WASHINGTON, DC-To screen children appropriately for tuberculosis, physicians need to be aware of the tuberculosis skin test's limitations. Not only is the tuberculosis skin test expensive to perform, but it is often inaccurate in low-risk populations, cautioned Jeffrey Starke, MD. "So if you test a lot of low-risk kids, the vast majority of your positive results will be false, and you'll be doing it all for nothing."

Dr. Starke, a professor of pediatrics at Baylor College of Medicine in Houston, discussed how best to screen children for tuberculosis at the recent 1999 Annual Meeting of the American Academy of Pediatrics (AAP). He also provided details about pending changes in the pharmacologic management of tuberculosis in both children and adults.

NO NEED FOR MASS SCREENING

"Mass tuberculosis screening is a bad idea," emphasized Dr. Starke in an interview with Respiratory Reviews, "because tuberculosis is not an issue for the vast majority of children in the United States." Indeed, the prevalence of tuberculosis is 1% or less in most parts of the country. Because the risk of tuberculosis is so low, the false-positive rate associated with mass skin-test screening is about 92%, said Dr. Starke. The false-positive rate would be similar in other low-risk areas, such as Canada, Australia, New Zealand, and all of Western Europe, he added.

However, the tuberculosis skin test is very accurate in high-risk populations-for example, when tuberculosis incidence is 90%, the false-positive rate is only 1%. Therefore, its most appropriate and cost-effective use in this country is to screen immigrant children born in such higher-risk regions as Latin America and Asia.

A POWERFUL RISK FACTOR

"There's no question that being foreign-born is the number one tuberculosis risk factor in children, for a couple of reasons," said Dr. Starke. "First, it's the most powerful risk factor associated with a positive skin test."

Furthermore, it is the only risk factor that can be linked directly to the child. "All the other tuberculosis risk factors are attributable to the adults in the child's environment," Dr. Starke explained. "For example, some of them may have human immunodeficiency virus (HIV) infection or the acquired immunodeficiency syndrome. Others may have been in prison." In addition, foreign-born adopted children are at particularly high risk for tuberculosis, he noted, because disease rates are often extremely high among the adult employees at foreign orphanages.

Rates of tuberculosis infection may reach 2% to 5% per year among children in high-risk nations. "And children who come to the United States from these areas have been found to have a 15% to 25% chance of being infected," estimated Dr. Starke. "That's a pretty darn good chance." He therefore urged that all of these children receive at least one tuberculosis skin test, preferably upon their arrival. Other pediatric risk factors for which tuberculosis testing is indicated include:

• Having a family history of tuberculosis infection going back two to three generations;
• Having traveled abroad or having been exposed to foreign visitors;
• Having contact with HIV-infected persons;
• Having contact with current or previous prison inmates; and
• Living in high-risk areas.

Foster children should be tested, too, added Dr. Starke, since they may have a history of tuberculosis exposure that their foster parents are unaware of. In all cases, physicians should interpret the test after 48 to 72 hours, recording not only whether it is positive or negative, but also the size of the induration.

Fortunately, foreign-born children at high risk for tuberculosis do not necessarily remain at high risk. "Over time, they assume the same risk as the surrounding population," Dr. Starke stated.

NEW TREATMENT RECOMMENDATIONS

Dr. Starke was on a Centers for Disease Control and Prevention/American Thoracic Society committee that recently reexamined the long-standing practice of treating adult tuberculosis infection with 6 months of isoniazid administration. Health departments favored this practice, he said, because European studies from the 1960s and 1970s found the 6-month duration to be more cost-efficient than 9 months of treatment, and effective enough-though not as clinically effective as the 9-month course of therapy.

"We reanalyzed the data and realized we were wrong," Dr. Starke reported. "Isoniazid treatment in adults should really be 9 months." The new recommendation is not yet official, but he expects it to appear soon in Morbidity and Mortality Weekly Report (MMWR) and other medical publications.

Nine months of isoniazid used to be the standard treatment for children with tuberculosis. However, physicians have increasingly switched to 6 months because the latest edition of the AAP Red Book suggested that the shorter treatment period was acceptable, Dr. Starke said. Because no data exist to support 6 months of treatment in children, the next edition of the Red Book, scheduled for 2000, will strongly recommend 9 months of isoniazid administration.

For children and adults, treatment options will now also include a rifampin/pyrazinamide combination, which has been found effective for preventing tuberculosis infection from developing into full-blown disease in animal models and HIV-positive adults. "This regimen is third on the list, though," noted Dr. Starke, "because there's less evidence to support it than there is for either 6 months or 9 months of isoniazid."

-Timothy M. Begany