protein is neutralizing, and antibody to the fusion protein also prevents cell-to-cell spread of the virus (Merz et al., 1980). The fusion proteins of measles and mumps viruses, closely related to paramyxoviruses, are antigenically inactivated by formalin, and it is possible that early measles and parainfluenza vaccines failed for this reason.
Reinfections are common with PIV-1, PIV-2, and PIV-3. They appear to be most frequent with PIV-3, another feature of this virus that resembles RSV (Chanock et al., 1963). There is also evidence from volunteer studies in adults that secretory neutralizing antibody correlates better than serum antibody with protection against challenge with PIV-1 (Smith et al., 1966). It is assumed that this rule also holds for PIV-2 and PIV-3, although this has not been proved, and information in children is scanty.
Although there is some cross-reactivity between these three PIV types, cross-protection probably does not occur. There does not appear to be any serologic variation within each type.
Descriptions and estimates given below are based on reports by Glezen et al. (1982, 1983), and the National Institute of Allergy and Infectious Diseases (in press), and on personal communications from Clyde (1983) and Glezen (1984).
Parainfluenza viruses cause acute respiratory disease in young children, although infection may be asymptomatic. The viruses often produce serious lower respiratory tract illnesses: types 1 and 2 are usually associated with croup, while type 3 may cause bronchiolitis and pneumonia. Infections with type 4 virus are rarely serious and are not considered further. Acute lower respiratory illness from parainfluenza virus infection may eventually contribute to chronic obstructive pulmonary disease (Glezen, in press), however, no attempt has been made here to include such a contribution in chronic morbidity estimates. This aspect of the disease burden of parainfluenza virus infection needs periodic reevaluation. For the disease comparison in this report, all clinically significant cases of parainfluenza infection are assumed to occur in children under five years of age.
It is estimated that 25 percent of children under five years of age experience a clinically significant parainfluenza infection (initial or reinfection) each year (Clyde, personal communication, 1983; Glezen et al., 1982). Applying this percentage to 1984 population projections (18,232,490 children under age 5) produces a total of 4,558,123 cases.
Mild upper respiratory tract parainfluenza illness is judged to fall into Morbidity Category A; an estimated 20 percent subsequently progress to lower tract involvement (Category B). Thirty-one percent of cases (in both categories) are assumed to occur under 1 year of age