Intravascular ultrasound to aid in the diagnosis and revision of an intra-aortic pedicle screw: illustrative case

Landon D. Ehlers Departments of Neurosurgery and

Search for other papers by Landon D. Ehlers in
Current site
jns
Google Scholar
PubMed
Close
 MD
,
Patrick J. Opperman Departments of Neurosurgery and

Search for other papers by Patrick J. Opperman in
Current site
jns
Google Scholar
PubMed
Close
 MD, MS
,
Jack E. Mordeson Departments of Neurosurgery and

Search for other papers by Jack E. Mordeson in
Current site
jns
Google Scholar
PubMed
Close
 BS
,
Jonathan R. Thompson Surgery, University of Nebraska, Omaha, Nebraska

Search for other papers by Jonathan R. Thompson in
Current site
jns
Google Scholar
PubMed
Close
 MD
, and
Daniel L. Surdell Departments of Neurosurgery and

Search for other papers by Daniel L. Surdell in
Current site
jns
Google Scholar
PubMed
Close
 MD
Open access

BACKGROUND

Pedicle screw impingement on vessel walls has the potential for complications due to pulsatile effects and wall erosion. Artifacts from spinal instrumentation create difficulty in accurately evaluating this interface. The authors present the first case of intravascular ultrasound (IVUS) used to characterize a pedicle screw breach into the aortic lumen.

OBSERVATIONS

A 21-year-old female with surgically corrected scoliosis underwent computed tomography angiography (CTA) 3 years postoperatively, which revealed a pedicle screw within the thoracic aorta lumen. Metal artifact distorted the CTA images, which prompted the decision to use intraoperative IVUS. The IVUS confirmed the noninvasive imaging findings and guided final decisions regarding aortic endograft size and location during spine hardware revision.

LESSONS

For asymptomatic patients presenting with pedicle screws malpositioned in or near the aorta, treatment decisions revolve around the extent of vessel wall penetration. Intraluminal depth can be obscured by artifact on computed tomography or magnetic resonance imaging or inadequately evaluated by a transesophageal echocardiogram. In our intraoperative experience, IVUS confirmed the depth of vessel lumen violation by a single pedicle screw and no wall penetration by two additional screws of concern. This was useful in deciding on thoracic endovascular aortic repair graft size and landing zone and facilitated safe spinal instrumentation removal and revision.

ABBREVIATIONS

CT = computed tomography; CTA = computed tomography angiography; IVUS = intravascular ultrasound; MRI = magnetic resonance imaging; TEE = transesophageal echo; TEVAR = transthoracic endovascular aortic repair

BACKGROUND

Pedicle screw impingement on vessel walls has the potential for complications due to pulsatile effects and wall erosion. Artifacts from spinal instrumentation create difficulty in accurately evaluating this interface. The authors present the first case of intravascular ultrasound (IVUS) used to characterize a pedicle screw breach into the aortic lumen.

OBSERVATIONS

A 21-year-old female with surgically corrected scoliosis underwent computed tomography angiography (CTA) 3 years postoperatively, which revealed a pedicle screw within the thoracic aorta lumen. Metal artifact distorted the CTA images, which prompted the decision to use intraoperative IVUS. The IVUS confirmed the noninvasive imaging findings and guided final decisions regarding aortic endograft size and location during spine hardware revision.

LESSONS

For asymptomatic patients presenting with pedicle screws malpositioned in or near the aorta, treatment decisions revolve around the extent of vessel wall penetration. Intraluminal depth can be obscured by artifact on computed tomography or magnetic resonance imaging or inadequately evaluated by a transesophageal echocardiogram. In our intraoperative experience, IVUS confirmed the depth of vessel lumen violation by a single pedicle screw and no wall penetration by two additional screws of concern. This was useful in deciding on thoracic endovascular aortic repair graft size and landing zone and facilitated safe spinal instrumentation removal and revision.

