Viral infections in workers in hospital and research laboratory settings: a comparative review of infection modes and respective biosafety aspects.

OBJECTIVES To compare modes and sources of infection and clinical and biosafety aspects of accidental viral infections in hospital workers and research laboratory staff reported in scientific articles. METHODS PubMed, Google Scholar, ISI Web of Knowledge, Scirus, and Scielo were searched (to December 2008) for reports of accidental viral infections, written in English, Portuguese, Spanish, or German; the authors' personal file of scientific articles and references from the articles retrieved in the initial search were also used. Systematic review was carried out with inclusion criteria of presence of accidental viral infection's cases information, and exclusion criteria of absence of information about the viral etiology, and at least probable mode of infection. RESULTS One hundred and forty-one scientific articles were obtained, 66 of which were included in the analysis. For arboviruses, 84% of the laboratory infections had aerosol as the source; for alphaviruses alone, aerosol exposure accounted for 94% of accidental infections. Of laboratory arboviral infections, 15.7% were acquired percutaneously, whereas 41.6% of hospital infections were percutaneous. For airborne viruses, 81% of the infections occurred in laboratories, with hantavirus the leading causative agent. Aerosol inhalation was implicated in 96% of lymphocytic choriomeningitis virus infections, 99% of hantavirus infections, and 50% of coxsackievirus infections, but infective droplet inhalation was the leading mode of infection for severe acute respiratory syndrome coronavirus and the mucocutaneous mode of infection was involved in the case of infection with influenza B. For blood-borne viruses, 92% of infections occurred in hospitals and 93% of these had percutaneous mode of infection, while among laboratory infections 77% were due to infective aerosol inhalation. Among blood-borne virus infections there were six cases of particular note: three cases of acute hepatitis following hepatitis C virus infection with a short period of incubation, one laboratory case of human immunodeficiency virus infection through aerosol inhalation, one case of hepatitis following hepatitis G virus infection, and one case of fulminant hepatitis with hepatitis B virus infection following exposure of the worker's conjunctiva to hepatitis B virus e antigen-negative patient saliva. Of the 12 infections with viruses with preferential mucocutaneous transmission, seven occurred percutaneously, aerosol was implicated as a possible source of infection in two cases, and one atypical infection with Macacine herpesvirus 1 with fatal encephalitis as the outcome occurred through a louse bite. One outbreak of norovirus infection among hospital staff had as its probable mode of infection the ingestion of inocula spread in the environment by fomites. CONCLUSIONS The currently accepted and practiced risk analysis of accidental viral infections based on the conventional dynamics of infection of the etiological agents is insufficient to cope with accidental viral infections in laboratories and to a lesser extent in hospitals, where unconventional modes of infection are less frequently present but still have relevant clinical and potential epidemiological consequences. Unconventional modes of infection, atypical clinical development, or extremely severe cases are frequently present together with high viral loads and high virulence of the agents manipulated in laboratories. In hospitals by contrast, the only possible association of atypical cases is with the individual resistance of the worker. Current standard precaution practices are insufficient to prevent most of the unconventional infections in hospitals analyzed in this study; it is recommended that special attention be given to flaviviruses in these settings.