N. Reynaert et Uo. Hafeli, Self-absorption correction for (32)p, (198) Au and Re-118 stents: Dose point kernel calculations versus Monte Carlo, MED PHYS, 28(9), 2001, pp. 1883-1897
Citations number
31
Categorie Soggetti
Radiology ,Nuclear Medicine & Imaging","Medical Research Diagnosis & Treatment
Monte Carlo simulations of dose distributions around radioactive stents are
very time intensive. Thus, in order to calculate the dose distribution aro
und a Re-188 stent, we chose to test a point kernel method, a method which
is known to be faster but the accuracy of which has not been established fo
r this application. The dose point kernel method, which takes into account
absorption in the strut material (=self-absorption). was based on different
beta-emitting point source distributions in water by itself and surrounded
by steel spheres of different thicknesses. This information was input into
an integration routine that modeled either a Palmaz-Schatz or Multilink st
ent. The dose distributions around Au-198 and P-32 stents calculated with t
he dose point kernel method were compared to those calculated using EGS4 an
d MCNP 4B Monte Carlo methods. The resulting correction for self-absorption
in steel was distance dependent and averaged 1.12 for P-32 and 1.25 for Au
-198 stents. The dose point kernel method gave nearly identical results to
these full Monte Carlo simulations and was thus used to calculate the dose
distributions around a (188) Re stent. Although Re-188 has a half-life of o
nly 17 hours, it is posited to be useful for radioactive restenosis prevent
ion, given that a recently developed rapid electrodeposition procedure allo
ws stents to be made radioactive, at predetermined activities. within 15 mi
nutes. The dose point kernel calculations of a Re-188-coated Multilink sten
t were compared to its radiochromic film measurements. The dose fall-off ag
reed with the calculations within 5% over 0.4 to 3.5 nim from the stent sur
face. The dose point kernel method is a valuable too] to determine depth do
se distributions around activated stents taking into account the detailed g
eometry and the self-absorption in the struts. It not only requires much le
ss processing time than Monte Carlo methods, but also allows the use of hig
her resolutions in modeling the geometry, which leads to more accurate self
-absorption correction factors. (C) 2001 American Association of Physicists
in Medicine.