Published online before print March 1, 2002, 10.1148/radiol.2231010900
(Radiology 2002;223:127-136.)
© RSNA, 2002
Interventional Musculoskeletal Procedures Performed by Using MR Imaging Guidance with a Vertically Open MR Unit: Assessment of Techniques and Applicability1
Justin W. Genant, BA,
Jan E. Vandevenne, MD,
A. Gabrielle Bergman, MD,
Christopher F. Beaulieu, MD, PhD,
Stephen T. Kee, MD,
Alexander M. Norbash, MD and
Philipp Lang, MD, MBA
1 From the Department of Radiology (J.E.V., A.G.B., C.F.B., S.T.K., A.M.N., P.L.) and School of Medicine (J.W.G.), Stanford University Medical Center, Stanford, Calif; Department of Radiology, Universitair Ziekenhuis Antwerpen, Belgium (J.E.V.), and Department of Radiology, Brigham and Women’s Hospital, 75 Francis St, ASB-1, Floor L-1, Rm 003E, Boston, MA 02115 (A.M.N., P.L.). Received May 9, 2001; revision requested June 15; revision received October 1; accepted October 22. Address correspondence to P.L. (e-mail: pklang@partners.org).
 |
ABSTRACT
|
|---|
PURPOSE: To evaluate the safety of and time required for a broad range of musculoskeletal interventional procedures performed by using magnetic resonance (MR) imaging guidance with a vertically open 0.5-T unit.
MATERIALS AND METHODS: Sixty-three MR imaging–guided procedures were performed. A vertically open MR unit equipped with in-room display monitors allowed interactive freehand MR guidance predominantly with fast spin-echo and gradient-echo sequences. Each procedure was classified in terms of the anatomic location, procedure type, and tissue type involved. The procedures were evaluated for success of needle placement, adequacy of tissue sampling, total procedural time, needle time, number of needle passes, and complications.
RESULTS: Procedures consisted of tissue sampling with core-needle (n = 6) or fine-needle aspiration (n = 20) biopsy, corticosteroid or contrast agent injection (n = 19), joint cyst aspiration (n = 7), and drainage (n = 11). Successful needle placement was achieved in all 63 cases. Cytologic and histologic tissue samples were sufficient for pathologic diagnosis in 24 of 26 cases. In two cases, complications occurred: transient local bleeding and a brief vasovagal episode. The mean total procedural time was 64.8 minutes; the mean needle time, 26.2 minutes; and the mean number of needle passes per patient, 1.6.
CONCLUSION: With use of a vertically open MR unit, MR-guided interventional procedures involving bone, soft tissue, intervertebral disks, and joints are safe and sufficiently rapid for use in clinical practice.
© RSNA, 2002
Index terms: Biopsies, 30.1261, 30.1262, 40.1261, 40.1262 • Interventional procedures, 30.1261, 40.1261 • Magnetic resonance (MR), guidance, 30.121411, 30.121412, 40.121411, 40.121412
|
INTRODUCTION
|
|---|
The qualities of magnetic resonance (MR) imaging, including superior soft-tissue contrast, multiplanar imaging capabilities, and the ability to enable tracking of temperature changes, have led to the high potential of MR imaging for use in guiding interventional procedures. However, limited access to the patient during MR imaging and the presence of a magnetic field and gradients have challenged engineers and physicians in their attempts to make MR imaging–guided interventional procedures widely acceptable. With the development of MR systems that allow better access to patients and the creation of an adequate array of MR-compatible instruments, several groups have described the results of MR imaging–guided interventional procedures (1–6), which have been performed mainly in the abdomen or the head and neck. Most of the experiences thus far have been with horizontally open magnets (2,4,7), and data on vertically open MR units are limited (1,8). Experience with MR imaging guidance of musculoskeletal procedures also is limited, and the reports to date either have focused on one type of musculoskeletal interventional procedure (7,9–11) or have been studies on subsets of other nonmusculoskeletal interventional procedures (1,2). The purpose of our study was to evaluate the safety and time requirements of MR imaging guidance in a broad range of musculoskeletal interventional procedures performed with a vertically open 0.5-T MR unit.
|
MATERIALS AND METHODS
|
|---|
Subjects
Fifty-five consecutive patients who were examined with a vertically open MR unit between January 1997 and March 2000 were included in the study. A total of 63 MR imaging–guided interventional musculoskeletal procedures were performed in these 55 patients, who ranged in age from 19 to 94 years. Patients with pacemakers, vascular clips, or other contraindications to MR imaging were excluded from the MR imaging–guided procedures. Informed consent was obtained from each patient before he or she underwent a procedure. At Stanford University Medical Center, neither institutional review board approval nor informed consent from the patient is required to review patient images or files for the purposes of unfunded research, provided that all patient identifiers are removed from the data, as was the case in our study.
MR Imaging Unit
All procedures were performed by using a 0.5-T open MR unit (Signa SP; GE Medical Systems, Milwaukee, Wis) (Fig 1). This unit has a double-doughnut configuration with a central vertical opening, which provides the operator with unrestricted vertical and side access to any region of the patient’s body. Liquid crystal panels within the vertical opening displayed the MR images and allowed the radiologist to perform the interventional procedure interactively and view the MR images as they were acquired. In our study, microphones and speakers in the MR suite and control room enabled continuous verbal communication between the radiologist and the technologist operating the MR unit. Flexible surface transmit-and-receive radio-frequency coils were used in all procedures. Prior to the MR imaging–guided interventions, all lesions had been imaged with ultrasonography (US), computed tomography (CT), or 1.5-T MR imaging.
