During outbreak investigations in Studies III and IV, the PFGE patterns of selected sporadic strains and strains related to previous outbreaks were also investigated (Table 12, Figure 3, Study IV, Figure 1). Outbreak A occurred in August, 1997, and affected 35 schoolchildren in Pirkkala, Tampere region. Four Y. pseudotuberculosis O:3 isolates available for typing from this study had indistinguishable genotyping patterns S31N31 in PFGE (Figure 3). Outbreak B occurred in August-September 1998, and affected 53 schoolchildren in Mänttä, also in Tampere region. As with outbreak A, the four Y. pseudotuberculosis O:3 isolates available for typing had indistinguishable genotyping patterns S31N31 in PFGE. The sources of outbreaks A and B could not be resolved, but epidemiological investigation suggested vegetables eaten in a school canteen as possibly related to these outbreaks.
Outbreak C started only a month after outbreak B in October 1998 and lasted for three weeks. This outbreak (Study III) comprised four geographically separate clusters of cases in Southern Finland among which 47 culture confirmed Y. pseudotuberculosis O:3 infections were identified. All 27 isolates from case patients available for PFGE had the indistinguishable genotyping patterns S32N32.
A population-based case control study identified iceberg lettuce as a potential source of the infections. During the trace-back investigations, 4 farms in the southwest archipelago were identified as possible sources of iceberg lettuce sold to the cafeterias associated with clusters of case patients. The implicated lot of iceberg lettuce was no longer available for culture. Y. pseudotuberculosis was isolated from one soil and one irrigation water sample taken from one of the suspected farms in November 1999. In subsequent sampling in October 2000, Y. pseudotuberculosis was additionally isolated from two iceberg lettuce samples. One strain isolated from iceberg lettuce was serotype O:2, but the other strains did not agglutinate with O:1-O:6 antisera. The PFGE patterns of environmental strains differed from the outbreak strain patterns.
Outbreak D in October, 1999, comprised three geographically separate clusters of 31 cases also in southern Finland. The 17 isolates available for typing from this outbreak had indistinguishable genotyping patterns S32N32. Although the source of this outbreak could not be identified, iceberg lettuce was again suspected. The outbreak A and B genotype S31N31 had 3 band difference with the corresponding outbreak C and D genotype S32N32.
During 2001, 89 culture-confirmed cases of Y. pseudotuberculosis were reported in Finland; 55 (62 %) were of serotype O:1, and 34 (38 %) were of serotype O:3.
After notification of a suspected food-borne outbreak among schoolchildren from one municipality on 30 May, a case control study and a PFGE study, including Y. pseudotuberculosis isolations from 1 May to 31 July 2001 were conducted.
90
Foodborne Yersinia
Research 11/2009
National Institute for Health and Welfare
Table 12. Distribution of PFGE types among Y. pseudotuberculosis outbreak strains (Studies III and IV) and selected non-outbreak strains (Hallanvuo et al., 2002)
77
Table 12. Distribution of PFGE types among Y. pseudotuberculosisoutbreak strains (Studies III and IV) and selected non-outbreak strains (Hallanvuo et al., 2002)
Origin of the strain Time of isolation Serotype
No of
strains PFGE type Study
Non-outbreak strains February 1994 O:3 1 S32 N32 Hallanvuo et al., 2002
March 1995 O:3 1 S34 N32
April 1997 O:3 1 S32 N32
Outbreak A August 1997 O:3 4 S31 N31 Hallanvuo et al., 2002
Outbreak B September 1998 O:3 4 S31 N31 Hallanvuo et al., 2002
Outbreak C November 1998 O:3 27 S32 N32 Study III
from 4 regions
Non-outbreak strain July 1999 O:3 1 S33 N33 Hallanvuo et al., 2002
Outbreak D October 1999 O:3 17 S32 N32 Hallanvuo et al., 2002
from 3 regions
Non-outbreak strains June 2000 O:3 1 S32 N32 Hallanvuo et al., 2002
May 2000 O:1 1 S11 N11
July 2000 O:1 1 S12 N12
Outbreak E Summer 2001 O:3 13 S32 N321 Study IV
from several regions 5 S33 N322
10 S33 N333
3 S32 N34
1 S34 N31
1 S34 N32
1 S33 N34
O:1 16 S11 N11
32 S12 N12
2 S13 N13
1 S12 N15
1 S15 N16
1 S17 N17
1 S121 N12
1 S122 N122
1Genotype of strains originating from the Western Finland infection clusters (see Figure 3).
