Definitions

A case was defined as laboratory-confirmed MERS-CoV infection in a person as determined by a PCR test using 2 gene targets (14). Cases were further classified into epidemiologic categories. A household case was illness in a person who spent >1 night or 8 continuous hours in the same home with a case-patient while that person was infected. A healthcare-associated case (HCA) was illness in a case-patient who had been exposed to another case-patient exclusively in a healthcare setting in the 14 days before their own onset of symptoms or specimen collection date. A work (or other) setting–related case was illness in a person who was exposed to a case-patient but not in a healthcare or household setting. An epidemiologically unlinked case was defined as illness in a person without any documented link to a patient with suspected or confirmed MERS-CoV infection. Contacts were persons who provided care for the patient, including healthcare personnel or family members; had close physical contact with the patient; or stayed in the same place with a laboratory-confirmed.

A household cluster was defined as >2 cases in the same household. An HCA cluster was >2 HCA cases in the same healthcare setting or healthcare interaction (e.g., hospital transport). A work cluster was >2 cases in the same workplace. To be considered part of the same cluster, secondary case-patients must have had a positive laboratory test result and share an epidemiologic link, such as a workplace, hospital room, or household. Suspected case-patients who had a negative laboratory test result for MERS-CoV were classified as test-negative suspected case-patients.

Asymptomatic case-patients were those who had no reported symptoms at the time of a positive test recorded by a healthcare provider in the medical chart. Mildly symptomatic case-patients reported symptoms, such as sore throat, rhinorrhea, or cough, and did not require oxygen during their hospital stay. Severely symptomatic case-patients required supplemental oxygenation during their hospitalization, ranging from nasal cannula to intubation.

Surveillance System

HAAD began surveillance for MERS-CoV in January 2013. Three surveillance systems were used to identify case-patients who tested positive for the virus: 1) Infectious Disease Electronic Notification System, in which physicians identified suspected case-patients and completed a basic form on the basis of patients’ clinical features; 2) Sheikh Khalifa Medical Center Laboratory surveillance, at the laboratory responsible for processing MERS-CoV specimens, which also records patient demographic characteristics, dates of collection, and results; and 3) Operations Center, which runs an active surveillance system that contacts 42 public and private hospitals in Abu Dhabi and records information regarding patients that were admitted or transferred to the intensive care unit because of primary respiratory failure or acute respiratory distress syndrome.

Because suspected case-patients may be admitted for primary respiratory failure or acute respiratory distress syndrome from multiple etiologies, HAAD authorities conducted investigations to determine whether an identified patient should be tested for MERS-CoV, if the patient had not already been captured by 1 of the surveillance systems. In addition, during April and May 2014, screenings for MERS-CoV were conducted, most under the supervision of HAAD, in various locations where MERS-CoV cases had been identified, including, but not limited to, hospitals and workplace settings.

For each case-patient, HAAD conducted detailed contact investigations within 24 hours of notification. HAAD officials interviewed case-patients if available. If the case-patient was unable to be interviewed, the patient’s epidemiologic information was collected from family members or other proxies. Staff at hospitals or local Disease and Prevention Screening Centers tested a sputum (preferred) or nasopharyngeal sample from all case-patient contacts for MERS-CoV. Samples from contacts were tested regardless of whether the person experienced symptoms. For each test-positive case-patient identified, we collected clinical information using the International Severe Acute Respiratory and Emerging Infection Consortium form, which was filled out in real time by healthcare providers and later verified by retrospective chart review.

All persons who tested positive for MERS-CoV, including asymptomatic persons, were admitted to a hospital. Before they could be discharged, confirmed case-patients were required to show negative results on 2 MERS-CoV PCR tests conducted at least 48 hours apart.

