Cardiac MR imaging genotoxicity?

Magnetic resonance (MR) imaging, including cardiac MR (CMR), is an important, widely used diagnostic tool, and is generally considered a safer alternative to X-ray and radioisotope imaging where there is a clear relationship between exposure to ionizing radiation and increased cancer risk. MR imaging exposes the patient to a mixture of static magnetic fields, time-varying gradient fields, and pulsed radiofrequency fields, while the exposure of the operator is typically dominated by the static fields. There have been a number of reviews by national and international committees on the potential long-term health risks, and the current consensus is that there is no clear link between fields associated with MR imaging and subsequent health risk. However, recently there has been increased concern about the potential risks associated with MR imaging due to a limited number of studies reporting an increase in DNA damage following exposure, and this has resulted in calls to limit its use. These and related studies on the potential of MR imaging to result in genotoxicity have also been the subject of a number of recent reviews specifically on genotoxicity. These also conclude that the ability for MR imaging to produce DNA lesions has yet to be robustly demonstrated, and identify the need for more carefully designed comprehensive studies. One of the main ways of studying the potential adverse effects associated with MR imaging has been to study DNA damage using cH2AX assays as a measure of DNA double-strand breaks (DSBs). The formation of a DSB results in the phosphorylation of surrounding H2AX histones, forming cH2AX foci. There have been two key in vivo studies reporting an increase in DNA damage following exposure using this assay. Fiechter et al. reported a statistically significant enhancement in DSBs (quantified using immunofluorescence microscopy and flow cytometry) in human lymphocytes taken from 20 patients directly following contrast-enhanced 1.5 T CMR, when compared with samples taken prior to imaging (see Table 1). Lancellotti et al. reported a significant enhancement in DSBs (quantified using flow cytometry) in T lymphocytes at day 2 [median fluorescence intensity (MFI) = 397 ± 215] and 1 month (MFI = 529 ± 424) postexposure to unenhanced 1.5 T CMR on 20 healthy men, when compared with samples taken prior to imaging (MFI = 238 ± 88). However, no enhancement was observed at 1 h (MFI = 275 ± 114) and 2 h (MFI = 282 ± 155) post-exposure. The more recent cH2AX studies of Brand et al. (45 patients) and Critchley et al. (64 patients), the latter found in this issue of the journal, did not show an enhancement 5 min post-1.5 T contrastenhanced CMR in vivo patient exposure (see Table 1). Interestingly, the mean number of foci per cell observed by Brand et al. using immunofluorescence microscopy is significantly smaller, and more consistent, than the corresponding data of Fiechter et al., varying from 0.09 to 0.17 (compared with pre-exposure values ranging from 0 to 0.661 for Fiechter et al.). These values are consistent with typical background levels of < 0.2 foci per lymphocyte reported for ionizing radiation biodosimetry studies. While significant differences (both positive and negative) were seen by Fiecheter et al. when directly comparing pre-exposure and post-exposure response of individuals, the mean excess foci observed by Brand et al. was 0.001 ± 0.001 DSBs per cell. The methodology used by Critchley et al. is similar to that used by Fiechter et al., with blood drawn before and after MR imaging of patients, quantified using flow cytometry, and analysed in a blinded fashion. This study was carried out on 64 consecutive consenting patients and this constitutes the largest study to date. All patients received 0.1 mmol/kg Gadovist and underwent a standard ‘viabilitytype’ clinical CMR using a 1.5 T scanner taking 42 ± 11 min (compared with 68 ± 22 min for Fiechter et al. and 30–60 min for Brand et al.). CMR was found not to be associated with a significant change of expression in both T cells and monocytes (see Table 1); however, there were significant inter-patient variations in expression, with both large increases and decreases observed following CMR. In vitro studies were also performed on blood samples from 22 healthy volunteers. Again no difference in cH2AX expression was observed in T cells or monocytes, between blood receiving CMR exposure and