JOURNAL TRANSCRIPT
Final presentation and evaluation of epidemiological findings in the
EHEC O104:H4 Outbreak Germany 2011
Map showing HUS incidence in the outbreak
EHEC bacteria of the outbreak strain O104:H4 Scanning electron microscope. Scale: 1 μm Source: Holland, Laue (Robert Koch Institute)
Edition information Final presentation and evaluation of the epidemiological findings in the EHEC O104:H4 outbreak, Germany 2011 Robert Koch-Institute, September 2011 Available online at www.rki.de Publisher Robert Koch-Institute (RKI) Nordufer 20 13353 Berlin - Germany Editorial team RKI, Department for Infectious Disease Epidemiology, Division 35 Press RKI-Print Shop Suggested citation Robert Koch Institute. Report: Final presentation and evaluation of epidemiological findings in the EHEC O104:H4 outbreak, Germany 2011. Berlin 2011.
Final report EHEC O104:H4 outbreak, Germany 2011
-1-
Contents Abstract......................................................................................................................... 2 1
Descriptive Epidemiology.................................................................................. 4
1.1 1.2 1.3 1.4 1.5 1.6 1.7
German Notification Data ............................................................................................ 4 Surveillance of bloody diarrhoea .................................................................................. 8 Case reports abroad (as of 18 August, 2011).............................................................. 10 Incubation period......................................................................................................... 11 Estimate of the exposure period.................................................................................. 13 Reporting delays .......................................................................................................... 14 “Now-Casting” ..............................................................................................................17
2
Investigations on the vehicle of infection ....................................................... 19
2.1 2.2 2.3 2.4 2.4.1 2.5 2.5.1 2.5.2
Early epidemiological studies ..................................................................................... 19 Analysis of a satellite outbreak in two canteens of a Frankfurt-based company .... 19 Recipe-based restaurant cohort study ........................................................................20 Findings from additional case-control studies on the consumption of sprouts ......21 “Raw Vegetables” Case-Control Study....................................................................... 22 Investigations of disease clusters ............................................................................... 23 Cooperation with the EHEC Task Force.................................................................... 23 Cohort studies of selected clusters.............................................................................24
3
Bacteriology of the outbreak strain ..................................................................25
3.1 3.2
Detection and characteristics of the pathogen .......................................................... 25 Laboratory tests at the NRC ........................................................................................26
4
Focus of other epidemiological studies........................................................... 28
4.1 4.2 4.3
Studies of population consumption patterns ............................................................28 “Late cases” ................................................................................................................... 31 Domestic environment and shedders.........................................................................33
5
External communication by the RKI during the outbreak...............................34
5.1 5.2
Forwarding of information within the Public Health Service (ÖGD) and to national and international public authorities........................................................................... 34 Media- and public relations ........................................................................................ 36
6
Appendix ...........................................................................................................38
6.1
Schedule of documents provided via Early Warning and Response System during the outbreak ................................................................................................................. 38 Schedule of documents provided during the outbreak on the RKI website (new or updated, as of 25 August 2011)................................................................................... 39
6.2
Supplement .................................................................................................................41
Final report EHEC O104:H4 outbreak, Germany 2011
-2-
