Study shines new light on worldwide co-infection

Data show high rate of TB in U.S. HIV-infected

The prevalence of tuberculosis among HIV-positive Americans is 40 times higher than in the general population, underscoring the need for more TB screening and preventive therapy for HIV-positive patients, according to a report from the Centers for Disease Control and Prevention (CDC).

"We know the toll of these interconnected epidemics, and now we need to strengthen our prevention efforts to include TB screening for all people infected with HIV, and if needed, preventive therapy to avoid developing TB disease," says Helene Gayle, MD, MPH, director of the CDC's Center for HIV, STD, and TB Prevention.

The study was one of more than a dozen presented during the 12th World AIDS Conference held June 28 to July 3 in Geneva. Among the more noteworthy findings presented was that TB infection rates between 1993 and 1996 were remarkably high in HIV-positive persons - 333 per 100,000. The rate among African Americans was three times higher than whites.

Overall, of the 588,245 adults with AIDS, 33,767 or 5.7% were reported with TB. A CDC analysis shows that people with AIDS who live in the United States but were born in other countries are more likely to be reported with TB than those born in North America.

Most cases of co-infection were reported in persons born in Africa (13.8%), followed by persons born in the Caribbean (13.7%), Central/South America (10.3%), Asia/Pacific region (10.1%), and Europe (4.2%). Asians born in the Asia/Pacific region were more than four times as likely to be co-infected than Asians born in this country.

The study concludes that these patterns of co-infection may reflect not only TB prevalence in countries of origin but also delayed access to preventive therapy for certain racial groups.

A survey of co-infection rates at U.S. correctional facilities throughout the same period found that 7% of TB patients were inmates at the time of TB diagnosis. Among these inmates, half (1,663) had their HIV status reported, and of that group, 56% were found to be HIV-positive. Compared to HIV-negative inmates, more HIV-positive inmates were African American (66% vs. 57%), born in the United States (93% vs. 86%), and had died before being diagnosed with TB (6% vs. 1%).

The correctional facility survey also found that HIV-positive inmates more often had normal chest radiographs (17% vs. 7%) and noncavitary disease (84% vs. 67%). Sputum smear positivity and culture positivity did not vary significantly, while anti-tuberculosis drug resistance was more prevalent among HIV-positive inmates (9% vs. 2% for rifampin resistance; 7% vs. 2% for isoniazid plus rifampin resistance).

In a study of co-infection among female patients in the United States, the CDC found that 29% of women with TB were HIV-positive. Co-infected patients were more likely to be young (37 years vs. 45 years) than HIV-negative women and more likely to be African American (72% vs. 40%).

There was little difference in rates of pulmonary TB in HIV-positive and HIV-negative women. However, HIV-positive women were twice as likely to have both pulmonary and extra-pulmonary disease than HIV-negative women. Also, HIV-positive women were less likely to have cavitary disease and abnormal chest radiograph, making diagnosis more difficult, the CDC notes.

Tracking the epidemiology of extrapulmonary TB among HIV-positive persons in the United States was the focus of another study presented at the conference. In an analysis of all extrapulmonary TB cases from 1993 to 1996, the CDC found that of those cases in which the patient's HIV status was known, 48% were HIV-positive. Approximately 37% had extrapulmonary TB only and 64% had both pulmonary and extrapulmonary TB. Among those co-infected, the most frequent disease sites were lymph nodes (47%), pleura (13%), miliary (11%), meninges (8%), and bone or joint (6%).

In addition, the study found that HIV-positive patients were more likely than HIV-negatives to have miliary disease (11% vs. 6%) and meningeal disease (8% vs. 5%). Drug resistance rates and positive culture results were also higher among HIV-positive patients. In another survey of pulmonary TB and drug susceptibility among HIV-positive persons in sentinel hospitals in the United States, the CDC found that HIV-positive patients receiving care at those hospitals were 14 times more likely than HIV-negative patients to have pulmonary TB. Moreover, HIV-positive patients were five times more likely to have multidrug resistant infections than those who were HIV-negative.

The impact of HIV infection on TB patients in selected HIV clinics in New York City appears to have decreased significantly in recent years, according to another study presented at the conference. At one time the center of the co-infection epidemic, New York City has seen the incidence of TB in HIV-positive patients decline 51% in five clinics at four hospital surveyed by the CDC.

Incidence of co-infection at the clinics decreased from a high of 43 cases per 100 person years in 1992 to 2.2 in 1996. Mortality among co-infection cases that were followed in the clinic was 30% one year after diagnosis and 54% after two years. Despite the dramatic decline, the city still has one of the highest incidence of co-infection in the country.

Unlike many urban communities in the North-east, New Haven, CT, has not seen HIV fuel its TB epidemic, indicating that the introduction of public health measures to control TB at a critical time in the AIDS epidemic may have thwarted the impact seen elsewhere.

Researchers from Yale University, also in New Haven, found that the incidence of TB has declined since 1987, despite significant drug use, poverty, and a high AIDS incidence. By using molecular epidemiological techniques, the researchers showed that only about one-fourth of recent TB cases in the city were the result of primary infection.

CDC study presented

Identifying the natural history of HIV and TB co-infected was the subject of a CDC study also presented at the conference. Cases were followed on patients in six states that reported 62% of AIDS cases nationally from 1992 to 1994 and that provided information on all cases of TB in persons with HIV infection for a mean follow-up of 29 months.

The study found that TB was the initial AIDS diagnosis in 72% of 2,595 HIV-infected persons reported during that period. Among patients whose AIDS-defining condition was TB first, 41% had at least one other opportunistic infection subsequently. The most common subsequent opportunistic infection was Pneumocystis carinii pneumonia (22%), wasting (12%), candida esophagitis (8%), and disseminated MAC (8%). The study also found that survival from AIDS was longer for those who had TB only at the initial AIDS diagnosis.

Treatment studies relevant to the United States were limited at the conference. One such study tried to evaluate which HIV-positive populations should be considered "high risk" and candidates for preventive therapy. Researchers at Boston Public Health Commission identified 245 anergic HIV-positive persons without prior positive PPD tests and evaluated how they progressed to TB disease.

Only one patient, a foreign-born male who was anergic in 1992, developed culture-confirmed pulmonary TB two years later. Also, only 16% of patients completed preventive therapy, while the incidence of TB during an average 32-month follow-up was only 1.6 per 1000 person years.

"We concluded that even in `high risk' populations, anergy testing failed to identify persons likely to progress to TB disease," the authors note. "Decisions regarding preventive therapy should be based on PPD testing alone, both in individuals and populations."

Follow-up data were presented on a randomized trial of rifampin plus pyrazinamide for two months vs. isoniazid for 12 months in HIV-positive patients. Patients in both groups had low rates of TB with no significant differences in efficacy, the authors noted. The two-drug, two-month regimen resulted in better overall adherence but more opiate withdrawal symptoms, they added.

In a meta-analysis of isoniazid preventive therapy in HIV-infected patients, researchers from Switzerland and Canada evaluated six randomized control trials, concluding that isoniazid prophylaxis reduced the risk of TB in skin-test positive patients with HIV infection. However, its impact on mortality needs further investigation.

In a final treatment study presented at the conference, researchers at Case Western Reserve University in Cleveland evaluated the safety, efficacy, and tolerance of ethambutol during the continuation phase of treatment in HIV-positive patients in Uganda. The researchers concluded that the drug was safe, and the relapse rate was comparable to the standard thiacetazone-containing regimen in HIV-positive patients.