State of Knowledge and Data Gaps of Middle East Respiratory Syndrome Coronavirus ( MERS-CoV ) in Humans

BACKGROUND: Between September 2012 and 22 October 2013, 144 laboratory-confirmed and 17 probable MERS-CoV cases from nine countries were notified to WHO. METHODS: We summarize what is known about the epidemiology, virology, phylogeny and emergence of MERS-CoV to inform public health policies. RESULTS: The median age of patients (n=161) was 50 years (range 14 months to 94 years), 64.5% were male and 63.4% experienced severe respiratory disease. 76.0% of patients were reported to have ?1 underlying medical condition and fatal cases, compared to recovered or asymptomatic cases were more likely to have an underlying condition (86.8% vs. 42.4%, p<0.001). Analysis of genetic sequence data suggests multiple independent introductions into human populations and modelled estimates using epidemiologic and genetic data suggest R<sub>0</sub> is <1, though the upper range of estimates may exceed 1. Index/sporadic cases (cases with no epidemiologic-link to other cases) were more likely to be older (median 59.0 years vs. 43.0 years, p<0.001) compared to secondary cases, although these proportions have declined over time. 80.9% vs. 67.2% of index/sporadic and secondary cases, respectively, reported ?1 underlying condition. Clinical presentation ranges from asymptomatic to severe pneumonia with acute respiratory distress syndrome and multi-organ failure. Nearly all symptomatic patients presented with respiratory symptoms and 1/3 of patients also had gastrointestinal symptoms. CONCLUSIONS: Sustained human-to-human transmission of MERS-CoV has not been observed. Outbreaks have been extinguished without overly aggressive isolation and quarantine suggesting that transmission of virus may be stopped with implementation of appropriate infection control measures.


Fig. 1: a) Country of probable exposure of laboratory-confirmed MERS-CoV cases b) Number of laboratory confirmed MERS-CoV cases by country of probable exposure
Bars do not represent the exact location of cases. The most detailed level of information for place was mapped when possible. For cases where only the province name was reported, the geographic centre of that province is mapped.

Epidemiology of MERS-CoV infection in humans
Our understanding of the epidemiology and clinical presentation of MERS-CoV infection is heavily influenced by the recommended surveillance strategies for case detection, which largely focus on severe illness and virologic testing 4 , and published detailed investigations of cases. Contact tracing activities allow for the detection of confirmed cases with a broader spectrum of illness. Confirmed cases include only those with a positive polymerase chain reaction (PCR) in accordance with the laboratory guidelines for virus genetic material 7 .

Fig. 1: a) Country of probable exposure of laboratory-confirmed MERS-CoV cases b) Number of laboratory confirmed MERS-CoV cases by country of probable exposure
Bars do not represent the exact location of cases. The most detailed level of information for place was mapped when possible. For cases where only the province name was reported, the geographic centre of that province is mapped.

Epidemiology of MERS-CoV infection in humans
Our understanding of the epidemiology and clinical presentation of MERS-CoV infection is heavily influenced by the recommended surveillance strategies for case detection, which largely focus on severe illness and virologic testing 4 , and published detailed investigations of cases. Contact tracing activities allow for the detection of confirmed cases with a broader spectrum of illness. Confirmed cases include only those with a positive polymerase chain reaction (PCR) in accordance with the laboratory guidelines for virus genetic material 7 .

Fig. 1: a) Country of probable exposure of laboratory-confirmed MERS-CoV cases b) Number of laboratory confirmed MERS-CoV cases by country of probable exposure
Bars do not represent the exact location of cases. The most detailed level of information for place was mapped when possible. For cases where only the province name was reported, the geographic centre of that province is mapped.

Fig. 2: Epidemiologic Curve of MERS-CoV Confirmed (n=144) and Probable Cases (n=17)
*After week 24 in 2013, 58 cases were not reported with date for symptom onset. For these 58 cases, the symptom onset date was estimated (date of reporting to WHO minus median of difference between onset date and reporting date of those cases that had both of these dates available; the median difference was calculated by country). Cases are reported by the location where infection is believed to have occurred. Overall, the median age of MERS-CoV patients is 50 years (range 14 months to 94 years) and 64.5% of patients were male ( Table 1). As of 22 October 2013, WHO has confirmation that 65 patients (40.4%), of whom 62 were confirmed and 3 probable, have died, of these 4 deaths could not be matched up to specific cases due to a lack of identifying information. Outcome is unknown for 45 patients, either because they are still in hospital at the time of writing, their outcome has not been reported, or we are unable to match a fatal case to our line list (n=4). Age and sex varied by outcome, with a higher median age and proportion of male sex among patients who died (58 years old and 80.0%, respectively) versus those who recovered or were asymptomatic (34 years old and 60.0%, respectively).
The majority (63.4%) of patients experienced severe respiratory disease while 29.8% were reported to have nonsevere disease, including 18 cases reported as asymptomatic. Severe disease was defined as admission to an intensive care unit [ICU]; use of extracorporeal membrane oxygenation (ECMO), mechanical ventilation, or vasopressors; or reported by the member state as "critical" or "severe"; or who died. Severity is unknown for 6.8% of patients. Fifty-three of the 114 hospitalized patients, for whom we had data, died.
Seventy-six per cent of patients are reported to have at least one underlying medical condition and fatal cases were more likely to have an underlying condition (86.8% among fatal cases vs. 42.4% among recovered or asymptomatic cases, p<0.001, Table 1). Of those who were reported to have at least one comorbid condition, specific comorbid diagnoses were not reported for 47.8% of patients. However, of those with data, the most commonly reported conditions include chronic renal failure (13.3%), diabetes (10.0%), and heart disease (7.5%). Of five patients who were immunocompromised, two reported use of immunosuppressive medication, two reported multiple myeloma, and solid tissue cancer was reported in one patient. A published analysis of 47

