« Previous | Contents | Next »
Listen
THE WATT GROUP REPORT
CHAPTER 3 NOSOCOMIAL OUTBREAKS OF SALMONELLA - BACKGROUND
This chapter is intended to provide background to the factors that contributed to the outbreak; it is not an extensive scientific review.
Salmonella species causing gastrointestinal infection in man are often derived from cattle or poultry in which they may be found as commensals in faeces. For infection to occur, the organisms must be ingested, usually through the faecal-oral route. In the main, infection is foodborne and occurs by ingestion of meat or animal products contaminated with these faecal salmonella. In a smaller proportion of outbreaks infection may occur due to person-to-person contact, either directly or indirectly via contamination of equipment or of a surface, which is then handled by a second person and the organism transferred to their mouth.
Ingestion of
Salmonella spp. may produce:
No effects (i.e. organisms destroyed by the digestive system);
Replication within the gut and consequent faecal excretion without clinical symptoms (asymptomatic carriage);
Replication within the gut and clinical (usually gastrointestinal) symptoms.
Both asymptomatic carriers and symptomatic patients can be a potential source of infection to others.
Salmonella spp. cause symptoms 6-72 hours (usually 12-36 hours) after ingestion.
1 The incubation period may be as long as 7 days. The illness is usually self limiting and short-lived but in a proportion of patients (usually at the extremes of life) the illness can be life threatening.
2
Although it had long been assumed, on the basis of volunteer studies, that it was necessary to ingest large numbers of salmonella organisms to produce infection, a review of the subject
3 concluded both that such studies had their limitations (e.g., few tested low challenge doses) and that, in those foodborne outbreaks where the challenge dose had been calculated, the numbers of organisms could be as low as 10_ or less. There also appears to be a clear dose-response relationship; ingested dose appears to be an important determinant of incubation period, symptoms and severity of disease.
4,5 All of these studies relate to foodborne infections; but it is reasonable to assume that these conclusions are also valid for person-to-person spread. There needs to be more detailed studies of non-foodborne nosocomial outbreaks of
Salmonella infection to establish how person to person spread occurs and whether non-contact routes (e.g., aerosolisation of organisms) play any or a significant part. Such studies would be assisted by comprehensive surveillance of nosocomial outbreaks.
Salmonella spp. are grouped into serotypes (sometimes still referred to as species). The commonest serotypes are
S.enteritidis and
S.typhimurium. Although there has been a year-on-year reduction in the number of laboratory isolates of salmonella,
S.enteritidis accounts for the majority of human isolates (see OCT report).
Strains of
S.enteritidis and
S.typhimurium can be subtyped by their susceptibility to a range of bacteriophages. The organism involved in this outbreak was
Salmonella enteritidis phage type 21, an uncommon phage type in Scotland (49 of 1,349 isolates in 1998), although, as in the rest of the UK, there has been a recent increase in frequency of isolation.
2
Strains can be further subtyped using molecular techniques such as pulsed field gel electrophoresis (PFGE) and plasmid profiling, as well as antibiotic susceptibility testing. These methods may have considerable use in the investigation of outbreaks.
Salmonella spp. can cause outbreaks of infection in the community and in healthcare premises such as hospitals. Exact information on the role of salmonella in the community and in hospitals is difficult to obtain; laboratory based surveillance is believed to detect only a small proportion of cases in the community.
6 In any case most laboratory reports refer to individual patient isolates, rather than outbreaks.
7
In a review of outbreaks of salmonella in hospitals in England and Wales over a 10 year period, Joseph and Palmer
8 showed that of 248 outbreaks, foodborne infection accounted for 57 (24%) outbreaks and presumed person-to-person spread for 70 (30%). No information was available for 107 (46%) of the outbreaks.
A more recent review
9 concluded that of 22 outbreaks in which
Salmonella spp. were implicated, person-to-person spread accounted for 12, foodborne infection for 8 and in the other 2 the route of infection was unknown. Frequently person-to-person transmission is established by the exclusion of foodborne infection.
Although it is sometimes possible to establish the source of person-to-person spread in nosocomial salmonella outbreaks
10 in most outbreaks the exact source(s) cannot be determined. For example, careful investigation of a nosocomial outbreak due to
S.enteritidis phage type 4, including the use of PFGE
11 established that person-to-person spread was the most likely cause, but failed to establish the source. The authors concluded that admission of a patient with an unrecognised community-acquired infection was the most likely source.
This was probably the case in an outbreak of nosocomial
S.typhimurium infection in a Scottish hospital, which resulted in transmission of infection to 2 patients, 13 health care workers and a relative of one of the health care workers. The causative organism carried two independent drug resistance plasmids, which the authors concluded might have contributed to the virulence and transmissibility of the organism.
12
« Previous | Contents | Next »