The surveillance of Salmonella in Scotland is based on laboratory reports to HPS from the Scottish Salmonella Reference Laboratory (SSRL). Isolates from routine diagnostic clinical and veterinary laboratories are sent to SSRL for confirmation and typing. In addition the laboratory receives isolates from environmental and poultry sources as part of its surveillance of the organism.
During 2004, 2203 reports of Salmonella (excluding Typhi and Paratyphi) were added to the national dataset at HPS. This involves a screening process to remove duplicate and repeat samples and does not fully represent the workload of SSRL who analyse many more isolates than are detailed here. The sources of these isolates are given in Table 1.
During 2004, 1143 cases of human Salmonella were reported to HPS. This represents a decrease of 9% on the previous year and brings the numbers to roughly that seen in 2002. Figure 1 shows the decreasing trend of these organisms since the mid 1990s.
As in previous years, most of the reports were received during the summer months (Figure 2).
Most cases were seen in children under five. When age-specific incidence rates were calculated this was even more evident (Figure 3).
Four serotypes were responsible for 77% of cases of salmonellosis in Scotland. The predominant serotype causing human disease was S. Enteritidis (53%) followed by Typhimurium (19%). The most frequently reported serotypes are shown in Table 2.
There was 37 serotypes reported on one occasion only. Four serotypes were identified for the first time in Scotland. Details of these are given in Table 3.
Salmonella Enteritidis phage types
S. Enteritidis continued to be the most frequently reported serotype of Salmonella from humans. Phage Type 1 (PT1) was the predominant phage type of Enteritidis in 2004. This was a continuation of the trend first seen in 2003 when, for the first time in Scotland, Phage Type 4 (PT4) was not the most commonly reported phage type of Enteritidis (Table 4).
Eight phage types were reported on one occasion only including phage type 21a, which was reported for the first time in Scotland in 2004. Twenty nine percent of PT1 and 25% of PT4 were believed to have been acquired abroad.
Typhimurium phage types
The number of reports of Typhimurium was roughly the same as the previous year (214 compared to 211 in 2003). DT104 remained the most frequently reported phage type of Typhimurium and accounted for 37% of all reports (table 5) – this represented an increase of 55% on the 51 reports made in 2003.
There was 12 phage types reported on one occasion only and three phage types (DT11, DT178 and DT19) were reported for the first time in Scotland. Isolates of phage types 11 and 19 were believed to be associated with travel abroad.
Twenty eight percent of cases of salmonellosis indicated that their infection may have been acquired abroad. Twenty nine percent of all Enteritidis and 15% of Typhimurium were believed to have been acquired abroad.
Five outbreaks of Salmonella infection were reported during 2004: one each of S. Newport, S. Saint-Paul, S. Enteritidis PT4, S. Enteritidis PT14b and S. Typhimurium, a decrease compared to the seven reported in 2003.
The largest of these outbreaks in 2004 was an outbreak of S. Enteritidis PT4 associated with a cruise ship, in which 115 persons were ill and 25 were microbiologically confirmed. A cohort study of all passengers and crew confirmed a statistical association between being a case and the consumption of chocolate mouse. The outbreak of S. Newport was part of a larger UK-wide outbreak (Current notes 38/3802 and 38/4101) with the 21 Scottish cases spread across a number of the NHS boards in Scotland. The outbreak of S. Saint-Paul also had cases in a number of NHS boards.
Extra-intestinal isolates of Salmonella
Most isolates reported were from faeces, however extra-intestinal isolates were also received (Table 6)
Salmonella Typhi and Paratyphi
Previous data are for cases of salmonellosis. During 2004, there were in addition 12 reports of Paratyphi A and eight reports of Typhi. Most (70%) indicated that the infection was believed to have been acquired outwith the UK.
Molecular typing of isolates of Salmonella from humans
The application of plasmid profile analysis (PPA) and pulsed-field gel electrophoresis (PFGE) for the typing of Salmonella has been under development in SSRL for a number of years and results are now held on a database. The adoption of internationally standardised protocols and the introduction of advanced analytical software allows the sharing and comparison of PFGE data between many European national reference laboratories, as well as laboratories which form the PULSENET network (including the USA, Canada, Australia).
From the beginning of 2003, most non-typhoidal isolates of Salmonella of human origin have been typed by PPA and PFGE. These molecular techniques will provide results that can enhance the information from the phenotypic methods currently employed for typing of Salmonella.
