Bangladesh MDR TB treatment regimen

What is the Bangladesh regimen?

TB clinic in Bangladesh where the shorter TB regimen was developed © Damien Foundation

TB clinic in Bangladesh where the shorter TB regimen was developed © Damien Foundation

The Bangladesh regimen is a short course 9 month regimen, for the treatment of “uncomplicated MDR TB”. “Uncomplicated MDR TB” is TB where the bacteria are only resistant to the first line drugs rifampicin and isoniazid.

The Bangladesh regimen, together with the STREAM trial, are largely now of historical interest. But both sets of trials were important in the development of shorter, all oral regimens, for the treatment of drug resistant TB.

The standard treatment for multi drug resistant TB used to require the use of “reserve” or second line TB drugs, and the drugs had to be taken for 24 months. The Bangladesh regimen only required treatment for nine months.

MDR treatment success

Success rates for the treatment of MDR-TB were until recently very low. Two of the main reasons for this were believed to be the length of the treatment regimen, and the side effects of many of the second line TB drugs.

Bangladesh studies

The search for a shorter course of drugs started some years ago, and included a series of studies carried out by the Damien Foundation. The series started in 1997 and consisted of six prospective observational cohort studies over a period of twelve years. The aim was to see if the two year duration of MDR-TB treatment could be shortened using novel combinations of existing drugs.

The studies were carried out in Bangladesh and the first regimen was largely based on the 1996 WHO guidelines. Periodic assessment of treatment outcomes, relapses and adverse drug reactions was used to guide the introduction of changes for subsequent cohorts. The most promising results were seen with the sixth treatment regimen. This was a group of 206 patients treated for a total of 9 to 12 months. The duration of the intensive phase was dependent on the response at four months to smear microscopy. Overall the Damien Foundation studies treated 515 patients between 2005 and 2011.

Drugs used for the Bangladesh regimen

The Bangladesh regimen included a drug, kanamycin, which had to be painfully injected every day © David Rochkind

The Bangladesh regimen includes a drug, kanamycin, which has to be painfully injected every day © David Rochkind

The aim was to develop an effective, safe and inexpensive treatment regimen, and the drugs to be used were chosen with extreme care.  So drugs such as PAS and cycloserine, whose activity is offset by toxicity and poor tolerability, were therefore avoided. More active drugs that are also difficult to tolerate, such as prothionamide, were limited to the intensive phase.

These drugs were added to a core of selected first line drugs that might be expected to work in at least some patients: ethambutol, isoniazid and pyrazinamide. High doses of certain drugs were used in an effort to maximize effectiveness. Other drugs such as high dose moxifloxacin, and a moderately high isoniazid dose were chosen to help overcome the selection of resistance. Clofazimine was chosen for its lack of known serious toxicity.

An additional over riding objective was to avoid amplification of resistance and the creation of XDR-TB. XDR-TB is TB that has developed resistance to rifampicin and isoniazid as well as to at least one of the second-line anti-TB injectable drugs kanamycin, capreomycin or amikacin and also to at least one of the fluoroquinolones. High cure rates of MDR-TB were achieved only with the introduction of the fluoroquinolones. So the second line injectable drugs were not considered as core drugs, but only as powerful companion drugs protecting the fluoroquinolones.

So these and other considerations led to seven drugs being used: gatifloxacin, clofazimine, ethambutol, and pyrazinamide. This was supplemented by high dose isoniazid, kanamycin and prothionamide during the first four months.

The results of the Bangladesh studies

The result of the sixth and most successful regimen, was that 87.9% of patients completed treatment without relapse. This was a remarkable outcome when MDR-TB cure rates usually range from 11 to 79%. There was great optimism about the results with it being said that

“We had hoped to save the lives of a few patients, but now it seems we have found the treatment regimen needed to control the growing MDR-TB epidemic. This is so long as the bacilli remain sensitive to its core drugs” Dr Armand Van Deun  1

However, when the results were published in 2010 under the title “Short, Highly Effective and Inexpensive Standardized Treatment of MDR-TB there was criticism as to whether it was a “highly effective standard” given that the study was observational. 2

It was also said that there were major flaws in the study design. Potential flaws in the Bangladesh regimen study included the fact that:

  1. It was an observational study and many patients were excluded. More eligible patients chose not to participate than to participate (578 vs. 486);
  2. No patients with HIV were included;
  3. Patients were treated in Damien Foundation centres with consequent individual & family support;
  4. Cohorts were enrolled sequentially, meaning that the various regimens were studied during different periods of time. In addition cohort sizes were not predefined.

Assessing the regimen in other settings

The promising results from Bangladesh prompted a 2008 external review of the Bangladesh study, led by the WHO.3 This review concluded that additional evidence from larger studies was needed in order to determine whether scale-up of a nine-month regimen could be recommended in less controlled conditions.

In particular many people considered that there was a need for a larger multinational trial to be carried out before the Bangladesh regimen could be widely adopted. And the review also concluded that this should ideally be done in the form of a randomised control trial. It was also necessary to prove that the regimen could be equally effective when used to treat patient populations elsewhere, particularly in countries with high rates of TB drug resistance and high rates of TB-HIV co-infection.

