Botswanan isoniazid preventative therapy trial screens out large numbers of people who are already ill

Roughly half of the people living with HIV in Botswana referred from HIV testing clinics or other sites to enrol into a large isoniazid preventive therapy (IPT) trial ended up being ineligible for the treatment because they were already too ill or had abnormal chest x-rays suggestive of tuberculosis (TB), according to a report presented at the Sixteehth International AIDS Conference on Monday in Toronto.

The study enrolled its quota of participants, but this finding may be of particular importance for Botswana's larger IPT programme. According to Dr Charles D Wells of the US Centers for Disease who made the presentation: “National IPT programmes may best target candidates for IPT by targeting HIV screening to healthier individuals.”

The study illustrates a shortcoming of preventative therapy — that people often do not seek care or treatment until they are already ill.

TB and HIV coinfection in Botswana

Botswana has one of the highest HIV rates in the world, with around 25% of the adults aged 15 to 49 HIV-infected. But many people are also coinfected with TB. The prevalence of HIV in TB patients in Botswana is between 60 and 86%.

Until the early 1990s, Botswana was making steady progress in controlling TB, but since the HIV and AIDS epidemic, there has been a resurgence of TB in the country. Botswana now has one of the highest TB case rates in the world — with a TB case rate of 623 per 100,000 people in 2002. “TB is the leading cause of death in people with Botswana as evidenced by autopsy studies,” said Dr Wells.

Botswana has one of the world’s most progressive governments in terms of responding to the HIV epidemic. In 2001, the country rolled out a nationwide prevention of mother-to-child transmission (PMTCT) programme which now reaches an estimated 83% of the country's pregnant HIV-infected women. Between 2001 and 2005, the country began providing free antiretroviral therapy (ART) to eligible people with HIV, and now has over 65,000 people on ART. The country began offering routine “opt-out” HIV testing in 2004, and has so far identified about 350,000 HIV-infected individuals.

The country has also been a leader in terms of IPT. Following World Health Organization's recommendations that people with HIV and positive TB skin tests be given IPT, Botswana began implementing a large programme in 2004 to provide IPT to people with HIV — although, the country does not use TB skin tests to screen or enrol participants (since TB exposure is so common, and TB skin tests are less reliable in people with HIV). As of June of 2006, roughly 50,000 people have been screened for the IPT programme with 35,000-40,000 on treatment.

Questions persist about isoniazid preventive therapy

Although studies have shown that six months of IPT reduces the risk of people with HIV with latent TB developing active TB by about 60%, it is quite easy for people to become reinfected in regions with a high burden of TB. Clinical studies in such settings report mixed results following six months of IPT — one study in Zambia suggested that the protective effect could last up to three years (Quigley 2001), but a study from Uganda found the effect only lasted one year (Johnson 2001).

So the CDC and researchers in Botswana decided to conduct a double-blinded placebo-controlled study to see whether longer treatment is better than six months of IPT. The study sought to randomise about two thousand participants to either 36 months of IPT or to six months of IPT then 30 months of placebo.

Since the trial was enrolling people at the same sites that run that national IPT programme, researchers performed an analysis of the enrolment process in order to better inform the programme about characteristics of people with HIV seeking IPT — and about usefulness of programme’s screening criteria.

There were two rounds of screening for the trial. The first round of screening was essentially the same clinical criteria as for the national IPT programme, which excludes anyone who has a current illness (cough or fever), terminal AIDS, active hepatitis, a history of poor treatment adherence, pregnancy, or who has been treated for TB in the last three years. Subjects are given a physical exam and anyone with any symptoms of TB is also excluded from the programme.

But a second round of screening involving chest x-rays and laboratory evaluations was added for the trial. Anyone with an abnormal chest x-ray (without special circumstances), elevated liver enzymes, or significant neutropenia, or anaemia was excluded from the study.

Enrolment and screening characteristics

From 2004 to 2006, 4,346 people were referred to the trial: 311 declined participation, and 1,247 were screened out of the trial during the first round of screening (a 36% loss). Of the remaining 2,768, 140 more declined participation, and 613 were screened out due to their lab work and chest x-rays (a 28% loss). 2,004 participants are now on treatment, and eleven more should start soon.

During the first round of screening, the most frequent reasons for exclusion were current illness or terminal AIDS (52%). However, the physical examination in the first round of screening didn’t identify everyone who might be a poor candidate for IPT. In the second round of screening, the most common cause for exclusion was abnormal chest x-rays (50%). 20% of the clinical exclusions were due to neutropenia, and 7.3% were due to elevated liver enzymes.

341 (12%) of the people who passed the first round of screening had abnormal chest x-rays consistent with TB. However, thus far, only 25 of these cases have been confirmed as TB.

However, Dr Wells wouldn’t go so far as to say that IPT programmes should screen patients with chest x-rays, saying that “the data so far suggest that if the patients are asymptomatic, they probably don’t have TB.”

Instead, Dr Wells focused on other characteristics of the enrolees, for example, their CD4 cells. The median CD4 cell count for the enrolees in the study was 308 cells/mm³, and 595 (35%) had a CD4 cell count below 200 cells/mm³. However, 45% of those who were screened out during the second screening process had CD4 cell counts below 200 cells/mm³, vs. 31% in those who were eligible (p<0.001).

The point is that, like other interventions for people with HIV, such as ART, it would be better not to wait until people have very advanced disease before starting IPT. Although the IPT programme in Botswana has had no trouble enrolling tens of thousands, the number of people living with HIV in the country who could potentially benefit from IPT is much, much larger.