Fig. 33.1
Needle enter right L5/S1 zygapophyseal joint
Fig. 33.2
L5/S1 zygapophyseal joint injection with intra-articular contrast
33.2.5 Medial Branch Blocks
Fluoroscopic imaging, starting with an AP view and transitioning to more oblique views as needed, is used to visualize the target (Fig. 33.3). The target is the space between the superior border of the transverse process and the mamillo-accessory notch at the junction of the superior articular process and the transverse process. The superficial skin can be anesthetized with a small amount of local anesthetic, followed by the use of a spinal needle, which is then introduced and directed toward the target until a bony stop is reached. Care must be taken to avoid needle placement that is too far posterior along the bulk of the superior articular process (SAP) itself or placement too far lateral along the transverse process itself. AP and lateral views must be used to confirm correct positioning (Fig. 33.4). A small amount of contrast is injected while visualizing under live fluoroscopy to assure proper placement and the absence of venous flow (Fig. 33.5). If vascular flow is encountered, the needle should be repositioned and contrast reinjected to assure that the flow has abated. Next a small amount of concentrated anesthetic (0.3–0.5 mL) is injected. Larger volumes should be avoided so as to minimize unintended spread of anesthetic. Targeting of the L5–S1 facet joint is slightly different as the L5 dorsal ramus crosses the sacra ala as opposed to the transverse process. The target for this injection is 5 mm below the superior junction between the sacral ala and the S1 SAP. When MBBs are performed according to the guidelines, no significant complications have been reported [23]. Technical complications such as thecal puncture only occur if the procedural guidelines are not followed and the needle is grossly misplaced. Full technical description of the procedure is available from ISIS [23].
Fig. 33.3
Oblique view with needles over L4 medial branch and L5 dorsal ramus to anesthetize the L5/S1 facet joint
Fig. 33.4
Depth view for diagnostic medial branch blocks
Fig. 33.5
AP view for diagnostic medial branch block with contrast placed to assure no aberrant vascular flow
The purpose of an MBB is to precisely deliver local anesthetic to anesthetize a medial branch and subsequently evaluate the patient’s response. Therefore, careful documentation of the patient’s pre- and post-procedure pain level is required. In general the patient must be experiencing his or her typical pain prior to the procedure. After the injection pain level should be recorded in a pain diary in a systematic fashion, at intervals of at least 30–60 min for the remainder of the day. Pain levels should also be recorded at less frequent time intervals over the course of the next few days. The utility in the test truly lies in the information obtained from a properly conducted block and not in the execution of the block itself [23]. Without an accurately maintained pain diary, a perfectly performed MBB loses all potential benefit. Maintaining an accurate pain diary in real time limits the potential for recall bias. The 2004 ISIS guidelines recommend a formal post-MBB assessment by an assessor other than the performing physician; however, this is not practical in most practices. Ideally relief based on a numeric or visual analog scale, patients should also track pain medication requirements and ability or inability to perform tasks that are typically pain limited.
33.2.6 Intra-articular Z-Joint Injections
In terms of diagnostic utility, intra-articular z-joint injections have fallen out of favor. Birkenmaier found that using pericapsular injection of anesthetic to predict response to denervation procedures had poorer outcomes than when MBBs were used [27]. Three randomized controlled trials that evaluated the use of intra-articular injections of local anesthetic as a prognostic tool have been published, two of which were equivocal [28, 29] and one that was definitively negative [30]. In addition to improved diagnostic utility, medial branch blocks are also used in place of intra-articular anesthetic blocks because they are easier to perform, safer to perform, more easily subjected to controls, and have proven therapeutic utility in that they predict response to radiofrequency neurotomy [23].
There is limited data on the therapeutic utility of intra-articular z-joint injections as well. In uncontrolled studies the success rate for lumbar intra-articular steroid injections has varied widely, with reported success rates between 18 and 63 % [31, 32]. Only two RCTs evaluating the efficacy of lumbar z-joint intra-articular steroid injections have been published. In one of the studies, 8 mL of injectate was used and compared intra-articular anesthetic, extra-articular anesthetic, or intra-articular saline and found no difference at 3 months between the groups [33]. However, since the z-joint can only hold 1–2 mL, the validity of this study is in question. The other study evaluated patients who had immediate pain relief from intra-articular anesthetic injection and randomized them to receive either intra-articular corticosteroid or intra-articular saline [34]. There was no difference between the groups at 1 and 3 months; however, at 6 months the intra-articular steroid group was statistically more likely to have improved. The authors attributed this to concurrent therapies received by certain patients in the study, even though the results at 6 months remained statistically significant even after assuming that all patients with concurrent treatment did not improve.
