Parotitis in a Child Infected with Triple-Reassortant Influenza A Virus in Canada in 2007

Abstract

Swine H3N2 influenza virus designated A/Ontario/1252/2007 was isolated from a child with parotitis. Diagnosis was confirmed by viral isolation and serological assays. A/Ontario/1252/2007 was related to H3N2 triple reassortants that emerged in swine in the United States in 1998. Three of five tested household members were also seropositive for A/Ontario/1252/2007.

Influenza A viruses infect a broad range of species, with avian and swine strains presenting the most potential for interspecies transmission. Since 1998, H3N2 triple reassortants with genes derived from human (HA, NA, and PB1), swine (M, NS, and NP), and avian (PA and PB2) viruses have caused outbreaks of respiratory disease in pigs throughout the United States, and in 2005, such infections spread across Canada (6, 9). Zoonotic infections of humans with H3N2 swine influenza viruses (SIV) have been reported (5). Here, we report a case of acute parotitis due to the transmission of an H3N2 triple reassortant from pigs to a child. This is the third human H3N2 SIV infection identified in Canada since 2005. These results indicate that SIV infection in humans may be less rare than previously thought and that atypical presentation of the disease may be occurring.

Case report.

A 6-year-old boy presented to a local hospital emergency department in August 2007 with left-sided facial swelling and nasal congestion. The accompanying parent reported that the swelling had begun the previous day, while malaise and anorexia had begun 2 days previously. He did not have fever, cough, or pharyngitis. On examination, the left parotid gland was enlarged, but no purulent discharge was observed. No other household members were ill at the time. The boy lived on a swine farm that housed 2,000 pigs, and he occasionally visited the barn. No influenza outbreak was occurring among the pigs. A throat swab taken for bacterial culture and sensitivity was positive for Streptococcus pyogenes, and the child was put on antibiotic therapy. Mumps was also included in the differential diagnosis because of the clinical presentation (i.e., parotitis). Results of molecular tests (negative reverse transcription-PCR on a buccal swab specimen) and serologic tests (no change in immunoglobulin G titers between acute- and convalescent-phase sera) were not indicative of an acute mumps infection. A viral isolate was obtained from the inoculation of the buccal swab on a rhesus monkey cell culture, and direct antibody testing of the isolate was positive for influenza virus. The isolate was sent to the National Microbiology Laboratory for influenza virus subtyping, where it was designated A/Ontario/1252/2007. The child made an uneventful recovery at home.

Antigenic and molecular analyses.

A hemagglutination inhibition (HI) assay performed using 4 HA units of virus and 0.5% (vol/vol) turkey red blood cells showed that the isolate was not inhibited by antisera against recent (A/Wisconsin/67/05 and A/New Caledonia/20/99) and earlier (A/Panama/2007/99 and A/Nanchang/933/95) human influenza A virus strains but was inhibited by antiserum to A/Swine/Texas/4199-2/98 (H3N2) virus with an HI titer of >512, indicating that the A/Ontario/1252/2007 virus was antigenically related to SIV (Table 1) . The results also indicate that the assay is specific, since no cross-reactivity was observed between the human reference strain antisera and the swine influenza viruses (Table 1).

TABLE 1.

HI reaction of A/Ontario/1252/2007 isolate with reference antisera against currently circulating human and swine influenza viruses

Isolate	HI reaction to reference antiseruma
	A/New Caledonia/20/99 (human H1N1)	A/Wisconsin/67/05 (human H3N2)	A/Panama/2007/99 (human H3N2)	A/Nanchang/933/95 (human H3N2)	A/Swine/Texas/4199-2/98 (swine H3N2)
Control
A/New Caledonia/20/99	1,024	<4	<4	<4	<4
A/Wisconsin/67/05	<4	2,048	256	<4	<4
A/Ontario/RV1273/05	<4	<4	4	8	>512
Patient
A/Ontario/1252/2007	<4	<4	<4	<4	>512

^aThe HI titer is defined as the reciprocal of the last dilution of antiserum that completely inhibits hemagglutination.

Nucleotide sequencing of the full-length coding regions of all eight RNA segments of the isolate further determined that it was most closely related to A/Swine/Ontario/33853/2005 (H3N2) virus, which has the same human/classical swine/avian triple reassortant genotype as the H3N2 viruses that emerged in swine in the United States in 1998 (6). Sequence analysis showed that nucleic acid homology between A/Ontario/1252/2007 and A/Swine/Ontario/33853/2005 ranges from 98.9% (HA and PA) to 99.4% (M) and that amino acid identities range from 97.8% (NS) to 100% (NP). Phylogenetic analysis showed that all of the genes of A/Ontario/1252/2007 clustered with Canadian swine isolates from 2005 to 2006 (data not shown).

Serology.

