Infection control and SARS transmission among healthcare workers, Taiwan

Infection control and SARS transmission among healthcare workers, Taiwan

Yee-Chun Chen

This study found infrequent transmission of severe acute respiratory syndrome (SARS) coronavirus to healthcare workers involved in the care of the first five case-patients in Taiwan, despite a substantial number of unprotected exposures. Nonetheless, given that SARS has been highly transmissible on some occasions, we still recommend strict precautions.


Healthcare workers may be unwittingly exposed to the severe acute respiratory syndrome-associated coronavirus (SARS-CoV) from patients with pneumonia at the onset of an epidemic (1,2). They are also at increased risk of acquiring SARS from known case-patients with a high viral load who require intensive respiratory care (1-3). The first case-patient in Taiwan was admitted to National Taiwan University Hospital on March 8, 2003, before the World Health Organization (WHO) issued the first global alert (4,5). The patient was intubated in the emergency room and was admitted to the intensive care unit. The second case-patient, his wife, was admitted to the emergency room with pneumonia on March 14. The occurrence of two cases of pneumonia in the same household within 6 days, together with the patients’ recent travel to Guangdong, China, through Hong Kong, led us to suspect a diagnosis of atypical pneumonia, which later came to be known as SARS.

Before the second case was detected, healthcare workers routinely used standard precautions. Specific infection-control measures, including droplet and contact precautions against SARS, were implemented after the second patient was admitted. The efficacy of these infection-control measures in protecting healthcare workers was determined by: 1) the occurrence of SARS symptoms as defined by WHO criteria (6) and 2) a rise in antibodies to SARS-CoV before and after specific infection-control measures were implemented.

The Study

From March 8 to March 28, the hospital admitted five patients in whom SARS-CoV infection was subsequently laboratory-confirmed. The patients were isolated in negative-pressure rooms. Patients 2 and 3 were family members of patient 1. Patients 4 and 5 were believed to have contracted SARS on a March 15 flight from Hong Kong to Beijing. Four of the patients progressed rapidly to respiratory failure and were intubated.

Healthcare workers caring for these patients were exposed during two periods. During March 8-14, before specific infection-control precautions were implemented, 73 healthcare workers were exposed to patients 1 and 2. During March 15-28, after specific precautions were implemented, an additional 150 healthcare workers were exposed to all five patients.

All healthcare workers who had contact with SARS patients used personal protective equipment, including gown, gloves, N95 respirators, disposable cap, and shoe covers. Healthcare workers exposed to SARS patients or their environments were monitored for signs or symptoms of SARS for 14 days after the last exposure. Healthcare workers who had high-risk exposures to SARS were excluded from new duty assignments. We considered performing any of the following to be a high-risk exposure: endotracheal intubation >30 min, cardiopulmonary resuscitation >30 min, pleurocentesis >30 min, or bedside care (such as chest care [including percussion and postual drainage] or feeding) >30 min. Any healthcare worker in whom fever developed (temperature [greater than or equal to] 38[degrees]C) was isolated in a specially designated ward.

A total of 223 healthcare workers exposed to SARS patients were interviewed by one of two researchers with a structured questionnaire designed by the Centers for Disease Control and Prevention (CDC), USA, and the Center for Disease Control, Taiwan. The following data were recorded on uniform case-report sheets: extent of personal protective equipment use during exposure, type of exposure (stay in the same room, direct patient contact, or exposure to respiratory droplets and secretions), disease phase of patients to whom they were exposed (during incubation period, early fever, fever and cough, or intubation period), occurrence of fever ([greater than or equal to] 38[degrees]C), and respiratory or gastrointestinal symptoms after exposure. Proportional data were tested by using [chi square] or Fisher exact test (EpiInfo 6, CDC, Atlanta, GA). A p value <0.05 was considered significant. The Ethics Committee of the hospital approved these studies.

Serum samples were collected twice from 206 healthcare workers during a 1-month period after the initial exposure to patients with SARS, with a minimum interval between collections of 2 weeks. Serologic response to SARS-CoV was determined by using an indirect immunofluorescence assay (IFA) as described previously (5) and the immunochromatographic test (ICT, Tyson BioResearch, Inc, Taiwan). ICT consists of a double-antigen (recombinant viral nucleocapsid antigen) sandwich. The test gives results within 15 min. Data obtained from 13 patients with severe SARS, as defined by using CDC criteria (7), showed that the sensitivity of the ICT test was >90% within 2 weeks of fever onset and 100% after 6 weeks. Data obtained from 51 cases of severe SARS demonstrated that the sensitivity of either IFA or ICT was 98% after 6 weeks (8). Furthermore, the specificity of each assay determined by 812 serum samples was 100%.

The Table compares the extent of personal protective equipment use before and after implementing specific infection-control measures. Healthcare workers during the “after” period were substantially more likely than the “before” period to have used full personal protective equipment (Table).