Fig. 2: Epidemiologic Curve of MERS-CoV Confirmed (n=144) and Probable Cases (n=17)
*After week 24 in 2013, 58 cases were not reported with date for symptom onset. For these 58 cases, the symptom onset date was estimated (date of reporting to WHO minus median of difference between onset date and reporting date of those cases that had both of these dates available; the median difference was calculated by country). Cases are reported by the location where infection is believed to have occurred. Overall, the median age of MERS-CoV patients is 50 years (range 14 months to 94 years) and 64.5% of patients were male ( Table 1). As of 22 October 2013, WHO has confirmation that 65 patients (40.4%), of whom 62 were confirmed and 3 probable, have died, of these 4 deaths could not be matched up to specific cases due to a lack of identifying information. Outcome is unknown for 45 patients, either because they are still in hospital at the time of writing, their outcome has not been reported, or we are unable to match a fatal case to our line list (n=4). Age and sex varied by outcome, with a higher median age and proportion of male sex among patients who died (58 years old and 80.0%, respectively) versus those who recovered or were asymptomatic (34 years old and 60.0%, respectively).
The majority (63.4%) of patients experienced severe respiratory disease while 29.8% were reported to have nonsevere disease, including 18 cases reported as asymptomatic. Severe disease was defined as admission to an intensive care unit [ICU]; use of extracorporeal membrane oxygenation (ECMO), mechanical ventilation, or vasopressors; or reported by the member state as "critical" or "severe"; or who died. Severity is unknown for 6.8% of patients. Fifty-three of the 114 hospitalized patients, for whom we had data, died.
Seventy-six per cent of patients are reported to have at least one underlying medical condition and fatal cases were more likely to have an underlying condition (86.8% among fatal cases vs. 42.4% among recovered or asymptomatic cases, p<0.001, Table 1). Of those who were reported to have at least one comorbid condition, specific comorbid diagnoses were not reported for 47.8% of patients. However, of those with data, the most commonly reported conditions include chronic renal failure (13.3%), diabetes (10.0%), and heart disease (7.5%). Of five patients who were immunocompromised, two reported use of immunosuppressive medication, two reported multiple myeloma, and solid tissue cancer was reported in one patient. A published analysis of 47

Fig. 2: Epidemiologic Curve of MERS-CoV Confirmed (n=144) and Probable Cases (n=17)
*After week 24 in 2013, 58 cases were not reported with date for symptom onset. For these 58 cases, the symptom onset date was estimated (date of reporting to WHO minus median of difference between onset date and reporting date of those cases that had both of these dates available; the median difference was calculated by country). Cases are reported by the location where infection is believed to have occurred. Overall, the median age of MERS-CoV patients is 50 years (range 14 months to 94 years) and 64.5% of patients were male ( Table 1). As of 22 October 2013, WHO has confirmation that 65 patients (40.4%), of whom 62 were confirmed and 3 probable, have died, of these 4 deaths could not be matched up to specific cases due to a lack of identifying information. Outcome is unknown for 45 patients, either because they are still in hospital at the time of writing, their outcome has not been reported, or we are unable to match a fatal case to our line list (n=4). Age and sex varied by outcome, with a higher median age and proportion of male sex among patients who died (58 years old and 80.0%, respectively) versus those who recovered or were asymptomatic (34 years old and 60.0%, respectively).
The majority (63.4%) of patients experienced severe respiratory disease while 29.8% were reported to have nonsevere disease, including 18 cases reported as asymptomatic. Severe disease was defined as admission to an intensive care unit [ICU]; use of extracorporeal membrane oxygenation (ECMO), mechanical ventilation, or vasopressors; or reported by the member state as "critical" or "severe"; or who died. Severity is unknown for confirmed patients from KSA, most of whom were part of the Al Hasa cluster, found that almost all (45/47) had at least one underlying condition, including diabetes, hypertension, chronic cardiac disease and chronic renal disease 9 .
We classified 51 patients as likely being sporadic (n=29) or "index" cases (n=22) based on the following criteria: 1) a report of having no exposure to other known cases, 2) occurring in an area with no previous cases or no cases within the last 2 months, or 3) reported as the index case in a cluster or being the first case reported with symptoms in a cluster. Ninety-five cases were classified as secondary cases with epidemiologic links to other confirmed MERS-CoV cases (Table 1). For 17 cases, epidemiological classifications remain unclear as no information about contact with other cases is available.
Seventy-three per cent of index/sporadic cases were male, as compared with 60.0% of secondary cases, though this difference was not significant. Notably, this proportion has declined over time; more than 90% of the earliest cases reported were male. Index/sporadic cases were older (median age 59 years vs. 43 years, p<0.001), and more likely to experience severe disease requiring hospitalization or advanced care (90.2% vs. 49.5%, p<0.001) ( Table 1). A similar proportion of index/sporadic cases and secondary cases reported at least one underlying condition (80.9% and 67.2%, respectively), and diabetes, heart disease and immunosuppression were most often reported among index/sporadic cases.
The majority (90.2%) of index/sporadic cases had severe or fatal disease (Table 1). However, the proportion with reported chronic renal failure was higher among secondary cases, which is related to the large outbreak related to haemodialysis units in hospitals in Al Hasa 10 . Only one of the secondary cases (n=95) that occurred outside of the Al Hasa outbreak reported chronic renal failure.
Among secondary cases, 13 MERS-CoV cases are believed to have been infected in household settings, 60 in health care settings (HCS), and one in a workplace other than a HCS. For twenty-one cases, the place of transmission was not reported. Cases associated with HCS in Jordan, France, KSA, UK, UAE, and Qatar 3 , 10 , 11 , 12 included HCW (30 cases) treating MERS-CoV patients, patients seeking treatment in hospitals for conditions unrelated to MERS-CoV (19 cases) and visitors (6 cases). The type of exposure of five additional secondary cases associated with HCS is unknown. In households and the non-HCS workplace, secondary cases occurred among family contacts or co-workers. The specific types of exposure resulting in transmission are currently unknown.
Only 49 cases have information on exposure to animals, including owning or visiting a farm with camels, goats, sheep, chickens, ducks or other animals. Of these, exposure to animals has been reported for only a few cases (n=7; Table 1). confirmed patients from KSA, most of whom were part of the Al Hasa cluster, found that almost all (45/47) had at least one underlying condition, including diabetes, hypertension, chronic cardiac disease and chronic renal disease 9 .
We classified 51 patients as likely being sporadic (n=29) or "index" cases (n=22) based on the following criteria: 1) a report of having no exposure to other known cases, 2) occurring in an area with no previous cases or no cases within the last 2 months, or 3) reported as the index case in a cluster or being the first case reported with symptoms in a cluster. Ninety-five cases were classified as secondary cases with epidemiologic links to other confirmed MERS-CoV cases (Table 1). For 17 cases, epidemiological classifications remain unclear as no information about contact with other cases is available.
Seventy-three per cent of index/sporadic cases were male, as compared with 60.0% of secondary cases, though this difference was not significant. Notably, this proportion has declined over time; more than 90% of the earliest cases reported were male. Index/sporadic cases were older (median age 59 years vs. 43 years, p<0.001), and more likely to experience severe disease requiring hospitalization or advanced care (90.2% vs. 49.5%, p<0.001) ( Table 1). A similar proportion of index/sporadic cases and secondary cases reported at least one underlying condition (80.9% and 67.2%, respectively), and diabetes, heart disease and immunosuppression were most often reported among index/sporadic cases.
The majority (90.2%) of index/sporadic cases had severe or fatal disease (Table 1). However, the proportion with reported chronic renal failure was higher among secondary cases, which is related to the large outbreak related to haemodialysis units in hospitals in Al Hasa 10 . Only one of the secondary cases (n=95) that occurred outside of the Al Hasa outbreak reported chronic renal failure.
Among secondary cases, 13 MERS-CoV cases are believed to have been infected in household settings, 60 in health care settings (HCS), and one in a workplace other than a HCS. For twenty-one cases, the place of transmission was not reported. Cases associated with HCS in Jordan, France, KSA, UK, UAE, and Qatar 3 , 10 , 11 , 12 included HCW (30 cases) treating MERS-CoV patients, patients seeking treatment in hospitals for conditions unrelated to MERS-CoV (19 cases) and visitors (6 cases). The type of exposure of five additional secondary cases associated with HCS is unknown. In households and the non-HCS workplace, secondary cases occurred among family contacts or co-workers. The specific types of exposure resulting in transmission are currently unknown.
Only 49 cases have information on exposure to animals, including owning or visiting a farm with camels, goats, sheep, chickens, ducks or other animals. Of these, exposure to animals has been reported for only a few cases (n=7; Table 1). confirmed patients from KSA, most of whom were part of the Al Hasa cluster, found that almost all (45/47) had at least one underlying condition, including diabetes, hypertension, chronic cardiac disease and chronic renal disease 9 .
We classified 51 patients as likely being sporadic (n=29) or "index" cases (n=22) based on the following criteria: 1) a report of having no exposure to other known cases, 2) occurring in an area with no previous cases or no cases within the last 2 months, or 3) reported as the index case in a cluster or being the first case reported with symptoms in a cluster. Ninety-five cases were classified as secondary cases with epidemiologic links to other confirmed MERS-CoV cases (Table 1). For 17 cases, epidemiological classifications remain unclear as no information about contact with other cases is available.
Seventy-three per cent of index/sporadic cases were male, as compared with 60.0% of secondary cases, though this difference was not significant. Notably, this proportion has declined over time; more than 90% of the earliest cases reported were male. Index/sporadic cases were older (median age 59 years vs. 43 years, p<0.001), and more likely to experience severe disease requiring hospitalization or advanced care (90.2% vs. 49.5%, p<0.001) ( Table 1). A similar proportion of index/sporadic cases and secondary cases reported at least one underlying condition (80.9% and 67.2%, respectively), and diabetes, heart disease and immunosuppression were most often reported among index/sporadic cases.
The majority (90.2%) of index/sporadic cases had severe or fatal disease (Table 1). However, the proportion with reported chronic renal failure was higher among secondary cases, which is related to the large outbreak Only 49 cases have information on exposure to animals, including owning or visiting a farm with camels, goats, sheep, chickens, ducks or other animals. Of these, exposure to animals has been reported for only a few cases (n=7; Table 1).

Clinical Features of MERS-CoV infection in Humans
Data suggest that the clinical presentation of MERS-CoV infection ranges from asymptomatic to very severe pneumonia with the acute respiratory distress syndrome (ARDS), septic shock and multi-organ failure resulting in death. At least two cases had a consumptive coagulopathy during the course of their illness. The clinical course is more severe in patients with immunocompromising conditions and more likely to be mild in individuals without underlying medical conditions (Table 1). Only three cases have been reported in children under 5 years of age 4 .
Typically, the disease starts with fever and cough 3 , 9 , 10 , 11 , chills, sore throat, myalgia and arthralgia 2 , 5 , 9 , 11 , followed by dyspnoea 5 , 9 , 10 , and rapidly progresses to pneumonia, often requiring ventilatory and other organ support. Nearly all symptomatic patients presented with respiratory symptoms; however, one immunocompromised patient was initially admitted to hospital with fever, chills and diarrhoea and later found to have pneumonia [11]. At least one-third of patients also had gastrointestinal symptoms, such as vomiting and diarrhoea 5 , 9 , 10 , 11 . Almost half of the patients developed pneumonia (44.1%), and 20 (12.4%) developed ARDS (Table 1).
Chest radiograph findings vary but are consistent with viral pneumonitis and ARDS: bilateral hilar infiltration, uni-or bilateral patchy densities or infiltrates, segmented or lobar opacities, ground glass appearance, and small pleural effusions have been described. Lower lobes tend to be affected more than upper lobes early in the course of illness; radiographic appearance progresses rapidly. Computed tomographic scans have shown interstitial infiltrates and consolidation compatible with ARDS in severe cases. In some severe cases, renal failure developed concurrently with respiratory failure.
Common laboratory findings include leucopoenia, particularly lymphopaenia 1 , 3 , 5 , 11 . Reports from a few cases found viral RNA in blood 11 , urine 13 and stool 13 but at much lower viral loads than in the respiratory tract.
The viral load in upper respiratory tract specimens is general lower than in the lower respiratory specimens, though data are limited. Co-infection with other respiratory viruses (e.g., parainfluenza, rhinovirus, influenza

Clinical Features of MERS-CoV infection in Humans
Data suggest that the clinical presentation of MERS-CoV infection ranges from asymptomatic to very severe pneumonia with the acute respiratory distress syndrome (ARDS), septic shock and multi-organ failure resulting in death. At least two cases had a consumptive coagulopathy during the course of their illness. The clinical course is more severe in patients with immunocompromising conditions and more likely to be mild in individuals without underlying medical conditions (Table 1). Only three cases have been reported in children under 5 years of age 4 .
Typically, the disease starts with fever and cough 3 , 9 , 10 , 11 , chills, sore throat, myalgia and arthralgia 2 , 5 , 9 , 11 , followed by dyspnoea 5 , 9 , 10 , and rapidly progresses to pneumonia, often requiring ventilatory and other organ support. Nearly all symptomatic patients presented with respiratory symptoms; however, one immunocompromised patient was initially admitted to hospital with fever, chills and diarrhoea and later found to have pneumonia [11]. At least one-third of patients also had gastrointestinal symptoms, such as vomiting and diarrhoea 5 , 9 , 10 , 11 . Almost half of the patients developed pneumonia (44.1%), and 20 (12.4%) developed ARDS (Table 1).
Chest radiograph findings vary but are consistent with viral pneumonitis and ARDS: bilateral hilar infiltration, uni-or bilateral patchy densities or infiltrates, segmented or lobar opacities, ground glass appearance, and small pleural effusions have been described. Lower lobes tend to be affected more than upper lobes early in the course of illness; radiographic appearance progresses rapidly. Computed tomographic scans have shown interstitial infiltrates and consolidation compatible with ARDS in severe cases. In some severe cases, renal failure developed concurrently with respiratory failure.
Common laboratory findings include leucopoenia, particularly lymphopaenia 1 , 3 , 5 , 11 . Reports from a few cases found viral RNA in blood 11 , urine 13 and stool 13 but at much lower viral loads than in the respiratory tract.
The viral load in upper respiratory tract specimens is general lower than in the lower respiratory specimens, though data are limited. Co-infection with other respiratory viruses (e.g., parainfluenza, rhinovirus, influenza