MR Imaging Sequences
Prior to needle placement, at least one conventional spin-echo, fast spin-echo, or gradient-echo sequence was used to localize the lesion or the anatomic region of interest. When needed, fat suppression was performed by using inversion recovery with a fast spin-echo MR sequence (4,000/100 [repetition time msec/echo time msec], echo train length of four, 256 x 192 matrix, three signals acquired, 4-mm section thickness); alternatively, fat- and water-selective MR images were obtained by using the three-point Dixon method (12) with a spin-echo sequence (400/28, 220-mm field of view, 256 x 192 matrix, 4-mm section thickness). The imaging planes were chosen on the basis of the expected location and contour of the lesion. For example, in the knee joint, lesions typically were imaged in the sagittal and transverse planes, unless the lesion was located far medially or far laterally. The skin entry site and biopsy trajectory were defined by using fast spin-echo T1- or T2-weighted images (400/19 or 4,000/108).
During needle placement, spoiled gradient-echo (120.0/6.6, 256 x 128 matrix, one signal acquired, 10-mm section thickness) or fast spin-echo (200/102, 256 x 128 matrix, one signal acquired, 5-mm section thickness) sequences were used to generate images in a three- or five-section mode. The fast gradient-echo sequence (14.1/6.6, 60° flip angle, 256 x 128 matrix) is a single-section mode in which an image was displayed approximately every 2 seconds. Orientation of the needle trajectory was most often transverse, and the imaging planes were mostly parallel and perpendicular to the needle. Oblique and double-oblique planes were used when needed. During needle placement, the radiologist used an MR imaging–guided freehand technique and relied on the images displayed within the MR suite for guidance.
After the needle was withdrawn, additional diagnostic sequences were performed for procedures that involved gadolinium-based contrast agent injection into a joint space or tendon sheath. These sequences included the acquisition of spin-echo T1-weighted images (800/21) and spin-echo three-point Dixon images in two or three orthogonal planes. For interventional procedures in which postprocedural hemorrhage was a concern, the radiologist performed an additional T1-weighted spin-echo or T2-weighted fast spin-echo sequence (4,000/100, three signals acquired, echo train length of four or 16) to rule out hemorrhage.
Needles
Several types of commercially available MR imaging–compatible needles were used. A 22-gauge MR imaging–compatible Lufkin aspiration cytology needle (E-Z-Em, Westbury, NY) was used in 11 cases involving soft tissues or joints, including all five tenograms. The lengths of this needle include 5, 10, 15, and 20 cm. For many of the fine-needle aspiration biopsies, a 21-gauge aspiration needle (Technicut; Manan, Northbrook, Ill) was used with a coaxial approach through an 18-gauge MR imaging–compatible histologic needle (E-Z-Em). This coaxial approach enabled the radiologist to obtain multiple aspiration samples without repositioning the outer coaxial needle. A 21-gauge biopsy needle (Surecut; TSK Laboratory, Tochigi, Japan), which can enable the acquisition of a cylindrical core tissue sample and a cytologic sample, was used in four cases, all of which also involved the use of an 18-gauge coaxial needle.
To obtain core-needle biopsy samples, two devices were used: a 14-gauge semiautomatic biopsy gun (Daum, Schwerin, Germany) and a 14-gauge side-cutting biopsy needle (Daum). Special needles were not required for sampling of bone lesions because all of the bone lesions in this series had either caused erosion of the cortex or extended into the adjacent soft tissue.
Although algorithms for the selection of biopsy techniques for imaging-guided sampling of tissue in the musculoskeletal system have been proposed (13), in the current study, the attending radiologist determined on a case-by-case basis whether core-needle biopsy, fine-needle aspiration biopsy, or both would be performed.
Description of Procedures
Patients were placed in a supine, prone, or decubitus position on the table of the MR unit. Each procedure was performed by one of five attending radiologists (A.G.B., C.F.B., S.T.K., A.M.N., P.L.) who had 5–15 years of clinical experience. To identify the needle entry site, the radiologist placed his or her fingertip at different positions on the patient’s skin while multiple fast gradient-echo or spoiled gradient-echo images were acquired at that level. The images were displayed within the MR unit; this allowed the radiologist to visualize and interactively adjust the position of the fingertip relative to both the lesion and the other nearby structures of interest. The skin at the needle entry site was cleaned with povidone-iodine solution and injected with 2–3 mL of 1% lidocaine by using a 25-gauge hypodermic needle. In many cases, the anesthetic agent was also injected into deeper tissues by using a 22-gauge 1.5-inch needle. The radiologist then advanced an MR imaging–compatible needle through the skin and into the area of interest by using a freehand technique. Biopsy, aspiration, injection, and/or drainage was then performed.
Each procedure was classified with respect to three properties: anatomic location, tissue type, and procedure type. The anatomic location of each procedure was categorized as spine and paraspinal, pelvic, upper extremity, ankle and foot, knee and leg, or miscellaneous. The tissue type was defined as bone, soft tissue, joint, or intervertebral disk. Tenograms were assigned to the soft-tissue category. The five procedure types were core-needle biopsy, fine-needle aspiration biopsy, injection, joint cyst aspiration, and drainage. Core-needle biopsies were distinct from fine-needle aspiration biopsies in that they required different needle types and yielded a measurable three-dimensional tissue sample. Cytologic and histologic evaluations of the fine-needle aspirates and core-needle biopsy samples, respectively, were performed.