In 1996, Dolan reported that positive findings on single-photon emission computed tomography (SPECT) scans may correlate with greater levels of pain relief for up to 3 months following intra-articular z-joint steroid injections [35]. Similarly, Ahmad and Ackerman published a study of patients with low back pain and isolated z-joint inflammation seen on SPECT imaging and compared outcomes at 12 weeks between those that received intra-articular steroid plus anesthetic and those that received medial branch perineural steroid and anesthetic. The number of patients with greater than 50 % pain relief at 12 weeks was 61 % in the intra-articular group, which was statistically significantly greater than in the perineural group (26 %) [36]. A study by Pneumaticos also found that the patients with SPECT positive imaging had significantly better outcomes at 1 month and 3 months with intra-articular injections compared to patients that did not have facet joint abnormalities on SPECT [37]. However there was no difference between the groups at 6 months. Another recent study that compared the effect of intra-articular z-joint steroid injection with intramuscular steroid injection for presumed z-joint-mediated low back pain found slight increase in outcomes with respect to physical function and reduction in NSAID use in the intra-articular group [38]. However the effect size of these results was small, there was no control arm to the study, and all subjects in the intra-articular group received bilateral injections at L3–L4, L4–L5, and L5–S1 as opposed to selecting the joints thought to be most likely involved. Others have reported that intra-articular steroids are no better than sham injections [21]. Given the mixed literature, it is likely that there is a subset of the population that does significantly respond to intra-articular steroids, that being those with joint inflammation seen on SPECT.
As opposed to steroids, hyaluronic acid is another injectate that has been studied for the treatment of z-joint-mediated low back pain. Hyaluronic acid has been used in other joints with osteoarthritis to theoretically improve the viscoelastic properties of the defective synovial fluid and thus decrease pain [39]. Fuchs et al. compared the efficacy of intra-articular hyaluronic acid versus intra-articular steroid with 6-month follow-up and found that both groups demonstrated significant improvement in pain scores and function over the 6-month period [40]. There were no significant differences between the two groups in any of the outcome measures except for faster onset of symptom relief within the steroid group. Unfortunately the study did not include a control arm, which given the limited data on the efficacy of intra-articular steroids limits the utility of such a non-inferiority design study. However, at the very least there is theoretical benefit of hyaluronic acid over steroids given the reduced side effect profile of hyaluronic acid compared to steroids.
Given the presence of a better diagnostic test and limited evidence on the therapeutic utility of therapeutic intra-articular facet injections, defined indications for facet joint injections have not been published.
33.2.7 Medial Branch Blocks
Medial branch blocks have a well-established diagnostic utility and are thus possibly indicated in a patient that has chronic or subacute low back pain thought to be mediated by structures innervated by the medial branches, in most cases the z-joints, and confirmation of this will alter management. Not all patients with low back pain require MBBs. Examples of patients not needing an MBB would be those with low-level pain that does not result in functional limitations, when the pain is thought to be due to other structures, when pain is still in the acute stages, or when the first step in treatment is conservative therapy regardless MBBs are not indicated [23]. It is also important to have ruled out more serious possible causes of low back pain such as infection and tumors, at least through history and examination if not through other diagnostic tests, prior to proceeding with MBBs as a diagnostic test.
To fully address the use of MBBs requires a more in-depth discussion about the false positives of MBBs, false negatives of MBBs, what constitutes a positive finding with a single MBB, the number of medial branch blocks required, and ultimately establishing cutoffs for what constitutes a true positive result from MBBs.
While a full discussion of radiofrequency ablation (RFA) is covered elsewhere in this textbook, the practical use of what level of response to MBB constitutes a positive MBB is directly tied to the ability of MBB to predict response to the efficacy of radiofrequency neurotomy. As the criteria for what constitutes a positive MBB become more stringent, the likelihood of successful RFA also increases. In general, when nearly complete relief of symptoms is required from MBB in order to proceed with RFA, outcome data on the efficacy of RFA is very strong [41, 42], whereas relaxed criteria for percent relief of symptoms from MBB have led to less profound outcomes in the subsequent RFA that were performed [28, 30, 43]. When looking at pain relief as a general topic however, studies have found that as little as 30 % pain relief is clinically meaningful in chronic pain conditions [44]. Other studies indicate that 50 % in pain reduction improves a patient’s quality of life [45]. Elsewhere in the pain literature, 50 % pain relief is the most commonly used dichotomous outcome measure [46]. So while the academic benefit of using more stringent criteria for what constitutes a positive MBB and how this then lends itself to having more robust positive findings when investigating RFA, at least some consideration must also be made to the proponents of less stringent criteria for what constitutes a positive MBB even if this results in less robust pain relief in the less ideal patients for RFA in light of the fact of what other literatures suggest that is clinically meaningful to the patient. Regardless of these theoretical differences, on the most basic level, pain relief from lumbar medial branch blocks implies that the patient’s pain is mediated by the anesthetized nerve, and if there is no pain relief, the target nerve is deemed not to be contributing to the patient’s pain. Nonetheless, additional safeguards must be in place to minimize the chance of false positives. Consideration must also be made about possibilities of false negatives.
33.2.7.1 False Positives
Depending on how a positive response is defined, rates of false-positive results from single MBBs in the lumbar spine have been reported to be between 17 and 41 % [5, 47, 48]. Many argued that such high false-positive rate when only single MBB is used has rendered them invalid [48–50]. To mitigate false-positive responses, both dual and triple blocks have been proposed. Dual anesthetic blocks include performing the procedure twice with two different anesthetic of differing duration and evaluating if the patient’s pain relief is concordant with the duration of each anesthetic. In order to prevent false positives due to the patients’ potential bias toward desiring a positive response, the patient should not be instructed as to the expected duration of relief if pain relief may be achieved. A triple block includes a dual block in addition to performing the procedure with saline as a placebo. Triple blocks are not commonly performed for a variety of reasons, including but not limited to cost, efficiency, and the ethics of performing invasive procedures with placebo medications. ISIS recommends dual comparative blocks as a viable alternative to placebo-controlled blocks [23]. True positive finding with comparative local anesthetic blocks is when a patient reports duration of pain relief that corresponds to the expected duration of action of the anesthetic used, in essence that the patient experiences pain relief with both blocks but longer relief when the longer-acting anesthetic is used [51].
