The possibility that a devastating human pandemic could arise, causing massive loss of human life, is discussed. Such a major threat to the human species is likely to be a virus, and would spread by the respiratory route. It need not necessarily cause massive loss of life, but if it caused serious illness or incapacity it would still have a major impact. A possible source is from an existing respiratory pathogen, but it would more probably arise from an infection that is maintained in an arthropod or vertebrate host, but which at present either does not infect humans, or if it does it fails to be effectively transmitted between them. More research should therefore focus on the pathogenetic factors and the viral determinants that promote respiratory transmission.

We describe the first cluster of cases of necrotizing fasciitis (NF) in this century in the United Kingdom (UK). Between 1 January and 30 June 1994 there were six cases (five confirmed, one probable) of Streptococcus pyogenes NF in west Gloucestershire, population 320000. Two cases died. The first two patients probably acquired their infections during the course of elective surgery performed in the same operating theatre, possibly from a nasopharyngeal carrier amongst the theatre staff. The remaining infections were community-acquired. Of 5 S. pyogenes isolates there were 2 M1 strains, 1 M3, 1 M5 and 1 M non-typeable strain. S. pyogenes NF had not been recorded in west Gloucestershire in the preceding 10 years and the incidence of S. pyogenes bacteraemia in England and Wales had not risen in the past 5 years.

The two presumably theatre-acquired infections raised several issues. The need for detailed bacteriological investigation of all cases of post-surgical NF was confirmed. Clusters of S. pyogenes infection following surgery should be managed by closure of the operating theatre until all staff have been screened for carriage. Closure of an operating theatre and screening of staff following a sporadic case is probably not justified because of the infrequency of surgical cross-infection with S. pyogenes. Regular, routine screening of theatre staff is neither practical nor necessary.

During March and April of 1992, the health surveillance system began detecting increasing numbers of cases of meningococcal disease (MCD) in the Islamic holy city of Makkah (Mecca). We identified 102 bacteriologically confirmed cases (CC) and 80 suspected cases (SC) of MCD. Neisseria meningitidis was identified as Group A. III-1 clone. The ratio of male:female cases was 2·9:1. All age groups of males were affected. There was only one case among women aged 10–30; 50% of the adult female cases were 55 or older. The case-fatality ratio (CFR) was 14·7% among CC. Pakistanis, who comprised about one-third of the CC. had a CFR of 26·7%. Fifty-nine percent of CC were religious visitors. CC in residents were most common in persons living near the Holy Mosque (Haram), where the carriage rate reached 86%. A mass vaccination program against MCD was instituted, using AC bivalent meningococcal vaccine (MCV). An abrupt drop, from a mean of 15 CC per week to 2 CC per week (only in visitors), coincided with vaccinating 600000 persons over 2 weeks. Makkah residents who had been vaccinated against MCD were less likely to have contracted MCD (OR = 0·17, 95% CI: 0·06–0·50). MCV was of no significant protective value if it had been administered 5 years before the outbreak. The main reason for not being vaccinated as stated by both cases (71%) and controls (45%) was not knowing about the disease. The age and sex differences probably relate to differences in exposures to crowded conditions. Health education should illuminate the seriousness of the disease and the importance of vaccination.

Vaccination against Neisseria meningitidis is not part of routine immunization schemes in any country; instead, targeted vaccination of groups at the highest risk is recommended during outbreaks and epidemics. After a long period of sporadic occurrence of meningococcal invasive disease, a new clone of Neisseria meningitidis C: 2a: P1.2, ET-15/37, occurred in the Czech Republic, and caused local outbreaks in two neighbouring districts, Olomouc and Bruntal, in spring 1993. In Olomouc, a mass campaign was conducted during which 6191 students were vaccinated (5·6 % of the total population of this locality and 96% of all students in the age group 15–19) within 2 weeks in June 1993. In Bruntal district, no such campaign was organized. In Olomouc, the incidence of invasive disease caused by Neisseria meningitidis C in the age group 10–24 decreased from 57 to 0 per 100000 (P < 0·001) during the post-vaccination period (July 1993–August 1994), but no such decrease was observed in Bruntal. Although other factors can affect the frequency of disease, these results support the current recommendations of targeted vaccination in outbreaks of meningococcal disease.

