| Sign In to gain access to subscriptions and/or personal tools. |
Artifacts from Dental Casting Alloys in Magnetic Resonance Imaging
1 Advanced Biomaterials, Correspondence: * corresponding author, shafii.orad{at}tmd.ac.jp
The potential advantage of magnetic resonance imaging (MRI) has been limited by artifacts due to the presence of metallic materials. For quantitative evaluation of the magnitude of artifacts from dental casting alloys and implant materials in MR imaging, 11 dental casting or implant materials were imaged by means of 1.5 T MRI apparatus with three different sequences. Mean and standard deviation of water signal intensity (SI) around the sample in the region of interest (1200 mm2) were determined, and the coefficient of variation was compared for evaluation of the homogeneity of the SI. A variety of artifacts with different magnitudes was observed. Only one of the samples, composed mainly of Pd, In, and Sb, showed no artifacts in all imaging sequences. We concluded that selection of specific dental casting alloys according to their elemental compositions could minimize the metal artifacts in MRI; however, titanium alloys currently pose a problem with respect to causing MRI artifacts.
Key Words: magnetic resonance imaging • metal artifact • dental casting alloy • titanium • implant
Magnetic resonance imaging (MRI) has become one of the most powerful diagnostic tools in radiology and diagnostic sciences. The advantages of MRI include its ability to provide sectional images of anatomic regions in any arbitrary plane and its excellent soft-tissue contrast resolution. MRI is based on the signal of nuclear magnetic resonance (NMR) emitted by the interaction of atomic nuclei that possess spin with incident radiofrequency within a static magnetic field. The NMR signal, which is primarily related to the proton density of the sample or tissue, also varies according to T1 or T2 relaxation time and velocity of fluid in the sample that is influenced by other internal and external factors (Vlaardingerbroek and den Boer, 1996). MRI thus gives excellent images of anatomical structures differing in proton density and other tissue characteristics. Despite these imaging characteristics, MRI has the shortcoming of being prone to magnetic susceptibility difference artifacts caused by the presence of metallic materials such as dental or orthopedic implants, dental cast restorations, and aneurysm clips (Bui et al., 2000). Artifacts in MR images can be defined as the pixels that do not faithfully represent the tissue components being studied (Brown and Semelka, 1995). All substances when placed in a magnetic field are magnetized to a degree which varies according to their magnetic susceptibility. There are three types of substances with different magnetic susceptibilities that need to be considered in MRI, namely, paramagnetic, diamagnetic, and ferromagnetic. Although the artifacts caused by ferromagnetic substances such as iron, nickel, and cobalt have been widely reported (Rudisch et al., 1998; Suh et al., 1998), the effects of dental casting alloys with compositions of various diamagnetic and paramagnetic substances or combinations of both have not yet been analyzed in detail. The purpose of our study was to test the hypothesis that, with selection of dental casting alloys or implant materials with specific composition of diamagnetic and paramagnetic substances, it is possible to minimize the metal artifacts in MRI.
Preparation of Samples To evaluate the effects of alloy composition on the magnitude of artifacts in MRI, we selected 11 kinds of dental casting alloys or implant materials. Currently, these dental casting alloys are clinically utilized in restorative and prosthodontics dentistry, and titanium alloys are widely used in implant dentistry and medical devices. Only one of these dental casting alloys contained ferromagnetic substances (Cobaltan). The others consisted of various percentages of diamagnets, paramagnets, or a combination of both. The compositions of these alloys with their casting conditions are listed in Table 1  .
To prepare a similar shape for all alloys, we used inlay wax to mold a cube with a 5.1-mm edge. The wax pattern was attached to a sprue positioned in the center of a crucible formed with a metallic casting ring and filled with the appropriate investment material (Table 1  ). Commercially pure titanium (Titan A) and titanium alloy were cast with the use of an argon arc melting centrifugal casting machine (Vulcant-T, Shofu, Kyoto, Japan) and the others by an electric resistance furnace motor drive centrifugal casting machine (Tiegelschleuder TS3, Degussa, Germany). After the casting process, all cast rings were kept at room temperature. All castings were divested, then separated from the sprue and ground to 5.0 mm from the edge by means of a cutting resin disc (#0, Ruby dental-products, Osaka, Japan). In all MR sequences, the cutting disc had been confirmed not to cause any artifacts. A cube (5-mm edge) was cut from the acrylic resin plate that was used as a control sample.