ABBREVIATIONS

CT = computed tomography; CTA = computed tomography angiography; IVUS = intravascular ultrasound; MRI = magnetic resonance imaging; TEE = transesophageal echo; TEVAR = transthoracic endovascular aortic repair

The vertebral pedicle screw is a mainstay of spine surgery because it provides superior stabilization by incorporating the entire breadth of the vertebrae. Freehand thoracic and lumbosacral pedicle screw placement has been shown to have a 2.3% rate of pedicle cortex breach by ≥2 mm.1 Neurovascular structures in proximity are at risk for damage if the pedicle is violated significantly.2 The descending aorta is primarily at risk throughout the length of the thoracic spinal column.3 Although the overall incidence of injury from posterior spine surgery is very rare, injury to the great vessels has up to a 61% mortality risk if not recognized.4,5

Acute recognition of aorta violation by a pedicle screw is dealt with immediately, typically with vascular surgery consultation and direct vessel repair or endovascular graft placement. However, when delayed recognition of pedicle screw impingement occurs, management of the asymptomatic patient is debated. Pedicle screws abutting the aortic wall in bovine models leads to chronic vessel changes from repeated pulsation.6 These changes are realized both histopathologically with wall thinning and collagen deposition as well as biomechanically through lower failure stress.6 Pedicle screw malposition results in morbidity and mortality concerns due to aortic wall erosion over time and is addressed in multiple case reports.7–10

The diagnosis of pedicle screw impingement on vascular structures is typically made by computed tomography (CT) or magnetic resonance imaging (MRI). Hardware artifacts in noninvasive imaging techniques may not allow clear display of screw impingement or erosion into the vessel wall. We describe the use of intravascular ultrasound (IVUS) to evaluate the depth of pedicle screw penetration of the thoracic aorta to guide treatment. To our knowledge, this is the first use of IVUS to evaluate the integrity of the blood vessel and confirm intraluminal violation of a malpositioned pedicle screw.

Illustrative Case

History and Examination

A 21-year-old female with scoliosis surgically treated by an outside provider 3 years earlier presented to the emergency department for acute chest pain. Work-up included chest CT angiography (CTA), which appeared to reveal aortic wall penetration by the left T5 pedicle screw (Fig. 1). Her chest pain resolved without incident, and the finding of the remaining work-up was negative, but, given the patient’s young age and concern for long-term complications from aortic erosion, surgical intervention was offered.

FIG. 1.
FIG. 1.

Preoperative CTA. Left: Axial view at the T5 level of the left T5 pedicle screw laterally breached and violating the aortic wall (white arrow). Right: Corresponding sagittal view of the left T5 pedicle screw.

The T5 pedicle screw appeared to breach the lumen based on the CTA, but there was no evidence of a pseudoaneurysm. The left-sided T4 and T7 pedicle screws were also noted to be impinging the aorta; however, metal artifact on the CTA limited accurate assessment of aortic involvement. Revision of the T5 pedicle screw would involve transthoracic endovascular aortic repair (TEVAR), so a multidisciplinary approach with vascular surgery and neurosurgery was planned. Alternative imaging modalities were considered not only to help define the length of screw transgressing the aortic wall but also to determine if the T4 and T7 screws had caused vessel erosion. A transesophageal echo (TEE) was suggested as an alternative to assess the integrity of the thoracic aorta. Uncertain of the utility and efficacy of TEE in characterizing aortic wall impingement and considering that a TEVAR was likely necessary for safe and definitive treatment, we elected to use intraoperative IVUS to more accurately evaluate all three levels involved.

Operation

The patient was intubated, and TEE was performed. This did not adequately visualize the aorta to show if the pedicle screw was intraluminal. For the first stage of the procedure, the patient was placed in a modified right lateral decubitus position to allow simultaneous femoral artery access and spinal access to expose the spinal hardware over the T5 screw. The left femoral artery was accessed by cutdown because of her small-caliber vessel (5–6 mm) and the need for a 20-French sheath. Cutdown access also allowed primary vessel repair before safely positioning the patient prone for final stabilization. While the vascular surgeons approached the femoral artery, the neurosurgery team exposed directly over the thoracic levels of concern. The rod was cut using a rod cutting burr and was removed from the saddle. The screw at T5 was identified anatomically and confirmed with fluoroscopy. Once the femoral artery was accessed with a 9-French sheath, the large-bore IVUS catheter (Visions PV 0.035, Philips Volcano Corp.) was guided into the thoracic aorta. IVUS confirmed left T5 screw intraluminal violation by 8 mm (Fig. 2). Aortic wall impingement by the T4 and T7 pedicle screws was confirmed with IVUS, but there was no evidence of intra-aortic violation at T4 or T7. Confirmation that there were no additional sites of aortic perforation aided in guiding vessel graft size and choice of spinal level to center the graft placement (T5).

FIG. 2.
FIG. 2.

Intraoperative IVUS image confirming erosion of the T5 pedicle screw (white arrow) through the aortic wall. Intraoperative measurement estimated the screw depth to be 8 mm within the aortic lumen.

