Prevalence of tuberculosis, Correlations among diagnostic scoring systems and tuberculosis associated mortality in children admitted at Bugando Medical Centre in Mwanza, Tanzania
Material type:
Item type | Current library | Collection | Status | Barcode | |
---|---|---|---|---|---|
POSTGRADUATE DISSERTATIONS | MWALIMU NYERERE LEARNING RESOURCES CENTRE-CUHAS BUGANDO | NFIC | 1 | PD0039 |
Abstract:
Introduction: TB in children is a significant cause of morbidity and mortality in the developing countries while its diagnosis remains an unmet challenge in clinical practice throughout the world. Thus, diagnostic systems were developed to deal with these difficulties and provide a clinician with a rationalized, stepwise tool to identify children in need of TB treatment. These systems have not been used extensively on the wards of BMC; and this may be contributing to the under-diagnosis or over-diagnosis of TB in our hospital.
Objectives: To determine prevalence, correlation among four diagnostic systems (Keith Edwards, MASA, Fourie and Brazilian MoH diagnostic scoring system), clinical indicators and in-hospital mortality of TB in children aged ≥2 months to < 12years.
Materials and Methods: A prospective cross-sectional study in children ≥ 2 months to < 12 years old of age, admitted in the pediatric wards from October to December, 2012. All were screened using clinical symptoms and signs for TB. Later physical examination, Mantoux testing, CXR imaging and HIV testing was conducted in all TB suspects (Keith Edwards score ≥ 4, Mantoux test excluded.
Results: The prevalence of TB cases was 6.2% (36/585) and varied widely with different diagnostic systems (3.2 – 14.4%). There was poor to moderate correlation (r≤0.7) and relatively fair agreement (k≤0.4) among the scoring systems with TB diagnosis. The clinical indicators strongly associated with TB diagnosis were low weight for age, positive history of adult TB contact, positive Mantoux test, suggestive CXR and HIV infection. TB related in-hospital mortality was 13.9% (5/36).
Conclusion and Recommendations: TB is common among children admitted to our hospital but not as hypothesized. Uncritical use of any of the four diagnostic systems shall not be supported in the diagnosis of TB in our setting since there is poor to moderate correlation. Mantoux test is critical and should be easily available and performed on all children suspected of TB. Also, active screening using clinical symptoms and signs for TB should be done to all children aged > 2 months to ≤ 12 years admitted to hospitals in Africa.
There are no comments on this title.