Skeletal Radiol (2015) 44:1–8 DOI 10.1007/s00256-014-2028-x
REVIEW ARTICLE
Cervical foraminal steroid injections under CT guidance: retrospective study of in situ contrast aspects in a serial of 248 cases Pierre Pottecher & Denis Krausé & Lucy Di Marco & Romaric Loffroy & Louis Estivalet & Romain Duhal & Xavier Demondion
Received: 6 November 2013 / Revised: 8 August 2014 / Accepted: 29 September 2014 / Published online: 16 October 2014 # ISS 2014
Abstract Objectives To describe all the CT findings after in situ contrast injection just before steroid injection and to recognize the abnormal aspects associated with intravascular contamination. Material and methods We retrospectively evaluated 248 cervical transforaminal steroid injections done at the university hospital in Dijon, France, in 2008–2012, to treat cervicobrachial neuralgia inadequately improved by optimal medical treatment for at least 3 weeks. Features describing the opacification patterns were recorded. Results Five main nonvascular opacification patterns were identified: clumps of contrast agent outside the foramen (16 %), a crab claw pattern surrounding the ganglion (13 %), a “French” circumflex accent pattern (15 %), reflux along the needle (7 %), and facet joint capsule opacification (22 %). Concerning the situations requiring a change in needle position, intravenous injection occurred in 26 % of the patients, with a crab claw pattern in half the cases and a clump pattern in half the cases. Intraarteriolar injection was noted in two patients. Conclusion CT after in situ contrast injection ensures proper needle positioning outside the blood vessels before steroid injection. Penetration of the needle tip into a vein is very common, whereas arteriolar puncture is extremely rare.
P. Pottecher (*) : D. Krausé : L. Di Marco : R. Loffroy : L. Estivalet Département de Radiologie et Imagerie Diagnostique et interventionnelle, CHU Dijon Bocage Central, Dijon, France e-mail: [email protected] R. Duhal : X. Demondion Service de radiologie musculosquelettique, CCIAL, laboratoire d’anatomie, faculté de médecine de Lille, Hôpital Roger Salengro, CHRU de Lille, Lille, France
Keywords Steroid infiltrations . Cervical radiculopathy . Contrast in situ control . Intervertebral foramen
Introduction Cervicobrachial neuralgia (CBN) in young individuals is a common condition usually ascribable to impingement on a nerve root of a soft disc herniation, most often at the entrance of or within the foramen [1]. The pain is due not only to nerve root compression by the disc, but also to local inflammation with release of enzymes and proinflammatory mediators. Severe pain present even at night and responsible for disability is common. Persistent pain despite optimal treatment for at least 3–6 weeks is a well-accepted indication for a local corticosteroid injection [2–4]. To be effective, the injection must place the corticosteroid in contact with the nerve root at the point of disc impingement, which is the site of proinflammatory mediator release [5]. The most widely used approaches are transforaminal, transfacet (which is followed by diffusion through the synovial joint), and posterior interlaminar (Fig. 1) [6–8]. Numerous cases of severe neurological complications of cervical corticosteroid injections have been reported [9–11]. These complications include spinal cord and brain infarction responsible for tetraplegia, paraplegia, or death. In 2011, the number of reported devastating neurological complications of cervical corticosteroid injections was about 14 in France and over 60 worldwide. The primary objective of this article is to establish, by a radioanatomical correlation, a precise description of all CT findings with contrast in situ before a transforaminal steroid injection for CBN. The population study was composed by 248 patients addressed to the Hospital of Dijon (Burgundy, France) between 2008 to 2012 for a CBN resistant to a medical treatment.
2
Fig. 1 Axial CT slice: C4–C5 foraminal herniation. Options for corticosteroid injection include the (a) transforaminal, (b) posterior articular, and (c) interlaminar approaches
Anatomy of the cervical foramina Boundaries Each cervical foramen is located between the pedicles of two adjoining vertebrae and has a 45° forward orientation parallel to the transverse processes. The foramen is bordered from cranial to caudal by the body of the suprajacent vertebra, intervertebral disc, and uncus of the suprajacent vertebra. The facet joint forms the posterior border. The foramen is also bordered in part by the facet joint synovial capsule, which extends as it recesses upward and forward toward the foramen and downward and backward toward the spinous process and ligamentum flavum (Figs. 2, 3, 4 and 5).
Skeletal Radiol (2015) 44:1–8
the uncinate process, and of the dorsal root ganglion, which often lies within a concavity of the superior articular process [12]. The nerve root exits the foramen at the immediately suprajacent intervertebral space (e.g., the C6 root exits at the C5–C6 space). The foramen also contains a radicular or radiculomedullary arteriole and the foraminal venous plexus surrounding the nerve root. The density of this plexus ensures good opacification of the epidural space and intervertebral foramen during contrast-enhanced CT or magnetic resonance imaging. The radiculomedullary arteries are fed by the subclavian artery via the vertebral artery or the ascending or deep cervical artery. In the most common configuration, about three radiculomedullary arteries arise from the vertebral artery at C3, another from the deep cervical artery at C6, and yet another from the ascending cervical artery at C8. Variants are common, however, and the configuration is asymmetric in most individuals [13]. Each radiculomedullary artery varies in diameter from 0.2 to 2 mm [14]. The radiculomedullary artery may course along either the anterior or the posterolateral part of the foramen. Consequently, the risk of accidental puncture is not predictable [13].
CT-guided transforaminal injection technique Availability of a team of skilled technicians is crucial to the safety and efficacy of transforaminal injections (Figs. 6–8).