Plasmid profile analysis
Plasmid profiling of an organism involves the isolation of extrachromosomal circular DNA molecules from the organism of interest, followed by the separation of these molecules, based on their size, by agarose gel electrophoresis. While plasmids have the ability to transfer between organisms, the presence or absence of particular plasmid molecules can be an important epidemiological marker.
The number of plasmid profile analyses carried out on human Salmonella isolates from 2004 together with the percentage of plasmid containing isolates, number of plasmid profiles and number of isolates within the most common profile for the most frequently reported serotypes in Scotland are given in Table 7.
A total of 1145 isolates were typed by plasmid profile analysis. Of these, only 177 (15.4%) isolates were found not to contain plasmids. This ‘plasmid free’ status is considered as a profile type.
Within S. Enteritidis, over 96% of isolates contained at least one plasmid. In 324 of these isolates, the only plasmid present was the 57-kilobase pair (kb) serotype associated virulence plasmid that was the most common profile type for Enteritidis. This profile was observed in isolates belonging to PT1, 2, 3, 4, 4b, 5c, 6, 6a, 7, 8, 13a, 15, 21, 31, 33, 35, 44, RDNC and untypable. Thirteen different plasmid profiles were observed within the 138 isolates of PT 1 examined and 14 profiles were observed within the 143 isolates of PT4 examined. Twenty isolates of Enteritidis were plasmid free, belonging to PT1, 4, 13a, RDNC and untypable. Excluding PT1 and PT4, the remaining 282 isolates of Enteritidis belonged to 81 different plasmid profiles, 45 of these profiles were represented by a single isolate.
For S. Typhimurium, 31 out of 205 isolates examined were found to be plasmid free. This was the most common plasmid profile and included isolates of DT12, 40, 56var, 104b, 120, 161, 170, U302 and RDNC. The 90kbp serotype associated virulence plasmid was identified as the only plasmid in 70 isolates including DT 1, 3, 8, 19, 41, 49, 56var, 94, 104, 193, U302, RDNC and untypable. A further 83 plasmid profiles were identified among the remaining 104 isolates of Typhimurium, 71 of these profiles being unique.
The plasmid profiles observed within isolates of other serotypes were extremely diverse. However, in many cases, outbreak strains could be linked by the possession of a characteristic plasmid profile e.g. a national outbreak of Newport was characterised by the possession of a plasmid of 100 kb.
Pulsed-field gel electrophoresis
Typing of an organism by PFGE involves the cutting of chromosomal DNA into large fragments by specific enzymes. The separation of these fragments by electrophoresis provides a ‘fingerprint’ which is dependent on the number and position of the specific sites identified by the enzyme. The primary enzyme used for Salmonella is XbaI but other enzymes, which may be able to separate within individual XbaI patterns, can be utilised in certain circumstances such as enhanced outbreak investigation. Indistinguishable patterns are assigned the same pattern designation. Some of these designations are given in the format agreed by the European laboratories which participated in the SalmGene project which ended in 2005. PFGE patterns submitted to the Salm-gene project are given a 10-character code as follows:
- A one-letter genus designation e.g. Salmonella = S.
- A three-letter serotype designation e.g. Enteritidis = ENT. Examples: Salmonella Enteritidis = SENT; Salmonella Typhimurium = STYM etc.
- A two-letter restriction enzyme designation e.g., Xba1 = XB; SpeI = SP; BlnI= BL.
- Thus an isolate of Salmonella Enteritidis digested with the enzyme Xba1 would read: SENTXB
- Each new pattern is given a four-digit numerical identifier starting with .0001 and finishing at .9999. For instance, the final designation for the first pattern of Salmonella Enteritidis digested with the enzyme Xba1 would be SENTXB.0001, and for the third pattern of Salmonella Typhimurium digested with the same enzyme STYMXB.0003 etc. Patterns found in Scottish isolates are given provisional local designations until such time as a match can be found in the European database e.g. EntXSco7.
The number of PFGE analyses carried out on human Salmonella isolates from 2004 together with the number of PFGE profiles and number of isolates within the most common profile for the ten most common serotypes are given in Table 8.
