Prevalence of non-pneumonic infections with SARS-correlated virus

Abstract

Patrick Woo and colleagues1Woo PC Lau SK Tsoi HW et al.Relative rates of non-pneumonic SARS coronavirus infection and SARS coronavirus pneumonia.Lancet. 2004; 363: 841-845Summary Full Text Full Text PDF PubMed Scopus (115) Google Scholar provide evidence of subclinical or non-pneumonic infections with the virus thought to cause SARS and propose that non-pneumonic infections are more common than SARS virus pneumonia in Hong Kong. However, I worry that the method used by Woo and colleagues could have led to a misleading concept. Although the authors claim that they used western blot to confirm the specificity of antibodies against SARS-correlated virus, the method they used, which was based on a recombinant nucleocapsid protein and a recombinant spike polypeptide blotted onto membranes to probe corresponding antibodies, is actually a dot-blot ELISA.Western blot only shows high specificity in detecting antigens when used with known specific antibodies. False-positive reactions in ELISA can be caused by the binding of low-affinity polyreactive antibodies. Therefore, I am not convinced that the antibodies detected in the four individuals without SARS in Woo and colleagues' report were specific to the SARS virus. The specificity of such antibodies in non-SARS cases needs to be confirmed by a viral neutralisation assay. Indeed, nine other samples that were weakly positive for antibodies against SARS virus by a dot-blot ELISA were all negative by a viral neutralisation test.2Chow PKH Ooi E-E Tan H-K et al.Healthcare worker seroconversion in SARS outbreak.Emerg Infect Dis. 2004; 10: 249-250Crossref PubMed Scopus (23) Google ScholarSeveral reports have documented subclinical and mild infections with SARS virus in individuals who had been in close contact with patients with SARS. Li and colleagues3Li G Chen X Xu A Profile of specific antibodies to the SARS-associated coronavirus.N Engl J Med. 2003; 349: 508-509Crossref PubMed Scopus (236) Google Scholar, 4Li G Zhao Z Chen L Zhou Y Mild severe acute respiratory syndrome.Emerg Infect Dis. 2003; 9: 1182-1183Crossref PubMed Scopus (21) Google Scholar reported that of 125 people exposed to SARS patients, 20 developed SARS with seroconversion of antibodies against SARS virus, 103 were negative for such antibodies, and only two had non-pneumonic infections confirmed by detection of the specific antibodies. Ho and colleagues5Ho KY Singh KS Habib AG et al.Mild illness associated with severe acute respiratory syndrome coronavirus infection: lessons from a prospective seroepidemiologic study of health-care workers in a teaching hospital in Singapore.J Infect Dis. 2004; 189: 642-647Crossref PubMed Scopus (49) Google Scholar also found that SARS-virus pneumonia was more common than non-pneumonic infections in people who had histories of contact with patients with SARS. The fact that the SARS outbreaks occurred in spring 2003 and were contained by mid-July after each of the cases of probable or suspect SARS was quarantined or sequestered also suggests that subclinical infections with SARS virus in the general population are rare.Subclinical or non-pneumonic SARS-virus infections do exist; however, such infections are less common than apparent infections after exposure to SARS. Patrick Woo and colleagues1Woo PC Lau SK Tsoi HW et al.Relative rates of non-pneumonic SARS coronavirus infection and SARS coronavirus pneumonia.Lancet. 2004; 363: 841-845Summary Full Text Full Text PDF PubMed Scopus (115) Google Scholar provide evidence of subclinical or non-pneumonic infections with the virus thought to cause SARS and propose that non-pneumonic infections are more common than SARS virus pneumonia in Hong Kong. However, I worry that the method used by Woo and colleagues could have led to a misleading concept. Although the authors claim that they used western blot to confirm the specificity of antibodies against SARS-correlated virus, the method they used, which was based on a recombinant nucleocapsid protein and a recombinant spike polypeptide blotted onto membranes to probe corresponding antibodies, is actually a dot-blot ELISA. Western blot only shows high specificity in detecting antigens when used with known specific antibodies. False-positive reactions in ELISA can be caused by the binding of low-affinity polyreactive antibodies. Therefore, I am not convinced that the antibodies detected in the four individuals without SARS in Woo and colleagues' report were specific to the SARS virus. The specificity of such antibodies in non-SARS cases needs to be confirmed by a viral neutralisation assay. Indeed, nine other samples that were weakly positive for antibodies against SARS virus by a dot-blot ELISA were all negative by a viral neutralisation test.2Chow PKH Ooi E-E Tan H-K et al.Healthcare worker seroconversion in SARS outbreak.Emerg Infect Dis. 2004; 10: 249-250Crossref PubMed Scopus (23) Google Scholar Several reports have documented subclinical and mild infections with SARS virus in individuals who had been in close contact with patients with SARS. Li and colleagues3Li G Chen X Xu A Profile of specific antibodies to the SARS-associated coronavirus.N Engl J Med. 2003; 349: 508-509Crossref PubMed Scopus (236) Google Scholar, 4Li G Zhao Z Chen L Zhou Y Mild severe acute respiratory syndrome.Emerg Infect Dis. 2003; 9: 1182-1183Crossref PubMed Scopus (21) Google Scholar reported that of 125 people exposed to SARS patients, 20 developed SARS with seroconversion of antibodies against SARS virus, 103 were negative for such antibodies, and only two had non-pneumonic infections confirmed by detection of the specific antibodies. Ho and colleagues5Ho KY Singh KS Habib AG et al.Mild illness associated with severe acute respiratory syndrome coronavirus infection: lessons from a prospective seroepidemiologic study of health-care workers in a teaching hospital in Singapore.J Infect Dis. 2004; 189: 642-647Crossref PubMed Scopus (49) Google Scholar also found that SARS-virus pneumonia was more common than non-pneumonic infections in people who had histories of contact with patients with SARS. The fact that the SARS outbreaks occurred in spring 2003 and were contained by mid-July after each of the cases of probable or suspect SARS was quarantined or sequestered also suggests that subclinical infections with SARS virus in the general population are rare. Subclinical or non-pneumonic SARS-virus infections do exist; however, such infections are less common than apparent infections after exposure to SARS. Prevalence of non-pneumonic infections with SARS-correlated virusAuthors' reply Full-Text PDF