In the escalating battle against extensively drug-resistant tuberculosis, conflicting findings from laboratory tests have hampered efforts to control the spread of the disease.
Some of the conflicts come from a lack of standardized testing methods and others from subtle but critical differences in the way the tests are performed.
The most celebrated example of such discordant findings involved Andrew Speaker, the Atlanta lawyer who caused an international health scare after traveling to Europe in May with what was believed to be extensively drug-resistant tuberculosis, known as XDR.
This month, Speaker's doctors downgraded his type of tuberculosis to multidrug-resistant, or MDR, after repeating tests that initially gave a different result.The reclassification drew widespread publicity partly because conflicting findings came from tests performed at two of the country's main reference laboratories for tuberculosis -- the Centers for Disease Control and Prevention in Atlanta and the National Jewish Medical and Research Center in Denver.
Full details of Speaker's case are not known. But the problems go beyond one case. Last week, as Speaker underwent surgery in Denver to remove the diseased upper lobe of his right lung, a panel of experts met in Geneva at the World Health Organization's headquarters to discuss resistant tuberculosis testing. The discussion focused on updating the recommendations for such procedures made by a similar agency panel in 2001.
The overwhelming majority of tuberculosis cases are caused by bacterial strains that yield to the standard, or first-line, anti-TB drugs. Newer, second-line drugs are used if a strain of tuberculosis is MDR or XDR, which are resistant to the first-line drugs. If tuberculosis strains are not tested for drug resistance as soon as they are found in a patient, the problem may be detected too late to permit a cure.
Tuberculosis resistance develops when drugs are misused or mismanaged. For example, patients may fail to complete their full course of treatment. Health care providers may prescribe the wrong treatment, the wrong dose or the wrong length of time for taking the drugs. Another problem occurs when drugs are not available, or when the drugs are of poor quality.
What alarms health officials is the potential for outbreaks of MDR to evolve into large XDR ones, creating the specter of an uncontrollable health menace. Officials are trying to control an outbreak in South Africa, in which XDR killed 52 of 53 infected people, all of whom were also infected with the AIDS virus.
Reliable tests to determine resistance to first-line drugs were developed when the drugs were first marketed about a half-century ago. Fewer resistance tests exist for the newer, second-line drugs needed to treat MDR and XDR tuberculosis, and many of them are difficult to perform for a number of reasons.
One reason is that some can become unstable under conditions of laboratory testing. Laboratories do not grind up pills for resistance testing because they contain substances that could lead to unreliable results. Instead, laboratories use pure powders of an anti-tuberculosis drug's active ingredient. Another reason is that the many steps involved in the laboratory process increase chances for human error.
In 2001, the WHO panel said the latest knowledge was "very incomplete regarding how to best perform" resistance tests of second-line drugs and the usefulness of the tests in treating such cases.
Since then, experience with newer laboratory techniques and a review of published scientific papers show their usefulness, participants said in interviews.
But countries have neglected investments in tuberculosis research and care for many years, and now they need to make greater efforts to improve testing for resistance, said Abigail Wright, a tuberculosis expert at WHO.
The panel also relied on quality-control checks of a number of laboratories, which showed their ability to correctly detect resistant strains.
Drug-resistant tuberculosis is not new. In 1962, Eleanor Roosevelt died from a strain of tuberculosis resistant to isoniazid and streptomycin, according to published research by Barron Lerner of Columbia University, where she was treated.
As more tuberculosis strains have become resistant to more drugs in recent years, health officials have come up with the MDR and XDR designations. Such forms are much costlier to control than standard tuberculosis.
In 2006, the Centers for Disease Control and the WHO revised the definition of XDR in part because of a lack of standardisation.