PCR is now the preferred method to detect FHV in biological samples. Viral isolation lost interest but is a valid method still used in several laboratories. The sensitivity and the specificity of the tests, and especially PCR, are good but may differ depending on the laboratory because there is no harmonisation. These tests, and immunofluorescence are described in this chapter.
PCR is currently used to detect FHV DNA in conjunctival, corneal or oropharyngeal swabs,
corneal scrapings, aqueous humor, corneal sequestra, blood or biopsies. Conventional PCR,nested-PCR and real-time PCR are used routinely to detect FHV DNA in diagnostic laboratories (Hara et al., 1996; Helps et al., 2003; Marsilio et al, 2004; Maggs et al., 1999a; Nasisse et al., 1997; Stiles et al., 1997a, 1997b; Sykes et al 2001, Vögtlin et al., 2002; Weigler
et al., 1997). Most PCR primers are based on the highly conserved thymidine kinase gene.
Molecular diagnostic methods appear to be more sensitive than virus isolation or indirect
immunofluorescence (Burgesser et al., 1999; Reubel et al., 1993; Stiles et al., 1997; Weigler
et al., 1997).
Because the very low amounts of viral nucleic acids detectable by PCR may not be associated
with disease, PCR positive results should be interpreted with caution. The sensitivity of PCR
depends on the test (Maggs and Clarke, 2005) and it is advisable to use a system that includes
a control that detects feline DNA to give an indication of how much material was on the
swab, and to check for substances that might inhibit PCR. Due to its high sensitivity, PCR
may also detect viral DNA in scrapings of the cornea and/or tonsils suggesting nonproductive
infection (Maggs et al., 1999b; Reubel et al., 1993; Stiles et al., 1997a).
Consequently its predictive value for clinical infection may be poor, depending on the test
sensitivity, the samples analysed (corneal scrapings and biopsies more frequently yield
positive results than conjunctival ones) and the population tested (e.g. shelter cats are more
likely to test positive than owned pet cats).
Additionally, many if not all PCR tests are able to detect FHV DNA in modified-live vaccines
(Maggs and Clarke, 2005) and it is not presently known if vaccinal strains may be detected in
recently vaccinated animals and if so, for how long after vaccination.
A positive PCR result may represent low level shedding or viral latency and does not mean
that the virus is responsible for clinical signs, although it indicates the possibility of recurring
signs in the future. However, when quantitative real-time PCR is used (Vögtlin et al., 2002),
the viral load present in the material tested may provide additional information on the
etiological importance of the agent. When high loads are present in the nasal secretion or
tears, this suggests active replication and therefore involvement of the virus in the clinical
signs. If low copy numbers are detected in corneal scrapings, this would indicate a latent
infection.
Molecular diagnosis may be more convenient for clinicians, because the use of fluorescein
does not interfere with specificity of the test and samples can be mailed over several days at
ambient temperature (Maggs 2005). It also allows the simultaneous detection in the same
samples of other feline pathogens frequently implicated in respiratory and ocular diseases,
especially Chlamydophila felis and, less reliably, feline calicivirus (Helps et al., 2003;
Marsilio et al., 2004).
If PCR is not available, virus isolation (VI) is an alternative method of diagnosing FHV
infection. Virus isolation is less sensitive than PCR but does indicate that replicationcompetent
virus, not just DNA is present. VI also allows the simultaneous detection of feline
calicivirus.
In cats undergoing primary FHV infection, the virus can be easily detected by isolation from
conjunctival, nasal or pharyngeal swabs or scrapings, or from post-mortem lung samples But
during chronic infections when the aetiological origin of disease has to be confirmed, VI may
be more difficult.
Asymptomatic carriers may also be detected by VI and both positive and negative predictive
values of VI appear to be low in some studies (Gaskell and Povey, 1977; Maggs et al.,
1999b). Samples must be collected before application of fluorescein or Rose Bengal stainwhich can inhibit viral replication in cell culture (Brooks et al., 1994; Storey et al., 2002) and clinical specimens should be sent quickly to the laboratory and is ideally refrigerated during shipping. For logistic reasons and despite its good sensitivity in acute disease, VI is not routinely used for FHV infection diagnosis.
FHV-specific proteins can be detected by immunofluorescent assay (IFA) on conjunctival or corneal smears or biopsy. As for VI, fluorescein instillation should be avoided before sampling. For IFA, this may give false positive results and interfere with the interpretation of the test. IFA has been reported to be less sensitive than VI or PCR, especially in chronic infections (Nasisse et al., 1993; Burgesser et al., 1999). Although no correlation between VI and IFA testing has been observed, combination of VI and IFA may predict the presence of virus better than either test alone (Nasisse et al., 1993; Maggs et al., 1999b). Because of lack of sensitivity and the interference with fluorescein, often used in ophthalmology practice, IFA is not the most suitable diagnostic test in chronic ocular diseases (Nasisse et al., 1993).
FHV antibodies can be detected by serum neutralization or ELISA in serum, aqueous humour
and cerebrospinal fluid (Dawson et al., 1998; Maggs et al., 1999b). The seroprevalence is
very high in cats due to natural infection and vaccination. Consequently, the presence of
specific antibodies does not correlate with disease and active infection (Maggs et al., 1999b).
Moreover, antibody detection does not allow differentiation between infected and vaccinated
animals, neutralizing antibodies are undetectable until 20 to 30 days after a primary infection
and antibody titres may be low in animals with either acute or chronic disease. Consequently
serology has a very limited value in the diagnosis of feline herpesvirus infection (Nasisse and
Weigler, 1997; Maggs et al., 1999b; Maggs, 2005).