Eurosurveillance

[dothedd]

Eurosurveillance, Volume 18, Issue 24, 13 June 2013

Rapid communications

First cases of Middle East Respiratory Syndrome Coronavirus (MERS-CoV) infections in France, investigations and implications for the prevention of human-to-human transmission, France, May 2013

A Mailles ()1, K Blanckaert2,3, P Chaud2,4, S van der Werf5, B Lina6, V Caro7, C Campese1, B Guéry8, H Prouvost4, X Lemaire9, M C Paty1, S Haeghebaert4, D Antoine1, N Ettahar10, H Noel1, S Behillil5, S Hendricx9, J C Manuguerra7, V Enouf6, G La Ruche1, Caroline Semaille1, B Coignard1, D Lévy-Bruhl1, F Weber1, C Saura1, D Che1, The investigation team11
1.Institut de veille sanitaire (InVS), Saint Maurice, France
2.These authors contributed equally to this work
3.Antenne Régionale de Lutte contre les Infections Nosocomiales (ARLIN), Lille, France
4.Institut de Veille Sanitaire, Lille, France
5.National Reference Center for influenza viruses (coordinating center) and Unit of Molecular Genetics of RNA Viruses, coordinating center, Institut Pasteur, Paris, France
6.National Reference Center for influenza viruses, Hospices Civils de Lyon and Virpath, Université Claude Bernard Lyon1, Lyon, France
7.Cellule d'Intervention Biologique d'Urgence (CIBU), Institut Pasteur, Paris, France
8.Centre Hospitalier Régional et Universitaire, Université de Lille 2, Lille, France
9.Centre hospitalier, Douai, France
10.Centre Hospitalier, Valenciennes, France
11.The members of the team are listed at the end of the article

--------------------------------------------------------------------------------

Citation style for this article: Mailles A, Blanckaert K, Chaud P, van der Werf S, Lina B, Caro V, Campese C, Guéry B, Prouvost H, Lemaire X, Paty MC, Haeghebaert S, Antoine D, Ettahar N, Noel H, Behillil S, Hendricx S, Manuguerra JC, Enouf V, La Ruche G, Semaille C, Coignard B, Lévy-Bruhl D, Weber F, Saura C, Che D, The investigation team. First cases of Middle East Respiratory Syndrome Coronavirus (MERS-CoV) infections in France, investigations and implications for the prevention of human-to-human transmission, France, May 2013. Euro Surveill. 2013;18(24):pii=20502. Available online: www.eurosurveillance.org/ViewArticle.aspx?ArticleId=20502
Date of submission: 29 May 2013
--------------------------------------------------------------------------------



In May 2013, Middle East Respiratory Syndrome Coronavirus (MERS-CoV) infection was diagnosed in an adult male in France with severe respiratory illness, who had travelled to the United Arab Emirates before symptom onset. Contact tracing identified a secondary case in a patient hospitalised in the same hospital room. No other cases of MERS-CoV infection were identified among the index case’s 123 contacts, nor among 39 contacts of the secondary case, during the 10-day follow-up period.

--------------------------------------------------------------------------------

On 7 May 2013, Middle East Respiratory syndrome-Coronavirus (MERS-CoV) infection was confirmed in France in a traveller who became ill after returning from the United Arab Emirates (index case). An investigation was immediately carried out among his contacts since onset of illness, as well as among individuals who had co-travelled with him to the United Arab Emirates. The aim of the investigation was to detect possible other cases and prevent human-to-human transmission. The secondary objective was to try to identify any likely circumstances of exposure to the virus during his travel.

MERS-CoV is a novel virus among the genus Betacoronavirus, which was initially identified in Saudi Arabia in September 2012, in two patients with severe pneumonia [1]. As of 7 May 2013, when the case in France was identified, 30 cases had been confirmed as infected with the virus worldwide, including four diagnosed in the United Kingdom (UK) and two in Germany [2,3].

Surveillance, contact tracing and case finding in France

French surveillance system
In France, suspected cases of MERS-CoV infection have to be reported by attending physicians to regional health agencies and hospital infection control teams. After validation of the classification as a possible case by a French Institute for Public Health Surveillance (InVS) regional office (CIRE), located in a regional health agency, a standardised notification form including socio-demographical information, clinical details, and history of travel in at-risk countries is completed for each possible case.

Up to 17 May, a possible case was defined as follows:

(i) any patient with a history of travel in an at-risk country, who presented with clinical signs and/or imaging consistent with acute respiratory distress syndrome (ARDS) or pulmonary infection, encompassing fever ≥38°C and cough within 10 days after return;
(ii) any contact of a symptomatic possible or confirmed case, presenting with acute respiratory infection, whatever the severity, with an onset of symptoms within 10 days of the last contact with a possible/confirmed case while symptomatic.

The list of at-risk countries, as defined in European Centre for Disease Prevention and Control (ECDC) rapid risk assessment dated 7 December 2012, included, Bahrain, Iran Iraq, Israel, Jordan, Kuwait, Lebanon, Palestine, Oman, Qatar, Saudi Arabia, Syria, United Arab Emirates, and Yemen [4].

For each possible case, respiratory samples (nasopharyngeal aspiration/swab, bronchoalveolar lavage (BAL) fluid when indicated, or induced sputum) are collected and sent to the National Reference Centres for influenza (Institut Pasteur, Paris (coordinating centre) or Hospices civils, Lyon) to be tested for the presence of MERS-CoV genome by real-time reverse transcriptase polymerase chain reaction (RT-PCR) [5,6].

A confirmed case is defined as a possible case with a positive MERS-CoV RT-PCR on respiratory samples [5,6].

Moreover, as part of the usual surveillance of both emerging or nosocomial infections, any cluster of hospitalised patients or healthcare workers (HCW) presenting with severe respiratory infections, regardless of any history of travel in at-risk countries, has to be notified to Public Health Authorities.