First serum samples were collected 12.4 [+ or -] 5.4 days (mean [+ or -] standard deviation) after initial exposure to SARS patients. Second serum samples were collected 37.2 [+ or -] 7.9 days after exposure. Ninety percent were collected [greater than or equal to] 30 days after exposure. None of the 73 healthcare workers exposed during the before period produced a positive result on serologic tests for SARS. This group included a physician who intubated patient 1 and wore two layers of surgical masks and used inline suction after intubation. SARS developed in 1 of 150 healthcare workers exposed during the after period. This healthcare worker was a chest physician. On March 17, he performed a 30-min chest sonogram on patient 2 in a negative-pressure isolation room and wore an N-95 respirator, double gloves, gown, disposable cap, and shoe covers. On the same day, he helped intubate patient 2 while positioned approximately 3 feet from the patient’s head. During this period, patient 2 was irritable and had a vigorous cough. The physician recalled that he had not tried the mask on or confirmed that it was air-tight before entering the isolation room. Fever developed 4 days later in this physician, designated as patient 6, and pneumonia developed 5 days after that. Both virus culture and reverse transcriptase-polymerase chain reaction (RT-PCR) demonstrated SARS-CoV in the sputum. Immunoglobulin (Ig) G against SARS-CoV determined by IFA was >1:1,000 (5). After this experience, the infection-control team reemphasized the importance of fit-testing facemasks and recommended wearing a face shield when in close contact with SARS patients. SARS did not develop in another physician who intubated patient 2 and four nurses who assisted the procedure in the same room.


In this study, a physician who intubated a patient with SARS while following standard precautions did not become ill, but SARS developed in another physician whose N95 respirator was not properly fit-tested. A serologic response to SARS-CoV could not be demonstrated in 205 healthcare workers who spent time in the same room as or had direct contact with SARS patients.

The major question that arises from this study is why 36 (50%) healthcare workers who stayed in the same room with SARS patients before the outbreak was recognized and who did not wear masks were not infected. Several possible explanations exist. Patient 2 wore a face mask when she visited the emergency room. The physician who intubated patient 1 was alert, wore two layers of surgical masks, and followed standard precautions. Inline suction was routinely performed at the hospital for intubated patients to prevent aerosol formation; therefore, unprotected healthcare workers might not have been exposed to a sufficient amount of SARS-CoV to produce a systemic infection. An alternate explanation could be that existing serologic assays are not sufficiently sensitive to identify subclinical infections. This explanation is unlikely, however, because the tests we used have been shown to be highly sensitive and specific in patients with SARS (5,8), and 90% of convalescent-phase serum samples were collected [greater than or equal to] 30 days after exposure. Yet another explanation could be that SARS-CoV is attenuated by serial passage in humans. This explanation is also unlikely since SARS developed in the five index patients admitted to the hospital in the early phase of the epidemic and in one physician with a poorly fitting mask. Further, phylogenetic tree analysis (9) indicates that patients 2, 3, and 6 were infected by strains related to the large outbreak in Amoy Gardens in Hong Kong (2), and patients 4 and 5 were infected by strains related to a large hospital outbreak in Taipei (10). A final explanation could be, simply, that the disease does not develop in all people exposed to the virus.

Transmission of SARS was limited initially at our hospital (attack rate 0.4%) when healthcare workers followed standard precautions or specific infection-control measures, including droplet and contact precautions. However, in later stages of the epidemic, SARS was more likely to develop in healthcare workers, despite similar or higher levels of personal protective equipment rise. Although one possible explanation for this could have been exposure to unrecognized SARS patients, contamination of the environment leading to indirect contact transmission may have also played a role (11).

In conclusion, while SARS-CoV can spread rapidly in a nonimmune human population (1-3), this study demonstrated infrequent transmission of SARS to healthcare workers caring for the first five SARS patients in Taiwan, despite a number of unprotected exposures. Nonetheless, given that SARS has, on other occasions, shown itself to be highly transmissible (1-3,10), we still recommend strict precautions (1-3,11-14).

Table. Personal protection before and after recognizing severe

acute respiratory syndrome (SARS) and implementing specific

infection-control measures at the National Taiwan University Hospital

Exposure type

In the same room (a)

Protective Before After

measures (n = 73) (n = 155) p value

Masks <0.001

None 36 0

Surgical mask, 37 155

N95 or P100


Gloves <0.001

None 57 7

One- or two-layer 16 148

Eye protection <0.001

None 73 117

Glasses, goggles, 0 38

or face shields

Gowns <0.001

None 66 6

One- or two-layer 7 149

Exposure type

Direct contact

Protective Before After

measures (n = 46) (n = 132) p value

Masks <0.001

None 20 0

Surgical mask, 26 132

N95 or P100


Gloves <0.001

None 28 4

One- or two-layer 18 128

Eye protection <0.001

None 46 99

Glasses, goggles, 0 33

or face shields

Gowns <0.001

None 38 6

One- or two-layer 8 126

Exposure type

Exposure to respiratory

droplets and secretions

Protective Before After

measures (n = 37) (n = 92) p value

Masks <0.001

None 17 0

Surgical mask, 20 92

N95 or P100


Gloves <0.001

None 17 2

One- or two-layer 20 90

Eye protection <0.001

None 37 66

Glasses, goggles, 0 26

or face shields

Gowns <0.001

None 30 3

One- or two-layer 7 89

(a) Five healthcare workers stayed in the same room with SARS

patients before and after implementation of specific

infection-control measures. Among 223 healthcare workers, 178

had direct contract to SARS patients or their environment, and

129 had exposure to respiratory droplets and secretions.


