Rapid Flu Tests Often Inaccurate

http://www.nytimes.com/2009/08/06/health/06flu.html?_r=1&ref=business

Dr. Ginocchio is a representative for a more expensive but slower test, but she is dead right about this. There have been fatal cases documented here on allnurses indicating the unreliability of these rapid tests months ago.

Dr. Tamerius, who has a financial interest in the rapid test, blames its inaccuracy on the performance of the collector, or the timing of the test. Technically this is correct, but it is only part of the problem. There is another issue not addressed here. Viruses mutate, and the primer used for the test has to be updated periodically as well to adjust for those changes.

The accuracy of these tests has been an issue for the past few years. It became especially evident in diagnosing H5N1 bird flu. Doctors in bird flu endemic countries such as Indonesia have learned the hard way through the deaths of patients, to take a good history looking for increased risk factors, and to look closely at the patient. Treatment with Tamiflu is most effective if started within the first 48 hours of showing symptoms. And, it is not just about the Tamiflu. If you have a sick patient, you want to know what you are dealing with, and what the clinical course might be going to look like. Bird flu cases and some of these swine flu cases have progressed very rapidly from a few symptoms to the need for critical care. These patients will need a lot more than just Tamiflu.The onus of whether to treat or not to treat is placed squarely on the practioner seeing these patients. CDC says not to rely on the test results alone. Doctors must look at the clinical picture.

Dr. Christine Ginocchio, of the North Shore-Long Island Jewish Health System, whose research show a flaw in rapid flu tests.

...the tests have a severe limitation: They may fail more than half the time to detect swine flu infections, according to newly published studies and to experts in medical testing.

The low sensitivity of the tests is becoming a concern to health authorities because a false negative reading might prompt a doctor not to prescribe antiflu drugs.

It is also one of the big issues laboratory directors face as they prepare for what is expected to be a crush of flu testing this fall and winter. Numerous diagnostics companies are hoping to capitalize on demand for influenza testing.

The rapid tests "are missing a ton of flu," said Christine C. Ginocchio, director of the division of microbiology, virology and molecular diagnostics at the North Shore-Long Island Jewish Health System in Lake Success, N.Y.

For seasonal flu, experts have long known about the low detection ability of the rapid tests. The new studies suggest the tests are no better, and possibly worse, at detecting the swine flu strain now spreading around the world, known formally as the novel H1N1 virus.

In a study published recently in The Journal of Clinical Virology, Dr. Ginocchio found that one rapid test detected only 10 percent of the swine flu infections that could be picked up by a more sophisticated laboratory culture. A different rapid test detected 40 percent. (Dr. Ginocchio is a consultant to Luminex, a company that makes a more accurate but slower test.)

The federal Centers for Disease Control and Prevention said Thursday that in its own study, three rapid tests detected 40 to 69 percent of the swine flu cases. The rapid tests performed better on the seasonal flu, picking up as many as 80 percent of the cases.

Last week, the C.D.C. updated guidance urging doctors to be cautious in relying on the tests.

"We're saying you need to understand the limitations of these tests," Dr. Timothy M. Uyeki, an author of the C.D.C. guidance, said in an interview. "The clinician should not base a decision to treat or not treat on the basis of a negative result."

But some doctors say there is no good substitute for the simplicity, speed and low cost of the rapid tests. Manufacturers of the tests say the products are helpful if used appropriately.

"When these tests are used properly, the performance is very, very good," said John D. Tamerius, senior vice president for clinical and regulatory affairs at Quidel, which describes itself as the leading maker of such tests.

But in a letter to The New England Journal of Medicine in June, Navy researchers said the Quidel test detected only half the swine flu infections caught by a more sensitive technique. In the C.D.C. study, Quidel's test picked up 69 percent of the swine flu cases, the best performance of the three tests studied.

More accurate tests are available but they generally require sophisticated laboratories. And results might not come for a day or more, making the tests of little use in deciding whether to prescribe drugs like Tamiflu, which are supposed to be started within 48 hours of the appearance of symptoms.

The rapid tests, by contrast, take only a few minutes to half an hour, and most can be done in the doctor's office or emergency room, without a laboratory. That is appealing to some doctors.

The rapid tests also cost only about $10 to $20, though some doctors might charge more, versus about $100 for a test using a sophisticated technique called the polymerase chain reaction, or P.C.R. While a negative result on a rapid test might not be reliable, a positive result, at least during flu season, usually does indicate a person has the flu.

The rapid tests do not tell if a patient has the swine flu. They say only if flu is present, or in some cases whether it is type A or type B influenza. The swine flu is type A, but so are many seasonal flu strains.

So more sophisticated tests, beside being used to double-check a negative rapid test result, are also needed to see if a positive test result is the swine flu.