| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Innovation |
1 Both authors: South Texas Radiology Group, 9150 Hueber Rd., Ste. 195, San Antonio, TX 78230.
Received May 15, 2005;
accepted after revision August 9, 2005.
Address correspondence to D. K. Smith
(dsbonerad-AJR{at}yahoo.com).
Abstract
 Top Abstract References |
|---|
CONCLUSION. This approach is easier to perform and results in improved cement distribution along the length of the sacral ala.
Keywords: injection technique • interventional radiology • musculoskeletal radiology • sacroplasty • spine • vertebrae • vertebroplasty
Insufficiency fractures of the sacral ala are common and produce functional disability in elderly individuals. Percutaneous vertebroplasty is a reliable and highly effective treatment for the pain associated with osteoporotic vertebral compression fractures of the sacrum [1, 2]. Previous technical reports describe a needle insertion technique along the short axis of the sacrum (perpendicular to the dorsal cortex of the sacrum) [1, 2]. In our experience, the short-axis technique presents three technical challenges: difficulty ensuring that the needle tip is located in the intramedullary space of the sacral ala before cement injection; production by cement injection of a round collection with only a short length of cement along the fracture line; and frequent extravasation of cement early in the injection, thereby limiting the volume of cement that can be injected.
We developed the long-axis technique to improve the distribution of cement along the long axis of the sacral alar fracture, to decrease the risk of cement extravasation produced by inadvertent anterior cortex perforation, and to use the confines of the intramedullary space to guide the needle rather than relying on identification of poorly seen anatomic features on the lateral radiographic projection.
The sacrum is a shield-shaped flat bone that forms the posterior wall of the pelvic ring (Figs. 1A, 1B, 1C, and 1D). The sacrum is suspended between the adjacent innominate bones by the sacroiliac joints and a complex of sacropelvic ligaments. The sacrum has an oval central canal that conducts the sacrococcygeal nerves. At each sacral level, the segmental nerve root travels laterally through the sacral ala and exits the sacrum through the dorsal and ventral neural foramina (Fig. 1A). The walls of the neural foramina are lined with cortical bone and are visible using fluoroscopy (Fig. 1B). These arcuate lines are the densest bone in the sacrum and are important buttresses against the shear stresses applied to the sacral ala. The 2-cm-wide segment of sacrum located between the neural foramina and the sacroiliac joint is weaker than the sacral ala medial or lateral to this segment (Fig. 1B).
|
|
|
|
Percutaneous sacroplasty has been performed in the United States for 4-5 years. Previous technical descriptions have advocated placement of needles along the short axis of the sacrum [1, 2]. We performed our first 10 sacroplasties using this short-axis approach and encountered difficulties in ensuring that the needles were appropriately positioned before cement injection. The operator must rely on the lateral fluoroscopic view to guide the needle into the intramedullary space of the sacrum without penetrating the anterior cortex. This cortex is difficult to see in the lateral projection because the fluoroscopic image is poor and the cortex overlies other bone structures, including the sacral body and the iliac wings. When we obtained an ideal needle position, we found that the injected cement formed a small spherical cement collection of only 1-2 cm before extravasating through vascular channels into the pelvis.
|
|
|
|
|
In the frontal plane, the image intensifier is rotated cephalad so that the beam parallels the L5-S1 disk space and is aligned perpendicular to the long axis of the upper sacrum (S1-S3). To treat the left sacral ala, the image intensifier is rotated approximately 25-30° to the right so that the beam is parallel to the left sacroiliac joint (Figs. 1A and 1B). Ring forceps are used to localize the skin insertion site at the mid-point between the inferior margin of the sacroiliac joint and the lateral margin of the left S3 neural foramen (Fig. 1B). The skin and periosteum are anesthetized with lidocaine or bupivacaine hydrochloride. A 22-gauge spinal needle is inserted into the skin and directed toward the midpoint between the superior margin of the sacroiliac joint and the lateral margin of the left S1 neural foramen (Fig. 1B). The image intensifier is rotated to the lateral plane (or lateral tube if biplane fluoroscopy is available). The position and orientation of the guide needle are adjusted so that the needle is directed toward the center of the S1 body (Figs. 1D and 2C). The needle must be cephalad enough to avoid penetrating the anterior cortex, especially in patients having a displaced transverse fracture line (Fig. 2C).