Abstract From May to July 2011, primarily in northern Germany, there was a large outbreak of illnesses characterized by haemolytic-uremic syndrome (HUS) and bloody diarrhoea associated with infections by enterohemorrhagic Escherichia coli (EHEC) of the serotype O104:H4. In this report, results from the surveillance, the epidemiological studies and microbiology of the Robert Koch-Institute (RKI) are presented. The contributions by the RKI to the identification of disease clusters and tracing of food items are found in a report by the EHEC Task Force at the Federal Office for Consumer Protection and Food Safety (BVL). Epidemiological characterization and course of the outbreak The outbreak (based on the date of onset of diarrhoea symptoms) began at the beginning of May 2011 and reached its peak on 22 May 2011. Since then, both the number of reported cases of EHEC gastroenteritis associated with the outbreak and the number of new HUS cases decreased. Since mid-June, only sporadic cases of HUS occurred. On 26 July, the Robert Koch-Institute declared that the outbreak had ended. At that point, there had been no new cases clearly associated with the outbreak for 3 weeks, since the last illness on 4 July. A total of 855 cases of HUS and 2,987 cases of acute gastroenteritis attributed to the outbreak were contracted (as of 16 August 2011). The number of cases affecting women outnumbered men in both HUS (68%) and EHEC (58%). The majority of cases involved adults. This is in stark contrast to the observed cases of EHEC gastroenteritis and HUS reported in the last years, in which small children were predominantly affected. Death was reported for 35 (4.1%) of the patients identified with HUS and 18 (0.6%) of the patients with EHEC gastroenteritis. Cases of illness were reported from all federal states, but the 5 most northern states were most affected, including Hamburg, Schleswig-Holstein, Bremen, Mecklenburg-Western Pomerania and Lower Saxony, with HUS incidence in these states up to 10 cases per 100,000 persons. Once the infection vehicle (sprouts) was identified and its distribution was stopped at the beginning of June, there were no further clusters associated with the consumption of this vehicle. In the late stages of the outbreak, cases of secondary transmission by infected persons via close contact within households occured, as well as distinct localized outbreaks that could be attributed to secondary contamination of food products by employees (EHEC shedders) in the food industry. There were also a few recorded laboratory infections. Intensive surveillance for EHEC O104 was continued after the official end of the outbreak in order to identify a potential transition of the infection to an endemic phase. After July 4, 7 additional infections from EHEC O104 were recorded, as yet exclusively cases of apparent transmission within households or from occupational exposure (data as of August 30, 2011). Overall, the frequency of EHEC and HUS reported after July 4 sharply decreased to a rate interpretable as “background”. Evidence for sprouts as the vehicle of infection Large-scale EHEC infection outbreaks typically originate from fecal contamination of vegetable or animal foods which are not sufficiently heated prior to consumption or are typically consumed raw. Evidence for sprouts as the responsible vehicle in this outbreak in Germany arose from epidemiological studies by the Robert KochInstitute in collaboration with regional and local public health authorities, as well as with clinics and the investigations of the federal food safety authorities. Epidemiological studies show a statistically
Final report EHEC O104:H4 outbreak, Germany 2011
-3-
significant association between the consumption of sprouts and risk of disease (e.g. recipe-based restaurant cohort study: relative risk 14.2; 95% CI 2.6 - ; all 31 cases in the cohort study explained by the consumption of sprouts). Investigations by the EHEC Task Force at the BVL revealed that 41 of 41 well-documented localities (e.g. restaurants) in each of which several cases were exposed (so-called clusters) can be traced back to sprouts from Company A in Lower Saxony. In the synopsis of the available results, the RKI, the Federal Institute for Risk Assessment (BfR) and the BVL concordantly concluded that the outbreak in Germany caused by EHEC O104:H4 was attributable to the consumption of contaminated sprouts from Company A. An outbreak involving EHEC O104:H4 reported in France (illness onset between 15 and 20 June) likewise indicated a connection with the consumption of locally produced sprouts. Investigations by national and international food authorities revealed that fenugreek seeds involved in France and in Company A could be traced back to the same supplier. Conclusions for further epidemiological surveillance and recommendations This outbreak of EHEC infections is the largest recorded up to now in Germany and, based on the number of cases of HUS, is the largest outbreak of this sort worldwide. Within a relatively short period of time, epidemiological studies and systematic tracing of food products led to the discovery of sprouts as the vehicle of infection. Currently there is no evidence to suggest that the EHEC O104:H4 pathogen has become endemically established in Germany after the end of the outbreak. The activities of epidemiological surveillance were constant and focussed on surveillance in accordance with the Protection Against Infection Act (IfSG) for EHEC and HUS, as well as emergency room monitoring of cases of bloody diarrhoea in selected hospitals. Furthermore, doctors and the public health services maintained increased vigilance for the occurrence of bloody diarrhoea and HUS including a rapid diagnosis (with differentiation with respect to the outbreak strain) and notification of in- and outpatients. Furthermore, within