Comments on potential health effects of MRI-induced DNA lesions: quality is more important to consider than quantity.

Magnetic resonance imaging (MRI) is increasingly being used in cardiology to detect heart disease and guide therapy. It is mooted to be a safer alternative to imaging techniques, such as computed tomography (CT) or coronary angiographic imaging. However, there has recently been an increased interest in the potential long-term health risks of MRI, especially in the light of the controversy resulting from a small number of research studies reporting an increase in DNA damage following exposure, with calls to limit its use and avoid unnecessary examination, according to the precautionary principle. Overall the published data are somewhat limited and inconsistent; the ability of MRI to produce DNA lesions has yet to be robustly demonstrated and future experiments should be carefully designed to optimize sensitivity and benchmarked to validate and assess reproducibility. The majority of the current studies have focussed on the initial induction of DNA damage, and this has led to comparisons between the reported induction of γH2AX and implied double-strand break (DSB) yields produced following MRI with induction by imaging techniques using ionizing radiation. However, γH2AX is not only a marker of classical double-ended DSB, but also a marker of stalled replication forks and in certain circumstances stalled DNA transcription. Additionally, ionizing radiation is efficient at producing complex DNA damage, unique to ionizing radiation, with an associated reduction in repairability. Even if the fields associated with MRI are capable of producing DNA damage, the lesions produced will in general be simple, similar to those produced by endogenous processes. It is therefore inappropriate to try and infer cancer risk by simply comparing the yields of γH2AX foci or DNA lesions potentially produced by MRI to those produced by a given exposure of ionizing radiation, which will generally be more biologically effective and have a greater probability of leading to long-term health effects. As a result, it is important to concentrate on more relevant downstream end points (e.g. chromosome aberration production), along with potential mechanisms by which MRI may lead to DNA lesions. This could potentially involve a perturbation in homeostasis of oxidative stress, modifying the background rate of endogenous DNA damage induction. In summary, what the field needs at the moment is more research and less fear mongering.