Technical Note
Thoracic Spine Localization Using Preoperative Placement of Fiducial Markers and Subsequent CT. A Technical Report Amjad Nasr Anaizi1 Christopher Kalhorn1 Faheem A. Sandhu1
Michael McCullough2
1 Department of Neurological Surgery, Georgetown University
Hospital, Washington, District of Columbia, United States 2 Department of Radiology, Georgetown University Hospital, Washington, District of Columbia, United States
Jean-Marc Voyadzis1
Address for correspondence Amjad Nasr Anaizi, MD, Department of Neurological Surgery, Georgetown University Hospital, 3800 Reservoir Road, Washington, DC 20007, United States (e-mail: [email protected]).
J Neurol Surg A 2015;76:66–71.
Abstract
Keywords
► ► ► ► ► ►
spine localization thoracic discectomy thoracic spine wrong level surgery thoracic disc
received August 30, 2012 accepted after revision January 3, 2014 published online July 21, 2014
Study Design A retrospective case series evaluating the use of fiducial markers with subsequent computed tomography (CT) or CT myelography for intraoperative localization. Objective To evaluate the safety and utility of preoperative fiducial placement, confirmed with CT myelography, for intraoperative localization of thoracic spinal levels. Summary of Background Data Thoracic spine surgery is associated with serious complications, not the least of which is the potential for wrong-level surgery. Intraoperative fluoroscopy is often used but can be unreliable due to the patient’s body habitus and anatomical variation. Methods Sixteen patients with thoracic spine pathology requiring surgical intervention underwent preoperative fiducial placement at the pedicle of the level of interest in the interventional radiology suite. CT or CT myelogram was then done to evaluate fiducial location relative to the level of pathology. Surgical treatment followed at a later date in all patients. Results All patients underwent preoperative fiducial placement and CT or CT myelography, which was done on an outpatient basis in 14 of the 16 patients. Intraoperatively, fiducial localization was easily and quickly done with intraoperative fluoroscopy leading to correct localization of spinal level in all cases. All patients had symptomatic improvement following surgery. There were no complications from preoperative localization or operative intervention. Conclusions Preoperative placement of fiducial markers confirmed with a CT or CT myelogram allows for reliable and fast intraoperative localization of the spinal level of interest with minimal risks and potential complications to the patient. In most cases, a noncontrast CT should be sufficient. This should be an equally reliable means of localization while further decreasing potential for complications. CT myelography should be reserved for pathology that is not evident on noncontrast CT. Accuracy of localization is independent of variations in rib number or vertebral segmentation. The technique is a safe, reliable, and rapid means of localizing spinal level during surgery.
© 2015 Georg Thieme Verlag KG Stuttgart · New York
DOI http://dx.doi.org/ 10.1055/s-0034-1371512. ISSN 2193-6315.
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
66
Introduction Thoracic spine surgery is associated with serious complications, not the least of which is the potential for wrong-level surgery. In a survey of 415 surgeons, 50% admitted to operating at the wrong level of the spine at some point in their career.1 Traditionally, surgeons use intraoperative fluoroscopy to count levels and confirm the correct operative level. Some surgeons count up from the sacrum on lateral fluoroscopy; others count ribs on anteroposterior fluoroscopy to localize the correct level. These methods, even in the hands of the most thorough surgeons, are subject to error, especially in patients with anatomical variations or those with osteopenia or a body habitus that makes visualization of bony anatomy on fluoroscopy difficult. One study evaluating variations in lumbosacral anatomy by whole-spine MR imaging of 750 patients revealed that 80% had five, 14.6% had six, and 5.3% had four lumbar vertebrae.2 These common variants can lead to serious errors when relying on intraoperative fluoroscopy to count levels for spinal localization. We describe a simple and reliable method for thoracic spine level localization that largely eliminates the shortcomings of traditional localization methods and should significantly reduce the potential for wrong-level spine surgery. This method combines placement of small radiopaque fiducial markers followed by computed tomography (CT) or CT myelography, clearly marking the site of pathology for rapid intraoperative localization.
Materials and Methods A series of patients who underwent this localization protocol were retrospectively reviewed. All 16 patients who
Anaizi et al.
underwent this localization protocol were included in this series. Of the 16 patients, 13 had thoracic disc herniations, one patient presented with a dorsal thoracic epidural tumor, one with a thoracic arachnoid cyst, and one patient presented with discitis refractory to medical therapy and two unsuccessful percutaneous biopsy attempts. Following diagnosis and discussion of the surgical plan, patients were sent for percutaneous fiducial placement under fluoroscopic guidance by an interventional radiologist followed by a CT or CT myelography. The fiducials were placed with patients in the prone position under local anesthesia. These were placed at the transverse processes in line with the pedicles at the level of interest. We believed that it was highly unlikely these fiducials would migrate. However, placing them bilaterally should ensure that any migration would be detected because the fiducial locations would not be symmetrical. Percutaneous fiducial placement and subsequent CT or CT myelography were performed on an outpatient basis in most patients. Location of pathology relative to the fiducial markers on CT was used for localization using intraoperative fluoroscopy. ►Table 1 lists the patient demographics, diagnosis, surgical intervention, and complications.