the context of the notification requirements for EHEC and HUS, all new cases of EHEC and HUS that meet the outbreak case definition continue to be scrutinized by local health authorities based on a questionnaire developed by the RKI (e.g. regarding secondary transmission, laboratory infection) in order to be able to identify the source of infection of these new cases. The explicit advice to consistently observe personal hygiene and food hygiene measures continues to be vital. Strict adherence to hand hygiene (http://www.bzga.de/?sid=663) and other standard measures of hygiene are of central importance. Stringent adherence to hygienic practices is generally essential in a household, but particularly in the presence of EHEC-infected persons or persons with diarrhoea. This means that the utmost cleanliness is especially imperative in the kitchen and bathroom. Apart from direct consumption of contaminated food, the bacteria can also be transmitted via hand contact or contaminated kitchen utensils. This is of particular importance if potentially contaminated food is not subsequently heated. The risk can be reduced if hands and kitchen utensils are washed thoroughly with water and soap/detergent and dried carefully before preparing food, especially food that will not be subsequently cooked. (The recommendations of the BfR can be found at: www.bfr.bund.de > A - Z Index > EHEC). Objects, clothing or surfaces contaminated with feces or vomit should be immediately washed or cleaned; typical household gloves should be worn if there is contact. Recommendations are available at www.rki.de > Infektionskrankheiten A-Z > EHEC.
Final report EHEC O104:H4 outbreak, Germany 2011
-4-
1 Descriptive Epidemiology On 19 May 2011, the RKI was invited by the Health and Consumer Protection Agency in Hamburg to assist the responsible authorities in the investigation of a cluster of three paediatric HUS cases. Upon arrival of the RKI team on 20 May, it quickly became clear that adults were atypically also affected by HUS, and that the number of cases continued to rise rapidly. An outbreak investigation was initiated.
1.1 German Notification Data The electronic reporting system in Germany has collected standardized data covering HUS and EHEC gastroenteritis cases since 2001. Suspicion of disease, disease and death from HUS are notifiable by the treating physician according to §6 of the IfSG (Protection Against Infection Act), and EHEC-detection is notifiable by laboratories according to §7 IfSG. All information pertaining to the cases is amalgamated by the local public health departments (a case is either an EHEC gastroenteritis case OR an HUS case). In addition to the routine surveillance, on 23 May local and state public health departments were requested in a newsletter to report HUS and HUS suspected cases to the RKI immediately after receiving the notification and to add subsequent case investigation results. In contrast to the usual surveillance reference definitions of EHEC gastroenteritis (toxinbased laboratory detection, serogroup optional, and illness with symptoms of gastroenteritis) and HUS (purely clinical case definition, EHEC detection by laboratory diagnostics is optional), the following restrictions were set in place in order to define the cases likely attributable to the outbreak: cases with onset of disease (the typical first symptom is diarrhoea) between 1 May and 4 July 2011 were included (“outbreak time period”). Cases with unknown illness dates were counted from reporting week 19 (beginning 9 May) to 28 (ending 17 July). Cases with evidence for EHEC sub types that do not correspond to the characteristics of the outbreak strain were excluded. This applies to EHEC of serogroups other than the outbreak strain O104:H4 (for details see section 3), as well as to EHEC without serogroup information reported to be only stx1-positive. With data as of 1pm on 16 August 2011, all reported HUS cases (including any remaining suspected cases) and all EHEC cases that met the clinical description were analysed. A total of 855 cases of HUS and 2,987 cases of EHEC gastroenteritis (without development of HUS) were reported, hence a total of 3,842 cases is attributable to the outbreak. For 5% of HUS cases and 9% of EHEC gastroenteritis cases, a date of disease onset is not available. An additional 19 HUS cases and 719 EHEC cases were recorded, which were not ascribed to the outbreak because of the exclusion criteria described above. In the same time period in the previous 5 years (2006-2010), a median of 13 HUS cases and 218 EHEC gastroenteritis cases were reported. For the 2011 outbreak period, this corresponds to a 67-fold increase in HUS and a 17-fold increase in EHEC. The increase in the number of 2011 EHEC cases not ascribed to the outbreak (n=719), well beyond the total EHEC cases in the previous year (n=218), reflects the strongly increased attention to and higher clinical investigation rates for EHEC. Among the HUS cases, 68% were female and the median age was 42 years (range from 0 to 91 years). Among EHEC cases, 58% were female and the median age was 46 years (range from 0 to 100 years). Figure 1 shows the incidence of reported HUS cases by age and gender. Hospitalization is likely in all HUS cases (for EHEC reported for 54%). Among HUS patients, 35 (4.1%) died and among the patients with EHEC gastroenteritis, 18 (0.6%) died.
