Paper # 845 
IRIS-related Mortality in HIV-infected Children Admitted after Initiating ART in Uganda
Judy Orikiiriza*1, S Bakeera-Kitaka1, D Boulware2, Y Mulumba3, V Musiime4, P Mugyenyi4, and E Mworozi1
1Makerere Univ Coll of Hlth Sci, Kampala, Uganda; 2Univ of Minnesota, Minneapolis, US; 3Uganda-Case Western Reserve Univ Res Collaboration, Kampala; and 4Joint Clin Res Ctr, Kampala, Uganda
Background: Mortality in resource-limited areas
appears to be particularly high during the early months after antiretroviral
therapy (ART) initiation. The causes for clinical deterioration after starting
ART and the mortality attributable to Immune Reconstitution Inflammatory
Syndrome (IRIS), however, have not been established. We investigated the
IRIS-related mortality in HIV+ children in Uganda in the first 72 hours of admission.
Methods: We conducted a prospective study of children
aged ≤18 years at Joint Clinical Research Center receiving ART for 0.5 to
6 months duration during December 2006 to October 2007. We assessed clinical
and laboratory data at time of ART initiation and study recruitment. The
children requiring hospitalization were followed for 72 hours, and acute
mortality was assessed. Subjects were prospectively evaluated for IRIS by a
standardized pediatric case definition. Data were analyzed using STATA version
10.
Results: In this study, 162 children (1:1.3 male:
female ratio) were evaluated with a median age of 6 years (Interquartile range:
2.5 to 11 years). The prevalence of IRIS was 38% (62/162, 95%CI: 31 to 46%)
with (77%) unmasking and (23%) paradoxical IRIS events. IRIS most frequently
occurred in the first month of ART (55%). Overall, 24% (39/162) of
subjects required hospitalization for IRIS which accounted for 62% (24/39) of
all hospital admissions while on ART (OR = 3.6, 95%CI 1.7 to 7.6, P =0.001).
Mild IRIS events managed as an outpatient occurred in 24% (38/162). The overall
survival was 94% (153/162). Of the 9 acute deaths, IRIS attributable mortality
was 44% (n = 4). Mortality occurred in the first one month of ART (OR = 5.6,
95%CI 1.04 to 30.0, P =0.07), and common characteristics of the
children who died included severe immunosuppression, oral candidiasis, and
malnutrition.
Conclusions: In resource-limited areas IRIS-related
mortality is most likely to occur soon after ART initiation in children and
adolescents with advanced immunosuppression and malnutrition. There is a need
to start ART early to prevent morbidity and mortality after ART initiation in
children. There is need for larger IRIS studies in children to assess the
mortality burden in resource-limited areas.
|
