The Rate of Detection of Intravascular Injection in Cervical Transforaminal Epidural Steroid Injections With and Without Digital Subtraction Angiography

Original Research

The Rate of Detection of Intravascular Injection in Cervical Transforaminal Epidural Steroid Injections With and Without Digital Subtraction Angiography James P. McLean, MD,† James D. Sigler, MD, Christopher T. Plastaras, MD, Cynthia Wilson Garvan, PhD, Joshua D. Rittenberg, MD Objective: To determine whether digital subtraction angiography (DSA) combined with real-time fluoroscopic imaging improves the detection rate of intravascular injection during cervical transforaminal epidural steroid injections (CTFESIs). Design: Retrospective analysis. Setting: Outpatient surgery center. Participants: A total of 134 subjects with cervical radicular pain who had CTFESIs performed by a single physician between June 9, 2004, and April 23, 2007. Interventions: One hundred seventy-seven CTFESIs performed at one or more cervical spinal levels either unilaterally or bilaterally. Procedures performed before September 12, 2005, used fluoroscopic guidance with contrast injection and live imaging to identify intravascular injection. All procedures performed after September 12, 2005, also included DSA. Main Outcome Measures: Intravascular injection detected during CTFESIs with and without DSA. Results: Intravascular injection was detected in 17.9% of CTFESIs performed without DSA. By adding DSA technology to the real-time fluoroscopic imaging procedure, the detection of vascular injection nearly doubled to 32.8%, which was statistically significant (P ⫽ .0471). Conclusions: The use of DSA improves the detection rate of intravascular injection during CTFESIs.

INTRODUCTION The primary indication for cervical epidural steroid injections (CESIs) is radicular pain secondary to cervical disk pathology or spinal stenosis [1-7]. The symptomatic relief of radicular pain is attributed to suppression of inflammation secondary to mechanical or chemical nerve root irritation [8]. Significant reduction in extremity pain after CESIs has been documented in prospective studies [3,4]. There are 2 major approaches for delivery of steroid into the cervical epidural space: interlaminar and transforaminal. The interlaminar approach has been used more often, based on retrospective surveys [9]. However, the transforaminal approach theoretically allows a higher concentration of medication to be injected closer to the site of the suspected pathology [10,11], potentially resulting in a more robust therapeutic response. Interlaminar and transforaminal CESIs carry the risk of serious injury to the central nervous system from such factors as incorrect needle placement and intravascular injection of particulate corticosteroid preparations. Potential complications include nerve root trauma, unintentional dural puncture, unintended spinal anesthesia with respiratory and hemodynamic compromise, and vertebral artery injury [12,13]. Central nervous system injury may occur as a result of direct injection of material into the spinal cord, infarction of the brain or spinal cord after injection of particulate steroids into an artery, or compression of the spinal cord from an epidural hematoma or abscess [1,3,9,14,15]. The complication most pertinent to this study is the inadvertent injection of particulate corticosteroids into a radicular artery. A growing body of evidence supports an embolic PM&R

636

1934-1482/09/$36.00 Printed in U.S.A.

J.P.M. Spine Center, Kansas University Medical Center; Kansas City, KS Disclosure: nothing to disclose J.D.S. Department of Physical Medicine & Rehabilitation, Kansas University Medical Center, Kansas City, KS Disclosure: nothing to disclose C.T.P. Department of Rehabilitation Medicine, University of Pennsylvania, Philadelphia, PA Disclosure: nothing to disclose C.W.G. College of Education, University of Florida; Gainesville, FL Disclosure: nothing to disclose J.D.R. Department of Physical Medicine and Rehabilitation, Northwestern University Feinberg School of Medicine; Rehabilitation Institute of Chicago, 1030 N. Clark Street, Suite 500, Chicago, IL 60610. Address correspondence to: J.D.R.; e-mail: [email protected] Disclosure: nothing to disclose Disclosure Key can be found on the Table of Contents and at www.pmrjournal.org Submitted for publication October 30, 2008; accepted March 29, 2009. †

Deceased.