A molecular typing method for Staphylococcus aureus based on coagulase gene polymorphisms (coagulase gene typing) was evaluated by examining a total of 240 isolates which comprised 210 methicillin-resistant S. aureus (MRSA) and 30 methicillin-susceptible S. aureus (MSSA) collected from a single hospital. By Alul restriction enzyme digestion of the PCR-amplified 3′-end region of the coagulase gene including 81-bp repeated units, the MRSA and MSSA isolates examined were divided into 6 and 12 restriction fragment length polymorphism (RFLP) patterns, respectively, whereas five patterns were commonly detected in MRSA and MSSA. MRSA isolates that showed a particular RFLP pattern were considered to be predominant in the hospital. Coagulase typing with type-specific antisera was also performed for all S. aureus isolates for comparison. Coagulase types II and VII were most frequently detected and included isolates with four and five different AluI RFLP patterns, respectively, whereas each of the other coagulase types corresponded to a single RFLP pattern. These results indicated that RFLP typing was more discriminatory than serological typing, for typing S. aureus and demonstrated its utility in epidemiologic investigation of S. aureus infection in hospitals.

Sixty-nine strains of Vibrio cholerae 01 isolated at different times were analysed to investigate if there were any differences among the 01 strains isolated before, during and after the advent of the O139 serogroup. Of the 69 01 strains examined, 68 belonged to the Ogawa serotype while one belonged to the Inaba serotype. With the exception of one strain all other strains of V. cholerae 01 belonged to the eltor biotype. A single O1 strain isolated before the emergence of the O139 serogroup could not be classified as either eltor or classical biotype because it was resistant to both classical and eltor specific bacteriophages. Marked variations in the susceptibility to antibiotics of V. cholerae O1 isolated during the different periods were observed. In addition, strains of V. cholerae isolated after the epidemic of serogroup O139 in Calcutta showed an expanding R-type with resistance to a variety of drugs as compared to the 01 strains isolated before the advent of the O139 serogroup. From this study, it is clear that there is a substantial mobility in genetic elements of V. cholerae Ol which necessitates a continuous monitoring to keep abreast of the changing traits of the etiologic agent of cholera.

Since 1977, Vibrio cholerae O1 has been isolated from the Australian aquatic environment and periodically cholera cases have occurred following exposure to these environments. To study the relationships between clinical isolates and environmental isolates from rivers and aquatic life, widely distributed throughout the country, a wide range of molecular typing methods were employed. In this paper we report the analysis of the 180 Australian isolates (10 clinical and 170 environmental) using ribotyping. Seven ribotype patterns were observed among the Australian inaba isolates, 2 of which included all clinical inaba isolates and 84% environmental inaba isolates collected from 9 rivers and creeks in eastern Australia during an 8-year period. Isolates from epidemiologically related clinical cases, asymptomatic household contacts and sewage were indistinguishable. The ogawa isolates were more diverse, with 9 ribotypes observed among 24 isolates from 8 rivers during the same period. Ribotype patterns were not shared between the serotypes with the exception of one ogawa isolate which could be distinguished using PFGE. Ribotyping has been useful in confirming an association between epidemiologically related clinical isolates and the aquatic environment and the persistence of several clones of the O1 serovar in the Australian environment during an 8-year period.