Laboratory Diagnostic Testing

The laboratory analyzed upper (e.g., nasopharyngeal, oropharyngeal) and lower respiratory tract samples (e.g., sputum, bronchoalveolar lavage fluid, tracheal aspirates) and serum. If hospital staff were not able to collect sputum spontaneously, the clinician could order that sputum be induced in a negative pressure room or that nasopharyngeal aspirates be obtained. Specimens were tested by using real-time reverse transcription PCR (rRT-PCR), upE, and open reading frame 1 assays in the Sheikh Khalifa Medical Center laboratory (15,16). A laboratory team from the US Centers for Disease Control and Prevention (CDC; Atlanta, GA, USA) used a nucleocapsid-based rRT-PCR assay to verify infection in a random sample of 23 specimens from 2014 (17). To calculate the length of positivity for each MERS-CoV case-patient, we calculated the difference between the date of the first positive test and the date the virus was last detected.

Data Management and Analysis

We merged results from Sheikh Khalifa Medical Center laboratory and HAAD’s epidemiologic databases, which contained information on patients’ demographic characteristics, symptoms, and exposure, using a combination of unique individual identifiers. We also retrospectively reviewed medical charts to collect additional clinical information about case-patients with positive test results.

Because changes occurred in the surveillance system during the study period, we compared demographic characteristics of MERS-CoV case-patients and those of suspected case-patients who were tested and had negative results during January 1, 2013–April 17, 2014. Suspected case-patients with negative test results identified during April 18, 2014–May 9, 2014, were not available for comparison. We included all MERS-CoV case-patients identified during January 1, 2013–May 9, 2014, in analyses of demographic characteristics, clusters, disease severity, sample type, and PCR positivity. Differences in proportions were contrasted by using Mantel-Haenszel χ² test. We defined statistical significance as a p value <0.05. Data were analyzed by using SAS version 9.3 (SAS Institute, Cary, NC, USA).

Ethics

Because these data were collected as part of a public health response, HAAD and CDC determined that data were nonresearch and not subject to review by an institutional review board. Secondary data analysis was then carried out for operational purposes.

Comparison of Laboratory-Confirmed Case-Patients and Test-Negative Suspected Case-Patients

For January 1, 2013–April 17, 2014, HAAD surveillance systems contained records for 1,586 unique persons, including 41 (3%) with confirmed MERS-CoV infection, 1,467 (92%) suspected case-patients with negative test results, and 78 (5%) whose test results were unknown or no test had been performed. Most case-patients were male (61%), Asian (54%), and 20–59 years of age (76%) (Technical Appendix Table 1). Case-patients more frequently reported exposure to a known or suspected MERS-CoV case-patient (30, 73%) and exposure to an animal (10, 24%) within the previous 2 weeks of illness onset or specimen collection than did suspected case-patients with negative test results. Fifty-one percent of MERS-CoV cases were healthcare associated.

Emiratis and nationals from other Gulf Cooperation Council countries (i.e., Saudi Arabia, United Arab Emirates, Qatar, Bahrain, and Oman) were the groups most often tested (855, 57%), whereas Asians had the highest proportion of positive test results (22/187, 12%). Of all reported and evaluated case-patients, 15% (6/41) had traveled internationally (Technical Appendix Table 1).

All MERS-CoV Cases

During April 17–May 9, 2014, surveillance activities identified an additional 24 MERS-CoV cases, for a total of 65 cases during the January 1, 2013–April 17, 2014, study period. We retrospectively reviewed charts of 64/65 (98%) case-patients; 1 case-patient was transferred to a hospital outside Abu Dhabi, and medical records were unavailable.

Figure

Figure. Epidemiologic curve showing confirmed cases of Middle East respiratory syndrome coronavirus (MERS-CoV) infection, Abu Dhabi, United Arab Emirates, January 1, 2013–May 9, 2014 (N = 65). Most cases were reported during...

The first case-patient with confirmed MERS-CoV infection in Abu Dhabi was identified in March 2013, and the number of case-patients identified (n = 41) peaked in April 2014 (Figure). During April 2014, the highest number of tests was performed, with 24 positive results, 323 negative results, and no test result available for 64 tests. Through contact tracing, HAAD identified 2,372 contacts for 56 case-patients, an average of 56 contacts (range 2–199 contacts) per positive case-patient.