© 2009 by the American Academy of Physical Medicine and Rehabilitation Vol. 1, 636-642, July 2009 DOI: 10.1016/j.pmrj.2009.03.017

PM&R

Vol. 1, Iss. 7, 2009

637

Figure 1. Depiction of a cervical transforaminal epidural steroid injection (CTFESI) and adjacent vascular structures. (From reference 22, with permission from the Mayo Foundation.)

mechanism of injury, whereby inadvertent intra-arterial injection of particulate corticosteroid causes a distal infarct [8,14-19]. In addition to intra-arterial injection, other potential mechanisms of infarction include vertebral artery perforation causing dissection/thrombosis and needle-induced vasospasm [16,20] (Figure 1). Considering these various complications that can arise from intravascular injection of corticosteroids, proper technique is critical to avoid injection of medication into venous and arterial structures. The use of fluoroscopy may diminish the risks associated with CTFESIs by confirming proper needle placement. Under fluoroscopic guidance, the needle must always remain in contact with the posterior wall of the foramen. This typically avoids contact with the spinal nerve, its roots, and accompanying vessels [11]. However, a wide range of anatomical variations may exist in the origin and location of radicular arteries [21]. Variable anastomoses can exist between vertebral and cervical arteries, which make it possible to introduce steroid particles into the vertebral circulation via the cervical arteries. Branches from ascending and deep cervical arteries have been found in the posterior aspect of the intervertebral foramen and may be vulnerable to injection or injury during cervical transforaminal epidural steroid injections (CTFESIs) [22]. Intravenous injection, seen commonly, is associated with much less dire outcomes. However, it may lead to loss of efficacy of a therapeutic procedure or a falsely negative diagnostic block.

Standard injection techniques, such as aspiration before injection to confirm the absence of blood in the needle hub, are not reliable. Furman et al demonstrated in 2 separate studies that aspiration before injection is unreliable for detecting intravascular needle placement [10,23]. The studies found that the use of observed blood in the needle hub to predict intravascular injections was only 44.7% sensitive for lumbosacral and 45.9% sensitive for CTFESIs. Injection of contrast medium is critically important to the safe execution of CTFESIs. Originally, this was used to indicate correct location of the injection and to exclude intrathecal injection, whereas it now also serves to identify inadvertent intravascular injection. This must be done under realtime imaging because spot films taken after the injection may not show contrast medium that has been rapidly cleared [11,24,25]. Digital subtraction angiography (DSA) is a relatively new addition to fluoroscopically guided, contrast-enhanced CTFESIs. Digital subtraction, the commonly accepted standard for documenting angiography and venography, may be a useful tool for documentation of or avoidance of inadvertent intravascular injection [24]. Digital subtraction enhances visualization of contrast distribution during injection by subtracting pixel values of an initial scout image preinjection mask from subsequent images after contrast injection [24]. The pixel subtraction algorithm is based on image movement, so pixel values of stationary background structures are

638

McLean et al

RATE OF DETECTION OF INTRAVASCULAR INJECTION IN CTFESI

detection rate of intravascular injection of contrast during CTFESIs. No previous study has investigated whether DSA lowers the threshold for the detection of intravascular injection. This study prospectively collected data for CTFESIs to compare the detection rates of intravascular injection of contrast with and without DSA. Additional data were collected to determine whether the rate of intravascular injection is affected by the following factors: age, gender, level of injection, side of injection, number of injections, and history of prior intravascular injection in the patient. These factors were analyzed for intravascular injection both with and without DSA to determine whether DSA is more effective for certain patient populations or anatomical injection sites.

METHODS Approval for this study was obtained through the Institutional Review Committee at Rehabilitation Institute of

Figure 2. Comparison of contrast flow with and without digital subtraction angiography (DSA). (a) Example of contrast flow documented with standard fluoroscopy without DSA. (b) Example of contrast flow documented with DSA.

subtracted, essentially leaving only the image of contrast dye that has been injected in real time (Figures 2 and 3). During injection with DSA, it is important that the patient remain still because his or her movement is detected as a change in image and impairs subtraction. Use of real-time DSA may also allow improved visualization of contrast material at lower volumes than with conventional fluoroscopy [26]. Disadvantages of real-time DSA include additional radiation exposure to the patient and medical personnel, cost of new or upgraded fluoroscopic equipment, and previous lack of clinical evidence supporting its efficacy in CTFESIs. The purpose of this study is to determine whether DSA combined with real-time fluoroscopic guidance improves the

Figure 3. Comparison of intravascular flow pattern with and without digital subtraction angiography (DSA). (a) Example of venous flow documented using standard fluoroscopy without DSA. (b) Example of venous flow with DSA.

