HIV and TB in Practice: encouraging data on IPT

This regular feature on HIV/TB integration is kindly supported by the Stop TB Department of the World Health Organization.

Isoniazid preventive therapy (IPT) to prevent active tuberculosis (TB) can be successfully delivered in two different — though limited — clinical contexts in sub-Saharan Africa, according to two studies presented at the HIV Implementers’ Meeting in Windhoek, Namibia, in June.

In the first study, Dr Gilbert Tene of the International Center for AIDS Care and Treatment Programs (ICAP) in Rwanda, described how the country has systematically scaled up the provision of IPT to the childhood contacts of adults with active TB. Meanwhile, another ICAP-supported project in Mozambique that is providing IPT to people with HIV attending an urban ART clinic, found that while adherence to IPT was initially poor, adherence to clinical visits could be dramatically improved with strengthened counselling.

Although neither project addressed outcome data such as a reduction in the burden of TB, and falls short of offering IPT for all HIV-positive people without active TB, they did provide some useful information about how to start to put an IPT programme in place. Even so, how and whether to prioritise IPT delivery as a service for people with HIV in resource-constrained settings was a much debated topic at the conference, with one leading South African clinician even asking whether IPT is really worth all the effort.

Background on IPT

Taking a course of IPT as directed reduces the risk of TB in people with HIV by about a third (and 64% in those known to have been exposed to TB) but the benefit may only last a couple of years in settings where people are frequently re-exposed to TB.

Since 1998, the WHO and UNAIDS have recommended six to nine months of isoniazid preventive therapy (IPT) as part of the essential package of care for people living with HIV — once active TB has been safely excluded. According to the policy, information about IPT should be made available to everyone living with HIV. The policy does not require that people have a tuberculin skin test (TST) to show whether they have been exposed to TB, but it does recommend a mandatory chest x-ray for TB.

More recently, the Core Group of the TB/HIV Working Group of the Stop TB Partnership began to strongly advocate for the roll-out of IPT as a public health intervention for people with HIV. And at the 3 I’s Meeting last year, it was recommended that the WHO release a new clearer policy statement on IPT, which would drop the requirement for a chest x-ray, and co-package IPT with intensified case finding (ICF) (or TB screening) as a part of the same intervention. The idea is that all people with HIV should be routinely screened for TB and that anyone with HIV who is not identified as a TB suspect would immediately be offered IPT. However, a formal policy is yet to be released, and as the last HIV and TB column reported, debate persists about whether chest x-ray is a necessary screening tool to exclude TB for IPT programmes.

Countries have responded to WHO policy by adopting IPT as policy, but most have been reluctant to scale up implementation because of the absence of successful models, lack of operational guidance and other concerns.

“There are many challenges with IPT. It is not yet part of many national programmes,” said Wafaa El-Sadr of ICAP at the Implementers’ Meeting.  She noted concerns including its effectiveness, the need to provide adherence support, the difficulty to exclude TB disease in people with HIV and the chance it might lead to isoniazid (INH) resistance if active TB cases are treated with mono-TB therapy.

But even before IPT was recommended for people with HIV, it was widely recommended for child contacts of TB cases. Yet very few countries in resource-constrained settings actually practice this intervention. In fact, a couple of years ago at the Union World Conference on Lung Health in Paris, Dr Hans Rieder, a leading TB expert, said that programmes shouldn’t even think of offering IPT to adults with HIV until they had begun to provide IPT to child contacts of known pulmonary TB cases:

“I tell the national programme, implement first where it is simple, and if that works we go to the next [stage]... For example, we recommend preventive therapy for children under the age of five who live in the same household as a newly diagnosed smear-positive case. That is the easiest group to do. Everybody agrees, it’s simple, you give it to the patient, who gives it to the kid. Toxicity is minimal, they are under five, even if you miss a lymphadenopathy it’s safe. It’s in every manual of the national TB programmes in Africa — I’ve seen it — but I have not seen a single country where they have systematically implemented it,” he said.

Which is why the report from Rwanda is so significant, because it clearly can be done.

Scaling up IPT for kids in Rwanda

Children under five years of age are at a much greater risk of developing tuberculosis, especially those with HIV. Globally, there are about one million paediatric TB cases each year, and half of them lead to death.

