John Bartlett, MD


July 23, 2003

In This Article

Severe Acute Respiratory Syndrome (SARS)

Ruan YJ, Wei CL, Ee AL, et al. Comparative full-length genome sequence analysis of 14 SARS coronavirus isolates and common mutations associated with putative origins of infection. Lancet. 2003;361:1779-1785. Abstract These researchers from Singapore sequenced the entire SARS viral genome from the Singapore index case, 3 primary contacts, and 1 secondary contact. They then compared these data with 9 SARS-CoV genomes sequenced by others. The results showed 129 sequence variations in the 14 viral isolates, including 16 recurrent variant sequences. The common variant sequences defined 2 distinctive genotypes that linked infections (in Hanoi, Hong Kong, Singapore, and Toronto) that originated in Hotel M in Hong Kong, and a second with isolates unrelated to Hotel M that was found in cases from Beijing, Guangzhou, and Hong Kong. Other common sequence variance distinguished the geographic origin of isolates. The authors conclude that this analysis shows "genetic signatures" that can be used to trace the source of infection.

Comment: Coronaviruses are enveloped RNA viruses with a size of 27-30 kilobase (kb). They represent the largest RNA viruses. Characteristic features are a wide range of hosts (rodents, cats, pigs, cattle, birds, and people) and diverse diseases (respiratory infections, gastroenteritis, and hepatitis). In people, pathology has been limited to the common cold. A characteristic feature is a high rate of genetic mutation. The study above shows a completely pathogenic strain that does not appear to arise from simple recombination of existing strains. It appears to be well adapted to the human host with substantial genetic conservation. Since it is a new human pathogen, it is assumed that most humans are vulnerable in that they are immunologically naive. In an accompanying editorial, Brown and Tetro[1] note similarities with influenza. Since both viruses appear to be zoonoses, they have tropism for both respiratory and GI tissue, and they have mechanisms to generate great genetic variability.

Normile D. Infectious diseases. Up close and personal with SARS. Science. 2003;300:886-887. Abstract The authors review the experience of virologist Malik Peiris of the University of Hong Kong who, with his colleagues, was the first to isolate the SARS coronavirus. Dr. Peiris came to the Hong Kong University microbiology laboratory in 1995 and was a key scientist in the 1997 avian flu (H5N1) outbreak. With news of the pneumonia outbreak in Guangdong, their first impression was that this was H5N1, an impression that was reinforced when there were 2 cases with proven H5N1 acquired in China in February 2003. However, this theory was disproven by the inability to find H5N1 in SARS cases. Early in the week of 3/17/03, viral cultures were positive using a cell line generally used for hepatitis A, and this virus was reactive to convalescent sera from SARS patients. Electron microscopy (EM) showed particles that suggested coronavirus. Dr. Peiris missed the daily WHO SARS network phone conference to attend a government function, but mailed EMs of his isolates claiming a suspected coronavirus. Meanwhile, the Centers for Disease Control and Prevention (CDC) made a similar claim without reference to the Hong Kong laboratory observations. In terms of publications, the Hong Kong group's paper appeared in Lancet online 4/8/03, and reports from the other groups appeared in The New England Journal of Medicine 4/10/03.

Booth CM, Matukas LM, Tomlinson GA, et al. Clinical features and short-term outcomes of 144 patients with SARS in the Greater Toronto area. JAMA. 2003;289:2801-2809. Abstract Twenty-one authors summarized the clinical features of 144 adult patients admitted to 10 academic and community hospitals in Toronto for the period 3/7/03 to 4/10/03. The index case was a 78-year-old woman of Hong Kong descent who visited Hong Kong from 2/13-23 in "Hotel M," where there was a cluster of 13 cases. She returned to her home in Toronto, developed symptoms on 2/25, and died at home on 3/5. Several family members developed symptoms and one was hospitalized in a community hospital that became the "epicenter for the Toronto outbreak."

Of the 144 cases, 111 (77%) acquired the disease through hospital exposure, including 73 (51%) who were healthcare workers. The epicenter hospital accounted for 82 of the 111 nosocomial cases. Other sources were contact in the home in 35 (24%) and travel to epidemic areas in 3 (2%). The median incubation time from exposure to onset of symptoms was 6 days, with a range of 3-10 days. The initial symptoms were fever alone (23%), fever plus the characteristic prodrome (headache, malaise, or myalgia), cough or dyspnea (20%), or the prodrome alone (13%). The clinical features including symptoms, signs, and laboratory test results are summarized in Table 1 .

The authors conclude that the majority of patients acquired the disease through hospital exposure. Major risk factors for adverse outcome were comorbidities including diabetes; nevertheless, 93.5% of the patients survived.

