© 2004 American Thoracic Society
Transmission of Severe Acute Respiratory Syndrome in Critical CareDo We Need a Change?University of Geneva Hospitals Geneva, Switzerland The 2003 Severe Acute Respiratory Syndrome (SARS) outbreak left us with over 8,000 infected individuals and 700 deaths. Hospitals acted as an amplifier of the outbreak, and healthcare workers paid a heavy toll as they represented about 20% of cases. The article by Fowler and coworkers (pp. 1198–1202) provides additional data to document the ease of transmission of the virus in critical care, and tackles the difficult issue of the risk carried by aerosol-generating procedures (1). Let us briefly recap on the modes of transmission of SARS. On average, each case transmitted the disease to 2 to 4 other individuals. This basic reproduction number is consistent with transmission essentially by contact and droplets. The transmissibility of the disease, however, is very heterogeneous, as many patients did not transmit the virus at all, whereas a few transmitted the virus to tens of individuals. Risk factors for these rare "superspreading events" are incompletely elucidated, but probably include aerosol transmission, as suspected in several clusters (2–5). Fowler and coworkers performed a retrospective cohort study to investigate the risk for healthcare workers to be infected when exposed to several procedures: endotracheal intubation, non-invasive positive-pressure ventilation, and high frequency oscillatory ventilation (1). During the outbreak period, 7 SARS patients were cared for in the ICU; 122 critical care staff were considered at risk, and 10 (8.2%) became infected, a figure similar to previous reports (2, 6). Attack rates varied across professional categories (7.6% among nurses, 11.1% among respiratory therapists, and 20.0% among physicians) but the small sample size precludes further interpretation of this finding. The risk of being infected among physicians and nurses who performed or assisted in endotracheal intubation was about 13 times higher than among those who did not. The small sample size did not allow multivariate analysis to account for confounding factors, but the magnitude of the association is rather convincing. Why did such a risk arise in this ICU despite the fact that patients were recognized as being infected with the SARS-associated coronavirus, were cared for in negative-pressure rooms, and the staff were using personal protection equipment? First, patients were severely ill, coinciding with high viral load and maximum infectiousness (7). Second, nurses who became ill were often exposed to SARS patients within 48 hours of their admission. Many factors surround the admission of a patient to the ICU: deterioration of clinical status with symptoms increasing the spread of droplets or aerosols (dyspnea, cough), close and prolonged contact with the patient, institution of ventilator support, high number of potentially infecting exposures, and all this in the context of a life-threatening emergency and, surely, stress resulting from caring for a SARS patient. All of the preceding factors increase the likelihood of breaches in infection control measures. Finally, it is reported that the use of goggles or face shield was variable, but, unfortunately, more precise data on the use of personal protective equipment are not provided (1). This issue, however, must be crucial. Does personal protective equipment work? Yes. Lau and coworkers compared 72 healthcare workers infected with SARS with 143 matched control subjects to investigate risk factors for infection (8). Inconsistent use of personal protective equipment was much more frequent among the cases, in particular the use of goggles, gown, gloves, and cap. Interestingly, there was no such association regarding the use of a mask (either surgical or N95). By contrast, seven healthcare workers became infected after having cared for a critically ill patient who was only later recognized as having SARS; none wore personal protective equipment as recommended for SARS (2). These findings (8) and those from other studies (9, 10) demonstrate the effectiveness of personal protection equipment and may, at least partly, explain the high infection risk described by Fowler and coworkers (1). To assess whether some procedures carry a higher risk of SARS transmission is surely important, but can absolutely not be dissociated from the use of personal protective equipment. Fowler and coworkers found an increased risk associated with endotracheal intubation, a (nonsignificant) increased risk associated with noninvasive positive-pressure ventilation, and no association with high frequency oscillatory ventilation (1). Caution, however, should be exercised in the interpretation of the latter two findings given the small sample size. Intubation, suctioning before intubation, and manipulating an oxygen mask were risk factors for infection in the study by Loeb and coworkers (10). In the latter study, however, none of the eight infected healthcare workers consistently used the appropriate personal protective equipment. In sharp contrast is the study by Lau and coworkers, which revealed no association between infection risk and high-risk procedures (8). There is no doubt that some procedures increase the risk for healthcare workers, but how this fact should be translated into infection control policies is the real question. Restricting the number of individuals involved during the procedure, requiring that the procedure be performed by the most qualified, avoiding some procedures when a safer and adequate alternative exists, and so on, appear to be reasonable solutions (1, 11). System changes (for example, ensure adequate staffing, especially around the patient's transfer to the ICU), or process changes (such as establishing the role of all involved staff) might contribute to risk reduction (5). Enhancing personal protection during high-risk procedures, such as recommending the use of a powered air-purifying ventilator, is problematic. First, the benefit in risk reduction is not demonstrated. Second, this equipment is tedious to wear and may require the help of an assistant, the process might be too long and incompatible with emergency care, and, finally, the removal of the device may put the healthcare worker at risk. Adequate use of standard personal protective equipment seems the cornerstone issue. To illustrate this again, feeling air entering around the sides of the mask (12) could be avoided by a simple and hardly ever performed action: mask fit-testing (11, 13). To achieve that, SARS preparedness should lay stress on some basic elements that in no way differ from regular infection control activities, such as infection control training (8). FOOTNOTES Conflict of Interest Statement: S.H. and D.P. do not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. REFERENCES
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
