|Image courtesy: WHO/H.M. Dias.|
Post by Tamilarasu Kadhiravan.
Kadhir is an academic physician based at the Jawaharlal Institute of Medical Education and Research (JIPMER) at Puducherry, in South India; interested in tropical infectious diseases research, particularly tuberculosis. Here, he blogs about the Cochrane review he recently co-authored: Rifamycins (rifampicin, rifabutin and rifapentine) compared to isoniazid for preventing tuberculosis in HIV-negative people at risk of active TB.
What did the Cochrane Review find?
Let me keep things short and straight:
1) We reconfirmed what everybody knows - that short course rifampicin-pyrazinamide combination is unacceptable for treating people with LTBI due to the increased risk of liver toxicity.
2) We consolidated the available evidence that shows that short course rifampicin for 3-4 months has less risk of liver toxicity (this is important; many folks, old and young alike, still believe that rifampicin is more toxic to the liver than INH), and better treatment completion than INH alone or the combination of INH and Rifampicin, with no evidence to demonstrate decreased comparative efficacy.
3) We found that no benefit was apparent with the rifampicin-INH combination in terms of adherence and side effects over INH alone; and
4) Finally, the new kid in the block, supervised weekly rifapentine-INH looks promising.
My thoughts on rifampicin-INH combination
When I say, no benefit was apparent with the rifampicin-INH combination, the obvious question that crosses one’s mind is, should guidelines recommend this regimen any more or not? Well, the answer to this question is not that straight forward. All other things being equal (I mean efficacy and side effects), although no clear superiority was demonstrated by medical researchers, as a common man I would prefer (for that matter anybody would, I hope!), a regimen that needs to be taken for just 3 months than a regimen that needs to be taken for 6-9 months.
What does this review tell us – were all things really equal? Strictly speaking, the quality of the evidence to draw such a conclusion was of low to very low quality, meaning that future research is very likely to change our faith in the currently available effect estimates. (Oops! Shouldn’t it be ‘confidence’, for there is no place for ‘faith’ here).
But, do we really need more research to find out whether rifampicin-INH for 3 months is equally efficacious as INH alone for 6-9 months, given the fact that we know that 3-4 months of rifampicin alone is good enough. If one depends on logic, perhaps would say, no we don’t. But, you know, logic is the last thing that works in clinical medicine. At least that’s what I believe. In this particular case, I would say that more is not necessarily better. In fact, available evidence indicates that co-formulation of rifampicin and INH decreases the bio-availability of the former . Thus, in my view, as far as efficacy is concerned, we cannot equate 3 months rifampicin-INH with 3-4 months of rifampicin, and we need specific evidence to prove that point. (Let’s posit a contrarian view - a common sense approach – when rifampicin alone for 3 months can do the job, why add INH to it?)
Another question pertinent to this discussion is whether 3 months of rifampicin-INH would be cost-saving compared to 6-9 months of INH. While the addition of rifampicin would drive the costs up, a potential decrease in the need for monitoring, physician visits, and work time lost would save on indirect costs. Good amount of information on cost-benefit is available for the 4 months of rifampicin regimen. But, for the rifampicin-INH regimen information on cost-benefit in HIV-negative people is limited.
So, where do we stand now - should guidelines recommend the rifampicin-INH regimen or not? The honest answer would be, we don’t know. The jury is still out. Particularly, see what happens when this review gets updated.
What is the relevance of this review to high-burden settings?
In countries where tuberculosis is rampant, even a lay man would know what ‘TB’ stands for. But, it is a strange fact that many medical personnel do not know what the term ‘LTBI’ means. (Not infrequently, one encounters patients with various illnesses/conditions ranging from fever, abdominal pain, weight loss, infertility, and like symptoms that were treated with a full-course of anti-TB treatment, just because the only abnormality the doctor managed to find was a positive Mantoux (tuberculin skin test; TST). That’s clearly an abuse of the TST as well as anti-TB drugs.) Certainly, I am not exaggerating things. All of us know well that necessity directs knowledge. They do not know because they need not know; they need not know because they do not treat LTBI!
Yes, it is true that physicians in high-burden countries don’t treat LTBI. In a sense, it is a form of therapeutic nihilism. The oft cited reason is that it is futile to treat people with LTBI since they would quickly get re-infected. True, there is a risk of re-infection in high-burden settings. But, the average annual risk of TB infection (ARTI) in India is 1.5% only and is currently falling. Hence, it is unreasonable to presume that all of them would necessarily get re-infected. In fact, somebody took the pains to do mathematical modelling and found that when the ARTI is 2% and the prevalence of LTBI is 20%-50%, the number needed to treat (NNT) to prevent one case of active TB would still be less than 25.
But, I strongly agree that treating the large pool of asymptomatic people with LTBI to control TB as a public health problem is not the intervention where we should be spending our limited resources. At the present moment, diagnosing and treating active TB is our priority in high-burden settings; there could be no different opinion about that. While treatment of LTBI has been an important component of the TB control strategy in low-burden countries in the West for a long time, the public health strategy in high-burden countries essentially should focus on effective treatment of active TB cases.
However, in my view, there are some select patient groups where treatment of LTBI should be considered even in a high-burden setting. These are,
· Possibly other patients on long-term immunosuppression, and
The justification for focusing on these select patient groups is the absolute clinical benefit that would accrue to individuals treated for LTBI rather than any presumed benefit to the population, which probably does not exist. I think, it is these select patient groups to whom our review is relevant in high-burden resource-limited settings, and certainly not as a general public health measure.