Treatment outcomes for children with multidrug-resistant tuberculosis: a systematic review and meta-analysis.

BACKGROUND

Paediatric multidrug-resistant (MDR) tuberculosis is a public health challenge of growing concern, accounting for an estimated 15% of all global cases of MDR tuberculosis. Clinical management is especially challenging, and recommendations are based on restricted evidence. We aimed to assess existing evidence for the treatment of MDR tuberculosis in children.

METHODS

We did a systematic review and meta-analysis of published and unpublished studies reporting treatment outcomes for children with MDR tuberculosis. We searched PubMed, Ovid, Embase, Cochrane Library, PsychINFO, and BioMedCentral databases up to Oct 31, 2011. Eligible studies included five or more children (aged ≤16 years) with MDR tuberculosis within a defined treatment cohort. The primary outcome was treatment success, defined as a composite of cure and treatment completion.

RESULTS

We identified eight studies, which reported treatment outcomes for a total of 315 patients. We recorded much variation in the characteristics of patients and programmes. Time to appropriate treatment varied from 2 days to 46 months. Average duration of treatment ranged from 6 months to 34 months, and duration of follow-up ranged from 12 months to 37 months. The pooled estimate for treatment success was 81·67% (95% CI 72·54-90·80). Across all studies, 5·9% (95% CI 1·3-10·5) died, 6·2% (2·3-10·2) defaulted, and 39·1% (28·7-49·4) had an adverse event. The most common drug-related adverse events were nausea and vomiting. Other serious adverse events were hearing loss, psychiatric effects, and hypothyroidism.

INTERPRETATION

The treatment of paediatric MDR tuberculosis has been neglected, but when children are treated outcomes can be achieved that are at least as good as those reported for adults. Programmes should be encouraged to report outcomes in children to improve the knowledge base for care, especially as new drugs become available.

FUNDING

None.