Magnetic resonance imaging plays a vital role in the preoperative diagnosis of and surgical planning for instability of the cervical vertebral column.1–4 However, implant-associated magnetic susceptibility artifacts detected after surgery raise concerns about the diagnostic use of MRI. Surgical implants made of metals with high magnetic susceptibilities relative to biological tissues introduce inhomogeneities in the static magnetic field, which causes artifacts attributable to signal loss from displaced resonance frequencies and signal dephasing.5,6 Susceptibility artifacts can obstruct or misrepresent portions of a displayed image, which can severely compromise an evaluator's diagnostic interpretation. The most commonly recognized susceptibility artifacts are represented by dark black regions indicative of a lack of signal, geometric distortions of anatomy, bright white regions of signal accumulation, and failure of fat saturation.7,8
Several imaging parameters, such as the type of pulse sequence, field strength, matrix size, field of view, slice thickness, bandwidth, and echo train length, can alter magnetic susceptibility artifacts.7,9–20 The ability to mitigate implant-related susceptibility artifacts by use of various imaging techniques, such as view-angle tilting, slice-encoding metal artifact correction, short echo-time projection reconstruction acquisitions, single-point imaging, prepolarized MRI, and postprocessing image correction, has been evaluated.21 In 1 veterinary study,22 a spin-warp TSE sequence improved depiction of the spinal cord margin by reducing the size of susceptibility artifacts related to stainless steel screws used to stabilize the atlantoaxial joint in small-breed dogs. Despite promising results for these proprietary sequences, many of these software packages are currently cost prohibitive and not readily available for use in veterinary medicine.
In addition to special sequences and adjustable parameters, susceptibility artifacts are also affected by the size, shape, orientation, and metallic composition of surgical implants.20,23–27 Although the titanium and stainless steel alloys commonly used in surgical implants are both within the paramagnetic range, the magnetic susceptibility of titanium is much more closely matched with that of biological tissues.28 Therefore, fewer local field inhomogeneities develop near titanium implants than near stainless steel implants. Studies9,22,29–34 have indicated that various titanium implants used in maxillofacial, orthopedic, and vertebral surgeries of humans induce less severe susceptibility artifacts than do stainless steel implants. Similar results have been obtained for the canine appendicular skeleton.32 However, evaluation of titanium-related MRI susceptibility artifacts in the cervical vertebrae of dogs is limited to the description of a single specimen that was part of the planning phase of 1 study.22 The authors of that study reported that the susceptibility artifacts related to titanium screws did not interfere with assessment of the spinal cord by use of conventional TSE sequences.22
The purpose of the study reported here was to characterize and compare magnetic susceptibility artifacts related to titanium and stainless steel monocortical screws in the cervical vertebrae of canine cadavers by use of routine TSE sequences. We hypothesized that titanium screws placed in a monocortical manner would cause mild susceptibility artifacts within the implanted vertebrae, but that evaluation of the spinal cord and intervertebral disks would not be impaired. Furthermore, we hypothesized that stainless steel screws would cause significantly greater susceptibility artifacts that would hinder interpretation of the neuroanatomic structures across 4 contiguous cervical vertebrae.
Materials and Methods
Sample
Twelve canine cadavers were obtained within 1 hour after the dogs were euthanized; dogs were euthanized for reasons unrelated to the present study. All dogs were adult mixed-breed pit bull–type dogs with a body weight between 26 and 38 kg. Lateral and ventrodorsal cervical radiographs were evaluated to confirm skeletal maturity. Cadavers with skeletal abnormalities, including spondylosis deformans, were excluded from the study. Whole-body fluoroscopy was used to eliminate specimens with metallic foreign material and microchips. Cadavers were stored at 3°C. The study was approved by the Institutional Animal Care and Use Committee at The Ohio State University College of Veterinary Medicine.
Cervical vertebral stabilization
Cadavers were allowed to warm to approximately 20°C for at least 1 hour prior to use in the study. Cadavers were arbitrarily allocated to 2 groups (6 dogs/group) for cervical vertebral stabilization. Dogs of one group received regular 3.5-mm stainless steel cortical screws,a and dogs of the other group received self-tapping 3.5-mm titanium cortical screws.a Surgical procedures were performed by a board-certified veterinary surgeon (BFH). Three monocortical screws were placed in a triangular manner into the ventral aspect of the vertebral bodies of both C4 and C5, as described elsewhere.35 Screws were incompletely inserted, which allowed the heads to be incorporated in 20 g of polymethylmethacrylate.b An appropriate amount of time was allowed for the polymethylmethacrylate to harden before the surgical site was closed in a routine manner. Implant positioning was confirmed by evaluation of postsurgical radiographs.
MRI sequences
Magnetic resonance imaging was performed with a 3.0-T magnetc and 16-channel head-spine coil. Imaging was performed with dogs positioned in dorsal recumbency in accordance with a standardized positioning protocol. Routine MRI sequences were acquired, including T1-weighted TSE, T2-weighted TSE, and T2-weighted STIR TSE, by use of imaging parameters (Appendix). Images for T1-weighted and T2-weighted sequences were obtained in the transverse, sagittal, and dorsal planes, and STIR sequences were acquired in the sagittal plane.
Anatomic ROIs
Magnetic susceptibility artifacts were scored by use of a modified Jarvik scoring system36 in 20 predetermined ROIs over 4 contiguous vertebrae (C3 through C6). Artifact scores were defined on the basis of severity of signal void or anatomic distortion (or both) as follows: 0 = no artifact, 1 = partly distinguishable with < 50% of the structure affected by artifact, 2 = partly distinguishable with > 50% of the structure affected by artifact, and 3 = completely indistinguishable. The ROIs included the intervertebral disks, vertebral end plates, ventral aspect of the vertebral canal, and multiple sections of the spinal cord. Images were evaluated by use of commercially available software.d Images were evaluated separately by 2 board-certified veterinary radiologists (EMG and AMH), 1 board-certified veterinary surgeon (BFH), and 1 senior resident in a veterinary radiology training program (BGJ).
Statistical analysis
For the purposes of statistical analysis, ROIs were allocated into 3 evaluation groups relative to the metallic implants and vertebrae (Figure 1). Groups were defined as cranial (C3), implanted (C4 and C5), and caudal (C6).
Schematic illustration of 20 anatomic ROIs in the cervical vertebrae and spinal cord of canine cadavers with stainless steel or titanium monocortical screws implanted in the ventral aspect of the body of both C4 and C5. The ROIs were allocated into 3 groups for statistical analysis as follows: cranial (C3), implanted (C4 and C5), and caudal (C6).
Citation: American Journal of Veterinary Research 78, 4; 10.2460/ajvr.78.4.458
Evaluator scores were assessed for normality by calculating descriptive statistics, plotting histograms, and performing the Anderson-Darling test for normality. Data violating the assumption of normality were reported as median and range. For statistical comparison, artifact scores (0 to 3) were summed for all 4 evaluators and 4 MRI sequences, which provided a total of 16 values with a range of 0 to 48. In this manner, a single total artifact score was calculated for each ROI. Artifact scores were compared between stainless steel and titanium screws at each ROI by use of Mann-Whitney U tests. Artifact scores at each ROI were summed for the 4 evaluators and compared among MRI sequences by use of Friedman tests. Values for κ statistics and their 95% confidence intervals were used to estimate agreement among evaluator-assigned artifact scores. Statistical analyses were performed by use of commercially available software.e,f Results were considered significant at P ≤ 0.05.
Results
Median values of the calculated total artifact scores for both the implanted and caudal groups were significantly higher for stainless steel screws than for titanium screws (Table 1; Figure 2). Thirteen ROIs for the titanium screws had median values of 0, whereas there was only 1 ROI (ROI 11.2) for the stainless steel screws that had a median value of 0. For the stainless steel screws, every median value for the cranial group was significantly lower, compared with the comparable location for the implanted or caudal groups. In spinal cords with titanium screws, all 5 ROIs for each cranial and caudal group had median values of 0, whereas the only ROI with a median value of 0 was for the spinal cord overlying the C4–5 intervertebral disk (ROI 6.1). Screw position was also evaluated on postsurgical radiographs. Typically, screws in C4 were 3.5 mm farther from the C3–4 intervertebral disk space, compared with the distance between screws in C5 and the C5–6 intervertebral disk space.
Representative T2-weighted images obtained in the sagittal plane of the cervical vertebral region of canine cadavers depicting susceptibility artifacts associated with stainless steel screws (A) and titanium screws (B). Notice the susceptibility artifacts (dashed arrows) and their location relative to the spinal cord. The artifacts associated with the stainless steel screws are worse, compared with the artifacts associated with the titanium screws. The hydrated unobstructed C4–5 intervertebral disk for the specimen implanted with titanium screws is indicated (asterisk). Cd = Caudal. Cr = Cranial. Dor = Dorsal. Vent = Ventral.
Citation: American Journal of Veterinary Research 78, 4; 10.2460/ajvr.78.4.458
Median (range) values for total artifact scores* of canine cadavers in which stainless steel (n = 6) or titanium (6) monocortical screws were implanted in the ventral aspect of the vertebral body of C4 and C5.
| Implant type | ||||
|---|---|---|---|---|
| Group† | ROI location | Stainless steel | Titanium | P value‡ |
| Implanted | 1 | 48.0 (48–48) | 11.5 (5–24) | 0.002 |
| 2 | 48.0 (48–48) | 19.0 (7–28) | 0.002 | |
| 3 | 48.0 (48–48) | 19.0 (10–25) | 0.002 | |
| 4 | 47.0 (46–48) | 15.0 (7–18) | 0.002 | |
| 5 | 48.0 (48–48) | 10.5 (8–19) | 0.002 | |
| 6.1 | 45.5 (36–47) | 0 (0–4) | 0.002 | |
| 6.2 | 44.5 (43–46) | 4.5 (4–10) | 0.002 | |
| 6.3 | 46.0 (43–48) | 4.0 (0–7) | 0.002 | |
| 7 | 32.5 (23–46) | 4.0 (0–9) | 0.002 | |
| 12 | 48.0 (46–48) | 5.5 (4–9) | 0.002 | |
| Total | 48.0 (23–48) | 8.0 (0 28) | 0.002 | |
| Cranial | 8 | 19.5 (12–39) | 0 (0–1) | 0.002 |
| 9 | 15.5 (5–30) | 0 (0–0) | 0.002 | |
| 10 | 0.5 (0–6) | 0 (0–0) | 0.180 | |
| 11.1 | 3.5 (0–13) | 0 (0–0) | 0.015 | |
| 11.2 | 0 (0–11) | 0 (0–0) | 0.394 | |
| Total | 5.0 (0–39) | 0 (0–1) | 0.002 | |
| Caudal | 13 | 48.0 (41–48) | 0 (0–1) | 0.002 |
| 14 | 47.0 (35–48) | 0 (0–0) | 0.002 | |
| 15 | 27.5 (8–36) | 0 (0–0) | 0.002 | |
| 16.1 | 43.0 (18–48) | 0 (0–0) | 0.002 | |
| 16.2 | 22.5 (3–34) | 0 (0–0) | 0.002 | |
| Total | 36.0 (3–48) | 0 (0–1) | 0.002 | |
| All | Total | 44.0 (0–48) | 0 (0–28) | 0.002 |
Magnetic susceptibility artifacts were scored in 20 predetermined ROIs over 4 contiguous vertebrae (C3 through C6) by use of the following scale: 0 = no artifact, 1 = partly distinguishable with < 50% of the structure affected by artifact, 2 = partly distinguishable with > 50% of the structure affected by artifact, and 3 = completely indistinguishable. Values reported were the sum of scores for 4 evaluators and 4 sequences (16 scores/vertebral specimen).
The ROIs were allocated into 3 evaluation groups relative to the metallic implants and vertebrae as follows: cranial (C3), implanted (C4 and C5), and caudal (C6).
Values were significant at P ≤ 0.05 (Mann-Whitney U test).
On MRI images obtained after the surgical procedures were completed, the most readily apparent susceptibility artifacts observed included black regions of signal void, bright white regions of signal accumulation, and geometric distortions of anatomic structures (Figure 3). Examination of postsurgical radiographs revealed that the screws implanted in C4 were consistently placed farther caudally in the vertebral body and from the C3–4 intervertebral disk space, compared with the placement of the screws in C5 in relationship to the C5–6 intervertebral disk space (Figure 4).
A T2-weighted TSE image obtained in the sagittal plane of the cervical vertebral region of a canine cadaver illustrating signal accumulation (solid arrows) and signal void attributable to misregistration or signal dephasing (dashed arrows).
Citation: American Journal of Veterinary Research 78, 4; 10.2460/ajvr.78.4.458
Schematic illustrations of the lateral (A) and ventral (B) views of C4 and C5 depicting variations in screw placement. Notice in each panel that the screw in C4 is farther from the C3–4 intervertebral disk space (dashed arrow) and the screw in C5 is closer to the C5–6 intervertebral disk space (solid arrow). Variations in screw angle are not depicted.
Citation: American Journal of Veterinary Research 78, 4; 10.2460/ajvr.78.4.458
Overall agreement among evaluators was good for all sequences for both the stainless steel (κ = 0.592; P < 0.001) and titanium (κ = 0.518; P < 0.001) screws. The highest agreement (κ = 0.627; P < 0.001) was for the T1-weighted sequence of vertebral columns implanted with stainless steel screws.
Discussion
Similar to results for studies26,37 of humans and a descriptive report22 for veterinary medicine, the study reported here found that titanium implants caused less severe magnetic susceptibility artifacts in the cervical vertebrae and spinal cord than did stainless steel implants. The most readily apparent susceptibility artifacts observed included black regions of signal void, bright white regions of signal accumulation, and geometric distortions of anatomic structures (Figure 3). Artifact scores and qualitative assessment within the type of metallic implant were similar among all TSE sequences, which is similar to results for a study32 conducted to evaluate implant-related artifacts in the canine stifle joint. Compared with artifacts for other readily available clinically applicable sequences, spin echo sequences are least affected by susceptibility artifacts.11,38 Spin echo sequences use a 180° refocusing pulse that reduces static field dephasing and decreases signal loss, compared with gradient echo sequences that do not use refocusing pulses.39 Because only TSE sequences were used in the present study, it was not surprising that there were no significant differences among sequences within the type of metallic implant. However, the sample size was not sufficient to provide post hoc multiple comparisons among MRI sequence, ROI, and type of metallic implant.
A decrease in magnitude of susceptibility artifacts related to the use of titanium implants has been clearly established; however, to our knowledge, the location of these artifacts relative to the major anatomic structures of the cervical vertebrae and spinal cord as determined by use of conventional TSE sequences has not been previously described in dogs. Analysis of results for the present study indicated that stainless steel screws caused significantly greater artifact scores than did titanium screws at every anatomic location evaluated, except for 2 ROIs. There were few, if any, artifacts detected at both of those ROIs irrespective of the type of metallic implant. Additionally, susceptibility artifacts related to both types of metallic implants were worse in the implanted vertebrae (ie, closer to the screws) than in the adjacent vertebrae. Titanium screws were associated with low artifact scores that were limited to the implanted vertebrae. In fact, the median values calculated for the total artifact scores of all ROIs outside vertebrae implanted with titanium screws were 0. Conversely, stainless steel screws were associated with high artifact scores that spanned 3 contiguous cervical vertebrae (C4 through C6). The location of the artifacts evident with stainless steel implants also differed from the location of artifacts for titanium implants.
Interestingly, fewer stainless steel–related artifacts were seen in C3 (cranial to the implanted site). In fact, artifact scores for the stainless steel implants at all ROIs cranial to C4 were significantly less than those for similar ROIs caudal to C5. The asymmetric distribution of susceptibility artifacts within the cervical vertebrae relative to the implanted sites was most likely caused by subtle differences in surgical screw placement. Examination of postsurgical radiographs revealed that the screws implanted in C4 were consistently placed farther caudally in the vertebral body and from the C3–4 intervertebral disk space, compared with placement of the screws in C5 in relationship to the C5–6 intervertebral disk space (Figure 4). Additional studies are needed to more accurately determine the artifact-free distance from cervical vertebral screws, which could help clinicians determine the diagnostic use of MRI prior to obtaining images.
In addition to screw position, there were differences in the angle of screws relative to the long axis of the vertebral column. It was the authors’ impression that these differences in the angular alignment of the screws correlated well with the minor variations between certain specimens within the same type of metallic implant. Screw angle can cause variations in susceptibility artifacts in humans.31 Susceptibility artifacts are elongated in the frequency encoding direction40; therefore, artifacts can be reduced by aligning important anatomic structures in the phase-encoding direction. In the present study, vertebrae of the cadavers were aligned parallel to the MRI table, irrespective of screw angle. Additional studies are needed to develop specific positioning techniques aimed at eliminating or diverting artifacts from important anatomic structures such as the spinal cord.
In the study reported here, titanium monocortical screws were associated with low artifact scores at the ROIs within the spinal cord of implanted vertebrae. Susceptibility artifacts were focally limited to the ventral aspect of the spinal cord immediately adjacent to the tips of the titanium screws. Despite these mild artifacts at the level of the implanted vertebrae, there was excellent overall agreement among evaluators for assessment of the spinal cord, which is similar to results of another report.22 However, additional studies are needed to confirm these results in clinically affected patients.
A limitation of the present study was that the metal-related artifacts evaluated on MRI images were not compared with artifacts for other imaging modalities (eg, CT). Images for CT are also subject to metal-related artifacts, albeit for extremely different physics reasons. Computed tomography beam-hardening artifacts cause streaks that radiate from metallic implants in the scan field and degrade image quality.41 Severity of streak artifacts is proportional to the proton density of the material, with stainless steel causing worse artifacts than are caused by titanium.33,42 Artifacts on CT can be reduced by use of a high kilovolts peak, high milliampere-seconds, narrow collimation, thinner slices, monoenergetic dual-energy techniques, or sonogram inpainting methods.43 The best 3-D imaging technique for postoperative assessment of patients with metallic implants is highly debated and depends on whether the evaluator is focused on bone or soft tissues.33,43–45 Therefore, CT and MRI should be considered complementary imaging modalities, rather than mutually exclusive imaging methods.
For the study reported here, titanium monocortical screws caused significantly fewer susceptibility artifacts, compared with results for stainless steel screws. Postsurgical cervical MRI for the evaluation of the spinal cord or adjacent intervertebral disks should not be avoided in patients with titanium monocortical screws. Stainless steel screws are more likely to cause moderate to severe susceptibility artifacts that will interfere with interpretation of important anatomic structures, including the spinal cord. Additionally, careful attention to the positioning and angle of screw insertion relative to key anatomy, as well as adjustable imaging factors, may help reduce overall susceptibility artifacts and improve image quality.
Acknowledgments
This manuscript represents a portion of a thesis submitted by Dr. Jones to the Department of Veterinary Sciences at The Ohio State University as partial fulfillment of the requirements for a Master of Science degree.
The authors thank Dr. Michael Knopp for facilitating MRI imaging.
ABBREVIATIONS
| ROI | Region of interest |
| STIR | Short tau inversion recovery |
| TSE | Turbo spin echo |
Footnotes
Synthes Vet, West Chester, Pa.
Simplex P bone cement, Stryker Corp, Mahwah, NJ.
Philips Achieva, Cleveland, Ohio.
eFilm, Merge Healthcare Inc, Chicago, Ill.
Stata, version 11.0, StataCorp, College Station, Tex.
IBM SPSS Statistics, version 21, IBM Corp, Armonk, NY.
References
1. Gutierrez-Quintana R, Penderis J. MRI features of cervical articular process degenerative joint disease in Great Dane dogs with cervical spondylomyelopathy. Vet Radiol Ultrasound 2012; 53: 304–311.
2. Lipsitz D, Levitski RE, Chauvet AE, et al. Magnetic resonance imaging features of cervical stenotic myelopathy in 21 dogs. Vet Radiol Ultrasound 2001; 42: 20–27.
3. De Decker S, Gielen IM, Duchateau L, et al. Low-field magnetic resonance imaging findings of the caudal portion of the cervical region in clinically normal Doberman Pinschers and Foxhounds. Am J Vet Res 2010; 71: 428–434.
4. da Costa RC, Parent J, Dobson H. Comparison of magnetic resonance imaging and myelography in 18 Doberman Pinscher dogs with cervical spondylomyelopathy. Vet Radiol Ultrasound 2006; 47: 523–531.
5. Lüdeke KM, Röschmann P, Tischler R. Susceptibility artefacts in NMR imaging. Magn Reson Imaging 1985; 3: 329–343.
6. Hargreaves BA, Worters PW, Pauly KB, et al. Metal-induced artifacts in MRI. AJR Am J Roentgenol 2011; 197: 547–555.
7. Hargreaves BA, Chen W, Lu W, et al. Accelerated slice encoding for metal artifact correction. J Magn Reson Imaging 2010; 31: 987–996.
8. Farahani K, Sinha U, Sinha S, et al. Effect of field strength on susceptibility artifacts in magnetic resonance imaging. Comput Med Imaging Graph 1990; 14: 409–413.
9. Lee MJ, Kim S, Lee SA, et al. Overcoming artifacts from metallic orthopedic implants at high-field-strength MR imaging and multi-detector CT. Radiographics 2007; 27: 791–803.
10. Harris CA, White LM. Metal artifact reduction in musculoskeletal magnetic resonance imaging. Orthop Clin North Am 2006; 37: 349–359.
11. Port JD, Pomper MG. Quantification and minimization of magnetic susceptibility artifacts on GRE images. J Comput Assist Tomogr 2000; 24: 958–964.
12. White LM, Kim JK, Mehta M, et al. Complications of total hip arthroplasty: MR imaging—initial experience 1. Radiology 2000; 215: 254–262.
13. Bagheri MH, Hosseini MM, Emami MJ, et al. Metallic artifact in MRI after removal of orthopedic implants. Eur J Radiol 2010; 81: 1–7.
14. Matsuura H, Inoue T, Ogasawara K, et al. Quantitative analysis of magnetic resonance imaging susceptibility artifacts caused by neurosurgical biomaterials: comparison of 0.5, 1.5, and 3.0 Tesla magnetic fields. Neurol Med Chir (Tokyo) 2005; 45: 395–399.
15. Lee MJ, Janzen DL, Munk PL, et al. Quantitative assessment of an MR technique for reducing metal artifact: application to spin-echo imaging in a phantom. Skeletal Radiol 2001; 30: 398–401.
16. Cha JG, Jin W, Lee MH, et al. Reducing metallic artifacts in postoperative spinal imaging: usefulness of IDEAL contrast-enhanced T1-and T2-weighted MR imaging—phantom and clinical studies. Radiology 2011; 259: 885–893.