Injection procedures were performed to deliver contrast agents, anesthetic agents, or corticosteroids into a joint or lesion. For example, dilute gadopentetate dimeglumine (Magnevist; Berlex Laboratories, Wayne, NJ) (1:200 ratio with sterile saline) was often injected into joints. Some injection procedures also involved the administration of 0.25% bupivacaine with or without a corticosteroid, such as dexamethasone sodium phosphate (Decadron; Merck, Whitehouse Station, NJ) or betamethasone (Celestone; Schering-Plough, Kenilworth, NJ). Joint cyst aspirations involved exclusively cysts of the hip or shoulder joint. Drainage procedures were performed to remove fluid from a lesion for therapeutic purposes or for microbiologic evaluation; no drainage procedure involved histopathologic evaluation.
There were four cases in which one patient underwent two procedures during the same visit. In each instance, a shoulder arthrogram was obtained with the patient in the supine position, and then spinoglenoid notch cyst aspiration was performed with the patient in the prone position. In these four instances, the single visit was counted as two distinct cases. In addition, four other patients underwent separate procedures on different days. These eight patients account for the fact that 63 procedures were performed in 55 patients.
This was partly a retrospective study, and the times for each case were measured by reviewing the MR images from each procedure, which were saved on optical disks. Total procedural time was the difference between the time of acquisition of the first image and the time of acquisition of the last image in a case. Total needle time for each procedure was the time from the insertion of the first MR imaging–compatible needle to the withdrawal of the last MR imaging–compatible needle. Multiple passes through a coaxial needle sheath were counted as a single pass, since a needle pass did not involve repositioning the outer coaxial needle.
To assess safety, patient records were reviewed with respect to three items: cases in which a patient asked that the procedure be stopped, complications occurring during or immediately after a procedure, and late-occurring complications. Needle placement was judged to be successful if images demonstrated the needle tip to be in the area of interest. Tissue sampling success rates were calculated on the basis of whether adequate tissue for diagnosis at pathologic analysis was obtained from a biopsy or aspiration. For each case, lesion size and lesion depth were measured on the images. We obtained pathologic and microbiologic analysis results by reviewing the patient files and notes kept by the MR technicians and nurses. The mean, SD, and median values for total procedural time, needle time, and number of passes were calculated for all procedures.
|
RESULTS
|
|---|
Sixty-three MR imaging–guided interventional musculoskeletal procedures were completed. None of the 55 patients asked that the procedure be stopped. Complications occurred in only two cases: One patient undergoing a sternoclavicular joint injection experienced a brief vasovagal episode that resolved spontaneously, and one patient had local bleeding during aspiration biopsy of a vertebral lesion, so the procedure was stopped. Neither of these two patients had ongoing complications related to the procedure. In the vertebral lesion biopsy case, the tissue sample obtained was sufficient for diagnosis with pathologic analysis. Successful needle placement, which was confirmed on images obtained during each procedure (Fig 2), was achieved in all cases.
View larger version (139K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2a. Transverse fast spin-echo MR images (4,000/108) demonstrate spinoglenoid notch cyst aspiration in a 46-year-old man who had been experiencing intermittent shoulder pain for several years. (a) MR image obtained with the patient supine shows a high-signal-intensity spinoglenoid notch cyst (arrow) posterior to the scapula. Relative mean signal intensities (±SD) of the infraspinatus (34 ± 7), supraspinatus (23 ± 4), and deltoid (19 ± 8) muscles indicated suprascapular nerve impingement with resulting denervation and/or fatty degeneration of the infraspinatus muscle and, to a lesser degree, the supraspinatus muscle. (b) MR image obtained with the patient prone shows the approach for cyst aspiration from the posterior view (top of image). Successful placement of the aspiration needle (arrowheads) is demonstrated: The tip of the needle is inside the cyst.
|
|
View larger version (147K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2b. Transverse fast spin-echo MR images (4,000/108) demonstrate spinoglenoid notch cyst aspiration in a 46-year-old man who had been experiencing intermittent shoulder pain for several years. (a) MR image obtained with the patient supine shows a high-signal-intensity spinoglenoid notch cyst (arrow) posterior to the scapula. Relative mean signal intensities (±SD) of the infraspinatus (34 ± 7), supraspinatus (23 ± 4), and deltoid (19 ± 8) muscles indicated suprascapular nerve impingement with resulting denervation and/or fatty degeneration of the infraspinatus muscle and, to a lesser degree, the supraspinatus muscle. (b) MR image obtained with the patient prone shows the approach for cyst aspiration from the posterior view (top of image). Successful placement of the aspiration needle (arrowheads) is demonstrated: The tip of the needle is inside the cyst.
|
|
Most of the 63 MR imaging–guided procedures were either fine-needle aspiration biopsy (n = 20) or injection (n = 19) procedures; fewer drainages (n = 11), joint cyst aspirations (n = 7), and core-needle biopsies (n = 6) were performed. The majority of the 63 procedures were performed in the pelvic region (n = 17), spine and paraspinal region (n = 14), and upper extremities (n = 13); fewer procedures were performed in foot and ankle (n = 7), knee and leg (n = 6), and miscellaneous regions (n = 6). The various procedures performed in each anatomic region are listed in Table 1. The mean total procedural time, mean total needle time, and mean number of needle passes per case for the MR imaging–guided musculoskeletal interventional procedures were 64.8 minutes (median duration, 58 minutes), 26.2 minutes, and 1.6, respectively. The total procedural times ranged from 10 to 115 minutes. The procedural times categorized according to anatomic location, tissue type, and procedure type are listed in Table 2.
During needle advancement, the fast spin-echo and spoiled gradient-echo sequences were used interchangeably (Fig 3). Both of these sequences were adjusted to facilitate the acquisition of three to five image sections. The fast gradient-echo sequence, with which only one imaging section was acquired, was used for needle advancement in six cases (Fig 4). In these six cases, the mean needle time was not substantially shorter than the overall mean needle time (25.2 minutes in cases involving fast gradient-echo imaging vs overall mean needle time of 26.2 minutes). The three-point Dixon technique was used when gadolinium-based contrast agent was injected into a joint or tendon sheath (Fig 5).