Clinical Features of MERS-CoV infection in Humans
Data suggest that the clinical presentation of MERS-CoV infection ranges from asymptomatic to very severe pneumonia with the acute respiratory distress syndrome (ARDS), septic shock and multi-organ failure resulting in death. At least two cases had a consumptive coagulopathy during the course of their illness. The clinical course is more severe in patients with immunocompromising conditions and more likely to be mild in individuals without underlying medical conditions (Table 1). Only three cases have been reported in children under 5 years of age 4 .
Typically, the disease starts with fever and cough 3 , 9 , 10 , 11 , chills, sore throat, myalgia and arthralgia 2 , 5 , 9 , 11 , followed by dyspnoea 5 , 9 , 10 , and rapidly progresses to pneumonia, often requiring ventilatory and other organ support. Nearly all symptomatic patients presented with respiratory symptoms; however, one immunocompromised patient was initially admitted to hospital with fever, chills and diarrhoea and later found to have pneumonia [11]. At least one-third of patients also had gastrointestinal symptoms, such as vomiting and diarrhoea 5 , 9 , 10 , 11 . Almost half of the patients developed pneumonia (44.1%), and 20 (12.4%) developed ARDS (Table 1).
Chest radiograph findings vary but are consistent with viral pneumonitis and ARDS: bilateral hilar infiltration, uni-or bilateral patchy densities or infiltrates, segmented or lobar opacities, ground glass appearance, and small pleural effusions have been described. Lower lobes tend to be affected more than upper lobes early in the course of illness; radiographic appearance progresses rapidly. Computed tomographic scans have shown interstitial infiltrates and consolidation compatible with ARDS in severe cases. In some severe cases, renal failure developed concurrently with respiratory failure.
Common laboratory findings include leucopoenia, particularly lymphopaenia 1 , 3 , 5 , 11 . Reports from a few cases found viral RNA in blood 11 , urine 13 and stool 13 but at much lower viral loads than in the respiratory tract.
The viral load in upper respiratory tract specimens is general lower than in the lower respiratory specimens, though data are limited. Co-infection with other respiratory viruses (e.g., parainfluenza, rhinovirus, influenza
The median times from symptom onset to hospitalization, admission to an ICU or to death are 4.0 days (range 0-16, n=62), 5.0 (1-15, n=35) and 11.5 days (4-298, n=40), respectively. The duration of hospitalization to either discharge or death was relatively short, with a median of 7.0 days (2-39, n=21) and 9.0 (2-293, n=24) days, respectively. One patient, who died after 298 days of symptom onset, had been treated with ECMO from the third week of illness 2 . Two additional cases died after 35 days or more in hospital, having undergone prolonged ECMO support.

Treatment
In the absence of pathogen-specific interventions, patient management largely depends on provision of organ support, and vigilance for and prevention of complications. In specific circumstances, additional interventions have included empiric use of broad-spectrum antimicrobial agents, antivirals (oseltamivir and/or acyclovir), and the addition of anti-fungal agents to minimize risk of co-infections with opportunistic pathogens.
Lung-protective ventilatory strategies for ARDS, cardiovascular support, antimicrobial therapy for co-infections, and renal replacement therapy for acute renal failure have been used 1 , 2 , 11 , 13 . Case reports of patients supported with ECMO are available for six patients, five of whom have died 2 , 11 , 12 . No case-control data exist to evaluate the effectiveness of such interventions.
Many patients with severe disease have been treated with systemic high-dose corticosteroids, which was intended to reverse the progression of respiratory distress and to prevent lung fibrosis. This appears to have been unsuccessful.
None of the antimicrobial agents used so far, including the antivirals, appear to be successful in improving [1] WHO defines a "cluster" as two or more persons with onset of symptoms within the same 14 day period, and who are associated with a specific setting, such as a classroom, workplace, household, extended family, hospital, other residential institution, military barracks or recreational camp (WHO Interim Surveillance Recommendations, 2013).
A number of the cases in France, Italy, Jordan, KSA, Tunisia, UAE, UK and Qatar have been reported in clusters 1 , 4 , providing evidence that human-to-human transmission has occurred in HCS, households, and the workplace ( Table 2) 3 , 5 , 10 , 11 , 12 , 64 . In France, Italy, Jordan, KSA, Qatar, Tunisia, UAE and UK, secondary cases were identified through intense case-finding and follow-up of contacts. Intensity of follow-up of close contacts of patients varies between countries and has included screening contacts for respiratory symptoms and evidence of infection either using reverse-transcriptase PCR (RT-PCR) or, in some situations, serologic assays ( Table 2).
As the exposure that results in sporadic infection is unknown, it is impossible to estimate the incubation period in these cases. However, data for human-to-human transmission in the clusters are available from France 11 , 19 , A(H1N1)pdm09, herpes simplex, influenza B) has been reported in some patients and secondary nosocomially acquired bacterial infections (Klebsiella pneumoniae, Staphylococcus aureus, Acinetobacter sp., Candida sp.) have been reported in patients who received mechanical ventilation 1 , 2 , 12 , 13 .
The median times from symptom onset to hospitalization, admission to an ICU or to death are 4.0 days (range 0-16, n=62), 5.0 (1-15, n=35) and 11.5 days (4-298, n=40), respectively. The duration of hospitalization to either discharge or death was relatively short, with a median of 7.0 days (2-39, n=21) and 9.0 (2-293, n=24) days, respectively. One patient, who died after 298 days of symptom onset, had been treated with ECMO from the third week of illness 2 . Two additional cases died after 35 days or more in hospital, having undergone prolonged ECMO support.