17. Graf H, Steidle G, Martirosian P, et al. Metal artifacts caused by gradient switching. Magn Reson Med 2005; 54: 231–234.
18. Graf H, Lauer UA, Berger A, et al. RF artifacts caused by metallic implants or instruments which get more prominent at 3 T: an in vitro study. Magn Reson Imaging 2005; 23: 493–499.
19. Chen CA, Chen W, Goodman SB, et al. New MR imaging methods for metallic implants in the knee: artifact correction and clinical impact. J Magn Reson Imaging 2011; 33: 1121–1127.
20. Kolind SH, MacKay AL, Munk PL, et al. Quantitative evaluation of metal artifact reduction techniques. J Magn Reson Imaging 2004; 20: 487–495.
21. Koch KM, Hargreaves BA, Pauly KB, et al. Magnetic resonance imaging near metal implants. J Magn Reson Imaging 2010; 32: 773–787.
22. Griffin IV JF, Archambault NS, Mankin JM, et al. Magnetic resonance imaging in cadaver dogs with metallic vertebral implants at 3 Tesla. Spine 2013; 38: E1548–E1553.
23. Petersilge CA, Lewin JS, Duerk JL, et al. Optimizing imaging parameters for MR evaluation of the spine with titanium pedicle screws. AJR Am J Roentgenol 1996; 166: 1213–1218.
24. Camacho CR, Plewes DB, Henkelman RM. Nonsusceptibil-ity artifacts due to metallic objects in MR imaging. J Magn Reson Imaging 1995; 5: 75–88.
25. Fache JS, Price C, Hawbolt EB, et al. MR imaging artifacts produced by dental materials. AJNR Am J Neuroradiol 1987; 8: 837–840.
26. Ernstberger T, Heidrich G, Buchhorn G. Postimplantation MRI with cylindric and cubic intervertebral test implants: evaluation of implant shape, material, and volume in MRI artifacting—an in vitro study. Spine J 2007; 7: 353–359.
27. David FH, Grierson J, Lamb CR. Reducing susceptibility artefacts in magnetic resonance images of the canine stifle following surgery for cranial cruciate ligament deficiency. Vet Comp Orthop Traumatol 2012; 25: 488–497.
28. Schenck JF. The role of magnetic susceptibility in magnetic resonance imaging: MRI magnetic compatibility of the first and second kinds. Med Phys 1996; 23: 815–850.
29. Eggers G, Rieker M, Kress B, et al. Artefacts in magnetic resonance imaging caused by dental material. MAGMA 2005; 18: 103–111.
30. Suh JS, Jeong EK, Shin KH, et al. Minimizing artifacts caused by metallic implants at MR imaging: experimental and clinical studies. AJR Am J Roentgenol 1998; 171: 1207–1213.
31. Ganapathi M, Joseph G, Savage R, et al. MRI susceptibility artefacts related to scaphoid screws: the effect of screw type, screw orientation and imaging parameters. J Hand Surg [Br] 2002; 27: 165–170.
32. David FH, Grierson J, Lamb CR. Effects of surgical implants on high-field magnetic resonance images of the normal canine stifle. Vet Radiol Ultrasound 2012; 53: 280–288.
33. Knott PT, Mardjetko SM, Kim RH, et al. A comparison of magnetic and radiographic imaging artifact after using three types of metal rods: stainless steel, titanium, and vitallium. Spine J 2010; 10: 789–794.
34. Ebraheim NA, Savolaine ER, Zeiss J, et al. Titanium hip implants for improved magnetic resonance and computed tomography examinations. Clin Orthop Relat Res 1992; 275: 194–198.
35. Hettlich BF, Allen MJ, Pascetta D, et al. Biomechanical comparison between bicortical pin and monocortical screw/polymethylmethacrylate constructs in the cadaveric canine cervical vertebral column. Vet Surg 2013; 42: 693–700.
