From ProMed:

Some Australians may have been falsely diagnosed with human swine flu while others with the disease may have been sent home with a negative
result because the laboratory test used to identify the potentially deadly virus is only about 90 percent accurate.

So far more than 1,200 Australians have tested positive to swine flu. [New South Wales] Health is among those using polymerase chain reaction testing, also known as PCR testing, to identify the H1N1 strain of influenza. Last night [9 Jun 2009] NSW Health declined to make any further comment. The latest information from the World Health Organization, supported by the US Food and Drug Administration, suggests the newly developed rRT-PCR “rapid” testing method gives only a “presumptive positive” rather than a “definitive positive” result for H1N1 influenza. According to reports from both organizations, the test can also provide a “false negative” result in cases where the patient actually has the virus. PCR testing is the global benchmark used to identify viruses and is endorsed by the WHO.

In April, the FDA issued an emergency use authorization to allow a “rapid” form of the method to be rushed to laboratories across the US to combat a potential swine flu epidemic, even though the method had not been fully developed or obtained approval. A document released by the FDA on 28 Apr [2009] states only that rRT-PCR swine flu testing “may be effective” in testing for swine flu. The FDA report states that “a positive result indicates that the patient is presumptively infected with swine flu virus, but not the stage of infection,” and that “a negative result does not, by itself, exclude the possibility of swine flu virus infection.”

The WHO has issued its protocols for using the new test method to laboratories around the world. That document states tests can provide only a “presumptive positive” result for swine flu. The WHO document also said “a false negative result may occur if inadequate numbers of organisms are present in the specimen due to improper collection, transport or handling.”

The rRT-PCR testing method was developed by the global pharmaceutical giant Roche — the company that also owns the rights to Tamiflu, one
of just 2 antiviral drugs used to treat the virus. George Koumantakis, the Scientific and Regulatory Affairs manager for Roche Diagnostics Australia, said the PCR testing method was “about 90 percent accurate.”

“It’s impossible to claim 100 percent accuracy for the method due to a number of reasons,” Dr Koumantakis said. “The primary one is that the virus only has to mutate slightly for the test to be completely ineffective.”

Last week Roche reported that stocks of Tamiflu had sold out in Australia due to unprecedented demand.

The number of Australians with the H1N1 virus increased from 1,207 to 1,224 yesterday [9 Jun 2009]. The health minister, Nicola Roxon, said the virus was mild in Australia, and only about 10 people had been hospitalized.

[Byline: Mark Hawthorne, Julia Medew, Ellie Harvey for the Sydney Morning Herald]

- –
Communicated by: ProMED-mail Rapporteur Mary Marshall

[With A(H1N1) influenza virus infections spreading rapidly around the world, it is not surprising that there is a real sense of urgency in developing tests to diagnose it. Ideally, a diagnostic test should be 100 per cent specific for the pathogen in question (A (H1N1) influenza virus in this case) and sensitive enough to detect all infected individuals without false positives. It should be inexpensive to conduct and suitable for processing a large number of
samples quickly. But diagnostic tests seldom, if ever, reach this ideal. The question, then, is if 90 per cent accuracy is helpful in providing timely treatment to patients and in tracking the disease in time and space. In this instance, it is much better than no diagnostic test at all, even if the positive results are considered
presumptive. — Moderator TY]