Case Example A 46-year old man presented with thoracic back pain, bilateral leg weakness, and myelopathy on neurologic exam. Magnetic resonance imaging (MRI) of the thoracic spine revealed a T8–T9 disc herniation. Surgical treatment of the disc herniation was planned. Prior to surgery, percutaneous fiducial placement (Civco, Orange City, Iowa, United States) under fluoroscopic guidance was performed at the level of
Table 1 Patient demographic information, thoracic level, pathology, surgical intervention, and complications Patient
Age, y
Sex
Pathology
Surgical intervention
Complications
1
52
F
T11–T12 HNP
Lateral transthoracic discectomy
None
2
30
F
T6–T7 HNP
Lateral transthoracic discectomy
None
3
38
M
T2–T3 HNP
Posterior transpedicular discectomy
None
4
54
M
T6–T7 HNP
Lateral transthoracic discectomy
None
5
44
F
T8–T9 HNP
Posterior transpedicular discectomy
None
6
34
M
T8–T9 HNP
Lateral transthoracic discectomy
None
7
46
M
T8–T9 HNP
Lateral transthoracic discectomy
None
8
50
M
T6–T7 epidural tumor
Posterior laminectomy for resection
None
9
58
F
T8–T9 discitis
Posterior transpedicular discectomy
None
10
55
F
T7–T8 HNP
Lateral transthoracic discectomy
None
11
28
M
T6–T7 HNP
Lateral transthoracic discectomy
None
12
48
F
T7–T8 arachnoid cyst
Posterior laminectomy for fenestration
None
13
51
M
T11–T12 HNP
Lateral transthoracic discectomy
None
14
64
M
T11–T12 HNP
Lateral transthoracic discectomy followed by a posterior transpedicular discectomy
None
15
47
F
T5–T6 HNP
Posterior transpedicular discectomy
None
16
45
M
T4–T5 HNP
Posterior transpedicular discectomy
None
Abbreviation: HNP, herniated nucleus pulposus.
Journal of Neurological Surgery—Part A
Vol. 76
No. A1/2015
67
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
Thoracic Spine Localization Using Fiducials
Thoracic Spine Localization Using Fiducials
Anaizi et al.
Fig. 3 Intraoperative anteroposterior X-ray visualizing the fiducial markers and confirming the correct operative level.
Fig. 1 (A) Photograph of fiducials used for localization. (B) Photograph of patient in prone position during fiducial placement under local anesthesia.
transthoracic discectomy with subsequent placement of an interbody graft was performed (►Fig. 3). The patient had excellent relief of his symptoms following surgery. Postoperative X-rays showed good placement of the intervertebral graft and confirmed treatment of the correct spinal level (►Fig. 4). No complications from fiducial placement, CT myelography, or surgery were encountered.
Results pathology in the interventional radiology suite using local anesthesia. The patient then immediately underwent a CT myelogram to confirm satisfactory placement of the fiducials and visualize their location relative to the level of pathology (►Fig. 1). During surgery, the fiducials were easily visible with fluoroscopy, allowing for fast and accurate identification of the correct spinal level (►Fig. 2). A minimally invasive
A total of 14 of the 16 patients had fiducial placement and a CT or CT myelogram performed on an outpatient basis. Fiducial placement for the patient described in the case example and one with an epidural tumor were done during their hospitalizations due to the urgent need for surgical intervention. A variety of approaches were used to perform the thoracic discectomies as deemed appropriate by the primary surgeon. The patient with the epidural tumor underwent posterior laminectomy for tumor resection. The patient with the arachnoid cyst underwent a laminectomy and cyst fenestration. All patients tolerated surgery well and experienced improvement of preoperative symptoms. The average fluoroscopy time needed for placement of the fiducials was 8 minutes with an average cumulative dose of 408.9 mGy. The average dose length product of the CT scan or CT myelogram performed following fiducial placement was 1313 mGy cm. The intraoperative fluoroscopy time was not available for inclusion in this study. There were no perioperative or postoperative complications associated with fiducial placement, CT myelography, or operative intervention.