Contact tracing and case finding
The contact tracing of all identified cases is implemented as soon as the diagnosis is confirmed. Contacts are defined as all people who provided healthcare to a confirmed case without individual protection, shared the same hospital room, lived in the same household or shared any leisure or professional activity with a confirmed case since this case’s onset of clinical symptoms of MERS-CoV infection (respiratory, digestive or even isolated fever ≥38°C). All contacts are followed-up during a 10-day period (equal to the maximum incubation period according to the knowledge of the disease at the time of the investigation described in this report) after their last contact with the confirmed case to check for clinical symptoms, and asked to measure their body temperature twice a day. The follow-up consists of daily calls from the InVS or CIRE for contacts who are not HCW or from the hospital infection control teams for HCW, to check for the occurrence of clinical symptoms and fever (≥38°C). Contacts are also provided with a hotline number to call anytime in case of any symptom.

For confirmed cases with a history of travel in an at-risk country, a contact tracing of all members of the travel group (co-travellers) is implemented. If the confirmed case had onset of symptoms during the travel, co-travellers are investigated as contacts. Because they potentially have been exposed to the same source of infection (co-exposed), co-travellers are followed-up during a 10-day period after their return from an at-risk country. They are interviewed about the nature and date of their activities, exposure to people presenting with respiratory symptoms, food consumption and exposures to animals, and to aerosols during the travel, in order to investigate the source of infection.

The investigations are carried out with respect to French regulations (authorisation of the Commission Nationale Informatique et Libertés n°341194v42).

Detected confirmed cases
The index case was a 64 year-old male patient with a history of renal transplant, who had returned from the United Arab Emirates on 17 April. He had onset of symptoms on 22 April consisting of fever (38.9°C) and diarrhoea but no respiratory signs. He was admitted in hospital A on 23 April where he was hospitalised until 29 April. On 26 April, the patient presented with dyspnoea and cough; he was transferred to hospital B for a single calendar day to undergo a BAL in a specialised respiratory unit and was re-admitted in hospital A. On 29 April, he was transferred to hospital C in an intensive care unit (ICU). All hospitals were in the same department, whereby hospitals A and B were in the same town, while C and D were in two other towns. Possible MERS-CoV infection was suspected on 1 May and the index case was isolated and individual precautions implemented for HCW and visitors. MERS-CoV infection was confirmed on 7 May. On 8 May, the index case was transferred to hospital D where he was admitted in ICU in a specialised unit with maximal precautions, including a negative pressure room. He died on 28 May 2013, 36 days after onset of symptoms.

Case 2 was identified during the contact tracing of the index case. He was a 51-year-old male patient treated with steroids for several months prior to hospitalisation. He had no history of travel during the weeks before his hospitalisation. He shared with the index case a 20m2 room with a single bathroom in hospital A from 26 to 29 April, while the index case presented with respiratory symptoms (Figure). The beds in the room were 1.5 m apart [7]. He was discharged on 30 April. Onset of symptoms suggestive of MERS-CoV infection occurred on 8 May, 12 days after first exposure. He first presented with malaise, muscle pain and fever (38.5°C) in the afternoon, and cough later that day. As case 2 was known as a contact of the index case, he was admitted in the infectious diseases ward in hospital D and isolated on 9 May. MERS-CoV infection was confirmed during the night of 11 to 12 May. Case 2 was admitted in ICU on 12 May where he is still isolated with the same precautions as the index case.

Figure. Timeline of epidemiological features of two cases of Middle East Respiratory Syndrome Coronavirus (MERS-CoV) infection and exposure and follow-up period of their contacts (n=162), France, April–May 2013

CONTINUED:

[dothedd]

Contact tracing
The index case had travelled in the United Arab Emirates from 9 to 17 April 2013 with 37 co-travellers and his spouse. All co-travellers were interviewed from 10 to 13 May, and none had had any respiratory or digestive symptoms or fever, neither during the journey nor since their return. Except for the spouse, as their interview took place 23 to 26 days after their last contact with the at the time asymptomatic index case, they were not followed-up. All had done the same itinerary and shared common activities with the index case. Their interview did not allow suggesting any hypothesis about the source of infection.

In total, 123 contacts exposed to the index case from his onset of symptoms (22 April) until his isolation (1 May) were identified and interviewed from 8 to 10 May. Six of them were family members who visited the index case in hospital A. Other contacts were 88 HCW and two patients (including case 2) in hospital A, four HCW in hospital B, 20 HCW and three patients in hospital C. Of the five contacts who were patients, only case 2 had shared a room with the index case. No contacts were identified in hospital D, as maximal infection control precautions had been immediately taken. Seven of the total 123 contacts matched the case definition for possible cases and were therefore tested for MERS-CoV infection (samples were taken between one and six days after contacts became symptomatic): only case 2 tested positive.

In total, 39 people were identified as contacts of case 2: 30 had attended a party with case 2 on 8 May, two had visited him at home on 9 May before admission to hospital D, and seven had visited him at home on 9 May and attended the party. Among those 39, 16 had a face-to-face conversation longer than 15 minutes with case 2 and were considered close contacts as described elsewhere [3]. All 39 contacts were interviewed on 12 May, and followed-up until 19 May for those with last contact on 9 May (n=9), and until 18 May for others (n=30). As of 19 May, all were asymptomatic.

Control measures

As soon a MERS-CoV infection was confirmed, the index case and case 2 were isolated, using airborne and contact precautions, in a negative pressure room with dedicated staff [8]. Case 2 had to wear a surgical mask until his medical condition required mechanical ventilation, and HCW who took care of the patients had to wear a filtering face piece (FFP)2 mask [8].

Close contacts of case 2 were asked not to return to work or school until the end of the follow-up, and were provided with surgical masks to wear when not alone and alcohol based hand rub. Other contacts could go on with their usual activities but had to carry a mask, and in case of symptoms, wear it and immediately go back home and call the dedicated hotline [8]. Particular measures for close contacts were implemented after case 2 was diagnosed, and were therefore not applied to contacts of the index case.