PM&R

Vol. 1, Iss. 7, 2009

Table 1. Sample description Variable Gender Male Female Age (y)* Total number of injections received 1 2 3 4 5

All Patients 59 (44.0%) 75 (56.0%) 46.8 (10.9) 97 (72.4%) 33 (24.6%) 3 (2.2%) 0 (0%) 1 (0.8%)

*Values given as mean (SD).

Chicago/Northwestern University. All subjects in the study were patients who had CTFESIs by the same physician at an outpatient surgery center between June 9, 2004, and April 23, 2007. All data for the CTFESIs were collected prospectively and input into a database. During the data collection period, the method of detecting intravascular injection was modified to include DSA. Results of this study are a retrospective analysis of intravascular injection with and without DSA, which was performed using the data collected over the entire timeframe of the study. All patients in the study were examined and treated for cervical radicular pain. Patients with axial pain were excluded from the study. No other specific exclusion criteria were used. The total number of patients in the study was 134. All procedures performed before September 12, 2005, used real-time fluoroscopic guidance with contrast enhancement to identify potential intravascular injection. All procedures performed after September 12, 2005, used real-time fluoroscopic guidance with contrast enhancement and DSA. All injections were performed by J.D.R. This physician was a fellowship-trained physiatrist with 5 years of experience at the beginning of the study period in performing CTFESIs. All injections were done as per the Practice Guidelines of the International Spine Intervention Society [27]. The patient was in a supine position and an oblique approach was used to localize the needle in the posteroinferior aspect of the intervertebral foramen. An anteroposterior view was used to verify the depth of the tip of the needle. The target position on the anteroposterior view was the sagittal midline of the articular pillar. The needle was not advanced beyond a vertical line connecting the uncinate processes. Anteroposterior and oblique radiographs were taken to document final needle position before injection of contrast. For patients in the non-DSA group, contrast was injected under real-time fluoroscopy without DSA. If intravenous injection was detected, the needle was repositioned. If intra-arterial injection was identified, the procedure was aborted. If a contrast pattern was seen outlining the exiting spinal nerve and flowing medially into the epidural space without intravascular location, anteroposterior and oblique radiographs were saved to document the contrast flow pattern. A test dose of preservative-free lidocaine without epinephrine was then adminis-

639

tered. After 1.5 minutes, a gross assessment of sensory and motor function in the lower and upper limbs was performed. If there was no change in sensory and motor function, and the patient was not exhibiting signs of local anesthetic toxicity of the central nervous system, 1 mL dexamethasone (10 mg/mL) was injected. The procedure was the same for the DSA group except that DSA was used when contrast was being administered. After the procedure, the physician’s documentation included whether or not an intravascular flow pattern was detected, with a distinction made between arterial and venous injection. Data collected for each patient in this study include: age, gender, level of injection, sides of injection, number of injections, and whether intravascular injection had been previously identified in the patient. To conduct valid statistical testing (ie, to ensure independence of observations), only data from the first injection of each subject were used in inferential procedures. Data were analyzed using SAS software version 9.1 (Cary, NC). ␹2 and Wilcoxon rank sum tests were used to compare groups on categorical and numerical data, respectively. A P value of .05 was considered statistically significant.