36. Jarvik JG, Robertson WD, Wessbecher F, et al. Variation in the quality of lumbar spine MR images in Washington State. Radiology 2000; 215: 483–490.
37. Ernstberger T, Heidrich G, Schultz W, et al. Implant detectibility of intervertebral disc spacers in post fusion MRI: evaluation of the MRI scan quality by using a scoring system—an in vitro study. Neuroradiology 2007; 49: 103–109.
38. Czervionke LF, Daniels DL, Wehrli FW, et al. Magnetic susceptibility artifacts in gradient-recalled echo MR imaging. AJNR Am J Neuroradiol 1988; 9: 1149–1155.
39. Tartaglino LM, Flanders AE, Vinitski S, et al. Metallic artifacts on MR images of the postoperative spine: reduction with fast spin-echo techniques. Radiology 1994; 190: 565–569.
40. Eustace S, Goldberg R, Williamson D, et al. MR imaging of soft tissues adjacent to orthopaedic hardware: techniques to minimize susceptibility artefact. Clin Radiol 1997; 52: 589–594.
41. Barrett JF, Keat N. Artifacts in CT: recognition and voidance. Radiographics 2004; 24: 1679–1691.
42. Cahir JG, Toms AP, Marshall TJ, et al. CT and MRI of hip arthroplasty. Clin Radiol 2007; 62: 1163–1171.
43. Gupta A, Subhas N, Primak AN, et al. Metal artifact reduction. Radiol Clin North Am 2015; 53: 531–547.
44. Tohtz SW, Rogalla P, Taupitz M, et al. Inter- and intraobserver variability in the postoperative evaluation of transpedicular stabilization: computed tomography versus magnetic resonance imaging. Spine J 2010; 10: 285–290.
45. Trammell TR, Flint K, Ramsey CJ. A comparison of MRI and CT imaging clarity of titanium alloy and titanium alloy with cobalt-chromium-alloy pedicle screw and rod implants in the lumbar spine. J Bone Joint Surg Am 2012; 94: 1479–1483.
Appendix
Summary of MRI sequences and imaging parameters used for the assessment of the cervical vertebrae and spinal cord of canine cadavers implanted with stainless steel or titanium screws.
| T2-weighted | T1-weighted | ||||||
|---|---|---|---|---|---|---|---|
| Parameter | Transverse | Sagittal | Dorsal | Transverse | Sagittal | Dorsal | Sagittal |
| TE (ms) | 120 | 110 | 110 | 8 | 8 | 8 | 60 |
| TR (ms) | 3205.89 | 4803.25 | 4803.25 | 591.81 | 528.95 | 528.95 | 5336.63 |
| TI (ms) | — | — | — | — | — | — | 210 |
| NEX | 2 | 2 | 2 | 2 | 2 | 2 | 2 |
| Slice (mm) | 3 | 3 | 3 | 3 | 3 | 3 | 3 |
| Interslice gap (mm) | 0 | 0 | 0 | 0 | 0 | 0 | 0.3 |
| Matrix | 180 × 180 | 280 × 264 | 280 × 198 | 180 × 180 | 280 × 280 | 200 × 200 | 280 × 265 |
| Pixel bandwidth | 193 | 400 | 423 | 222 | 334 | 335 | 334 |
| TSE factor (ETL) | 24 | 22 | 22 | 3 | 5 | 5 | 21 |
| FOV (mm) | 384 × 384 | 560 × 560 | 400 × 400 | 384 × 384 | 560 × 560 | 400 × 400 | 560 × 560 |
| Acquisition duration (s) | 205.18 | 240.16 | 182.52 | 365.78 | 131.71 | 131.71 | 202.79 |
— = Not applicable. ETL = Echo train length. FOV = Field of view. NEX = Number of excitations. TE = Echo time. TI = Inversion time. TR = Repetition time.