Phantom
MR Imaging
A region of interest (ROI) was drawn around the image of sample with the proprietary software installed for data acquisition, data processing, and image viewing in the MRI system (Fig. 1
Statistical Analysis Difference of mean SI per pixel in ROI between samples and control was analyzed by the Welch modified t test. The mean value was obtained from an average of 5 measurements of each sample. The difference between mean CV of the 5 measurements was evaluated by t test. A P value of less than 0.05 was considered significant.
Typical examples of MR images are shown for acrylic resin control, KIK-Wing, and Cobaltan with the use of T2 FSE and GE pulse sequences (Fig. 1 ). Artifacts appear as signal intensifications, signal voids, and blurring on nonlinear distortions (Figs. 1c, 1f). Both the commercially pure titanium (Titan A) and titanium alloys show artifacts in all 3 sequences.
Distribution of SI per pixel in ROI is shown in Fig. 2
Different magnitudes of artifacts were observed for individual dental casting alloys or implant materials, depending on pulse sequences. The mean values of SI per pixel in the ROI and SD obtained from the average of the 5 measurements and the CV with 95% confidence interval for longitudinal MR images of the samples are shown in Table 2 . The mean SI per pixel for Cobaltan is significantly different from that of acrylic resin control, but the difference was small in T1 and T2 FSE compared with a large difference in the CV. The mean value of KIK-Noble and Bior17 in T1 FSE sequences was significantly different in the SI from that of control but not different in CVs. The differences of SI in T1 FS and T2 FS between BiO17 and control and those between KIK-Noble and the control were not large; however, the differences were significant because the number of pixels was great. Nevertheless, the differences of CV in T1 FS and T2 FS between BiO17 and control and those between KIK-Noble and the control were not significant. However, the mean value of New Silver in T2 FSE was not different from control but was significantly different in CVs. In T1 FSE, there was no statistical difference in the CV between control and each of 6 dental casting alloys (New Silver, PGA3, Bior17, KIK-Noble, KIK-Wing, K14 Inlay) (p > 0.05). However, in T2 FSE and GE, only KIK-Wing showed no significant difference in the mean CV from control in both sequences and showed minimum metal artifact in the images.
These findings indicate that CV is a good measure for evaluating the magnitude of artifacts. All CV values showing a significant difference from the control value corresponded to the apparent metal artifact in the image recognized by visual examination. It was also reconfirmed that in T1 FSE, histograms of 6 dental casting alloys could be superimposed on that of the control, and in T2 FSE and GE, only KIK-Wing had a superimposable histogram.
MR imaging artifacts associated with a metallic object are primarily dependent on the inhomogeneity of the magnetic field and the magnetic susceptibility of the specific materials used to make the object, as well as on the amount of metal, shape, orientation, and position of the object in situ (Shellock and Kanal, 1998). Magnetic susceptibility is one of the physical properties of material and can be defined as the ratio of magnetic response of a material to the applied magnetic field (White, 1970). There are three types of substances, according to their magnetic susceptibility. Diamagnetic substances have no unpaired orbital electrons. When such a substance is placed in an external magnetic field, a weak magnetic field is induced in the direction opposite the magnetic field. Thus, diamagnetic substances have a small negative magnetic susceptibility and are basically non-magnetic. Paramagnetic substances have unpaired orbital electrons. Their induced magnetic field, under the external magnetic field, has the same direction relative to the external magnetic field. Consequently, their presence causes an increase in the effective magnetic field. Ferromagnetic substances are strongly attracted by a magnetic field and thus have high potential for MRI artifacts. Iron, cobalt, and nickel are three types of ferromagnets (Hashemi and Bradley, 1997). Currently, most clips and many implants are made of non-ferromagnetic materials such as titanium. Even so, when a patient with a "non-ferromagnetic" metal in the body is subjected to MRI, an artifact is produced which causes a drop-out of signal near the metallic surface (Bennett et al., 1996). Dental implants, orthopedic screws, and aneurysm clips are some examples of titanium alloys that produce metal artifacts on MR images, thus obscuring the images of tissues near the metallic objects. To overcome the "metal artifacts" problem and to improve image resolution