key: cord-295971-jtv1jj2z authors: Cho, Sun Young; Kang, Ji-Man; Ha, Young Eun; Park, Ga Eun; Lee, Ji Yeon; Ko, Jae-Hoon; Lee, Ji Yong; Kim, Jong Min; Kang, Cheol-In; Jo, Ik Joon; Ryu, Jae Geum; Choi, Jong Rim; Kim, Seonwoo; Huh, Hee Jae; Ki, Chang-Seok; Kang, Eun-Suk; Peck, Kyong Ran; Dhong, Hun-Jong; Song, Jae-Hoon; Chung, Doo Ryeon; Kim, Yae-Jean title: MERS-CoV outbreak following a single patient exposure in an emergency room in South Korea: an epidemiological outbreak study date: 2016-07-09 journal: Lancet DOI: 10.1016/s0140-6736(16)30623-7 sha: doc_id: 295971 cord_uid: jtv1jj2z BACKGROUND: In 2015, a large outbreak of Middle East respiratory syndrome coronavirus (MERS-CoV) infection occurred following a single patient exposure in an emergency room at the Samsung Medical Center, a tertiary-care hospital in Seoul, South Korea. We aimed to investigate the epidemiology of MERS-CoV outbreak in our hospital. METHODS: We identified all patients and health-care workers who had been in the emergency room with the index case between May 27 and May 29, 2015. Patients were categorised on the basis of their exposure in the emergency room: in the same zone as the index case (group A), in different zones except for overlap at the registration area or the radiology suite (group B), and in different zones (group C). We documented cases of MERS-CoV infection, confirmed by real-time PCR testing of sputum samples. We analysed attack rates, incubation periods of the virus, and risk factors for transmission. FINDINGS: 675 patients and 218 health-care workers were identified as contacts. MERS-CoV infection was confirmed in 82 individuals (33 patients, eight health-care workers, and 41 visitors). The attack rate was highest in group A (20% [23/117] vs 5% [3/58] in group B vs 1% [4/500] in group C; p<0·0001), and was 2% (5/218) in health-care workers. After excluding nine cases (because of inability to determine the date of symptom onset in six cases and lack of data from three visitors), the median incubation period was 7 days (range 2–17, IQR 5–10). The median incubation period was significantly shorter in group A than in group C (5 days [IQR 4–8] vs 11 days [6–12]; p<0·0001). There were no confirmed cases in patients and visitors who visited the emergency room on May 29 and who were exposed only to potentially contaminated environment without direct contact with the index case. The main risk factor for transmission of MERS-CoV was the location of exposure. INTERPRETATION: Our results showed increased transmission potential of MERS-CoV from a single patient in an overcrowded emergency room and provide compelling evidence that health-care facilities worldwide need to be prepared for emerging infectious diseases. FUNDING: None. Since the fi rst identifi cation of Middle East respiratory syndrome coronavirus (MERS-CoV) infection in 2012, 1 most patients infected with the virus have been exposed in the Middle East. As of March 23, 2016, 1698 laboratoryconfi rmed cases have been reported to WHO. 2 On the basis of previous epidemiological fi ndings, 3 the potential of MERS-CoV to spread to large numbers of people has been considered low, by contrast with severe acute respiratory syndrome coronavirus (SARS-CoV). The basic reproductive number of MERS-CoV was estimated to be less than 1·0, suggesting low transmissibility. 4, 5 However, a 2013 outbreak of MERS-CoV infection in Al Hasa, Saudi Arabia, where one patient infected seven other patients in dialysis and intensive care units, 6 raised concerns about potential so-called super-spreaders 7 that were reported during the SARS epidemic. 8, 9 From May to July, 2015, a large outbreak of MERS-CoV infection occurred in South Korea from a traveller returning from the Middle East, which led to 186 confi rmed cases (Patient 1 to Patient 186) in the country. 10 Patient 1 was diagnosed at our hospital (Samsung Medical Center, Seoul, South Korea) after transmitting the virus at several health-care facilities before he came to our hospital. Patient 14 was exposed to Patient 1 outside the hospital and sought additional care at our hospital without knowing he was infected with MERS-CoV. Therefore, we experienced both South Korea's fi rst MERS-CoV case and the case of highest transmission of MERS-CoV following a single patient exposure in an emergency room. We aimed to investigate the epidemiology of MERS-CoV infection in a crowded emergency room outside of the Middle East and the presence of multiple super-spreaders. In May, 2015, two patients with MERS-CoV infection (Patient 1 and Patient 14) sought care in our emergency room at the Samsung Medical Center without knowing