Spine Injections

Introduction

More than 80% of adults will suffer from low back pain at some point in their lives. The majority of back pain is attributed to muscular and ligamentous strain and spasm, which cannot be reliably attributed to a specific disease or spinal pathology. Mechanical back pain is typically somatic pain. More complex, severe, and chronic pain may arise from the facet joints and paraspinal muscles in the dorsal compartment innervated by the medial and lateral branches of the dorsal primary rami. Disorders involving the anterior and posterior longitudinal ligaments and the annulus of the disc in the ventral compartment may also lead to low back pain. An annular tear may have continued leakage of the nucleus pulposus material, which can lead to chronic inflammation. Radicular pain can result from chemical irritation and inflammation of the nerve root, which may become swollen and edematous. Disc herniation results in release of large amounts of phospholipase A2, which results in production of prostaglandins and leukotrienes from cell membranes. This leads to inflammation, sensitization of nerve endings and resultant pain. Bone overgrowth or age-related changes producing annular and ligamentous bulging can lead to external pressure on the nerve roots, resulting in venous obstruction and neural edema.

Epidural Injection

The primary indication for an epidural steroid injection is radicular pain caused by nerve root inflammation, irritation, and edema. The most common use of epidural injections is to place steroids into the epidural space, where the medications spread to bathe the spinal nerves. Studies and reviews have been done in the past to help identify which patients are most likely to benefit from epidural steroid injections. Some 304 patients with low back pain were assessed with regards to their response to epidural steroid injections, and the findings were then correlated with the specific pathology causing the low back pain. The best response was observed in patients with recent onset of nerve root irritation symptoms and the absence of psychopathology. Epidural steroid injections were also therapeutic for patients with herniated discs and either nerve root irritation or compression, and patients with spondylolisthesis or stenosis. Other studies have reported therapeutic benefit in patients with radicular pain syndromes or herniated nucleus pulposus, as well as lumbar spinal stenosis with or without neurogenic claudication. ^,

Anatomy

Similar to the variability in the anatomy of the bony structures of the vertebrae from the cervical to the sacral levels, the anatomy of the epidural space varies as well. The epidural space is defined as the area between the dura mater and the vertebral bones ( Fig. 73.1 ). The epidural space begins at the foramen magnum and extends to the sacrococcygeal ligament. It is filled with epidural fat, a venous plexus, and areolar tissue. The epidural space is smaller in the cervical and thoracic spine when compared with the lumbar spine. In the case of the interlaminar approach, the epidural space is typically accessed in the midline via the space between adjacent laminae. A loss of resistance technique is used to advance an epidural needle through the dense ligamentum flavum. Loss of resistance signifies entry into the epidural space. The ligamentum flavum is shaped like a tent, as is the dorsal epidural fat collection, with the peak of the tent’s roof in the midline and typically associated with the greatest depth of epidural fat. The lateral aspect of the ligamentum flavum can be as much as 1 cm deeper or more anterior than the midline. Therefore, if the epidural needle is not midline, the epidural space will occur at a deeper level, and the epidural fat collection will be smaller and in some cases absent. The ligamentum flavum is thickest at the lumbar and thoracic levels and thinnest at the cervical levels. Similarly, the anterior-posterior dimension of the epidural space itself is narrowest in the cervical region, ranging from 2 to 3 mm in thickness, increases at the thoracic level to about 3 to 5 mm in thickness, and is the largest at the lumbar level, ranging from 5 to 6 mm. ^,

