World AIDS Day 2013 – but don't forget TB…
"If we are to do something about AIDS, then we have to do something about TB. If we are to do something about TB, we are going to have to do something about AIDS. As we have overcome apartheid, so we shall defeat TB and HIV/AIDS, these ungodly twin killers."
Archbishop Desmond Tutu
The combination of these two diseases together is quite simply the deadliest double-whammy of disease of our age. In Africa TB is the AIDS defining illness, unlike anywhere else, and it is why TB has gone AWOL in Africa since HIV emerged there in any epidemic sense.
Whilst the “normal” risk of developing active TB disease from a latent infection is generally accepted as being 5-10% in a lifetime, in circumstances of co-infection the estimated risk is revised to be 5-15% in every year that the patient might survive the HIV infection. The more regularly-quoted likelihood of developing active disease if someone is HIV positive and has a latent infection is that it is rendered between twenty-one and thirty-four times more likely following HIV infection than it was beforehand. Whatever the true probabilities may be (and they may well vary dramatically in different countries), it is quite clear that the likelihood of developing active TB if you are latently infected and contract HIV (or vice versa, of course) is exceptionally and terrifyingly high.
The two diseases then seem to work both bi-directionally and synergistically, each progressing more quickly in the presence of the other than they generally do on their own. One dramatic estimation has suggested that the life expectancy of a TB/HIV patient is as little as six weeks following the diagnosis of the second disease, although it is hard to believe or substantiate such a claim. A reference in the Global Plan for TB 2011-2015, however, casts the risk from the two diseases in somewhat of a similarly dramatic light: people living with HIV who develop TB but who do not receive effective TB and HIV care have only a one-in-ten chance of surviving for three months or more.
Often with co-infection the patient’s sputum unfortunately fails to show the bacilli when it is stained and examined in the microscope. This leads to delayed diagnosis or a complete misdiagnosis. This pauci-bacillary nature of any co-infected sputum is thought to be because of the lowered host immune response to the disease which results in fewer visible bacilli in the sputum. This resultant diagnostic delay may go some way to further explain the massively reduced life expectancy suggested above. What also seems to happen is that, whilst treating one disease, the other will naturally accelerate, meaning that treating both diseases together is difficult and can be hazardous to the patient.
If drugs to treat HIV are available (and unfortunately they often still are not in Africa) they frequently have to be withheld until the TB is brought under control; but the drugs for TB immediately exacerbate the HIV-provoked immune deficiency and so accelerate the progression of HIV immune deficiency. Furthermore, once the drugs for both diseases can be safely applied together, regimes of reduced efficacy often have to be implemented because of the interactions between the drugs for the two diseases some of which are still to be fully recognised and documented.
TB is also known to increase HIV infectivity, as well as reduce HIV treatment efficacy. What is undeniable is that people living with HIV (PLWH) need the earliest possible diagnosis and treatment of active TB disease if it emerges.
TB in co-infected cases also exhibits higher-than-normal rates of extra-pulmonary disease (at least 25% instead of a more normal 10% and often much higher) – a factor which unfortunately makes active disease far more difficult to recognise because symptoms may be atypical, and because sputum tests will again generally show positive even in cases of active disease.
Co-infection of HIV and TB is also thought to result in more rapid potential development of MDR- and XDR-TB if the drugs are at all mismanaged. This poses a huge epidemiological risk, which is particularly acute for Africa.
It is estimated that at least one-third of the current 34 million people living with HIV worldwide are latently infected with TB bacteria, although most are not yet ill with active TB. Looking at the two diseases from the opposite direction, about one fifth of tested TB patients also show up as HIV-positive.
TB is a leading cause of death in people with HIV infection. Almost 25% of deaths among people with HIV are due to TB – in 2011 about 430,000 people died of HIV-associated TB. In Africa, however, TB is the leading cause of death among PLWH, and most HIV-positive patients die from it.
In 2012 there were an estimated 1.1 million cases of HIV-positive new TB cases. 75% of these were in Africa.
In 2011, the percentage of TB patients found to be HIV-positive in the 28 African countries in the list of 41 priority HIV countries ranged from 8% in Ethiopia to 77% in Swaziland. Besides Swaziland, the countries with more than 50% of TB patients who were tested having an HIV test which was HIV-positive were Botswana, Lesotho, Malawi, Mozambique, Namibia, South Africa, Uganda, Zambia and Zimbabwe.