they were infected with MERS-CoV. While these patients were in the emergency room, a large number of patients, visitors, and health-care workers were exposed during both events. When MERS-CoV infection was suspected in Patient 1 and Patient 14, contact investigation was immediately initiated. Since no one developed MERS among contacts who were exposed to Patient 1, only contacts of Patient 14 are reported here. We identifi ed, from electronic medical record review and security video footage, all patients who had been in the emergency room with Patient 14 as contacts, regardless of the location and duration of exposure. We categorised patient contacts into three groups on the basis of their maximum exposure: patients who were in the same zone in the emergency room (group A; considered close contacts), those who were in diff erent zones but had time overlap with Patient 14 in the registration area or radiology suite (30 min before and 2 h after; group B), and those who were in diff erent zones (group C). Patients who were admitted to hospital for treatment of their primary illness after exposure in the emergency room were quarantined in private rooms for 14 days from the last exposure or discharged home after treatment was fi nished and continued isolation at home. Patients and their family members who were already discharged home were reached by telephone, informed about possible MERS-CoV exposure, and provided with hotline numbers for any inquiries. Health-care workers who were exposed were identifi ed through interviews and review of employees' duty schedules, electronic signature on medical records of Patient 14 and patient contacts, security video footage, and self-report. Health-care workers who provided direct care to Patient 14 were initially considered close contacts and were placed into quarantine at home for 14 days from the last day of exposure. Other health-care workers who worked in the emergency room during the same time period continued to work with monitoring and were removed immediately from duty if symptoms developed. A confi rmed case was defi ned as a person with laboratory confi rmation of MERS-CoV infection from sputum samples, initially by real-time RT-PCR testing with amplifi cation targeting the upstream E region (upE) and then confi rmed by subsequent amplifi cation of open reading frame 1a (ORF1a) using PowerChek MERS realtime PCR kits (Kogene Biotech, Seoul, Korea). Patients' demographic information, underlying disease, dates of emergency room visit, duration of stay with exact arrival and departure times, and location within the emergency room were collected. If radiographic examinations were done, the time of examination was collected. For healthcare workers, age, sex, occupation, history of patient assignments and working or visiting zone, and dates and time of duty or emergency room visits were collected. The attack rate was calculated by dividing the number of confi rmed cases by the total number of exposed individuals in the emergency room in each group. Because the total list of visitors was unavailable, we estimated the number of visitors who were in the emergency room by assuming that one patient had at least one visitor during their stay; we also simulated the scenarios of two and four visitors per patient. To avoid underestimation, we chose the assumption of one visitor per patient, which would give the highest attack rate among the scenarios. The incubation period was defi ned as the time of fi rst exposure to the onset of clinical symptoms of MERS-CoV infections. Categorical variables were presented with frequency (percentage) and continuous variables were summarised with median (range, IQR). We calculated overall comparison of attack rates across the groups with χ² test and across zones with Fisher's exact test. Incubation Evidence before this study Little information on nosocomial outbreaks caused by Middle East respiratory syndrome coronavirus (MERS-CoV) outside the Middle East had been available before the large MERS-CoV outbreak in South Korea in 2015, for which global alert was issued. We searched PubMed for reports published in English from May 1, 2015, to Dec 31, 2015, using the terms "MERS-CoV" and "Korea". We identifi ed 38 reports, none of which provided detailed description for the contact investigation of massive transmission of MERS-CoV from a super-spreader in an overcrowded emergency room setting. To our knowledge, this study is the fi rst to categorise exposed patients into groups according to the type of exposure and to document group-specifi c incubation periods and attack rates. Furthermore, this study provides detailed