RESULTS This study included a total of 177 CTESIs performed on 134 patients. The patient population comprised 59 males and 75 females. The age range of patients was 25 to 77 years of age, with an average age of 46.8 years (SD ⫽ 10.9). As shown in Table 1, 72% of the patients had only 1 injection, 25% had 2 injections, and 3% had 3 or more injections. Three patients received bilateral injections, with the rest of the injections being unilateral. Slightly more injections were performed on the right side. Injections were performed at various cervical levels as shown in Table 2. Most of the injections were at level C5-6 (41%) or C6-7 (51%). Table 2 describes the side, level, and injection method (with and without DSA) for first injection and all injections to demonstrate the potential for extrapolating results to multiple injections in a patient. Fifty per-

Table 2. Description of injections Variable Side Left Right Bilateral Level C3-4 C4-5 C5-6 C6-7 C7-T1 Method Without DSA With DSA

First Injection

All Injections

55 (41.0%) 79 (59.0%) 3 (1.5%)

75 (42.4%) 102 (57.6%) 5 (2.8%)

0 (0%) 9 (6.7%) 55 (41.0%) 68 (50.8%) 2 (1.5%)

1 (0.6%) 12 (6.8%) 72 (40.7%) 90 (50.9%) 2 (1.1%)

67 (50.0%) 67 (50.0%)

93 (52.5%) 84 (47.5%)

DSA ⫽ digital subtraction angiography. Bilateral injections were included in both left and right.

640

McLean et al

RATE OF DETECTION OF INTRAVASCULAR INJECTION IN CTFESI

Table 3. Comparison of factors between the with DSA and without DSA groups (first injections) Without DSA n ⴝ 67

Variable Gender Male Female Age (y)* Mean (SD) Side of injection Left Right Level of injection C4-5 C5-6 C6-7 Venous injection Not detected Detected Arterial injection Not detected Detected

27 (40.3%) 40 (59.7%) 45.7 (11.3)

With DSA n ⴝ 67 32 (47.8%) 35 (52.2%) 48.0 (10.3)

P Value*

.3842

Number of Subjects

Subjects with Vascular Injection

67 67

12 (17.9%) 22 (32.8%)

Without DSA With DSA

P Value

DSA ⫽ digital subtraction angiography.

30 (44.8%) 37 (55.2%)

.3799

5 (7.5%) 27 (40.3%) 35 (52.2%)

4 (6.0%) 28 (41.8%) 33 (49.3%)

.7300 .8606 .7297

55 (82.1%) 12 (17.9%)

45 (67.2%) 22 (32.8%)

.0471

67 (100%) 0 (0%)

67 (100%) 0 (0%)

cent of first injections and 48% of all injections in this study used DSA. Table 3 compares subjects whose first injections used traditional fluoroscopic guidance with those whose first injections incorporated DSA technology to identify vascular injection. Groups were not found to differ on gender, age, or both side and level of injections. All intravascular injections were found to have intravenous flow patterns. Intra-arterial injection was not detected in any of the subjects. There was a significant difference in detection of intravascular injection associated with the use of DSA. CTFESIs performed without DSA detected intravascular injection in 17.9% of the injections. By adding DSA technology to the procedure, the detection of intravascular injection nearly doubled, with 32.8% of the injections showing intravascular penetration (Table 4). Thus, intravascular injection was detected in 32.8% of injections when DSA was used compared with only 17.9% of

70 60 50 40 Injections 30 20 10 0

Without DSA

.0471

.1346

25 (37.3%) 42 (62.7%)

*Value given as mean (SD).

Table 4. Effect of DSA on the detection of vascular injection

With DSA # w/ Vascular uptake # w/o uptake

Figure 4. Percentage of intravascular injection detected with and without digital subtraction angiography (DSA).

PM&R

Vol. 1, Iss. 7, 2009

Table 5. Comparison of factors between the groups with and without vascular injections (first injection)

Variable Gender Male Female Age (y)* Side Left Right Level C4-5 C5-6 C6-7 C7-T1 Method Without DSA With DSA

Without Vascular Injection n ⴝ 100

With Vascular Injection n ⴝ 34

47 (47.0%) 53 (53.0%) 46.8 (11.1)

12 (35.3%) 22 (64.7%) 47.0 (10.2)

41 (41.0%) 59 (59.0%)

14 (41.2%) 20 (58.8%)

.9856

8 (8.0%) 38 (38.0%) 53 (53.0%) 1 (1.0%)

1 (2.9%) 17 (50.0%) 15 (44.1%) 1 (2.9%)

.3087 .2191 .3708 .5584

55 (55.0%) 45 (45.0%)