Both confirmed cases were notified to the ECDC and the World Health Organization (WHO), respectively on 8 May and 12 May.

Information about the disease and the outbreak was released to the public through the media, and to travellers via flyers and posters disseminated in airports. Specific information about the patients’ management was disseminated to healthcare professionals through mailing lists and institutions’ websites.

Discussion and conclusion

We report the investigation of the first two cases of MERS-CoV diagnosed in France since the emergence of the virus was first described in Saudi Arabia in 2012 [1]. The index case diagnosed in France was imported from the United Arab Emirates, and the second case resulted from a nosocomial infection. Considering that both cases spent four days (26 to 29 April) in the same hospital room, the incubation period of case 2 ranged from nine to 12 days. This emphasises the need for gathering more clinical information from future and past cases to be able to determine precisely the incubation period.

As of 7 June 2013, 55 cases were identified worldwide since the beginning of the worldwide outbreak [9], suggesting a limited human-to-human transmission, even if we assume that some cases may have not been diagnosed.

The index case was initially admitted with an atypical presentation consisting of digestive symptoms but no respiratory signs. Therefore, MERS-CoV infection was not suspected until the patient was in ICU with severe pneumonia. This finding raised the importance of disseminating information about emerging diseases in all hospital settings, including those wards that are not specialised in infectious diseases or critical care.

In-hospital transmission has previously been described in England, in a family member who visited a confirmed case in hospital [10]. A hospital cluster suggestive of nosocomial transmission has also been reported in Saudi Arabia, although the details of the transmission are still under investigation [11]. In France, a secondary infection was diagnosed in another hospitalised patient with underlying condition and long-term steroid treatment. The respiratory presentation of the index case strongly suggests an airborne transmission in the hospital room shared by both patients. However, some questions remain about the possible infectiousness of other body fluids or clinical samples, including stools as the index case presented with diarrhoea at an early stage of his disease, and a cross transmission through contaminated surfaces, medical devices or hands of HCW cannot be ruled out. During the severe acute respiratory syndrome (SARS) outbreak in 2003, a cluster of infections was detected in inhabitants of the same building. Virus aerosols originating from a flat where the index case of the cluster had had digestive symptoms, spread by drainage pipes, were assumed to be the origin of the infection of other cases in the cluster [12].

The large majority of reported MERS-CoV cases worldwide had underlying conditions and presented with severe respiratory infection requiring hospitalisation in ICU. Atypical presentations in immunocompromised patients may be really challenging for clinicians, especially as digestive symptoms are very common in travellers. Based on the index case’s clinical presentation and on knowledge acquired from the SARS outbreak [13], the French case definition for possible cases was extended on 17 May to improve the sensitivity of the surveillance system. It now includes severe febrile clinical signs or febrile diarrhoea in immunocompromised persons or in those with chronic underlying conditions, returning from an at-risk country [14].

Despite the identification of few infections since 2012, MERS-CoV has demonstrated a real potential for nosocomial transmission, and stringent recommendations have to be implemented around possible cases as soon as MERS-CoV infection is suspected. The challenge presented by possible atypical presentations highlights the need for a better knowledge about both the virus and the disease.

Useful knowledge about the infection by MERS-CoV might be obtained from serological investigation in people who shared exposures of confirmed cases, or in contacts of confirmed cases. Such studies might help raising hypothesis about the extent of transmission and risk factors for infection and fatal outcome and must be encouraged.

[dothedd]

 CONTINUED:

www.eurosurveillance.org/ViewArticle.aspx?ArticleId=20502

[dothedd]

Eurosurveillance, Volume 18, Issue 24, 13 June 2013

Perspectives

Transmission scenarios for Middle East Respiratory Syndrome Coronavirus (MERS-CoV) and how to tell them apart

S Cauchemez ()1, M D Van Kerkhove1, S Riley1, C A Donnelly1, C Fraser1, N M Ferguson1
1.MRC Centre for Outbreak Analysis and Modelling, Department of Infectious Disease Epidemiology, Imperial College London, London, United Kingdom

--------------------------------------------------------------------------------

Citation style for this article: Cauchemez S, Van Kerkhove MD, Riley S, Donnelly CA, Fraser C, Ferguson NM. Transmission scenarios for Middle East Respiratory Syndrome Coronavirus (MERS-CoV) and how to tell them apart . Euro Surveill. 2013;18(24):pii=20503. Available online: www.eurosurveillance.org/ViewArticle.aspx?ArticleId=20503

Date of submission: 02 May 2013
--------------------------------------------------------------------------------

Detection of human cases of Middle East Respiratory Syndrome Coronavirus (MERS-CoV) infection internationally is a global public health concern. Rigorous risk assessment is particularly challenging in a context where surveillance may be subject to under-ascertainment and a selection bias towards more severe cases. We would like to assess whether the virus is capable of causing widespread human epidemics, and whether self-sustaining transmission is already under way. Here we review possible transmission scenarios for MERS-CoV and their implications for risk assessment and control. We discuss how existing data, future investigations and analyses may help in reducing uncertainty and refining the public health risk assessment and present analytical approaches that allow robust assessment of epidemiological characteristics, even from partial and biased surveillance data. Finally, we urge that adequate data be collected on future cases to permit rigorous assessment of the transmission characteristics and severity of MERS-CoV, and the public health threat it may pose. Going beyond minimal case reporting, open international collaboration, under the guidance of the World Health Organization and the International Health Regulations, will impact on how this potential epidemic unfolds and prospects for control.

