March 23, 2009
Tuberculosis under control? Definitely not
Many people in rich countries think of tuberculosis as a disease of the past. Indeed well until the 1980s, experts thought that tuberculosis (TB) could be eliminated in a matter of decades.1 TB was seemingly under control.
Antibiotics, developed from the 1940s on, appeared to be effective in treating the disease. And as long as the strategy introduced by the World Health Organization (WHO), known as DOTS (Directly Observed Treatment, Short-Course), was correctly and efficiently rolled out, policy makers were convinced that TB would one day be a scourge of the past.
But now the international community recognizes that with around nine million new cases appearing every year, TB is far from defeated. TB is a deadly killer, responsible for 1.7 million deaths in 2006 – that is almost four lives claimed every single minute. The vast majority of TB cases occur in developing countries, with 22 high burden countries (mostly low and middle income countries) carrying approximately 80 percent of the global burden. Two billion people – one third of the world’s population – are infected and carry the tubercle bacillus.2
Worse, this ‘disease of the past’ has returned with new faces that are stretching our capacities to breaking point. The rapid spread of TB among people living with HIV, coupled with the emergence and spread of strains of TB that are resistant to the most common and effective drugs used to treat the disease, have led to a situation where far from being contained, TB is in fact spiraling out of control.
After calling victory too early, the world is waking up to the fact that TB has re-emerged as a major threat to global health.
Fueled by the HIV pandemic
People living with HIV are particularly vulnerable to developing TB. For that reason, TB incidence rates have shot up dramatically in the wake of the HIV epidemic, in particular in sub-Saharan Africa.3
In countries characterized by high HIV prevalence, the number of TB cases has almost tripled in the last 15 years. In South Africa, 44 percent of newly detected TB cases are estimated to be HIV positive. Globally, the figure stands at eight percent.4 In developing countries, TB is the leading cause of death among people who are HIV infected.5
Drug-resistant strains spreading globally
What is TB?
Tuberculosis is a contagious airborne disease, and spreads like a common cold. It is caused by a bacterium called Mycobacterium tuberculosis (or M. tuberculosis) which usually infects the lungs.
Only one in 10 people infected by the bacterium will actually develop the disease, since a healthy immune system will keep the infection dormant. But these infections can be reactivated years, even decades later, if the immune system is weak. This explains why people living with HIV whose immune system has been suppressed by the virus are so vulnerable to TB.
The pulmonary form of TB is characterized by a persistent cough, shortness of breath and chest pain. Each person with the infectious form of TB, if untreated, will go on to infect between 10 and 15 other people each year.
The Mycobacterium can also infect almost any part of the body, such as the lymph nodes, the spine or bones. This is the extra-pulmonary form of TB, and is more common in HIV infected patients and children. Although extra-pulmonary TB may not be contagious, it is equally vital to diagnose and treat it rapidly, as all forms of the disease can be deadly if adequate treatment is not given.
The emergence and geographical spread of strains of TB that are resistant to treatment by the standard anti-TB drugs is a second major concern. Resistance to any one drug is the result of naturally occurring genetic mutation within the Mycobacterium, but resistance to multiple drugs emerges through incorrect, interrupted or frequently repeated treatment.
The WHO estimates that there are now almost 490,000 new cases of multidrug-resistant (MDR) TB every year, and that the amount of drug resistance has been ever-increasing in countries as diverse as Peru, South Africa, China and India.6 An estimated 120,000 people die annually from multidrug-resistant TB.7
Diagnosis, treatment and prevention of drug-susceptible TB are difficult enough. But when it comes to tackling the disease in patients who are also infected with HIV or those with resistant strains of the disease, the medical challenges are multiplied.
In this document, through our work in Africa, the Caucasus and Asia, we illustrate our encounters with the new faces of TB and the tough challenges that we face in treating patients with drug-resistant TB and those infected with both TB and HIV. We also shed light on the limits of our capabilities, specifically linked to the continuing neglect of research into the development of newer and better vaccines, diagnostics and drugs to prevent, detect and treat tuberculosis.