Discussion Fig. 2 Sagittal computed tomography (CT) myelogram illustrating (A) the level of pathology and (B) axial CT myelogram showing fiducial markers at the level of the disc herniation. Journal of Neurological Surgery—Part A
Vol. 76
No. A1/2015
Thoracic spine localization can be challenging due to the lack of reliable landmarks, scapular shadows, anatomical
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
68
Fig. 4 (A, B) Intraoperative anteroposterior and lateral X-rays showing instrumentation placed following the discectomy and fiducial markers at the same level.
variations, and differences in body habitus. These difficulties can lead to wrong-level surgery and unacceptable patient morbidity. Traditional techniques using intraoperative fluoroscopy have multiple shortcomings making them vulnerable to error. In an attempt to reduce the potential for wrong-level surgery in the thoracic spine, several localization methods have been described in the literature. Singh et al described the use of an oblique modification of the standard crosstable lateral to count levels intraoperatively.3 The use of intraoperative computer-assisted image guidance for level localization has been proposed, but widespread implementation of this approach is limited by the lack of availability at most hospitals and the expense of this technological
Anaizi et al.
tool.4 Rosahl et al described the use of radiographic skin markers placed preoperatively at the level of pathology.5 This method is not associated with any additional risk to the patient; however, it is susceptible to significant error due to the inevitable skin and soft tissue shifts that occur during positioning and surgical exposure. The preoperative injection of methylene blue at the spinous process of the pathologic level has also been used for localization.6 This again is associated with minimal risk to the patient but lacks the precision necessary for a reproducible and reliable localization technique, given the potential for spread of the dye to surrounding soft tissue. In 2007, Cornips et al described the use of CT-guided transpleural placement of a wire into the lung parenchyma after CT myelography to localize the level of pathology prior to a thorascopic discectomy.7 Although a CT myelogram was performed prior to localization in each case, there was still one instance of wrong-level surgery in their series of 86 patients due to fracture and retraction of the localizing needle. Their technique is limited to the transthoracic approach and carries associated risks of pneumothorax and bleeding. More recently, Hsu et al described the preoperative injection of polymethylmethacrylate (PMMA) into the vertebral body adjacent to spinal pathology as a method of thoracic level localization.8 This was then followed by a noncontrast CT to confirm level. This localization method is subject to the risks associated with a standard vertebroplasty including the potential for cement extravasation into the spinal canal or neural foramen, radiculopathy, hematoma formation, cement embolus, pedicle fracture, and infection. The risk for cement leakage is likely lower than reported in the literature, given the small volume of cement needed, but the possibility for extravasation remains. Additionally, kyphoplasty/vertebroplasty procedures are generally not performed above the T5 level due to limited fluoroscopic visualization of the vertebral body and pedicles. Finally, placement of PMMA into the treated pedicle and vertebral body precludes the ability to place pedicle screws in the future. In 2013, Chin et al described the use of sterile spinal needles docked on the transverse process of L3 and T12.9 The thoracic level of interest was then found by counting from the T12 vertebra, which remained marked with a spinal needle even after the incision was made. This method should improve accuracy but remains subject to poor visualization of bony anatomy due to osteopenia or body habitus and potential difficulties with accurate localization in patients with anatomical variations. The method described here allows for reliable localization of the level of pathology with minimal additional risk to the patient. Once the pathologic lesion is diagnosed using conventional imaging and surgical intervention is planned, the patient is scheduled for fiducial placement in the interventional radiology suite as an outpatient. In our series, all patients had fiducials placed bilaterally at the transverse processes in line with the pedicle at the level of pathology. Fiducial placement was done percutaneously under Journal of Neurological Surgery—Part A
Vol. 76
No. A1/2015
69
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
Thoracic Spine Localization Using Fiducials
Thoracic Spine Localization Using Fiducials
Anaizi et al.
fluoroscopic guidance. We believe it is highly unlikely that these fiducials would migrate, and there were no instances of fiducial migration in our series. Bilateral symmetrical placement adds a level of confidence because in the unlikely event of fiducial migration, these fiducials would no longer be symmetrical on fluoroscopic imaging in the operating room. In addition, when approaching the pathology posteriorly, it is not uncommon to encounter and inadvertently remove the fiducial. In such an instance, the contralateral fiducial will remain to confirm localization of the correct level. There is no need to remove these fiducials intraoperatively or postoperatively because they have not been associated with any complications. Once fiducials were placed, a CT or CT myelogram was done to confirm placement of the fiducial and visualize its location relative to the level of pathology. Once this relationship is established, the actual level of pathology becomes irrelevant because it can be identified simply by the position of the fiducial marker. In our series, fiducials were placed without complications and with correct-level localization in all cases. Potential complications of this localization method include infectious and bleeding complications related to the fiducial placement. These patients are also subjected to a CT scan that they otherwise would not undertake. This CT is associated with additional radiation exposure for the patient, although this would reduce the intraoperative radiation exposure. If a CT myelogram is necessary, the myelogram carries unique potential complications