Several reports described an age-related accumulation of a particular mitochondrial DNA (mtDNA) deletion ('common deletion') in post-mitotic tissues. These findings led to the hypothesis that free radicals generated inside the mitochondria could damage mtDNA during a normal life span. The impaired electron transfer function resulting from mtDNA damage would increase the production of free radicals creating a vicious cycle. If this vicious cycle is an important player in the somatic accumulation of mtDNA deletions, patients with impaired oxidative phosphorylation (regardless of the primary defect) should have an accelerated accumulation of mtDNA deletions. We tested this hypothesis by performing three analyses: (a) comparing the amounts of the mtDNA 'common deletion' in normal controls and patients with genetically characterized mitochondrial disorders associated with pathogenic mtDNA point mutations or deletions other than the common deletion; (b) analyzing the co-segregation of the age-related mtDNA common deletion with a pathogenic mtDNA point mutation; and (c) by the detection of multiple mtDNA deletions by long PCR in controls and patients with mitochondrial disorders. We observed a positive correlation between age and common deletion levels in controls (r = 0.80) and patients (r = 0.69). The slopes of the curves were similar, suggesting that the rate of accumulation of the age-related common deletion was the same in both groups. We could not find a co-segregation of the pathogenic point mutated mtDNA molecules with the common deletion nor increased number of age-related deletions in patients. Our data do not support the hypothesis that a vicious cycle (damage to mtDNA would affect the respiratory function, leading to the generation of more free radicals, which in turn would provoke additional mtDNA damage) is an important factor in the accumulation of age-related mtDNA deletions.