Cervical Epidural Steroid Injections

Patient positioning is important for all epidural injections. For cervical epidural steroid injections, the patient typically lies prone with the head in a small headrest. The C-arm is rotated 15 to 20 degrees caudally from the axial plane. This allows visualization of the interlaminar space and allows for needle advancement between adjacent spinous processes. The skin is prepped and draped in a sterile manner. The skin overlying the interspace is anesthetized with 1% lidocaine. The largest interlaminar distance and epidural fat collection is found between C6‒C7 and C7‒T1. Given this anatomy, needle entry is most commonly directed at one of these two levels, and spread of medication is relied upon to reach the level of patient pathology. This technique may be performed at interspaces C3‒C4 and below, although the epidural space may be significantly smaller and the risk higher, and has not been described at or above the C2‒C3 level. An 18- or 20-gauge Tuohy needle is placed through the skin and advanced until it is engaged in the interspinous ligament. Anteroposterior (AP) images are used to determine midline placement of the needle ( Fig. 73.2 ). Once the Tuohy needle is engaged in the ligament, a syringe containing a small amount of preservative-free saline is attached to the needle, and the needle is slowly advanced in 1- to 2-mm increments until a loss of resistance is experienced. Images should be taken repeatedly after 5 to 10 mm of advancement. Understanding the anatomy is crucial to avoiding injury to the patient, given the proximity of the spinal cord at these levels. Contralateral oblique imaging, which involves rotating the C-arm to 45 to 55 degrees oblique to either side of midline, allows visualization of the laminae and the posterior extent of the spinal canal as the needle is advanced ( Fig. 73.3 ). The contralateral oblique technique is advantageous in obese patients, where bony and soft tissues may obscure the important images in the lateral view. Conversely, the C-arm may be rotated to a full lateral view to continue the injection ( Fig. 73.4 ). The lateral view of the cervicothoracic junction may be obscured by the shoulders, torso, and/or arms. After the needle tip enters the epidural space, 1 to 1.5 mL of nonionic radiographic contrast is injected to confirm epidural spread, and this spread is verified in the AP and lateral views. Following confirmation of correct needle placement, local anesthetic or saline and steroid are injected slowly.

Thoracic Epidural Steroid Injections

Patients lie prone with the head either straight, as during cervical epidural steroid injection, or turned to one side. The C-arm is rotated 40 to 50 degrees caudally from the axial plane to allow visualization of the interlaminar space and to guide needle advancement. The skin is prepped and draped in a sterile manner. The skin is anesthetized at the level of entry with 1% lidocaine, and an 18- or 20-gauge Tuohy needle is advanced until it is firmly seated in the tissue. AP imaging is then used to direct the needle toward the midline, as well as to the margin of the inferior lamina of the interspace that is to be entered between the spinous process and the lamina. The midline approach may be used at the thoracic levels, but given the steep angle of the spinous processes true coaxial placement is difficult, especially at the midthoracic level. Repeat images are taken as the needle is advanced in 2- to 3-mm increments. The needle tip is directed over the lamina until gentle contact is made with the bone; then the needle is advanced superiorly into the interlaminar space at 1- to 2-mm increments until a loss of resistance occurs. Once in the epidural space, contrast is injected to confirm placement in the AP and lateral views. Then local anesthetic or saline and steroid is injected slowly ( Figs. 73.5 and 73.6 ).

Lumbar Epidural Steroid Injections

Patients lie prone with the head either turned to the side or straight as with thoracic epidurals. To increase the separation between adjacent spinous processes and reduce the lumbar lordosis, a pillow is placed under the mid and lower abdomen. To optimize visualization of the interlaminar space, the C-arm is rotated 15 to 20 degrees caudally from the axial plane. The skin is prepped and draped in a sterile manner. The skin is anesthetized at the level of entry using 1% lidocaine, and an 18- or 20-gauge Tuohy needle is placed until it is firmly seated in the interspinous ligament. AP imaging is used to direct the needle toward midline. The C-arm is then rotated for a lateral imaging, and the needle is advanced until it lies posterior to the junction of the spinous process and the laminae. The loss of resistance technique is then employed, using 1 to 3 mL of preservative-free saline in a glass syringe attached to the Tuohy needle. Correct needle position is confirmed with injection of nonionic radiographic contrast, and spread is confirmed in the AP and lateral planes ( Figs. 73.7 and 73.8 ). Now local anesthetic or saline and steroid may be injected slowly.

Caudal Epidural Steroid Injection

The caudal approach to access the epidural space may be used to target the sacral nerves, as well as for patients in whom direct access via the lumbar epidural approach is difficult or in whom the epidural space in the lumbar region may be compromised, that is, those with severe osteoarthritis and/or previous lumbar surgeries, respectively. The patient lies prone with the head either turned to the side or straight, as with thoracic and lumbar epidural injections. The C-arm is rotated to properly visualize the sacrum, sacral hiatus, and coccyx. It is often difficult to identify the sacral hiatus in the AP view, and the sacral hiatus may be identified in the lateral view. In the lateral view, the sacral hiatus may be identified by following the posterior sacral plane to its inferiormost extent at the superior aspect of the sacral hiatus. In some cases the sacral hiatus may be difficult to identify by fluoroscopy, and palpating the sacral cornu in the midline near the superior extent of the gluteal cleft is helpful in locating the sacral hiatus. The skin is prepped and draped in a sterile manner. After the sacral hiatus is identified the skin overlying this area is anesthetized with 1% lidocaine. An 18- or 20-gauge Tuohy needle or smaller 22-gauge 3.5-inch spinal needle may be used to advance toward the sacrococcygeal ligament and into the caudal epidural space. The angle of the needle is decreased to lie closer to the plane of the sacrum and advanced 1 to 2 cm into the caudal epidural space. AP and lateral imaging are used to confirm the needle’s position. At this point 1 to 2 mL of nonionic radiographic contrast is injected to confirm correct needle position ( Figs. 73.9 and 73.10 ). In some settings, a catheter may be used with the caudal technique. In this case a 17-gauge Tuohy needle would be placed in a similar manner until it is in the caudal canal, as described earlier. Once in place, a soft flexible catheter with a stylet is advanced through the Tuohy needle until the desired level is reached; typically in the lower lumbar region. Once the catheter is at the desired level, contrast dye is once again injected to confirm catheter position in the epidural space. The medication is then injected; the catheter may be retracted as the therapeutic medication is injected for possible spread at multiple levels.