including seizures, contrast reactions, renal insufficiency, bleeding complications, and potential for puncture of the conus.10 These are all very rare, and there were no such complications in our series. We believe the additional risks of complications and added cost are outweighed by the benefits of this localization technique. In this series, most patients underwent a CT myelogram following fiducial placement. Thoracic disc herniations and other thoracic spine pathology can often be seen on routine CT without the need for myelography. For this reason we recommend acquisition of a noncontrast CT immediately following fiducial placement. If the pathology cannot be reliably identified on CT, then a CT myelogram should be performed to visualize the pathology relative to the fiducial. Furthermore, CT is an important adjunct to MRI in the diagnosis and treatment of thoracic disc herniations because it can provide useful additional information regarding disc calcification.11 Fiducial placement and the subsequent CT or CT myelogram can be done at any time before surgery without compromising the localization. In the operating room, the CT images are available for review, and the surgeon is able to immediately locate the fiducial, and hence the level of pathology, using intraoperative fluoroscopy. In posterior cases, the fiducial was routinely encountered during the exposure adding further confidence regarding correct site localization. This localization method can be utilized regardless of the surgical approach (anterior, lateral, or posterior) and does not require any counting of ribs or Journal of Neurological Surgery—Part A
Vol. 76
No. A1/2015
vertebral bodies during surgery. Given the need for two additional procedures, this may not be appropriate when emergent intervention is needed, but it can be performed the same day as surgery when urgent intervention is needed. Weaknesses of this study include the relatively small number of patients. Thoracic pathology is not as commonly encountered as cervical or lumbar spine pathology. In addition, patients with prior spine instrumentation, severe compression fractures, or lytic bony lesions requiring surgical intervention were not included in this study because their pathology can be easily localized with intraoperative fluoroscopic images without the need for fiducial localization. We believe, based on our experience, that intraoperative fluoroscopy use and radiation exposure to the surgeon and operating room staff is minimized with this technique. However, this study does not directly compare intraoperative radiation exposure when utilizing fiducial marker localization with any other localization technique. Further investigation is necessary to resolve these issues.
Conclusion We describe a method for localization of thoracic spine pathology that combines the placement of small fiducial markers with subsequent CT imaging. This technique is easy to perform, very reliable, and has the potential to significantly reduce the use of intraoperative fluoroscopy. It is applicable to the entire thoracic spine, and with proper care and attention nearly eliminates localization errors during thoracic spine surgery. We believe this study functions as a useful feasibility study, and larger investigations are needed to more clearly define its reliability and identify any potential complications.
References 1 Mody MG, Nourbakhsh A, Stahl DL, Gibbs M, Alfawareh M, Garges
2
3
4
5
6
KJ. The prevalence of wrong level surgery among spine surgeons. Spine 2008;33(2):194–198 Hanson EH, Mishra RK, Chang DS, et al. Sagittal whole-spine magnetic resonance imaging in 750 consecutive outpatients: accurate determination of the number of lumbar vertebral bodies. J Neurosurg Spine 2010;12(1):47–55 Singh H, Meyer SA, Hecht AC, Jenkins AL III. Novel fluoroscopic technique for localization at cervicothoracic levels. J Spinal Disord Tech 2009;22(8):615–618 Nowitzke A, Wood M, Cooney K. Improving accuracy and reducing errors in spinal surgery—a new technique for thoracolumbar-level localization using computer-assisted image guidance. Spine J 2008;8(4):597–604 Rosahl SK, Gharabaghi A, Liebig T, Feste CD, Tatagiba M, Samii M. Skin markers for surgical planning for intradural lesions of the thoracic spine. Technical note. Surg Neurol 2002;58(5): 346–348 Paolini S, Ciappetta P, Missori P, Raco A, Delfini R. Spinous process marking: a reliable method for preoperative surface localization of intradural lesions of the high thoracic spine. Br J Neurosurg 2005; 19(1):74–76
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
70
Thoracic Spine Localization Using Fiducials
Anaizi et al.
9 Chin KR, Seale JM, Cumming VM. Avoidance of wrong-level
Preoperative localization of herniated thoracic discs using myeloCT guided transpleural puncture: technical note. Childs Nerv Syst 2007;23(1):21–26 8 Hsu W, Sciubba DM, Sasson AD, et al. Intraoperative localization of thoracic spine level with preoperative percutaneous placement of intravertebral polymethylmethacrylate. J Spinal Disord Tech 2008; 21(1):72–75
thoracic spine surgery using sterile spinal needles. J Spinal Disord Tech 2013; October 31 (Epub ahead of print) 10 Sandow BA, Donnal JF. Myelography complications and current practice patterns. AJR Am J Roentgenol 2005;185(3): 768–771 11 Dietze DD Jr, Fessler RG. Thoracic disc herniations. Neurosurg Clin N Am 1993;4(1):75–90
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
7 Cornips E, Beuls E, Geskes G, Janssens M, van Aalst J, Hofman P.
71
Journal of Neurological Surgery—Part A
Vol. 76
No. A1/2015
Copyright of Journal of Neurological Surgery. Part A. Central European Neurosurgery is the property of Georg Thieme Verlag Stuttgart and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
Thoracic Spine Localization Using Preoperative Placement of Fiducial Markers and Subsequent CT. A Technical Report Amjad Nasr Anaizi1 Christopher Kalhorn1 Faheem A. Sandhu1
Michael McCullough2
1 Department of Neurological Surgery, Georgetown University
Hospital, Washington, District of Columbia, United States 2 Department of Radiology, Georgetown University Hospital, Washington, District of Columbia, United States
Jean-Marc Voyadzis1
Address for correspondence Amjad Nasr Anaizi, MD, Department of Neurological Surgery, Georgetown University Hospital, 3800 Reservoir Road, Washington, DC 20007, United States (e-mail: [email protected]).
J Neurol Surg A 2015;76:66–71.
Abstract
Keywords
► ► ► ► ► ►
spine localization thoracic discectomy thoracic spine wrong level surgery thoracic disc
received August 30, 2012 accepted after revision January 3, 2014 published online July 21, 2014
Study Design A retrospective case series evaluating the use of fiducial markers with subsequent computed tomography (CT) or CT myelography for intraoperative localization. Objective To evaluate the safety and utility of preoperative fiducial placement, confirmed with CT myelography, for intraoperative localization of thoracic spinal levels. Summary of Background Data Thoracic spine surgery is associated with serious complications, not the least of which is the potential for wrong-level surgery. Intraoperative fluoroscopy is often used but can be unreliable due to the patient’s body habitus and anatomical variation. Methods Sixteen patients with thoracic spine pathology requiring surgical intervention underwent preoperative fiducial placement at the pedicle of the level of interest in the interventional radiology suite. CT or CT myelogram was then done to evaluate fiducial location relative to the level of pathology. Surgical treatment followed at a later date in all patients. Results All patients underwent preoperative fiducial placement and CT or CT myelography, which was done on an outpatient basis in 14 of the 16 patients. Intraoperatively, fiducial localization was easily and quickly done with intraoperative fluoroscopy leading to correct localization of spinal level in all cases. All patients had symptomatic improvement following surgery. There were no complications from preoperative localization or operative intervention. Conclusions Preoperative placement of fiducial markers confirmed with a CT or CT myelogram allows for reliable and fast intraoperative localization of the spinal level of interest with minimal risks and potential complications to the patient. In most cases, a noncontrast CT should be sufficient. This should be an equally reliable means of localization while further decreasing potential for complications. CT myelography should be reserved for pathology that is not evident on noncontrast CT. Accuracy of localization is independent of variations in rib number or vertebral segmentation. The technique is a safe, reliable, and rapid means of localizing spinal level during surgery.
© 2015 Georg Thieme Verlag KG Stuttgart · New York
DOI http://dx.doi.org/ 10.1055/s-0034-1371512. ISSN 2193-6315.
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
66
Introduction Thoracic spine surgery is associated with serious complications, not the least of which is the potential for wrong-level surgery. In a survey of 415 surgeons, 50% admitted to operating at the wrong level of the spine at some point in their career.1 Traditionally, surgeons use intraoperative fluoroscopy to count levels and confirm the correct operative level. Some surgeons count up from the sacrum on lateral fluoroscopy; others count ribs on anteroposterior fluoroscopy to localize the correct level. These methods, even in the hands of the most thorough surgeons, are subject to error, especially in patients with anatomical variations or those with osteopenia or a body habitus that makes visualization of bony anatomy on fluoroscopy difficult. One study evaluating variations in lumbosacral anatomy by whole-spine MR imaging of 750 patients revealed that 80% had five, 14.6% had six, and 5.3% had four lumbar vertebrae.2 These common variants can lead to serious errors when relying on intraoperative fluoroscopy to count levels for spinal localization. We describe a simple and reliable method for thoracic spine level localization that largely eliminates the shortcomings of traditional localization methods and should significantly reduce the potential for wrong-level spine surgery. This method combines placement of small radiopaque fiducial markers followed by computed tomography (CT) or CT myelography, clearly marking the site of pathology for rapid intraoperative localization.
Materials and Methods A series of patients who underwent this localization protocol were retrospectively reviewed. All 16 patients who
Anaizi et al.
underwent this localization protocol were included in this series. Of the 16 patients, 13 had thoracic disc herniations, one patient presented with a dorsal thoracic epidural tumor, one with a thoracic arachnoid cyst, and one patient presented with discitis refractory to medical therapy and two unsuccessful percutaneous biopsy attempts. Following diagnosis and discussion of the surgical plan, patients were sent for percutaneous fiducial placement under fluoroscopic guidance by an interventional radiologist followed by a CT or CT myelography. The fiducials were placed with patients in the prone position under local anesthesia. These were placed at the transverse processes in line with the pedicles at the level of interest. We believed that it was highly unlikely these fiducials would migrate. However, placing them bilaterally should ensure that any migration would be detected because the fiducial locations would not be symmetrical. Percutaneous fiducial placement and subsequent CT or CT myelography were performed on an outpatient basis in most patients. Location of pathology relative to the fiducial markers on CT was used for localization using intraoperative fluoroscopy. ►Table 1 lists the patient demographics, diagnosis, surgical intervention, and complications.
Case Example A 46-year old man presented with thoracic back pain, bilateral leg weakness, and myelopathy on neurologic exam. Magnetic resonance imaging (MRI) of the thoracic spine revealed a T8–T9 disc herniation. Surgical treatment of the disc herniation was planned. Prior to surgery, percutaneous fiducial placement (Civco, Orange City, Iowa, United States) under fluoroscopic guidance was performed at the level of
Table 1 Patient demographic information, thoracic level, pathology, surgical intervention, and complications Patient
Age, y
Sex
Pathology
Surgical intervention
Complications
1
52
F
T11–T12 HNP
Lateral transthoracic discectomy
None
2
30
F
T6–T7 HNP
Lateral transthoracic discectomy
None
3
38
M
T2–T3 HNP
Posterior transpedicular discectomy
None
4
54
M
T6–T7 HNP
Lateral transthoracic discectomy
None
5
44
F
T8–T9 HNP
Posterior transpedicular discectomy
None
6
34
M
T8–T9 HNP
Lateral transthoracic discectomy
None
7
46
M
T8–T9 HNP
Lateral transthoracic discectomy
None
8
50
M
T6–T7 epidural tumor
Posterior laminectomy for resection
None
9
58
F
T8–T9 discitis
Posterior transpedicular discectomy
None
10
55
F
T7–T8 HNP
Lateral transthoracic discectomy
None
11
28
M
T6–T7 HNP
Lateral transthoracic discectomy
None
12
48
F
T7–T8 arachnoid cyst
Posterior laminectomy for fenestration
None
13
51
M
T11–T12 HNP
Lateral transthoracic discectomy
None
14
64
M
T11–T12 HNP
Lateral transthoracic discectomy followed by a posterior transpedicular discectomy
None
15
47
F
T5–T6 HNP
Posterior transpedicular discectomy
None
16
45
M
T4–T5 HNP
Posterior transpedicular discectomy
None
Abbreviation: HNP, herniated nucleus pulposus.
Journal of Neurological Surgery—Part A
Vol. 76
No. A1/2015
67
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
Thoracic Spine Localization Using Fiducials
Thoracic Spine Localization Using Fiducials
Anaizi et al.
Fig. 3 Intraoperative anteroposterior X-ray visualizing the fiducial markers and confirming the correct operative level.
Fig. 1 (A) Photograph of fiducials used for localization. (B) Photograph of patient in prone position during fiducial placement under local anesthesia.
transthoracic discectomy with subsequent placement of an interbody graft was performed (►Fig. 3). The patient had excellent relief of his symptoms following surgery. Postoperative X-rays showed good placement of the intervertebral graft and confirmed treatment of the correct spinal level (►Fig. 4). No complications from fiducial placement, CT myelography, or surgery were encountered.
Results pathology in the interventional radiology suite using local anesthesia. The patient then immediately underwent a CT myelogram to confirm satisfactory placement of the fiducials and visualize their location relative to the level of pathology (►Fig. 1). During surgery, the fiducials were easily visible with fluoroscopy, allowing for fast and accurate identification of the correct spinal level (►Fig. 2). A minimally invasive
A total of 14 of the 16 patients had fiducial placement and a CT or CT myelogram performed on an outpatient basis. Fiducial placement for the patient described in the case example and one with an epidural tumor were done during their hospitalizations due to the urgent need for surgical intervention. A variety of approaches were used to perform the thoracic discectomies as deemed appropriate by the primary surgeon. The patient with the epidural tumor underwent posterior laminectomy for tumor resection. The patient with the arachnoid cyst underwent a laminectomy and cyst fenestration. All patients tolerated surgery well and experienced improvement of preoperative symptoms. The average fluoroscopy time needed for placement of the fiducials was 8 minutes with an average cumulative dose of 408.9 mGy. The average dose length product of the CT scan or CT myelogram performed following fiducial placement was 1313 mGy cm. The intraoperative fluoroscopy time was not available for inclusion in this study. There were no perioperative or postoperative complications associated with fiducial placement, CT myelography, or operative intervention.