A study that looked at cervical facet pain found that while the specificity of dual comparative MBBs was 88 % specific, it was only 54 % sensitive [52]. Accordingly, a 12 % false-positive rate is much lower than when single blocks are used and is well within acceptable levels when compared to other diagnostic tests. However, the low sensitivity also suggests that false negatives must also be considered when performing MBBs. And while the risk of a false-positive test is that a patient may undergo an un-needed radiofrequency neurotomy, false negatives are potentially worse than false positives as they withhold potentially beneficial treatment from a patient.
33.2.7.2 False Negatives
False negatives will theoretically increase as more stringent criteria are applied for what constitutes a positive MBB. There are many reasons a false-negative response may occur, including but not limited to concurrent pain generators, inaccurate technique, excessive or inadequate use of superficial local anesthetic, and vascular uptake of injectate [53]. Procedural-related pain may preclude the patient from properly identifying that their typical pain has been alleviated. Fortunately, there is evidence that a single-needle approach to block multiple medial branches reduces procedural-related pain compared to conventional multiple needle site entry techniques [54]. There may also only be partial pain relief in the instance that there are multiple pain generators. Pain can concurrently be originating from the contralateral side, additional segmental levels, or an alternate painful structure. The patient may have difficulty in recognizing relief of their z-joint-mediated pain if these other possible pain generators continue to cause pain, resulting in a false-negative response. Some argue that it is unreasonable to expect complete pain relief from MBB as z-joint degeneration and pain rarely occurs in isolation of other potential sources of LBP. Radiologic studies demonstrate that significant z-joint degeneration never occurs in the absent of disk degeneration [55]. Even more, loss of disk height can accelerate or precipitate z-joint degeneration [56]. Conversely other studies have suggested that multiple pain generators simultaneously contributing to pain occur in less than 5 % of low back pain patients [57]. Regardless, the argument can be made that if only partial pain relief is achieved by a MBB, simultaneous anesthetization of the other painful structure should still enable total pain relief and confirmation of pain generators [21]. Unfortunately, the literature and techniques with respect to anesthetic injections for diagnosis of other sources of low back pain are not established enough to put this theoretical argument into practice.
If the procedure is performed without the patient in their usual state of pain, eliciting enough pain relief to be significantly noticeable and thus be considered a positive test may be difficult or even impossible. Vascular uptake of the anesthetic during MBB can also contribute to false negatives, which has been documented in 3.5 % of lumbar MBBs [58]. The risk of vascular uptake can be mitigated by use of real-time contrast injection [23]. It has also been hypothesized that there are potentially anomalous innervations of the facet joint other than the medial branch blocks. In the original Kaplan study that found that anesthetizing the medial branches blocked painful responses from facet joint capsular distention in eight of the nine patients, the other patient has been postulated to have anomalous innervations [3]. If the target nerve is missed and not bathed in anesthetic, a false negative may also occur. One study that compared cervical (not lumbar) low-volume MBBs (0.25 cc vs 0.5 cc) found that the target nerve was missed 7 % of the time in both cases, but that unintentional aberrant spread occurred twice as often (38 % compared to 16 %) in the high-volume group [59]. Alternatively, another study that performed CT scans after lumbar MBB to evaluate location of contrast found that the target nerve was bathed in injectate in all 120 injections [25]. Regardless all MBBs should be done with low volumes (0.25–0.3 cc) with live fluoroscopy to assure the lack of venous uptake.
33.2.8 Facet Cyst Injections
Facet joints can develop cysts that can be readily seen on MRI imaging and if in the anterior aspect of the joint may result in radicular pain. If symptomatic, symptoms are often that of radicular pain or stenosis, likely because of proximity of the cyst to the spinal nerve. Rupture of the cysts can be attempted. One study achieved 72 % success defined by avoidance of surgery and improvement in symptoms. In the same study there was a 37.5 % recurrence rate, but 45 % of recurrences responded to repeat cyst rupture [60]. Another study found that 46 % of the time surgery could be avoided [61]. Technique of cyst rupture involves anesthetizing the spinal nerve using a transforaminal approach, followed by high-volume facet intra-articular injection with anesthetic and contrast.
33.3 Epidural Steroid Injections
By definition epidural steroid injections target the spinal nerves. They are thus indicated for radicular pain. The most common causes of radicular pain are intervertebral disk herniations and spinal stenosis. It has been shown that the degree of nerve root compression does not correlate to the level of pain [62–64]. It has also been shown that pure mechanical compression of the spinal nerves produces paresthesia and motor weakness but not pain [65]. As such, it reasons that radicular pain must be caused by additional factors in addition to, if not exclusive of, root compression. Multiple studies have shown that inflammation is an essential component to the painful symptoms experienced in patients with radiculopathy [13, 66–68]. Inflammatory mediators such as phospholipase A2, prostaglandin E2, leukotrienes, nitric oxide, immunoglobulins, and cytokines such as interleukin-6 and tumor necrosis factor alpha are all involved in the inflammatory component of radiculopathy [69–71]. Various biochemical inflammatory markers must be present in order for the dorsal root ganglion to generate painful discharges [72]. Moreover, many of these inflammatory mediators such as phospholipase 2 have been found within the nucleus pulposus itself and are found in high concentrations along with inflammatory cells such as macrophages at sites of disk herniation [68, 73, 74]. Histopathological findings consistent with inflammation have also been found in nerve root specimens taken from decompression surgery [68, 75]. Corticosteroids inhibit phospholipase 2 and leukocyte aggregation at sites of inflammation; prevent degranulation of granulocytes, mast cells, and macrophages and transmission of nociceptive C-fibers; and stabilize ectopic discharge of neuronal membranes [70, 76, 77]. As such local administration of corticosteroid can theoretically result in symptom relief [78].