The patient was heparinized to a goal activated clotting time of 250, and the vascular team partially deployed the Zenith TX2 dissection endovascular graft (22 mm × 79 mm; Cook Medical) starting distal to the left subclavian artery, placing the center of the graft over the T5 pedicle screw location. The Cook TX2 graft was chosen because of the short graft length, small graft diameter, and relatively small delivery profile. Once partial deployment reached the level of the T5 pedicle screw, the offending screw was removed (Fig. 3). The remainder of the graft was deployed without incident. The patient remained hemodynamically stable throughout the procedure, and a completion aortogram showed no evidence of endoleak or extravasation (Fig. 4). The sheath was removed, the femoral artery was primarily repaired, and the groin incision was closed in layers. Flow was confirmed to the feet with a handheld Doppler. Subsequently, the patient was placed prone to allow restabilization of the spine with additional hardware. The impinging T4 and T7 pedicle screws were revised, and the corresponding rod was placed without incident.

FIG. 3.
FIG. 3.

Intraoperative fluoroscopy showing partial deployment of the TEVAR graft starting distal to the left subclavian artery and centered over the T5 pedicle screw. Driver is seen to be engaged in the screw head to allow tandem removal while the remainder of the graft is deployed.

FIG. 4.
FIG. 4.

Completion angiography after TEVAR graft deployment showing good wall apposition and no evidence of endoleak or extravasation.

Outcome

Postoperatively, the patient was monitored in the intensive care unit and did very well. No cardiovascular concerns developed during her stay, and she required minimal pain management. She was discharged on postoperative day 3. At her 1-month follow-up, she was clinically asymptomatic (Fig. 5).

FIG. 5.
FIG. 5.

Follow-up CTA at 1 month showing revised spinal hardware at T5 and stable vascular graft wall apposition without evidence of endoleak or pseudoaneurysm development.

Patient Informed Consent

The necessary patient informed consent was obtained in this study.

Discussion

Observations

The risk of a pedicle screw malposition is increased in the setting of scoliosis due to the complex curvature, which distorts the normal anatomical relationships. When scoliosis is treated in the pediatric population, the use of fluoroscopy is often minimized or abandoned to reduce the risk associated with cumulative radiation dose at a young age. Inherently, this increases the risk of pedicle screw malposition. A minilaminectomy has been described as a means of localizing the pedicle for optimizing pedicle screw placement in scoliosis cases to avoid using fluoroscopy; however, severe scoliosis can also make this process difficult.11 The position of the thoracic aorta is more lateral and posterior in patients with scoliosis relative to their counterparts with a normal spine.12 This change in anatomical relationship places the vessel at increased risk with lateral trajectory pedicle screws.

Penetration of the aorta has obvious acute consequences, but, even with impingement, there is concern about erosion of the vascular wall through chronic microtrauma from aortic pulsation.3 Chronic aortic violation from a pedicle screw does not have a well-defined history, which leads to debate about when it is appropriate to treat an asymptomatic patient. Without intervention, the concern is for delayed manifestation secondary to vessel wall injury and/or pseudoaneurysm formation.6,10

An in vivo bovine model involved intentionally abutting thoracic pedicle screws to the aorta to evaluate vessel erosion.6 Follow-up microscopic evaluation revealed wall erosion into the tunica intima as early as 3 months postoperatively. Histopathologic analysis showed disorganized replacement of the tunica media with collagen tissue. Over 12 months, the remodeling led to a thickened arterial wall but biomechanically had 33% lower failure stress. This study confirms that pedicle screw–aorta interaction is not benign. However, the erosive effects of pedicle screw–aortic wall interactions can be difficult to diagnose and are often overlooked until symptoms arise or vascular imaging is obtained. Despite the delayed diagnosis, several groups have recommended revision of the spinal hardware and repair of the aortic wall by open or endovascular methods.3,7

Spinal hardware revision can be completed with vascular surgery to repair the aorta or at a separate occurrence if patient positioning prohibits feasibility in the same operative setting. Options to repair the aorta include 1) primary repair, 2) patch angioplasty, 3) TEVAR, and 4) interposition grafting.3 Most cases are performed using endovascular methods, unless a pseudoaneurysm is found; then, open surgical primary repair is preferred.3,7

Recommendations regarding treatment have centered on the depth of pedicle screw penetration into the aorta. Screws impinging <5 mm into the aorta and without evidence of extravasation on vascular studies could be removed without aortic repair.3 The decision to replace a pedicle screw takes into consideration the screw depth into the aorta, location of the screw, and spine stability when the screw is removed. If the screw has >5 mm of aortic impingement, then repair of the aorta in conjunction with removal of the offending screw is recommended.3 However, it can be difficult to appreciate aortic impingement on noninvasive imaging, because CT metallic scatter beams or MRI metal artifact distorts the screw–aorta interface. This may also cause measurement errors, which could be problematic if using the 5-mm cutoff for guiding treatment.