Treatment
In the absence of pathogen-specific interventions, patient management largely depends on provision of organ support, and vigilance for and prevention of complications. In specific circumstances, additional interventions have included empiric use of broad-spectrum antimicrobial agents, antivirals (oseltamivir and/or acyclovir), and the addition of anti-fungal agents to minimize risk of co-infections with opportunistic pathogens.
Lung-protective ventilatory strategies for ARDS, cardiovascular support, antimicrobial therapy for co-infections, and renal replacement therapy for acute renal failure have been used 1 , 2 , 11 , 13 . Case reports of patients supported with ECMO are available for six patients, five of whom have died 2 , 11 , 12 . No case-control data exist to evaluate the effectiveness of such interventions.
Many patients with severe disease have been treated with systemic high-dose corticosteroids, which was intended to reverse the progression of respiratory distress and to prevent lung fibrosis. This appears to have been unsuccessful.
None of the antimicrobial agents used so far, including the antivirals, appear to be successful in improving [1] WHO defines a "cluster" as two or more persons with onset of symptoms within the same 14 day period, and who are associated with a specific setting, such as a classroom, workplace, household, extended family, hospital, other residential institution, military barracks or recreational camp (WHO Interim Surveillance Recommendations, 2013).
A number of the cases in France, Italy, Jordan, KSA, Tunisia, UAE, UK and Qatar have been reported in clusters 1 , 4 , providing evidence that human-to-human transmission has occurred in HCS, households, and the workplace ( Table 2) 3 , 5 , 10 , 11 , 12 , 64 . In France, Italy, Jordan, KSA, Qatar, Tunisia, UAE and UK, secondary cases were identified through intense case-finding and follow-up of contacts. Intensity of follow-up of close contacts of patients varies between countries and has included screening contacts for respiratory symptoms and evidence of infection either using reverse-transcriptase PCR (RT-PCR) or, in some situations, serologic assays ( Table 2).
As the exposure that results in sporadic infection is unknown, it is impossible to estimate the incubation period in these cases. However, data for human-to-human transmission in the clusters are available from France 11 , 19 , A(H1N1)pdm09, herpes simplex, influenza B) has been reported in some patients and secondary nosocomially acquired bacterial infections (Klebsiella pneumoniae, Staphylococcus aureus, Acinetobacter sp., Candida sp.) have been reported in patients who received mechanical ventilation 1 , 2 , 12 , 13 .
The median times from symptom onset to hospitalization, admission to an ICU or to death are 4.0 days (range 0-16, n=62), 5.0 (1-15, n=35) and 11.5 days (4-298, n=40), respectively. The duration of hospitalization to either discharge or death was relatively short, with a median of 7.0 days (2-39, n=21) and 9.0 (2-293, n=24) days, respectively. One patient, who died after 298 days of symptom onset, had been treated with ECMO from the third week of illness 2 . Two additional cases died after 35 days or more in hospital, having undergone prolonged ECMO support.

Treatment
In the absence of pathogen-specific interventions, patient management largely depends on provision of organ support, and vigilance for and prevention of complications. In specific circumstances, additional interventions have included empiric use of broad-spectrum antimicrobial agents, antivirals (oseltamivir and/or acyclovir), and the addition of anti-fungal agents to minimize risk of co-infections with opportunistic pathogens.
Lung-protective ventilatory strategies for ARDS, cardiovascular support, antimicrobial therapy for co-infections, and renal replacement therapy for acute renal failure have been used 1 , 2 , 11 , 13 . Case reports of patients supported with ECMO are available for six patients, five of whom have died 2 , 11 , 12 . No case-control data exist to evaluate the effectiveness of such interventions.
Many patients with severe disease have been treated with systemic high-dose corticosteroids, which was intended to reverse the progression of respiratory distress and to prevent lung fibrosis. This appears to have been unsuccessful.
None of the antimicrobial agents used so far, including the antivirals, appear to be successful in improving [1] WHO defines a "cluster" as two or more persons with onset of symptoms within the same 14 day period, and who are associated with a specific setting, such as a classroom, workplace, household, extended family, hospital, other residential institution, military barracks or recreational camp (WHO Interim Surveillance Recommendations, 2013).
A number of the cases in France, Italy, Jordan, KSA, Tunisia, UAE, UK and Qatar have been reported in clusters 1 , 4 , providing evidence that human-to-human transmission has occurred in HCS, households, and the workplace ( Table 2) 3 , 5 , 10 , 11 , 12 , 64 . In France, Italy, Jordan, KSA, Qatar, Tunisia, UAE and UK, secondary cases were identified through intense case-finding and follow-up of contacts. Intensity of follow-up of close contacts of patients varies between countries and has included screening contacts for respiratory symptoms and evidence of infection either using reverse-transcriptase PCR (RT-PCR) or, in some situations, serologic assays ( Table 2).
As the exposure that results in sporadic infection is unknown, it is impossible to estimate the incubation period in these cases. However, data for human-to-human transmission in the clusters are available from France 11 , 19 , UK 12 , Italy 20 , Tunisia 21 , KSA 10 and Jordan. The incubation period has been estimated to be just over five days, but could be as long as two weeks (median 5.2 days (95% CI: 1.9 to 14.7) 10 ; 5.5 (95% CI: 3.6-10.2) 22 ).
The incubation period for the primary cases who acquired infection from environmental or animal sources is unknown.
Transmission in all reported clusters has been observed to be limited, and current evidence from contact tracing suggests that transmission did not extend beyond close contacts into the community. Among clusters around exported cases travelling to France, UK, Italy, Germany and Tunisia from the Middle East, transmission to close contacts has been limited 6 , 11 , 12 , 20 , 23 , and secondary attack rates among family members of patients in other clusters appear to be low 3 , 5 , 6 , 10 , 12 , 24 , 25 . Systematic implementation of infection prevention and control measures in reported clusters involving HCS has appeared to limit onward transmission to HCW and hospitalized patients 10 , 11 , 12 .
In HCWs infected by exposure to patients, fewer underlying conditions and milder presentations have been described; however, 37.5% experienced severe disease and at least three died. None of the HCW secondary cases had reported underlying immunosuppression; however, two had hypertension and one had an unspecified underlying comorbid condition. Among secondary cases in HCS who were patients admitted to hospital for other conditions at the time of MERS-CoV infection (n=19), all had underlying conditions, 5.3% reported immunosuppressing conditions, all experienced severe disease, and 89.5% died. UK 12 , Italy 20 , Tunisia 21 , KSA 10 and Jordan. The incubation period has been estimated to be just over five days, but could be as long as two weeks (median 5.2 days (95% CI: 1.9 to 14.7) 10 ; 5.5 (95% CI: 3.6-10.2) 22 ).
The incubation period for the primary cases who acquired infection from environmental or animal sources is unknown.
Transmission in all reported clusters has been observed to be limited, and current evidence from contact tracing suggests that transmission did not extend beyond close contacts into the community. Among clusters around exported cases travelling to France, UK, Italy, Germany and Tunisia from the Middle East, transmission to close contacts has been limited 6 , 11 , 12 , 20 , 23 , and secondary attack rates among family members of patients in other clusters appear to be low 3 , 5 , 6 , 10 , 12 , 24 , 25 . Systematic implementation of infection prevention and control measures in reported clusters involving HCS has appeared to limit onward transmission to HCW and hospitalized patients 10 , 11 , 12 .
In HCWs infected by exposure to patients, fewer underlying conditions and milder presentations have been described; however, 37.5% experienced severe disease and at least three died. None of the HCW secondary cases had reported underlying immunosuppression; however, two had hypertension and one had an unspecified underlying comorbid condition. Among secondary cases in HCS who were patients admitted to hospital for other conditions at the time of MERS-CoV infection (n=19), all had underlying conditions, 5.3% reported immunosuppressing conditions, all experienced severe disease, and 89.5% died. UK 12 , Italy 20 , Tunisia 21 , KSA 10 and Jordan. The incubation period has been estimated to be just over five days, but could be as long as two weeks (median 5.2 days (95% CI: 1.9 to 14.7) 10 ; 5.5 (95% CI: 3.6-10.2) 22 ). The incubation period for the primary cases who acquired infection from environmental or animal sources is unknown.
Transmission in all reported clusters has been observed to be limited, and current evidence from contact tracing suggests that transmission did not extend beyond close contacts into the community. Among clusters around exported cases travelling to France, UK, Italy, Germany and Tunisia from the Middle East, transmission to close contacts has been limited 6 , 11 , 12 , 20 , 23 , and secondary attack rates among family members of patients in other clusters appear to be low 3 , 5 , 6 , 10 , 12 , 24 , 25 . Systematic implementation of infection prevention and control measures in reported clusters involving HCS has appeared to limit onward transmission to HCW and hospitalized patients 10 , 11 , 12 .
In HCWs infected by exposure to patients, fewer underlying conditions and milder presentations have been Islands, 100% of sera from unrelated retired racing camels from Oman and more than 90% of dromedary camels from Egypt were found to have antibody titres reactive with MERS-CoV using a variety of methods, including neutralization. No viral genetic material was detected in a limited number of tested camel sera and faecal samples. Islands, 100% of sera from unrelated retired racing camels from Oman and more than 90% of dromedary camels from Egypt were found to have antibody titres reactive with MERS-CoV using a variety of methods, including neutralization. No viral genetic material was detected in a limited number of tested camel sera and faecal samples.

