Rapid HIV tests are reducing unneeded PEP

High accuracy rates, fast results

New highly accurate rapid HIV tests are reshaping the way hospitals are handling bloodborne pathogen exposures.

"The specificity and sensitivity of these tests are at the 99% level," says Ronald H. Goldschmidt, MD, director of the family practice inpatient service at San Francisco General Hospital and co-director of National Post-Exposure Warmline, PEPline, Perinatal Hotline [(888) 448-4911].

"False positives and negatives are extraordinarily rare. That’s what characterizes these new tests as being different from the old ones, because the first generation of rapid tests were sufficiently inaccurate that they created the potential for lots of problems," he notes.

Within minutes, employee health professionals can reassure employees that they do not need to worry — or start them on antiviral medication to prevent infection. Positive results still must be confirmed by the Western blot or immunofluorescent antibody.

"The testing can alleviate unnecessary anxiety and avert unnecessary treatment," Goldschmidt says.

The Centers for Disease Control and Prevention (CDC) has not issued recommendations that specifically pertain to the recent generation of tests. But the 2001 Guidelines for the Management of Occupational Exposure to HIV, HCV, and HBV state: "Testing to determine the HBV, HCV, and HIV infection status of an exposure source should be performed as soon as possible. . . . An FDA-approved rapid HIV-antibody test kit should be considered for use in this situation, particularly if testing by EIA cannot be completed within 24 to 48 hours."1

In fact, 14% of health care workers who were exposed to blood or body fluid from a patient later found to be HIV-negative had started taking antiviral medication, CDC data show.1

Vanderbilt University Medical Center in Nashville, TN, began using the rapid tests during the daytime, weekday hours. The proportion of exposed employees taking post-exposure prophylaxis dropped from 23% to 15%.

"Typically, they would have to take (PEP) for two or three days while we waited for an Elisa [enzyme immunoassay] to come back from a source," says Melanie Swift, MD, medical director of the Vanderbilt Occupational Health Clinic.

In September 2004, the hospital expanded the use of rapid tests to cover all shifts. Swift expects the use of PEP to drop further.

Employees on the antiviral medication need monitoring and often experience side effects that cause them to miss work. But beyond the direct and indirect costs of PEP, there is great anxiety.

"We’ve had individuals lose sleep and be incredibly stressed out," says Swift. "We’ve had people on the weekend before their wedding considering postponing their wedding until they get results. It’s a major thing in their lives. [With rapid HIV tests], before your head hits your pillow that night, you know whether you’ve been exposed to HIV. You can’t put a price on that."

Even if you tell an employee that the source patient is not high-risk for HIV, the anxiety and uncertainty lingers, says Sue Sebazco, RN, CIC, director of infection control and employee health at Arlington [TX] Memorial Hospital and president of the Association for Professionals in Infection Control and Epidemiology (APIC) in Washington, DC.

"Often, we know the person is unlikely to have HIV, but that thread of doubt tends to carry the day," she says. "People quite understandably are worried and end up being treated unnecessarily until results come back days or weeks later."

Currently, four rapid HIV tests are available in the United States: OraQuick Rapid HIV Antibody Test, Reveal G-2 Rapid HIV-1 Antibody Test, Uni-Gold Recombigen HIV-1 Test, and the Multispot HIV-1/HIV-2 Rapid Test. (See charts.)

Hospitals are deciding which test to use based on the test’s characteristics and their specific needs. For example, Arlington Memorial prefers a plasma or serum test, which would be conducted in the laboratory, Sebazco says. Maintaining laboratory involvement allows the hospital to have greater control over confidentiality of results.

"We want the information as rapidly as possible, but we want it performed through our laboratory," she says.

Other hospitals may take advantage of tests that are CLIA-waived, which means they do not have to be performed in a lab. That would be particularly useful for smaller hospitals that do not have internal lab resources on a 24/7 basis.

"The most important thing is that people who require testing actually get tested, not who does the test," Goldschmidt says. "It would be the mistake to not do the testing because the proper laboratory personnel and training were not available."

Eileen Couture, DO, FACEP, clinical chair of the emergency department at Oak Forest [IL] Hospital, studied OraQuick as a point-of-care test in the ED.

The study was performed under strict guidelines and controls, and the tests were kept in a locked cabinet. The rapid testing provided results for patients while they still were being cared for in the emergency department. (Couture’s study did not include use of the test following occupational exposure.)

"We had very high specificity and sensitivity," she says. "It was easy to use, with a very quick turnaround."

While all the rapid HIV tests are highly accurate, they vary somewhat.

For example, OraQuick Advance can be used to test for HIV-1 and HIV-2. The whole blood version has no false negatives (specificity=100%), and the plasma and oral fluid test produces one false negative for every 1,000 tests performed (specificity=99.9%). It will result in four false positives per 1,000 tests (sensitivity=99.6%) in the whole blood and plasma versions, and seven false positives per 1,000 (sensitivity=99.3%) for oral fluid.

The new MultiSpot HIV-1/HIV-2 Rapid Test produces no false positives (sensitivity=100%) and seven false negatives per 10,000 tests (specificity=99.93%) for serum and nine false negatives per 10,000 tests (specificity=99.91%) for plasma.

Hospitals also should consider the probable predictive value of tests — the probability that a positive result is a true infection. The positive predictive value is lower when the disease prevalence is low. For example, if the HIV prevalence among the patient population is 0.1%, then only one in 1,000 patients has HIV infection. If the specificity of the test is 99.9%, then one in 1,000 test results is false negative. So for every true positive result, the test produces one false negative result — a rate of 50%.

Some hospitals have decided to keep their current Elisa testing, followed by Western blot confirmation, particularly if their lab offers a quick turnaround time. A test that could be conducted at the bedside creates potential quality control and even reporting issues, as some clinicians might just perform a rapid test and neglect to follow up, says James Garb, MD, director of occupational health and safety at Baystate Health System in Springfield, MA.