Review Article

Timing of Antiretroviral Therapy for HIV in the Setting of TB Treatment

Table 1

Studies on initiation of highly active antiretroviral therapy (HAART) in HIV/TB-coinfected individuals.

Study (author, year)Study characteristicsTiming of HAART relative to TB treatmentOutcomesAdverse events

Leonard  et  al.  2002 [51]- Retrospective cohort
- 644 adults and children (60% pulmonary TB, 25% pulmonary and extrapulmonary TB, 15% extrapulmonary TB)
- Single center — Atlanta, GA
- Comparison of 3 HIV/TB coinfected cohorts: 1991, 1994, 1997
HAART at time of diagnosis versus HAART post-TB diagnosis — not definedDecreased 1 year mortality in 1994 and 1997 cohort compared to 1991 cohortN/A

Sungkanuparph et  al.  2006 [52]- Retrospective, observational
- 29 adult patients (69% pulmonary TB, 31% extrapulmonary TB)
- Jan 2002–Dec 2002
- Single center—Thailand
Median 8 weeks (range 4–12 weeks)Virological suppression
-65% at 24 weeks
-76% at 48 weeks
TB outcomes
-26/29 completed therapy
-No relapse
-No new OI
IRIS—1/29
Death—1/29 (CMV encephalitis)
Rash—2/29 (NVP)
EFV switched to NVP 1/29 (dizziness)
AZT switched to D4T 3/29 (anemia)

Manosuthi  et  al. 2006  [53]- Retrospective cohort
- 1003 adult patients
- Jan 2000–Dec 2004
- Single center — Thailand
- Comparison of HAART+ to HAART− group
2 mth versus 4 mth versus
6 mth versus 9 mth versus
12 mth (subgroup analysis)
Increased survival for pts. receiving HAART as compared to no HAART
Increased mortality with HAART initiation at >6 mths. (11.3%) compared to <6 mths (4.1%) ( )
N/A

Sanguanwongse  et al. 2008 [54]- Observational cohort
- 1269 patients (adult and pediatric) (54% pulmonary TB, 35% extrapulmonary TB, 12% both)
- Oct 2004–March 2006
- Multicenter — Thailand National Surveillance Network
Not definedDecreased mortality in group receiving HAART (11%) compared to that not receiving HAART (46%) (relative risk 0.24, 95% confidence interval: 0.19 to 0.30)N/A

Tabarsi  et  al.  2009 [55]- Retrospective cohort
- 69 patients
- 2002–2006
Group I: HAART after 8 wks if CD4 < 100
Group II: HAART at 2 wks if CD4 < 100
Group I & II: HAART after 8 wk if CD4 101–200
Increased mortality when HAART deferred after 8 weeks if CD4 <100 Group  I versus Group II mortality = 27.7% versus 4.5% ( )No difference between groups I and II regarding Grade 3 or 4 events, IRIS, or new OI

Velasco  et  al.  2009  [56]- Mixed retrospective/ prospective study
- 313 patients
- 1996–2004
- Simultaneous: HAART within 2 mths of TB diagnosis
- Nonsimultaneous: HAART after 3 mths of TB diagnosis
- Decreased mortality in simultaneous group (9.3%) versus nonsimultaneous group (19.7%) ( )
- No difference in virological/immunological outcomes
N/A

Torok  et  al.  2009 [57]- Randomized, double blind, placebo-controlled trial
- 253 patients
- Vietnam
- Clinical diagnosis of TB meningitis
HAART started within 7 days (immediate arm) or at 2 mths (deferred arm) after initiation of TB treatmentMortality:
-Immediate arm: 76 deaths/127 pts
-Deferred arm: 70 deaths/126 pts ( , )
Incidence of grade 3 or 4 adverse events first 2 mths:
-Immediate: 86%
-Deferred: 75%  

Abdool  Karim  et al.  2010 [58]- Open-label, randomized, controlled trial
- 642 patients
- Durban, South Africa
- Only patients with positive sputum smear for acid-fast bacilli and CD4 count < 500/mm3 included
HAART started either during TB treatment (in two integrated-therapy groups) or after completion of TB treatment (in one sequential-therapy group)Mortality rate per 100 py:
-Integrated arm: 5.4
-Sequential arm: 12.1 ( ) Virological suppression 6 mths after HAART initiation:
-Integrated arm: 91.1%
-Sequential arm: 86.7% ( )
Incidence of IRIS:
-Integrated arm: 12.4%
-Sequential arm: 3.8%
Grade 3 or 4 adverse events:
Integrated arm: 30/100 py
Sequential arm: 32/100 py ( )

Blanc  et  al.  2010 [59]- Open label randomized,
controlled trial
- 661 patients
- Cambodia
- Only patients with positive smear for acid-fast bacilli and CD4 count < 200/mm3 included
HAART started at 2 weeks (early arm) or 8 weeks (late arm) after initiation of TB treatmentMortality rate per 100 py:
-Early arm: 8.28
-Late arm: 13.77 ( ) Virological suppression 50 wks after HAART initiation:
-Early arm: 95.6%
-Late arm: 95.6% ( )
Incidence of IRIS per 100 pm:
-Early arm: 4.03
-Late arm: 1.44

OI: Opportunistic infections IRIS: Immune reconstitution inflammatory syndrome EFV: Efavirenz NVP: Nevirapine pts: Patients HR: Hazard ratio py: Person-years pm: Person-months N/A: Not available.