We are grateful to members of the infection-control team for their important contributions to this investigation and Calvin Kunin for his critical review of this manuscript.

This study was supported by a grant from the National Science Council, R.O.C. (NSC 92-3112-B-002-043).


(1.) Lee N, Hui D, Wu A, Chan P, Cameron P, Joynt GM, et al. A major outbreak of severe acute respiratory syndrome in Hong Kong. N Engl J Med 2003;348:1986-94.

(2.) Tomlinson B, Cockram C. SARS: experience at Prince of Wales Hospital, Hong Kong. Lancet 2003;361:1486-7.

(3.) Peiris JSM, Chu CM, Cheng VCC, Chan KS, Hung IFN, Poon LLM, et al. Clinical progression and viral load in a community outbreak of coronavirus-associated SARS pneumonia: a prospective study. Lancet 2003;361:1767-72.

(4.) Twu SJ, Chen TJ, Chen CJ, Olsen SJ, Lee LT, Fisk T, et al. Control measures for severe acute respiratory syndrome (SARS) in Taiwan. Emerg Infect Dis 2003;9:718-20.

(5.) Hsueh PR, Hsiao CH, Yeh SH, Wang WK, Chen PJ, Wang JT, et al. Microbiologic characteristics, serologic responses, and clinical manifestations in severe acute respiratory syndrome, Taiwan. Emerg Infect Dis 2003;9:1163-7.

(6.) World Health Organization. Case definitions for surveillance of severe acute respiratory syndrome (SARS) [monograph on the Internet]. 2003 May 1 [cited 2003 May 4]. Available from:

(7.) Centers for Disease Control and Prevention. Updated interim U.S. case definition for SARS [monograph on the Internet]. 2003 Jul 18 [cited 2003 Sep 1]. Available from:

(8.) Wang JT, Sheng WH, Fang CT, Chen YC, Wang JL, Yu CJ, et al. Clinical manifestations, laboratory findings, and treatment outcomes of SARS patients. Emerg Infect Dis 2004;10:818-24.

(9.) Yeh SH, Wang HI, Tsai CY, Kao CL, Yang JY, Liu HW, et al. Characterization of severe acute respiratory syndrome coronavirus genomes in Taiwan: molecular epidemiology and genome evolution. Proc Natl Acad Sci U S A 2004;101:2542-7.

(10.) Lee ML, Chen CJ, Su IJ, Chen KT, Yeh CC, King CC, et al. Severe acute respiratory syndrome–Taiwan, 2003. MMWR Morb Mortal Wkly Rep 2003;52:461-6.

(11.) Chen YC, Huang LM, Chan CC, Su CP, Chang SC, Chang YY, et al. SARS in hospital emergency room. Emerg Infect Dis 2004;10: 782-8.

(12.) Seto WH, Tsang D, Yung RWH, Ching TY, Ng TK, Ho M, et al. Effectiveness of precautions against droplets and contact in prevention of nosocomial transmission of severe acute respiratory syndrome (SARS). Lancet 2003;361:1519-20.

(13.) Chan PKS, Ip M, Ng KC, Chan RCW, Wu A, Lee N, et al. Severe acute respiratory syndrome-associated coronavirus infection. Emerg Infect Dis 2003;9:1453-4.

(14.) Centers for Disease Control and Prevention. Cluster of severe acute respiratory syndrome cases among protected health-care workers–Toronto, Canada, April 2003, MMWR Morb Mortal Wkly Rep 2003;52:433-6.

Dr. Chen is a physician in the departments of Internal Medicine, National Taiwan University Hospital and National Taiwan University College of Medicine. Her research interests include molecular epidemiology, pathogenesis, and in vitro susceptibility testing of medically important fungal pathogens. She is a member of the Infection Control Committee of National Taiwan University Hospital.

Address for correspondence: Pan-Chyr Yang, Department of Internal Medicine, National Taiwan University Hospital, No. 7, Chung-Shan South Road, Taipei, Taiwan 10016; fax: 886-2-2393-4176: email:

Yee-Chun Chen, * Pei-Jer Chen, * Shan-Chwen Chang, * Chiang-Lian Kao, * Shiou-Hwa Wang, * Li-Hua Wang, * Pan-Chyr Yang, * and the SARS Research Group of National Taiwan University College of Medicine and National Taiwan University Hospital (1)

* National Taiwan University Hospital, National Taiwan University College of Medicine, Taipei, Taiwan

(1) Ding-Shinn Chen, Yuan-Teh Lee, Che-Ming Teng, Pan-Chyr Yang, Hong-Nerng Ho, Pei-Jer Chen, Ming-Fu Chang, Jin-Town Wang, Shan-Chwen Chang, Chuan-Liang Kao, Wei-Kung Wang, Cheng-Hsiang Hsiao, and Po-Ren Hsueh.

COPYRIGHT 2004 U.S. National Center for Infectious Diseases

COPYRIGHT 2004 Gale Group