Medications

Corticosteroids have been studied extensively for their ability to suppress inflammation. The mechanism of action is most likely related to steroids’ ability to decrease or inhibit phospholipase A2 activity. Steroids may also have a local anesthetic and antinociceptive effect.

Methylprednisolone acetate and triamcinolone diacetate are the most well-studied steroids for epidural injection. The concentration of both of these medications is typically 40 to 80 mg/mL, which is also the most common therapeutic dose range. Both have been reported to be effective and safe. The steroids are typically combined with preservative-free saline or local anesthetic. In the cervical, thoracic, and lumbar regions, typically a total volume of up to 3 to 5 mL is injected, with smaller volumes used in the cervical region. In the caudal region a minimum total volume of 5 mL is injected.

Use of preservative-free local anesthetic is common with epidurals. However, with an unintended intrathecal injection and the volumes and doses used for epidurals, there is concern for high spinal anesthesia as a result of the local anesthetic injected. This is especially concerning in the cervical and thoracic regions, above the level of T4. The effects are of variable severity, depending on the level that is involved, but may include cardiovascular and/or respiratory compromise. If the local anesthetic spreads to the brainstem, a total spinal may occur and result in a complete loss of consciousness.

In 2014, the US Food and Drug Administration issued a drug safety communication about epidural injections of steroids, noting the potential for rare but serious adverse effects including loss of vision, stroke, paralysis, and death. These complications are most commonly related to intraarterial injection of particulate steroid. Brain and spinal cord infarction have been reported in such cases as well.

Complications

Dural puncture can occur, which may result in a post–dural puncture headache. Anatomically, the spinal dura mater extends from the foramen magnum to the second segment of the sacrum. It consists of a dense connective tissue matrix of collagen and elastic fibers. Dural puncture may occur if the epidural needle inadvertently punctures the dura itself. The pathophysiology of a post–dural puncture headache is thought to be related to the leakage of cerebrospinal fluid (CSF) through the dural hole created by the needle. Low CSF pressure can result, worsened at the level of the brain in the upright position; thus the headache is typically positional, in that it is worse when the patient is upright and improved with lying flat. Low CSF pressure results in meningeal venodilation, and a headache may be caused by acute venous distension. Secondly, a headache may be caused by traction of the nerves, including the cranial nerves, resulting from the low CSF pressure. Headaches are typically managed conservatively with fluids and oral analgesics; although in the lumbar region, if symptoms fail to improve after 24 to 48 hours, an epidural blood patch using autologous blood is a safe and effective treatment that relieves headache symptoms promptly in the majority of patients. The incidence of post–dural puncture headaches is highest in the lumbar region and more so in patients with prior surgery. Cervical and thoracic epidural blood patches have been described, but are not commonly used. Direct trauma to the spinal cord in the cervical and thoracic region, with consequences such as paraplegia and quadriplegia, has been described. Deep sedation leads to higher risk, and sedation, if used, should allow direct conversation between the patient and the practitioner to ensure communication in the event of contact with neural elements before the occurrence of significant traumatic injury. This is especially true with cervical epidural injections, where sedation has been directly correlated with adverse outcomes. For cervical procedures, regardless of whether they are interlaminar or transforaminal, closed claims analysis has revealed that patients being heavily sedated or unresponsive at the time of injection is significantly associated with an increased risk of spinal cord injury. There is agreement that, if sedation is used, it should be light enough to allow the patient to communicate pain or other sensations felt during the procedure. Direct trauma to the cauda equina or spinal nerves is unlikely with careful use of radiographic guidance. A careful review of imaging studies should be performed before injections, as any level where there may be effacement of the epidural space is high risk for an interlaminar epidural steroid injection. All interlaminar epidural injections at any level may results in epidural bleeding or infection; hematoma or abscess may lead to significant spinal cord or neural element compression. ^,

