An outbreak of gastroenteritis occurred in a village on the Island of Crete, with 1479 inhabitants. One hundred and thirty–eight symptomatic patients from 57 different families were examined. Thirty percent of children under 12 years were affected compared with 4% of adolescents and adults (P < 0.0001). Thirty–five out of 105 stool cultures (33%) grew Shigella sonnei. Thirty–four isolates had the same susceptibility pattern and were sensitive to ampicillin, while one was resistant to this antibiotic. Thirty–three isolates had the same plasmid profile. Based on epidemiologic and environmental findings, it was concluded that the transmission of shigellosis was waterborne. Shigella were not isolated from water, but one of the three village springs was found to be contaminated with faecal coliforms. This spring was probably contaminated by sewage material, due to its proximity to a source of untreated sewage effluent. Implementation of environmental and control measures brought the epidemic to an end.

Transfer of shigella R–plasmids in vivo has seldom been demonstrated. Strains of Shigella dysenteriae type 1 and Shigella flexneri type 5b were isolated from a Bulgarian traveller who visited Vietnam and developed dysentery, which was treated with trimethoprim/sulfamethoxazole (TMP/SMZ) for a short time. Both species of shigellae are unusual in Bulgaria where strains of S. sonnei predominate. Both shigella strains were multiresistant to the same antimicrobial agents. Each strain contained a 48–kilobase plasmid that conferred the entire resistance phenotype to a susceptible Escherichia coli. Restriction endonuclease patterns of plasmid DNA from the respective strains were identical. Transmissible plasmids of the same resistance phenotypes and restriction patterns were isolated from the patient's colonic E. coli. Transconjugants hybridized to a dihydrofolate reductase type I–DNA probe. These studies support the hypothesis that R–plasmid transfer may occur between non-pathogenic, faecal strains and pathogenic shigellae, a process that may have been facilitated by inadequate treatment with TMP/SMZ at the onset of the illness.

When Pst I–generated digests of genomic DNA from each of the type strains of 49 of the Vi phage types of Salmonella typhi were probed with a PCR–amplified IS200 gene probe, all strains were found to possess at least 11 IS200 elements carried on fragments in the range 24.2–1.2 kb. Fourteen fingerprints were identified but two patterns designated IS200Sty1 and IS200Sty2 predominated. In one strain, a plasmid–mediated IS200 element was identified. When IS200 fingerprinting was applied to epidemiologically–unrelated strains of S. typhi isolated in Ecuador, 3 patterns were identified in 10 strains belonging to 9 different phage types. It is concluded that Vi phage typing remains the method of choice for the primary differentiation of S. typhi but that IS200 fingerprinting may be of limited use in laboratories which do not have access to phage typing.

Restriction fragment length polymorphism of ribosomal RNA genes was analysed among 133 Escherichia coli strains predominantly from blood and urine, including 21 isolates from faeces of healthy persons. The strains had also been characterized for their O:K:H serotypes, for the presence of P, S and type 1C fimbriae, non–P, non-S mannose–resistant haemagglutinins and haemolysin production. Hind III–digested genomic DNA was subjected to Southern blot analysis with either plasmid pKK3535 containing E. coli rRNA operon or purified rRNA as a probe. Among the 133 strains 20 ribotypes were obtained. The distribution of strains into different ribotypes generally correlated with their O:K:H serotype. Ribotype variation within serotypes was mainly seen among strains with the K5 capsule. The origin of the strains or the presence of virulence-associated factors did not correlate with the ribotype. In conclusion, ribotyping appears to be a valuable method in epidemiologic studies especially when the serotyping methods are not available.

The general bacterial flora of 38 Tasmanian SIDS infants was examined together with faecal flora of 134 comparison infants ranging in age from birth to 6 months. The microflora of all specimens received was investigated with special emphasis on the toxigenic Escherichia coli (TEC). Samples were examined for verocytotoxigenic E. coli, free faecal verocytotoxin (FVT), heat labile toxin (LT) and heat stable toxin (ST) producers with the use of a Vero cell assay and commercial kits. The findings of this study revealed a high isolation rate (39%) of TEC from SIDS infants as compared to 1.5% from the healthy comparison infants. Atypical E. coli strains were also identified during the study, including E. coli A–D. An analysis of the same specimens for rotaviral and adenoviral antigens indicated that 30% of the SIDS cases were positive as compared to 20% in the comparison group.