Rwanda is a country of 10 million, 83% of the population is rural and 57% lives below the poverty line. The pulmonary TB case detection rate is relatively low (compared to southern African countries) at about 48 per 100,000, in 2008. In the same year, there were 349 cases in children (4.4% of the total TB cases). About a third of Rwanda’s TB cases are also HIV-infected.

“The Rwandan Ministry of Health (MOH) has included IPT in the national TB guidelines for many years but until 2006, this intervention had not been systematically implemented,” said Dr Tene.

In 2005, the TB unit of the MOH collaborated with the Association of Pediatricians in Rwanda, ICAP and other international partners to revise the TB programme guidelines and training materials to include a chapter on paediatric TB, contact tracing and IPT. Tools were also revised to include information on IPT (including TB treatment cards, IPT registers, quarterly report forms) and information education and communication (IEC) materials were developed and distributed to the districts.¹

Contact tracing was done through actively asking adult TB patients about their contacts and inviting them in for consultation, and through home visits to smear-positive pulmonary TB patients by the TB nurse, and referral of any children under five to the local health facility.

At the health facility, an algorithm was designed for nurses to identify and diagnose latent TB infection in these children based on a symptom-screen for TB and physical examination. This algorithm does not include systematic chest X-ray nor tuberculosis skin testing (TSTs), unless the child has symptoms suggesting that further evaluation is necessary to exclude active TB disease.

The algorithm asks whether the child has had fever or cough for more than 15 days (in spite of an empiric course of antibiotics, and if malaria has been excluded). TB treatment is started in children with persistent cough or fever if they are smear-positive, have an x-ray suggestive of TB or a positive TST. Children with persistent cough or fever but without these signs of TB might also be started on a full course of TB treatment if they are both household contacts of a pulmonary TB patient and HIV-infected or suffering from severe malnutrition.

Any child contacts of a pulmonary TB patient who are found to be free of signs and symptoms suggestive of TB or in whom active TB has been ruled out are treated with IPT for six months. IPT is given daily to the child by their parent, with monthly follow-up at the health facilities. Any child who has an HIV- infected parent is also offered an HIV test and if infected, enrolled for HIV care and treatment.

To ensure that IPT is scaled up country-wide, a TB training curriculum was developed and implemented at all districts between 2006 and 2008. The national programme officers provided intensive supervision and mentorship to districts and sites nationally, with quarterly evaluation meetings at the district level.

Rwandan results

Since 2006, all 187 of the country’s TB Detection and Treatment Centres and all 30 districts have begun implementing IPT for kids and are collecting and reporting IPT data. IPT uptake has increased nationwide from 815 children in 2006 to 1507 in 2007 and 1349 in 2008. In 2007 and 2008, 14-15% of the parents were found to be HIV-positive — uptake of HIV testing among the children was very high (around 97%). In 2008, 38% (513/1349) of children overall on IPT were tested for HIV and 6% (32/513) of those tested were found to be infected.

Dr Tene stressed that outreach has to be “constantly improved” to be certain that they are reaching all the children who need IPT. The country also needs to collect data on adherence and completion of IPT, and to assess the impact of the programme.

“In Rwanda, national implementation of IPT for childhood contacts of adults with smear-positive TB is feasible,” he concluded. “Nevertheless, further evaluation is required to know better its impact on childhood mortality and morbidity.”

Even so, getting a nationwide programme off the ground is no small feat.

“The implementation of IPT for child contacts globally has been very very dismal, so hearing this presentation is very encouraging,” said Dr Haileyesus Getahun of WHO’s Stop TB Department.

However, Rwanda is still resisting offering IPT to adults with HIV.

Mozambique starts offering IPT to patients on ART

Dr Anna Scardigli, also with ICAP, described some of the challenges to beginning an IPT programme in Mozambique. The country has a high HIV prevalence (~16%) (though it is around 23% in Maputo), a high TB incidence: 431 cases per 100,000 population and a high burden of co-infection — about 60% of TB patients tested for HIV in 2008 were HIV-positive. Mozambique began to scale up TB/HIV collaborative activities in 2006. Last year, it began to pilot an IPT programme.²

“There were various constraints getting the programme started, first of all the availability of isoniazid as a single drug (rather than coformulated with other TB drugs), and then the coordination between the TB services and ART facilities,” she said. TB services and ART facilities are often physically separated, with the TB service storing isoniazid. On top of this, there were the traditional concerns about the ability to rule out active TB and that well patients would not be adherent to IPT.