--------------------------------------------------------------------------------
As of 30 May 2013, 50 laboratory-confirmed cases of Middle East Respiratory Syndrome Coronavirus (MERS-CoV) infection have occurred worldwide [1]. An apparently high case-fatality ratio (60%; 30 deaths as of 30 May 2013 [1]) and growing evidence that human-to-human transmission is occurring [2] make MERS-CoV a threat to global health. The current situation has already been compared to the early stages of the severe acute respiratory syndrome (SARS) epidemic in 2003 [3,4].
No animal reservoir has yet been identified for MERS-CoV, and yet human cases, mostly severe, have been detected over a wide geographical area in the Middle East and Europe. If most human cases to date have arisen from animal exposure, this implies a large but as yet uncharacterised zoonotic epidemic is under way in animal species to which humans have frequent exposure (Figure 1A). In this scenario, we might expect relatively small numbers of human cases overall, though with the limited surveillance data available to date, we cannot rule out the possibility that substantial numbers of human cases, with milder disease, have gone undetected.

Even if most human cases to date have been infected through zoonotic exposure, is it possible that MERS-CoV already has the potential to support sustained human-to-human transmission but has by chance so far failed to do so?

Figure 1. Two illustrative scenarios for transmission of Middle East Respiratory Syndrome Coronavirus (MERS-CoV)

Each of these scenarios has very different implications for the assessment of severity, relevance of reservoir-targeted strategies and potential impact of MERS-CoV globally. Although it may not be possible to completely rule out any of the scenarios with the data currently available, it is timely to consider the priorities for data collection and analysis as cases accrue, so as to best be able to reduce uncertainty and refine the public health risk assessment.

Transmission scenarios for an emerging infection

The human-to-human transmissibility (and thus epidemic potential) of an emerging pathogen is quantified by the (effective) reproduction number, R, the average number of secondary infections caused by an index human infection. Depending on the value of R, different transmission scenarios are possible, as described below.

Scenario 1: subcritical outbreaks (R<1)
If R<1, a single spill-over event from a reservoir into human populations may generate a cluster of cases via human-to-human transmission, but cannot generate a disseminated, self-sustaining epidemic in humans. The number of human infections expected under this scenario is roughly proportional to the number of zoonotic introductions of the virus into the human population, with a multiplier, 1/(1−R), that increases with R (twofold if R=0.5, but 10-fold if R=0.9).

In this scenario, human infections can be mitigated by controlling the epidemic in the reservoir and/or preventing human exposure to the reservoir. Examples of this scenario are A(H5N1) and A(H7N9) avian influenzas.

Scenario 2: supercritical outbreaks (R>1 but epidemic has not yet become self-sustaining in human populations)
If R>1, a self-sustaining epidemic in humans is possible but emergence following introduction is a chance event: many chains of transmission may extinguish themselves by chance, especially if R is close to 1. In the case of SARS, for example, where ‘super-spreading’ events played an important role in transmission (i.e. a small proportion of cases were responsible for a large proportion of onward transmission), it has been estimated that there was only a 24% probability that a single introduction would generate a self-sustaining epidemic [5] (following [5], we technically define ‘super-spreading’ events by an over-dispersion parameter k=0.16; the absence of super-spreading events is defined by k=0.5). This is because if the first cases were not part of a super-spreading event, they would be unlikely to generate further cases. However, in this scenario, a self-sustaining epidemic is eventually inevitable if zoonotic introductions into the human population continue (Figure 2). As with the subcritical scenario (R<1), reducing infections from the reservoir is critical to reducing the public health risk.

Figure 2. Probability that the epidemic has become self-sustaining in humans after n introductions from the reservoir if R>1

Scenario 3: self-sustaining epidemic (R>1)
If R>1 and the epidemic has become self-sustaining in humans, the number of human cases is expected to grow exponentially over time. The rate of growth increases with R, but decreases with the mean generation time (GT), the time lag from infection of an index case to infection of those they infect. For example, for an eight-day GT – similar to that of SARS – once self-sustaining, the number of human cases is expected to double about every week if R=2, but only about every month if R=1.2. Although chance effects may mask exponential growth early in the epidemic, a clear signal of increasing incidence would be expected once the number of prevalent infections increases sufficiently [6]. If case ascertainment remains constant over time, the incidence of detected cases would be expected to track that of underlying infections, even if only a small proportion of cases are detected. Once the epidemic is self-sustaining, control of the epidemic in the reservoir would have limited impact on the epidemic in humans.

Publicly available data

As of 30 May 2013, 50 confirmed cases of MERS-CoV have been reported with symptom onset since April 2012 from Saudi Arabia, Jordan, Qatar, United Arab Emirates, the United Kingdom (UK), France and Tunisia [1,2,7-24]. There are additional probable cases from Jordan, Saudi Arabia and Tunisia [1,12,14]. Information on animal exposures is limited and the animal reservoir has not yet been identified. However, we suspect that some of the cases may have arisen from zoonotic exposure in the Arabian Peninsula. Human-to-human transmission is suspected in several familial and healthcare facility clusters in Saudi Arabia, Jordan UK and France. We understand that follow-up investigations of contacts of the confirmed MERS-CoV cases have taken place by Ministry of Health officials in affected countries, finding no evidence of additional symptomatic infection [7-10,15-19]. At this stage, it is difficult to ascertain whether other primary zoonotic or secondary human-to-human cases have been missed. Most cases have been reported as severe disease (40 of 44 with documented severity) and 30 (as of 30 May 2013) have been fatal [25]. Table 1 summarises data for each cluster.

Table 1. Summary information per cluster of Middle East Respiratory Syndrome Coronavirus (MERS-CoV) infection, as of 30 May 2013

Urgent data needs

Existing and additional data will help characterise the MERS-CoV transmission scenario. Many appeals for data have been brought forward by several experts and institutions such as the World Health Organization (WHO). We support this and summarise data requirements and the studies required to collect such data are summarised in Table 2. We illustrate here how these data may be analysed and interpreted with adequate statistical techniques [26-28].