A skin incision is made at the base of the guide needle, and the bone cannula is inserted alongside the guide needle to the level of the periosteum. If the cannula is directed alongside the guide needle, we have not found it necessary to check the orientation of the cannula until the cortex is penetrated. The cannula is advanced into the intramedullary cavity approximately 1 cm, and the position and orientation are confirmed using frontal and lateral projections. The cannula is advanced using the lateral projection until the needle tip is located 1 cm inferior to the geometric center of the S1 vertebral body (Figs. 2A and 2C). The anterior cortex of the ala is located at the geometric center of the vertebral body and should not be penetrated (Figs. 1D, 2C, and 2D). The needle position is confirmed in both projections, and a similar approach is followed to place the cannula in the contralateral ala.
A standard mixture of PMMA cement is prepared and injected into the cannula using fluoroscopic visualization in both the frontal and lateral projections. We rely primarily on the frontal projection to watch for cement extravasation toward the sacral neural foramina. As the space at S1 is filled, the cannula is slowly withdrawn while the cement is being injected to fill along the course of the fracture (Figs. 2B and 2D). When the needle approaches the inferior margin of the sacroiliac joint, cement injection is stopped and the cannula is removed. If desired, an epidural steroid injection can be performed at the same sitting using an approach through the sacral hiatus or the dorsal neural foramina.
The potential complications of this needle insertion technique include penetration of the anterior cortex of the sacrum, penetration of the cephalad or superior margin of the sacral ala, and cement extrusion into the sacral neuroforamina. The intramedullary cavity of the S1-S4 level is a rectangular space and can be easily traversed if the needle is inserted at the site described and the orientation of the needle remains parallel to the long axis of the upper sacrum. When a displaced transverse sacral fracture is present, the S1-S2 fragment displaces anteriorly with respect to the lower sacrum. In these patients, the needle insertion site and the orientation of the needle must be adjusted to avoid penetrating the anterior sacral cortex (Fig. 2C).
We have not experienced any penetrations of the anterior sacral cortex. If the needle is advanced past the geometric center of the S1 vertebral body, there is a risk of penetrating the cephalad cortex of the S1 sacral ala with resulting cement extrusion into the soft tissues adjacent to the site of cortical penetration (Fig. 2E). We experienced this problem in the first two or three patients before we adopted the approach described in this article. In those patients, the cement extrusion was localized along the superior margin of S1, an area in which no significant neurovascular structures are located (Fig. 2E). These patients experienced no appreciable symptoms related to the cement extrusion.
Regardless of whether the long- or the short-axis needle insertion technique is used, potential also exists for cement extrusion into the neuroforamina. We obtain a CT scan of the sacrum before the procedure to determine if a fracture line extends into one or more of the neuroforamina. If so, we try to limit the amount of cement that we inject near the involved foramen. The neuroforamen is a tubular structure and cement may be contained in the medullary cavity but may appear to be in the neural foramen because of the obliquity of the image. In our experience, cement has not extruded into the neuroforamen unless the osseous walls of the foramen are fractured.
We have found that this approach is technically easier and that it more consistently produces a vertical column of cement distribution along the fracture line than the short-axis approach. The long-axis approach also uses the boundaries of the intramedullary space of the sacrum to guide the bone cannula and relies less on identification of poorly defined anatomic landmarks. This procedure can be a useful tool for treating patients with sacral insufficiency fractures.
|
TopAbstractReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  M.V. Jayaraman, H. Chang, and S.H. Ahn An Easily Identifiable Anatomic Landmark For Fluoroscopically Guided Sacroplasty: Anatomic Description and Validation with Treatment in 13 Patients AJNR Am. J. Neuroradiol., May 1, 2009; 30(5): 1070 - 1073. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C.T. Whitlow, S.K. Yazdani, M.L. Reedy, S.E. Kaminsky, J.L. Berry, and P.P. Morris Investigating Sacroplasty: Technical Considerations and Finite Element Analysis of Polymethylmethacrylate Infusion into Cadaveric Sacrum AJNR Am. J. Neuroradiol., June 1, 2007; 28(6): 1036 - 1041. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