Three hundred and sixty-three of the 563 isolates of S.Enteritidis examined had the PFGE profile SENTXB.0001. These included a wide range of phage types including PT1, 1c, 3, 4, 5a, 5c, 6, 6a, 7, 8, 14b, 21, 21a, 29, 31, 35, untypable and RDNC, and reflect the high degree of clonality found within this serotype. A further 20 isolates had the profile SENTXB.0014 which is identical with SENTXB.0001 apart from the absence of a single fragment which correlates with the absence of the serotype associated virulence plasmid. Sixty-one isolates had the profile SENTXB.0002. These belonged to the phage types 8, 13a, 14b, 22, untypable and RDNC and this profile is representative of a different clonal lineage. The numbers of isolates observed with the PFGE patterns most frequently found in human isolates of Enteritidis in Scotland in 2003 and 2004 are detailed in Table 9.
Isolates of Enteritidis PT14b are found to occur with PFGE profile SENTXB.0001 and SENTXB.0002. PT14b isolates with profile SENTXB.0002 are suspected to have been acquired following the consumption of food products containing imported eggs or associated with travel to the Spain or Portugal. In 2004, 22 isolates of PT14b were of PFGE profile SENTXB.0001 and 19 were SENTXB.0002. A single isolate was of type SENTXB.0040, a possible variant of SENTXB.0002.
Three different PFGE profiles were identified among the 140 isolates of PT1 examined. There were seven PFGE profiles for PT4 (146 isolates), four profiles for PT21 (46 isolates), five profiles for PT8 (41 isolates), 10 profiles for PT6a (22 isolates) and 12 profiles for RDNC isolates (35 isolates). A total of 23 PFGE profiles were identified among the remaining 91 isolates examined.
Serovar Typhimurium is a more diverse serotype than Enteritidis and this was reflected in the diversity of PFGE profiles and lower number of isolates belonging to the most common patterns. STYMXB.0001 was the predominant PFG profile for Typhimurium and included isolates of PT104, 104b, U302, U310 and untypable.
Eight different PFGE profiles were observed among the 78 isolates of DT104 examined. There were eight profiles observed among isolates of DT104b (17 isolates), eight profiles for U302 (11 isolates), four profiles for DT170 (eight isolates) and 23 profiles for RDNC isolates (35 isolates). A total of 35 PFGE profiles were identified among the remaining 54 isolates of Typhimurium examined.
The numbers of isolates observed with the PFGE patterns most frequently found in human isolates of Typhimurium in Scotland in 2003 and 2004 are detailed in Table 10.
PFGE analysis of serotypes other than Enteritidis and Typhimurium has identified many different PFGE profiles within individual serotypes. In some cases, the PFGE profile has been used to identify outbreak strains, e.g. the strains involved in the national outbreak of Newport were confirmed by their distinctive PFGE profile. PFGE profiles were used to confirm the identity of isolates of Group B strains involved in a hospital outbreak. Some isolates were rough and/or non-motile and could only be confirmed by the use of genotypic methods.
PFGE profiling provides a valuable means of subtyping phage types of Typhimurium and Enteritidis, improving the epidemiological information available and allowing better targeting of resources. Similarly, the ability to subtype those serotypes for which no phage typing scheme is available has proven to be of great value in outbreak investigation. Continued monitoring of plasmid and PFGE profiles present in Salmonella from human disease, and extension of these techniques to isolates from veterinary, food and environmental sources will provide new insights in the epidemiology of Salmonella in Scotland.
During 2004 there were 453 reports of Salmonella from animal sources. This represents a slight decrease (4%) on the 471 reports made in 2003 (Figure 4)
Isolates from cattle, pigs and sheep decreased while reports from birds increased (Table 11).
As seen in the previous 2 years, S. Dublin was the most common serotype reported from animals (Table 12).
Of the 241 reports of Salmonella from bovine sources, 179 (74%) were S. Dublin. This represented a decrease on the previous year where 93% of cattle isolates were Dublin. Reports of Typhimurium (16%) increased on the previous year, where Typhimurium represented 4% of cattle isolates. This increase was mostly due to an increase in DT104 (20 reports compared to seven in 2003).
Of the 118 reports of Salmonella from avian sources, 110 (93%) were S. Typhimurium. The majority of isolates were either phage type 40 or 56 Var and were associated with wild birds.
The Scottish Salmonella Reference Laboratory would like to thank the staff in Clinical and Veterinary Laboratories for their continued and valued support.