33.3.1 Contraindications
Contraindications include bleeding diathesis, local infection at the injection site, systemic infection, cardiovascular instability, uncontrolled diabetes, uncontrolled glaucoma, cauda equina syndrome, pregnancy, and allergy to local anesthetic or steroid medication.
33.3.1.1 Types of Epidural Steroid Injections
Caudal and Interlaminar Epidural Steroid Injection
Caudal epidural steroid injections (CESI) involves administration of steroid into the epidural space through the sacral hiatus. In 1957 Cyriax reported the first use of caudal epidural steroid injection for pain relief [79]. The first interlaminar epidural injection was described by Pages in 1921 [80]. Interlaminar injections involve delivery of medication into the posterior epidural space between the dura anteriorly and ligamentum flavum posteriorly. Both of these techniques have limited literature on their efficacy and have thus somewhat fallen out of favor when compared to transforaminal epidural steroid injection. In light of this a full description of their techniques is not warranted in this chapter.
Transforaminal Epidural Steroid Injection
The first reported TFESI was described in 1952 by Robecchi and Capra [81]. TFESI delivers steroid to the epidural space in close proximity to the affected nerve root. Compared to other approaches, it is more targeted in bathing the affected spinal nerve root close to its dorsal root ganglion and theoretically maximizes the therapeutic effect. Derby first postulated that a transforaminal approach theoretically is superior because it can provide a high concentration of injectate directly to the posterior annulus and the ventral epidural space [82].
General Technique
The patient is positioned prone and sterilely draped. The skin is anesthetized with local anesthetic. Using the AP view the transforaminal space that is the target is identified by first squaring off the inferior end plate of the target level. Then using an oblique approach the spinal needle is advanced under the pedicle, making sure not to pass medial to the 6 o’clock position of the pedicle when viewed from a true anterior-posterior (AP) view (Figs. 33.6 and 33.7). The needle should be deep to the lamina on the lateral view and in the safe triangle on the AP view. The safe triangle is composed of the base of the corresponding pedicle, lateral border of the vertebral body, and lateral border of the exiting nerve root. Needle placement must be confirmed by both AP and lateral views. The AP view is used to confirm that the needle has not been placed too medially which increases risk of dural puncture (Fig. 33.7). The lateral view is used to confirm depth (Fig. 33.8). Contrast medium is then injected using live fluoroscopy and evaluated for transforaminal epidurogram spread and assuring that no intravascular or intrathecal spread has occurred (Fig. 33.9). After confirming appropriate contrast, spread small-volume injectate, usually consisting of 1–2 cc of anesthetic followed by 1–2 cc of steroid [23, 78].
Fig. 33.6
(a) Transforaminal oblique view (without needle). (b) Transforaminal oblique view (with needle)
Fig. 33.7
TF AP view
Fig. 33.8
TF lateral view
Fig. 33.9
TF AP post-contrast injection
Complications and Side Effects
Overall incidence of complications for TFESIs ranges between 5.5 and 9.6 % [83, 84]. Rates are higher for multilevel injections versus single-level injections [83]. The most common side effects include injection site pain, vasovagal reaction (3.5 %), increased radicular pain, light-headedness, increased pain caused by direct trauma to the spinal nerve, nausea, non-positional headache, vomiting, facial flushing, and elevated blood pressure [78, 84–86]. Anesthetic medications, contrast, and steroids can all cause allergic reaction. Steroids can also cause myopathy, fluid retention, hypertension, mood abnormalities, menstrual irregularities, hyperglycemia, and iatrogenic Cushing syndrome [87–90].
Bleeding is another possible complication. Patients with coagulopathy or on anticoagulation medications are at increased risk of bleeding complications [91–93]. Very rarely does bleeding result in epidural hematoma and compression of the spinal cord and spinal nerves, reported to occur 1 in 150,000 injections [94, 95]. Surgical evacuation is warranted in the rare event it does occur.
Infection is another known risk of all spine injections. Infection risk includes epidural abscess, diskitis, osteomyelitis, and meningitis [96–100]. Given proximity to the pelvic and abdominal cavity, gram-negative infections are more likely. If the needle is advanced too far ventral or lateral, there is also a risk of abdominal cavity puncture leading to infection [101]. Serious infection is extremely rare, occurring only 0.001–0.1 % of the time. If present, serious infections require surgical intervention 70 % of the time and often do not fully recover [102]. Fifty-three percent of the time, an infection presents as worsening pain, most often around 7 days postinjection [102].