Content
Study population
The cervical foramina contains the same nervous and vascular structures as the thoracolumbar foramina but with a smaller proportion of fat. The nervous structures are located in the posteroinferior part of the foramen and consist of the anterior root, located immediately posterior to the posterior margin of
The study population consisted of 248 patients (99 women, 149 men) recruited at the Hospital of Dijon (Burgundy, France) for a CBN resistant to a medical treatment from 4 to 6 weeks. All the patients had been treated by a general practice physician, rheumatologist or neurologist. The decision for the
Fig. 2 Cadaver study: cervical foramen in the oblique sagittal plane (a) and axial plane (b). Posterior roots (PRs) and anterior roots (ARs) in the lower part of the posterior foramen, with radicular arteries in contact with
the roots before (solid arrow). The joint cavity is delineated by the broken arrows. VB, vertebral body; D, intervertebral disc; Ped, pedicle; Sup art, superior articular process; Inf art, inferior articular process
Skeletal Radiol (2015) 44:1–8
3
Fig. 3 Cadaver study: axial slice through the left C7–T1 foramen. Connection of the joint capsule (broken arrows) with the foramen. VB, vertebral body; L, lamina; K1, anterior arc of the left first rib
transforaminal injection was made after a CT scan or MRI of the cervical spine demonstrated the existence of a hernia according with the topography of the radiculalgia. Concerning the preparation of the patient, a careful interview is mandatory to detect contraindications, such as a history of allergic reaction to contrast agents or anesthetics, and to evaluate the risk:benefit ratio. The use of medications that affect hemostasis, such as antiplatelet drugs, should be noted; clotting tests should be performed (prothrombin time, activated partial thromboplastin time, and international normalized ratio). The patient should receive detailed information about the procedure and its expected benefits and risks, then asked to provide consent. After the procedure, the patient is monitored closely in the hospital for 1 h. Course of action and equipment before the infiltration Installation of the patient The patient is supine with the head in neutral rotation and traction on the wrists to ensure fixed extension of the cervical spine and to improve the visibility of the C6–C7 and C7–T1 foramina.
Fig. 4 Cadaver study: upper (a) and lower (b) axial slices through C4– C5 showing a dense foraminal venous network. The epidural space is comparatively full as shown in Fig. 2b where the epidural network is dissected. Comparison with an axial CT slice (c) after contrast injection. The epidural and foraminal veins (broken arrows) form a dense venous network. VB, vertebral body; ART, articular process; SC, spinal cord; DRG, dorsal root ganglion
A thin-slice helical (1.25-mm) CT image centered on the level to be injected is acquired to visualize the disc-root impingement (Fig. 6) and to determine the needle trajectory toward the foramen that will avoid the vascular structures (jugular vein, carotid artery, and external jugular vein). A technician marks the skin. A local anesthetic is injected subcutaneously using a 21-G needle, which is the optimal caliber for obtaining a straight trajectory through the soft tissues and good anesthetic distribution at a distance from the foramen.
This diameter is considered optimal by several authors. The needle is inserted on the side of, and anterolateral to, the impingement, then advanced under CT guidance with 1.25mm slices until the tip reaches the foramen exit, in contact with the anterolateral edge of the articular process, far behind the vertebral artery. The ideal site is behind the ganglion, against the articular process.
The needle and its position
Contrast agent
A 22-G needle (diameter, about 400 μ) is used to minimize the risk of accidentally cannulating a vessel near the foramen.
With a 22-G needle, 1–1,2 ml is then injected. We use a nonionic, low-osmolality, iodinated contrast agent such
4
Skeletal Radiol (2015) 44:1–8
Fig. 5 Cadaver study of the cervical spine: radiculomedullary artery (arrow) included in the spinal anterior arterial axis. VB, vertebral body; DRG, dorsal root ganglion; SC, spinal cord
as iopamidol (Iopamiron 300, authorized for intrathecal use, Bracco Imaging, Courcouronnes, France). This fundamental step ensures good opacification of the distal foramen containing the ganglion and often the joint capsule. Vascular opacification by the contrast agent indicates faulty needle position within a blood vessel. The contrast agent may diffuse to the external and/or internal foramen or remain outside the foramen, where it penetrates the usual diffusion spaces (Fig. 7). Fig. 7 Axial CT slices showing diffusion at the needle tip (a) and foramen exit (b)
Results: in situ contrast and radiological-anatomical correlations
Normal findings
Study population and level of transforaminal injection
Clump pattern of diffusion: 16 % of cases
The descriptive statistics of the population are summarized in Table 1 and the levels involved in Table 2.
Close to the needle tip and articular process, the contrast agent formed polycyclic clumps, with moderate diffusion to the exit of the foramen.
Fig. 6 Axial CT slice with contrast injection showing a left C5–C6 disc herniation
Fig. 8 Axial CT slice showing a clump diffusion pattern of the contrast agent, with diffusion to the transverse foramen and foramen exit
Skeletal Radiol (2015) 44:1–8
5
Table 1 Descriptive statistics of the study population. The mean of age, visual analog scale (VAS) pain, and duration of cervicobrachial neuralgia (CBN) are presented with the standard deviation in parentheses Population (n=248) Age (years) Visual analog scale pain Duration of CBN (days)
54.5 (15.3) 5.8 (1.2) 41 (13.4)
Mean and standard deviation in parentheses
Diffusion around the ganglion: 28 % of cases (Figs. 9 and 10) Diffusion predominated at the foramen exit. The ganglion footprint resembled a crab’s claw (Fig. 9) or a circumflex accent (Fig. 10). Reflux along the needle: 7 % of cases (Fig. 11) Near the needle tip, retrograde contrast agent diffusion mimicked venous opacification. However, the symmetrical pattern of distribution on either side of the needle suggested reflux rather than venous opacification (Fig. 11). Other radiological and anatomical findings Joint capsule opacification: 22 % of cases (Fig. 12) When positioning the needle at the posterosuperior part of the foramen exit, the facet joint capsule may be contacted or perforated. The contrast agent may therefore diffuse up to the spinous process and/or contralaterally, mimicking venous opacification. This capsule forms recesses upward and forward toward the intervertebral foramen and downward and backward to the spinous process and ligamentum flavum (Fig. 12). However, corticosteroid injection can be performed safely [7].