View larger version (129K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3a. (a, b) Transverse T2-weighted fast spin-echo MR images (2,000/108) demonstrate fine-needle aspiration biopsy of a paraspinal psoas muscle soft-tissue mass in a 74-year-old man with low back pain and chronic lymphocytic leukemia. (a) A small (<1-cm) nodular lesion (arrowheads) is shown to have higher signal intensity than the adjacent psoas muscle. Initially, the needle (arrows) was aimed too far laterally. (b) The trajectory of the needle (arrows) was easily corrected with guidance from images acquired during the same sequence as in a; this resulted in accurate positioning of the needle tip (arrowhead) in the central aspect of the lesion. (c) On the transverse spoiled gradient-echo MR image (2,000/108) obtained in the same patient, the needle susceptibility artifact (arrows) is larger than that in a or b, and the aorta (black arrowhead) and inferior vena cava (white arrowhead) are more conspicuous. However, the lesion is not well visualized. Tissue sampling revealed nonmalignant tissue.
|
|
View larger version (127K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3b. (a, b) Transverse T2-weighted fast spin-echo MR images (2,000/108) demonstrate fine-needle aspiration biopsy of a paraspinal psoas muscle soft-tissue mass in a 74-year-old man with low back pain and chronic lymphocytic leukemia. (a) A small (<1-cm) nodular lesion (arrowheads) is shown to have higher signal intensity than the adjacent psoas muscle. Initially, the needle (arrows) was aimed too far laterally. (b) The trajectory of the needle (arrows) was easily corrected with guidance from images acquired during the same sequence as in a; this resulted in accurate positioning of the needle tip (arrowhead) in the central aspect of the lesion. (c) On the transverse spoiled gradient-echo MR image (2,000/108) obtained in the same patient, the needle susceptibility artifact (arrows) is larger than that in a or b, and the aorta (black arrowhead) and inferior vena cava (white arrowhead) are more conspicuous. However, the lesion is not well visualized. Tissue sampling revealed nonmalignant tissue.
|
|
View larger version (110K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3c. (a, b) Transverse T2-weighted fast spin-echo MR images (2,000/108) demonstrate fine-needle aspiration biopsy of a paraspinal psoas muscle soft-tissue mass in a 74-year-old man with low back pain and chronic lymphocytic leukemia. (a) A small (<1-cm) nodular lesion (arrowheads) is shown to have higher signal intensity than the adjacent psoas muscle. Initially, the needle (arrows) was aimed too far laterally. (b) The trajectory of the needle (arrows) was easily corrected with guidance from images acquired during the same sequence as in a; this resulted in accurate positioning of the needle tip (arrowhead) in the central aspect of the lesion. (c) On the transverse spoiled gradient-echo MR image (2,000/108) obtained in the same patient, the needle susceptibility artifact (arrows) is larger than that in a or b, and the aorta (black arrowhead) and inferior vena cava (white arrowhead) are more conspicuous. However, the lesion is not well visualized. Tissue sampling revealed nonmalignant tissue.
|
|
View larger version (111K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4a. Sagittal fast gradient-echo MR images (14.7/7.0, 60° flip angle) demonstrate fine-needle aspiration biopsy of an upper extremity soft-tissue mass in a 38-year-old pregnant woman with a history of previously treated synovial sarcoma. (a) Finger placement (arrows) helps identify the initial needle entry site on the skin. The mass (arrowheads) is adjacent to and partially extending into the humerus. (b) Successful trajectory of the needle (arrows) is confirmed for aspiration. The diagnosis of the mass (arrowheads) at pathologic analysis was sarcoma.
|
|
View larger version (109K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4b. Sagittal fast gradient-echo MR images (14.7/7.0, 60° flip angle) demonstrate fine-needle aspiration biopsy of an upper extremity soft-tissue mass in a 38-year-old pregnant woman with a history of previously treated synovial sarcoma. (a) Finger placement (arrows) helps identify the initial needle entry site on the skin. The mass (arrowheads) is adjacent to and partially extending into the humerus. (b) Successful trajectory of the needle (arrows) is confirmed for aspiration. The diagnosis of the mass (arrowheads) at pathologic analysis was sarcoma.
|
|
View larger version (117K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5a. Fine-needle aspiration biopsy of the sacroiliac joint in a woman with sacral pain that developed 4 weeks postpartum. (a) Transverse fast spin-echo MR image (4,000/102) shows the tip of the needle (arrow) approaching the sacroiliac joint. (b) Transverse three-point Dixon MR image (400/28) shows the needle (arrows) in the sacroiliac joint (arrowhead). Clear yellow liquid (0.5 mL) was aspirated from the sacroiliac joint, and then diluted gadopentetate dimeglumine was injected to document the needle location in the joint. Fluid cultures were negative for microorganisms. The patient was found to have a seronegative spondyloarthropathy.
|
|
View larger version (97K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5b. Fine-needle aspiration biopsy of the sacroiliac joint in a woman with sacral pain that developed 4 weeks postpartum. (a) Transverse fast spin-echo MR image (4,000/102) shows the tip of the needle (arrow) approaching the sacroiliac joint. (b) Transverse three-point Dixon MR image (400/28) shows the needle (arrows) in the sacroiliac joint (arrowhead). Clear yellow liquid (0.5 mL) was aspirated from the sacroiliac joint, and then diluted gadopentetate dimeglumine was injected to document the needle location in the joint. Fluid cultures were negative for microorganisms. The patient was found to have a seronegative spondyloarthropathy.
|
|
The lesions targeted at the MR imaging–guided interventional procedures ranged from 8 to 80 mm in largest dimension, and the depth of the lesions ranged from 15 mm (sternoclavicular joint injections) to 121 mm (fine-needle aspiration biopsy of a retroperitoneal mass).