Dural puncture is another potential complication of TFESI [103]. Dural puncture can result in positional headache. If not identified, as evidenced by flashback of CSF or by recognition of poor positioning on fluoroscopy, intrathecal administration of anesthetic can cause cauda equina, arachnoiditis, or meningitis [101]. Intradiscal injection can occur during TFESI [104–106]. The primary concern with intradiscal injection is diskitis, and prophylactic antibiotics are usually given if this complication is encountered.
Rate of intravascular injection has been reported as high as 11.2 % for all lumbar TFESIs and as high as 21.3 % for S1 TFESIs [107] (Fig. 33.10). The risk of intravascular injection is double in patients over 50 years old [108]. Most of these injections are venous in nature. The real concern is intra-arterial injection into the spinal radiculomedullary arteries. The artery of Adamkiewicz, which supplies the anterior third of the spinal cord, is often implicated in intra-arterial injections due to its location in the neural foramen. There is variability in the anatomy of the artery of Adamkiewicz as it is located on the left 63 % of the time and is between the T9 and L2 level only 85 % of the time [109–111]. Intra-arterial injection with a particulate corticosteroid during TFESI has been reported to cause spinal infarction and subsequent paraplegia [109–111].
Fig. 33.10
Venous flow on S1 TF ESI
Smuck reported that intermittent fluoroscopy only identifies 57 % of vascular injections as opposed to continuous fluoroscopy [112]. Even more concerning is that confirmation of epidural spread does not rule out concomitant vascular uptake [107, 112]. Digital subtraction angiography can also be used in adjunct with continuous fluoroscopy to further enhance the ability to detect intravascular flow [101, 113].
An anesthetic test dose can also be used to reduce the risk of intra-arterial injection. And anesthetic challenge dose involves administration of anesthetic such as lidocaine after needle position has been confirmed with contrast injection and evaluating for patient response. Reported symptoms such as tinnitus, metallic taste in mouth, headache, dizziness, and sensorimotor changes in either all four or bilateral lower extremities are suggested of intra-arterial infiltration. If positive the procedure should be terminated. Despite these safeguards, irreversible paraplegia has been reported even when continuous fluoroscopy, digital subtraction angiography, and anesthetic test dose have all been implemented [114]. In addition to potentially catastrophic events, intravascular uptake may also reduce the efficacy of TFESIs [108, 115]. Additionally the use of a non-particulate, preservative-free corticosteroid such as dexamethasone could also reduce the risks of inadvertent intra-arterial injection.
33.3.2 Dosing and Number of Injections
No standard dose of steroid exists for TFESI though in a recent comprehensive review of the literature MacVicar reported in most studies that investigated TFESI used either low (40 mg)-dose methylprednisolone or high (80 mg)-dose methylprednisolone, equivalent dosing of triamcinolone or betamethasone, and that less extensive use of dexamethasone has been found in the literature [116].
There is no current literature that specifically investigates the ideal number of injections to achieve maximal benefit. However, MacVicar pooled the number of injection data from all studies that reported categorical data on patients that achieved at least 50 % pain relief (totaling 9 studies with a total of 727 injection included) and revealed that 94 ± 2 % of patients with successful outcomes from TFESI did so with only 1 injection [116]. Of the 15 patients that had relief in a study by Ghahreman, only 5 required a second injection, and none required more than 2 [117]
33.3.3 Evidence-Based ESI for Radicular Pain Due To Disk Herniation
33.3.3.1 CESI Efficacy
The data on CESI for disk herniation is quite limited. Even more problematic is that the majority of available studies utilize blind CESI. Blind interlaminar and blind caudal approaches demonstrate a 30–40 % rate of missing the epidural space [2, 118]. Current standard of care dictates that fluoroscopy be used for such injections. This further minimized the usefulness of available literature. Also worth considering is that spread of injectate via CESI is at best up to L3–L4 and more likely only up to the L4–L5 level and that L4–L5 is the most cephalic level of pain generation that has been reported to be amenable to treatment with caudal injection [119–121].
The first evidence that CESI may be beneficial for radicular pain was published in 1971 [122]. In 1987 Matthews published results of a series of three blind CESI in patients and showed benefit in pain reduction at 3 months but not 1, 6, or 12 months [123]. It was not until Bush and Hiller published a randomized placebo-controlled study in 1991 that more significant evidence became available. They reported significant gains in mobility and quality of life at 4 weeks in the group that received CESI compared to placebo [124]. Unfortunately, the study did not differentiate between radicular pain due to stenosis and disk herniation and was limited by a very small sample size (n = 23). Moreover, the procedures were performed without fluoroscopy. At 1 year differences between the groups were no longer present, as the anesthetic-only control group demonstrated similar gains by that time [124].
Dincer et al. studied the efficacy of blind CESI compared to 1 month of NSAID therapy for radicular pain due to disk herniation in 64 patients and found that the CESI group had statistically greater improvement in VAS at 2 weeks, 4 weeks, and 12 weeks and in Oswestry scores at 2 and 4 weeks [125].
Another study was designed to evaluate if targeted placement of steroid using endoscopically placed steroid around the affected nerve root had greater effect compared to less targeted steroid placement via fluoroscopically guided CESI. It evaluated patients with radicular pain but excluded those with “chronic stenosis” as defined by symptoms of 18 months or longer. Both groups showed significant improvement in pain at 6 weeks, 3 months, and 6 months but no difference between the two groups [126]. While there was no control group, the results of the CESI group can be evaluated as a cohort independent of the endoscopic group, and the study can be used as evidence that CESI can produce favorable improvements in pain for up to 6 months in patients with radicular pain of less than 18 months.