epidemiological data, including a fl oor plan of the emergency room, to understand how MERS-CoV spread by a single super-spreader through several modes of transmission. Results from our contact investigation showed increased transmission potential of MERS-CoV from a single spreader, as has been documented in the severe acute respiratory syndrome epidemic. The potential for similar outbreaks anywhere in the world should be noted, as long as MERS-CoV transmission continues in the Middle East. Our study provides evidence that hospitals, laboratories, and governmental agencies should be prepared for MERS-CoV infection. period and exposure time were compared among groups with Kruskal-Wallis test, followed by Tukey's test using ranks for multiple comparisons. To assess the risk factors for MERS-CoV infection among all patient contacts, we did a multiple logistic regression analysis based on likelihood ratio, by regressing on age, sex, underlying disease, and groups. In a subgroup analysis of patients in group A, the length of stay in the same zone and location were included. For these analyses, odds ratios and 95% CIs were reported. p values and 95% CIs were adjusted with Bonferroni's correction for multiple comparisons if necessary. Two-sided p values of less than 0·05 were considered signifi cant. We used SAS version 9.4 and GraphPad Prism version 6.04 for statistical analyses. Samsung Medical Center is a modern 1982-bed university-affi liated tertiary hospital providing referral care in South Korea (total population roughly 50 million), with roughly 9000 staff , including more than 1400 physicians and 2600 nurses. The emergency room entrance is located on the ground fl oor near the south gate of the main hospital building. More than 200 patients are seen in the emergency room each day; the average duration of stay in the emergency room was 15 h before the MERS-CoV outbreak (see appendix p 2 for details on emergency room overcrowding index). The emergency room has seven patient care areas, including zones I to IV for See Online for appendix adults, a trauma zone, a resuscitation room, and a paediatric zone (fi gure 1). The paediatric zone and zone IV are separated from the rest of the main areas. Two negative-pressure rooms are located in the paediatric zone and two are in zone IV. The emergency room has its own radiology suite for emergency room patients only. The sizes of each zone were as follows: zone I 121·7 m², zone II 64·2 m², zone III 168·3 m², and zone IV 223·9 m². Zones I and II included seating areas (50 chairs in zone I and 26 chairs in zone II), where stable patients received treatment and waited for test results. Seriously ill patients, who required close observation and needed a designated bed, were moved to zone III (17 beds) or zone IV (23 beds). Zone IV was used for patients being admitted. Beds in zones III and IV were spaced roughly 1·8 m apart, with curtains in between. Nurses were assigned to work in designated zones, whereas physicians and transfer agents took care of patients in several zones. All zones in the emergency room were covered by the same air handling units. There was no funding source for this study. The corresponding author had full access to all the data in the study and had fi nal responsibility for the decision to submit for publication. Upon arrival at our emergency room, he denied having travel history to the Middle East and any possible exposure to people infected with MERS-CoV. He was treated for possible bacterial pneumonia on the basis of partial improvement from previous antibiotic treatment and increased C-reactive protein concentration of 13·0 mg/dL (normal <0·3 mg/dL; appendix p 4). During his stay in the emergency room, he was provided with a mask but frequently could not hold it because of severe respiratory symptoms. He was not isolated in a separate room; a negative-pressure room was not considered at that time. As his dyspnoea aggravated on May 28, supplemental oxygen was administered at 2 L per min via a nasal cannula (up to 5 L per min). However, no aerosol-producing procedures, including nebuliser treatments, were given. On the night of May 29, he received a notifi cation call from the health authorities about possible exposure to Patient 1, notifi ed our hospital, and was immediately transferred from the emergency room to isolation in a negativepressure isolation room. MERS-CoV infection was confi rmed on May 30, and he was transferred to the nationally designated health-care facility. From May 27 to May 29, he stayed in three zones in our emergency room: zone II for roughly 10 h on May 27, zone III for 19 h from May 27 to May 28, and zone IV for 25 h from May 28 to May 29 (fi gure 1). Additionally, from May 27 to May 29, he went to the radiology suites four times. On May 27, he walked around and outside the emergency room and went to the toilet several times because of diarrhoea. Between May 27 and May 29, 2015, the average ventilation rate in the emergency room was maintained at three air changes per h, taking 2 h to remove airborne contaminant with a 99·9% effi ciency. 