Burden of disease and risk factors for death among children treated for tuberculosis in Malawi

Abstract

SUMMARY SETTING Tuberculosis is a leading cause of childhood death. There is limited patient-level data on pediatric TB in Malawi that can be used to guide programmatic interventions. OBJECTIVE To describe pediatric tuberculosis case burden, disease pattern, treatment outcomes, and risk factors for death and poor outcome. DESIGN We conducted a retrospective cohort study utilizing routinely collected data in treatment registers. Odds ratios for factors associated with poor outcome and death were calculated separately using generalized estimating equations. RESULTS Children represented 8% (371/4,642) of tuberculosis diagnoses. Median age was 7(IQR 2.8–11) and 32.8%(113/345) were HIV-infected. Of these, 54.0% were on antiretroviral therapy (ART) at time of anti-tuberculosis treatment (ATT) initiation, 21.2% started ART during ATT, and 24.8% had no documented ART. The treatment success rate was 77.3% (11.2% cured, 66.1% completed treatment) with 22.7% experiencing a poor outcome (9.5% died, 13.2% were lost to follow-up). Being on ART at time of ATT initiation was associated with an increased odds of death compared to initiating ART during ATT(aOR=2.75,95%CI 1.27–5.96). CONCLUSION Children represent a small proportion of diagnosed tuberculosis cases and experience poor outcomes. Higher odds of death among children already on ART raises concerns over the management of these children in routine settings. Further discussion and research is warranted on pediatric-specific strategies to improve outcomes among these children.

Deaths during tuberculosis treatment among paediatric patients in a large tertiary hospital in Nigeria

“…[ 12 ] In Malawi, a nation-wide survey of programme data in 1998 reported that 17% of the children died; [ 13 ] while analysis of 3-year (2011–2013) programme data from treatment centres in central Malawi reported 9.5% deaths. [ 14 ] Likely inclusion of more severe cases with higher mortality risk, would at least in part explain the difference between these studies and our findings.…”

Epidemiology of Pediatric Tuberculosis in Kenya and Risk Factors for Mortality during Treatment: A National Retrospective Cohort Study

“…We found that more than one-quarter of children with TB in Kenya were HIV infected, which is slightly lower than the 31% co-infection rate among adults with TB in Kenya and the 32% co-infection rate reported in a sample of Malawian children with TB, 19,20 but still substantial. Prevention of TB among HIV-infected children is a key priority in Kenya.…”

“…27 However, unlike a study from Malawi, we did not find that children on ART at the time of starting TB treatment were more likely to die than children who initiated ART after being diagnosed with TB. 20 The use of ART among TB/HIV co-infected children reduces morbidity and mortality. 22,28 Our observation that HIV-infected children on ART still had nearly 4 times the likelihood of death as HIV-negative children could be due to early mortality among children who were started on ART when their disease was already advanced, virologic failure, immune reconstitution syndrome, or poor adherence to ART.…”

Loss to Follow-Up (LTFU) during Tuberculosis Treatment

“…Attention should also be paid toward LTFU among certain special populations. The proportion of LTFU among childhood TB patients ranges from 4 to 37% [52][53][54][55][56][57]. Among the children with drug-resistant TB, it ranges from 5 to 19.09% [58][59][60].…”