Table 2. Assessing the transmission scenario of a zoonotic virus: data requirements, suggested investigations, parameter estimation and policy implications

TABLE GRAPHS NOT AVAILABLE


Line-list data on confirmed cases
The spatio-temporal dynamics of cases may be used to ascertain whether the epidemic is self-sustaining and if so, to characterise human-to-human transmission [27-29]. It is therefore important that detailed epidemiological information is recorded for all confirmed and probable cases.

Identification of the reservoir species and exposure data
The importance of identifying animal reservoir(s) and understanding human exposure to reservoir species (e.g. direct contact, contact via contaminated food) is well recognised. Once the reservoir has been identified, any exposure of MERS-CoV human cases to that reservoir should be documented in epidemiological investigations. Currently, the uncertainty regarding reservoirs and modes of transmission mean that only five of 50 cases can reliably be classified as ‘human-to-human’ transmission, with the source of infection unclear for the remainder.

If none of the MERS-CoV cases detected by routine surveillance had exposure to the reservoir(s), this would clearly indicate that an epidemic in humans is already self-sustaining [26]. By contrast, if a substantial proportion of cases have been exposed to the reservoir(s), it may be possible to rule out the hypothesis that R≥1.

A similar analytical approach can be used to assess local levels of transmission in countries where MERS-CoV cases are imported from abroad. We can determine if there is self-sustaining transmission in a country by monitoring the proportion of cases detected by routine surveillance with a travel history to other affected countries [26].

If reservoir exposure cannot be found in spite of detailed epidemiological investigations, this may indicate that the epidemic is already self-sustaining in humans. It is therefore important that efforts to identify the reservoir are documented even if they are unsuccessful. To date, very few of the 50 cases have reported contact with animals [1].

Thorough epidemiological investigations of clusters of human cases
Thorough and systematic epidemiological investigations – including contact tracing of all household, familial, social and occupational contacts, with virological and immunological testing – permits assessment of the extent of human infection with MERS-CoV among contacts of confirmed cases [29]. In this context, virological and serological testing is important for ascertaining secondary infections.

As stated above, if R>1, human-to-human transmission will eventually become self-sustaining after a sufficiently large number of virus introductions. So, if thorough cluster investigations indicate that all introductions to date have failed to generate large outbreaks, we can derive an upper bound for R (Figure 3). The distribution of cluster sizes can also be used to estimate R [30,31].

Figure 3. Upper bound for the reproduction number R as a function of the number of introductions from the reservoir that failed to generate self-sustaining epidemics


TABLE GRAPHS NOT AVAILABLE

Routine surveillance is likely to be biased towards severe cases. As a consequence, the case-fatality ratio estimated from cases detected by routine surveillance may be a substantial overestimate. Secondary cases detected during thorough epidemiological investigations of human clusters are expected to constitute a more representative sample of cases in general, meaning more reliable estimates of severity will be obtained by recording clinical outcomes in this subset of cases. Seroepidemiological studies allow for better characterisation of the spectrum of disease, and for the calculation of the proportion of asymptomatic or subclinical infections [29].

Population-level data
Once reliable serological assays are available to measure levels of antibodies to MERS-CoV, it will be important to undertake serological surveys in communities affected early to assess the prevalence of MERS-CoV infection. Should MERS-CoV cases continue to arise in those communities, a rapid follow-up study to collect paired serum samples would be highly valuable. Even a relatively small number of paired sera (about 1,000) could be used to estimate underlying infection rates and refine estimates of severity [32].

Conclusions

We have described three possible transmission scenarios for the emergence of a novel human pathogen from a suspected zoonotic reservoir, with different implications for risk assessment and control.

The most optimistic scenario is that R<1, and thus there is no immediate threat of a large-scale human epidemic. In this scenario, identifying the reservoir will inform efforts to limit human exposure. Detailed genetic investigations and estimation of R are also important for determining the selection pressure and opportunity for the virus to evolve higher human transmissibility [33].

If R>1 but by chance MERS-CoV has not yet generated a self-sustaining epidemic, the total number of animal-to-human infections must have been relatively small. This would suggest that the severe cases that have been detected are not the tip of the iceberg and that disease severity is therefore high.

The final possibility is that R>1 and that human-to-human transmission is already self-sustaining. If this is the case, R must still be relatively low (i.e. <2) unless transmission only began to be self-sustaining in the recent past (e.g. early 2013). In this scenario, overall human case numbers might already be relatively large, suggesting that severity may be substantially lower than it appears from current case reports. Rapid implementation of infection control measures upon detection of MERS-CoV cases may be limiting onward spread beyond close contacts, and may explain the lack of clear-cut evidence from the epidemiological data available thus far that human-to-human transmission is self-sustaining.

Given the current level of uncertainty around MERS-CoV, it is important that adequate data are collected on future cases to underpin rigorous assessment of the transmission characteristics and severity of MERS-CoV, and the public health threat it may pose. This paper has reviewed the epidemiological investigations needed (Table 2); use of standard protocols – being developed by several groups; see available protocols from WHO [34], the Consortium for the Standardization of Influenza Seroepidemiology (CONSISE) [35] and International Severe Acute Respiratory and Emerging Infection Consortium (ISARIC) [36]) – where possible, would be beneficial. Going beyond minimal case reporting, open international collaboration, guided by the International Health Regulations, will impact how this potential epidemic unfolds and prospects for control.

www.eurosurveillance.org/ViewArticle.aspx?ArticleId=20503

[dothedd]

Eurosurveillance, Volume 18, Issue 11, 14 March 2013

Rapid communications

Evidence of person-to-person transmission within a family cluster of novel coronavirus infections, United Kingdom, February 2013

The Health Protection Agency (HPA) UK Novel Coronavirus Investigation team ()1
1.The members of the team are listed at the end of the article

--------------------------------------------------------------------------------

Citation style for this article: The Health Protection Agency (HPA) UK Novel Coronavirus Investigation team. Evidence of person-to-person transmission within a family cluster of novel coronavirus infections, United Kingdom, February 2013 . Euro Surveill. 2013;18(11):pii=20427. Available online: www.eurosurveillance.org/ViewArticle.aspx?ArticleId=20427
Date of submission: 05 March 2013
--------------------------------------------------------------------------------



In February 2013, novel coronavirus (nCoV) infection was diagnosed in an adult male in the United Kingdom with severe respiratory illness, who had travelled to Pakistan and Saudi Arabia 10 days before symptom onset. Contact tracing identified two secondary cases among family members without recent travel: one developed severe respiratory illness and died, the other an influenza-like illness. No other severe cases were identified or nCoV detected in respiratory samples among 135 contacts followed for 10 days.