A total of 26 tissue samples were obtained for cytologic or histologic evaluation. Six core-needle biopsy samples were evaluated, with the following resultant pathologic diagnoses: metastatic adenocarcinoma (n = 1), postradiation changes (n = 1), and negative for malignancy (n = 4). Twenty fine-needle aspiration biopsy samples yielded the following diagnoses: metastatic adenocarcinoma (n = 4), high-grade sarcoma (n = 2), non-Hodgkin lymphoma (n = 2), metastatic transitional cell carcinoma (n = 1), plasmacytoma (n = 1) (Fig 6), inflammatory changes (n = 4), and negative for malignancy (n = 4). Two of the 26 tissue samples, both fine-needle aspirates, were deemed to be insufficient to make an accurate pathologic diagnosis; this led to a tissue sampling success rate of 92% (24 of 26 samples).
View larger version (139K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6a. Fine-needle aspiration biopsy of the T11 vertebral body in a 62-year-old man with spinal cord compression. (a) Sagittal T2-weighted fast spin-echo localizer MR image (4,000/125) obtained with the patient prone shows the lesion (arrowheads) extending into the spinal canal, displacing the cerebrospinal fluid. (b) Transverse T1-weighted fast spin-echo MR image (550/17) obtained with the patient prone shows the needle tip (arrow) within the mass (arrowheads), which involves the posterior elements of the vertebral body. Tissue sampling revealed probable plasmacytoma, which was confirmed surgically as multiple myeloma.
|
|
View larger version (130K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6b. Fine-needle aspiration biopsy of the T11 vertebral body in a 62-year-old man with spinal cord compression. (a) Sagittal T2-weighted fast spin-echo localizer MR image (4,000/125) obtained with the patient prone shows the lesion (arrowheads) extending into the spinal canal, displacing the cerebrospinal fluid. (b) Transverse T1-weighted fast spin-echo MR image (550/17) obtained with the patient prone shows the needle tip (arrow) within the mass (arrowheads), which involves the posterior elements of the vertebral body. Tissue sampling revealed probable plasmacytoma, which was confirmed surgically as multiple myeloma.
|
|
A total of 20 tissue or fluid samples were sent for culture and/or Gram stain: seven from drainage or aspiration, seven from fine-needle aspiration biopsy, five from joint cyst aspiration, and one from core-needle biopsy. Five of these 20 samples resulted in positive cultures: acid-fast bacilli from a prevertebral abscess (Fig 7) and from a knee lesion (Fig 8), Bacteroides species from a pelvic fluid collection, Staphylococcus aureus from an intervertebral disk aspiration, and Propionibacterium acnes from a separate intervertebral disk aspiration.
View larger version (140K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 7a. Drainage of a cervical prevertebral abscess in a 29-year-old Asian-Indian man with a history of dry cough, fevers, and right arm pain and numbness. (a) Sagittal T2-weighted fast spin-echo MR image (4,000/100) demonstrates a high-signal-intensity prevertebral mass (arrowheads) extending caudally from the C6-7 intervertebral disks. Also note the collapse of the C7 vertebra (arrow). (b) Transverse fast spin-echo MR image (2,500/105) shows the finger placement (white arrows) technique used to identify a suitable skin entry site for the needle. The prevertebral lesion (arrowheads) is behind the trachea. Vertical linear artifacts (gray arrows in b and c) produced from the display monitors are seen on this image and in c. (c) Transverse fast spin-echo MR image (2,500/105) shows the needle (white arrows) penetrating the prevertebral mass (arrowheads). Gram stain of the aspirate revealed acid-fast bacilli, and cultures grew Mycobacterium tuberculosis.
|
|
View larger version (137K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 7b. Drainage of a cervical prevertebral abscess in a 29-year-old Asian-Indian man with a history of dry cough, fevers, and right arm pain and numbness. (a) Sagittal T2-weighted fast spin-echo MR image (4,000/100) demonstrates a high-signal-intensity prevertebral mass (arrowheads) extending caudally from the C6-7 intervertebral disks. Also note the collapse of the C7 vertebra (arrow). (b) Transverse fast spin-echo MR image (2,500/105) shows the finger placement (white arrows) technique used to identify a suitable skin entry site for the needle. The prevertebral lesion (arrowheads) is behind the trachea. Vertical linear artifacts (gray arrows in b and c) produced from the display monitors are seen on this image and in c. (c) Transverse fast spin-echo MR image (2,500/105) shows the needle (white arrows) penetrating the prevertebral mass (arrowheads). Gram stain of the aspirate revealed acid-fast bacilli, and cultures grew Mycobacterium tuberculosis.
|
|
View larger version (126K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 7c. Drainage of a cervical prevertebral abscess in a 29-year-old Asian-Indian man with a history of dry cough, fevers, and right arm pain and numbness. (a) Sagittal T2-weighted fast spin-echo MR image (4,000/100) demonstrates a high-signal-intensity prevertebral mass (arrowheads) extending caudally from the C6-7 intervertebral disks. Also note the collapse of the C7 vertebra (arrow). (b) Transverse fast spin-echo MR image (2,500/105) shows the finger placement (white arrows) technique used to identify a suitable skin entry site for the needle. The prevertebral lesion (arrowheads) is behind the trachea. Vertical linear artifacts (gray arrows in b and c) produced from the display monitors are seen on this image and in c. (c) Transverse fast spin-echo MR image (2,500/105) shows the needle (white arrows) penetrating the prevertebral mass (arrowheads). Gram stain of the aspirate revealed acid-fast bacilli, and cultures grew Mycobacterium tuberculosis.