11 The median temperature was 23·8°C (range 14·4-32·2), and the median relative humidity was 32·9% (range 27·1-36·8). 675 patients (117 in group A, 58 in group B, and 500 in group C), an estimated 683 visitors, and 218 health-care workers were identifi ed as contacts of Patient 14 (table 1). We assumed that each patient had one visitor and added eight extra visitors (fi ve to group A and three to group C) The epidemic curve of this emergency room-associated outbreak is shown in fi gure 2. The incubation period was determined from 73 confi rmed MERS-CoV cases: six cases were excluded because we could not determine the date of symptom onset, and data were not available from three visitors. The median incubation period was 7 days (range 2-17, IQR 5-10). Among 59 patients and visitors in groups A-C (excluding six who were not initially identifi ed as contacts), the median incubation period was signifi cantly shorter in group A than in group C (fi gure 2). Excluding three patients with confi rmed MERS-CoV infection who were not identifi ed in the initial patient contact investigation (appendix p 5), the overall attack rate for patients in the emergency room was 4% (30 of 675). Patients in group A had the highest attack rate (20% [23 of 117]), compared with 5% (three of 58) in group B and 1% (four of 500) in group C (fi gure 3). After adjusting for age, sex, underlying disease, and groups, patients in group A had the highest risk for MERS-CoV infection (table 2). In group B, all three patients who had MERS-CoV infection had time overlap in the radiology suite with Patient 14. The median exposure time for patients in group A to Patient 14 was 3·0 h in zone II (range 0·5-10·3, IQR 1·9-4·5), 13·9 h in zone III (0·6-18·9, 6·3-18·4), and 17·4 h in zone IV (0·2-23·2, 9·2-21·4). The attack rates were 23% (13 of 57) in zone II, 32% (seven of 22) in zone III, and 8% (three of 38) in zone IV (fi gure 3). After adjusting for age, sex, underlying disease, and exposure time, staying in zone II was associated with a signifi cantly higher risk for MERS-CoV infection than staying in zone IV (table 2) . MERS-CoV transmission occurred in zone III, despite the fact that the distance from Patient 14's bed to the beds of other patients were as far as 6 m (fi gure 4). In zone IV, Patient 14 moved from bed 12 to bed 23, and six additional cases were documented in patients and visitors occupying beds in the middle of this zone, which were not adjacent to Patient 14's bed. No MERS-CoV infection was reported in patients and visitors who had been in the emergency room on May 29 during the time period when they were exposed only to zones II (n=81) or III (n=15), while Patient 14 was confi ned to zone IV. These patients were exposed to areas that were potentially environmentally contaminated but not to Patient 14 himself (fi gure 4). Under the assumption of one visitor per patient and excluding three visitors with confi rmed MERS-CoV infection who were not identifi ed in the initial visitor contact investigation (appendix p 5), the overall attack rate for visitors was 6% (38 of 683). All patient contacts (n=675) Any underlying disease 1·97 (0·79-5·41) 0·12 Error bars represent 95% CI. MERS-CoV=Middle East respiratory syndrome coronavirus. The attack rates for patients and visitors were 20% (47 of 239) in group A, 5% (six of 116) in group B, and 2% (15 of 1003) in group C. Under the assumptions of two visitors per patient and four visitors per patient, the overall attack rates for visitors were 3% and 1%, respectively. 