Final report EHEC O104:H4 outbreak, Germany 2011
80
Incidence female Incidence male Proportion female
70
3
60
50 2
40 30
1
20 10
0
Line: Proportion of female case patients (%)
Columns: HUS Inzidence (Cases/100,000 pop.)
4
-5-
0 0-9
10-19
20-29
30-39
40-49
50-59
60-69
70-79
80+
Age group
Figure 1: Incidence of HUS by age category and gender (left y axis) and the proportion of cases that were female (right y axis) in each age category (n=855 HUS cases).
Disease onset (based on diarrhoea symptoms) of the first case of HUS in an adult was 8 May, which is also the date of the first case of HUS with detection of EHEC O104. On 1 May, a 45-year-old male from Aachen became ill with EHEC. In that case, other diarrhoea pathogens in addition to EHEC O104 were detected, so it is not clear whether the disease onset with respect to EHEC really is 1 May. In the next EHEC case with detection of O104, onset of illness was 8 May, the same as for HUS. This case involved a 42-year-old female from Lower Saxony. Among the HUS cases, the proportion of bloody diarrhoea reported was 79%, and among EHEC cases this proportion was 56%. These values are taken as minimum percentages, since the input screen of the current electronic reporting system only offers the options “bloody diarrhoea” and “diarrhoea, not specified”. Data on laboratory detection of EHEC O104 in known cases is currently still forwarded to RKI. Currently there is information confirming the outbreak strain EHEC O104 for 42% of laboratory-confirmed cases of HUS and 21% of EHEC diarrhoea cases. Based on the distinct excess of cases due to the outbreak compared to the previous year as outlined above, it must however be assumed that almost all these HUS cases can be attributed to the outbreak, and a large percentage of the EHEC cases, the size of which cannot be determined precisely because of the lack of microbiological data. Figure 2 shows the epidemiological curve of HUS and EHEC cases. Both curves sharply rise from 8 May on, peak on the 21 and 22 May, respectively, and then decline quickly at first and more slowly later on. Considering only cases with detection of O104, the peaks are on 20 May (HUS) and 22 May (EHEC gastroenteritis).
Final report EHEC O104:H4 outbreak, Germany 2011
-6-
As of 27 July, no new cases that could be clearly attributed to the outbreak were reported (criteria: HUS in an adult with O104 or no serogroup confirmed, or EHEC with O104 confirmed). Therefore, whe the outbreak was deemed ended as of 5 July the active phase of the outbreak investigation ended and the post-outbreak surveillance phase began. Under these conditions (as of 30 August 2011), 7 additional cases with disease onset dates between 17 July and 14 August came to the attention of the RKI, in which O104 was detected and/or infection with it was probably: one household secondary HUS case in an adult with detection of O104 in the primary case only, as well as 6 EHEC cases with detection of O104.
250
EHEC-gastroenteritis
200
Number of cases
HUS
150
100
50
0 1 2 3 4 5 6 7 8 9 10111213141516171819202122232425262728293031 1 2 3 4 5 6 7 8 9 101112131415161718192021222324252627282930 1 2 3 4 May
June
July
Date of disease onset (diarrhea), 2011
Figure 2: Epidemiological curve for HUS and EHEC outbreak cases (809 HUS and 2,717 EHEC cases with known date of disease onset (diarrhoea) within the outbreak time period.
Within the outbreak time period, cases were reported from all 16 states; however, the five most northern federal states (Hamburg, Schleswig-Holstein, Bremen, Lower Saxony and Mecklenburg-Western Pomerania) were most affected (HUS disease incidences from 1.8 to 10.0 cases per 100,000 persons – all other states had incidence rates