my personal edition > sars > news


  E-Mail this DGReview to a colleague

DGReview

Epidemiological and Airflow Simulation Data Suggest Airborne Transmission of SARS During 2003 Housing Complex Outbreak in Hong Kong

Combined analyses of epidemiological data and building airflow simulations within the Amoy Gardens housing complex suggest airborne transmission of the severe acute respiratory syndrome (SARS) virus during the 2003 outbreak in Hong Kong, China.

"The epidemic curve supports the hypothesis of a common source of the outbreak in Amoy Gardens, and the spatial distribution of the cases conformed to the hypothesis that virus-laden aerosols spread from a single source (the index apartment unit), as shown in our model made with the use of airflow-dynamics data," reports Ignatius T. S. Yu, MB, BS, MPH, Department of Community and Family Medicine, Chinese University of Hong Kong, and colleagues.

The investigators assessed the date of onset of symptoms; the location of the residences; associations between building, floor, and the direction each apartment faced; and the probability of infection using logistic regression for the initial 187 SARS cases. These combined data suggested a common source of the outbreak, since 97% of the patients lived in the 7 buildings identified as A through G. Additionally, the index patient and 99 of the other SARS patients lived in building E.

The investigators then used computational fluid dynamics analyses to predict detailed airflow patterns in the air shafts and around the buildings of the housing complex at the time of the SARS outbreak. Similar computational models were previously used to study the spread of foot-and-mouth disease and an anthrax outbreak, the authors note.

Faeces and urine with extremely high concentrations of SARS-associated coronavirus from the index patient produced virus-laden aerosols in the drainage pipes of building E, which returned to the bathroom of unit 7 through dried up seals of the floor-drain traps. The virus-laden aerosols then moved through and upward in the building air shafts due to the buoyancy of the warm, humid air in the air shaft and the negative pressure created by exhaust fans or the action of the winds blowing around the building. Apartment units on the middle and upper floors of building E had higher probabilities of infection than units on the lower floors - odds ratio of 5.15 on the middle floors and 3.1 on the upper floors.

Smaller concentrations of virus-laden aerosols then spread to adjacent buildings by wind flow after reaching the top of the building E airshaft. SARS cases occurred in buildings B, C, and D from 1 to 3 days after the first cases occurred in building E. "The concentrations of the aerosols decayed as the plume travelled away from the source, and the decay corresponded to lower attack rates (and, possibly, to a longer incubation period) in other apartment units of building E and in other buildings," the authors write.

"Our epidemiologic analysis, experimental studies, and air flow simulations support the probability of an airborne spread of the SARS virus in the outbreak in Amoy Gardens," the authors conclude.
N Engl J Med 2004;350:1731-9

E-Mail this DGReview to a colleague

To print, use this version