PREVALENCE AND CORRELATES OF CYTOMEGALOVIRUS VIREMIA IN KENYAN CHILDREN AT HOSPITAL DISCHARGE
Loading...
Date
Authors
Li, Vanessa
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Background CMV viremia is common in immunocompetent adults admitted to intensive care units, and individuals with detectable CMV viremia are at a higher risk for death and other poor outcomes. Given the high prevalence and early acquisition of CMV in Sub-Saharan Africa, hospitalized African children may also be at high risk for CMV viremia. However, while CMV has been well-studied in severely immunosuppressed populations including those with congenital infection or organ transplant, no studies have examined the association between CMV viremia and mortality in post-discharge children in sub-Saharan Africa; nor, documented the prevalence of CMV viremia in children admitted or discharged from the hospital. This cross-sectional study utilizes baseline clinical, demographic, and CMV DNA data from Kenyan children aged 1 to 59 months enrolled in a randomized controlled trial in Western Kenya, to determine the prevalence and correlates of CMV viremia at hospital discharge. Utilizing quantitative CMV PCR, it also determines the relationship between clinical and laboratory risk factors and the quantity of CMV DNA in blood. Methods To determine the prevalence of CMV viremia, quantitative real-time CMV PCR was used to measure CMV viral loads in stored plasma specimens from 1024 children aged 1-59 months at hospital discharge (with a detection limit of 50 copies/ml of plasma). CMV viral loads were compared among children by HIV exposure and HIV-infection status using the 2-sample Wilcoxon rank-sum (Mann-Whitney). Among HIV-uninfected children, the general linear model with Poisson distribution was used to calculate prevalence ratios for hypothesized correlates of CMV viremia and correlates of high CMV viral load (≥1000 copies/ml). Prevalence ratios were adjusted for age and HIV exposure status. Results The study population consisted of 1024 children 1 to 59 months old who were enrolled at hospital discharge into an ongoing randomized controlled trial of azithromycin to prevent post-discharge morbidity and mortality. The median age was 18 months, and over half of children were under the age of 2 years old. Malnutrition was common; 24% were stunted, 6% wasted and 12% underweight. Most children were HIV unexposed (HUU) (85%), while a number were HIV-exposed uninfected (HEU) (9%) and HIV-infected (HIV+) (2%). CMV viremia was detected in 32% of children at enrollment; 30% in HUU children, 40% in HEU children, and 40% in HIV+ children. The median viral load for all included children was 2.4 log10 copies/mL (IQR: 2.0 to 2.9); it was 2.4 log10 copies/mL (95% CI: 2.0 to 2.9) in HIV-unexposed children, 2.3 log10 copies/mL (95% CI: 2.0 to 2.8; p-val.=0.8) in HEU, and 3.3 log10 copies/mL (95% CI: 2.6 to 5.4) HIV+ children. HIV+ children had significantly higher log10 copies/mL CMV viral load than HEU (p=0.01) or HUU children (p=0.02). Among all HIV-uninfected children (HEU and HUU), age younger than 2 years was associated with a 40% (PR=0.6; 95% CI: 0.5 to 0.8) increased prevalence of CMV viremia at discharge adjusting for HIV-exposure. Adjusting for age, children who were HEU had a 30% (PR=1.3; 95% CI=1.0 to 1.9) higher prevalence of CMV viremia. Among 314 children with detectable CMV, potential correlates including length of hospital stay, discharge diagnosis, and anthropometric indices were not associated with CMV prevalence. Study site and age were the only predictors of high CMV viral load in adjusted models. Conclusions Among children being discharged from hospitals in western Kenya, detectable CMV viremia was common, with a prevalence of CMV viremia similar to that previously reported in adults admitted to ICU in high income settings. Geographical region and young age were correlates of viremia in this population of children. Understanding the potential clinical implications of CMV viremia in these recovering children may offer opportunity for targeted interventions to reduce CMV viral load and improve outcomes. Keywords CMV reactivation, Cytomegalovirus, human herpesvirus, immunosuppressed, hospitalized children, Kenya
Description
Thesis (Master's)--University of Washington, 2019