There are no studies that directly evaluate the efficacy of fluoroscopically guided CESI for radicular pain due to disk herniation. At best, reviewing the available literature including the blind CESI and group subset analysis of other studies shows that there is evidence, while minimal, that fluoroscopically guided CESI may provide short-term pain relief for acute and subacute radicular pain.
33.3.3.2 Interlaminar Epidural Steroid Injection (ILESI) Efficacy
There is limited literature that supports the use of ILESI. In perhaps the best designed study aimed at evaluating ILESI, Carette published in the NEJM a randomized double-blind trial in 158 patients with radicular pain due to herniated nucleus pulposus that compared blind ILESI of methylprednisolone to interlaminar injections with normal saline and found that there was significant improvement in leg pain in the steroid group at 3 weeks and 6 weeks, but by 3 months there was no difference between the groups. All other parameters evaluated including ODI did not reveal any differences between the two groups at 3, 6, or 12 weeks. Only group mean data was evaluated; no categorical data was included in the study [127]. In 2003 Valat et al. published a randomized double-blind control trial comparing blind interlaminar saline to blind interlaminar steroid for radicular pain due to “presumed” disk herniation with 85 patients total. Primary outcome was whether or not patient deemed their symptoms “resolved” or “markedly improved” at day 20, whereas “worse” and “slight improvement” were deemed as failure. Use of NSAIDs or surgery was also considered failure. At day 20 there was no significant difference between the groups with an intention to treat analysis; however, once patients that were lost to follow-up or excluded (11 of the 85 subjects) there was a strong trend (p = 0.054) toward treatment “success” in the steroid group compared to the saline group. By day 35 the groups were found to have equal outcome [128]. In 2009 Parr reviewed the best available literature for interlaminar epidural injections for low back pain, including the two studies mentioned above, and concluded that the evidence is “limited for blind interlaminar epidurals in managing all types of pain except for short-term relief of pain secondary to disk herniation and radiculitis” [129]. Perhaps more importantly though, Parr noted that the evidence identified does not represent contemporary interventional pain management practices given that none of the studies identified utilized fluoroscopy. Strictly speaking, the evidence may not be extrapolated to fluoroscopically directed lumbar interlaminar epidural injections. Currently there continues to be a paucity of literature that is investigating, much less in support of, fluoroscopically guided interlaminar steroid injection. This is most likely in large part due to the major shift in clinical practice toward transforaminal epidural steroid injections.
33.3.3.3 TFESI Efficacy
Certainly the literature for caudal and interlaminar epidural steroid injections is limited and in many instances has not found these interventions to be more effective than sham treatments [127, 128, 130]. However, careful review of the available literature that specifically investigates transforaminal injections for radicular pain reveals significant and positive findings, most dramatically for herniated disk pathology. Unfortunately, some systematic reviews consider all types of epidural steroid injections as equivalent and have thus inappropriately dismissed the reported effectiveness of a TFESI [131]. Promising research evaluating the use and efficacy of TFESIs has been more abundant over the last 20 years. Some of the earlier studies demonstrated a significant surgical sparing effect of TFESI for herniated disks causing radicular pain. In 1997 Weiner and Fraser reported that 27/30 patients with severe lumbar radiculopathy had immediate pain relief after TFESI, and 22 of the 28 (79 %) patients available for longer-term follow-up had considerable and sustained relief [132]. In a randomized controlled double-blind study by Riew in 2000 that evaluated the surgical sparing effect of transforaminal anesthetic compared to transforaminal anesthetic plus steroid in 55 patients with lumbar radicular pain due to either canal stenosis or disk herniation that were scheduled for surgery, the group that received anesthetic plus steroid deferred surgical intervention 71 % of the time compared to only 33 % of the patients that received anesthetic alone [133]. This effect was maintained for 5 years [133]. The publication reported stratifying the data based on patients with stenosis and those with lumbar disk herniation but did not present the raw data nor did they comment on the efficacy of TFESI in preventing surgery based on pathology (stenosis vs disk herniation). However, they did report that in the group that avoided surgery, which was predominantly composed of patients that received both steroid and anesthetic, that the patients with stenosis had “significant relief of low back pain” (p < 0.008) and those with disk herniation trended toward “significant relief of low back pain” (p < 0.07) [133]. The surgical sparing effect of TFESI was also demonstrated by Wang in 2002 [134]. In a retrospective review of 69 patients with symptomatic herniated lumbar disks whom had failed conservative management with anti-inflammatories and physical therapy who were now requesting diskectomy, only 16 (23 %) eventually went on to having surgery after receiving between 1 and 6 TFESI with a mean follow-up of 1.5 years [134]. All three studies combined provide strong evidence that for patients with radicular pain due to herniated nucleus pulposus, TFESI is an effective means of preventing surgical intervention in a significant amount of patients.