New Recommendations from WHO

In 2016 WHO recommended the shortened nine month Bangladesh treatment regimen for MDR-TB patients.4

West & Central Africa Francophone Study

After the Bangladesh studies took place, the first multi country MDR TB trial using the nine months regimen was one taking place in nine countries. The countries were the French speaking Benin, Burkina-Faso, Burundi, Cameroon, Cote d’Ivoire, Central African Republic, Niger, Democratic Republic of Congo and Rwanda.

Initial results showed that among a cohort of 507 adult patients, 80.9% had treatment success, 7.7% died, 6.5% were lost to follow-up and 4.9% were treatment failures.5

It was said that:

“These preliminary results from using a 9 month MDR-TB treatment regimen are excellent. Implementing the shortened regimen is proving feasible and with improved outcomes compared with the standard MDR-TB treatment regimen. Dr Arnaud Trebucq

However, it was still the case that no randomised control trial had been done.

STREAM trial

The STREAM study (Standardised Treatment Regimen of Anti-Tuberculosis Drugs for Patients with MDR TB) was a subsequent study supported by USAID. STREAM was a multi centre international randomized control trial which had the aim of evaluating shortened regimens for patients with MDR TB.

STREAM was originally designed to validate a modified version of the Bangladesh regimen, by comparing it with the current WHO standard of care, under the more stringent criteria of a randomized controlled trial. The modification of the regimen was the replacement of gatifloxacin with moxifloxacin.

The first trial was started in Ethiopia, South Africa, Vietnam and Mongolia. On 30th June 2015 the trial enrolled its 424th patient, the last patient to be recruited for phase 1 of STREAM.

Modification of STREAM

The STREAM study has since been expanded to test two additional shortened treatment regimens using the new TB drug bedaquiline. So the STREAM study will now consist of two stages.

The first stage will evaluate the non inferiority of the Bangladesh regimen when compared to the WHO standard of care. That is, is the Bangladesh regimen at least as good or indeed better than the current WHO standard of care?

The second stage will assess whether a nine month bedaquiline containing regimen is better than the Bangladesh regimen. It will also look at whether a six month bedaquiline containing regimen is at least as good as or is better than the Bangladesh regimen. The nine month regimen was to be an all oral regimen that did not require painful injections, and the six month regimen was a simplified regimen.

TB drugs being sent out for use in the STREAM study

TB drugs being sent out for use in the STREAM study

Janssen the manufacturer of bedaquiline suggested that this second stage of STREAM would replace a separate phase 3 study of bedaquiline. It would also test whether the addition of bedaquiline could improve the Bangladesh regimen by replacing the injectable drug kanamycin. It would also establish if the duration of MDR-TB treatment could be reduced to just six months with kanamycin being given for just two months. It would also obtain important safety information about bedaquiline.

However, these proposals resulted in a number of ethical and scientific concerns being raised. Following a number of discussions it was agreed in December 2014, to alter the design of STREAM stage 2. So if Stage 1 fails to show that the modified Bangladesh regimen is at least as effective as the WHO standard of care, then bedaquiline containing regimens can still be compared to the current WHO standard of care in Stage 2.

In March 2016 stage 2 of STREAM began enrolling a total of 1,155 patients. The results of the study are expected in 2021. By November 2016 three countries were recruiting and 28 participants had been enrolled in the study. But it was not possible to say that the nine month regimen was as good as or equal (i.e. not inferior) to the longer regimen.6

STREAM stage 1 results

In October 2017 preliminary results were reported of the first phase of the STREAM trial. It was found that the nine month regimen is very close to the effectiveness of the standard 20 month regimen recommended by WHO.

STREAM trial final results

In early 2019 the final results from the first stage of the STREAM trial showed that a 9-11 month treatment regimen is as effective in treating MDR-TB (when given under trial conditions) as the 20-24 month treatment.

Page Reviewed

This page was last reviewed in May 2021.
Author Annabel Kanabus

Social Media

If you have found this page useful please tell other people about and if you have a website please link to us at

  1. “World Health Organization recommends shortened nine-month treatment regimen for MDR-TB patients, The Union,[
  2. Van Deun, A et. al. “Short, highly effective and inexpensive standardized treatment of multidrug-resistant tuberculosis, Am J Respir Crit Care Med 2010; 182:684-692
  3. Nunn, A et. al. “Evaluation of a standardized treatment regimen of anti-tuberculosis drugs for patients with multi-drug-resistant tuberculosis (STREAM): study protocol for a randomized controlled trial, Trials 2014 ; 15:353
  4. "World Health Organisation recommends shortened nine-month treatment regimen for MDR-TB patients", 2016,[]
  5. “New cohort study shows promise of dramatically shorter treatment option for multidrug-resistant TB, 46th Union Conference on Lung Health,
  6. “Preliminary results from STREAM trial provide insight into shorter treatment for multidrug-resistant tuberculosis”,

Would you like to send us a comment about this page?

You might like to read more about: