Philippe L. Pereira, MD

 Accurate needle placement requires instrumentationfor good imaging contrast and high spatial resolution 

Interventional radiology plays an increasingly important role in the diagnosis and treatment of tumors. The trend toward minimally invasive surgical techniques brings a greater need for intraoperative guidance and/or monitoring with CT, MRI, or angiography. Many experts expect the typical operating room of the future to be based around such imaging modalities.

Both CT and MRI can be used to guide MR-compatible instruments within the body with considerable precision.1 Interventional MRI is especially likely to become well accepted as a clinical tool. It offers many advantages for anatomic visualization, including excellent soft-tissue contrast, capability for volume, temperature-sensitive, and near-real-time imaging, as well as for high-resolution imaging of vessel walls, flow measurement, and imaging of vessel conspicuity. Multiplanar imaging is possible in any plane with arbitrary slice thickness, without the need for patient repositioning. MRI contrast agents have good safety profiles, and neither operators nor patients are exposed to ionizing radiation (Figure 1).

Many of the first interventions to be performed under MR guidance were biopsies.2,3 MRI is now regarded as an attractive method of guiding, monitoring, and controlling tumor ablation, aspiration cytology, and percutaneous biopsies. Scanners may be combined with C-arm fluoroscopy and angiographic facilities when required.

Drawbacks of using MRI for intervention include the relatively long time required to acquire imaging information (especially for low-field systems), limited physical access, and the relatively high costs of hardware maintenance. Acquisition times can be reduced considerably, however, if modern scanners with strong gradient systems are used with appropriate MR fluoroscopic sequences.4 All interventional instruments and devices must be MR-compatible, as must any integrated therapy delivery modules.

The high costs of MRI are largely compensated for by the ease of interventions it guides and the ability to perform procedures that might otherwise be impossible. MR fluoroscopy may acquire one image per second. Tumors that can hardly be seen using ultrasound guidance may be well identified on MRI (Figure 2). CT and MRI complement each other rather than compete for applications (see table).

The introduction of open-bore and short closed-bore magnets, coupled with the availability of fast imaging sequences, has allowed MRI to become a valuable tool for the guidance and control of percutaneous, nonvascular, and vascular procedures. Open style scanners allow the interventionalist to have direct access to patients and permit precise localization, targeting, and near-real-time monitoring.

A number of options exist for guiding needle tip placement under MR fluoroscopy. These include sequences based on spin-echo imaging, such as the Local Look or Zoom Imaging technique,5 on fast T1-weighted gradient-echo or T2-weighted FISP imaging. The choice of sequence will depend on the target region.

Reports of MR-guided biopsies in the musculoskeletal system, liver, brain, and breast can now be found in the literature.6-16 One group reported the mean duration of biopsy to be 21 minutes.9 Another showed that information from MR-guided breast biopsies matched with histological diagnosis in 98% of cases. MR-guided interventions of the breast have been routinely performed in cases at a rate of 3% to 5% for all patients undergoing diagnostic contrast-enhanced breast MRI.13

Depiction of the needle, especially the tip, is extremely important in MR-guided biopsies. Either active or passive imaging can be used. An active imaging approach might involve the use of miniaturized coils at the needle tip. but such procedures are technically difficult and best avoided if passive imaging techniques could offer comparable accuracy. While passive tracking is much simpler to perform, it generally suffers from artifacts. These vary according to the needle's orientation, material composition, and shape, as well as the sequence type and magnetic field strength.17,18 Our own investigation into differing needle tips for passive tracking has revealed the variance in artifacts (Figure 3).

Artifacts induced by the susceptibility of instruments that are brought into the scanner's field-of-view can also cause a significant problem.19 Because ferromagnetic materials generally cause huge artifacts, instruments for MR-based interventions are made of less magnetic materials such as titanium, nitinol, or carbon fiber to minimize artifact size.

Metal biopsy needles can disrupt the imaging process in two ways. First, the excited slice is distorted around the needle. Second, variations in local magnetic field from the needle confound the frequency encoding process, resulting in signal mismapping in the frequency encoding direction. Intravoxel dephasing results in signal voids.

A study of needle depiction at 0.2T and 1.5T, examining multiple pulse sequences and needle orientations, showed that needle tips could be localized within 1 mm.17 Excellent image quality could be obtained for both field strengths.

Artifacts are generally less severe for low field strengths and for spin-echo techniques. They are most problematic when gradient-echo techniques with a long echo time are used. Good results have, however, been obtained with fast T1-weighted FLASH sequences. Needle design is another important factor. Dedicated MRI biopsy needles could indicate the tip position in fluoroscopic sequences by provoking artifacts at well-defined distances from the needle tip.

Likely artifacts from metallic devices on gradient-echo imaging can be estimated from calculations of the local magnetic field and intravoxel dephasing.18 This method makes it possible to predict the blooming ball-shaped signal void seen on MRI when a biopsy needle is parallel to the external magnetic field. Such a situation is actually quite unusual. Biopsy needles are more likely to be perpendicular to the exterior magnetic field when used in open MR scanners.

A study using computer simulations to predict artifacts concluded that needle composition, needle orientation, and pulse sequence can limit the accuracy of MR-guided needle tip placement.19 Methods to improve tip positioning under MRI are needed.

The dipole model is one way of predicting needle artifacts.18 This method, which computes artifacts on a subgrid, is intrinsically stable and can be used on large regions of interest. Computing the field distribution for a needle object space of size 30 x 30 x 400 on a grid sized 300 x 3 x 400 cells or more will take approximately six hours on a PC with a Pentium 4 processor. The computed artifact sizes are reliable estimates for artifact sizes based on the intravoxel dephasing effect. This approach addressess the situation within MRI using gradient-echo sequences.

The use of diamagnetic material could compensate for the effects of artifacts from paramagnetic materials. Diamagnetic material could also be used to create markers to indicate the true position of a needle tip. Modeling can help indicate the required geometry for such needles, without the need to manufacture prototypes. The dipole model can predict artifact behavior and needle depiction on MRI and is easy to implement. Model computations have also shown that negative markers, which consist of ring-shaped, spared-out regions in an additional diamagnetic coating of the needle tip, are superior to positive markers, which consist of diamagnetic bismuth rings added to the needle tip geometry.20

Radiologists should become familiar with likely artifacts prior to MR-guided biopsy procedures. This can be achieved by studying the simulated artifact geometry for various angles between the needle and exterior field. The effect of positioning the inner stylet can also be studied for puncture needles.

We obtained a multiple-point artifact with a prototype biopsy needle equipped with negative markers in an open 0.2T scanner (Concerto, Siemens Medical Solutions) and at 1.5T (Sonata, Siemens) using a fast FLASH T1-weighted sequence. Such an artifact might be an appropriate guide to tip placement when biopsying well-defined target regions such as small tumors. Because computations are not affected by noise, artifact creation can be studied on a more basic level using the numerical model.

The use of MR-guided interventional procedures is greatly supported by modern imaging facilities. Instruments used during such procedures need to be visualized without obscuring or distorting the underlying anatomy. Good imaging contrast and high spatial resolution have to be maintained for accurate needle placement. Use of the dipole model can help reduce the cost of designing new instruments for interventional MRI. Selection of optimal materials and geometry for developing new needles should be possible prior to prototype manufacturing with the use of computer models.

DR. MÜLLER-BIERL is project manager of the German Ministry of Research and Education (BMBF, project no. FKZ 01EZ0410, "Innovations in Medical Technology"), DR. KÖNIG is an interventional radiologist, PROF. DR. PEREIRA is research director of radiology and vice chair of diagnostic radiology, and PROF. DR. SCHICK is head of the section on experimental radiology, department of diagnostic radiology, all at the University Clinic Tübingen in Germany.