Comment: In this series, as with the Hong Kong experience, there was extensive use of ribavirin (88%) and steroids (40%). This has been a controversial issue since ribavirin does not appear to have in vitro activity against the SARS coronavirus and there are no clinical trials. This report from Toronto also notes that ribavirin was associated with significant toxicity: 71 (49%) had evidence of hemolysis with a decrease in hemoglobin of at least 2 g/dL, and 40% had at least 1.5-fold increases in transaminase levels. An accompanying editorial by H. Masur and colleagues[2]) compares this series with 2 others reported from Hong Kong.[3,4]). The 3 series show that the frequency of diarrhea with presentation was 6% to 20%, the median age was about 45 years, the frequency in healthcare workers was 28% to 51%, the frequency of ICU admissions was 21% to 38%, and the frequency of mechanical ventilation was 14% to 28%. In all 3 series, risk factors for more severe disease and death were older age, diabetes, smoking, and other comorbid conditions.

Revisions to the Interim U.S. Case Definition for Severe Acute Respiratory Syndromes. CDC: Health Alert Network June 5, 2003. The CDC has now provided an updated definition of probable or suspected cases of SARS based on clinical, epidemiologic, and laboratory criteria ( Table 2 ). These are summarized as follows:

Clinical criteria

  • Asymptomatic or mild respiratory illness

  • Moderate: Temperature > 100.4° F plus respiratory illness with cough, dyspnea, or hypoxia

  • Severe: Temperature > 100.4° F plus respiratory illness (as above) plus radiographic evidence of pneumonia, acute respiratory distress syndrome (ARDS), or autopsy showing pneumonia or ARDS

Epidemiologic criteria

  • Travel (including transit in airport) within 10 days of onset to area with current or previously documented or suspected community transmission of SARS (Table 3)

  • Close contact within 10 days of onset of symptoms with person with known or suspected SARS


  • Confirmed: (1) Detection of antibody to SARS-CoV in specimens obtained during acute illness or > 21 days post illness onset; (2) detection of SARS-CoV RNA by RT-PCR confirmed by second PCR using a second aliquot of the specimen or a different set of primers; or (3) culture of SARS-CoV

  • Negative: Negative SARS-CoV in convalescent serum obtained > 21 days post onset of symptoms

  • Undetermined: Lab tests not done or not completed

Enserink M. Infectious diseases. Clues to the animal origins of SARS. Science. 2003;300:1351. Abstract This report from the Focus section of Science summarizes the May 23, 2003, announcement by Hong Kong microbiologist Yuen Kwok-Yung and colleagues on finding the SARS coronavirus in civet cats. The clue was that a disproportionate number of patients who had early cases were working in southern China's food industry, where civet cats are sold for food. The investigators obtained 25 animals representing 8 species from a market in Guangdong. Coronavirus was recovered from each of 6 masked palm civets. This virus was nearly identical to the human SARS agent except for the deletion of 29 nucleotides. This isolate matches closely with 18 of the 19 human coronavirus genomes posted on the Web to date.

Cooke FJ, Shapiro DS. Global outbreak of severe acute respiratory syndrome (SARS). Int J Infect Dis. 2003;7:80. Abstract The authors review the global outbreak of SARS, but the most interesting facet of the report concerns the epidemiology of transmission within Hotel M and secondarily in hospitals in 4 regions of the world. The initial cases were in Guangdong Province in South China. The index case in Hotel M, patient A, noted the onset of symptoms on 2/15/03. He traveled to Hong Kong to visit his family and stayed on the ninth floor of Hotel M on 2/21/03. He was hospitalized the next day in Hong Kong Hospital #2 and died 2/23/03. There were 12 other guests in Hotel M who then became cases in Hanoi, Iceland, Singapore, Toronto, and the United States, and 4 hospitals in Hong Kong. The epidemiology of several of these patients designated by letter are summarized in Table 4 , which indicates their location after leaving Hotel M and the epidemiology of SARS at their subsequent location after leaving Hotel M.

Comment: This is a remarkable epidemic with an index case and 12 hotel guests who acquired infection, including 8 who stayed on the ninth floor where the patient was. These 13 patients then became the source for at least 270 secondary cases in 8 widely disbursed geographic areas.

Holmes KV, Enjuanes L. Virology. The SARS coronavirus: a postgenomic era. Science. 2003;300:1377.Abstract This issue of Science includes 2 reports of the 30,000 nucleotide RNA genomes of 2 isolates of the SARS coronavirus.[5,6] The summary of these 2 papers emphasizes the following observations:

  • The genome has the features that characterize coronavirus, but is sufficiently different from prior coronaviruses to represent a new coronavirus group.

  • The 2 strains represent 1 from Toronto (SARS-CoV Tor2 ) and the Urbani strain from Vietnam; these differ by just 8 nucleotides, suggesting substantial stability with human passage.

  • The genome shows that this strain is not a "host-range mutant of a known coronavirus nor a recombinant between known coronaviruses."

The authors note that the strain is unlikely to be the product of genetic engineering.

