Yotebieng M, Van Rie A, Moultrie H, et al. Effect on mortality and virological response of delaying antiretroviral therapy initiation in children receiving tuberculosis treatment. AIDS. 2010 Jun 1;24(9):1341-9.
Rates of co-infection with HIV and tuberculosis (TB) are high among children in resource-constrained areas. Children are particularly vulnerable to progression from TB infection to active disease, with increased risk of disease progression and death among those also infected with HIV.(1, 2) Recommendations for treating children coinfected with TB and HIV have called for delaying initiation of antiretroviral therapy (ART) anywhere from several weeks after initiating TB treatment to the completion of TB treatment as a way of minimizing drug interactions and differentiating side effects between TB and HIV medications.(3, 4) However, delays in initiation of ART until TB treatment has been completed have been associated with increased mortality among adults.(5)
To estimate the effect of delays in ART among children initiating TB treatment on mortality and viral suppression
Outpatient pediatric HIV clinic, Soweto, South Africa
Pediatric patients (aged less than 16 years) co-infected with TB and HIV who had their first clinic visit and received care between April 1, 2004 and March 31, 2008 and initiated TB treatment prior to starting ART.
Survival and viral suppression
Diagnosis of TB was based upon clinical criteria; TB treatment consisted of rifampicin, isoniazid, and pyrazinamide for the first two months, followed by rifampicin and isoniazid for an additional four months. First-line ART consisted of stavudine, lamivudine, and ritonavir-boosted lopinavir for children who were aged three years or younger and who weighed >10 kg. During TB treatment, the ritonavir doses were doubled.
Children receiving ART were assessed at one and three months and every three months thereafter. CD4 cell and viral load measurements were obtained at baseline and then every six months.
The time from starting anti-tuberculin therapy to initiation of ART was categorized as more than 15, more than 30, or more than 60 days. Mortality was analyzed using inverse probability-of-treatment and censoring weighting of marginal structural models that controlled for confounding from various times of ART initiation that would be expected to occur based, in part, on clinical factors. Patients were censored at 36 months or at last clinic visit, whichever came first, or when lost to follow-up (defined as at least six months without a clinic visit or response to contact tracing by telephone). The probability of initiating ART was estimated using pooled logistic regression models. The Cox proportional hazards model was used to estimate the relative hazard (RH) of viral suppression with adjustment for baseline variables.
There were 573 patients included in the analysis. Median follow-up time was 9.6 months (interquartile range [IQR] 1.9-23.1). Thirteen percent were lost to follow-up and the baseline demographic characteristics of the children lost to follow-up did not differ from those who remained in care.
The median age was 3.5 years (IQR 1.4-6.8), median CD4% was 11.9% (IQR 6.6-18.3), and median viral load
was 5.2 log copies/mL (IQR 4.5-5.9). A total of 461 (80.5%) children started ART, and the median time between
the start of TB treatment and ART was 17 days after TB treatment began (IQR 0-49 days). Initiation of ART was
>15 days after TB treatment in 55% of children, >30 days in 42% of children, and >60 days in 23% of children.
Thirty-one percent of the children began ART and anti-TB treatment the same day.
Seven (1.2%) children died before they started ART (three died within 30 days of starting TB treatment and four
died after 60 days). Thirty children died after starting ART. The HR for starting ART more than 15 days after
treatment was 0.82 (95% confidence interval [CI]: 0.48-1.63), for starting ART more than 30 days after TB
treatment was 0.86 (95% CI: 0.46-1.60), and for starting ART more than 60 days after TB treatment was 1.32
(95% CI: 0.55-3.16)
There were 324 children who had at least one viral load measurement after ART, and of these, 291 (90%)
achieved viral load below 400 copies/mL. Virologic suppression was more likely among older children and those
with a higher weight for age and lower viral loads. Delaying ART for >15 or >30 days had no effect on viral
suppression (HR 0.98, 95% CI: 0.76, 1.26 and HR 0.95, 95% CI: 0.73, 1.23), respectively. The HR for delays
greater than 60 months was 0.84 (95% CI: 0.54-1.10).
Delaying ART for more than two months after initiating anti-TB treatment increases the risk of death and is associated with worse virologic response to ART.
This was a high quality study; sample size was sufficient, participants were representative of children in care, follow-up time was adequate, and loss to follow-up was accounted for and reasonable in number.
This study provides further evidence in support of not delaying ART for more than 30 days among children receiving anti-TB treatment. Similar findings have been reported for adults coinfected with TB and HIV. Although additional studies, such as randomized trials in children, may be considered, given the survival benefit of earlier initiating of ART, particularly among children most severely immunocompromised, changes in treatment recommendations should be considered in light of current data.
- Chakraborty R. HIV-1 infection in children: a clinical and immunologic overview. Curr HIV Res. 2005;3:31-41.
- Newton SM, Brent AJ, Anderson S, Whittaker E, Kampmann B. Paediatric tuberculosis. Lancet Infect Dis. 2008; 8:498-510.
- World Health Organization, Antiretroviral Therapy of HIV Infection in Infants and Children in Resource-Limited Settings: Towards Universal Access, W.H. Organization, Editor. 2006,World Health Organization: Geneva, Switzerland.
- Mofenson LM, Oleske J, Serchuck L, Van Dyke R, Wilfert C. Treating opportunistic infections among HIV-exposed and infected children: recommendations from CDC, the National Institutes of Health, and the Infectious Diseases Society of America. MMWR Recomm Rep 2004; 53:1-92.