Sera from the study case and the case's household members were tested by HI assay for antibody against the currently circulating human strains and the isolate from the study case, A/Ontario/1252/2007. A microneutralization (seroneutralization [SN]) assay was also performed as previously described using A/Ontario/1252/2007 (7). Specimens were considered seropositive for influenza virus at an HI or neutralization titer of ≥32. Serologic analyses on paired sera from the study case, collected 14 days apart, revealed seroconversion, with elevated titers to A/Ontario/1252/2007 by either the HI (1/128) or SN (1/64) assay (Table 2). Seropositivity to A/Ontario/1252/2007 was also demonstrated for the sera from the father and 3-year-old sibling but not for the sera from the mother and 5-year-old sibling (Table 2). Acute- and convalescent-phase sera from the study case and the serum from the 5-year-old sibling were also positive for A/New Caledonia/20/99, suggesting previous exposure to this epidemic virus. One sibling was not tested because of its young age (8 weeks).

TABLE 2.

Clinical features and HI and SN assay results with antisera from the study case and case's household members with recently circulating human influenza virus strains and swine influenza H3N2 virus A/Ontario/1252/2007

Household member (sample collection date [yr-mo-day])	Age (yr)	Gendera	HI assay titerb			SN titer for A/Ontario/1252/2007	Clinical symptoms
			A/Wisconsin/67/05	A/New Caledonia/20/99	A/Ontario/1252/2007
Study case
Acute-phase serum (2007-08-15)	6	M	<4	64	<4	<4	Facial swelling, nasal congestion
Convalescent-phase serum (2007-08-29)	6	M	<4	64	128	64	None
Father (2007-09-27)	36	M	<4	<4	64	32	None
Mother (2007-09-27)	29	F	<4	8	16	16	None
Sibling (2007-09-27)	5	M	<4	64	<4	<4	None
Sibling (2007-09-27)	3	M	<4	<4	256	128	None

^aM, male; F, female.

^bThe HI titer is defined as the reciprocal of the last dilution of antiserum that completely inhibits hemagglutination.

Swine investigation.

Influenza viruses H1N1 and H3N2 are known to be widespread across swine herds in Ontario, Canada. The records of the Animal Health Laboratory at the University of Guelph show that specimens from five separate swine farms, submitted by veterinarians from swine-producing areas in southwestern Ontario from early May 2007 to the present, were positive for influenza virus. Influenza virus subtypes included H1N1, H3N2, and some mixed H1N1/H3N2 within herds. Influenza virus infection does not usually have a significant economic impact on swine herds, and producers must pay for veterinary and laboratory services. Therefore, few observed cases are submitted to the laboratory database. Private swine veterinarians were not aware of any unusual influenza activity either.

Conclusion.

In this study, we report for the first time a case of H3N2 SIV infection in a child with acute parotitis. Serologic evidence of H3N2 infection was also found in three of six household members. Seropositive household members did not present any signs or symptoms of viral infections, suggesting that unrecognized human infection with SIV is also occurring.

Several cases of SIV infection in humans have been reported (5). In 4 of these cases, H3N2 viruses were isolated, and 46 of these cases involved H1N1 viruses. All case patients were described as having pneumonia, upper respiratory infections, or acute respiratory disease (5). To our knowledge, acute parotitis has not been associated with any previous human SIV infection. The most common cause of parotid gland swelling in childhood is mumps infection. However, viral parotitis can also be caused by Epstein-Barr virus, coxsackievirus, influenza A virus, and parainfluenza viruses (3). Bacterial parotitis is generally caused by Staphylococcus aureus and, less commonly, by Streptococcus species, such as Streptococcus pyogenes, and typically occurs in patients who are dehydrated, elderly, or immunocompromised (3). It is generally differentiated from viral parotitis by the expression of pus from the parotid gland. In this case, the lack of pus secretion from the parotid gland, the isolation of SIV from a buccal swab, and further confirmation of SIV infection by serology, as well as the absence of laboratory evidence of an acute mumps infection, suggest that the acute parotitis may have been caused by SIV. At the very least, SIV infection may have presented favorable conditions for secondary infection with Streptococcus pyogenes through dehydration in the child. Furthermore, the finding of Streptococcus pyogenes is not that surprising, since it can be part of the normal oral flora. Cases of human influenza A virus parotitis have been reported in Massachusetts (12 cases) during the 1975-1976 influenza season, Chile (1 case) in 2003, and New York in 1984 (1 case) and 2004 (1 case) (1, 2, 4, 8). Like the child in this study, patients presented with one or more symptoms of viral illness, but most (14/15) did not have a range of clinical symptoms typically associated with influenza (fever, malaise, myalgias, and cough). Our results show that SIV infection in humans can be involved in the occurrence of sporadic cases of parotitis, similarly to what has been observed for human influenza A. As the threat of an influenza pandemic due to highly pathogenic H5N1 avian influenza virus looms, a better understanding of all clinical manifestations of influenza is essential for the proper management of atypical cases of influenza. Therefore, in the proper clinical and epidemiological settings, acute parotitis should be considered in the differential diagnosis of influenza. Prompt identification of a novel influenza virus strain increases the lead time for development of a vaccine and implementation of prevention and control measures. Standard methods for influenza virus detection and subtyping are mostly optimized for the currently circulating human strains (H1N1 and H3N2). Molecular diagnostic methods such as reverse transcription-PCR would allow for the rapid detection of and differentiation between epidemic and pandemic viruses.