 1 for 32 and 102 days respectively. One child was simultaneously shedding rotavirus and the other child was excreting adenovirus.

A rubella serological survey of 476 individuals selected by cluster sampling technique from Caieiras, a small town located in the outskirts of São Paulo city, southeastern Brazil, was carried out over the period November 1990–January 1991. The aim of the study was to characterize rubella epidemiology in a representative non-immunized community in south east Brazil. The survey comprised a seroprevalence study, stratified by age (0–40 years) and a seroconversion study of rubella vaccine in non-infected children below 2 years of age. Mathematical techniques were applied to resultant data sets to determine the age dependent rates of decay in the proportion of individuals with maternally derived antibodies, vaccine seroconversion, and infection of susceptibles, termed the force of infection, and to estimate the average age at first infection.

Serological evidence was used to confirm an outbreak of Akabane disease in cattle in the Turkish Province of Aydin in 1980. Thereafter, serum collections from the Middle East were screened for the presence of neutralizing antibodies to Akabane virus. The results indicate that the virus was present in a number of provinces on the south Turkish coast in 1979 and 1980 but that it probably did not persist into 1981; the virus had also been present on Cyprus in 1980 and on at least one previous occasion. There was also evidence of limited virus transmission in the Orontes river valley in Syria in 1979 and less precise evidence to show that occasional infection occurred in the lower Jordan river valley. The failure of Akabane virus to persist in southern Turkey for more than two years indicates that this area is open to epidemic rather than endemic infection. The presence of neutralizing antibodies in the eastern Turkish Provinces of Gaziantep and Diyarbakir suggests that this might be the route whereby Akabane virus occasionally invades the Middle East region.