12 (35.3%) 22 (64.7%)

.0471

P Value*

.2349

Abbreviation: DSA, digital subtraction angiography. *Values given as mean (SD).

injections without DSA (Figure 4). This increase in detection of intravascular injection during first injection was statistically significant, with a P value of .0471. Increased detection of intravascular injection using DSA technology was similar when considering all injections (no DSA⫽ 18.3%; with DSA ⫽ 33.3%), first injection (no DSA ⫽ 17.9%; DSA ⫽ 32.8%), and second or subsequent injection (no DSA ⫽ 15.0%; with DSA ⫽ 30.8%). Other variables were analyzed to determine their effect on intravascular injection, including gender, age, side of injection, and cervical level of injection. Apart from DSA, no other factors were found to have a significant association with the presence of intravascular injection during injections. Table 5 compares factors between the with-vascular-injection and without-vascular-injection groups.

DISCUSSION The results of this study clearly show that use of DSA is associated with an increased detection rate of intravascular injection during CTFESIs. Because other variables, such as gender, age, location, and level of injection did not have statistical association with intravascular injection, it is reasonable to conclude that the addition of DSA to real-time fluoroscopy can increase the detection rate of intravascular injection in CTFESIs. The improved visualization of contrast with DSA allows physicians to observe intravascular injection that may have been missed using conventional fluoroscopy. The physician can then reposition the needle and confirm proper placement prior to injection of medications. Several limitations of this study are apparent. Subjects were limited to those seen by a single physician in 1 medical center; thus, extrapolation to the general population may not be accurate. The power of the study is only moderate because of a relatively low number of subjects involved. Temporal

641

relationships may be a confounding factor, because all injections without DSA preceded those with DSA. Therefore, factors other than DSA may have influenced the increase in intravascular injection detection. For example, it is possible that the physician performing the CTFESIs improved his procedural effectiveness by gaining experience throughout the study period. There are additional patient characteristics that were not measured that may have affected the incidence of intravascular injection. For example, the rate of foraminal stenosis, prior history of surgery, or other anatomic factors may not have been equivalent between the 2 groups. An important limitation of this study is that no incidences of intra-arterial injection were detected during the study period. The most likely reason for this is that intra-arterial injection occurs rarely and our sample size was too small. However, it is also possible that the clinician failed to recognize an intra-arterial flow pattern if it did occur. Theoretically, improved detection of intravenous injection implies improved detection of intra-arterial injection. Improved ability to detect intravascular injection should decrease the risk of CTFESIs by lowering the rate of complications related to inadvertent injection of corticosteroid into radicular arteries. However, additional clinical studies need to be performed to statistically validate this hypothesis. A direct correlation between rate of complications and detection rate of intravascular injection could not be made using the results of our study. Additionally, detection rate of intra-arterial injection could not be measured with this study. Nevertheless, it is reasonable to use DSA technology with the goal of reducing the rate of complications during CTFESIs.

CONCLUSION The risk of complications cannot be completely eliminated in any procedure. The use of proper technique can minimize the risks associated with CTFESIs. Based on the present findings, DSA along with real-time fluoroscopy can improve the ability to detect intravascular injection and thus has the potential to reduce the risk of complications associated with inadvertent intravascular injection.

REFERENCES 1. 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;5:319-325. 2. Castagnera L, Maurette P, Pointillart V, Vital JM, Erny P, Senegas J. Long-term results of cervical epidural steroid injection with and without morphine in chronic cervical radicular pain. Pain 1994;58:239243. 3. Cicala RS, Westbrook L, Angel JJ. Side effects and complications of cervical epidural steroid injections. J Pain Symptom Manage 1989;4:64-66. 4. Rowlingson JC, Kirschenbaum LP. Epidural analgesic techniques in the management of cervical pain. Anesth Analg 1986;65:938-942. 5. Slipman CW, Lipetz JS, Jackson HB, Rogers DP, Vresilovic EJ. Therapeutic selective nerve root block in the nonsurgical treatment of atraumatic cervical spondylotic radicular pain: a retrospective analysis with independent clinical review. Arch Phys Med Rehabil 2000;81:741-746.