“As most PLWH who start IPT are in good health and do not require frequent visits, strategies to improve adherence to monthly follow-up visits for a preventive therapy are needed,” said Dr Scardigli.

They decided to limit the size of the initial programme to the amount of isoniazid that was available, and to store it at the pharmacy of the health unit. But the programme was first piloted in one ART facility (at Mavalane Hospital in Maputo) supported by ICAP after using a TB screening checklist that ICAP had also helped develop.

In July 2008, a workshop was held for all 20 staff members at the ART clinic that focused on IPT eligibility (how to rule out active TB), isoniazid delivery, IPT register completion and follow-up of patients. The nurse designated as the TB/HIV focal point was responsible for coordination of the IPT programme.

The clinic had already had experience using the symptom checklist to screen for TB in all the patients enrolled at the ART facility. With the launch of the IPT programme, doctors began prescribing IPT for all eligible patients who screened negative for TB, and monthly follow-up was performed by nurses. Patients were told to return to the clinic monthly for follow-up visits and to pick up their next month’s supply of isoniazid (rather than for their next ART consultation).

Almost all of the patients had stage one or two HIV disease. All were on ART (which is because doctors who prescribe isoniazid only see patients on ART rather than pre-ART patients).

During the first six months of implementation, 109 HIV patients initiated IPT. No patient discontinued due to toxicity, and only two quit drug because of pregnancy.

But as Dr Scardigli anticipated, adherence was not great. After the first month, only 13/34 (38%) patients who had started IPT returned for their follow-up visit and to get their supply of isoniazid.

When many of the patients who defaulted on IPT returned for their ART consultation, they reported that they had ‘forgotten’ about their IPT appointment. Dr Scardigli said that one possible reason for this was that the ART consultation was with the clinician, but the IPT consultation was with a nurse in another room.

So they decided to strengthen the counselling at IPT initiation to emphasise the importance of adherence. In addition, IPT was also recorded on the patient card and file envelope, in addition to the IPT register, so that IPT patients could be tracked better and more healthcare staff such as the receptionist and pharmacist could be involved in patient follow-up and remind the patient of the importance of adhering to the programme.

Gradually, adherence improved. By month three, 78% returned for their follow-up visits — although about 42% came in about a week late. The team worked more on strengthening counselling and staff commitment to the programme. By month six, the proportion returning for follow-up reached 91% (99/109) and only 32% came in somewhat late. This was more likely to occur when they had another hospital consultation scheduled soon afterwards

Overall, 92 (84.4%) are believed to have completed six months of IPT. There were 15 (13.8%) who were defaulters to some point of follow-up.

“So IPT implementation in an ART facility (or HIV/AIDS consultation) is feasible, but commitment of the whole staff involved in the care of the patient — and not just one person — is needed to increase adherence and keep better track of the patient, and of course intensive counselling and education of patients especially prior to initiating IPT,” she said.

She said that getting patients to come back to the first follow-up visit is the most challenging, but after that point they rarely miss further visits. Even so she stressed that “combining IPT visits with other visits and ART pick-up can reduce missed visits and avoid delays.”

They now plan on expanding TB screening in all people with HIV (at all entry points to care). They plan to strengthen follow-up even beyond six months on IPT to evaluate the effectiveness of the programme.

During the question and answer session, Dr Scardigli was asked about the timing of starting IPT in people on ART.  “It’s important to consider the risk of interaction between ART and IPT and the risk of toxicity, “ she said, “and recent studies suggest waiting about four months on ART before starting IPT. In our case, most of the patients had already been on ART for a long time. But it was the clinician’s decision rather than a criteria. Now they are thinking to give a clear recommendation to wait for six months before starting on IPT.“

She was also asked about the limited roll-out of the programme at one ART facility.

“The TB screening tool is now used in all the supported facilities,” she said, “but IPT has only been started in one facility which was also the first to begin using the TB screening tool. I think it is important because we can’t just start to implement IPT at any site. We had to select sites with the appropriate condition, both in terms of human resources and in the uptake of the TB screening tool. There are provinces in Mozambique where they have decided that they won’t start IPT yet, because they want to make sure that they are doing better at TB screening before starting IPT.”