|
|
View larger version (146K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 8. Transverse T2-weighted fast spin-echo MR image (2,500/102) demonstrates fine-needle aspiration biopsy of a soft-tissue mass of the knee in a 48-year-old man with a recent history of melanoma. A coaxial approach is seen. The outer 18-gauge needle (white arrows) is penetrating the near portion of the lesion (arrowheads), and the inner 21-gauge aspiration needle (black arrows) is extending into the far portion of the lesion for aspiration sampling. Granulomatous inflammation was seen at pathologic evaluation, and cultures grew M tuberculosis.
|
|
|
DISCUSSION
|
|---|
The earliest MR systems used for MR imaging–guided therapies were closed-bore magnets, which provide limited access to the patient (5,14). Closed-bore MR systems, which require removing the patient from the magnet bore during needle placement, have been used recently for MR imaging–guided arthrography of the shoulder (11) and abdominal biopsy (15) and in combination with a fluoroscopic C-arm unit (16). Although closed-bore magnets are appealing because they are more readily available than open magnets, limited patient access remains a great obstacle to interventional procedures.
MR units with a C-arm design provide improved patient access, but they still have drawbacks. The small superoinferior gap of the C arm allows limited but sufficient access for many biopsy procedures (2). This small gap size, along with the vertically directed external magnetic field, may prohibit needle placement in the superoinferior direction, and less conventional needle approaches, as have been needed during MR imaging–guided arthrography of the shoulder, may be required (9).
In the current study, we performed MR imaging–guided interventional musculoskeletal procedures by using a vertically open 0.5-T MR unit with a double-doughnut configuration. This type of unit was designed with the intent of guiding interventions, and to our knowledge, it is currently the only system that permits unrestricted vertical and side access to the patient at the most homogeneous portion of the magnetic field (1,8,17). The reported successful percutaneous MR imaging–guided biopsies performed with this type of unit have been mainly in the abdominal organs (1). In our experience with this type of vertically open MR unit, structures throughout the musculoskeletal system were accessed without difficulty. Furthermore, the external magnetic field of the MR unit is directed horizontally along the long access of the magnet bore, and in this study, the desired needle trajectory was never parallel to the magnetic field. This design is important because a needle positioned parallel to the external magnetic field has a markedly reduced apparent width at MR imaging, and this may make accurate needle positioning difficult and thus decrease the accuracy of needle tip localization (18).
Our experience with MR imaging–guided percutaneous procedures in the musculoskeletal system is based on the freehand needle technique; there was no dependence on external computed-image guidance systems, as have been described for this unit previously (1). Such systems may be more important during MR imaging–guided procedures in the brain, where redirecting the needle during advancement may be substantially more detrimental (1).
We had a 92% (24 of 26 samples) tissue sampling success rate for 26 fine-needle aspiration and core-needle biopsies. Although there are different methods of reporting success rates for MR imaging–guided tissue sampling (4,10), the most common method is based on adequate tissue collection for pathologic diagnosis. With this method, our success rate of 92% compares favorably with that in previous reports of MR imaging–guided biopsies and aspirations, in which the tissue sampling success rates varied between 71% and 90% (1,2,4,10). It should be noted that case selection, which could have a substantial effect on tissue sampling success rates, was not accounted for in these comparisons.
In our study, all injections, joint cyst aspirations, and drainages were completed successfully, as determined by performing imaging at the time of the procedure. The safety of each procedure was evaluated on the basis of adverse events during, immediately after, or any time following the procedure. Although it is unlikely that any occurred, late-occurring adverse events that were not reported in the patients’ medical files were not recognized in this study. Only two adverse events occurred: a transient vasovagal reaction in one patient and local bleeding that occurred at the lesion site and was easily controlled with direct pressure in another patient. This low rate of adverse events or serious complications is typical for imaging-guided interventional procedures (1,2,4,19,20).
With use of the 0.5-T unit during MR imaging–guided procedures, both the gradient-echo sequences (spoiled gradient-echo and fast gradient-echo) and the fast spin-echo sequences were useful in guiding needle insertion. The fast spin-echo sequence enabled better lesion visibility, whereas the spoiled gradient-echo sequence enabled better needle visibility. If the needle was not well visualized, a longer echo time or increased receiver bandwidth was used to increase the susceptibility artifact of the needle. The spoiled gradient-echo and fast spin-echo sequences were used in a multisection mode that facilitated easy detection of needle deviation from the desired plane. The use of a small number of sections, between three and five, shortened the imaging time. It was believed that at least three sections were necessary to detect any needle deviation from the imaging plane. In contrast, the fast gradient-echo sequence enables the acquisition of images in a single-section mode. This sequence provided the highest temporal resolution (approximately one image every 2 seconds), but only one section was obtained, and this created the potential for underestimation of needle deviation; however, we observed no complications in this respect. The use of the fast gradient-echo sequence in six cases did not result in a substantial decrease in mean needle time compared with the overall mean needle time (25.2 vs 26.2 minutes).
In our study, the three-point Dixon sequence (12,21) was used following the injection of gadopentetate dimeglumine into either a tendon sheath or a joint space. The three-point Dixon sequence enables the acquisition of fat-selective, water-selective, and both water- and fat-selective images. The three-point Dixon sequence enables low-field-strength magnets to produce images that are similar to fat-saturated images, which otherwise might be difficult to produce since spectral fat saturation is available only on high-field-strength magnets and the use of inversion-recovery sequences for fat suppression after gadolinium-based contrast agent injection is not recommended.