Contraindications

Absolute contraindications to epidural injections include systemic infection or local injection at the site of the infection; this may lead to abscess formation in the epidural space. Full anticoagulation or a history of bleeding disorders may lead to epidural hematoma and is also an absolute contraindication. Significant allergies to contrast, anesthetic, or the steroid used in the injection itself may require modification of technique. Local malignancy is also an absolute contraindication to proceeding with an epidural injection. Finally, patient refusal to any given elective procedure is an absolute contraindication. ^,

Interlaminar Versus Transforaminal Approach

Theoretically the transforaminal approach should result in better delineation of the nerve root, as well as improved spread in the anterior aspect of the epidural space. There is increased concern with regards to neurological injury and catastrophic complications, including spinal cord injury or stroke associated with transforaminal injections. The use of live fluoroscopic imaging or digital subtraction angiography (DSA) may reduce the risk to some degree. Previous surgical interventions leading to the loss of the posterior epidural space are a limitation of the interlaminar approach.

Transforaminal Epidural Steroid Injections and Selective Nerve Root Blocks

The transforaminal approach is based on the idea that injecting steroid around inflamed neural structures will provide superior relief of radicular pain when compared with injecting steroids into the dorsal epidural space alone. This is consistent with the theory that radicular pain occurs because of pathology in the ventral epidural space from disc protrusion, extrusion, leakage of the nucleus pulposus, or mechanical compression.

The techniques of selective nerve root blocks (SNRBs) and transforaminal epidural steroid injections (TFESIs) are the same, with the exception of the final needle position relative to the foramen itself. The purpose of the injection is different as well; SNRBs are typically used as diagnostic tools as opposed to therapeutic tools. Similar to the interlaminar approach, the approach to the injection differs based on cervical, thoracic, or lumbar locations, given the anatomy of the neural foramen and the surrounding structures. The highest risk is within the cervical region, given the high vascularity in the foramen and surrounding the foramen.

Indications

The indications for TFESIs or SNRBs are similar to those for interlaminar epidural steroid injections and include radiculitis/radiculopathy lumbar disc displacement, diagnostic blocks for vague symptoms or multilevel pathology, postlaminectomy patients with recurrent pain, and spinal or foraminal stenosis. Contraindications for these injections are similar to those for lumbar epidural steroid injections and include patient refusal, bleeding disorders, anticoagulation, infection, and local malignancy. The equipment required is similar to that required for a lumbar epidural steroid injection and includes a C-arm fluoroscope (DSA capability is desirable, and computed tomography [CT] may also be used) and fluoroscopy table, vital sign monitors, and 22- or 25-gauge spinal or blunt-tip needles varying in length from 2 to 7 inches, depending on patient size. Commonly used medications include corticosteroids, which include methylprednisolone, triamcinolone, betamethasone, and dexamethasone, and nonionic contrast dye.

Anatomy

The roof and floor of the cervical foramina are composed of the pedicles of consecutive vertebrae. Dorsally the cervical foramina is made of the superior articular process (SAP) of the lower vertebra and ventrally by the lower part of the upper vertebral body, the uncinate process of the lower vertebra, and the intervertebral disc. The epiradicular veins are in the superior portion of the foramen, and the spinal nerve is in the lower part of the foramen. The radicular and medullary arteries are made up of the vertebral arteries or the deep ascending cervical arteries, which supply the nerve roots and the spinal cord, respectively. The vertebral artery rises in a cephalad direction just anterior to the articular pillars of the zygapophyseal joint, immediately lateral to the foramen. These branches are at risk of injection or injury during cervical TFESIs or SNRB.

The boundaries of the thoracic foramina are similar to those at the cervical level; however, the ribs, pleura, and mediastinum surround the foramina and are at risk of puncture during the procedure. The artery of Adamkiewicz is also at risk of injury by puncture at the lower thoracic levels.

The anterior border of the lumbar foramina includes the upper vertebra and intervertebral disc. Posteriorly, the superior border is made up of the inferior articular process (IAP), and the inferior border is made up of the superior articular border. The pedicles form the roof and the floor. The artery of Adamkiewicz is the main arterial supply to the lower two-thirds of the spinal canal and enters on the left side between T9 and L1 in 80% of individuals, but can enter anywhere from T7 to L4. ^, Injury to this artery may lead to anterior spinal artery syndrome and paraplegia.