HIV programme in Western Kenya successfully delivers isoniazid preventive therapy to thousands

This article is more than 16 years old. Click here for more recent articles on this topic

Isoniazid preventive therapy (IPT) can be successfully delivered by HIV treatment programmes to thousands of people according to a report from the USAID-AMPATH care programme in Western Kenya presented Monday at AIDS 2008 in Mexico City. Between September 2004, and February 2007, the programme put close to 10,000 people with HIV on IPT, achieving high rates of treatment completion with lower rates of TB in people who completed treatment.

“IPT within a large HIV treatment program in sub-Saharan Africa appears both feasible and effective in preventing incident episodes of active tuberculosis,” said Dr Dr Lameck Diero of Moi University, who reported the findings.

IPT background

Isoniazid preventive therapy (IPT) is one of the three essential services known as the Three I’s, that WHO’s HIV and Stop TB Departments recommend that HIV programmes provide in order to reduce the burden of TB in people with HIV. IPT reduces the risk of active disease in people who have a recent or latent TB infection (see HATIP # 96).

There have been quite a few pilot studies of IPT, where the drug has been given to hundreds of patients. For example, in a poster presentation at this conference a programme in Thailand reported a trend towards a reduced rate of TB in around 1100 people with HIV who received IPT, and a 77% completion rate, very similar to the completion rate reported in Kenya (see below for further details of the Thai study).


active TB

Active disease caused by Mycobacterium tuberculosis, as evidenced by a confirmatory culture, or, in the absence of culture, suggestive clinical symptoms.


An antibiotic that works by stopping the growth of bacteria. It is used with other medications to treat active tuberculosis (TB) infections, and on its own to prevent active TB in people who may be infected with the bacteria without showing any symptoms (latent TB). 

person years

In a study “100 person years of follow-up” could mean that information was collected on 100 people for one year, or on 50 people for two years each, or on ten people over ten years. In practice, each person’s duration of follow-up is likely to be different.


The result of a statistical test which tells us whether the results of a study are likely to be due to chance and would not be confirmed if the study was repeated. All p-values are between 0 and 1; the most reliable studies have p-values very close to 0. A p-value of 0.001 means that there is a 1 in 1000 probability that the results are due to chance and do not reflect a real difference. A p-value of 0.05 means there is a 1 in 20 probability that the results are due to chance. When a p-value is 0.05 or below, the result is considered to be ‘statistically significant’. Confidence intervals give similar information to p-values but are easier to interpret. 

multivariate analysis

An extension of multivariable analysis that is used to model two or more outcomes at the same time.

But wide scale implementation has been lagging in many resource-limited settings because of concerns about the ability to monitor for drug toxicity and the ability to adequately screen for active disease (if active disease is missed, monotherapy with IPT could potentially lead to some isoniazid resistance).

Many countries have been waiting for more programmatic experience from other resource-constrained settings. But so far, only a large IPT programme in Botswana, and the large-scale “Thibela TB” trial of IPT in gold mines in South Africa have provided significant programmatic data. Unfortunately, recording and reporting problems have limited the useful data from the Botswana programme. Meanwhile, Thibela TB is an ongoing clinical trial that targets a specific setting (the mines) with exceptional adherence support mechanisms (incentives that might be difficult to replicate in a routine clinical setting).

However, AMPATH has been routinely providing IPT to its patients for a few years, so its team decided to review its electronic records between September 2004 and February 2007 to assess the outcomes and feasibility of the service.

AMPATH’s model

The AMPATH partnership is a collaboration of Moi University in Kenya and Indiana Universities supported by USAID/PEPFAR funding. Its first clinics opened in September 2001, and since that time, it has expanded to 18 main clinics and 8 satellite clinics in Western Kenya. AMPATH serves over 60,000 people, about half of whom are on ART. But TB has also been a major problem for the programme participants.

Like many sub-Saharan African countries, Kenya’s TB burden increased dramatically as a result of the HIV epidemic. Currently the national TB case notification rate is 329 per 100,000. 116,000 TB cases were reported in 2007, and about 50% of the TB cases are HIV-infected — and the case rate of TB in people with HIV is dramatically higher.

So AMPATH began screening every new person entering the programme for TB. The screening process involves a review of symptoms: looking for cough, fever, night sweats, weight loss, shortness of breath, sputum production and chest pain; a history of TB diagnosis or treatment; and a physical examination with a chest examination and lymph node exam. All patients also receive a chest x-ray.

If there are any signs or symptoms of TB or an abnormal chest x-ray, a sputum exam is ordered — and the diagnostic process continues until TB has either been diagnosed or excluded. IPT is given for nine months to every client who screens negative for TB and who has not previously been on TB treatment.

The programme uses standard data collection forms, and the data are later entered into the AMPATH electronic medical record system. The key endpoints recorded are either IPT completion or premature discontinuation (stopping before 9 months).

For this analysis, the investigators performed a retrospective record review. Analysis of data included comparison of categorical factors through chi-square tests, estimation of survival times by Kaplan-Meier method and multivariate analysis using Cox proportional hazards model.


9633 (33%) of the programme’s 29,197 clients who enrolled during the study period began IPT — almost none of those who qualified for IPT refused it. 7096 have met IPT endpoints. 5387 (76%) completed IPT, while 1,709 (24%) prematurely discontinued (the remainder are still on treatment).