Discussion Fig. 2 Sagittal computed tomography (CT) myelogram illustrating (A) the level of pathology and (B) axial CT myelogram showing fiducial markers at the level of the disc herniation. Journal of Neurological Surgery—Part A
Vol. 76
No. A1/2015
Thoracic spine localization can be challenging due to the lack of reliable landmarks, scapular shadows, anatomical
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
68
Fig. 4 (A, B) Intraoperative anteroposterior and lateral X-rays showing instrumentation placed following the discectomy and fiducial markers at the same level.
variations, and differences in body habitus. These difficulties can lead to wrong-level surgery and unacceptable patient morbidity. Traditional techniques using intraoperative fluoroscopy have multiple shortcomings making them vulnerable to error. In an attempt to reduce the potential for wrong-level surgery in the thoracic spine, several localization methods have been described in the literature. Singh et al described the use of an oblique modification of the standard crosstable lateral to count levels intraoperatively.3 The use of intraoperative computer-assisted image guidance for level localization has been proposed, but widespread implementation of this approach is limited by the lack of availability at most hospitals and the expense of this technological
Anaizi et al.
tool.4 Rosahl et al described the use of radiographic skin markers placed preoperatively at the level of pathology.5 This method is not associated with any additional risk to the patient; however, it is susceptible to significant error due to the inevitable skin and soft tissue shifts that occur during positioning and surgical exposure. The preoperative injection of methylene blue at the spinous process of the pathologic level has also been used for localization.6 This again is associated with minimal risk to the patient but lacks the precision necessary for a reproducible and reliable localization technique, given the potential for spread of the dye to surrounding soft tissue. In 2007, Cornips et al described the use of CT-guided transpleural placement of a wire into the lung parenchyma after CT myelography to localize the level of pathology prior to a thorascopic discectomy.7 Although a CT myelogram was performed prior to localization in each case, there was still one instance of wrong-level surgery in their series of 86 patients due to fracture and retraction of the localizing needle. Their technique is limited to the transthoracic approach and carries associated risks of pneumothorax and bleeding. More recently, Hsu et al described the preoperative injection of polymethylmethacrylate (PMMA) into the vertebral body adjacent to spinal pathology as a method of thoracic level localization.8 This was then followed by a noncontrast CT to confirm level. This localization method is subject to the risks associated with a standard vertebroplasty including the potential for cement extravasation into the spinal canal or neural foramen, radiculopathy, hematoma formation, cement embolus, pedicle fracture, and infection. The risk for cement leakage is likely lower than reported in the literature, given the small volume of cement needed, but the possibility for extravasation remains. Additionally, kyphoplasty/vertebroplasty procedures are generally not performed above the T5 level due to limited fluoroscopic visualization of the vertebral body and pedicles. Finally, placement of PMMA into the treated pedicle and vertebral body precludes the ability to place pedicle screws in the future. In 2013, Chin et al described the use of sterile spinal needles docked on the transverse process of L3 and T12.9 The thoracic level of interest was then found by counting from the T12 vertebra, which remained marked with a spinal needle even after the incision was made. This method should improve accuracy but remains subject to poor visualization of bony anatomy due to osteopenia or body habitus and potential difficulties with accurate localization in patients with anatomical variations. The method described here allows for reliable localization of the level of pathology with minimal additional risk to the patient. Once the pathologic lesion is diagnosed using conventional imaging and surgical intervention is planned, the patient is scheduled for fiducial placement in the interventional radiology suite as an outpatient. In our series, all patients had fiducials placed bilaterally at the transverse processes in line with the pedicle at the level of pathology. Fiducial placement was done percutaneously under Journal of Neurological Surgery—Part A
Vol. 76
No. A1/2015
69
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
Thoracic Spine Localization Using Fiducials
Thoracic Spine Localization Using Fiducials
Anaizi et al.
fluoroscopic guidance. We believe it is highly unlikely that these fiducials would migrate, and there were no instances of fiducial migration in our series. Bilateral symmetrical placement adds a level of confidence because in the unlikely event of fiducial migration, these fiducials would no longer be symmetrical on fluoroscopic imaging in the operating room. In addition, when approaching the pathology posteriorly, it is not uncommon to encounter and inadvertently remove the fiducial. In such an instance, the contralateral fiducial will remain to confirm localization of the correct level. There is no need to remove these fiducials intraoperatively or postoperatively because they have not been associated with any complications. Once fiducials were placed, a CT or CT myelogram was done to confirm placement of the fiducial and visualize its location relative to the level of pathology. Once this relationship is established, the actual level of pathology becomes irrelevant because it can be identified simply by the position of the fiducial marker. In our series, fiducials were placed without complications and with correct-level localization in all cases. Potential complications of this localization method include infectious and bleeding complications related to the fiducial placement. These patients