218 health-care worker contacts were identifi ed, and fi ve (2%) developed MERS-CoV infection. Three healthcare workers who were not initially identifi ed as contacts (one security guard, one physician, and one patient transfer agent) developed MERS-CoV infection. Although they were not involved in the direct care for Patient 14, they visited the emergency room between May 27 and May 29. Only close contacts were furloughed and other health-care workers were isolated when they developed symptoms. There were no secondary cases from health-care workers among contacts during the their duty hours. We did a contact investigation of the MERS outbreak at the Samsung Medical Center by grouping exposed individuals on the basis of the extent of exposure to Patients 1 and 14. To our knowledge, we are the fi rst to document group-specifi c incubation periods and attack rates. Our results showed the increased transmission potential of MERS-CoV from a single patient in an overcrowded emergency room setting. Overcrowding is an important issue for this outbreak and is also a common feature of modern medicine. This study is unique because the index exposure occurred in a large emergency room in a tertiary-care centre, with electronic medical record information available to track the location and duration of exposure, thus enabling near-complete tracing of exposed contacts. The classic defi nitions of close contact as being within roughly 6 feet (1·8 m) or within the same room or care area for a prolonged period of time were diffi cult to apply to an emergency room setting with high patient volumes, ongoing traffi c within the emergency room and to and from the radiology suite, and large numbers of visitors and family members. We considered all patients who visited the emergency room during the stay of Patients 1 and 14 as exposed contacts, developed criteria for close contacts by expanding on the defi nitions of the US Centers for Disease Control and Prevention (CDC), 12 and categorised patients into diff erent groups. Therefore, we could establish group-specifi c viral incubation periods and attack rates during the outbreak. In close contacts who stayed in the same zone, the incubation period was shorter and attack rate was higher than patients who stayed in diff erent zones. Additionally, in zones III and IV, patients were infected even when they were separated by curtains most of the time and were apart as far as 6 m (for beds on either side of the nurse's station; fi gure 4). Similar to the SARS outbreak, we observed so-called superspreaders among patients with MERS-CoV infection, and these super-spreaders can cause large outbreaks through several modes of transmission similar to those in the SARS outbreak. 13 Among patients who stayed in various locations, those who overlapped with Patient 14 at the radiology suite or registration area had higher attack rates (5%) than the rest of the patients (1%), suggesting that transmission might occur by even brief exposures to recently contaminated objects or encounters with individuals carrying a super-spreader. Comparisons of environmental exposure and patient exposure also revealed unique fi ndings. No patient developed MERS-CoV infection after exposure on May 29 only to the environment that had been potentially contaminated on May 27 (zone II) and May 28 (zone III) while Patient 14 was confi ned to zone IV on May 29. It is plausible that even if the environment was heavily contaminated by a super-spreader, the virus might not persist long enough in the environment to be capable of causing any new infection. Although patient exposure is clearly the most important factor in the spread of MERS-CoV, more research is needed to address the potential of environmental spread. 14 Increased viral load and larger amounts of respiratory secretions have been suggested as the factors for SARS-CoV super-spreaders. 15, 16 In this MERS outbreak, frequent ambulation of the index case could be considered as one factor related to high levels of viral transmission, in addition to large amounts of respiratory secretions and high viral load (cycle thresholds 18·6 for upE and 19·3 for ORF1a from Patient 1's sputum, 17 and 16·2 for upE and 19·9 for ORF1a from Patient 14's sputum). Of note, Patient 1 infected 28 patients in another hospital but caused no confi rmed secondary cases in our hospital, whereas Patient 14 caused an additional 82 cases in our hospital. The diff erence of transmissibility between these two individuals could be caused by a combination of factors such as the time from onset of disease, clinical symptoms, duration of contact exposure, pattern of behaviour inside and near the emergency room, and kinetics of viral shedding. We showed that obtaining a travel history from patients is an important element of history taking by all physicians, and not only those specialising in infectious diseases or those working in infection control. Suspicion for unusual infections should be maintained if patients do not or cannot report accurate histories. Readiness of laboratory support is essential for initial investigation and for control of outbreaks, and overly rigorous requirements for laboratory testing have the potential to delay diagnosis and further spread disease. Hospital leadership needs to lead in preparedness for disaster management