Fig. 9 Axial CT slice showing a crab claw pattern of diffusion produced by posteroanterior molding of the contrast agent around the dorsal root ganglion at the foramen exit
or contralateral intraductal and/or foraminal veins. Venous contamination was consistently combined with a clump or crab claw pattern. The CT slices must be extended to the levels above and below the injection to look for distal contamination of the spinal muscle veins (Fig. 13). Arteriolar contamination: 0.8 % of cases Of the 248 procedures, 2 (0.8 %) resulted in arterial opacification. The immediate diagnosis is the absence of contrast at the contact of the distal end of the needle, which can make us think of an incorrect injection of anesthesia instead of contrast (Figs. 14 and 15)
Findings requiring needle repositioning Intravenous injection: 26 % of cases (Fig. 13) Among 248 cervical transforaminal injections, 64 (26 %) resulted in foraminal vein opacification manifesting as abnormal persistence of the contrast agent within the ipsilateral
Table 2 Percentage of the cervical spine levels involved in cervicobrachial neuralgia (CBN) by hernia
Levels involved
Percentage
C4–C5 C5–C6 C6–C7 C7–T1
3% 27 % 65 % 5%
Fig. 10 Axial CT slice showing a circumflex accent pattern of diffusion (∧) produced by posteroanterior molding of the contrast agent around the dorsal root ganglion at the foramen exit
6
Skeletal Radiol (2015) 44:1–8
Fig. 11 Axial CT slice showing retrograde diffusion along the needle mimicking intravenous injection. The symmetrical distribution on both sides of the needle suggests diffusion along the needle. Note the concomitant opacification of the joint capsule
Fig. 13 Axial CT slices showing venous contamination after injection of 1 ml of contrast agent. In contact with the needle, posterior epidural spread of contrast agent opacification associated with the paraspinal veins (B) and contralateral foraminal veins (black arrow Fig. 12C)
Discussion
Fig. 12 Axial CT slices showing joint capsule opacification with contrast agent diffusion along the articular recess (b) extending upward and forward toward the intervertebral foramen and downward and backward toward the lamina (c) and ligamentum flavum
CBN generates a huge burden of suffering, work disability, and costs to society. Optimal treatment is therefore crucial. When CBN due to disc impingement on a nerve root persists despite optimal medical treatment for at least 3 weeks, transforaminal corticosteroid injections have been proven effective [2–4]. In France, 52,000 intraspinal corticosteroid injections were performed each year in 2009/2010 [15], usually to treat sciatica, femoral neuralgia, or CBN. In our experience over a 12-year period, transforaminal corticosteroid injection was satisfaying in 75 % of patients before 1 month (visual analog scale pain score less than 2). Reports of devastating neurological complications after transforaminal corticosteroid injections have generated considerable reluctance among interventional radiologists to use this method at the cervical or lumbar spine [9–11, 16, 17]. A survey among US physicians collected 30 cases of brain and/or spinal cord infarct with 13 deaths [18]. However, we contend that
Skeletal Radiol (2015) 44:1–8
7
Fig. 14 Axial CT slices showing arteriolar opacification. In contact with the tip of the needle (a), despite the injection of 0.8–1 ml of contrast agent, no diffusion is visible, indicating a strong epidural arteriolar washout effect (b)
transforaminal injection is safe provided CT guidance and a number of precautions are used. The transforaminal approach is well suited to the treatment of CBN due to disc impingement on a nerve root, as the corticosteroid is delivered at the site of release of proinflammatory mediators, which play a major role in pain generation. The contrast medium and therefore the corticosteroid diffuse around the ganglion [19]. Other effective approaches include transfacet injection with diffusion along the synovial membrane extensions to the impingement site [7]. The interlaminar approach is more painful and has been reported to induce local hematoma formation [20, 21]. Fluoroscopic guidance with subtraction angiography may provide better arteriolar contrast and similar efficacy compared to CT, but requires considerable experience on the part of the interventional radiologist [22]. Fluoroscopy ensures immediate detection of faulty needle position in the radiculomedullary artery or, more often, foraminal vein. In a 2-year retrospective study of 177 cervical transforaminal injections, adding digital subtraction angiography to fluoroscopic guidance doubled the proportion of patients with detected intravascular injection [23], which occurred in 33 % of patients and was consistently intravenous. CT guidance shows needle progression to the foramen exit, thereby allowing the operator to avoid the vessels (common carotid artery, internal and external jugular veins, and vertebral artery). Accidental puncture of an artery can be fatal. Many variants of vertebral artery anatomy exist, making CT guidance invaluable [24–26]. An interesting alternative is Doppler ultrasound,
Fig. 15 Axial CT slices showing arteriolar opacification after injection of 1–1.5 ml of contrast agent. A small amount of contrast agent is visible at the foramen exit. Arteriolar washout explains why the observed amount of contrast is smaller than the injected amount: (a) and (b). The volume of contrast agent distribution increased after needle repositioning (c), indicating that the injection could be performed safely
which visualizes the vessels located outside the foramen, especially in case of posterior distribution (as described in anatomy [13]), thus avoiding any vascular cannulation [27]. Using a 21-G needle for soft tissue anesthesia ensures better and faster diffusion all around the foramen, 4–5 cm under the skin, at the level of the disc herniation, compared to thinner needles. After anesthesia, insertion of a 22-G needle under CT guidance is painless. The needle tip is positioned at the distal foramen, just behind the ganglion, against the articular process. The 22-G diameter is the most widely used in France. Thinner needles (24–25 G) are deformable and do not eliminate the risk of intravascular injection with spinal cord infarction [28]. The contrast agent diffusion pattern showed no intravascular contamination in 73 % of our patients. Venous contamination occurred in 26 % of patients, half of whom had a crab claw and half a clump pattern of diffusion. In the remaining two patients, intraarterial injection occurred, with little or no visible contrast agent, indicating an immediate washout effect. Injection into a vein or artery requires needle repositioning under CT guidance.
8
Skeletal Radiol (2015) 44:1–8
The mechanisms responsible for ischemic neurological complications are well documented. Abnormalities of the arterioles may consist of dissection, spasm, or thrombosis by steroid emulsion aggregates [17]. Blood reflux into the needle is never observed, a fact that makes CT guidance particularly important to detect intravascular injection. We obtained six 1.25-mm slices, including a slice that clearly showed the needle tip at the foramen. This well-standardized procedure carried out by a highly trained team requires a CT acquisition time of only 10 min.
Conclusion Ambulatory transforaminal corticosteroid injection under CT guidance with routine contrast agent injection immediately before the corticosteroid injection plays a key role in the treatment of CBN due to disc impingement. Thorough understanding of contrast agent diffusion is essential to ensure the recognition of faulty needle position within a vein or arteriole. In every day practice, venous contaminations are very frequent; on the other hand, an arteriolar injection is exceptional. Considerable experience and regular practice on the part of the interventional radiologist are needed to ensure patient safety.
References 1. Radhakhrishnan K, Litchy WJ, O’Fallon WM, Kurland LT. Epidemiology of cervical radiculopathy: a population-based study from Rochester, Minnesota, 1976 through 1990. Brain. 1994;117: 325–35. 2. Bush K, Hillier S. Outcome of cervical radiculopathy treated with periradicular/epidural corticosteroid injections: a prospective study with independent clinical review. Eur Spine J. 1996;5:319–25. 3. Van Zundert J, Harney D, Joosten EA, Durieux ME, Patijn J, Prins MH, et al. The role of the dorsal root ganglion in cervical radicular pain: diagnosis, pathophysiology, and rationale for treatment. Reg Anesth Pain Med. 2006;31:152–67. 4. Cyteval C, Thomas E, Decoux E, Sarrabere MP, Cottin A, Blotman F, et al. Cervical radiculopathy: Open study on percutaneous periradicular foraminal steroid infiltration performed under CT control in 30 patients. AJNR. 2004;25:441–5. 5. Anderberg L, Annertz M, Persson L, Brandt L, Saveland H. Transforaminal steroid injections for the treatment of cervical radiculopathy: a prospective and randomised study. Eur Spine J. 2007;16:321–8. 6. Rathmell JP, Aprill C, Bogduk N. Cervical transforaminal injection of steroids. Anesthesiology. 2004;100:1595–600. 7. Drapé JL, Feydy A, Guérini H, Campagna R. Infiltrations articulaires postérieures cervicales en vue d’une épidurale. J Radiol. 2009;90: 1414. 8. Brunner P, Amoretti N, Soares F, Brunner E, Cazaux E, Brocq O, et al. Approaches in injections for radicular pain: the transforaminal, epidural and transfacet approaches. Diagn Interv Imaging. 2012;93: 711–22.
9. Rosenkranz M. Anterior spinal artery syndrome following periradicular cervical nerve root therapy. J Neurol. 2004;251:229– 31. 10. Brouwers PJ, Kottink EJ, Simon MA, et al. A cervical anterior spinal artery syndrome after diagnostic blockade of the right C6-nerve root. Pain. 2001;91:397–9. 11. Recommandations AFSSAPS Risque de paraplégie/tétraplégie lié aux injections radioguidées de glucocorticoides au rachis lombaire ou cervical Mars 2011. www.afsapps.fr 12. Pech P, Daniels DL, Williams AL, Haughton VM. The cervical neural foramina: correlation of microtomy and CT anatomy. Radiology. 1985;155:143–6. 13. Huntoon MA. Anatomy of the cervical intervertebral foramina: vulnerable arteries and ischemic neurologic injuries after transforaminal epidural injections. Pain. 2005;117:104–11. 14. Demondion X, Lefebvre G, Fisch O, Vandenbussche L, Cepparo J, Balbi V. Radioanatomie des foramens intervertébraux cervicaux et lombaires (vaisseaux, variantes). J Radiol. 2012;93:733–40. 15. Krausé D, Drapé JL. Spinal infiltration: have you modified your practice? Diagn Interv Imaging. 2013;2014(11):1065–7. 16. Botwin KP, Gruber RD, Bouchlas CG, Torres-Ramos FM, Freeman TL, Slaten WK. Complications of fluoroscopically guided transforaminal lumbar epidural injections. Arch Phys Med Rehabil. 2000;81:1045–105. 17. Wybier M. Cervical spinal steroid injections under fluoroscopic guidance (what I am still doing, what I do no longer). Rev Rhum. 2008;75:755–62. 18. Scanlon GC, Moeller-Bertram T, Romanowsky SM, Wallace MS. Cervical transforaminal epidural steroid injections: more dangerous than we think? Spine Phila Pa 1976). 2007;32:1249–56 19. Lasbleiz J, Siegfried D, Chales G, Marin F, Sighetti M, Duvauferrier R. Evaluation of CT guided cervical epidural injections in patients with mechanical cervicobrachial neuralgia. J Radiol. 2008;89:317– 23. 20. Lee JY, Nassr A, Ponnappan RK. Epidural hematoma causing paraplegia after a fluroscopically guided cervical nerve root injection. J Bone Joint Surg Am. 2007;89:2037–9. 21. Stoll A, Sanchez M. Epidural hematoma after epidural block: implications for its use in pain management. Surg Neurol. 2002;57:235–40. 22. Engel A, King W, Macvicar J. The effectiveness and risks of fluoroscopically guided cervical transforaminal injections of steroids: A systematic review with comprehensive analysis of the published data. Pain Med. 2014;15:386–402. 23. McLean JP, Sigler JD, Plastaras CT, Garvan CW, Rittenberg JD. The rate of detection of intravascular injection in cervical transforaminal epidural steroid injections with and without digital subtraction angiography. PM R. 2009;1:636–42. 24. Fitzgerald RT, Bartynski WS, Collins HR. Vertebral artery position in the setting of cervical degenerative disease: implications for selective transforaminal epidural injections. Interv Neuroradiol. 2013;19:425– 31. 25. Jung H, Lim JA, Park KB, Hong SW, Kwak KH, Park JM. Computed tomography-guided cervical selective transforaminal epidural block for an anatomical variations of vertebral artery—a case report. Korean J Anesthesiol. 2013;65:468–72. 26. Beckworth WJ, Sood R, Katzer AF, Wu B. Anomalous location of the vertebral artery in relation to the neural foramen. Implications in transforaminal epidural steroid injections. Pain Med. 2013;14:119–25. 27. Jee H, Lee JH, Kim J, Park KD, Lee WY, Park Y. Ultrasound-guided selective nerve root block versus fluoroscopy-guided transforaminal block for the treatment of radicular pain in the lower cervical spine: a randomized, blinded, controlled study. Skeletal Radiol. 2013;42:69– 78. 28. Ludwig MA, Burns SP. Spinal CORD infarction following cervical transforaminal epidural injection: a case report. Spine. 2005;30:266– 8.
REVIEW ARTICLE
Cervical foraminal steroid injections under CT guidance: retrospective study of in situ contrast aspects in a serial of 248 cases Pierre Pottecher & Denis Krausé & Lucy Di Marco & Romaric Loffroy & Louis Estivalet & Romain Duhal & Xavier Demondion
Received: 6 November 2013 / Revised: 8 August 2014 / Accepted: 29 September 2014 / Published online: 16 October 2014 # ISS 2014
Abstract Objectives To describe all the CT findings after in situ contrast injection just before steroid injection and to recognize the abnormal aspects associated with intravascular contamination. Material and methods We retrospectively evaluated 248 cervical transforaminal steroid injections done at the university hospital in Dijon, France, in 2008–2012, to treat cervicobrachial neuralgia inadequately improved by optimal medical treatment for at least 3 weeks. Features describing the opacification patterns were recorded. Results Five main nonvascular opacification patterns were identified: clumps of contrast agent outside the foramen (16 %), a crab claw pattern surrounding the ganglion (13 %), a “French” circumflex accent pattern (15 %), reflux along the needle (7 %), and facet joint capsule opacification (22 %). Concerning the situations requiring a change in needle position, intravenous injection occurred in 26 % of the patients, with a crab claw pattern in half the cases and a clump pattern in half the cases. Intraarteriolar injection was noted in two patients. Conclusion CT after in situ contrast injection ensures proper needle positioning outside the blood vessels before steroid injection. Penetration of the needle tip into a vein is very common, whereas arteriolar puncture is extremely rare.
P. Pottecher (*) : D. Krausé : L. Di Marco : R. Loffroy : L. Estivalet Département de Radiologie et Imagerie Diagnostique et interventionnelle, CHU Dijon Bocage Central, Dijon, France e-mail: [email protected] R. Duhal : X. Demondion Service de radiologie musculosquelettique, CCIAL, laboratoire d’anatomie, faculté de médecine de Lille, Hôpital Roger Salengro, CHRU de Lille, Lille, France
Keywords Steroid infiltrations . Cervical radiculopathy . Contrast in situ control . Intervertebral foramen
Introduction Cervicobrachial neuralgia (CBN) in young individuals is a common condition usually ascribable to impingement on a nerve root of a soft disc herniation, most often at the entrance of or within the foramen [1]. The pain is due not only to nerve root compression by the disc, but also to local inflammation with release of enzymes and proinflammatory mediators. Severe pain present even at night and responsible for disability is common. Persistent pain despite optimal treatment for at least 3–6 weeks is a well-accepted indication for a local corticosteroid injection [2–4]. To be effective, the injection must place the corticosteroid in contact with the nerve root at the point of disc impingement, which is the site of proinflammatory mediator release [5]. The most widely used approaches are transforaminal, transfacet (which is followed by diffusion through the synovial joint), and posterior interlaminar (Fig. 1) [6–8]. Numerous cases of severe neurological complications of cervical corticosteroid injections have been reported [9–11]. These complications include spinal cord and brain infarction responsible for tetraplegia, paraplegia, or death. In 2011, the number of reported devastating neurological complications of cervical corticosteroid injections was about 14 in France and over 60 worldwide. The primary objective of this article is to establish, by a radioanatomical correlation, a precise description of all CT findings with contrast in situ before a transforaminal steroid injection for CBN. The population study was composed by 248 patients addressed to the Hospital of Dijon (Burgundy, France) between 2008 to 2012 for a CBN resistant to a medical treatment.
2
Fig. 1 Axial CT slice: C4–C5 foraminal herniation. Options for corticosteroid injection include the (a) transforaminal, (b) posterior articular, and (c) interlaminar approaches
Anatomy of the cervical foramina Boundaries Each cervical foramen is located between the pedicles of two adjoining vertebrae and has a 45° forward orientation parallel to the transverse processes. The foramen is bordered from cranial to caudal by the body of the suprajacent vertebra, intervertebral disc, and uncus of the suprajacent vertebra. The facet joint forms the posterior border. The foramen is also bordered in part by the facet joint synovial capsule, which extends as it recesses upward and forward toward the foramen and downward and backward toward the spinous process and ligamentum flavum (Figs. 2, 3, 4 and 5).
Skeletal Radiol (2015) 44:1–8
the uncinate process, and of the dorsal root ganglion, which often lies within a concavity of the superior articular process [12]. The nerve root exits the foramen at the immediately suprajacent intervertebral space (e.g., the C6 root exits at the C5–C6 space). The foramen also contains a radicular or radiculomedullary arteriole and the foraminal venous plexus surrounding the nerve root. The density of this plexus ensures good opacification of the epidural space and intervertebral foramen during contrast-enhanced CT or magnetic resonance imaging. The radiculomedullary arteries are fed by the subclavian artery via the vertebral artery or the ascending or deep cervical artery. In the most common configuration, about three radiculomedullary arteries arise from the vertebral artery at C3, another from the deep cervical artery at C6, and yet another from the ascending cervical artery at C8. Variants are common, however, and the configuration is asymmetric in most individuals [13]. Each radiculomedullary artery varies in diameter from 0.2 to 2 mm [14]. The radiculomedullary artery may course along either the anterior or the posterolateral part of the foramen. Consequently, the risk of accidental puncture is not predictable [13].
CT-guided transforaminal injection technique Availability of a team of skilled technicians is crucial to the safety and efficacy of transforaminal injections (Figs. 6–8).
Content
Study population
The cervical foramina contains the same nervous and vascular structures as the thoracolumbar foramina but with a smaller proportion of fat. The nervous structures are located in the posteroinferior part of the foramen and consist of the anterior root, located immediately posterior to the posterior margin of
The study population consisted of 248 patients (99 women, 149 men) recruited at the Hospital of Dijon (Burgundy, France) for a CBN resistant to a medical treatment from 4 to 6 weeks. All the patients had been treated by a general practice physician, rheumatologist or neurologist. The decision for the
Fig. 2 Cadaver study: cervical foramen in the oblique sagittal plane (a) and axial plane (b). Posterior roots (PRs) and anterior roots (ARs) in the lower part of the posterior foramen, with radicular arteries in contact with
the roots before (solid arrow). The joint cavity is delineated by the broken arrows. VB, vertebral body; D, intervertebral disc; Ped, pedicle; Sup art, superior articular process; Inf art, inferior articular process
Skeletal Radiol (2015) 44:1–8
3
Fig. 3 Cadaver study: axial slice through the left C7–T1 foramen. Connection of the joint capsule (broken arrows) with the foramen. VB, vertebral body; L, lamina; K1, anterior arc of the left first rib
transforaminal injection was made after a CT scan or MRI of the cervical spine demonstrated the existence of a hernia according with the topography of the radiculalgia. Concerning the preparation of the patient, a careful interview is mandatory to detect contraindications, such as a history of allergic reaction to contrast agents or anesthetics, and to evaluate the risk:benefit ratio. The use of medications that affect hemostasis, such as antiplatelet drugs, should be noted; clotting tests should be performed (prothrombin time, activated partial thromboplastin time, and international normalized ratio). The patient should receive detailed information about the procedure and its expected benefits and risks, then asked to provide consent. After the procedure, the patient is monitored closely in the hospital for 1 h. Course of action and equipment before the infiltration Installation of the patient The patient is supine with the head in neutral rotation and traction on the wrists to ensure fixed extension of the cervical spine and to improve the visibility of the C6–C7 and C7–T1 foramina.
Fig. 4 Cadaver study: upper (a) and lower (b) axial slices through C4– C5 showing a dense foraminal venous network. The epidural space is comparatively full as shown in Fig. 2b where the epidural network is dissected. Comparison with an axial CT slice (c) after contrast injection. The epidural and foraminal veins (broken arrows) form a dense venous network. VB, vertebral body; ART, articular process; SC, spinal cord; DRG, dorsal root ganglion
A thin-slice helical (1.25-mm) CT image centered on the level to be injected is acquired to visualize the disc-root impingement (Fig. 6) and to determine the needle trajectory toward the foramen that will avoid the vascular structures (jugular vein, carotid artery, and external jugular vein). A technician marks the skin. A local anesthetic is injected subcutaneously using a 21-G needle, which is the optimal caliber for obtaining a straight trajectory through the soft tissues and good anesthetic distribution at a distance from the foramen.
This diameter is considered optimal by several authors. The needle is inserted on the side of, and anterolateral to, the impingement, then advanced under CT guidance with 1.25mm slices until the tip reaches the foramen exit, in contact with the anterolateral edge of the articular process, far behind the vertebral artery. The ideal site is behind the ganglion, against the articular process.
The needle and its position
Contrast agent
A 22-G needle (diameter, about 400 μ) is used to minimize the risk of accidentally cannulating a vessel near the foramen.
With a 22-G needle, 1–1,2 ml is then injected. We use a nonionic, low-osmolality, iodinated contrast agent such
4
Skeletal Radiol (2015) 44:1–8
Fig. 5 Cadaver study of the cervical spine: radiculomedullary artery (arrow) included in the spinal anterior arterial axis. VB, vertebral body; DRG, dorsal root ganglion; SC, spinal cord
as iopamidol (Iopamiron 300, authorized for intrathecal use, Bracco Imaging, Courcouronnes, France). This fundamental step ensures good opacification of the distal foramen containing the ganglion and often the joint capsule. Vascular opacification by the contrast agent indicates faulty needle position within a blood vessel. The contrast agent may diffuse to the external and/or internal foramen or remain outside the foramen, where it penetrates the usual diffusion spaces (Fig. 7). Fig. 7 Axial CT slices showing diffusion at the needle tip (a) and foramen exit (b)
Results: in situ contrast and radiological-anatomical correlations
Normal findings
Study population and level of transforaminal injection
Clump pattern of diffusion: 16 % of cases
The descriptive statistics of the population are summarized in Table 1 and the levels involved in Table 2.
Close to the needle tip and articular process, the contrast agent formed polycyclic clumps, with moderate diffusion to the exit of the foramen.
Fig. 6 Axial CT slice with contrast injection showing a left C5–C6 disc herniation
Fig. 8 Axial CT slice showing a clump diffusion pattern of the contrast agent, with diffusion to the transverse foramen and foramen exit
Skeletal Radiol (2015) 44:1–8
5
Table 1 Descriptive statistics of the study population. The mean of age, visual analog scale (VAS) pain, and duration of cervicobrachial neuralgia (CBN) are presented with the standard deviation in parentheses Population (n=248) Age (years) Visual analog scale pain Duration of CBN (days)
54.5 (15.3) 5.8 (1.2) 41 (13.4)
Mean and standard deviation in parentheses
Diffusion around the ganglion: 28 % of cases (Figs. 9 and 10) Diffusion predominated at the foramen exit. The ganglion footprint resembled a crab’s claw (Fig. 9) or a circumflex accent (Fig. 10). Reflux along the needle: 7 % of cases (Fig. 11) Near the needle tip, retrograde contrast agent diffusion mimicked venous opacification. However, the symmetrical pattern of distribution on either side of the needle suggested reflux rather than venous opacification (Fig. 11). Other radiological and anatomical findings Joint capsule opacification: 22 % of cases (Fig. 12) When positioning the needle at the posterosuperior part of the foramen exit, the facet joint capsule may be contacted or perforated. The contrast agent may therefore diffuse up to the spinous process and/or contralaterally, mimicking venous opacification. This capsule forms recesses upward and forward toward the intervertebral foramen and downward and backward to the spinous process and ligamentum flavum (Fig. 12). However, corticosteroid injection can be performed safely [7].
Fig. 9 Axial CT slice showing a crab claw pattern of diffusion produced by posteroanterior molding of the contrast agent around the dorsal root ganglion at the foramen exit
or contralateral intraductal and/or foraminal veins. Venous contamination was consistently combined with a clump or crab claw pattern. The CT slices must be extended to the levels above and below the injection to look for distal contamination of the spinal muscle veins (Fig. 13). Arteriolar contamination: 0.8 % of cases Of the 248 procedures, 2 (0.8 %) resulted in arterial opacification. The immediate diagnosis is the absence of contrast at the contact of the distal end of the needle, which can make us think of an incorrect injection of anesthesia instead of contrast (Figs. 14 and 15)
Findings requiring needle repositioning Intravenous injection: 26 % of cases (Fig. 13) Among 248 cervical transforaminal injections, 64 (26 %) resulted in foraminal vein opacification manifesting as abnormal persistence of the contrast agent within the ipsilateral
Table 2 Percentage of the cervical spine levels involved in cervicobrachial neuralgia (CBN) by hernia
Levels involved
Percentage
C4–C5 C5–C6 C6–C7 C7–T1
3% 27 % 65 % 5%
Fig. 10 Axial CT slice showing a circumflex accent pattern of diffusion (∧) produced by posteroanterior molding of the contrast agent around the dorsal root ganglion at the foramen exit
6
Skeletal Radiol (2015) 44:1–8
Fig. 11 Axial CT slice showing retrograde diffusion along the needle mimicking intravenous injection. The symmetrical distribution on both sides of the needle suggests diffusion along the needle. Note the concomitant opacification of the joint capsule
Fig. 13 Axial CT slices showing venous contamination after injection of 1 ml of contrast agent. In contact with the needle, posterior epidural spread of contrast agent opacification associated with the paraspinal veins (B) and contralateral foraminal veins (black arrow Fig. 12C)
Discussion
Fig. 12 Axial CT slices showing joint capsule opacification with contrast agent diffusion along the articular recess (b) extending upward and forward toward the intervertebral foramen and downward and backward toward the lamina (c) and ligamentum flavum
CBN generates a huge burden of suffering, work disability, and costs to society. Optimal treatment is therefore crucial. When CBN due to disc impingement on a nerve root persists despite optimal medical treatment for at least 3 weeks, transforaminal corticosteroid injections have been proven effective [2–4]. In France, 52,000 intraspinal corticosteroid injections were performed each year in 2009/2010 [15], usually to treat sciatica, femoral neuralgia, or CBN. In our experience over a 12-year period, transforaminal corticosteroid injection was satisfaying in 75 % of patients before 1 month (visual analog scale pain score less than 2). Reports of devastating neurological complications after transforaminal corticosteroid injections have generated considerable reluctance among interventional radiologists to use this method at the cervical or lumbar spine [9–11, 16, 17]. A survey among US physicians collected 30 cases of brain and/or spinal cord infarct with 13 deaths [18]. However, we contend that
Skeletal Radiol (2015) 44:1–8
7
Fig. 14 Axial CT slices showing arteriolar opacification. In contact with the tip of the needle (a), despite the injection of 0.8–1 ml of contrast agent, no diffusion is visible, indicating a strong epidural arteriolar washout effect (b)
transforaminal injection is safe provided CT guidance and a number of precautions are used. The transforaminal approach is well suited to the treatment of CBN due to disc impingement on a nerve root, as the corticosteroid is delivered at the site of release of proinflammatory mediators, which play a major role in pain generation. The contrast medium and therefore the corticosteroid diffuse around the ganglion [19]. Other effective approaches include transfacet injection with diffusion along the synovial membrane extensions to the impingement site [7]. The interlaminar approach is more painful and has been reported to induce local hematoma formation [20, 21]. Fluoroscopic guidance with subtraction angiography may provide better arteriolar contrast and similar efficacy compared to CT, but requires considerable experience on the part of the interventional radiologist [22]. Fluoroscopy ensures immediate detection of faulty needle position in the radiculomedullary artery or, more often, foraminal vein. In a 2-year retrospective study of 177 cervical transforaminal injections, adding digital subtraction angiography to fluoroscopic guidance doubled the proportion of patients with detected intravascular injection [23], which occurred in 33 % of patients and was consistently intravenous. CT guidance shows needle progression to the foramen exit, thereby allowing the operator to avoid the vessels (common carotid artery, internal and external jugular veins, and vertebral artery). Accidental puncture of an artery can be fatal. Many variants of vertebral artery anatomy exist, making CT guidance invaluable [24–26]. An interesting alternative is Doppler ultrasound,
Fig. 15 Axial CT slices showing arteriolar opacification after injection of 1–1.5 ml of contrast agent. A small amount of contrast agent is visible at the foramen exit. Arteriolar washout explains why the observed amount of contrast is smaller than the injected amount: (a) and (b). The volume of contrast agent distribution increased after needle repositioning (c), indicating that the injection could be performed safely
which visualizes the vessels located outside the foramen, especially in case of posterior distribution (as described in anatomy [13]), thus avoiding any vascular cannulation [27]. Using a 21-G needle for soft tissue anesthesia ensures better and faster diffusion all around the foramen, 4–5 cm under the skin, at the level of the disc herniation, compared to thinner needles. After anesthesia, insertion of a 22-G needle under CT guidance is painless. The needle tip is positioned at the distal foramen, just behind the ganglion, against the articular process. The 22-G diameter is the most widely used in France. Thinner needles (24–25 G) are deformable and do not eliminate the risk of intravascular injection with spinal cord infarction [28]. The contrast agent diffusion pattern showed no intravascular contamination in 73 % of our patients. Venous contamination occurred in 26 % of patients, half of whom had a crab claw and half a clump pattern of diffusion. In the remaining two patients, intraarterial injection occurred, with little or no visible contrast agent, indicating an immediate washout effect. Injection into a vein or artery requires needle repositioning under CT guidance.
8
Skeletal Radiol (2015) 44:1–8
The mechanisms responsible for ischemic neurological complications are well documented. Abnormalities of the arterioles may consist of dissection, spasm, or thrombosis by steroid emulsion aggregates [17]. Blood reflux into the needle is never observed, a fact that makes CT guidance particularly important to detect intravascular injection. We obtained six 1.25-mm slices, including a slice that clearly showed the needle tip at the foramen. This well-standardized procedure carried out by a highly trained team requires a CT acquisition time of only 10 min.
Conclusion Ambulatory transforaminal corticosteroid injection under CT guidance with routine contrast agent injection immediately before the corticosteroid injection plays a key role in the treatment of CBN due to disc impingement. Thorough understanding of contrast agent diffusion is essential to ensure the recognition of faulty needle position within a vein or arteriole. In every day practice, venous contaminations are very frequent; on the other hand, an arteriolar injection is exceptional. Considerable experience and regular practice on the part of the interventional radiologist are needed to ensure patient safety.
References 1. Radhakhrishnan K, Litchy WJ, O’Fallon WM, Kurland LT. Epidemiology of cervical radiculopathy: a population-based study from Rochester, Minnesota, 1976 through 1990. Brain. 1994;117: 325–35. 2. Bush K, Hillier S. Outcome of cervical radiculopathy treated with periradicular/epidural corticosteroid injections: a prospective study with independent clinical review. Eur Spine J. 1996;5:319–25. 3. Van Zundert J, Harney D, Joosten EA, Durieux ME, Patijn J, Prins MH, et al. The role of the dorsal root ganglion in cervical radicular pain: diagnosis, pathophysiology, and rationale for treatment. Reg Anesth Pain Med. 2006;31:152–67. 4. Cyteval C, Thomas E, Decoux E, Sarrabere MP, Cottin A, Blotman F, et al. Cervical radiculopathy: Open study on percutaneous periradicular foraminal steroid infiltration performed under CT control in 30 patients. AJNR. 2004;25:441–5. 5. Anderberg L, Annertz M, Persson L, Brandt L, Saveland H. Transforaminal steroid injections for the treatment of cervical radiculopathy: a prospective and randomised study. Eur Spine J. 2007;16:321–8. 6. Rathmell JP, Aprill C, Bogduk N. Cervical transforaminal injection of steroids. Anesthesiology. 2004;100:1595–600. 7. Drapé JL, Feydy A, Guérini H, Campagna R. Infiltrations articulaires postérieures cervicales en vue d’une épidurale. J Radiol. 2009;90: 1414. 8. Brunner P, Amoretti N, Soares F, Brunner E, Cazaux E, Brocq O, et al. Approaches in injections for radicular pain: the transforaminal, epidural and transfacet approaches. Diagn Interv Imaging. 2012;93: 711–22.
9. Rosenkranz M. Anterior spinal artery syndrome following periradicular cervical nerve root therapy. J Neurol. 2004;251:229– 31. 10. Brouwers PJ, Kottink EJ, Simon MA, et al. A cervical anterior spinal artery syndrome after diagnostic blockade of the right C6-nerve root. Pain. 2001;91:397–9. 11. Recommandations AFSSAPS Risque de paraplégie/tétraplégie lié aux injections radioguidées de glucocorticoides au rachis lombaire ou cervical Mars 2011. www.afsapps.fr 12. Pech P, Daniels DL, Williams AL, Haughton VM. The cervical neural foramina: correlation of microtomy and CT anatomy. Radiology. 1985;155:143–6. 13. Huntoon MA. Anatomy of the cervical intervertebral foramina: vulnerable arteries and ischemic neurologic injuries after transforaminal epidural injections. Pain. 2005;117:104–11. 14. Demondion X, Lefebvre G, Fisch O, Vandenbussche L, Cepparo J, Balbi V. Radioanatomie des foramens intervertébraux cervicaux et lombaires (vaisseaux, variantes). J Radiol. 2012;93:733–40. 15. Krausé D, Drapé JL. Spinal infiltration: have you modified your practice? Diagn Interv Imaging. 2013;2014(11):1065–7. 16. Botwin KP, Gruber RD, Bouchlas CG, Torres-Ramos FM, Freeman TL, Slaten WK. Complications of fluoroscopically guided transforaminal lumbar epidural injections. Arch Phys Med Rehabil. 2000;81:1045–105. 17. Wybier M. Cervical spinal steroid injections under fluoroscopic guidance (what I am still doing, what I do no longer). Rev Rhum. 2008;75:755–62. 18. Scanlon GC, Moeller-Bertram T, Romanowsky SM, Wallace MS. Cervical transforaminal epidural steroid injections: more dangerous than we think? Spine Phila Pa 1976). 2007;32:1249–56 19. Lasbleiz J, Siegfried D, Chales G, Marin F, Sighetti M, Duvauferrier R. Evaluation of CT guided cervical epidural injections in patients with mechanical cervicobrachial neuralgia. J Radiol. 2008;89:317– 23. 20. Lee JY, Nassr A, Ponnappan RK. Epidural hematoma causing paraplegia after a fluroscopically guided cervical nerve root injection. J Bone Joint Surg Am. 2007;89:2037–9. 21. Stoll A, Sanchez M. Epidural hematoma after epidural block: implications for its use in pain management. Surg Neurol. 2002;57:235–40. 22. Engel A, King W, Macvicar J. The effectiveness and risks of fluoroscopically guided cervical transforaminal injections of steroids: A systematic review with comprehensive analysis of the published data. Pain Med. 2014;15:386–402. 23. McLean JP, Sigler JD, Plastaras CT, Garvan CW, Rittenberg JD. The rate of detection of intravascular injection in cervical transforaminal epidural steroid injections with and without digital subtraction angiography. PM R. 2009;1:636–42. 24. Fitzgerald RT, Bartynski WS, Collins HR. Vertebral artery position in the setting of cervical degenerative disease: implications for selective transforaminal epidural injections. Interv Neuroradiol. 2013;19:425– 31. 25. Jung H, Lim JA, Park KB, Hong SW, Kwak KH, Park JM. Computed tomography-guided cervical selective transforaminal epidural block for an anatomical variations of vertebral artery—a case report. Korean J Anesthesiol. 2013;65:468–72. 26. Beckworth WJ, Sood R, Katzer AF, Wu B. Anomalous location of the vertebral artery in relation to the neural foramen. Implications in transforaminal epidural steroid injections. Pain Med. 2013;14:119–25. 27. Jee H, Lee JH, Kim J, Park KD, Lee WY, Park Y. Ultrasound-guided selective nerve root block versus fluoroscopy-guided transforaminal block for the treatment of radicular pain in the lower cervical spine: a randomized, blinded, controlled study. Skeletal Radiol. 2013;42:69– 78. 28. Ludwig MA, Burns SP. Spinal CORD infarction following cervical transforaminal epidural injection: a case report. Spine. 2005;30:266– 8.