At baseline, of the people starting IPT, 72.2% were women; the median age was 36.8 (interquartile range (IQR) 30.8-43.7), 44.5% attended urban clinic; median CD4 count 280 (IQR 153-432); 22.5% were WHO Stage III/IV and 32.9% were taking antiretroviral therapy.

Those who had less severe HIV disease (higher CD4 cell counts and WHO Stage I/II) or who attended an urban clinic were significantly more likely to complete IPT treatment (p<0.001).

53 (3%) of the 1709 people who quit IPT prematurely did so because they developed active TB, while another 73 developed TB after quitting for a 1 and 2-year incidence of 4.6% and 7.7%, respectively.

Dr Diero noted that a small number of discontinuations were due to adverse events, in particular hepatitis and peripheral neuropathy. He also noted that the TB cases that emerged despite taking IPT were treated with standard TB therapy containing isoniazid — although he said that optimally, such individuals should receive drug sensitivity testing. (Also, when he previously presented these findings at the HIV Implementers meeting, he said that treatment response to standard TB therapy was quite good.)

Among the 5387 people who completed IPT, 89 (1.7%) developed active TB: a one-year incidence of 2.1% and 4.6% at 2 years.

The median time for developing TB in these subjects was 480 days. “The non-completers developed TB much earlier,” said Dr Diero.

He also noted that even though perfect adherence is clearly important, it did not dramatically affect TB outcomes in this study.

“The proportion of people without TB was much higher in those with perfect adherence. However the difference between perfect and imperfect adherence did not reach statistical significance,” said Dr Diero.

However, in a multivariate analysis, several factors were associated with subsequently developing TB: having a low CD4 count at baseline (if below 50 cells/mm3, the Hazard Ratio (HR) was 2.575; p<0.009, if 50-99, the HR was 2.527; p=0.003); quitting IPT early (HR 1.79; p=0.001) or having WHO stage III/IV disease (HR 1.717, p-value 0.004).

Higher than usual treatment completion

Some in the audience noted that this programme achieved much higher rates of treatment completion than has been reported in other studies or settings.

“We do have a very strong outreach and social support programme that helps with adherence. And if people do not come in for their next visit, we have people who visit them in their homes and pull them back to the clinics,” said Dr Diero.

IPT in Thailand

Due to the prevalence and risks presented by co-infection, isoniazid preventive therapy (IPT) is recommended for all people living with HIV in Thailand. However, IPT is only provided by 20% of Thai physicians who care for persons with HIV.

In a poster discussion session, Nittaya Phanuphak of the Thai Red Cross AIDS Research Centre, Bangkok, Thailand, reported on the feasibility and benefits of providing IPT to HIV-positive persons in Thailand. This study was a four-year follow-up of patients in the Thai treatment program, MTCT Plus (MTCT+). Tuberculin skin testing (TST) is offered to all MTCT+ program patients at enrollment (if they had not had previous latent or active TB), and annually thereafter. Patients with latent TB (defined as a positive TST with a normal chest X-ray and no clinical symptoms of TB) are given a nine-month course of IPT.

A total of 1100 adult patients (290 men, 810 women) were enrolled in the study between February 2003 and January 2007 and followed until October 2007. Patients with active TB at enrolment (4.5%) or already receiving IPT (0.5%) were excluded, leaving the remaining 1011 patients in the analysis. The mean age was 29 years and median baseline CD4 count was 330 cells/mm3. One hundred and forty-five (15.7%) had been on antiretroviral therapy (ART) for at least three months at baseline, and 317 started ART thereafter.

Tuberculin skin test results

Tuberculin skin tests (TST) were performed in 924 of the 1011 study participants (91.4%). One hundred and forty-five (15.7%) baseline TSTs were positive. Positive results were associated with being on ART for three months or more (p=.018). There was also a borderline association for pregnancy status (p=0.049): pregnant women were less likely to have a positive TST than women who were not pregnant (odds ratio [OR] 0.5, 95% confidence interval [CI] 0.2 to 0.9).

Outcomes of isoniazid preventive therapy

IPT was completed in 151 (77.4%) of the latent TB cases, and showed a trend (not statistically significant) toward reducing active TB incidence. Active TB occurred in 1/151 (0.3 per 100 person-years) people who completed the course of IPT, versus 1/44 (1.5 per 100 person-years) non-completed IPT cases (relative risk, 3.4; 95% confidence interval [CI], 0.1 to 26.8), 6/533 (6.2 per 100 person-years) people with negative TSTs (relative risk, 1.7; 95% CI, 0.2 to 10.4), and 5/87 (38.0 per 100 person-years) people who were eligible but did not have a TST performed (relative risk, 8.7; 95% CI, 0.3 to 21.5).

Although the differences were not statistically significant due to the relatively low numbers of active TB cases in the study group, the researchers nevertheless stated that "isoniazid preventive therapy (IPT), even in non-completers, has a strong trend to reduce the incidence of active TB in tuberculin skin test-positive individuals," and that IPT "may benefit all HIV-infected individuals irrespective of TST results."


Diero L et al. . The experiences and outcomes of Isoniazid preventive therapy in an HIV treatment program in Western Kenya. AIDS 2008, Mexico City, abstract MOAB0306.

Phanuphak N et al. Incidences of latent and active tuberculosis among MTCT-Plus patients who received annual tuberculin skin test with isoniazid preventive therapy in Thailand: a 4-year follow up. Seventeenth International AIDS Conference, Mexico City, abstract MOPDB202, 2008.