More than two-thirds (69%) of all those people currently living with HIV (23.5 million people) live in sub-Saharan Africa — a total which includes a terrible 91 percent of the world’s HIV-positive children. The percentage of HIV-positive TB cases occurring in Africa in 2012 was estimated to be 75% of the total global number.
In 2011, an estimated 1.8 million people in the region became newly infected with HIV, and an estimated 1.2 million adults and children died as HIV-positive on the continent. This accounted for 71% of the total world’s AIDS deaths in that year.
In 2011, an estimated 330,000 children under 15 contracted HIV and an estimated 230,000 HIV-positive children died that year: most of them died from TB, and almost all died in Africa.
Clear progress has been being made in this concerted campaign against the co-epidemic. Cambodia, particularly, is identified as being especially successful. Globally, HIV associated TB deaths have been falling since 2003, but there was still at least 320,000 deaths from HIV associated TB in 2012. The UNAIDS and Stop TB Partnership’s target is actually to halve TB mortality rates among people living with HIV (PLWH) by 2015 compared to 2004. The estimated high point in 2005-6 was a little over half a million, so, while there is clearly work to do, this target is in sight.
There has also been a very encouraging increase in Anti-Retroviral Therapy (ART) among HIV-positive patients, up to 57% in 2012, but given the WHO recommendation that all HIV-positive TB patients should be automatically eligible for ART irrespective of their cell count there is still work to do. The early initiation of ART after the initiation of TB therapy is now recommended, being at is seen as being able to seriously reduce the risk of morbidity and mortality in PLWH. By itself, it is estimated to reduce the risk of re-activated or primary TB disease by as much as 65%. In the latest WHO recommendations the threshold CD4 count at which ART is recommended to be commenced has been raised from less than 350 to less than 500 CD4/mm. (ART is already recommended for all TB patients living with HIV whatever their cell count). This raising of the threshold should mean that many more PLWH should be eligible to start treatment and that fewer PLWH will be seen to develop TB that. As a result, all rates should correspondingly improve further.
There is still unfortunately little in the way of a real evidence base for the pharmaceutical treatment of both diseases together. Most treatments have been developed by expert opinion only, and much of the work has been done in Europe, where arguably different disease profiles exist, so that what has been proposed may not actually be so applicable where the co-epidemic thrives. Also, expert opinion often varies between authorities. Such a dearth of data may seem particularly troubling given that, worldwide, TB is the biggest killer of people living with HIV today.
Considering that nearly all of the rates in the co-epidemic are moving in the right direction, however, it certainly looks like expert opinion has been getting it right, stoked by the fires of HIV advocacy groups who recognise the threats to the campaign against HIV from tuberculosis. It may not be, however, that things will turn out quite as hoped, particularly as the possible impact from MDR- and XDR-TB enters the mix. This is because these deficiencies of knowledge are most particularly acute with the second line drugs, the ones used for treating DR-TB, since they have yet to be acceptably tested for interactions with HIV medication. Doing this has simply not been a priority issue for developers of HIV drugs since co-infection with TB is still relatively so much less common in wealthier countries and co-infection with drug-resistant strains even less so.
There exists a gold-standard model for the campaign against HIV/TB. It is called the “Three I’s for HIV/TB”. These are –
• Intensified case finding for TB,
• Isoniazid preventive therapy, and
• Infection control.
These are intended to reduce the burden of TB among people living with HIV and therefore should be being urgently implemented by all HIV services. All of these too are moving in the right direction, but only slowly where resources are poor. Encouragingly, the number of people living with HIV who were screened for TB (an element of the "intensified case finding") almost quadrupled from 600,000 in 2007 to 2.3 million in 2010. By 2012 this figure rose to over 4 million. However this still represents a small proportion of the 34 million people estimated to be living with HIV.
It is encouraged that, if TB infection is confirmed but active TB is confirmed as not being present after screening (in other words, the patient is latently infected), patients should automatically receive isoniazid preventive therapy (IPT). The treatment is intended to be free of charge at point of care and not to be prohibitively expensive for any health system. Of the 1.5 million people reported to have newly enrolled in HIV care in 2010 just 12% were actually put on isoniazid preventive therapy (IPT). In 2012, however, 1.6 million were reported to be newly put on HIV treatment, and 30% of them were put on IPT, so there is dramatic progress being made. Furthermore, 42 countries are now reporting data, whereas, a year earlier, just 29 were doing so. There are very clearly immense efforts being made to ramp up this part of the Global Plan.