The seasonal variation in the occurrence of V. vulnificus in relation to water temperature and salinity was studied along the Dutch coast. In two consecutive years V. vulnificus strains could be isolated in August when the water temperature was highest. The indole-positive strains isolated from North Sea water samples were identical to most strains isolated from human disease and from the environment. However, strains isolated from four of five patients living in countries around the North Sea were different from the North Sea isolates in that they were indole-negative and have a lower NaCl tolerance.

Aeromonas caviae, an enteropathogen associated with gastroenteritis, displays several virulence characteristics. Studies on the kinetics of growth of A. caviae and expression of β–haemolytic toxin revealed that A. caviae produced maximum haemolytic activity extracellularly during the stationary phase. Preliminary studies on the properties of A. caviae haemolysin suggested that divalent cations (Mg2+ and Ca2+) and thiol compounds, dithiothreitol and mercaptoethanol enhanced the haemolytic activity. Addition of L–cysteine, glutathione and EDTA reduced the haemolytic activity. The iron chelator, 2–2' bipyridyl, significantly inhibited the growth of A. caviae possibly by iron limitation, with parallel enhancement of haemolysin production compared to A. caviae grown in excess of iron. These results suggest that A. caviae produces only β-haemolysin, which resembles the haemolysins reported for several other bacteria and the activity might be regulated by environmental factors especially iron.

In recent years, methicillin–resistant Staphylococcus aureus (MRSA) isolates in Spain have increased dramatically; in 1986 there were only 1.2% MRSA amongst all nosocomial Staphylococcus aureus (SA) isolates, by 1989 this percentage had risen to 44% in some hospital causing a very serious epidemic situation in the country. We have characterized these isolates by direct, reverse and Fisk phage typing and we have also looked for an additional local set of phages to help us to differentiate these strains. We have been able to differentiate an epidemic strain from other MRSA strains which cause sporadic hospital outbreaks, and we have also distinguished between some variants of the epidemic strain.

Non-specific and specific mechanisms of adherence have been examined in two collections of methicillin–resistant Staphylococcus aureus (MRSA). Determination of hydrophobicity by salt aggregation, hydrophobicity indices and of adherence to the extra–cellular matrix proteins fibronectin, vitronectin, laminin and collagen type 1 have failed to reveal any correlation with phage-type, plasmid profile or antibiogram. Further, the strain collections, made over a period of years in two countries, differ markedly in their adherence characteristics; MRSA are heterogeneous in this respect. Such heterogeneity may explain the polarization of views on the epidemicity or ‘virulence’ of MRSA. With the exception of adherence to collagen a small group of methicillin sensitive S. aureus had characteristics intermediate between the two groups of MRSA.

This case control study investigated environmental factors in 74 confirmed cases of meningococcal disease (MD). In children aged under 5, passive smoking in the home (30 or more cigarettes daily) was associated with an odds ratio (OR) of 7.5 (95% confidence interval (CI) 1.46–38.66). ORs increased both with the numbers of cigarettes smoked and with the number of smokers in the household, suggesting a dose–response relationship. MD in this age group was also significantly associated with household overcrowding (more than 1.5 persons per room) (OR 6.0, 95% CI 1.10–32.8), with kisses on the mouth with 4 or more contacts in the previous 2 weeks (OR 2.46, 95% CI 1.09–5.56), with exposure to dust from plaster, brick or stone in the previous 2 weeks (OR 2.24, 95% CI 1.07–4.65); and with changes in residence (OR 3.0, 95% CI 1.0–8.99), marital arguments (OR 3.0, 95 % CI 1.26–7.17) and legal disputes in the previous 6 months (OR 3.10, 95% CI 1.24–7.78). These associations were independent of social class. Public health measures to lower the prevalence of cigarette smoking by parents of young children may reduce the incidence of MD. The influence of building dust and stressful life events merits further investigation.

Two hundred and eighteen nosocomial cases of Legionnaires' disease with 68 deaths were reported to the National Surveillance Scheme for Legionnaires Disease between 1980 and 1992, representing 15% of the reported infections acquired in England and Wales. Twenty–two nosocomial outbreaks accounted for 135 (62%) of these cases, the remainder occurring as single cases either in hospitals where other single cases or outbreaks had been reported in different years or as ‘sporadic’ cases in hospitals from which no other cases were reported. A clinical history prior to onset of Legionnaires' disease was available for 124 patients, 61 of whom had undergone recent transplant therapy or were immunosuppressed for other reasons.

