154 Laboratory Findings

Serology can either detect total antibodies or be class-specific (IgG, IgA, or IgM).

Different serological techniques are available for influenza diagnosis: haemagglutination inhibition (HI), compliment fixation (CF), enzyme immunoassays (EIA) and indirect immunofluorescence.

Serological diagnosis has little value in diagnosing acute influenza. In order to diagnose acute infection, an at least four-fold rise in titre needs to be demonstrate, which necessitates both an acute and a convalescent specimen. However it may have value in diagnosing recently infected patients.

Serology is also used to determine the response to influenza vaccination (Prince 2003).

Serology has greater clinical value in paediatric patients without previous exposure to influenza since previous exposure can lead to heterologous antibody responses (Steininger 2002).

Haemagglutination inhibition (HI)

HI assays are labour intensive and time consuming assays that require several controls for standardisation. However the assay reagents are cheap and widely available. Various red blood cells such as guinea pig, fowl and human blood group “O” erythrocytes are used. An 0.4−0.5% red blood cell dilution is generally used. Serum is pre-treated to remove non-specific haemagglutinins and inhibitors. A viral haemagglutinin preparation that produces visible haemagglutination (usually 4 haemagglutination units) is then pre-incubated with two-fold dilutions of the serum specimen. The lowest dilution of serum that inhibits haemagglutination is the HI titre. HI is more sensitive than complement fixation (Julkunen 1985, Prince 2003) and has the added advantage that it is more specific in differentiating between HA subtypes (Julkunen 1985).

Complement fixation (CF)

Complement fixation tests are based on the ability of antigen-antibody complexes to consume complement – which results in no compliment being available to lyse sensitised sheep red blood cells. These assays are labour intensive and necessitate controls for each procedure but reagents are cheap and widely available. CF assays are less sensitive than HI both in the diagnosis of acute infection and the determination of immunity after vaccination (Prince 2003).

Ezyme immuno assays (EIA)

EIAs are more sensitive than HI or CF assays (Bishai 1978, Julkunen 1985). Various commercial EIAs are available. Assays that detect IgG and IgA are more sensitive than IgM assays (Julkunen 1985) but are not indicative of acute infection.

Indirect immunofluorescence

Indirect immunofluorescence is not commonly used as a method to detect influenza virus antibodies.

Rapid tests

The clinical value of a diagnostic test for influenza is to a large extent dependent on the particular test’s turnaround time. The first diagnostic tests that were developed

Laboratory Tests 155

for influenza diagnosis were virus isolation and serological assays. At that stage it took more than two weeks to exclude influenza infection. Although shell vial tests have reduced the turn-around time of isolation, they are not generally regarded as rapid tests.

The development of direct tests such as immunofluorescence enabled the diagnosis within a few hours (1 to 2 incubation and wash steps). Immunofluorescence tests however necessitate skilled laboratory workers and the availability of immunofluorescence microscopes.

The revolution in rapid diagnosis of influenza was brought about by the development of rapid antigen assays (most of which work on an EIA or immunochromatography principle). These assays enable the diagnosis of influenza within 10-30 minutes. Some of these tests are so easy to perform that even non-laboratory trained people can perform these tests in the clinic, which is referred to as bedside or point- of-care testing.

RT-PCR reactions that required a gel electrophoresis step were initially time consuming but the relatively recent development of real-time technology made RTPCR diagnosis within about two hours possible. Although antigen assays are generally the most user-friendly, they are not as sensitive as direct immunofluorescence, isolation or RT-PCR.

Table 1 compares the characteristics of the different test methods available for influenza diagnosis.

Table 1: Comparison of test characteristics*

	Test	Sensitivity	Turnaround	Ease of per-	Affordability
			time	formance
	Direct detection
	Rapid tests (EIA /	-2	+2	+2	0
	chromatography)
	Immunofluorescence	0	+1	+1	+1
	Gel electrophoresis	+2	0	-1	-2
	RT-PCR
	Real-time RT-PCR	+2	+1	-1	-2
	Viral culture
	Routine viral culture	+2	-2	-1	+2
	Shell vial culture	+1	0	-1	+1
	Serology
	EIA	+2	-2	+1	+1
	Haemagglutination	+1	-2	-1	+2
	inhibition
	Complement fixation	0	-2	-2	+2

*Relative criteria for favourability of tests (5 point ordinal scale) -2: very unfavourable characteristic

-1: unfavourable characteristic 0: average characteristic

+1: favourable characteristic

+2: very favourable characteristic