Beyond a surgical sparing effect, the literature has also supported the use of TFESI as a means of providing symptomatic relief. 50 % pain relief has been established as what patients considered a “much improved” for pain in general as well as the minimal clinically important change for radicular pain [44, 135]. Subsequently, much of the literature on ESI has appropriately used this to define what is categorically considered to be a “positive response.” Another important consideration to make when evaluating studies with ordinal data such as VAS is that if mean data is used for statistical analysis of outcomes, the outcomes must be in a normal distribution. Otherwise the data should have the mode and interquartile range evaluated, not the mean. The lumping of responders and nonresponders can result in group mean scores that are worthless. Rarely are pain scores distributed in a normal bell-shaped distribution. This emphasizes that the need or importance of using categorical outcome, predefining what success is and who achieves success, is a vital step in evaluating these types of studies.
There are multiple studies that evaluated patient cohorts of various sizes that demonstrated significant pain relief in varying percentages of patients ranging from 38 to 75 % [136–140]. In 1998, Lutz reported a prospective case series of 69 patients in which “52 of 69 (75 %) patients had a successful long-term outcome, reporting at least a >50 % reduction between pre-injection and post-injection pain scores, as well as an ability to return to or near their previous levels of functioning after only 1.8 injections per patient” with a mean follow-up of 80 weeks [136]. More recently, a large retrospective study of 2,024 patients undergoing single lumbar TFESI for radicular pain either due to disk herniation or foraminal stenosis reported that 45.6 % had at least 50 % pain relief at 2 months. For patients with less than 3 months of pain, the success rate increased to 68.3 % at 2 months [137]. Unfortunately, data specifically on disk herniations or lumbar stenosis was not presented separately. In a study by Cyteval, when looking only at the subgroup of patients with radiculopathy solely due to disk herniation who had failed conservative therapy, 65 of 172 patients (38 %) had at least 50 % pain relief at 2 weeks, and 88 % of these had continued relief through 1 year [138]. Yet another study, by Narozny, found that 12 of 20 (60 %) patients with radicular pain due to disk herniation had at least 60 % pain relief at 6 months after TFESI [139]. Jeong reported on two different transforaminal approaches for lumbar radicular pain due to either central canal stenosis or disk herniation. Overall, at 4 weeks 99 of 122 (89 %) in the preganglionic approach group had good or excellent results (defined by at least 50 % reduction in pain) compared to 90 out of 127 (70 %) in the ganglionic group [140]. The difference between the two groups was no longer significant at 6 months. Overall analysis of the entire group as a single cohort of patients with radicular pain due to disk herniation that received TFESI is valuable in this case though. Overall there was still good or excellent response in 142 of the 222 subjects (64 %) for 6 months [140]. In the subset of patients with radicular pain due to disk herniation, 118 of 191 patients (62 %) had at least 50 % pain relief at 6-month follow-up [140].
As mentioned, dangerous pitfalls arise in randomized controlled trials when comparing mean pain scores from studies with non-normally distributed data as opposed to reporting predefined categorical data. A randomized controlled trial that evaluated the efficacy of transforaminal steroid versus transforaminal saline, first published in 2001, did not find benefit for TFESI at 3 or 6 months when analyzing mean group data [92]. However with subgroup analysis of the same data evaluating radicular pain due to contained herniations, the steroid group was found to have short-term benefit for radicular pain and disability [141]. At 1 year, TFESI was found to prevent progression to surgery within the same subgroup of disk herniations and when compared to the control group found a saving of $12,666 per responder on average [141]. Additional well-designed studies that consider predefined categorical outcomes have repeatedly demonstrated convincing data in support of the use of TFESI for disk herniation causing radicular pain. In 2002 Vad published a randomized (patient selected) controlled trial that compared the efficacy of TFESIs to trigger point injections for lumbosacral radiculopathy secondary to herniated nucleus pulposus [142]. Successful outcome was categorized as predefined improvement in all three categories of patient satisfaction, Roland Morris score, and VAS pain reduction. “After an average follow-up period of 1.4 years, the group receiving transforaminal epidural steroid injections had a success rate of 84 %, as compared with 48 % for the group receiving trigger point injections”[142]. The best designed study to date that investigated the efficacy of TFESI was by Ghahreman et al. in 2010. The study was a prospective, randomized study with five arms that compared TFESI to TF local anesthetic, to TF saline, to intramuscular steroid, and to intramuscular saline and used categorical outcomes to evaluate success. They found that a significantly greater proportion of patients in the TFESI group (54 %) achieved at least 50 % pain relief at 1 month compared to the other four arms (ranging between 7 and 21 %) [117]. Pain relief was “corroborated by significant improvements in function and disability and reduction in the use of other health care” in the TFESI group. Long-term relief at 1 year was also greater in the TFESI group (25 %) than the other four groups; however, the results did not reach statistical significance. The authors of the study argued that 25 % success rate at 1 year is “patently cost-effective” considering that the cost of the alternative (surgery) is much greater than the cost of a single injection. The study also found that using transforaminal saline as a control/placebo compared to TFESI, the number needed to treat to obtain at least 50 % pain relief at 4 weeks is only three [117]! Lastly, the design of the study also makes evident that both the medication (steroid) and route of delivery (transforaminal) combine to form a compound intervention that is unique and different from systemic steroids and from transforaminal administration of other compounds [117].