are also subjected to a CT scan that they otherwise would not undertake. This CT is associated with additional radiation exposure for the patient, although this would reduce the intraoperative radiation exposure. If a CT myelogram is necessary, the myelogram carries unique potential complications including seizures, contrast reactions, renal insufficiency, bleeding complications, and potential for puncture of the conus.10 These are all very rare, and there were no such complications in our series. We believe the additional risks of complications and added cost are outweighed by the benefits of this localization technique. In this series, most patients underwent a CT myelogram following fiducial placement. Thoracic disc herniations and other thoracic spine pathology can often be seen on routine CT without the need for myelography. For this reason we recommend acquisition of a noncontrast CT immediately following fiducial placement. If the pathology cannot be reliably identified on CT, then a CT myelogram should be performed to visualize the pathology relative to the fiducial. Furthermore, CT is an important adjunct to MRI in the diagnosis and treatment of thoracic disc herniations because it can provide useful additional information regarding disc calcification.11 Fiducial placement and the subsequent CT or CT myelogram can be done at any time before surgery without compromising the localization. In the operating room, the CT images are available for review, and the surgeon is able to immediately locate the fiducial, and hence the level of pathology, using intraoperative fluoroscopy. In posterior cases, the fiducial was routinely encountered during the exposure adding further confidence regarding correct site localization. This localization method can be utilized regardless of the surgical approach (anterior, lateral, or posterior) and does not require any counting of ribs or Journal of Neurological Surgery—Part A
Vol. 76
No. A1/2015
vertebral bodies during surgery. Given the need for two additional procedures, this may not be appropriate when emergent intervention is needed, but it can be performed the same day as surgery when urgent intervention is needed. Weaknesses of this study include the relatively small number of patients. Thoracic pathology is not as commonly encountered as cervical or lumbar spine pathology. In addition, patients with prior spine instrumentation, severe compression fractures, or lytic bony lesions requiring surgical intervention were not included in this study because their pathology can be easily localized with intraoperative fluoroscopic images without the need for fiducial localization. We believe, based on our experience, that intraoperative fluoroscopy use and radiation exposure to the surgeon and operating room staff is minimized with this technique. However, this study does not directly compare intraoperative radiation exposure when utilizing fiducial marker localization with any other localization technique. Further investigation is necessary to resolve these issues.
Conclusion We describe a method for localization of thoracic spine pathology that combines the placement of small fiducial markers with subsequent CT imaging. This technique is easy to perform, very reliable, and has the potential to significantly reduce the use of intraoperative fluoroscopy. It is applicable to the entire thoracic spine, and with proper care and attention nearly eliminates localization errors during thoracic spine surgery. We believe this study functions as a useful feasibility study, and larger investigations are needed to more clearly define its reliability and identify any potential complications.
References 1 Mody MG, Nourbakhsh A, Stahl DL, Gibbs M, Alfawareh M, Garges
2
3
4
5
6
KJ. The prevalence of wrong level surgery among spine surgeons. Spine 2008;33(2):194–198 Hanson EH, Mishra RK, Chang DS, et al. Sagittal whole-spine magnetic resonance imaging in 750 consecutive outpatients: accurate determination of the number of lumbar vertebral bodies. J Neurosurg Spine 2010;12(1):47–55 Singh H, Meyer SA, Hecht AC, Jenkins AL III. Novel fluoroscopic technique for localization at cervicothoracic levels. J Spinal Disord Tech 2009;22(8):615–618 Nowitzke A, Wood M, Cooney K. Improving accuracy and reducing errors in spinal surgery—a new technique for thoracolumbar-level localization using computer-assisted image guidance. Spine J 2008;8(4):597–604 Rosahl SK, Gharabaghi A, Liebig T, Feste CD, Tatagiba M, Samii M. Skin markers for surgical planning for intradural lesions of the thoracic spine. Technical note. Surg Neurol 2002;58(5): 346–348 Paolini S, Ciappetta P, Missori P, Raco A, Delfini R. Spinous process marking: a reliable method for preoperative surface localization of intradural lesions of the high thoracic spine. Br J Neurosurg 2005; 19(1):74–76
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
70
Thoracic Spine Localization Using Fiducials
Anaizi et al.
9 Chin KR, Seale JM, Cumming VM. Avoidance of wrong-level
Preoperative localization of herniated thoracic discs using myeloCT guided transpleural puncture: technical note. Childs Nerv Syst 2007;23(1):21–26 8 Hsu W, Sciubba DM, Sasson AD, et al. Intraoperative localization of thoracic spine level with preoperative percutaneous placement of intravertebral polymethylmethacrylate. J Spinal Disord Tech 2008; 21(1):72–75
thoracic spine surgery using sterile spinal needles. J Spinal Disord Tech 2013; October 31 (Epub ahead of print) 10 Sandow BA, Donnal JF. Myelography complications and current practice patterns. AJR Am J Roentgenol 2005;185(3): 768–771 11 Dietze DD Jr, Fessler RG. Thoracic disc herniations. Neurosurg Clin N Am 1993;4(1):75–90
This document was downloaded for personal use only. Unauthorized distribution is strictly prohibited.
7 Cornips E, Beuls E, Geskes G, Janssens M, van Aalst J, Hofman P.
71
Journal of Neurological Surgery—Part A
Vol. 76
No. A1/2015
Copyright of Journal of Neurological Surgery. Part A. Central European Neurosurgery is the property of Georg Thieme Verlag Stuttgart and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