Médecins Sans Frontières (MSF) and TB
We have been treating tuberculosis since our first day of operations more than 30 years ago. Today, MSF – often working alongside national health authorities – treats patients in 31 countries in a wide variety of settings, ranging from ...urban slums to rural areas, prisons or refugee camps.
MSF has been working to integrate care for patients with TB and HIV infection across many of its projects.
MSF has also increased the numbers of people it treats with multidrug-resistant tuberculosis from 11 patients in 2001 to 574 patients in 2007 in 12 different projects in countries including Uzbekistan, Georgia, Armenia, Kenya and South Africa.
In the period between 1999 and 2005, 52 percent of patients with drug-resistant TB under MSF’s care in projects in the Caucasus, Central Asia and Thailand either completed their treatment or were cured. 12 percent died and 18 percent defaulted on their treatment due to its length and toxicity. For the remaining 18 percent, treatment either failed or patients had yet to complete their treatment. These figures illustrate the considerable difficulties of treating drug-resistant TB even when this treatment is provided with considerable support and resources.
HIV/TB co-infection: Fighting a Losing Battle?
People living with HIV/AIDS, whose immune systems are suppressed, are particularly susceptible to TB. Not only are they much more likely to develop active TB, but the disease also progresses much more rapidly in HIV positive patients. TB causes up to half of all deaths of people with HIV.
This vulnerability caused by weakened immune systems explains why TB has been ripping through the populations in Sub-Saharan Africa where there is a high prevalence of HIV.
In the past 15 years, new TB cases have tripled in such countries.8 In Lesotho, for instance, where MSF runs an HIV care project in a rural health clinic, of the 221 patients started on TB treatment in 2006, 92 percent were also infected with HIV.
Diagnosis: Falling through the net
People infected with both TB and HIV often present unclear clinical symptoms and are frequently missed by existing diagnostic tools.
The most widely used technique for diagnosing TB in developing countries is no more sophisticated than examining a suspected patient’s sputum sample under a microscope to assess whether it contains TB mycobacteria. This method, called sputum-smear microscopy, was developed well over a century ago.
Although relatively fast and easy to implement in resource-limited settings, the method has significant limitations: it detects less than half of all TB cases, and it is, by definition, not able to identify TB in people, such as children or many people living with HIV, who either have difficulties producing enough sputum from their lungs for a sample for analysis, or don’t have sufficient or any mycobacteria in their sputum to be detected under the microscope.9 It also completely misses the extra-pulmonary form of TB.
The other possibility is to X-ray the patient’s chest. But for people living with HIV/AIDS, the X-ray often doesn’t show up the typical changes in the lung associated with TB infection. The result is that TB infection in many patients who are HIV positive goes undiagnosed – they are falling through the net.
The need for speed
We therefore desperately need better diagnostic techniques. Currently, a technique known as culture, also known as “gold-standard”, is considered the best alternative to microscopy. Culture consists of incubating (or ‘growing’) a sputum sample in a reagent or medium to see whether it contains live TB mycobacteria. It gives results that are far more accurate than microscopy. But as the mycobacteria are such slow-growing organisms, it can be up to eight weeks before a result comes through.
Swift diagnosis of tuberculosis is crucial – not only so that the patient can be put on appropriate treatment as soon as possible, but also to prevent the spread of the disease in the community.
Faster culture techniques do exist: one technique, known as MGIT (Mycobacterium growth indicator tube), is based on a liquid culture rather than the usual solid culture medium. Liquid culture methods have been endorsed by the WHO since 2007.10 MGIT, however, requires very skilled staff, a constant power supply and high safety standards to protect the laboratory staff handling the samples from contamination – things that simply do not exist in many of the more remote settings where we work. Getting hold of the component parts – new tubes and liquid reagent – also requires a reliable supply-chain which cannot always be assured.
In a pilot project in Kenya, where up to 80 percent of TB patients are co-infected with HIV, MSF is working with another improved culture medium. Called thin layer agar (TLA), it shortens the time to diagnosis down to eight to 10 days and shows similar accuracy to that achieved by MGIT. TLA is cheaper and less logistically challenging than the MGIT culture technique. Yet although these factors combine to make TLA potentially a very interesting tool for the sustainable scale-up of TB diagnosis, it is still a rather complex technique that has to be carried out in a laboratory with proper staff training and protection.
