SURVIVAL AND LOSS TO FOLLOW UP AMONG HIV AND TUBERCULOSIS CO-INFECTED CHILDREN BELOW 3 YEARS OF AGE INITIATING ANTI-TUBERCULOSIS TREATMENT IN WESTERN KENYA

The optimal strategy for treating young children co-infected with HIV and tuberculosis is unknown. We compared mortality and loss to follow-up (LTFU) among HIV-infected children aged three years and below who received anti-tuberculosis treatment (anti-TB) and combination antiretroviral therapy (cART) in various sequences. Methods: We conducted retrospective analysis of longitudinal data of HIV-infected children initiating anti-TB at less than 3 years who had not been initiated on cART at anti-TB initiation enrolled in the USAID-Academic Model Providing Access to Healthcare (AMPATH) Partnership in Kenya between 2001 and 2009. Survival outcomes were analyzed by Kaplan-Meier methods. Cox proportional hazard models were used to evaluate the risk factors associated with death or a composite outcome of death and/or LTFU. Results: Of the 365 eligible subjects, 52% were female, 66% were initiated on cART, 15.9% were LTFU and 6.9% died. Sixty-four were initiated on cART less than 8 weeks after starting on anti-TB, 64 between 8-24 weeks, and 237 after 24 weeks. Being female was associated with a higher chance of initiating cART. Children on cART and anti-TB had better survival compared to those not on anti-TB but never on cART, (HR 0.336 95%CI 0.149-0.757). There was no survival or LTFU difference between the three cART initiation timing groups. Conclusions: For young HIV-TB co-infected children, there was no statistical difference in mortality, LTFU or mortality & LTFU when initiation on cART was staggered depending on the severity of clinical or immunologic stage within 8 weeks, 8-24 weeks or after completion of anti-TB. HIV-TB co-infected children on both cART and anti-TB had better survival.