We calculated the total procedural time retrospectively by measuring the time between the acquisition of the first image and the acquisition of the last image in a case. This did not include the time for positioning the patient in the MR unit and therefore may have been an underestimation of the actual physician time spent performing a procedure. However, others (10) have reported an average time for MR imaging–guided aspiration biopsy of noncortical bone lesions of 45 minutes, which is comparable to our mean time of 50 minutes for six fine-needle aspiration biopsies of noncortical bone lesions. With increasing experience, we expect that the time requirements for MR imaging–guided procedures will decrease.
In this series, the bone lesions in which biopsy or aspiration was performed had either permeated the cortex or extended into the adjacent soft tissue. The available armamentarium to perform MR imaging–guided transcortical bone biopsy is limited, and experience in performing MR imaging guidance in such procedures remains minimal; however, a recently developed MR imaging–compatible coaxial bone biopsy system has been reported on (10). Developing efficient MR imaging–compatible transcortical bone biopsy instruments is important, because many pathologic processes in the bone marrow are more readily visible with MR imaging than with other imaging modalities.
MR imaging–guided percutaneous interventional procedures have the potential to help solve several clinical challenges. The lack of ionizing radiation is a benefit, as is the use of gadolinium-based rather than iodinated contrast agents, which reduces the possibility of allergic complications. As mentioned herein earlier, MR imaging can enable one to target intramedullary bone lesions that are not readily seen with other imaging modalities. Likewise, in the breast, MR imaging guidance has been shown to be useful for sampling of lesions that are not well visualized with other imaging techniques (22).
In addition, MR imaging–guided interventional procedures have the potential to enable temperature monitoring, which introduces new opportunities for treatment of lesions, such as thermal (ie, laser) ablation (23) and cryoablation (24). Whether MR imaging guidance will become widely used for interventional musculoskeletal procedures depends on several factors. Such factors include the availability of suitable MR units, MR imaging–compatible instruments, and trained personnel to operate them; the appropriate reimbursement patterns of health care payers; and the outcomes of subsequent studies to investigate the comparative accuracy, time requirements, complication rates, and costs of MR imaging–guided interventional procedures.
A brief evaluation of the literature reveals that the tissue sampling success rates for needle aspirations and biopsies of the musculoskeletal system with CT guidance, 98% (13), and US guidance, 98% (19), are slightly higher than the success rates that we achieved in this series. It should be noted, however, that case selection was not taken into account in this brief comparison. Furthermore, the costs of MR imaging–guided procedures are generally higher than those of other imaging modalities: The costs of MR imaging–guided interventional procedures have been estimated to be 10%–15% higher than the costs of comparable procedures performed with CT guidance (17). US guidance with comparable procedures is less expensive, has the advantages of nonionizing multiplanar imaging similar to MR imaging guidance, and enables imaging in real time.
Thus, US, CT, and conventional fluoroscopy remain the preferred imaging modalities for the majority of interventional procedures, and currently, the number of absolute indications for MR imaging guidance of interventional procedures remains small (3). MR imaging–guided interventional procedures seem to be best suited for difficult clinical cases that are not easily resolved with other imaging modalities. With regard to musculoskeletal procedures, MR imaging guidance is particularly attractive for deeply embedded soft-tissue lesions and for bone marrow lesions that have not breached the cortical bone and thus are difficult to find during open surgical biopsy. It is possible that as the procedural times and costs of MR imaging–guided procedures decrease and the familiarity and expertise with suitable interventional MR systems expand, MR imaging will become the modality of choice for guidance of certain musculoskeletal interventional procedures.
To summarize our experiences, MR imaging–guided interventional procedures involving bone, soft tissue, intervertebral disks, and joints that are performed by using a 0.5-T vertically open MR unit are safe and enable sufficiently rapid and successful needle placement for use in clinical practice.
|
ACKNOWLEDGMENTS
|
|---|
We express our gratitude to Harry K. Genant, MD, for his review of our manuscript. We also thank Claudia Cooper, MR technician, and Gail Riener-French, RN, MR nurse, for their valuable assistance during the MR imaging examinations.
|
FOOTNOTES
|
|---|
Author contributions: Guarantors of integrity of entire study, J.W.G., J.E.V., P.L.; study concepts, J.E.V., P.L.; study design, J.E.V., J.W.G., P.L.; literature research, J.W.G., J.E.V.; clinical studies, J.E.V., A.G.B., C.F.B., S.T.K., A.M.N., P.L.; data acquisition, all authors; data analysis/interpretation, J.W.G., J.E.V.; manuscript preparation, J.W.G., J.E.V.; manuscript definition of intellectual content, J.W.G., J.E.V., P.L.; manuscript editing, P.L., J.W.G.; manuscript revision/review and final version approval, all authors.
|
REFERENCES
|
|---|
-
Silverman SG, Collick BD, Figueira MR, et al. Interactive MR-guided biopsy in an open-configuration MR imaging system. Radiology 1995; 197:175-181.
-
Lewin JS, Petersilge CA, Hatem SF, et al. Interactive MR imaging–guided biopsy and aspiration with a modified clinical C-arm system. AJR Am J Roentgenol 1998; 170:1593-1601.
-
Adam G, Bucker A, Nolte-Ernsting C, Tacke J, Gunther RW. Interventional MR imaging: percutaneous abdominal and skeletal biopsies and drainages of the abdomen. Eur Radiol 1999; 9:1471-1478.
