Mitochondrial diseases can be caused by mutations of mitochondrial DNA(mtDNA) or nuclear genes, commonly defined by a lack of cellular energy. In cells, we can find co-existence of wild-type and mutated mtDNA molecules known as heteroplasmy. Detecting a low-heteroplasmic variance is crucial for a diagnosis of mitochondrial diseases, as a result of the bottleneck effect. The aim of this study was to verify the ability of next-generation sequencing for the detection of low-level DNA heteroplasmy. Sanger sequencing is a standard method of heteroplasmic variation detection, yet the lowest heteroplasmy ratio detected by the Sanger sequencing is around 15%. With an analysis of mitochondrial genome isolated from human blood we determined that next-generation sequencing is able to detect more than 5 % mtDNA heteroplasmy. Transfer of next-generation sequencing to clinical diagnostics is important, due to its ability to quickly detect mtDNA mutations. Moreover, a diagnosis of low-level heteroplasmy is very important for making difficult reproductive choices, since pathogenic mtDNA mutations can cause progressive and lethal diseases with no available cure.