Back to contents

HIV and TB in Practice

Theo Smart
Published: 21 May 2009

Are there indications of increased isoniazid resistance coming out of IPT programmes and studies?

Despite evidence showing that isoniazid preventive therapy (IPT) should markedly reduce TB-associated morbidity and mortality in people with HIV, many HIV programmes and healthcare providers remain reluctant to administer IPT because they fear that it could increase resistance to isoniazid (INH) among people with HIV who have active TB disease that goes undetected by symptom screening.

Reports that could further fuel the debate were presented at ‘the 3I’s Satellite Symposium’ sponsored by the Aurum Institute for Health Research and the Consortium to Respond Effectively to the AIDS TB Epidemic (CREATE), held just prior to the 4th South African AIDS Conference (SAAC) on March 31st. Several speakers reviewed the currently available IPT data.

New findings include preliminary data on INH resistance among some of the participants taking IPT in the large Thibela TB study; further evidence on the value of chest X-rays before going onto IPT, as well as discussions of other possible ways to increase case finding of active TB.

Published reports suggest that people who develop resistance to isoniazid alone should respond to standard four-drug therapy – but IPT programmes should implement good patient follow-up systems to make certain that these cases are diagnosed, and monitored closely.


At present there is very little evidence to suggest that IPT promotes drug-resistant disease, according to Professor Harry Hausler of the TB/HIV Care Association, who chaired the session on IPT at the symposium. He cited evidence from one 2006 meta-analysis that included over 18,000 people on IPT showing a slight increase in INH resistance, but the relative risk was fairly low (1.45, 95% CI 0.85-2.47).1 However, the authors concluded that ongoing surveillance was essential because it was impossible to exclude an increase in resistance with the rollout of large-scale IPT programmes being recommended.

“If you think about ART versus IPT, 10-20% of people who are HIV-positive are eligible for antiretroviral therapy, but close to 40% of people who are HIV-positive are eligible for IPT,” said Prof Hausler.  “So we should be looking at 6 million people on IPT globally rather than 29,000 [put on IPT last year].”

To date though, aside from the Botswana IPT programme, there have yet to be many large-scale programmes delivering IPT to people with HIV in high TB burden settings.

Botswana is undertaking countrywide surveillance for TB drug resistance but those data are not yet available. According to a couple of speakers at the symposium, recent studies have not reported much IPT resistance. For instance, in the Botswana IPT trial that is running concurrently with the IPT programme, only seven out of 1995 subjects taking six months of IPT have been diagnosed with active TB “but there has been no evidence of INH resistance,” Dr Tefera Agizew of Botswana’s Ministry of Health said at the symposium.

Similarly, Dr Neil Martinson of the Perinatal HIV Research Unit reported that IPT did not drive resistance in a study he conducted comparing several different TB preventive therapy regimens, including two IPT arms, (IPT for six months or continuous IPT). In a report earlier this year at the Conference on Retroviruses and Opportunistic Infections, there was no resistance in 14 out of 19 TB cases that occurred in 328 patients who had received six months of isoniazid (specimens were not available for five), and only one case of MDR-TB in the seven breakthrough cases out of 164 people taking INH continuously.

Thibela TB study

Thibela TB, however, is by far the largest study of IPT to date. It is a cluster-randomised trial to measure the impact of community-wide isoniazid preventive therapy (IPT) on TB incidence at the community level, in a population with a high prevalence of HIV. In the study, entire mine-shafts, rather than individuals, have been randomised to standard TB control with or without the addition of IPT. The total study population will include around 70,000 people – with about 38,000 on IPT.

Professor Gavin Churchyard presented some of the first resistance data from the study so far (see table). The impact of IPT on resistance is most apparent among first episodes of TB with mono-INH resistance, since resistance among retreatment cases could have been acquired during prior TB treatment.

Preliminary drug susceptibility of TB on IPT

Active TB

IPT (n=66)

Comparison (n=129)


First episodes, n=53

% (95% CI)

Retreatment, n=13

% (95% CI)

First episodes, n=97

% (95% CI)

Retreatment, n=32

% (95% CI)



13.2% (5.5-25.3)


7.4% (1.9-36.0)


8.2% (3.6-15.6)


25% (11.5-43.4)











“We would expect to see a higher proportion of INH resistance in people taking IPT,” said Prof Churchyard. “And although it is slightly higher in terms of INH resistance, it is not significantly higher.”

Indeed, the preliminary differences do not seem to be significant. But the fact that it is higher at all is worth noting, particularly because Thibela TB has one of the most rigorous screening processes for screening out active TB – including chest X-rays. Programmes that don’t do as good a job of screening for active TB before IPT may see more resistance. Furthermore, breakthrough cases were also more likely to be quickly diagnosed in Thibela, which may not be the case in the ‘real world.’

Given that IPT could still cut the number of active TB cases by 30 to 50%, a slight increase in resistance is still not a good excuse to prevent roll-out of IPT. However, it does suggest that more attention should be given to optimising the screening process – to reduce the number of active TB cases that are missed – and to the management of people with breakthrough disease.

Screening for active disease

“The difficulties screening for TB in HIV-infected patients are well known,” Dr Salome Charalambous of the Aurum Institute said at the symposium, and described several studies exploring optimal screening methods. Programmes should use such screens to identify people with symptoms since only ‘the well’ should be put on IPT. However, several studies have found that some people with HIV have no symptoms.

Dr Charalambous noted conflicting data on usefulness of chest X-rays. A pilot study for Botswana’s IPT programme suggested X-rays didn’t add much to screening, but representatives of the programme later reported subsequent experience that X-rays do indeed pick-up some cases of TB in asymptomatic patients. Dr Charalambous described one study in South African gold miners, where adding chest X-ray to a symptom-screening tool (night sweats, cough and weight loss) improved sensitivity from around 60 to 90%. 