After instituting laboratory screening for Escherichia coli O157. H7, a Connecticut hospital isolated the organism from four persons in September 1993. As a result, an outbreak of E. coli O157.H7 associated with a country club was detected. The club had served hamburger from the same shipment at two picnics. Attendees of two picnics were interviewed, stool cultures were obtained from symptomatic persons, and the remaining hamburger was cultured. Twenty (22%) of 89 persons who ate hamburger became ill, compared with 1 of 60 who did not eat hamburger (relative risk = 13·5, 95% confidence interval 3·2–56·3). Among persons who ate hamburgers, illness was strongly associated with eating hamburger that was not thoroughly cooked (P < 0·001). All 20 samples from 5 remaining boxes of patties yielded E. coli O157.H7. Isolates from hamburger and case-patients were indistinguishable by pulsed-field gel electrophoresis. Heightened surveillance can rapidly identify outbreaks and may mitigate their impact. However, continued review of food safety issues is necessary if E. coli O157.H7 outbreaks are to be prevented.

Specimens from 181 patients with diarrhoea were examined by a Military General Hospital in a 3-month period during deployment of troops to Saudi Arabia in 1990/1. DNA probes for heat labile (LT) and heat stable (ST) enterotoxin genes identified enterotoxigenic Escherichia coli (ETEC) in 47 of the specimens (26%) and 49 ETEC strains were isolated. The majority (55%) belonged to a novel ETEC serotype having the O-antigen 159 and a flagellar antigen designated as a provisional new type. They produced ST and the coli surface associated antigen (CS)6. Strains of serotype O6:H16 represented 22% of the ETEC examined. They produced ST, LT and CSS together with either CS1 or CS2. The remaining ETEC belonged to seven O:H serotypes. Overall, ST was the only enterotoxin gene identified in 73% of the ETEC and 67% of the strains expressed CS6 in the absence of other colonization antigens. Resistance to three or more antibiotics was observed in 53% of the ETEC, including most of the O159 strains.

Adhesion to HEp-2 cells has been shown to correlate with enteropathogenicity for Aeromonas species. Such adhesion is thought to reflect the ability of strains to adhere to human intestinal enterocytes, although HEp-2 cells are not of intestinal origin. In this study strains of Aeromonas veronii biotype sobria isolated from various sources were investigated in parallel assays for their ability to adhere to HEp-2 cells and to an intestinal cell line (Caco-2). Quantitative assays showed identical adhesion values were obtained with both cell lines. Adhesion was best when bacteria were grown at 22 °C compared with 37 °C and 7 °C. Some environmental isolates showed greater adhesion when grown at 7 °C than when grown at 37 °C. Filamentous structures on these strains are also optimally expressed under the above conditions (reported elsewhere). Mechanical shearing or trypsin treatment to remove surface structures from several adhesive strains grown at 22 °C decreased adhesion to cell lines by 50–80% providing further indirect evidence that filamentous adhesins may play a role in cell adhesion for this Aeromonas species.

In this study, the in vitro susceptibility of 209 campylobacter strains to the quinolones nalidixic acid, flumequine, ciprofloxacin, enrofloxacin, and to ampicillin, tetracycline and erythromycin was tested by the disk diffusion method. The strains were isolated from poultry abattoir effluent (DWA) and two sewage purification plants (SPA and SPB). Sewage purification plant SPA received mixed sewage, including that from a poultry abattoir, whereas SPB did not receive sewage from any meat-processing industry. The quinolone resistance of the DWA isolates ranged from 28% for enrofloxacin to 50% for nalidixic acid. The strains isolated from the sewage purification plants were more susceptible to the quinolones with a range of 11–18% quinolone resistance for SPB isolates to 17–33% quinolone resistance for SPA isolates. The susceptibility criteria as recommended by National Committee Clinical Laboratory Standards (USA) cannot readily be employed for campylobacter isolates. This investigation shows that the resistance of campylobacter bacteria is highest in the plant receiving sewage from a poultry slaughterhouse. Monitoring of antibiotic resistance of aquatic Campylobacter spp. is important, as surface waters are recognized as possible sources of infection.