In 2013 MacVicar et al. published a thorough and comprehensive review of the literature with accompanying systematic analysis of all published data regarding TFESI [116]. After plotting the success rates of outcome studies, pragmatic studies, and explanatory studies, MacVicar et al. summarized it best in saying that with regard to TFESI for radicular pain due to disk herniation, the literature is “abundant” and of “higher quality” and reveals that “about 60 % of patients seem to achieve at least 50 % relief of pain at between 1 and 2 months but that only 40 % maintain this outcome for 12 months” [116]. Specifically, their conclusions included that TFESIs are effective (more so in patients with contained disk herniations, low-grade compression, and acute symptom duration) [140, 143–145], are statistically more than placebo effects [117, 133], reduce the burden of disease by improving function [117, 142, 146] and reducing need for surgery [117, 133, 134], and ultimately are cost-effective [141].
33.3.4 Predictors of Response to TFESI
In the prospective case series published by Lutz, in patients who had pain for less than 36 weeks, the success rate was nearly 80 % versus only 65 % in patients with symptoms that present longer for longer durations [136]. Similarly, in the study by Jeong, univariant analysis revealed that pain of less than 6 months had better therapeutic effect than those with greater than 6 months of symptom duration regardless of symptoms being due to stenosis or disk herniation [140]. In a large retrospective study of 2,024 patients undergoing single lumbar TFESI for radicular pain either due to disk herniation or foraminal stenosis, the proportion of responders was significantly higher when there was less than 3 months of pain that was present (odds ratio 2.42) [137]. In the study by Cyteval on TFESI for lumbar radicular pain due to either foraminal stenosis or disk herniation, patients that had at least 75 % pain relief had a mean duration of symptoms of 3 months compared to the group that had less than 25 % pain relief who had a mean duration of symptoms of 8 months at 2-week follow-up [138]. The review article by MacVicar also pooled data from three studies that provided data on duration of radicular symptoms, in addition to other inclusion criteria, and its effect on success rates of TFESI. They concluded that while there is a statistical difference that exists of patients with pain present less than 6 month being more likely to have a positive response, the clinical effect is negligible, concluding that 70 % of patients with acute pain can expect to benefit, but up to 60 % of patients with chronic pain can still benefit as well [116]. Also of note though is that the 95 % confidence intervals between the two groups in the combined data overlapped [116].
In a review of 71 patients with lumbar radicular pain due to disk herniation treated with TFESI, clinical and radiological features were assessed for predictors of positive response, defined by at least 50 % reduction in pain 1 month postinjection [147]. The only feature that was found to be significant was grade of nerve root compression. Low-grade nerve root compression responded favorably 75 % of the time as compared to only a 26 % response rate with high-grade compression. For paracentral disk herniation, high-grade compression that is associated with poor response was defined as obliteration of periradicular CSF or fat or morphologically distorted nerve root. For foraminal/far lateral herniation, high-grade compression was defined as perineural fat obliteration in all four directions or morphologic distortion of the nerve root. Duration of symptoms, presence of neurologic symptoms, abnormal neurologic findings on exam, level of herniation, location of herniation, and morphology of disk herniation were all evaluated and not found to be a predictive response to TFESI [147]. Alternatively, in the original Ghahreman study, there was no association between the need to progress to surgery and the size of the disk herniation [117].
33.3.5 Comparative Studies
It is clear that the evidence in support of TFESI for relief of radicular pain is robust and definitive compared to the level of evidence available in support of CESI and ILESI. This has driven clinical practice largely toward predominately using this approach. Additionally, there is evidence that has attempted to directly compare the efficacy of TFESI, CESI, and ILESI.
In a retrospective review of 40 patients with radicular pain due to herniated lumbar disk, Schaufele found that 14 of 20 (70 %) of patients that received TFESI had improvement of at least 2 on a 0–10 pain score compared to only 9 of 20 (45 %) in those that received fluoroscopically guided ILESI. Follow-up was limited to 19 days only [148]. A prospective randomized trial by Thomas that compared fluoroscopically guided TFESI to blind ILESI for radicular pain secondary to lumbar disk herniation in 31 patients found that mean VAS pain score was statistically significantly lower (p < 0.04) in the TFESI group (VAS improvement 56.8 mm) than in the ILESI group (VAS improvement 40.3 mm) at 6 months [119]. The study is limited by small sample size, ILESI being performed blind, and lack of categorical data. Nonetheless, this is valuable evidence in support of the superiority of TFESI over ILESI given with a prospective and randomized design. A similar prospective randomized trial with 64 patients that compared fluoroscopically guided TFESI to fluoroscopically guided ILESI in patients with radicular pain due to lumbar disk herniation was published by Rados in 2011. Comparing the mean pain scores and mean Oswestry scores between the two groups found that while both groups improved at 6 months, there was no statistical difference between the groups [149]. At 6 months, when comparing the percentage of patients that improved by at least two points on VAS (TFESI 84 % vs ILESI 75 %), by at least 50 % on VAS (63 % TFESI vs ILESI 53 %) or ten-point improvement on Oswestry (TFESI 66 % vs ILESI 50 %), there was again no statistically significant difference, but there was a noticeable trend toward more favorable outcomes with TFESI [149]. Collectively, the three studies support the use of TFESI over ILESI for radicular pain.
Related posts:
Epidemiology of Lumbar Degenerative Disk Disease
To Fuse or Not to Fuse: That’s the Question
Lessons from a Life: The Journey of Spinal Neurosurgery in the United States
Minimally Invasive Transforaminal Lumbar Interbody Fusion (TLIF): Indications and Techniques
Oblique Lumbar Interbody Fusion
Facet Joint Pain: Presentation and Treatment, Is It a Myth?
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