Updated Interim Surveillance Case Definition for Severe Acute Respiratory Syndrome (SARS) - United States, April 29, 2003. MMWR Morb Mortal Wkly Rep. 2003;52:391-393. Abstract The following represents the updated interim US surveillance case definition of SARS:

Clinical criteria

  • Asymptomatic or mild respiratory illness

  • Moderate respiratory illness

  • -- temperature > 100.4° F (> 38° C), and

    -- 1 or more clinical findings of respiratory illness (eg, cough, shortness of breath, difficulty breathing, or hypoxia)

  • Severe respiratory illness

  • -- temperature > 100.4° F (> 38° C), and

    -- 1 or more clinical findings of respiratory illness (eg, cough, shortness of breath, difficulty breathing, or hypoxia), and

    -- radiographic evidence of pneumonia, or

    -- respiratory distress syndrome, or

    -- autopsy findings consistent with pneumonia or respiratory distress syndrome without an identifiable cause

Epidemiologic criteria

  • Travel (including transit in an airport) within 10 days of onset of symptoms to an area with current or recently documented or suspected community transmission of SARS (China, Hanoi, Hong Kong, Singapore, Taiwan, Toronto)

  • Close contact (cared for or lived with) within 10 days of onset of symptoms with a person known or suspected to have SARS infection

  • Laboratory criteria (enzyme immunoassay (EIA), indirect fluorescent antibody immunofluorescent assay (IFA), or reverse transcription polymerase chain reaction (RT-PCR)

  • Confirmed

  • -- detection of antibody to SARS-coronavirus (CoV) in specimens obtained during acute illness or > 21 days after illness onset, or

    -- detection of SARS-CoV RNA by RT-PCR confirmed by a second PCR assay, by using a second aliquot of the specimen and a different set of PCR primers, or

    -- isolation of SARS-CoV

  • Negative: absence of antibody to SARS-CoV in convalescent serum obtained > 21 days after symptom onset

  • Undetermined: laboratory testing either not performed or incomplete

Case classification

  • Probable case: meets the clinical criteria for severe respiratory illness of unknown etiology with onset since February 1, 2003, and epidemiologic criteria; laboratory criteria confirmed, negative, or undetermined

  • Suspect case: meets the clinical criteria for moderate respiratory illness of unknown etiology with onset since February 1, 2003, and epidemiologic criteria; laboratory criteria confirmed, negative, or undetermined

Dealing with SARS in a hospital. Hosp Infect Control. 2003;30:77. The hospital has been the major source for SARS in both patients and healthcare workers and the following summarizes the major points made from an audio conference "SARS: What US Hospitals Must Learn From the Canadian Outbreak,"[7] which can be purchased by calling 1-800-688-2421.

  • Case definition: The CDC case definition has changed 3 times; the most recent version is available at

  • Isolation vs quarantine: These must be distinguished. Isolation refers to sick people who are separated from others. Quarantine applies to healthy persons who are believed to have been exposed to a communicable disease and have a requested restriction in interactions during the incubation period. (During the Toronto outbreak, at least 10,000 persons were quarantined, and in China, the total exceeds 30,000).

  • Triage: The recommendation is to post signs in emergency departments and clinics asking persons with epidemiologic risks (travel to designated areas or close contact with cases) combined with fever, cough, or dyspnea to: (1) wear a surgical mask; (2) be evaluated in a room with negative pressure or high-efficiency particulate air (HEPA) filtration; and (3) use contact precautions. If negative pressure rooms and HEPA filtration are not available, there should be a private room for evaluation. Healthcare workers should use hand hygiene, follow CDC guidelines for tuberculosis with aerosol-generating procedures, use contact procedures, and use N-95 masks according to recommendations below.

  • N-95 masks: There should be training and fit-testing to ensure adequate sealing (see /etools/respiratory). SARS is unlike tuberculosis in that exposure is by contact with infected secretions as well as by airborne route. Once worn in the presence of SARS, the outside should not be touched, and the device should be discarded followed by hand hygiene. Reuse may be considered if the device has not been obviously soiled or damaged. To increase safety with reuse, the following are recommended: consider wearing a surgical mask or face shield over the N95 respirator, consider labeling respirators for the user, and use hand hygiene after placing the respirator on the face. The recommended respiratory protective devices are particle filter efficiency of 95% (N-95) or greater (N-99 or N-100). An alternative is powered air-purifying respirator (PAPR). If none of these are available, a surgical mask should be worn since this will improve barrier protection with large droplets, but the disadvantage is substantial leakage.

  • Contact isolation: Contact precautions include gloves, gown, and eye protection for all patient/environment contact.

  • Visitors: Screen visitors who are suspected SARS patients themselves and educate visitors about avoiding public places if they have fever or a respiratory illness. Specifically, they should call the hospital before visiting to implement this screening.

  • Management of exposed healthcare workers who were unprotected: Monitor for signs and symptoms of SARS for 10 days postexposure. The recommendation is not to furlough an employee in the absence of symptoms. With symptoms, they should contact the healthcare facility by telephone, avoid interactions outside the home, they should not work, and they should implement infection-control precautions including mask and hand hygiene. If symptoms do not progress in 72 hours, they may return for evaluation by employee health or infection control and then work if clear. Workers with suspected SARS should not return to work until 10 days after being afebrile and asymptomatic and cleared by infection control or employee health.