As close to the bedside as possible
Many of today’s existing tools are excessively complex. The use of culture as the main diagnostic tool – though it gives more accurate results than microscopy – still presents serious drawbacks in many of the settings where we work: access to culture remains very limited as the vast majority of TB patients (an estimated 85 percent) seek care in small clinics and health posts where either no test or only sputum-smear microscopy is available. Only the remaining 15 percent is seen at better equipped health structures, where it is possible to perform TB diagnosis using culture techniques.11
Current diagnostic methods thus need improving – we need tools that give better results, but are also faster, and we need tools that are as low-tech as possible.
Treatment: two diseases, one patient
Current TB treatment is complex and long-lasting, involving a combination of antibiotics that were developed more than 35 years ago. Lengthy treatment timelines (six or eight months of antibiotics therapy) and drug side effects make it difficult for patients to adhere to treatment through to its completion. Currently, the recommended strategy to maintain adherence is Directly Observed Treatment (DOT), where a health worker or a community volunteer supervises the patient taking his or her medication. While this can improve cure rates, it remains highly labor intensive for health staff, and places considerable strain on patients who sometimes have to travel several kilometers, every day for several months, to a health centre in order to receive treatment.12
Treating patients who are infected with both TB and HIV is even more difficult. When the drugs for the two diseases are taken in combination, there can be drug interactions, which may lead to increased side effects or reduce the effectiveness of treatment. For example, one of the main TB drugs, rifampicin, lowers the efficacy of one of the most common antiretrovirals used to treat AIDS, nevirapine. As a result, a more complicated treatment regimen is required. The patient must take a vastly increased number of pills every day, which when taken together can have toxic effects, notably on the liver.
Integrating HIV and TB treatment
Given the high risks of co-infection in places where large numbers of people with HIV live, it has become increasingly clear that treatment for the two diseases should be integrated. Treatment integration would allow patients to benefit from early diagnosis of either disease and ensure effective monitoring of the combined treatments.
MSF has worked on setting up ‘one stop shops’, in effect clinics where people are treated for both diseases in one place. This brings benefits to both health care staff and patients – patients are assured swifter diagnosis of the two diseases, and staff can effectively monitor them for both treatments at the same time.
However, despite the clear statement from the WHO and others of the importance of implementing an integrated approach13, in most places TB and HIV programs continue to operate in isolation from each other. In 2006, worldwide less than 1 percent of people living with HIV/AIDS were reported to be also screened for tuberculosis.14
Living with two diseases: HIV/TB co-infection
Drug-resistant tuberculosis: entering the mainstream
What is drug resistance and how does it develop?
In the case of TB, resistance to drugs – when the effectiveness of a drug against a pathogen is reduced – develops through a mutation of genes in the bacteria. Although this is a natural phenomenon, resistant bacteria will only multiply in the presence of the drug in question.
Resistance can also develop if the bacteria are underexposed to drugs, because of under-dosing or if the treatment is interrupted or not continued for long enough. In all of these cases, treatment is likely to fail and the disease will re-emerge in a more resistant form, meaning that fewer drugs will be effective against the bacteria.
But direct infection with resistant strains is also possible. Because TB is an airborne disease, a patient can contract drug-resistant TB directly through contact with another person already ill with the drug-resistant strain. There are concerns that the number of those directly infected this way is rapidly increasing. For instance in Tashkent in Uzbekistan, 15 percent of TB cases who have never been previously treated for TB have MDR-TB, that is to say the resistant strain was caught directly from another person with MDR-TB.15
Diagnosis: establishing which drugs work, and which don’t
Many countries in the developing world are not aware of the incidence of drug resistance among their populations affected with TB. A major factor is the difficulty associated with diagnosing drug-resistant strains.
Detecting drug resistance means finding out not only whether a person has TB or not, but just as importantly establishing to which drugs the patient’s TB strain has become resistant. For each individual patient, there will be a different pattern of resistance. Some programs provide a standard drug combination to all patients with MDR-TB but to ensure a patient is given the best possible treatment, doctors need to get an accurate profile of the specific drug resistance for each of their patients.