around the metallic object, scientists have investigated several new sequences or signal manipulations which are not yet able to solve the problem completely (Rudisch et al., 1998; Viano et al., 2000). To our knowledge, no quantitative studies evaluating the magnitude of metal artifacts in MRI have been reported. There have been some paradoxical results reported in the literatures about metal artifacts from titanium alloys in spin-echo sequences. Some studies reported that titanium had no significant metal artifacts in spin-echo sequences (Teitelbaum et al., 1988; Vaccaro et al., 1994). In contrast, several authors have reported that titanium alloys produced high- to moderate-magnitude artifacts in the spin-echo sequence (Suh et al., 1998; Malik et al., 2001; Ganapathi et al., 2002). In the present study, we confirmed the appearance of moderate-magnitude artifacts from commercially pure titanium and titanium alloys in the T1 FSE sequence, which is least sensitive to metal artifacts (Suh et al., 1998). We also confirmed the appearance of high-magnitude artifacts in T2 FSE and GE sequences, which are highly sensitive to metal artifacts. The contradictory results reported in the literature might be due to differences in the parameters used in MRI, such as magnetic field intensity and specific sequences, trace amounts of ferromagnetic substances from the samples, and geometric factors in the imaging. All of these factors are known to have a significant influence on the metal artifacts in MR. One reason for the absence of artifacts in spin-echo sequences from titanium (Vaccaro et al., 1994) may be that the sample was with an extremely small amount of metallic debris. It seems difficult to detect any signal void or signal intensification in spin-echo sequences for that amount. In fact, even a stainless steel sample did not show any artifacts in spin-echo sequences, but apparently did in GE. Titanium alloy imaged in a low-magnetic field intensity of 0.35 T also showed no artifact (Teitelbaum et al., 1988). However, diagnostic efficacy of MRI in the low-magnetic field is generally limited. Trace amounts of iron (0.15–0.25%) in the commercially pure titanium sample used in the present study, confirmed by the manufacturer, could be another reason for differences in the magnitude of artifacts. Six dental casting alloys in the T1 FSE sequence had shown minimum artifacts. These 6 alloys were composed of one paramagnetic element with other diamagnetic substances (KIK-Noble, KIK-Wing, Bior17, PGA3) or only pure diamagnetic substances (New Silver and K14 Inlay). New Silver and K14 Inlay showed artifacts in both T2 FSE and GE sequences. Bior17, titanium with a high percentage of gold, and PGA3 platinum, with high percentages of gold and silver, have showed minimum artifacts in T1 FSE but showed artifacts of only moderate magnitude in T2 FSE and GE. The main elemental composition of 2 other dental casting alloys (KIK-Noble, KIK-Wing) is palladium. Pure palladium is one of the elements that has the highest positive magnetic susceptibility among paramagnets (Weast et al., 1984). However, in the present study, a combination of palladium with diamagnetic substances such as indium, antimony (KIK-Wing), and gold (KIK-Noble) showed little artifact in T1 FSE. Only KIK- Wing showed no artifact in two other sequences, T2 FSE and GE.
In the GE sequence—which superimposes a small magnetic field onto the main field and is therefore the most metal-sensitive sequence (Hashemi and Bradley, 1997)—all dental casting alloys and implant materials showed high-magnitude artifacts except KIK-Wing (Fig. 1e There would thus be no difficulty in the diagnostic interpretation of MRIs from head and neck regions in patients with dental casting alloys that do not disturb the magnetic field. Therefore, materials for prosthetic restoration should be selected based not only on their biological compatibility and functional and esthetic qualities, but also on whether they generate minimum artifacts in MRI.
The authors acknowledge Prof. Akimasa Ishida, Biomedical Information, Institute of Biomaterials and Bioengineering, for his discussion in the statistical analysis. This study was supported by Grants-in-Aid (13470398) from the Japan Society for the Promotion of Science. This paper is based on a thesis submitted to the Graduate School, Tokyo Medical and Dental University, in partial fulfillment of the requirements for the PhD degree. Received for publication October 15, 2002. Revision received March 31, 2003. Accepted for publication May 28, 2003.
Journal of Dental Research, Vol. 82, No. 8,
602-606 (2003) This article has been cited by other articles:
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
|
Sign In to gain access to subscriptions and/or personal tools.
TOP
ABSTRACT
INTRODUCTION
