To the Editor —In the 15 October 2005 issue of the Journal Puzelli et al. reported the first serological evidence for transmission of low-pathogenicity avian influenza (LPAI) viruses to humans and suggested that all previous human infections have involved only high-pathogenicity avian influenza (HPAI) strains [1]. We would like to respond to this assertion, on the basis of our experience with the H7N3 subtype of avian influenza virus in British Columbia, Canada, during 2004

Between 16 February and 18 May 2004, an outbreak of HPAI subtype H7 occurred among poultry in the Fraser Valley of British Columbia. The index farm had 2 flocks of chickens—an older flock of egg layers and a younger flock of broilers, housed in separate barns. On 4 February, the older flock experienced a transient reduction in egg production and a slight increase in mortality (0.5% over 72 h), which subsequently resolved. On 16 February, the younger flock, in the adjacent barn, experienced a sudden substantial increase in mortality (>25% over 48 h). Virus from the first flock was characterized as LPAI subtype H7N3 (intravenous pathogenicity index [IVPI], 0), and that from the second flock was characterized as HPAI subtype H7N3 (IVPI, 2.96). Conversion of this strain’s pathogenicity was attributed to a recombination event in which a 21-nt sequence from the matrix gene was inserted into the protease-cleavage site of the hemagglutinin gene. By 5 April, 20 commercial flocks in the region had become infected, and preemptive depopulation of all poultry (population, ∼19 million) in the Fraser Valley was initiated. Ultimately, 42 commercial farms and 11 backyard flocks, representing ∼1.3 million birds, were considered to have been infected with HPAI subtype H7N3 [2–4]

Enhanced surveillance for conjunctivitis and influenza-like illness was conducted among persons involved in controlling the outbreak, and recommended precautions were predicated on the experience with subtype H7N7 in The Netherlands during 2003; Despite recommended precautions, 2 poultry workers—1 on 16 March and 1 on 24 March—experienced the onset of unilateral conjunctivitis (the first case was accompanied by coryza, the second was accompanied by headache), after separate incidents of direct conjunctival contact, on 13 March and 22–23 March, respectively, with poultry that had died from influenza infection. Neither worker had been wearing goggles or taking oseltamivir as had been recommended. Influenza A virus was isolated from both workers and was diagnosed as subtype H7, by reverse-transcription polymerase chain reaction and sequencing of the hemagglutinin gene, and was further characterized as subtype H7N3, by serotyping. On the basis of sequencing of the genomes of the isolates, IVPI testing, and clinical findings (in poultry and in ferrets), the isolate from a nasal specimen from the first case (collected 3 days after exposure) was characterized as being LPAI, and that from a conjunctival swab specimen from the second case (collected 2–3 days after exposure) was characterized as being HPAI. Isolates from poultry from the farms where these infections were acquired were characterized as being HPAI, by sequencing and IVPI testing. These findings do not preclude the possibility that, in the first case, the isolate characterized as being LPAI represented a back mutation from the HPAI that had caused the outbreaks in poultry. Alternatively, both LPAI and HPAI viruses may have been cocirculating among poultry [2–4]

Interestingly, development of antibody to the H7N3 subtype could not be demonstrated in either of these 2 human cases, regardless of the serological tests employed. Acute- and convalescent-phase serum samples collected several weeks after the onset of illness were tested by microneutralization assay and Western blot, as well as by routine and modified hemagglutination-inhibition assays. Both cases occurred after an incubation period of 1–3 days and were unlikely to have been the result of contamination, irritation, or trauma. Mild, primarily conjunctival infections induced by some H7 subtypes may not induce a strong systemic antibody response. In light of uncertainties in the sensitivity and specificity of serological assays, the definitive evidence for infection is based on the isolation of virus and the detection of viral RNA, particularly from sites where symptoms are manifest

The pathogenicity index is a measure used to characterize virulence in poultry. No evidence is available to support extrapolation of the significance of IVPI in poultry to an estimate of risk in humans. All outbreaks in poultry that bring humans into contact with new avian influenza subtypes should be cautiously managed, and appropriate protective measures should be applied to prevent adaptation to the human host

References

1.
Puzelli
S
Di Trani
L
Fabiani
C
, et al.  . 
Serological analysis of serum samples from humans exposed to avian H7 influenza viruses in Italy between 1999 and 2003
J Infect Dis
 , 
2005
, vol. 
192
 (pg. 
1318
-
22
)
2.
Pasick
J
Handel
K
Robinson
J
, et al.  . 
Intersegmental recombination between the haemagglutinin and matrix genes was responsible for the emergence of a highly pathogenic H7N3 avian influenza virus in British Columbia
J Gen Virol
 , 
2005
, vol. 
86
 (pg. 
727
-
31
)
3.
Hirst
M
Astell
CR
Griffith
M
, et al.  . 
Novel avian influenza H7N3 strain outbreak, British Columbia
Emerg Infect Dis
 , 
2004
, vol. 
10
 (pg. 
2192
-
5
)
4.
Tweed
SA
Skowronski
DM
David
ST
, et al.  . 
Human illness from avian influenza H7N3, British Columbia
Emerg Infect Dis
 , 
2004
, vol. 
10
 (pg. 
2196
-
9
)
Potential conflicts of interest: none reported
Financial support: BC Centre for Disease Control and Public Health Agency of Canada