This is the first study sponsored by National AIDS Control Organization, Ministry of Health & Family Welfare, Government of India in this topic. The results of this study show that once highly active antiretroviral therapy (HAART) is started, there is no significant difference in mortality between HIV-TB patients who were started on HAART 2-4 weeks after initiation of antituberculosis treatment (early ART group), and those who received it 8-12 weeks after starting ATT (delayed ART group). However, the risk for progression of HIV disease, as estimated by onset of failure of ARV therapy, increases significantly if initiation of HAART is deferred from 2-4 weeks to 8-12 weeks (16% vs. 31% risk, p = 0.045). The incidence of IRIS and other adverse events, and outcome of TB treatment are similar between the groups.
To decide the optimal timing for initiation of HAART after starting ATT in HIV-TB patients has been listed as a priority research question for resource limited settings by WHO . Several studies have been involved with the task of finding a definitive answer to it so that mortality and serious morbidity could be minimized in this patient population [12, 21]. A recent trial by Abdool Karim et al shows that starting antiretroviral therapy at CD4 cell count <500/mm3 during treatment for AFB smear-positive TB (integrated therapy) reduces mortality by 56% in HIV-TB cases, as compared to delaying it until the completion of TB therapy (sequential therapy) . They have reported an all-cause mortality rate of 5.4 per 100 person-years in the integrated therapy arm as compared to 12.1/100 per year in the sequential therapy arm (p = 0.003). Their integrated therapy arm corresponds roughly with both the groups of the present study taken together, where the combined mortality rate was observed to be 13.4/100 per year. The main reason for this apparently higher mortality observed here may be that the current study had a much higher percentage (77.3%) of patients in WHO stage four HIV disease at baseline than that (4.9%) in the South African study.
Another trial at Cambodia has shown in its preliminary report that there is a significant survival advantage of 34% in starting HAART at two weeks (early arm) rather than at eight weeks (late arm) after initiation of TB treatment [19, 23]. They observed a mortality of 17.8% in the early arm against 27.4% in the late arm (p = 0.002). On comparison, it can be seen that the Cambodian study reports a much higher mortality rate in both the groups than the present study despite both following a similar protocol in terms of timing of ART initiation. One of the important reasons for this difference may be their exclusive enrolment of patients with CD4 count <200/mm3. For the same reason, it can be argued that results of the current study are more generalizable than the Cambodian one. Moreover, while the present research work has excluded the deaths that might have occurred between the starting of ATT and ART in both the groups, this fraction was taken into account while calculating mortality in the Cambodian project.
The current research was designed to assess and compare the response of HAART in HIV-TB patients who started it 2-4 weeks after starting TB therapy and those who received it 8-12 weeks after starting TB treatment. CD4 cell counts and plasma viral load have been considered important markers for assessment of response to ARV therapy and disease progression [24
]. Previous studies have shown that co-infection with TB results in reduced survival, increased risk for other opportunistic infections and elevations in HIV replication [27
]. Increased HIV replication is attributed to activation of latently infected cells, and promotion of infection in uninfected lymphocytes and macrophages. HIV genetic diversity is also increased in the presence of active TB infection [29
]. The results of this study indicate that these processes may be significant enough to cause failure of ARV therapy after a few months, warranting a switch to second line ART. Some previous studies have demonstrated that death in HIV-TB patients within the first few months of TB treatment may be related to TB, whereas late deaths are attributable to HIV disease progression [32
]. Therefore, it can be extrapolated that in due course of time a differential disease progression might even translate into a significant difference in mortality between the two groups of the study. In light of this observation, the present work supports 2010 WHO recommendations for the management of HIV-TB cases: [35
Start ART in all HIV-infected individuals with active TB, irrespective of the CD4 cell count.
Start TB treatment first, followed by ART as soon as possible afterwards (and within the first eight weeks).
The overall incidence of Immune Reconstitution Inflammatory Syndrome observed here (10.0%) is close to that reported for the integrated therapy arm (12.4%) by Abdool Karim et al. The Cambodian trial has reported a significantly higher incidence of IRIS in the early ART arm (4.03 per 100 patient months) than in the late arm (1.44/100 pm, p <0.0001) , but no such difference could be seen in the present study.
This research work has a few limitations. Here all cause mortality was taken as the primary outcome instead of disease-specific death rate. While the latter would be more suitable for assessment of response to ART, there were logistic difficulties in ascertaining the exact cause of most of the deaths, as they happened away from the hospital. Another limitation is introduced by the uneven number of lost to follow up patients (11 in early ART and one in delayed ART) in the two groups of the study, which could affect its results. But seven of them from early ART arm were known to be alive by the end of 12 months since they were initiated on ART, and the consideration of the worst outcome for the five (four in early ART arm and one in delayed ART arm) untraceable patients did not alter the overall results of the study significantly. In addition, as decided in the approved protocol, sample size of the study was based on a target of convenience rather than on power calculations. In this study given a baseline rate of 10% mortality, there was an 80% power to detect a difference of 18% mortality between early and delayed ART groups. However, despite these limitations, being one of the first studies on this topic from India, the present work will serve as one of the resources for early starting of ARV treatment under National AIDS Control Programme. It will also help for any future metaanalysis for the timing of HAART initiation in HIV-TB cases.
This study presents important research findings regarding optimal time to initiate ART in HIV-TB cases so that detrimental effects of both the diseases can be minimized. Since it included HIV patients irrespective of baseline CD4 cell counts and with different types of TB, its results can be fairly generalized. It brought to light the observation that starting early ART in HIV-TB patients helps control the progression of HIV disease later on, and hence, it suggests that all such patients must be started on HAART as soon as possible after initiation of TB treatment.