In this study the distribution of phenotypes of campylobacter strains in sewage and surface waters was investigated by subtyping and by speciation of isolates from various aquatic environments. These environments included two municipal sewage plants (SPA and SPB) and waste water from a poultry abattoir (WWA). Both the sewage plants SPA and SPB collected domestic and industrial waste, and SPA received drain water from WWA. SPB received no waste water from any meat-processing plant. The isolates were speciated by PCR and subtyped by PCR/RFLP based on the flagellin PCR products.

From all three reservoirs, no Campylobacter lari was isolated, and approximately 80% of the isolates could be identified as C. jejuni and the rest belonged to the C. coli species. The PCR/RFLP typing technique has a high discrimination level and was reproducible between two separate laboratories. The 182 isolates tested yielded 22 distinct Dde I profiles. The results indicate that strains with profiles found in poultry are also detectable in waste water presumed to be solely from domestic and human sources. In addition some strains were unique to the known poultry-related sources, suggesting that avian-specific strains, non-pathogenic to man, may exist in the environment. In contrast some strains were unique to human waste indicating the potential importance of non-poultry sources of infection. No seasonality was observed in the profile distribution. So, at least in the Netherlands, it is unlikely that infections caused by contaminated surface waters contribute to the seasonality of human campylobacteriosis.

Examination of neck skin and caecal samples taken at a commercial processing plant from 15 randomly chosen poultry flocks showed that all flocks were contaminated initially with thermophilic Campylobacter spp., even in the apparent absence of caecal carriage. During processing, numbers of campylobacter on skin samples were reduced by between 10 and 1000-fold.

To improve hygiene control generally, chlorinated-water sprays were used to limit microbial contamination on equipment and working surfaces. In addition, chlorine concentrations in process water were increased and any unnecessary carcass contact surfaces in the processing plant were removed. When comparing flocks before and after the changes, it was found that numbers of campylobacter on packaged carcasses were significantly lower after the changes had been made (P 0·001). In practice, however, the reduction would be likely to have little impact on consumer exposure to campylobacter infection.

Between April and September 1993, a nationwide outbreak of salmonellosis occurred in Germany which was traced to contaminated paprika and paprika-powdered potato chips. Of the estimated 1000 cases, children below 14 years were principally affected. Levels of 0·04–0·45 organisms per gram were found in the snacks. The infective dose was estimated at 4–45 organisms with an attack rate of 1 in 10000 exposed persons. The unique feature of the outbreak was the variety of serovars involved. S. saintpaul, S. rubislaw and S. javiana were isolated during the same time period from paprika powder, spice mixtures, snacks and patients. Their clonal identity was confirmed by molecular typing methods. Furthermore, monophasic and non-motile strains of rare salmonella O-groups were isolated from both paprika products and patients. This is the largest documented outbreak due to contaminated spices which proved that even extremely low numbers of salmonellae adapted to the dry state were able to cause illness.

An outbreak of Salmonella enteritidis PT 6B food poisoning, the first reported of this recently described phage type, resulted in illness among 46 of 49 members of a camping group in North Wales. 33 of whom were hospitalized. Epidemiological evidence (P < 0·0001) indicated that a lemon meringue pie was the vehicle of infection. Fresh shell eggs, stored after purchase at ambient temperature, appear to be the most likely source of infection, with multiplication during preparation and subsequent storage of the pie a significant contributory factor. Campers may be at greater risk than others and should consider the use of cold boxes for the transport and storage of eggs, and avoid the preparation of lightly cooked egg products under these basic conditions.

Over 8000 ready to eat foods were examined for the presence of Listeria species. Overall. 5% of foods were found to contain these organisms. Higher occurrence was found in some foods such as chicken (11%) and fish (14%). Most of the Listeria species isolated were L. monocytogenes (49%) and L. innocua (36%) with lower numbers of other species. No seasonal pattern in the recovery of L. monocytogenes was found. Unsatisfactory or potentially hazardous levels of L. monocytogenes were found in 14 products (< 0·2%). mostly cooked meats. Undercooked chicken products appeared to present the greatest risk for the duration of this survey. The small number of samples which were potentially hazardous suggests that the risk to consumers is not high, and this is confirmed by the absence of clinical cases in the region during the period of study.

