Increased Risk of Restenosis After Placement of Gold-Coated Stents
To the Editor:
The article “Increased Risk of Restenosis After Placement of Gold-Coated Stents,” by Kastrati, et al1 concludes that “One-year event-free survival was significantly less favorable in the gold-stent group” (versus the steel stent group), although in the introduction they suggest in vitro studies indicated a more favorable result. I think that a possible explanation of their unexpected result lies in a seldom-recognized mechanism of radiation deposition at the boundary between a high atomic number (Z) material (gold) and a relatively low Z material (tissue). I think that this mechanism is an uncontrolled and unaccounted-for source of potentially significant radiation energy deposition (through fluoroscopy) in the intima of the artery contacting the material of the stent and may explain their unexpected result.
X-ray photons are not directly absorbed by the materials through which they pass; instead, they interact with the electrons bound to the atoms of the material. The photons are scattered by the electrons; thus, the photons lose energy and the electrons are accelerated. The accelerated electrons interact with other electrons in a cascade effect, until eventually all lose enough energy to be recaptured by the atoms of the material. At the interface between a high Z material (stent) and a low Z material (tissue) exposed to x-rays, the absorption of the x-rays in the high Z material is much greater than that in the low Z material. At this interface, the current of electrons (accelerated by the x-ray photons) leaves the surface of the high Z material and propagates through the low Z material, so that the energy deposition in the low Z material near the interface with the high Z material is close to that expected in the high Z material.
I estimate that for the gold-plated stent exposed to fluoroscopy, the energy absorbed by tissue in contact with the gold (due to the electron current emitted by the gold) would be ≈1000 times that expected to be absorbed by the tissue from fluoroscopy alone. For the steel stent, the energy absorbed by tissue in contact with the stent would be ≈130 times that due to fluoroscopy alone. In each case, the range of the electron current emitted from the metal of the stent into the bordering tissue is a strong function of the electron energy spectrum (which is a function of the incident x-ray spectrum), but for simplifying assumptions, the range would be ≈80 microns. Thus, although the increase in energy deposition in the tissue bordering the metal of the stent is very large, it only occurs in a small portion of the tissue, but a portion that is significant relative to the dimensions of the endothelium. It is significant that a thin (≈100 micron) coating of a low Z material (eg, a polymer) on the stent would absorb the current from the metal of the stent and eliminate the enhanced radiation deposition in the bordering tissue.
Although experimentation is necessary to establish whether the enhanced radiation deposition in the endothelium is a clinically significant issue, I think it is clear from the estimates I have presented that it is at least an uncontrolled variable in any study of stents, especially those that include polymer-coated or drug-eluting stents, or in brachytherapy accompanying placement of stents.
An appendix detailing the rationale and assumptions for the calculations presented here is available on request from the author.