Most case-patients were male (43/65, 66%) and 20 to 59 years of age (51/65, 78%) (Technical Appendix Table 2). All but 3 expatriate case-patients had been in the UAE for >1 month before diagnosis. However, those 3 had traveled to or lived in the Arabian Peninsula in the month before diagnosis. Forty-two percent of cases were healthcare associated, 34% were epidemiologically unlinked, 19% were household associated, and 6% were related to work or another setting. Eight (12%) cases resulted in death.

Cluster Identification

We identified 6 clusters. The number of MERS-CoV cases detected by surveillance in 2014 was greater than the number detected in 2013 because of a large cluster that involved both healthcare and household settings in the Eastern Region of UAE during April 2014 (Figure) (18). This resulted in 28 epidemiologically-linked MERS-CoV cases. All other clusters in Abu Dhabi—2 healthcare-associated clusters, 1 household-associated cluster, and 2 work-related clusters—involved <5 persons.

Epidemiologically Unlinked Cases

We identified 22 (34%) epidemiologically unlinked cases in Abu Dhabi (Technical Appendix Table 2); 86% were diagnosed in 2014, and the same percentage occurred in men. Most (59%) epidemiologically unlinked case-patients were 20–59 years of age; median age was 45 years (interquartile range [IQR] 36–66 years). Eight (36%) had traveled outside UAE within 14 days of initial symptoms (3 to Saudi Arabia, 3 to Oman, 1 to Bahrain, and 1 to Thailand). Six (27%) case-patients had documented contact with camels within 14 days before symptom onset or, if asymptomatic, before the date of specimen collection. Eight (36%) case-patients had visited a healthcare setting, such as a dialysis clinic, health clinic, or emergency department, before onset of illness, but these cases were not connected with a known healthcare-associated cluster.

Disease Severity

A higher proportion of case-patients were male in all 3 disease severity categories (Table 1). The gender disparity appeared to be lower among mildly symptomatic case patients; it was not significantly different for asymptomatic and for severely symptomatic case-patients (p = 0.25; Table 1). Patients who had mild or no symptoms were of similar ages, whereas those patients with severe symptoms were older (median age 37, 42, and 60 years, respectively) (Table 1). Of the 31 healthcare-personnel case-patients, 12 (39%) were asymptomatic. Among case-patients with mild and severe symptoms, 46% and 73%, respectively, had symptoms 1 to 7 days before being hospitalized (Table 1). Duration of patients’ PCR-positivity lengthened as disease severity increased (p = 0.03; Table 2). Of 15 case-patients who had positive test results for >2 weeks, 1 (7%) was asymptomatic.

Sample Type and Positivity

Samples from case-patients were tested, at physician’s discretion, while patients were in isolation. Case-patients with severe symptoms had a median of 4 (IQR 2–8, total 103) MERS-CoV rRT-PCR tests, patients with mild symptoms a median of 7 (IQR 3–14, total 214) tests, and patients with no reported symptoms a median of 5 (IQR 4–8, total 147) tests. Most (249/297, 84%) positive results were from lower respiratory tract specimens. Across all disease severity categories, 34% (85/249) of lower respiratory tract samples were positive >14 days after initial positive test for MERS-CoV, whereas 13% (6/48) of upper respiratory tract samples were positive >14 days (p<0.01) (Table 2). Specimens from 6% of case-patients (4/65) remained positive for >3 weeks. These positive specimens included those from 1 case-patient with severe symptoms and from 1 case-patient with mild symptoms whose samples remained positive for 23 days; 1 case-patient with mild symptoms whose samples remained positive for 27 days; and 1 case-patient with mild symptoms whose samples remained positive for 28 days.

Overall, upper respiratory samples were positive less frequently than lower respiratory tract samples (Table 2). In addition, upper respiratory samples were positive less frequently as disease severity lessened (Table 2). Lower respiratory tract samples, however, had a longer duration of PCR positivity than upper respiratory tract samples. For case-patients with severe and mild symptoms, a higher percentage of lower respiratory samples than upper respiratory samples was positive >14 days (for patients with severe cases, 39% vs. 8%; for patients with mild symptoms, 37% vs. 1%) (Table 2).