Sixty cases (27%) were diagnosed by culture of the organism and isolates from 56 patients were typed; 25 (42%) were non L. pneumophila serogroup 1 infections.

Methods for prevention and control of nosocomial outbreaks are discussed, in particular the susceptibility to Legionnaires' disease of certain groups of hospital patients.

Sera from 83 patients with campylobacter gastroenteritis were examined for the presence of legionella antibodies by indirect immunofluorescence. Twenty–one patients (25%) had positive titres (≥ 16) including 11 patients with titres of ≥ 128. Legionella seropositivity persisted in 5 of 9 patients (55%) studied for 6–9 months. Campylobacter isolates were serotyped by the Penner scheme. Isolates associated with legionella seropositivity included Penner types 1, 2 and 4, the common endemic serotypes in England. Campylobacter blocking fluids were prepared from a range of Penner reference strains. The blocking fluid prepared from Penner type 11 was the most efficient at inhibiting the false-positive legionella titres. Using this absorption step legionella titres were inhibited from 24 of 26 patients (92%) with campylobacter but not from 8 patients with cultureproven legionnaires' disease. We recommend that this method is incorporated into routine diagnostic legionella serology in order to eliminate false-positive reactions due to campylobacter.

We studied the relationship between IgA anti–campylobacter flagellin antibodies in breast milk samples and protection of breastfed infants living in a rural Mexican village from campylobacter infection. There were fewer episodes of campylobacter infection (symptomatic and asymptomatic combined) in infants breastfed with milk containing specific anti-flagellin antibodies (1.2/child/year, 95% CI 0.6–1.8) versus non–breastfed children (3.3/child/year, 95% CI 1.8–4.8; P < 0.01). Infants breastfed with milk that was antiflagellin antibody negative by ELISA also had fewer episodes of infection compared with non-breastfed children, but the difference did not reach statistical significance (1.8/child/year, 95% CI 0.7–3.0 versus 3.3/child/year, 95% CI 1.8–4.8, P > 0.05). Breastfeeding has a protective effect against campylobacter infection and is associated with the presence of specific antibodies directed against campylobacter flagellin.

Experimental data suggest a role for the microflora in Systemic Lupus Erythematosus (SLE). Anti–ds–DNA antibodies may be pathogenic in SLE by forming immune complexes with DNA. Foreign bacteria in the intestines could constitute the stimulus for anti–ds–DNA antibody production in SLE. Colonization Resistance (CR) is the defence capacity of the indigenous microflora against colonization of the intestines by foreign bacteria. A low CR implies increase of translocation of bacteria and a higher chance of subsequent, possibly DNA–cross–reacting antibacterial antibody production.

We measured CR by a comprehensive biotyping technique in healthy individuals and patients with inactive and active SLE. CR tended to be lower in active SLE patients than in healthy individuals (P = 0.09, Wilcoxon one sided, with correction for ties). This could indicate that in SLE more and different bacteria translocate across the gut wall due to a lower CR. Some of these may serve as polyconal B cell activators or as antigens cross-reacting with DNA.

Attempts were made to identify the causative organism of Lyme disease in Australia from possible tick vectors.

Ticks were collected in coastal areas of New South Wales, Australia, from localities associated with putative human infections. The ticks were dissected; a portion of the gut contents was examined for spirochaetes by microscopy, the remaining portion inoculated into culture media. The detection of spirochaetes in culture was performed using microscopy, and immunochemical and molecular (PCR) techniques. Additionally, whole ticks were tested with PCR for spirochaetes.

From 1990 to 1992, approximately 12000 ticks were processed for spirochaetes. No evidence of Borrelia burgdorferi or any other spirochaete was recovered from or detected in likely tick vectors. Some spirochaete–like objects detected in the cultures were shown to be artifacts, probably aggregates of bacterial flagellae.

There is no definitive evidence for the existence in Australia of B. burgdorferi the causative agent of true Lyme disease, or for any other tick–borne spirochaete that may be responsible for a local syndrome being reported as Lyme disease.

Botulism due to Clostridium botulinum type C causes considerable mortality in gulls in the UK, and refuse disposal sites are suspected as a major source of toxin. C. botulinum types B, C and D were each found in 12 (63.2%) of 19 landfill sites examined. Type E was detected in only one (5.2%) and types A, F and G were not found. The prevalence of type C spores was much higher than that demonstrated in the UK environment by earlier surveys. The presence of these spores, together with the rotting organic matter and generated heat associated with landfill sites, undoubtedly leads to bacterial proliferation and toxigenesis. This is likely to result in botulism in scavenging gulls unless skilled landfill management prevents the ingestion of toxic material. Type D spores were previously shown to be rare in the UK environment and their high prevalence on landfill sites was therefore surprising. Four composite samples of refuse collected before distribution on a landfill gave negative results for C. botulinum and it seems likely that the gulls themselves play a major role in introducing contamination.

A cluster of disseminated Candida albicans infections, which occurred at the Intensive Care Unit of the Department of Heart Surgery, was investigated. Ten patients became infected and seven died. A wide microbiological surveillance was carried out. A total of 14 isolates of Candida albicans, four environmental and ten human, were examined using the Restriction Endonuclease Analysis (REA) of DNA. The isolates were classified into five different main groups. Five of the clinical isolates had the predominant pattern Ab and two more clinical strains were very closely related. Two more isolates from the emergency kit desk and the hands of a nurse gave the same REA profile. Such a relationship proved the epidemic nature of the cluster, with most of the patients cross-infected, and strongly suggested transmission on the hands of the staff as a determinant of the epidemic. Thus, REA has the potential to address many important questions in the study of nosocomial epidemiology of Candida albicans.

Current methods to establish the duration of toxoplasma infection in pregnant women and for the diagnosis of toxoplasmosis in the neonate or HIV infected patient have significant limitations. We assessed the precision of a commercial ELISA for the detection of toxoplasma specific IgG and adapted the assay to measure avidity using an elution agent washing step. The sensitivity and specificity of the ELISA were 100 and 75 % respectively and optimal measurement of avidity was achieved using 6 M urea as the elution agent.

Toxoplasma lymphadenopathy of less than 3 months duration was associated with low avidity specific IgG but some discordant findings were recorded. Serial measurement of IgG avidity assisted the distinction between actively produced antibody in infants with congenital toxoplasmosis and passively acquired antibody of maternal origin in uninfected babies. There was no significant difference between avidity levels in HIV infected patients with or without cerebral toxoplasmosis.

In 1964, the Medical Research Council undertook a trial of measles vaccine in over 36000 United Kingdom children; 9577 of whom received live vaccine, 10625 received inactivated followed by live vaccine, and 16328 acted as unvaccinated controls. Participants in this study have been followed to determine the long term protection from measles vaccine and follow-up data were available on 4194, 4638 and 274 respectively. During the 5–year period 1986–90, the protective efficacy of live measles vaccine has remained high at 87%, but the 95% confidence interval was wide ( –43 to 99%) due to the small numbers of cases. Between 1976 and 1990, however, the overall efficacy of the live vaccine was 92% (95% confidence interval 86 to 95%) and there was no evidence of a decline in efficacy (P = 0.13) over the 15–year period. This study suggests that the protection from live measles vaccine persists for up to 27 years after vaccination, and that no change in the current United Kingdom measles immunization policy should be made on the grounds of waning immunity.

A trial of protective efficacy which compared medium–titre Edmonston-Zagreb (EZ) measles vaccine (104.6 p.f.u.) from the age of 4 months with the standard Schwarz (SW) measles vaccine given from the age of 9 months was started in an urban community in Guinea-Bissau in 1985. Because trials of high–titre measles vaccine have found increased mortality among female recipients, we examined whether EZ medium–titre vaccine was associated with any long-term impact on mortality, suppression of T–cells, or growth. The mortality rate ratio over 5 years of follow–up was 1.12 for EZ children compared with children in the standard group (P = 0.63). Seventy–five percent of the children still residing in the area at 5 years of age took part in an immunological and anthropometric examination. There was no difference in T–cell subsets between the two groups. There was no difference in mid–upper–arm circumference, but EZ children were significantly shorter than the children in the standard group. In conclusion, medium–titre EZ was not associated with reduced survival or persistent immunosuppression.