1. Lufkin RB, Gronemeyer DH, Seibel RM. Interventional MRI: update. Europ Radiol 1997;7(suppl 5):187-200.

2. Mueller PR, Stark DD, Simeone JF, et al. MR-guided aspiration biopsy: needle design and clinical trials. Radiology 1986;161(3):605-609.

3. Gehl HB, Frahm C. [MR-controlled biopsies]. Radiologe 1998;38(3):194-199. German.

4. Lewin JS, Nour SG, Duerk JL. Magnetic resonance image-guided biopsy and aspiration. Top Magn Reson Imaging 2000;11(3):173-183.

5. Buecker A, Adam G, Neuerburg JM, et al. MR-guided biopsy using a T2-weighted single-shot zoom imaging sequence (Local Look technique). J Mag Res Imaging 1998;8(4):955-959.

6. König CW, Trubenbach J, Fritz J, et al. Contrast enhanced MR-guided biopsy of hepatocellular carcinoma. Abdom Imaging 2004;29(1):71-76.

7. König CW, Pereira PL, Trubenbach J, et al. MR imaging-guided adrenal biopsy using an open low-field-strength scanner and MR fluoroscopy. AJR 2003;180(6):1567-1570.

8. Koenig CW, Duda SH, Truebenbach J, et al. MR-guided biopsy of musculoskeletal lesions in a low-field system. J Mag Res Imaging 2001;13(5):761-768.

9. Salomonowitz E. MR imaging-guided biopsy and therapeutic intervention in a closed-configuration magnet: single-center series of 361 punctures. AJR 2001;177(1):159-163.

10. Pretzsch M, Scholz R, Moche M, et al. [Musculoskeletal biopsies in an open 0.5-T-MR system]. Z Orthop Ihre Grenzgeb 2005;143(3):365-374. German.

11. Zangos S, Kiefl D, Eichler K, et al. [MR-guided biopsies of undetermined liver lesions: technique and results]. Rofo 2003;175(5):688-694. German.

12. Tronnier V, Staubert A, Wirtz R, et al. MRI-guided brain biopsies using a 0.2 Tesla open magnet. Minim Invasive Neurosurg 1999;42(3):118-122.

13. Fischer U, Kopka L, Grabbe E. Magnetic resonance guided localization and biopsy of suspicious breast lesions. Top Mag Res Imaging 1998;9(1):44-59.

14. Kuhl CK, Morakkabati N, Leutner CC, et al. MR imaging-guided large-core (14-gauge) needle biopsy of small lesions visible at breast MR imaging alone. Radiology 2001;220(1):31-39.

15. Schneider JP, Schulz T, Ruger S, et al. [MR-guided preoperative localization and percutaneous core biopsy of suspected breast lesions. Possibilities and experience on the vertically open 0.5-T-system]. Radiologe 2002;42(1):33-41. German.

16. Heywang-Kobrunner SH, Heinig A, Spielmann RP, et al. MR-guided needle biopsies of the breast. Sein 2001;11(3):219-222.

17. Lewin JS, Duerk JL, Jain JR, et al. Needle localization in MR guided biopsy and aspiration: effect of field strength, sequence design, and magnetic field orientation. AJR 1996;166(6):1337-1345.

18. Müller-Bierl B, Graf H, Steidle G, Schick F. Compensation of magnetic field distortions from paramagnetic instruments by added diamagnetic material: measurements and numerical simulation. Med Physics 2005;32(1):76-84.

19. Ladd ME, Erhart, P, Debatin JF, et al. Biopsy needle susceptibility artifacts. Magn Reson Med 1996;36(4):646-651.

20. Müller-Bierl B, Graf H, Pereira P, Schick F. Biopsy needles with markers: A numerical approach in assessing the influence of diamagnetic markers on needle artifacts. Presented at 14th International Conference of Medical Physics/39th Annual Congress of the German Society for Biomedical Engineering (BMT), Nurnberg, Germany; September 2005:368.

Articles

Accurate needle placement requires instrumentationfor good imaging contrast and high spatial resolution

Philippe L. Pereira, MD

Articles

MR-guided biopsies pose major challenges