Escherichia coli O104:H4 from 2011 European Outbreak and Strain from South Korea

To the Editor: Beginning in early May 2011, an outbreak caused by Shiga toxin–producing Escherichia coli O104:H4 was reported in Germany and other countries in Europe. In this outbreak, the number of hemolytic uremic syndrome (HUS) cases has been unusually high (1). As of June 9, 2011, a total of 722 cases of HUS, 19 deaths, and 2,745 cases of enterohemorrhagic E. coli (EHEC) infection were reported (2). 
 
A case of HUS caused by E. coli O104:H4 was first reported in South Korea in 2004 (3). Because infections caused by E. coli O104:H4 have been reported rarely, interest has arisen in the E. coli O104:H4 strain from South Korea. We characterized the E. coli O104:H4 strain isolated in South Korea (EC0417119) in 2004 and compared it with the E. coli O104:H4 strain associated with the current EHEC outbreak in Europe. 
 
The serotype EC0417119, isolated from a patient with HUS in 2004, was reconfirmed as E. coli O104:H4. The strain was positive for stx1 and stx2 by PCR (4) but negative for aggR by PCR (5). In the antimicrobial drug susceptibility test using VITEK 2 AST-N169 test kit (bioMerieux, Marcy L’Etoile, France), the strain was resistant to ampicillin, ampicillin/sulbactam, and trimethoprim/sulfamethoxazole but susceptible to ceftriaxone, cefotaxime, nalidixic acid, and tetracycline. 
 
We also performed pulsed-field gel electrophoresis (PFGE) for EC0417119, according to the PulseNet standard protocol (6), and compared its PFGE profile with that of the current outbreak strain E. coli O104:H4, which was obtained from the PulseNet Asia Pacific network. PFGE profiles resolved by either XbaI or BlnI did not match each other. The percentage similarity of XbaI- and BlnI-digested PFGE profiles of the 2 isolates was 75% and 66.7%, respectively, as shown in the Figure. 
 
 
 
Figure 
 
Clustering of A) XbaI- and B) BlnI-digested DNA fragments by pulsed-field gel electrophoresis (PFGE) for Escherichia coli O104:H4 2011 outbreak strain in Europe and isolate obtained in South Korea in 2004. 
 
 
 
Infections with the EHEC O104 strain were reported several times worldwide. In Europe, such occurrence was rare, and before the current outbreak, the EHEC O104:H4 strain was documented only once in South Korea. For this reason, it was logical to examine the possible relatedness of the EC0417119 strain and the strain causing the current outbreak. However, the EC0417119 strain has many different characteristics compared with the current outbreak strain: not possessing enteroaggregative E. coli determinant, not producing extended-spectrum β-lactamases, and not showing indistinguishable PFGE patterns. In conclusion, there is no evidence that the E. coli O104:H4 strain isolated in South Korea in 2004 is related to the strain that has a caused the massive and unprecedented EHEC outbreak in Europe.

We also performed pulsedfi eld gel electrophoresis (PFGE) for EC0417119, according to the PulseNet standard protocol (6), and compared its PFGE profi le with that of the current outbreak strain E. coli O104:H4, which was obtained from the PulseNet Asia Pacifi c network. PFGE profi les resolved by either XbaI or BlnI did not match each other. The percentage similarity of XbaIand BlnI-digested PFGE profi les of the 2 isolates was 75% and 66.7%, respectively, as shown in the Figure. Infections with the EHEC O104 strain were reported several times worldwide. In Europe, such occurrence was rare, and before the current outbreak, the EHEC O104:H4 strain was documented only once in South Korea. For this reason, it was logical to examine the possible relatedness of the EC0417119 strain and the strain causing the current outbreak. However, the EC0417119 strain has many different characteristics compared with the current outbreak strain: not possessing enteroaggregative E. coli determinant, not producing extended-spectrum β-lactamases, and

Respiratory Illness in Households of School-Dismissed Students during Pandemic (H1N1) 2009
To the Editor: In response to the emergence of pandemic (H1N1) 2009 virus (1), the Centers for Disease Control and Prevention (CDC) issued interim guidance for preventing spread of the pandemic virus in schools. Initial guidance recommended that dismissal of students be considered for schools with confi rmed cases of pandemic (H1N1) 2009 infection. The guidance was subsequently revised to recommend monitoring for respiratory illness and exclusion of ill students until they were noninfectious, rather than dismissal.
In Chicago, Illinois, USA, the fi rst cases of pandemic (H1N1) 2009 infection were identifi ed on April 28, 2009, of which 1 occurred in an elementary school student (2). In accordance with CDC guidance at the time, the school (school A) was closed for 1 week, April 29-May 5, 2009. CDC and the Chicago Department of Public Health investigated respiratory illnesses among students and their households during the period surrounding the school closure.
A telephone survey of students' households was conducted during May 15-20, 2009 (3). One adult member of each household was asked whether any household members had been "sick with cold or fl u symptoms or fever" since April 12. Age, date of illness onset, and symptoms and signs (fever, cough, sore throat, rhinorrhea or nasal congestion [runny or stuffy nose]) were recorded. Acute respiratory illness was defi ned as >1 symptom or sign from the list provided. Infl uenza-like illness was defi ned as fever plus cough or sore throat. Reports were excluded if onset date was before April 12 or unknown. Descriptive analysis was performed, and household attack rates were calculated. Dates of onset were used to evaluate timing of illness in relation to school closure and possible transmission within households. The investigation was approved as nonresearch by CDC.
Median age was 10 years (range <1-48 years). Of 57 household members for whom age and student status were recorded, 42 (74%) were students at school A. Thirty-four (60%) reported onset of symptoms before or on the day of school dismissal ( Figure).
Household attack rates ranged from 10% to 100% (median 25%). Five (13%) households reporting illness had no ill students who attended school A. In 4 of 11 households reporting >2 illnesses, students became ill before nonstudent household members. In the remaining 7 households, onset dates did not suggest student-to-nonstudent transmission.
Even though the school was closed almost immediately after the fi rst pandemic (H1N1) 2009 case was confi rmed in a student, onset of ≈60% of reported illnesses occurred before or on the day of school dismissal, suggesting that unrecognized transmission was already occurring in the school or community. These results are supported by data on confi rmed cases of pandemic (H1N1) 2009 in Chicago, which suggest that community transmission was high during the survey period (2). Our results also indicated that at least some