The diagnostic processes that allow medical workers to establish which drugs will work against the particular TB bacilli in a patient, and which won’t, are known as drug sensitivity testing – or DST. Results cannot be achieved with microscopy or by looking at chest X-rays but they can be obtained through certain culture methods.
A sputum sample is cultured a first time to grow the TB mycobacteria – as described in the previous chapter. The bacteria are then exposed to various TB drugs. After a period of time, the bacteria are then re-examined to see which drugs have had an effect. If the mycobacteria have continued to grow then we can conclude that they are resistant to the effects of the drug. If, however, the mycobacteria have been killed, we can conclude the drugs are still effective.
This might allow caregivers to determine a patient’s resistance patterns, but the method comes with all the problems associated with culture described above: it relies on sputum samples, so the technique is of limited use to those unable to produce sputum or with extra-pulmonary TB. It is complex, relying on a well-equipped laboratory and skilled staff, so much so that it is rarely available in developing countries, and not at all in more remote settings. And it is lengthy, taking up to eight to 12 weeks – time that patients needing to start treatment can ill-afford.
More modern techniques that work on analyzing the DNA of the Mycobacterium do address the last problem at least, in that they can give results in less than 48 hours. But they demand highly sophisticated pieces of equipment, meaning that we are still very far from a TB diagnostic tool that can be used as close to the patient’s bedside, however remote the setting, as possible.
Treatment: a terrible burden
Treating drug-resistant TB is notoriously arduous for patients and presents huge difficulties for health programs. Most of the second-line TB drugs used to treat drug-resistant TB are known for their relative ineffectiveness against the bacilli, meaning a lengthy treatment of up to two years. Patients must receive daily injections for up to six months and take a handful of different drugs once or twice a day for a further eighteen months or more. Treatment is also fraught with numerous side effects which require additional medical management.16
Such intense treatment places very big demands on patients. Many have to give up work in order to see their treatment through. Some patients who are hospitalized for periods of their treatment are isolated from their families, which can again give rise to psychological problems and major loss of income.
The toxicity of the drugs is perhaps the most striking feature of drug-resistant TB treatment (see table “The grim reality of DR-TB treatment”). Indeed, the severity of the side effects has been compared to cancer chemotherapy, with the difference that MDR-TB therapy is not administered in cycles, but in a continuum over two years. With a long list of commonly experienced side effects added to such a lengthy treatment, it is not surprising that many patients give up, some considering the treatment worse than the disease.
MSF project data from Georgia, Armenia, Uzbekistan and Thailand show a patient defaulter rate of 18 percent. Faced with these challenges, MSF has tried to improve the situation by introducing ambulatory and outpatient methods of treatment together with offering psychological and some economic support to patients. We have learned that we can often help patients overcome the social and economic barriers to continuing treatment.
Other caregivers experience similar difficulties. The proportion of patients cured and completing their treatment remains below 50 percent in many programs, especially when the patient is HIV/TB-co-infected.17 However, success rates of up to 66 percent have been reported in some programs.
What is particularly concerning is that even when the best treatment is available, some drug-resistant patients will go on to develop yet further resistance to their drugs. Data from our project in Sukhumi, Georgia, show that 13 percent of MDR-TB patients went on to develop extensively drug-resistant TB, or XDR-TB despite all of MSF’s efforts to provide the highest standards of care. Similarly, in Uzbekistan, the analysis of an MSF cohort of MDR-TB patients revealed that six percent of patients developed XDR-TB while under treatment.18
Practical difficulties make expanding treatment challenging
For healthcare programs, there are also logistical difficulties to contend with. Very often there is only one supplier of the second-line drugs used to treat drug-resistant TB. This can mean that caregivers often have to wait for new supplies of a drug and stocks can actually run out – any treatment interruption bringing disastrous consequences for patients and further fuelling resistance.
DR, PDR, MDR, XDR:
© 2013 Doctors Without Borders/Médecins Sans Frontières (MSF)