--------------------------------------------------------------------------------
On 8 February 2013, the Health Protection Agency (HPA) in London, United Kingdom (UK), confirmed infection with novel coronavirus (nCoV) in a patient in an intensive care unit, who had travelled to both Pakistan and Saudi Arabia in the 10 days before the onset of symptoms [1]. This patient (hereafter referred to as Case 1) was the 10th confirmed case reported internationally of a severe acute respiratory illness caused by nCoV. Two secondary cases of nCoV were subsequently detected. We describe the public health investigation of this cluster and the clinical and virological follow-up of their close contacts.
The nCoV was first described in September 2012 in a Saudi Arabian national who died in June 2012 [2,3]. The UK detected its first case of nCoV infection in a male foreign national transferred from Qatar to London in September 2012 [4]. By February 2013, a total of two clusters had been described globally: one cluster (n=2) among staff in a hospital in Jordan and a family cluster (n=3) in Saudi Arabia [5]. No clear evidence of person-to-person transmission was documented in either cluster [6].

Index case exposure history and laboratory investigations

The index case was a middle-aged UK resident, who had travelled to Pakistan for five weeks. He then travelled directly to Saudi Arabia on 20 January where he remained until his return to the UK on 28 January 2013. During his stay in Saudi Arabia, he spent time in Mecca and Medina on pilgrimage. On 24 January, while in Saudi Arabia, he developed fever and upper respiratory tract symptoms (Figure 1). No direct contact with animals or with persons with severe respiratory illness was reported in the 10 days before the onset of illness.

When back in the UK, the patient’s respiratory symptoms worsened and he visited his GP on 30 January; he was admitted to hospital on 31 January. He rapidly deteriorated and required invasive ventilation for respiratory support. Due to further deterioration, he needed extracorporeal membrane oxygenation (ECMO) and was thus transferred to a tertiary centre on 5 February, where he remains severely ill on ECMO as of 1 March.

Initial laboratory investigation included a respiratory virus screen, with confirmation of influenza A infection on 1 February. This was subsequently characterised as influenza A(H1N1)pdm09. As the patient’s clinical condition failed to improve following administration of influenza-specific antiviral drugs, he was subsequently investigated for nCoV infection in line with HPA guidance [7]. On 7 February, nCoV was detected initially in a throat swab with a real-time PCR assay at a local laboratory, and nCoV was confirmed on 8 February by the HPA Respiratory Virus Reference Unit.

Figure 1. Timeline of three novel coronavirus cases, United Kingdom, December 2012 to February 2013




Public health management

Following the confirmation of this imported nCoV case, the UK public health authorities implemented enhanced infection control measures to minimise possible onward transmission of infection: identification and follow-up of contacts to investigate whether transmission had occurred and prompt diagnosis and appropriate management of any further cases. The HPA protocol for investigation of nCoV cases and their close contacts was used [8]. For the purpose of the investigation, a close contact was defined as:
•Aeroplane setting: the aircraft passengers in the same row and the two rows in front and behind a symptomatic case;
•Household setting: any person who had prolonged (>15 minutes) face-to-face contact with the confirmed case(s) any time during the illness in a household setting;
•Healthcare setting: either (i) a worker who provided direct clinical or personal care to or examined a symptomatic confirmed case or was within close vicinity of an aerosol-generating procedure AND who was not wearing full personal protective equipment (PPE) at the time; or (ii) a visitor to the hospital who was not wearing PPE at the bedside of a confirmed case; full PPE was defined as correctly fitted high filtration mask (FFP3), gown, gloves and eye protection;
•Other setting: any person who had prolonged (>15 minutes) face-to-face contact with a confirmed symptomatic case in any other enclosed setting.

Identification and follow-up of individuals who had close contact with the index case from entry into the UK at any time during his symptomatic period was rapidly initiated by the HPA together with staff from the two hospitals the patient had attended (including the Infection Prevention and Control Teams and Occupational Health).

Close contacts were followed up for a minimum period of 10 days after last exposure to the index case. Following the identification of two secondary nCoV cases among symptomatic family contacts of the index case, contact tracing was initiated for their respective additional contacts. Follow-up included collection of information on the date and setting of contact with the index case, PPE use (healthcare workers) and any symptoms of respiratory infection in the 10 days after last exposure. Contacts who developed any symptoms of acute respiratory infection in this period were asked to self-isolate in their homes (or were isolated in hospital if admitted) until asymptomatic.

The airline provided details of passengers to the HPA to allow follow-up of those persons in the same row as the case and the two adjacent rows to the patient as per World Health Organization (WHO) guidance for severe acute respiratory syndrome (SARS) [9]. Passengers who were in the UK were followed up by the HPA to inform them of the potential exposure and determine whether they had developed symptoms of acute respiratory illness in the 10 days post exposure. UK authorities informed relevant overseas national authorities directly about non-UK resident contacts on the flight through International Health Regulation mechanisms.

Laboratory investigation

Symptomatic contacts had respiratory samples taken (nose and throat swab, and sputum if they had a productive cough) for testing for a panel of respiratory viruses (influenza virus, respiratory syncytial virus, parainfluenza virus types 1,2,3 and 4, adenovirus, rhinovirus, human metapneumovirus) and for nCoV. Criteria for laboratory confirmation of nCoV were Up E real-time PCR detection in two different laboratories [3] and detection of two other regions of the nCoV genome [3, HPA unpublished data].

In addition, nose and throat swabs were taken from a group of asymptomatic contacts of the three confirmed cases for nCoV testing to determine if there was evidence of asymptomatic carriage.

Paired serum samples are being taken from all household and healthcare contacts regardless of symptoms with the initial sample taken within seven days of last exposure and the second at least 21 days after the first. Once collected, samples will be tested for serological reactivity to nCoV.

Initial epidemiological investigation of cluster

By 28 February, tracing of contacts of the index case (Case 1) had identified 103 close contacts in the UK, including 59 healthcare workers in the two hospitals, 20 household contacts of whom 15 also visited him at the hospital, 13 family and friends who visited the case in hospital, and 11 contacts during the flight who were UK residents or nationals. In addition there were nine non-UK flight contacts.

Based on available information, a number of healthcare workers with direct contact with Case 1 did not have full PPE, e.g. were not wearing an FFP3 mask. Seven of 59 healthcare workers developed mild, self-limiting respiratory symptoms in the 10 days after last contact. The nCoV was not detected by PCR in the respiratory samples of any of these seven symptomatic contacts (Figure 2).

Six of the 20 household contacts of the index case developed acute respiratory symptoms in the 10 days since last exposure, of whom one progressed to severe illness requiring hospitalisation. This single hospitalised contact was subsequently confirmed to have nCoV infection (hereafter referred to as Case 2), and was also positive for type 2 parainfluenza virus. The remaining five symptomatic household contacts had mild self-limiting disease, and nCoV was not detected from their respiratory samples nor in any of the asymptomatic household contacts of Case 1 that were tested (Figure 2).

One of the 13 non-household contacts visiting Case 1 at the hospital, hereafter referred to as Case 3, developed an acute mild, respiratory illness, and nCoV was detected in a respiratory sample, as was type 2 parainfluenza virus.

Two of the 11 UK-based passengers reported respiratory symptoms: one had recovered by the time of interview and did not have respiratory samples taken. In the other, nCoV was not detected from respiratory samples.

Figure 2. Outcome of contacta follow-up for 10 days after last exposure to index case for respiratory illness and nCoV infection, after entry to the United Kingdom, February 2013 (n=92)




The periods of exposure of Case 2 and Case 3 to Case 1 and the timelines of their illnesses are represented in Figure 1.

Case 2 and his contacts
Case 2 was a male household member, who had an underlying malignant condition, the treatment of which is likely to have resulted in immunosuppression. He had not travelled overseas. Contact with the index case in a household setting occurred from the arrival of Case 1 in the UK until Case 1 was admitted to hospital on 31 January. Case 2 reportedly became unwell on 6 February and was admitted to hospital on 9 February. He required intensive care and ECMO treatment. In a nose and throat swab taken on 10 February, nCoV and type 2 parainfluenza virus were detected. His respiratory condition deteriorated and he died on 17 February.

A number of household contacts (four of 10), hospital visitors (one of one) and healthcare contacts (one of six) of Case 2 developed mild self-limiting respiratory illness in the 10 days after last exposure. In addition, case 2 had one neighbouring patient contact in the hospital, who did not develop symptoms. None had nCoV detected in respiratory samples (Figure 3).

Figure 3. Outcome of contact follow-up for 10 days after last exposure to Case 2 (secondary case) for respiratory illness and nCoV infection, United Kingdom, February 2013 (n=18)




Case 3 and her contacts
Case 3 is an adult female family member of Case 1 who lived in a different household and had not recently travelled abroad. She was exposed to Case 1 only while visiting him in hospital on three separate occasions from 1 to 4 February for a cumulative period of 2.5 hours, during which full PPE was not worn. During these visits Case 1 was intubated on a closed ventilator circuit. Case 3 had no contact with Case 2 while he was unwell. Case 3 developed a self-limiting influenza-like illness starting on 5 February, one day after her last contact with Case 1. She did not require medical attendance for her illness and fully recovered after nine days. She tested positive for nCoV on a single sputum sample taken on 13 February and positive for type 2 parainfluenza virus on a nose and throat swab taken on 15 February. Serology results are awaited.

A total of 25 close contacts of Case 3 were identified (nine household contacts, 14 other contacts, and two healthcare workers) of whom three developed mild self-limiting respiratory illness in the 10 days post exposure. None of these, nor the asymptomatic contacts that were tested, were found to have nCoV in respiratory samples (Figure 4).

Figure 4. Outcome of contact follow-up for 10 days after last exposure to Case 3 (secondary case) for respiratory illness and nCoV infection, United Kingdom, February 2013 (n=25)




Of the 44 contacts of Cases 1, 2 and 3 who were swabbed, 11 had another respiratory virus detected in respiratory samples: rhinovirus (n=7), influenza A(H3) and type 2 parainfluenza virus (n=1), type 2 parainfluenza virus (n=1), type 3 parainfluenza virus (n=1) and metapneumovirus (n=2).

Public health implications

We present evidence of limited person-to-person transmission of nCoV following contact with an index case returning to the UK from travel to Pakistan and Saudi Arabia. Neither of the two secondary cases that were detected had recently travelled and must therefore have acquired their infection in the UK. Both were extended family members and reported contact with the index case. One probably acquired the infection in a household setting and the other while visiting the index case in hospital. The nCoV was not detected among an additional 92 close contacts of the index case, or among the close contacts of the two secondary cases. These findings suggest that although person-to-person infection is possible, there is no evidence at present of sustained person-to-person transmission of nCoV in the UK in relation to this cluster. The limited transmissibility is consistent with the data available to date, with only two other reports of small, self-limited clusters of severe disease in the Middle East: one in a healthcare setting and the other in a household setting [5]. Furthermore, intensive follow-up of close contacts of two other cases imported to European countries has failed to demonstrate onward transmission [10,11].

We found that the index case in this cluster was co-infected with influenza. Type 2 parainfluenza virus was detected in the two secondary cases. This raises questions about what roles these other infections might play in relation to nCoV transmissibility and/or the severity of the illness. In addition, as the index case was diagnosed initially with influenza, this lead to a delay in recognition of nCoV. This highlights the importance of considering a diagnosis of nCoV in atypical cases (in this case the poor response to antiviral drugs), even if a putative alternative diagnosis has already been made. HPA guidance has been adapted accordingly [7].

Although the transmissibility patterns of nCoV and SARS have been different to date, confirmed cases of nCoV reported globally have suggested a clinical picture similar to SARS, in particular the presentation with severe respiratory illness, with nine of the 15 cases reported globally to date having died [12]. Two of the three cases we describe fit this clinical picture: two required ECMO treatment and one of them died. However, the third case presented with an acute self-limiting respiratory infection that did not require hospitalisation or medical attention. This first reported case of a milder nCoV illness raises the possibility that the spectrum of clinical disease maybe wider than initially envisaged, and that a significant proportion of cases now or in the future might be milder or even asymptomatic. This highlights the importance of intensive contact tracing and virological and serological follow-up around all confirmed cases of nCoV. The application of recently developed serological assays in one case¬–contact study did not provide evidence of asymptomatic infection, although the contacts investigated were exposed late in the case’s illness, when the viral load might be lower [11]. Paired sera are being gathered from contacts in this current investigation to determine whether there may have been more widespread mild or asymptomatic infection.

The fact that the two secondary cases acquired their infection from an imported sporadic case has enabled a preliminary estimation of the incubation and serial intervals. The timing of onset of symptoms in the index and the two secondary cases and of exposure suggests a putative incubation period ranging from one to nine days and a serial interval (time between onset of illness in index case and secondary case) of 13 to 14 days. Although the data are extremely limited, the observed upper range of the incubation period is perhaps more similar to that seen for SARS (usual range: two to 10 days) rather than seasonal coronavirus infection (usual range: two to five days) [13]. It is therefore not possible to ascertain with certainty whether the index case acquired his infection in Saudi Arabia or in Pakistan, although previous nCoV cases have been linked to the Middle East. This highlights the importance of gathering more information to determine risk factors for acquisition of infection.

All confirmed nCoV cases detected to date, apart from the two secondary cases in the UK cluster, spent time in the Middle East during the putative incubation period. This, together with our observations of limited secondary transmission, highlights the importance of ongoing vigilance and rapid investigation of cases of severe respiratory illness in residents of and travellers from that area. Further work is required to determine how widely nCoV is circulating globally. In particular serological investigations are needed on the extent of recent infection in various populations, as well as virological investigation of cases of severe undiagnosed respiratory illness in settings both in and beyond the Middle East.

--------------------------------------------------------------------------------
Acknowledgements
We would like to acknowledge Critical Care, Infection Prevention and Control and Occupational Health staff involved in patient care, and all health protection staff involved in the public health response.
--------------------------------------------------------------------------------
Members of the
HPA West Midlands East Health Protection Unit (HPU): Mamoona Tahir, Roger Gajraj, Madhu Bardhan, Huda Mohammed, Louise Dyke, Petra Charlemagne, Rea Alves.
HPA West Midlands West HPU: David Kirrage, Dan Killalea, Kate James, Melinda Kemp.
HPA West Midlands North HPU: Harsh Duggal, Robert Carr, Musarrat Afza, Nicholas Aigbogun, Bharat Sibal.
HPA West Midlands Regional Epidemiology Unit: Ruth Harrell, Obaghe Edeghere, Keith Neal.
HPA West Midlands Regional Director’s Office: Sue Ibbotson.
Birmingham City Hospital: Nimal Wickramasinghe, Nick Sherwood.
University Hospitals Birmingham: Beryl Oppenheim, Louise Hopton.
HPA Specialist Microbiology Network Public Health Laboratory Birmingham: Husam Osman, Erasmus Smit, Sowsan Atabani, Judith Workman, Steve Wilson, Clair Overton-Lewis, Margaret Logan.
HPA Greater Manchester HPU: Rosemary McCann, Marko Petrovic, Vinay Bothra, William Welfare.
University Hospital of South Manchester: Barbara Isalska, Julian Barker, Alan Ashworth, Igor Fedor.
HPA North West London HPU: Claude Seng, Deepti Kumar.
HPA South Yorkshire HPU: Suzanna Matthews.
HPA London Regional Director’s Office: Brian McCloskey.
University of Nottingham: Jonathan Nguyen-Van-Tam.
HPA Health Protection Services – Central Office: Paul Cosford.
HPA Reference Microbiology Services: Alison Bermingham, Joanna Ellis, Monica Galiano, Angie Lackenby, Richard Myers, Robin Gopal, Maria Zambon.
HPA Colindale Health Protection Services: *Richard Pebody, Lucy Thomas, Nicki Boddington, Helen K Green, Hongxin Zhao, Iain Kennedy, Ibrahim Abubakar, Jane Jones, Nick Phin, Mike Catchpole, John M Watson.
Conflict of interest
None declared.

Authors’ contributions
The HPA HPU and regional teams and NHS hospital teams were responsible for the collection of data and samples on cases and their contacts. The HPA Microbiology service teams were responsible for testing and interpretation of results from respiratory samples. National co-ordination of the investigation including design, data collation and analysis was undertaken by the HPS Colindale team in collaboration with other team members. HPS Colindale were responsible for the initial draft of the article. All co-authors provided comments and approved the final version.
www.eurosurveillance.org/ViewArticle.aspx?ArticleId=20427