Other noninvasive imaging modalities could be considered, such as TEE, which can typically identify aortic pathology with great reliability. However, this may not be as effective in identifying aortic breach from a pedicle screw, as we noted in our case. IVUS was an effective tool to reveal not only that the screw was truly intravascular but also how deep it had violated the vessel. Additionally, our patient’s preoperative CTA showed impingement of the left T4 and T7 pedicle screws, but IVUS confirmed that there was no compromise to the vessel wall. This information was useful to decide on the TEVAR graft size, landing zone, and proceeding with revision of the screws in the traditional prone position without the need for further angiography.

Lessons

The incidence of aortic injury from pedicle screws is believed to be low, but even delayed diagnosis in an asymptomatic patient has the risk of late vessel injury. With this in mind, surgeons should consider revision of the spinal hardware and vascular repair. Because traditional spine imaging modalities may not clearly define the extent of pedicle screw–aorta interaction, we present the first use of IVUS to assist in the diagnosis and treatment of thoracic aorta violation by pedicle screw. With endovascular repair being a common treatment choice, initially using IVUS can identify the depth of screw penetration in a reliable and radiation-free manner. If the screw is not violating the vessel lumen, surgeons may elect to remove the screw without the need for a graft. This would afford the patient one less implant and obviate the need for long-term imaging surveillance.

Disclosures

Dr. Thompson is the program director for the vascular surgery fellowship, which receives an annual educational grant from Gore Medical for simulation-based training.

Author Contributions

Conception and design: Thompson, Surdell. Acquisition of data: Ehlers, Thompson. Analysis and interpretation of data: Mordeson, Thompson. Drafting the article: Mordeson, Ehlers, Opperman. Critically revising the article: Mordeson, Ehlers, Opperman, Thompson. Reviewed submitted version of manuscript: Mordeson, Opperman, Thompson. Approved the final version of the manuscript on behalf of all authors: Mordeson. Administrative/technical/material support: Opperman, Thompson. Study supervision: Thompson.

References

  • 1

    Alomari S, Lubelski D, Lehner K, et al. Safety and accuracy of freehand pedicle screw placement and the role of intraoperative O-arm: a single-institution experience. Spine (Phila Pa 1976). 2023;48(3):180188.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 2

    Modi HN, Suh SW, Fernandez H, Yang JH, Song HR. Accuracy and safety of pedicle screw placement in neuromuscular scoliosis with free-hand technique. Eur Spine J. 2008;17(12):16861696.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 3

    Kayacı S, Cakir T, Dolgun M, et al. Aortic injury by thoracic pedicle screw. When is aortic repair required? Literature review and three new cases. World Neurosurg. 2019;128:216224.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 4

    Szolar DH, Preidler KW, Steiner H, et al. Vascular complications in lumbar disk surgery: report of four cases. Neuroradiology. 1996;38(6):521525.

  • 5

    Bingol H, Cingoz F, Yilmaz AT, Yasar M, Tatar H. Vascular complications related to lumbar disc surgery. J Neurosurg. 2004;100(3 Suppl Spine):249253.

  • 6

    Faro FD, Farnsworth CL, Shapiro GS, et al. Thoracic vertebral screw impingement on the aorta in an in vivo bovine model. Spine (Phila Pa 1976). 2005;30(21):24062413.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 7

    Zerati AE, Leiderman DB, Teixeira WG, et al. Endovascular treatment of late aortic erosive lesion by pedicle screw without screw removal: case report and literature review. Ann Vasc Surg. 2017;39:285.e17285.e21.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 8

    Jendrisak MD. Spontaneous abdominal aortic rupture from erosion by a lumbar spine fixation device: a case report. Surgery. 1986;99(5):631633.