There are still worrying complexities to this part of the policy however. The major risk is identified to be from using IPT on already active disease since doing so may stoke drug-resistance. The WHO recommends a four-symptom algorithm to rule out this risk, as a result of which they expect approximately 50% of PLWH to be eligible for IPT. In 2012 the estimated number of people living with HIV was 35 million, implying that the current target should be 17.5 million PLWH being swept up and put on IPT. How foolproof is the algorithm? Truthfully, knowing that TB has a habit of not showing symptoms until sometime after it has re-activated, it’s difficult for anyone to be certain. With these sort of numbers, only a small percentage of “misses” (i.e. PLWH being put on IPT when they already have asymptomatic re-activated TB) could make for a lot in terms of absolute numbers.
The rate of increase of implementation of IPT is, in fact, greater than the rate of improving screening for HIV, and the Global Plan intends that all of those eligible for IPT should be being put on the treatment but 2015. In fact, it is just possible that what is being done may actually ultimately prove to have been worse than doing nothing.
Where there are already high rates of MDR-TB, despite IPT being indicated for HIV positive people, it is epidemiologically almost certainly a risky treatment and it is arguable that it should not be embarked upon at all as a general policy in such countries. Remember: co-infection of HIV and TB is also thought to result in more rapid potential development of MDR-TB and XDR-TB if the drugs are at all mismanaged. The data on the efficacy of IPT has been harvested mainly in low incidence countries, without high rates of co-infection and without high rates of drug-resistance.
If an HIV patient who presents as a candidate for IPT is already isoniazid resistant, what exactly will the effects be of putting her on a long term course of the drug? It’s far from clear that anyone knows. The eight countries with the biggest proportion of HIV patients being put on IPT are (in order): South Africa (370,000), Ethiopia (30,000), Malawi (21,000), Mozambique (17,000), Lesotho (16,000), Haiti (15,000), Ukraine (14,000) and Namibia (12,000). Whatever the official figures may say, it’s unlikely that any of these countries do not have rates of MDR-TB that are well above the global average. This is particularly probable within the pool of latent infection where drug-resistance is undiagnosable undetectable and will only emerge in the future. All of these countries re almost certainly also already being affected by the insidious growing threat of XDR-TB, which will also be in there in their respective pools of latent disease. (South Africa, of course, is well recognised as suffering from XDR-TB, already carrying around 70% of the known pandemic, but four of these listed countries supply much of the migrant workforce that comprises the labour for in the HIV and TB-ridden mining industry of South Africa itself).
There is also the factor of re-infection in high incidence environments. Without any shadow of a doubt rates of re-infection are high in many high incidence countries although they will be reduced with ART. This risk dictates that the preventative treatment needs to be rolled out in really big numbers in order to make any impression on the latent pool and bring the numbers of re-activating disease down as intended.
The question has to be asked, however – will it turn out to be a circle of much less than is being expected of it – with the accompanying risk of stoking the DR epidemic in the process by unwittingly treating asymptomatic re-activated disease?
In the meantime, exactly how is the success or failure of the initiative being measured? It can’t be measured by skin testing. The only way, it would seem, is by long-term follow-up to see whether these patients have really avoided re-activated disease. Will there be provision allowed for doing this?
It seems almost certain that this progress in IPT has been due to the relative strength of the HIV advocacy lobby. The concomitant neglect of the associated issues relating to drug-resistant TB strongly suggests this – the priority being more intensely focused on the epidemiology of HIV rather than on the two diseases equally. It is absolutely vital that the risks from drug resistance are not neglected or glossed over.
Bishop Tutu, at the start of this section, offered his special wisdom and insight on this problem. “The Bish” himself was hospitalised for twenty months as a fourteen-year-old with TB so he has more reason than many to be interested in this disease. As importantly, he also appears to see the more complex issues: that looking at one disease through the lens of the other without doing exactly the same in reverse will miss vital things and just could create further problems; that neither disease can be resolved on their own; nor can both diseases be resolved together without both of them being resolved individually. We can add, however, that the complexities of both diseases must be fully considered and given equal respect in this process. All this can be done, just as it was done with apartheid, but it is an immense challenge and is fraught with further complications and risks, not the least of which is that they have to be viewed together with every possible angle being considered.