A collection of Salmonella enteritidis strains isolated in Switzerland (1965–90) was characterized. The phage type and plasmid profile of isolates were compared with the copy number and insertion loci of the DNA insertion element IS200. Three clonal lines of S. enteritidis were identified by IS200 profile; the various phage types were subtypes reproducibly associated with one of these lines. All human and poultry isolates contained a 38 Mda plasmid which hybridized with a mouse virulence-associated gene probe. In S. enteritidis, the IS200 profile is a race-specific molecular marker of the chromosome, and may be particularly applicable for studying the epidemiology of less common serovars.

An outbreak of typhoid caused by Salmonella typhi of the same Vi-phage type (D1) and of the same antibiogram was reported in a large psychiatric institution in Singapore. A total of 95 (4·8%) of the 1965 inmates were infected, 47 with symptoms and 48 asymptomatic. Transmission was through close person-to-person contact and not through contaminated food or water. The source of infection could not be established. The outbreak was brought under control by maintaining a high standard of environmental sanitation, active search for fever and diarrhoeal cases, identification of asymptomatic cases by rectal swabbing, and isolation of those found to be infected. Mass immunization with two doses of heat-phenol inactivated typhoid vaccine was also carried out concurrently. The vaccine was found to have an efficacy of 65·8% in preventing clinical illness.

An indirect ELISA developed for the serological detection of Salmonella typhimurium in chickens using lipopolysaccharide as detecting antigen has been evaluated further in experimental infections. Following oral infection of 24-week-old laying hens with an invasive strain of S. typhimurium, high titres of specific circulating IgG were induced which were maintained for 20 weeks. Similar IgG titres were found in egg yolk. When 4-day-old chickens were infected high antibody titres persisted for 45 weeks. Chickens inoculated orally or intramuscularly with different numbers of S. typhimurium organisms showed graded serum IgG responses to LPS. The IgG titres in experimentally infected in-bred lines of chickens which showed greater genetic resistance to salmonella infection were significantly lower than those found in more susceptible lines. Oral and intramuscular infection with 18 different types of enterobacteria, including avian pathogenic E. coli, Citrobacter spp., Klebsiella spp., Proteus spp. and citrobacter-like organisms possessing some salmonella LPS (none possessed the 0–4 antigen) and flagella antigens, did not induce S. typhimurium LPS-specific IgG responses. Chickens infected orally with rough or non-flagellate mutants of S. typhimurium did not induce high titres of LPS or flagella-specific IgG respectively. Sera obtained from S. typhimurium-infected chickens showed much higher titres against S. typhimurium LPS than with those antigens from other serotypes, including S. enteritidis.

Strains of Salmonella enterica serovar berta (S. berta) from Denmark and seven other countries have been characterized with the aim of developing a rational typing strategy in connection with outbreak investigations.

Biotyping divided the strains into H2S-positive (90%) and H2S-negative (10%) biovars. Six percent of the strains were resistant to one or more antimicrobial agents. Eighty-eight percent of the strains carried plasmids and 52 different plasmid profiles were recognized. Six of the common plasmid sizes in these profiles were shown by restriction enzyme analyses to contain more than one plasmid species. More than 90% of the strains had the same ribotype with the restriction enzymes Sma I and EcoR I and the same whole cell protein profile. Outer membrane protein profiles and isoenzyme profiles were identical in all S. berta analysed.

Plasmid profiling in combination with restriction enzyme analysis of plasmids seemed to be the most rational typing strategy for S. berta. The results indicated that S. berta strains regardless of geographical source or host are possibly clonal in nature.

We determined the natural history of the colonization of our hospital's potable water by culturing water approximately biweekly from 20 sites throughout the hospital for 4 years. Overall, 545 (24.7%) of the 2200 samples grew Legionella pneumophila. During hyperchlorination, 11·7% of the samples were positive while 41·6% were positive in the absence of chlorination. There was no seasonal trend towards positivity, but there was marked inter-site variation in the semi-quantitative culture results. However, a single strain of legionella (as defined by plasmid profiling) tended to persist at a site. Such a site was a unique ecological niche in that different sites in the same wing were populated by distinct strains.

The two wings of our hospital had a significantly different distribution of strains of legionella – plasmid profile type III predominated in the Victoria Wing while types II and VI predominated in Centennial Wing. Twenty-four of our 28 cases of nosocomial Legionnaires' disease occurred in the Centennial Wing. Three of the four cases in the Victoria Wing were caused by plasmid profile type III while 18 of the 24 isolates from patients who acquired their infection in the Centennial Wing were type II. We conclude that each water outlet serves as its own ecological niche of L. pneumophila.

An outbreak of salmonellosis in calves was monitored for persistence of Salmonella typhimurium excretion in faeces and the effect of treatment with apramycin. Prior to treatment apramycin-resistant Escherichia coli were present but all S. typhimurium isolates were sensitive. Following the treatment of six calves with apramycin, apramycin-resistant S. typhimurium were isolated from two treated calves and one untreated calf. Plasmid profiles of E. coli and S. typhimurium were compared and plasmids conferring resistance to apramycin and several other antibiotics were transferred by conjugation in vitro from calf E. coli and S. typhimurium isolates to E. coli K-12 and from E. coli to S. typhimurium. The plasmids conjugated with high frequency in vitro from E. coli to S. typhimurium, and hybridized to a DNA probe specific for the gene encoding aminoglycoside acetyltransferase 3-IV (AAC(3)-IV) which confers resistance to apramycin, gentamicin, netilmicin and tobramycin.

A discrimination index was applied to assess the value of biotyping and resistotyping, used alone or together, for the subspecific discrimination of Escherichia coli. The index was high when 599 strains from a wide variety of sources were examined by full biotyping with 10 tests. Although the discrimination achieved was lower when a sub-collection of 333 strains was examined, that finding probably indicated that many of the latter strains were of urinary origin and represented a limited number of uropathogenic clones. Combined biotyping and resistotyping provided a higher level of strain discrimination than either method used on its own. Results suggest that these typing methods may be used with confidence for the discrimination of strains of E. coli.

During 1990, a strain of methicillin-resistant Staphylococcus aureus became epidemic in Spain and spread in a manner analogous to that of EMRSA-1 in England. Isolates of this strain produced little protein A and were resistant to a number of antibiotics including ciprofloxacin. Beta-lactamase production was encoded by a c. 39 kb plasmid, which also conferred resistance to mercury, cadmium, ethidium bromide and propamidine isethionate.

Investigation showed that two variants, separable by supplementary and Fisk phage typing, were circulating. The B variant appeared to spread more readily than the A variant.

The opportunity was taken to compare the discriminatory power of traditional typing methods with molecular techniques. The discriminatory power of the molecular techniques used only reached the same level as the traditional methods when double enzyme digestion of total cellular DNA by EcoR I and Cla I was performed.

Results are presented from a number of epidemiological studies using enzyme immunoassays (EIA) based on the purified anthrax toxin antigens, protective antigen, lethal factor and oedema factor. Studies on sera from a group of 62 human anthrax patients in Turkey and from cattle in Britain following two unrelated outbreaks of anthrax show that EIA using protective antigen can be a useful diagnostic aid and will detect subclinical infections in appropriate circumstances. A serological survey on wildlife in the Etosha National Park, Namibia, where anthrax is endemic, showed that naturally acquired anthrax-specific antibodies are rare in herbivores but common in carnivores; in carnivores, titres appear to reflect the prevalence of anthrax in their ranges. Problems, as yet unresolved, were encountered in studies on sera from pigs following an outbreak of anthrax on a farm in Wales.

Clinical details, including treatment, of the human and one of the bovine outbreaks are summarized and discussed in relation to the serological findings.

Fifteen and five of 20 volunteers challenged with the enterotoxigenic Escherichia coli strain O 78·H11 showed a fourfold titre increase of serum ELISA antibody to the homologous O 78 and the heterologous O 8 lipopolysaccharide antigen, respectively. Sixty-three of 191 sera from 1- to 48-month-old German children showed serum antibody reactive with O 78 antigen, all but two of these O 78-positive sera also showed reactivity with at least one further O antigen. Only 14 of the O 78 reactive sera also showed antibody to heat-labile enterotoxin. In addition, soluble O 8 antigen could inhibit the binding of serum antibody to absorbed O 78 in 68% of the German children. Antibody reactive with O 78 antigen is thus not a reliable serological marker for enterotoxigenic E. coli infection in German children.

A total of 121 uropathogenic Pseudomonas aeruginosa strains were examined for production of several virulence-related factors. These strains were distributed in five predominant O-serotypes, i.e. O 4, O 12, O 11, O 6 and O 5, which accounted respectively for 23·9, 23·1, 12·3, 8·2 and 5·7% of isolates. Pyochelin and pyoverdin siderophores were produced by most of the isolates, defective variants occurring at very low frequency (2·4% for pyochelin and 7·4% for pyoverdin). Adherence to uroepithelial cells and production of cytotoxins was demonstrated in 52·8 and 67·7% of the strains, respectively, with higher frequencies for epidemiologically related strains belonging to serotypes O 4 and O 12. Titration of total proteases, elastase and phospholipase C revealed a high degree of heterogeneity among isolates. However, examination of individual O-serotypes by exoenzyme production showed that elevated levels of total proteases and elastase were characteristics of serotypes of minor numerical importance, i.e. O 1, O 10, O 11 and O 17, whilst low levels of elastase were produced by strains belonging to the predominant serotypes, namely O 4 and O 12. Moreover, epidemiologically related strains belonging to serotypes O 4 and O 12 appeared more homogeneous than the whole serogroup, when compared with other groups on the basis of exoenzyme levels.

Since its development in 1988, a serologic typing scheme for Xanthomonas maltophilia, based on 31 O antigens, has been successfully used to serotype isolates involved in nosocomial outbreaks in the United States. To determine if this serotyping scheme would be useful in typing X. maltophilia isolates from worldwide sources, we obtained additional isolates from 10 countries; of 900 isolates tested, 795 (88·3%) were typable. In order of predominance, the three most common serotypes were 10, 3 and 19. These three serotypes were most frequently associated with respiratory and blood isolates. This serotyping system is useful as an epidemiologic screening method for universal typing of outbreaks of X. maltophilia infections.

In order to investigate risk factors for the acquisition of trachoma, and to study the effect of continued exposure to ocular chlamydial infection on the severity of inflammatory trachoma and its cicatricial sequelae, a longitudinal study was conducted in a Gambian village. Over a 20-month period, the incidence of active (inflammatory) trachoma was significantly higher among those sharing a bedroom with an active case (64/561, 11·4%) than among those who were not exposed in this way (37/658, 5·6%) (relative risk 1·97, 95% confidence interval 1·33–2.90). There was a positive trend in the odds ratio for severe to moderate inflammatory disease versus mild disease as the number of active cases in the bedroom increased, but this failed to achieve statistical significance (P = 0·0506). Individuals with inflammatory trachoma of moderate or severe intensity at one survey were significantly more likely than others to have moderate or severe inflammatory changes at a previous or subsequent survey (odds ratio 14·9, 95% confidence interval 3·9–68·0), implying that host factors may be more important determinants of severity than the frequency of exposure to reinfection.

Eight strains of influenza C virus isolated in various areas of Japan between January 1985 and January 1989 were compared using monoclonal antibodies to the haemagglutinin-esterase (HE) glycoproteins and by oligonucleotide mapping of total vRNA. Five of six strains isolated during 1986–9 were closely related to one another and also resembled the virus, C/Aichi/1/81, isolated in 1981 in Aichi prefecture. This suggests that the C/Aichi/l/81-related viruses had an epidemiological advantage over any co-circulating viruses at least during that period. One of two 1985 isolates (C/Nara/1/85) was antigenically indistinguishable from the C/Mississippi/1/80 strain though their oligonucleotide patterns were markedly different from each other. This raises the possibility that C/Nara/1/85 may be a recombinant virus which receives its HE gene from the C/Mississippi/l/80-related parent.

Accurate incidence information is required to plan and evaluate screening programmes which have been proposed for the detection of primary toxoplasmosis and cytomegalovirus infection in pregnancy. Appropriate statistical methods are described for deriving incidence rates and their confidence intervals from three types of data: change in age-specific seroprevalence, seroconversion, and IgM studies. These methods are applied to seven published studies on toxoplasmosis and cytomegalovirus carried out in the UK. In these publications only one estimate of the infection rate per pregnancy was correctly derived, and none were accompanied by confidence intervals. Using the proposed methods, most estimates of the primary toxoplasmosis rate in these studies were between 2·5 and 5·5 per 1000 pregnancies, compared to the 2 per 1000 usually cited. Most cytomegalovirus incidence estimates were between 4 and 10 per 1000 pregnancies.

Infectious diseases establish in a population of wildlife hosts when the number of secondary infections is greater than or equal to one. To estimate whether establishment will occur requires extensive experience or a mathematical model of disease dynamics and estimates of the parameters of the disease model. The latter approach is explored here. Methods for estimating key model parameters, the transmission coefficient (β) and the basic reproductive rate (RDRS), are described using classical swine fever (hog cholera) in wild pigs as an example. The tentative results indicate that an acute infection of classical swine fever will establish in a small population of wild pigs. Data required for estimation of disease transmission rates are reviewed and sources of bias and alternative methods discussed. A comprehensive evaluation of the biases and efficiencies of the methods is needed.