  • 9

    Choi JB, Han JO, Jeong JW. False aneurysm of the thoracic aorta associated with an aorto-chest wall fistula after spinal instrumentation. J Trauma. 2001;50(1):140143.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 10

    Ohnishi T, Neo M, Matsushita M, Komeda M, Koyama T, Nakamura T. Delayed aortic rupture caused by an implanted anterior spinal device. Case report. J Neurosurg. 2001;95(2 suppl):253256.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 11

    Di Silvestre M, Parisini P, Lolli F, Bakaloudis G. Complications of thoracic pedicle screws in scoliosis treatment. Spine (Phila Pa 1976). 2007;32(15):16551661.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 12

    Sucato DJ, Duchene C. The position of the aorta relative to the spine: a comparison of patients with and without idiopathic scoliosis. J Bone Joint Surg Am. 2003;85(8):14611469.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • Collapse
  • Expand
  • FIG. 1.

    Preoperative CTA. Left: Axial view at the T5 level of the left T5 pedicle screw laterally breached and violating the aortic wall (white arrow). Right: Corresponding sagittal view of the left T5 pedicle screw.

  • FIG. 2.

    Intraoperative IVUS image confirming erosion of the T5 pedicle screw (white arrow) through the aortic wall. Intraoperative measurement estimated the screw depth to be 8 mm within the aortic lumen.

  • FIG. 3.

    Intraoperative fluoroscopy showing partial deployment of the TEVAR graft starting distal to the left subclavian artery and centered over the T5 pedicle screw. Driver is seen to be engaged in the screw head to allow tandem removal while the remainder of the graft is deployed.

  • FIG. 4.

    Completion angiography after TEVAR graft deployment showing good wall apposition and no evidence of endoleak or extravasation.

  • FIG. 5.

    Follow-up CTA at 1 month showing revised spinal hardware at T5 and stable vascular graft wall apposition without evidence of endoleak or pseudoaneurysm development.

  • 1

    Alomari S, Lubelski D, Lehner K, et al. Safety and accuracy of freehand pedicle screw placement and the role of intraoperative O-arm: a single-institution experience. Spine (Phila Pa 1976). 2023;48(3):180188.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 2

    Modi HN, Suh SW, Fernandez H, Yang JH, Song HR. Accuracy and safety of pedicle screw placement in neuromuscular scoliosis with free-hand technique. Eur Spine J. 2008;17(12):16861696.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 3

    Kayacı S, Cakir T, Dolgun M, et al. Aortic injury by thoracic pedicle screw. When is aortic repair required? Literature review and three new cases. World Neurosurg. 2019;128:216224.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 4

    Szolar DH, Preidler KW, Steiner H, et al. Vascular complications in lumbar disk surgery: report of four cases. Neuroradiology. 1996;38(6):521525.

  • 5

    Bingol H, Cingoz F, Yilmaz AT, Yasar M, Tatar H. Vascular complications related to lumbar disc surgery. J Neurosurg. 2004;100(3 Suppl Spine):249253.

  • 6

    Faro FD, Farnsworth CL, Shapiro GS, et al. Thoracic vertebral screw impingement on the aorta in an in vivo bovine model. Spine (Phila Pa 1976). 2005;30(21):24062413.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 7

    Zerati AE, Leiderman DB, Teixeira WG, et al. Endovascular treatment of late aortic erosive lesion by pedicle screw without screw removal: case report and literature review. Ann Vasc Surg. 2017;39:285.e17285.e21.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 8

    Jendrisak MD. Spontaneous abdominal aortic rupture from erosion by a lumbar spine fixation device: a case report. Surgery. 1986;99(5):631633.

  • 9

    Choi JB, Han JO, Jeong JW. False aneurysm of the thoracic aorta associated with an aorto-chest wall fistula after spinal instrumentation. J Trauma. 2001;50(1):140143.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 10

    Ohnishi T, Neo M, Matsushita M, Komeda M, Koyama T, Nakamura T. Delayed aortic rupture caused by an implanted anterior spinal device. Case report. J Neurosurg. 2001;95(2 suppl):253256.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 11

    Di Silvestre M, Parisini P, Lolli F, Bakaloudis G. Complications of thoracic pedicle screws in scoliosis treatment. Spine (Phila Pa 1976). 2007;32(15):16551661.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 12

    Sucato DJ, Duchene C. The position of the aorta relative to the spine: a comparison of patients with and without idiopathic scoliosis. J Bone Joint Surg Am. 2003;85(8):14611469.

    • PubMed
    • Search Google Scholar
    • Export Citation

Metrics

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 1295 1295 181
PDF Downloads 274 274 17
EPUB Downloads 0 0 0