642

McLean et al

RATE OF DETECTION OF INTRAVASCULAR INJECTION IN CTFESI

6. Stav A, Ovadia L, Sternberg A, Kaadan M, Weksler N. Cervical epidural steroid injection for cervicobrachialgia. Acta Anaesthesiol Scand 1993; 37:562-566. 7. Stojanovic MP, Vu TN, Caneris O, Slezak J, Cohen SP, Sang CN. The role of fluoroscopy in cervical epidural steroid injections: an analysis of contrast dispersal patterns. Spine 2002;27:509-514. 8. Marshall LL, Trethewie ER, Curtain CC. Chemical radiculitis. A clinical, physiological and immunological study. Clin Orthop Relat Res 1977; 129:61-67. 9. Derby R, Lee SH, Kim BJ, Chen Y, Seo KS. Complications following cervical epidural steroid injections by expert interventionalists in 2003. Pain Physician 2004;7:445-449. 10. Furman MB, Giovanniello MT, O’Brien EM. Incidence of intravascular penetration in transforaminal cervical epidural steroid injections. Spine 2003;28:21-25. 11. Rathmell JP, Aprill C, Bogduk N. Cervical transforaminal injection of steroids. Anesthesiology 2004;100:1595-1600. 12. McMillan MR, Crumpton C. Cortical blindness and neurologic injury complicating cervical transforaminal injection for cervical radiculopathy. Anesthesiology 2003;99:509-511. 13. 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;81:1045-1050. 14. Baker R, Dreyfuss P, Mercer S, Bogduk N. Cervical transforaminal injection of corticosteroids into a radicular artery: a possible mechanism for spinal cord injury. Pain 2003;103:211-215. 15. Brouwers PJ, Kottink EJ, Simon MA, Prevo RL. A cervical anterior spinal artery syndrome after diagnostic blockade of the right C6-nerve root. Pain 2001;91:397-399. 16. Scanlon GC, Moeller-Bertram T, Romanowsky SM, Wallace MS. Cervical transforaminal epidural steroid injections: more dangerous than we think? Spine 2007;32:1249-1256.

17. Beckman WA, Mendez RJ, Paine GF, Mazzilli MA. Cerebellar herniation after cervical transforaminal epidural injection. Reg Anesth Pain Med 2006;31:282-285. 18. Ludwig MA, Burns SP. Spinal cord infarction following cervical transforaminal epidural injection: a case report. Spine 2005;30: E266-E268. 19. Muro K, O’Shaughnessy B, Ganju A. Infarction of the cervical spinal cord following multilevel transforaminal epidural steroid injection: case report and review of the literature. J Spinal Cord Med 2007;30: 385-388. 20. Rozin L, Rozin R, Koehler SA, et al. Death during transforaminal epidural steroid nerve root block (C7) due to perforation of the left vertebral artery. Am J Forensic Med Pathol 2003;24:351-355. 21. Hoeft MA, Rathmell JP, Monsey RD, Fonda BJ. Cervical transforaminal injection and the radicular artery: variation in anatomical location within the cervical intervertebral foramina. Reg Anesth Pain Med 2006;31:270-274. 22. Huntoon MA. Anatomy of the cervical intervertebral foramina: vulnerable arteries and ischemic neurologic injuries after transforaminal epidural injections. Pain 2005;117:104-11. 23. Furman MB, O’Brien EM, Zgleszewski TM. Incidence of intravascular penetration in transforaminal lumbosacral epidural steroid injections. Spine 2000;25:2628-2632. 24. Jasper JF. Role of digital subtraction fluoroscopic imaging in detecting intravascular injections. Pain Physician 2003;6:369-372. 25. Smuck M, Fuller BJ, Yoder B, Huerta J. Incidence of simultaneous epidural and vascular injection during lumbosacral transforaminal epidural injections. Spine J 2007;7:79-82. 26. Sherry RG, Anderson RE. Real-time digitally subtracted fluoroscopy for cervical myelography. Radiology 1984;151:243-244. 27. Bogduk N, ed. ISIS Practice Guidelines for Spinal Diagnostic and Treatment Procedures. International Spine Intervention Society, San Francisco, CA; 2004.