-
Wang SJ, Sercarz JA, Lufkin RB, Borges A, Wang MB. MRI-guided needle localization in the head and neck using contemporaneous imaging in an open configuration system. Head Neck 2000; 22:355-359.
-
Mueller PR, Stark DD, Simeone JF, et al. MR-guided aspiration biopsy: needle design and clinical trials. Radiology 1986; 161:605-609.
-
Duckwiler G, Lufkin RB, Teresi L, et al. Head and neck lesions: MR-guided aspiration biopsy. Radiology 1989; 170:519-522.
-
Pereira PL, Gunaydin I, Trubenbach J, et al. Interventional MR imaging for injection of sacroiliac joints in patients with sacroiliitis. AJR Am J Roentgenol 2000; 175:265-266.
-
Schenck JF, Jolesz FA, Roemer PB, et al. Superconducting open-configuration MR imaging system for image-guided therapy. Radiology 1995; 195:805-814.
-
Petersilge CA, Lewin JS, Duerk JL, Hatem SF. MR arthrography of the shoulder: rethinking traditional imaging procedures to meet the technical requirements of MR imaging guidance. AJR Am J Roentgenol 1997; 169:1453-1457.
-
Neuerburg JM, Adam G, Buecker A, et al. MRI-guided biopsy of bone in a hybrid system. J Magn Reson Imaging 1998; 8:85-90.
-
Trattnig S, Breitenseher M, Rand T, et al. MR imaging–guided MR arthrography of the shoulder: clinical experience on a conventional closed high-field system. AJR Am J Roentgenol 1999; 172:1572-1574.
-
Glover GH, Schneider E. Three-point Dixon technique for true water/fat decomposition with B0 inhomogeneity correction. Magn Reson Med 1991; 18:371-383.
-
Logan PM, Connell DG, O’Connell JX, Munk PL, Janzen DL. Image-guided percutaneous biopsy of musculoskeletal tumors: an algorithm for selection of specific biopsy techniques. AJR Am J Roentgenol 1996; 166:137-141.
-
Lufkin R, Teresi L, Chiu L, Hanafee W. A technique for MR-guided needle placement. AJR Am J Roentgenol 1988; 151:193-196.
-
Salomonowitz EK, Cejna M, Dewey C. Simple and effective technique of guided biopsy in a closed MRI system. Abdom Imaging 2000; 25:638-642.
-
Adam G, Neuerburg J, Bucker A, et al. Interventional magnetic resonance: initial clinical experience with a 1.5-tesla magnetic resonance system combined with c-arm fluoroscopy. Invest Radiol 1997; 32:191-197.
-
Lewin JS. Interventional MR imaging: concepts, systems, and applications in neuroradiology. AJNR Am J Neuroradiol 1999; 20:735-748.
-
Lewin JS, Duerk JL, Jain VR, Petersilge CA, Chao CP, Haaga JR. Needle localization in MR-guided biopsy and aspiration: effects of field strength, sequence design, and magnetic field orientation. AJR Am J Roentgenol 1996; 166:1337-1345.
-
Rubens DJ, Fultz PJ, Gottlieb RH, Rubin SJ. Effective ultrasonographically guided intervention for diagnosis of musculoskeletal lesions. J Ultrasound Med 1997; 16:831-842.
-
Staatz G, Adam GB, Keulers P, Vorwerk D, Gunther RW. Spondylodiskitic abscesses: CT-guided percutaneous catheter drainage. Radiology 1998; 208:363-367.
-
Daniel BL, Butts K, Glover GH, Cooper C, Herfkens RJ. Breast cancer: gadolinium-enhanced MR imaging with a 0.5-T open imager and three-point Dixon technique. Radiology 1998; 207:183-190.
-
Daniel BL, Birdwell RL, Ikeda DM, et al. Breast lesion localization: a freehand, interactive MR imaging-guided technique. Radiology 1998; 207:455-463.
-
Vogl TJ, Mack MG, Muller PK, Straub R, Engelmann K, Eichler K. Interventional MR: interstitial therapy. Eur Radiol 1999; 9:1479-1487.
-
Tacke J, Speetzen R, Adam G, et al. Experimental MR imaging-guided interstitial cryotherapy of the brain. AJNR Am J Neuroradiol 2001; 22:431-440.
This article has been cited by other articles:
|
|
|
|
 
J. A. Carrino, B. Khurana, J. E. Ready, S. G. Silverman, and C. S. Winalski
Magnetic Resonance Imaging-Guided Percutaneous Biopsy of Musculoskeletal Lesions
J. Bone Joint Surg. Am.,
October 1, 2007;
89(10):
2179 - 2187.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
K. Tuncali, P. R. Morrison, C. S. Winalski, J. A. Carrino, S. Shankar, J. E. Ready, E. vanSonnenberg, and S. G. Silverman
MRI-Guided Percutaneous Cryotherapy for Soft-Tissue and Bone Metastases: Initial Experience
Am. J. Roentgenol.,
July 1, 2007;
189(1):
232 - 239.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
F. K. Wacker, S. Vogt, A. Khamene, J. A. Jesberger, S. G. Nour, D. R. Elgort, F. Sauer, J. L. Duerk, and J. S. Lewin
An Augmented Reality System for MR Image-guided Needle Biopsy: Initial Results in a Swine Model
Radiology,
February 1, 2006;
238(2):
497 - 504.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
S Hall and R Buchbinder
Do imaging methods that guide needle placement improve outcome?
Ann Rheum Dis,
September 1, 2004;
63(9):
1007 - 1008.
[Full Text]
[PDF]
|
|
|
TOP
ABSTRACT
INTRODUCTION