Gastro-intestinal diseases continue to be a major health problem in primary schools in the UK. This study, which took place in 20 primary schools in the Leeds area, investigated the presence of faecal indicator bacteria on children's hands and environmental surfaces. Faecal streptococci were used as an indicator of faecal contamination.

A handwashing knowledge score was developed for each child. Those children with good hygiene knowledge had less faecal contamination on their hands (relative risk: 1·4. 95% CI = 1·09–1·81, P = 0·005). Those schools with higher hand counts were more likely to have had a reported outbreak of gastroenteritis in the past. Values of the Townsend Deprivation Index, an indicator of deprivation, were compared with the hand results and those schools in high deprivation areas had higher hand counts. Of the swabs taken from surfaces in the toilet areas and classrooms, the carpets in the classrooms were the most frequently contaminated surfaces.

The prevalence of antibodies against Entamoeba histolytica was studied in the Mexican population using an immunoenzyme assay in solid phase (ELISA) and semiautomatic equipment. The antigen was a mixture of membrane proteins obtained by Triton X-100 extraction from an axenic culture of Entamoeba histolytica HM1-IMSS. The method was standardized by comparing serum samples from amoebic liver abscess patients with healthy volunteers. From the 60538 samples supplied by the National Seroepidemiology Survey, antibodies were found in 4·49 % (4·32–;4·65% at 95 % confidence limit). More significant titres occurred in the central region of the country. The ratio female to male was 1·25:1. The population living in metropolitan areas had probably been infected at a younger age than those living in the country.

Important differences were found in the seroprevalence obtained by ELISA compared with a study which used indirect haemagglutination (IHA) in the same sample frame.

An outbreak of cyptosporidiosis occurred in late April 1993 among resort hotel guests which was temporally associated with, but geographically distant from, a massive waterborne outbreak of cryptosporidiosis in Milwaukee, Wisconsin, that occurred in late March and early April of 1993. A case-control study was performed among groups with members who reported illness and among a systematic sample of groups who stayed at the resort hotel during the risk period. Of 120 persons interviewed, 51 (43%) met the case definition. Swimming in the resort hotel's pool was significantly associated with case status (OR = 9·8; 95% Cl 3·4, 29·7), as was consumption of ice from the hotel's ice machines (OR = 2·3; 95% Cl 1·01, 5·2). When analysis was restricted only to laboratory-confirmed cases and controls, swimming pool use was the only risk factor significantly associated with illness (OR=13·0; 95%C1 2·6, 88·7). Following waterborne outbreaks of cryptosporidiosis associated with water supplies, swimming pools should be considered as possible ongoing sources for transmission regionally.

In an outbreak of cryptosporidiosis in Warrington. a town in North-West England. 47 cases were recorded between November 1992 and February 1993. most within the first month. There was a strong statistical association between cases and residence in an area supplied from two groundwater sources. In a case-control study, a strong association between having drunk unboiled tap water from these sources, and a dose-response relationship were found. Oocysts were not detected in the water supply. During very heavy rainfall one source of water was found to drain surface water directly from a field containing livestock faeces, thereby bypassing natural sandstone filtration. Exceptionally heavy rainfall occurred at the probable time of infection. After withdrawal of the original water supply, the outbreak rapidly subsided. It was concluded that there was very strong evidence that this outbreak was waterborne. This, the second documented outbreak of cryptosporidiosis attributable to a groundwater supply, demonstrates that infection can be transmitted from a disinfected groundwater source despite apparently satisfactory treated water quality. We recommend that guidelines for protection of groundwater are implemented, raw groundwater should be routinely monitored for microbiological contamination, and the structure of all sources and waterworks should be assessed in risk surveys of water catchment areas.