Limited scale up of IPT in other settings

Indeed this continues to be the pattern in most countries.

“There are still restrictive national policies on IPT in many countries,” said Dr Getahun during an informal session on TB/HIV. “Although IPT is policy in many countries, the countries that actually report providing it are very, very few.”

He presented an example from Uganda that he said “Suggests that the only institutions that should provide IPT should be very high tech; it should only be given by medical doctors; there should be laboratory assistants, trained counsellors, pharmacy technicians, etc, so it really makes it a highly sophisticated intervention.” 

Other key issues

If an organisation has a TB default rate of greater than 5%, it will not be eligible to provide IPT

Other speakers at the meeting described similar constraints on rolling out IPT.

Dr Endris Mohammed, described the experience in Ethiopia, which began TB/HIV collaboration activities in earnest in 2004.

“To decrease the burden of TB in PLHIV: the 3 I’s [(IPT, Intensive Case Finding (ICF) and infection control] are being practiced. All patients with HIV are screened for TB using a screening tool during each visit, and those who are not showing signs and symptoms of TB can be provided with INH prophylaxis,” he said. But although monitoring and evaluation data show that screening and diagnosis of TB has improved over the last few years — the number put on IPT nevertheless remains low (and even seems to be falling).

“There is a low uptake of IPT. IPT is mostly provided at hospitals, it is not provided in health centres,” he said, though his slides indicated that IPT provision requires a chest x-ray to rule out active TB. “There should be clear guidance from WHO on how to rule out active TB and in whom to initiate IPT — that has been the bottleneck to scale up IPT.”

Dr Fadare Omoniyi from WHO said that in Nigeria, they had concluded there was little point in scaling up IPT “to protect somebody from getting TB when we are still exposing the patient through poor infection control to TB. So one of the criteria we have put in place is that before any facility can expand to offering IPT, there must be TB infection control measures in place. We really must do it comprehensively so that patients are not exposed to TB.”

Kenya is one of the most progressive countries in terms of implementing TB/HIV collaborative activities, and at the Implementers’ Meeting this year, Dr Bernard Langat of the Division of Leprosy TB and Lung Disease in Kenya described how the country has started implementing the 3 I’s to reduce the burden of TB among people with HIV.

But as in many other countries, the IPT policy is limited to the provision of IPT to children under five years of age in contact with someone with pulmonary TB, to research settings and to facilities with the capacity to offer adherence support and to rule out active TB (comprehensive care clinics with adequate capacity to diagnose TB and provide follow-up).

“IPT scale up calls for caution and adequate resource allocation. If you are going to start on isoniazid preventive therapy (IPT), you should be able to follow that patient until he completes it, and in case he fails, you should be able to start him on TB treatment,” said Dr Langat.

Recently there has been some expansion in IPT access in Kenya, however. Two HIV programme partners currently implementing IPT in 23 sites, in over 12,000 clients, are reporting treatment completion rates of 70%. “These are programmes that have resources for defaulter tracking and screening [including chest x-ray],” he said.

There is also an ongoing national IPT study looking at providing IPT to all the household contacts of pulmonary TB index cases at the community level.

But in addition to resources to track defaulters, he believes that the capacity to diagnose TB in people with HIV has to be improved.

Of note, Dr Amy Bloom of USAID sounded a similar note of concern and said that PEPFAR was prioritising support to intensified case finding (ICF) and laboratory strengthening for TB/HIV.

“There’s been a lot of talk about IPT. There have been a lot of pilot programmes, some of which have been more successful than others, but I don’t think that many of them have been that successful,” she said. “There are a number of countries who are concerned about implementing IPT because of very real fears that their labs and screening won’t pick up active cases, which makes it very hard to say, ‘oh, we’re going to be giving single drug therapy to people who may have active disease.’ These are very real concerns, so it really behoves us to make sure that we have really good intensified case finding from the clinical and laboratory side, so that we can move forward on things like IPT.  That’s not to say that you have to wait and wait before doing IPT, but we have to think about these things.”

Is IPT really worth all the trouble?

Dr Francois Venter of Johannesburg Hospital said he thinks that there has been a lack of critical thinking about IPT.

“Why is it that IPT is reaching so few people with HIV? Is it maybe because it was a badly thought-through programme?” he said. “One of the major themes of this meeting has been making sure that we get the most bang for our bucks in our programmes, and I have to ask about IPT, what is the bang for your buck?”

