Comparison of MIRU-VNTR genotyping between old and fresh clinical samples in tuberculosis.

Twenty-four loci mycobacterial interspersed repetitive unit-variable number tandem repeat analysis (MIRU-VNTR) is extensively used for genotyping and detection of polyclonal infections in tuberculosis. The aim of the present study was to compare the direct and indirect MIRU-VNTR genotyping and detection of polyclonal infections between old and fresh clinical samples. Two series of TB samples were collected for comparison. After genomic DNA extraction from clinical samples and their respective cultures, 24 loci MIRU-VNTR was performed. In the 14 old samples, no mixed infections were observed, in clinical samples and their respective cultures. In nine fresh samples, 44.4% of mixed infection was observed in the clinical samples, but no mixed infections were observed in their respective cultures. Surprisingly, in the old samples, 92.86% of samples (13/14) had an allelic change between clinical samples and their respective cultures. On the other hand, in fresh samples, only one sample (1/9) had an allelic change between clinical samples and their respective cultures. We concluded that 24 loci MIRU-VNTR undoubtedly is successful in direct genotyping of clinical samples, especially for the fresh samples. However, selecting starting material, such as clinical sample or respective culture can be controversial for the old samples. Regarding polyclonal infections, the fresh samples gives us a better view to detect these infections, especially in the clinical sample.