Virological Characteristics of MERS-CoV
MERS-CoV is an enveloped, single-stranded, positive-sense RNA virus that is a newly recognized species in lineage C of the genus Betacoronavirus within the subfamily Coronavirinae 1 , 28 . The genome is approximately 30.1kb long and contains at least 10 predicted open reading frames (ORF) common to betacoronaviruses, which are expressed from seven subgenomic mRNAs 28 . These ORFs mainly include ORF 1a/1ab, which encode for large replicase polyproteins containing conserved functional domains and several non-structural (NS) proteins of CoV, the spike-surface glycoprotein (S), the small-envelope (E) protein, the matrix (M) protein, and the nucleocapsid (N) protein.

Laboratory Testing of MERS-CoV
Initial detection of viral genome in human clinical samples was made using pancorona primers targeting highly conserved regions of the coronavirus genome. Specific assays for the detection of acute infection with MERS-CoV by real-time RT-PCR (rRT-PCR) have subsequently been developed. Several assays are now in widespread use, including those targeting a region upstream of the E gene (upE) or regions within ORF 1b (nsp14 protein) 37 , ORF 1a (nsp6 protein) 39 , and the nucleocapsid protein gene 38 . The assays for the upE and the ORF 1a targets are considered equally sensitive, while the ORF 1b assay is considered less sensitive than the ORF 1a assay.
The upE assay has been recommended for screening specimens with the ORF 1a assay or other specific gene targets being used for confirmation 39 .

Genetics and Emergence of MERS-CoV
Twenty-two whole sequences, and partial genome sequences from 9 additional cases, have been published in GenBank, from thirty different MERS-CoV infected patients (Table 3) Table 3). The study using most available sequences, estimates emergence in July 2011 with a broad credible interval, from July 2007 to June 2012 27 (Table 3). Future studies providing more genome sequences from humans and animals should add to our understanding of transmission patterns and increase the accuracy of the estimated genomic evolutionary rate. large replicase polyproteins containing conserved functional domains and several non-structural (NS) proteins of CoV, the spike-surface glycoprotein (S), the small-envelope (E) protein, the matrix (M) protein, and the nucleocapsid (N) protein.

Laboratory Testing of MERS-CoV
Initial detection of viral genome in human clinical samples was made using pancorona primers targeting highly conserved regions of the coronavirus genome. Specific assays for the detection of acute infection with MERS-CoV by real-time RT-PCR (rRT-PCR) have subsequently been developed. Several assays are now in widespread use, including those targeting a region upstream of the E gene (upE) or regions within ORF 1b (nsp14 protein) 37 , ORF 1a (nsp6 protein) 39 , and the nucleocapsid protein gene 38 . The assays for the upE and the ORF 1a targets are considered equally sensitive, while the ORF 1b assay is considered less sensitive than the ORF 1a assay.
The upE assay has been recommended for screening specimens with the ORF 1a assay or other specific gene targets being used for confirmation 39 .

Genetics and Emergence of MERS-CoV
Twenty-two whole sequences, and partial genome sequences from 9 additional cases, have been published in GenBank, from thirty different MERS-CoV infected patients (Table 3) Table 3). The study using most available sequences, estimates emergence in July 2011 with a broad credible interval, from July 2007 to June 2012 27 (Table 3). Future studies providing more genome sequences from humans and animals should add to our understanding of transmission patterns and increase the accuracy of the estimated genomic evolutionary rate. large replicase polyproteins containing conserved functional domains and several non-structural (NS) proteins of CoV, the spike-surface glycoprotein (S), the small-envelope (E) protein, the matrix (M) protein, and the nucleocapsid (N) protein.

Laboratory Testing of MERS-CoV
Initial detection of viral genome in human clinical samples was made using pancorona primers targeting highly conserved regions of the coronavirus genome. Specific assays for the detection of acute infection with MERS-CoV by real-time RT-PCR (rRT-PCR) have subsequently been developed. Several assays are now in widespread use, including those targeting a region upstream of the E gene (upE) or regions within ORF 1b (nsp14 protein) 37 , ORF 1a (nsp6 protein) 39 , and the nucleocapsid protein gene 38 . The assays for the upE and the ORF 1a targets are considered equally sensitive, while the ORF 1b assay is considered less sensitive than the ORF 1a assay.
The upE assay has been recommended for screening specimens with the ORF 1a assay or other specific gene targets being used for confirmation 39 .

Genetics and Emergence of MERS-CoV
Twenty-two whole sequences, and partial genome sequences from 9 additional cases, have been published in GenBank, from thirty different MERS-CoV infected patients (Table 3) Table 3). The study using most available sequences, estimates emergence in July 2011 with a broad credible interval, from July 2007 to June 2012 27 (Table 3). Future studies providing more genome sequences from humans and animals should add to our understanding of transmission patterns and increase the accuracy of the estimated genomic evolutionary rate.

MERS-CoV -Host Interactions
The S protein of coronaviruses is responsible for binding to the host cell receptor. The DPP4 (dipeptidyl peptidase 4, also known as CD26) 40  In humans, DPP4 is primarily expressed on the epithelial cells in kidney, small intestine, liver and prostate, and on activated leukocytes 13 , 40 . MERS-CoV has been shown to infect non-ciliated bronchial epithelium, bronchiolar epithelial cells, alveolar epithelial cells, endothelial cells, and lung ex vivo organ cultures at least as well as

MERS-CoV -Host Interactions
The S protein of coronaviruses is responsible for binding to the host cell receptor. The DPP4 (dipeptidyl peptidase 4, also known as CD26) 40  In humans, DPP4 is primarily expressed on the epithelial cells in kidney, small intestine, liver and prostate, and on activated leukocytes 13 , 40 . MERS-CoV has been shown to infect non-ciliated bronchial epithelium, bronchiolar epithelial cells, alveolar epithelial cells, endothelial cells, and lung ex vivo organ cultures at least as well as

MERS-CoV -Host Interactions
The S protein of coronaviruses is responsible for binding to the host cell receptor. The DPP4 (dipeptidyl peptidase 4, also known as CD26) 40

Discussion and gaps in knowledge
Much has been learned about the MERS-CoV and the disease syndrome it causes in humans since the first case was reported in September 2012, but many critical questions remain unanswered. From the time of the first cases reported to WHO, it was suspected that the virus was of animal origin and subsequent observations since that time have reinforced this view. Current evidence suggests that the virus is most closely related genetically to viruses found in a number of species of old world and new world insectivorous bats 1 , 28 , 29 , 30 , 31 , 32 , 33 , 34 .
The finding of a gene fragment in a Saudi bat 34 , which is homologous with that of viruses from humans, suggests that bats may be the ultimate reservoir of virus; however, it is very possible that another intermediate animal reservoir could be involved, as was the case with SARS and with Nipah virus in Malaysia. Serological data suggest infections with MERS-CoV or a closely related virus in camels, however the virus itself has not been detected in camels 36 , 35 . Currently, all known cases have a direct or indirect link to the Middle East.
Although surveillance is more limited in some parts of the world, large clusters like those seen in Al Hasa, KSA and Zarqa, Jordan have not been observed elsewhere. This apparent restriction of transmission from non-human sources to one area of the world may provide clues to the source of exposures. The most likely explanations are either a restriction in the range of the putative host species or exposures related to human behaviours that occur only in the area. Regardless of the reservoir, however, the primary question to be answered is the route of transmission from animal to human. Sporadic and index cases provide the best opportunity for identifying exposures of interest as these are the infections most likely acquired from non-human exposures. The high predominance of older individuals among these cases could be a clue to exposures and demands further investigation using structured case-control studies 53 . The recent increase in numbers of sporadic cases associated with a change in the proportion of male and female cases is concerning. While it could represent changes in surveillance strategies, it may also signal a change in the behaviour of the virus, an expansion of the reservoir, or changes in human exposures.
It is now clear that the virus does transmit from person to person. However, sustained human-to-human transmission has not been observed. The relative ease with which outbreaks have been extinguished without resorting to overly aggressive isolation and quarantine measures and the limited secondary transmission around clusters, suggest that transmission of virus may be readily stopped.
Genetic sequence data indicate that multiple independent introductions into human populations have occurred.
In addition, modelled estimates of R 0 suggest that R 0 is <1, though the upper range of estimates may slightly SARS-CoV 40 , 48 , 49 , 50 . Receptor preference and underlying tissue tropism may influence the severity of highly pathogenic coronaviruses compared with HCoV-229E infection, which is usually milder and has limited infectivity of these lower respiratory tract cells. However, cellular tropism does not completely account for severity of disease.
Cell-based studies have shown that MERS-CoV, like other coronaviruses, evades innate immune recognition, perhaps accounting for the high proportion of cases observed with severe disease 16 , 18 , 48 , 49 . However, MERS-CoV was observed to have higher sensitivity to pegylated interferon treatment, compared with SARS-CoV. This has been hypothesized to be due to the lack of a SARS-CoV ORF6 homologue in MERS-CoV, which has been shown to inhibit the nuclear translocation of p-STAT1 and activation of downstream antiviral genes 28   It is now clear that the virus does transmit from person to person. However, sustained human-to-human transmission has not been observed. The relative ease with which outbreaks have been extinguished without resorting to overly aggressive isolation and quarantine measures and the limited secondary transmission around clusters, suggest that transmission of virus may be readily stopped.

Discussion and gaps in knowledge
Genetic sequence data indicate that multiple independent introductions into human populations have occurred.
In addition, modelled estimates of R 0 suggest that R 0 is <1, though the upper range of estimates may slightly SARS-CoV 40 , 48 , 49 , 50 . Receptor preference and underlying tissue tropism may influence the severity of highly pathogenic coronaviruses compared with HCoV-229E infection, which is usually milder and has limited infectivity of these lower respiratory tract cells. However, cellular tropism does not completely account for severity of disease.
Cell-based studies have shown that MERS-CoV, like other coronaviruses, evades innate immune recognition, perhaps accounting for the high proportion of cases observed with severe disease 16 , 18 , 48 , 49 . However, MERS-CoV was observed to have higher sensitivity to pegylated interferon treatment, compared with SARS-CoV. This has been hypothesized to be due to the lack of a SARS-CoV ORF6 homologue in MERS-CoV, which has been shown to inhibit the nuclear translocation of p-STAT1 and activation of downstream antiviral genes 28   Although surveillance is more limited in some parts of the world, large clusters like those seen in Al Hasa, KSA and Zarqa, Jordan have not been observed elsewhere. This apparent restriction of transmission from non-human sources to one area of the world may provide clues to the source of exposures. The most likely explanations are either a restriction in the range of the putative host species or exposures related to human behaviours that occur only in the area. Regardless of the reservoir, however, the primary question to be answered is the route of transmission from animal to human. Sporadic and index cases provide the best opportunity for identifying exposures of interest as these are the infections most likely acquired from non-human exposures. The high predominance of older individuals among these cases could be a clue to exposures and demands further investigation using structured case-control studies 53 . The recent increase in numbers of sporadic cases associated with a change in the proportion of male and female cases is concerning. While it could represent changes in surveillance strategies, it may also signal a change in the behaviour of the virus, an expansion of the reservoir, or changes in human exposures.

Discussion and gaps in knowledge
It is now clear that the virus does transmit from person to person. However, sustained human-to-human transmission has not been observed. The relative ease with which outbreaks have been extinguished without resorting to overly aggressive isolation and quarantine measures and the limited secondary transmission around clusters, suggest that transmission of virus may be readily stopped.
Genetic sequence data indicate that multiple independent introductions into human populations have occurred.
In addition, modelled estimates of R 0 suggest that R 0 is <1, though the upper range of estimates may slightly exceed 1 22 , 26 . Reports of mild and asymptomatic cases discovered through investigations and testing of contacts of confirmed cases, suggest that the focus on severe disease as a surveillance strategy may miss significant numbers of milder or asymptomatic cases 22 . The importance of these milder cases in transmission, however, is uncertain and needs to be assess with seroepidemiologic studies 54 . Evidence from SARS, however, suggests that asymptomatic cases did not transmit infection 55 . It appears likely that the presence of comorbid conditions, particularly when found in high concentrations such as in HCS, can facilitate transmission.
The basic clinical appearance and course of severe cases seen since the earliest reports has not substantially changed; however, it is now clear that milder and even asymptomatic presentations do occur and that the casefatality rate is lower than observed. The acute renal failure observed in many cases, especially those reported early on, has been primarily associated with hemodynamic compromise, suggesting that it is likely a secondary process. However, receptors for the virus do occur in renal tissue and viral RNA has been detected at low titres in urine. It is important to note that although the primary site of infection has been the respiratory tract, approximately one-third of patients have experienced gastrointestinal symptoms and low titres of virus have been detected in faeces. While generally this is not likely useful for diagnostic purposes, this finding supports current recommendations for adding contact precautions in HCS and may have implications for infection control in the community. The observed differences in proportion of cases that are severe between primary (index/sporadic) and secondary cases is likely a reflection of the surveillance strategy that focuses on severe disease for identifying primary cases and the additional case-finding carried out around them. However, the possibility that the virulence of the virus is different when transmitted from humans than it is when transmitted from non-human sources cannot be excluded based on available evidence. The presence of a pre-existing chronic medical condition likely increases the chance of severe outcome and death. including consumption of a contaminated food product, or contact with a contaminated fomite. Thus far, spread among humans appears to be relatively easy to extinguish with conventional infection control measures and has not required the aggressive measures that were used to contain SARS. Prevention of transmission from the putative animal reservoir to humans is the critical step needed to halt the ongoing spread of this virus. The answer to how to achieve this and other remaining questions requires formal, structured, multinational collaborative research studies, including the development and implementation of effective surveillance exceed 1 22 , 26 . Reports of mild and asymptomatic cases discovered through investigations and testing of contacts of confirmed cases, suggest that the focus on severe disease as a surveillance strategy may miss significant numbers of milder or asymptomatic cases 22 . The importance of these milder cases in transmission, however, is uncertain and needs to be assess with seroepidemiologic studies 54 . Evidence from SARS, however, suggests that asymptomatic cases did not transmit infection 55 . It appears likely that the presence of comorbid conditions, particularly when found in high concentrations such as in HCS, can facilitate transmission.
The basic clinical appearance and course of severe cases seen since the earliest reports has not substantially changed; however, it is now clear that milder and even asymptomatic presentations do occur and that the casefatality rate is lower than observed. The acute renal failure observed in many cases, especially those reported early on, has been primarily associated with hemodynamic compromise, suggesting that it is likely a secondary process. However, receptors for the virus do occur in renal tissue and viral RNA has been detected at low titres in urine. It is important to note that although the primary site of infection has been the respiratory tract, approximately one-third of patients have experienced gastrointestinal symptoms and low titres of virus have been detected in faeces. While generally this is not likely useful for diagnostic purposes, this finding supports current recommendations for adding contact precautions in HCS and may have implications for infection control in the community. The observed differences in proportion of cases that are severe between primary (index/sporadic) and secondary cases is likely a reflection of the surveillance strategy that focuses on severe disease for identifying primary cases and the additional case-finding carried out around them. However, the possibility that the virulence of the virus is different when transmitted from humans than it is when transmitted from non-human sources cannot be excluded based on available evidence. The presence of a pre-existing chronic medical condition likely increases the chance of severe outcome and death. Many critical questions about MERS-CoV and its related disease remain. First among these is the nature of the exposures that result in human infection. Although an animal reservoir of the virus is the most likely hypothesis, the route of transmission could be either direct or indirect contact with the reservoir or an intermediate host, including consumption of a contaminated food product, or contact with a contaminated fomite. Thus far, spread among humans appears to be relatively easy to extinguish with conventional infection control measures and has not required the aggressive measures that were used to contain SARS. Prevention of transmission from the putative animal reservoir to humans is the critical step needed to halt the ongoing spread of this virus. The answer to how to achieve this and other remaining questions requires formal, structured, multinational collaborative research studies, including the development and implementation of effective surveillance exceed 1 22 , 26 . Reports of mild and asymptomatic cases discovered through investigations and testing of contacts of confirmed cases, suggest that the focus on severe disease as a surveillance strategy may miss significant numbers of milder or asymptomatic cases 22 . The importance of these milder cases in transmission, however, is uncertain and needs to be assess with seroepidemiologic studies 54 . Evidence from SARS, however, suggests that asymptomatic cases did not transmit infection 55 . It appears likely that the presence of comorbid conditions, particularly when found in high concentrations such as in HCS, can facilitate transmission.
The basic clinical appearance and course of severe cases seen since the earliest reports has not substantially changed; however, it is now clear that milder and even asymptomatic presentations do occur and that the casefatality rate is lower than observed. The acute renal failure observed in many cases, especially those reported early on, has been primarily associated with hemodynamic compromise, suggesting that it is likely a secondary process. However, receptors for the virus do occur in renal tissue and viral RNA has been detected at low titres in urine. It is important to note that although the primary site of infection has been the respiratory tract, approximately one-third of patients have experienced gastrointestinal symptoms and low titres of virus have been detected in faeces. While generally this is not likely useful for diagnostic purposes, this finding supports current recommendations for adding contact precautions in HCS and may have implications for infection control in the community. The observed differences in proportion of cases that are severe between primary (index/sporadic) and secondary cases is likely a reflection of the surveillance strategy that focuses on severe disease for identifying primary cases and the additional case-finding carried out around them. However, the possibility that the virulence of the virus is different when transmitted from humans than it is when transmitted from non-human sources cannot be excluded based on available evidence. The presence of a pre-existing chronic medical condition likely increases the chance of severe outcome and death. Many critical questions about MERS-CoV and its related disease remain. First among these is the nature of the exposures that result in human infection. Although an animal reservoir of the virus is the most likely hypothesis, the route of transmission could be either direct or indirect contact with the reservoir or an intermediate host, including consumption of a contaminated food product, or contact with a contaminated fomite. Thus far, spread among humans appears to be relatively easy to extinguish with conventional infection control measures and has not required the aggressive measures that were used to contain SARS. Prevention of transmission from the putative animal reservoir to humans is the critical step needed to halt the ongoing spread of this virus. The answer to how to achieve this and other remaining questions requires formal, structured, multinational collaborative research studies, including the development and implementation of effective surveillance