Dr Venter described how his unit developed an IPT programme and implemented it in a region of Johannesburg.

“I’ve realised that when it comes to TB activities, if my academic unit, a group of very dedicated people, don’t do it, it just doesn’t get done. We screened I don’t know how many people with HIV for TB to get people onto IPT, and we finally got a thousand people on it. Almost every single person on IPT in the area was in the four clinics that we worked in. Nobody else did it; it was only us. And the experience that I have had is that it is always these dedicated PEPFAR-funded projects that are doing all the work to get people onto IPT in these tiny programmes. We designed these really pretty posters, piloted ICF material, trained until we were blue in the face. But, if you ask me how much benefit it is, I’m not sure.”

He also noted that there is still a huge amount of resistance on the part of the nurses, and said that the minute his unit stops supporting the nurse, “I guarantee they will stop giving IPT.”

“We worked out that we had to screen 40 people with HIV before we put one on IPT — not stopped an infection or finished IPT, but merely to put one person on IPT.  You might argue that you got all those people screened and into the system: I can do that in my sleep with a much more cost-effective programme.”

IPT should be given to the well — people without symptoms or signs of TB. But Dr Venter isn’t sure how programmes can really keep these people in care:

“Our wellness programmes are so shocking, not just in developing countries but in developed countries. We cannot retain people in the system unless we have them on antiretrovirals. We all know that. But we take these people, we test them, and we try to insert them into a system where all we are offering them is INH prophylaxis. But outside of very highly specialised programmes, every single large-scale programme I see seems to haemorrhage people out of the system if you don’t give them antiretrovirals. We need to engage with that before we start offering INH in my opinion. And the reason is because these programmes do not sit naturally within the healthcare system. Healthy HIV positive people do not have an easy place to go within the public healthcare system.”

Dr Venter believes that operational interventions and greatly intensified case finding may work better by getting people with HIV and TB into care sooner.

“You don’t have to convince me of the science, you have to convince me that putting in the amount of effort into trying to put people onto IPT is actually worth it. If we put a whole lot of counsellors into the front in casualty and just pulled out all of the coughing patients and got their sputums to the lab quicker, I bet you that would be money better used,” he said.

His comments during the informal TB/HIV session caused considerable controversy among the participants.

“When it comes to the 3 I’s, focusing on the 3 I’s is important from the patient’s point of view, and also from the health system point of view,” said Dr Yared Kebede Haile of TB CAB — who stressed that people with HIV are at very high risk of dying of TB before receiving a diagnosis. “If a family member of mine is HIV-positive, and they have been exposed to TB, but don’t have active disease, I would put that family member on IPT, and I think everyone would do that. So that’s really a requirement from the patient point of view — it is an important strategy to prevent the development of TB. Now how can we translate that into programmes? “

“The major problem with IPT, is that it is being planned to be delivered as an isolated intervention. And that’s wrong. It needs to be an integral part of TB control,” he said.

Dr Christian Gunneberg of WHO said that one of the key reasons to stress IPT is to get nurses to routinely screen for TB.

“A positive example from Botswana’s national IPT programme: the IPT programme there led to HIV-positive people who were just diagnosed being immediately screened for TB — and if they were referred, those with TB being immediately put on TB treatment. IPT cannot be disconnected from ICF; IPT is just an extension of case detection in people with HIV, particularly in southern Africa. We have to do IPT in order to encourage intensified case finding,” he said.

Concerns may only be allayed once programme data from Botswana — where chest x-rays are not required for IPT —and other settings begin to report on the effectiveness of the IPT programmes in field settings as well as whether they have led to an increase in isoniazid resistance.

More data should be coming soon. Despite the persistent concerns, the number of countries reporting data on IPT to the WHO has quadrupled in the last four years, and within the last year, there has been a dramatic increase in the number of people with HIV put on IPT (outside of Botswana).

Dr El-Sadr said that she was beginning to be more optimistic about IPT — and that even though they may be a lot of work, the importance of small vanguard programmes may be greater than is immediately apparent.

“I do see some rays of hope that IPT may become an option, she said. “Finally, I think that the passion of people who are really committed to HIV/TB, establishing these programmes and really seeking quality is pushing the envelope to strengthen components of the system. It actually may pull up the rest of the healthcare system along the way.”