2Genotype of strains originating from the city of Mikkeli infection cluster.
3Genotype of strains originating from the Kainuu (region in Northern Finland) infection cluster.
91
Research 11/2009 National Institute for Health and Welfare Materials and methods
The sources of the multiple outbreaks during summer 2001 were not identified.
Iceberg lettuce was again suspected, but eating outside the home was most strongly associated with the infections.
During the summer 2001 outbreak E, the four most common Y.
pseudotuberculosis serotype O:1 and O:3 profiles (S11N11, S12N12, S32N32, and S33N33) accounted for 80% (16 + 32 +13 + 10/89) of the isolates. Conversely, 20% of the strains were divided mostly into single strain genotypes (Table 12).
Clustering of serotype O:3 cases were noticed in Western Finland, Mikkeli region and Kainuu region (Figure 3). The serotype O:3 PFGE type S32N32 that was involved in the previous outbreaks, C and D, appeared in strains representing infections scattered around Western Finland. These infections occurred within two weeks (data not shown). Genotype S33N33 that differed by two bands (SpeI enzyme) and by four bands (NotI enzyme) from the Western Finland genotype appeared in a regional cluster of infections in Kainuu (region in Northern Finland).
Almost simultaneously, genotype S33N32, that differed by two bands (SpeI) from the Western Finland genotype, appeared in a cluster of infections around the city of Mikkeli in mid Eastern Finland. All the serotype O:3 major cluster genotypes differed from each other by 2-4 bands when the results from both enzymes were considered separately (Figure 3). The dominating type S32N32 was also found among sporadic strains isolated before the outbreaks (in 1994 and 1997).
Strains from serotype O:1 infections diverged into two main PFGE types (S11N11 and S12N12) which differed from each other by only one band (Figure 3;
Study IV, Figure 1). These two types had been detected previously among sporadic strains in summer 2000. The infections were concentrated mainly in Southern Finland, but were more separate than O:3 infections in time and geography. An exception to this was a cluster of five infections in the Tampere region. Among the rest of serotype O:1 strains in summer 2001, the diversity was higher. For example, even up to 6 and 10 band differences were seen between SpeI profile S13 and major types S11 and S12, respectively (Table 12, data not shown).
92
Foodborne Yersinia
Research 11/2009
National Institute for Health and Welfare
Figure 3. Left panel: PFGE banding patterns of outbreak and non-outbreak strains of Y.
pseudotuberculosis (Table 12). Mw, Low Range PFG Marker (NEB); S, pattern with enzyme pseudotuberculosis (Table 12). Mw, Low Range PFG Marker (NEB); S, pattern with enzyme pseudotuberculosis
SpeI; N, pattern with enzyme NotI. Right panel: Different clusters of cases in outbreaks of Y.
pseudotuberculosis in Finland. A, outbreak A in Tampere region (August 1997); B, outbreak B in Tampere region (September 1998); C, outbreak C in four different regions in southern Finland (November 1998, Study III); D, outbreak D in three different regions in southern Finland (October 1999); E, outbreak E (summer 2001, Study IV), serotype O:3 infection clusters of Western Finland (W), Mikkeli region (M) and Kainuu region (K) are shown (See also Study IV, Figure 1).
Finland Sweden
Norway
North Sea Barents Sea
Russia A B
C E(W)
E(W)
E(W)
E(K)
E(M)
D
93
Research 11/2009 National Institute for Health and Welfare Discussion