Abstract:
uberculosis continues to be a major problem especially in Sub-Saharan
Africa where the spread is enhanced by HIV infection. TB/HIV co-infection has a high
morbidity and mortality, therefore a quick TB diagnosis for early initiation of therapy
is necessary. The application of Determine TB LAM strip test for non-sputum clinical
samples in the diagnosis of suspected tuberculosis in Western Kenya population has
not been evaluated. We are reporting on the use of urinary LAM test for presumptive
TB patients who are -HIV-1 infected in MTRH, Kenya. This was a cross-sectional
analyses of morning urine samples from 140 suspected HIV-1/TB co-infected adults
who are not on medication with CD4+ count <250 cells/mm3 for the presence of
Mycobacterium tuberculosis (MTB) LAM in urine. One hundred and fourty patients
were recruited; 37 (27.8%) tested positive for tuberculosis based on LAM test.
Diagnostic accuracy was based on urine culture and sputum microscopy which was
included for comparison. The sensitivity of LAM test against Sputum Microscopy was
36.4%, with a PPV of 22.0% while the specificity of LAM test was 73.9%, with a
NPV of 85.0%. Sensitivity increased to 60% for those with CD4≤100 cells/mm3 whereas the specificity slightly increased to 76.4% for those with
CD4>100cells/mm3.The comparison of LAM test against urine culture for TB were
similar and it increased the sensitivity and specificity to 100% and 80.7% respectively.
Stratified by CD4 categories, LAM test against urine TB culture increased the
sensitivity to 100% and specificity to 85.5% for HIV-infected persons with CD4
>100cells/mm3. This study showed that urinary LAM can be used as an adjunct test
for diagnosis of active TB and in combination with other tests in the diagnosis
platform.