According to preliminary results from another study, 381 patients at Tshepong Wellness Clinic in the Northwest Province were screened for active TB prior to starting ART. TB was diagnosed in 31.6%.

“But using a symptom screen alone would miss one quarter of the TB cases, while chest radiography improved sensitivity substantially,” said Dr Charalambous. “One thing I’d like to emphasize, is that [in this study] 62.5% of patients had TB symptoms, so even if we are just screening with symptoms, we would still have to further screen these patients with sputum microscopy and culture – which would lead to overstretching our laboratory facilities.”

But in Thibela, even more cases would have been missed if they had relied on symptom screening alone, as more than half, 54.6% of the TB cases detected, were asymptomatic. “In our setting, X-ray detected a significant proportion of all TB,” said Prof Churchyard.

Dr Martinson stressed that it was very important to exclude active disease – but he was not certain that chest X-rays would really be appropriate in the field.

“I would recommend not doing a chest X-ray. I think it’s really an excuse just to keep people away from receiving isoniazid preventive treatment, especially when we consider that IPT is meant for well people,” he said. “Clearly asymptomatic TB in HIV-infected individuals is a concern, but my experience is that TB is a fairly malignant disease in people who are HIV-infected. If you have at least two visits, one month apart, before giving IPT, most people who are HIV-infected who have got active TB are not going to be walking around feeling good for more than a month.” However, he suggested an additional option “Anyone who is HIV-infected should have a TB culture,” he said.

“Active screening with culture detects even more disease,” said Dr Stephen Lawn of the Desmond Tutu HIV Centre. In another study of 235 people enrolling into the ART programme, Dr Lawn and colleagues found that 25% (n=58) had culture positive TB.2 But in the absence of culture (specimens were sent to the lab for every participant), it would have been difficult to identify everyone with TB – 22% of those with culture positive TB had not TB symptoms (any one or more of cough, fever, night sweats or weight loss) and 30% had no sign of TB on chest X-ray.

Dr Lawn said that despite having no symptoms, it was possible to obtain induced sputum from people with HIV and get a diagnosis.

Unfortunately, in most settings, it takes weeks to get culture results back – and the capacity to do culture in everyone with HIV simply does not exist.

“We must be careful not to overstretch our laboratories, and with chest X-rays, there are also issues with access and interpretation,” said Dr Charalambous.

If anything, these observations once again underscore the need for an inexpensive and highly sensitive point-of-care diagnostic.

In the meantime, programmes rolling out IPT should make certain that they have good patient follow-up mechanisms, so that active cases that are missed and breakthrough cases on IPT are detected quickly and put on effective treatment.

Managing breakthrough disease

Published data suggest standard treatment ought to work.

“When active TB occurs among those given IPT, standard four-drug first-line therapy works,” said Professor Hausler citing two studies.3,4 However, the number of patients involved in these studies was not great – and mostly they were not performed in resource-constrained settings with a high burden of HIV and without good systems for patient adherence and follow-up.

It will be especially important for programmes to closely monitor treatment in patients with mono-INH resistance, because there is a chance that less than optimal responses could lead to resistance to other TB drugs as well. 

Operational plans for IPT programmes should include systems to document responses to standard TB therapy in people with HIV and INH-resistant TB. Another possibility would be to design studies to investigate what the best treatment options for these cases would be, comparing outcomes on standard therapy to those when people are given prolonged treatment, higher doses of INH (16-20 mg/kg) (several studies suggest that high doses remain active against the low level INH resistance most commonly be seen in first episodes of TB) or 5-drug regimens. In the end, more aggressive regimens may not be necessary, but it would be prudent to start asking these questions sooner than later.


Lois Eldred, Project Director, CREATE, Johns Hopkins University Center for Tuberculosis Research, Haileyesus Getahun, Stop TB department, WHO.


[1] Balcells ME et al. Isoniazid preventive therapy and risk for resistant tuberculosis. Emerg Infect Dis May;12(5):744-51, 2006.

[2] Edwards D et al. Baseline Screening for TB among Patients Enrolling in an ART Service in South Africa. 16th Conference on Retroviruses and Opportunistic Infections, Montreal, abstract 780, 2009.

[3] Mitchison DA, Nunn AJ. Influence of initial drug resistance on the response to short-course chemotherapy of pulmonary tuberculosis. Am Rev Respir Dis Mar;133(3):423-30, 1986.

[4] Nolan CM, Goldberg SV. Treatment of isoniazid-resistant tuberculosis with isoniazid, rifampin, ethambutol, and pyrazinamide for 6 months. Int J Tuberc Lung Dis, Nov;6(11):952-8, 2002.

This content was checked for accuracy at the time it was written. It may have been superseded by more recent developments. NAM recommends checking whether this is the most current information when making decisions that may affect your health.
Community Consensus Statement on Access to HIV Treatment and its Use for Prevention

Together, we can make it happen

We can end HIV soon if people have equal access to HIV drugs as treatment and as PrEP, and have free choice over whether to take them.

Launched today, the Community Consensus Statement is a basic set of principles aimed at making sure that happens.

The Community Consensus Statement is a joint initiative of AVAC, EATG, MSMGF, GNP+, HIV i-Base, the International HIV/AIDS Alliance, ITPC and NAM/aidsmap

This content was checked for accuracy at the time it was written. It may have been superseded by more recent developments. NAM recommends checking whether this is the most current information when making decisions that may affect your health.

NAM’s information is intended to support, rather than replace, consultation with a healthcare professional. Talk to your doctor or another member of your healthcare team for advice tailored to your situation.