Author + information
- Damien Kenny, MB, BCh, MRCPCH,
- Mark C.K. Hamilton, MB, ChB and
- Robin P. Martin, MB, ChB⁎ ()
- ↵⁎Bristol Congenital Heart Centre, Bristol Royal Hospital for Children and Bristol Royal Infirmary, Upper Maudlin Street, Bristol BS2 8BJ, United Kingdom
We commend Thanopoulos et al. (1) for their complete follow up of 46 adult patients undergoing aortic stenting for both native and recurrent coarctation of the aorta (CoA). Although endovascular stenting is the treatment of choice in many centers for CoA in older children and adults, there have been no complete longer-term follow-up studies evaluating stent durability, stenosis, and aneurysm formation, and this study adds some reassuring data to the published reports. However, we have concerns about the use of magnetic resonance imaging (MRI) after implantation of stainless steel stents (Palmaz stents, Johnson and Johnson International Systems, Warren, New Jersey) to evaluate stent integrity and effects on the aortic wall. There have been numerous reports demonstrating almost complete loss of signal with stainless steel stents both ex vivo and in vivo when imaging with magnetic resonance, leading to significant image artefacts and obscuring of the vessel lumen (2–4). These artefacts are most problematic with steady state free precession (SSFP) and gadolinium angiography. Typical metallic stent artifact on MRI causes signal dropout due to magnetic susceptibility and radiofrequency shielding. Magnetic susceptibility scrambles the phases of individual spins leading to signal void, which is almost complete with stainless steel—particularly when compared with other alloys such as nitinol and platinum (2). Radiofrequency shielding refers to current induced in the stent wall that opposes the original magnetic field and leads to reduction in overall signal. This current increases with the resonance frequency, and thus shielding becomes more pronounced with high field strengths used in clinical imaging. Thus, MRI is not equipped to identify in-stent stenosis or aneurysm formation, and indeed lack of signal might give falsely reassuring appearances within the vessel lumen.
The authors used multislice computed tomography at the end of their 5-year follow-up, and we feel this imaging modality is not susceptible to the same artefact and signal loss as MRI (5). Although the authors did not demonstrate aortic aneurysm formation in their series, other larger albeit less complete series have demonstrated aneurysm formation of up to 9% (6), and this might have been missed with MRI, leading to potentially serious consequences. Indeed most aneurysm formation secondary to stenting is likely to occur soon after the procedure, and thus a 5-year delay for accurate imaging might lead to unnecessary patient risk. The magnetic resonance scanning, although safe, is expensive; thus to ensure both clinically relevant and cost-effective follow-up of patients undergoing aortic stenting in the setting of CoA, we suggest early (3 months) post-procedural computerized tomography imaging. Some might argue a significant radiation load accompanies this form of imaging; however, with limited scan length, nonelectrocardiogram gating, and use of tube modulation to reduce unnecessary current, it is possible to ensure that this is kept to a minimum.
- American College of Cardiology Foundation