Meta-analysis of TB & HIV co-infection mortality rate in sub-Saharan African children, youth, and adolescents

Abstract

Background: Despite the effectiveness of antiretroviral treatment (ART) in reducing morbidity and mortality, children and adolescents with co-infections face an elevated risk of death due to their young age and compromised immune systems. While risk factors for tuberculosis (TB) and adverse TB outcomes in HIV-infected adults are well-documented for mortality estimation, understanding mortality risks among HIV-infected children and adolescents, especially in the era of test and treatment and universal ART for all HIV-infected persons, remains limited. This study aimed to estimate the mortality rate among TB and HIV-co-infected children in Sub-Saharan African countries using SRM.

Methods: We systematically searched relevant studies from seven international electronic databases. Articles were searched using Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. Article searching included six electronic databases including PubMed/MEDLINE (N = 1287), Scopus (N = 447), Web of Science (N = 174), Science Direct (N = 749, Cochran (N = 57), and Google Scholar and research repository bases searching (N = 42). The quality of primary studies was evaluated using Joanna Briggs Institute (JBI) checklist. The pooled mortality rate was estimated using a weighted inverse variance random-effect meta-analysis. Heterogeneity among studies was assessed using Cochran's Q test and estimated using I2 statistic. This document is registered in Prospero (CRD420251012913).

Result: In this SRM, 16 individual studies were included. During the co-infected mortality screening of 5,098 participants, 657 deaths were reported after co-treatment started. The pooled mortality burden was estimated at 12.96% (95% CI: 8.94 to 16.98, I2 = 92.6%, P = 0.001). The majority of TB co-infected cases were newly diagnosed after ART started. The final weighted inverse variance random-effect regression indicated WHO stages III and IV (pooled HR = 4.34), poor/ fair ART adherence (pooled HR = 3.11), missed Isoniazid preventive therapy (IPT) (pooled HR = 3.07), hemoglobin levels ≤ 10 mg/dL (pooled HR = 2.84), bedridden functional status (pooled HR = 3.19), below threshold CD4 count (pooled HR = 1.80), and missed cotrimoxazole preventive therapy (CPT) (pooled AOR = 1.58) were predictors of premature death during co-infection.

Conclusion: In this review, the overall pooled burden of mortality in HIV-infected children in SSA countries was high compared with the End TB Strategy target estimation. Significant predictors of mortality included WHO clinical stages III and IV, poor or fair ART adherence, missed Isoniazid preventive therapy (IPT), and hemoglobin levels ≤ 10 mg/dL. Therefore, counseling on antiretroviral therapy adherence should be strengthened; early screening and treating of anemia, screening and scaling up of IPT, critical ART drug, and nutritional counseling should be done during regular visits for caregivers to prevent premature deaths among children, youths, and adolescents during co-infection in SSA.

Keywords: Children; Ethiopia; HIV; Mortality; Pediatrics; Predictors; TB.

Fig. 1

PRISMA Flow diagram for TB and HIV co-infected mortality

Fig. 2

Pooled mortality rate during TB and HIV co-infection for children, youth and adolescents in SSA. This review included 16 studies of TB and HIV co-infected children and adolescents from Sub-Saharan Africa. Of 5,098 assessed cases, 657 deaths were reported, resulting in a pooled mortality burden of 12.96% (95% CI: 8.94 to 16.98, I² = 92.6%, P = 0.001). Most TB cases were newly diagnosed after ART started

Fig. 3

TB treatment cure rate during co-infection with HIV/AIDS among children, youth and adolescents in SSA countries

Fig. 4

Pooled attrition rate from TB during HIV co-infection treatment in SSA countries

Fig. 5

Subgroup analysis by age of co-infected death rate of children, youths and adolescents in SSA

Fig. 6

Subgroup analysis by years of publication during co-infection of children, youths and adolescents in SSA

Fig. 7

Funnel plot for assessing of publication biases of included studies for final meta-analysis

Fig. 8

Biggs regression test to assessed test of effects of small study sample

Fig. 9

aA-Leave-one-out sensitivity analysis of TB and HIV co-infected mortality in SSA countries

Fig. 10

Galbriath plot for identification of outlier studies during met-analysis