of high-risk communicable infectious diseases. Emergency preparedness at a national level and communication and support from government agencies are imperative to prevent and control any serious outbreak. The results of this study need to be interpreted with caution because the study was not suffi ciently powered to study risk factors for transmission. Some of our data were collected retrospectively. Analysis on visitors was limited because we did not have detailed data. Serological tests were not done simultaneously and attack rates were calculated on the basis of results from real-time RT-PCR of mainly symptomatic individuals. The potential transmission of MERS-CoV by asymptomatic carriers is under investigation. In conclusion, we report a large nosocomial MERS outbreak that occurred outside the Middle East. The potential for similar outbreaks anywhere in the world from a single traveller should be noted, as long as MERS-CoV transmission continues in the Middle East. Emergency preparedness and vigilance are crucial to the prevention of further large outbreaks in the future. Our report serves as an international alarm that preparedness in hospitals, laboratories, and governmental agencies is the key not only for MERS-CoV infections but also for other new emerging infectious diseases. SYC and J-MK designed the study and data collection methods, did the initial data analyses, drafted the manuscript, and approved the fi nal manuscript as submitted. YEH, who suspected and diagnosed the fi rst case of MERS-CoV infection in South Korea, engaged in the management of MERS-CoV outbreak control at the Samsung Medical Center as an infectious disease specialist and reviewed the manuscript. GEP, JYeL, J-HK, JYoL, JMK, JGR, and JRC coordinated data collection, engaged in the management of MERS-CoV outbreak control at the Samsung Medical Center, and reviewed the manuscript. SK supervised data collection and analysis, analysed the data, and reviewed the manuscript. HJH, C-SK, and E-SK did laboratory tests for MERS-CoV detection, coordinated laboratory data collection, and reviewed the manuscript. C-IK, IJJ, KRP, H-JD, and J-HS supervised data collection and reviewed the manuscript. Y-JK and DRC conceptualised and designed the study, and critically reviewed and revised the manuscript. All authors approved the fi nal submitted manuscript. We declare no competing interests. Isolation of a novel coronavirus from a man with pneumonia in Saudi Arabia Middle East respiratory syndrome coronavirus (MERS-CoV)-Saudi Arabia. Geneva: World Health Organization Middle East respiratory syndrome coronavirus: quantifi cation of the extent of the epidemic, surveillance biases, and transmissibility Interhuman transmissibility of Middle East respiratory syndrome coronavirus: estimation of pandemic risk Synthesizing data and models for the spread of MERS-CoV, 2013: key role of index cases and hospital transmission Hospital outbreak of Middle East respiratory syndrome coronavirus Superspreading and the eff ect of individual variation on disease emergence Transmission dynamics of the etiological agent of SARS in Hong Kong: impact of public health interventions Superspreading SARS events Middle East respiratory syndrome coronavirus outbreak in the Republic of Korea Guidelines for environmental infection control in health-care facilities. Recommendations of CDC and the Healthcare Infection Control Practices Advisory Committee (HICPAC) Middle East respiratory syndrome coronavirus (MERS) Evidence of airborne transmission of the severe acute respiratory syndrome virus Stability of Middle East respiratory syndrome coronavirus (MERS-CoV) under diff erent environmental conditions Severe acute respiratory syndrome-Singapore Severe acute respiratory syndrome Middle East respiratory syndrome in 3 persons, South Korea We express our sincere consolation for the patients and their families who had MERS-CoV infection. We greatly appreciate the eff orts of all the hospital employees and their families at the Samsung Medical Center, who worked tirelessly during this outbreak. We also appreciate the cooperation of all other hospitals in South Korea that worked together to overcome the nationwide outbreak. We acknowledge the consultation and support of the Korea Centers for Disease Control and Prevention, the MERS Rapid Response Team of the Public-Private Joint MERS Task Force, WHO, and the US Centers for Disease Control and Prevention. We also sincerely appreciate the discussion and critical feedback from Michael T Osterholm (University of Minnesota, Minneapolis, MN, USA) and Janet A Englund (Seattle Children's Hospital, Seattle, WA, USA).