Neurologic Deterioration After Spinal Surgery

Abstract

Neurological deterioration can occur for a number of reasons following spinal surgery. This work details some of the most common causes of neurologic deterioration. There is a focus on preventing deterioration as well as an outline of the appropriate workup which should be pursued if deterioration occurs. A number of treatment strategies are summarized. It is important to approach neurologic duration following spinal operations in a swift, systematic way to minimize the likelihood of permanent deficit.

Keywords

intraoperative spinal cord monitoring, spinal cord ischemia, spinal surgery, spinal hematoma, neurological deficit

Highlights

•

Postoperative neurologic decline includes a wide variety of etiologies.
•

Care must be taken preoperatively to minimize neurologic risk to the patient.
•

A swift, systematic approach should be taken to identify potential causes of postoperative neurologic decline.
•

Once the etiology is identified, immediate action should be taken to remedy the situation and give the patient the best chance of neurologic recovery.

Background

Neurosurgery, at its very core, is a dangerous undertaking. When operating within and around vital neurologic structures, even with the utmost care and a deft touch, the risk of neurologic injury is always present. Care must always be taken to minimize the risk of neurologic injury to the patient. The causes of postoperative neurologic decline are numerous and varied. They include hemorrhagic and vascular etiologies and hardware/implant complications, or they could be secondary to systemic complications such as cardiovascular or respiratory issues. In this chapter we will discuss neurologic decline after spinal surgery, including what the potential risk factors are, how to minimize risk to the patient, how to work up a patient experiencing neurologic compromise after spine surgery, and how to manage these complications. Because of the wide variety of potential etiologies of postoperative neurologic decline, this chapter cannot be an all-encompassing text; however, we will attempt to point out the highlights of the aforementioned topics.

Anatomic Insights

The epidural venous plexus can be subdivided into two parts, the anterior venous plexus and the posterior venous plexus, the latter being larger and the most clinically significant because it is what is encountered during posterior approaches to the spine. There is often an increased convolution of veins at the cervicothoracic junction. This may put patients undergoing surgery in this region at a higher risk of epidural hematoma postoperatively, and therefore extra care should be taken to ensure adequate hemostasis. The epidural venous plexus is postulated to be the source for many postoperative epidural hematomas, although other etiologies may exist.

The spinal cord receives blood flow mainly from one anterior spinal artery and two posterior spinal arteries. The anterior spinal artery, which is formed by anterior spinal branches of the vertebral arteries at the level of the foramen magnum, lies over the anterior sulcus of the spinal cord and supplies the anterior two-thirds of the spinal cord. Whether the anterior spinal artery is continuous or segmental is considered to be controversial. The anterior spinal artery relies heavily on segmental feeders as it descends the spinal cord. The posterior spinal arteries are formed by branches of the posterior inferior cerebellar arteries (PICA) and supply the posterior one-third of the spinal cord in the cervical region. As one moves more caudally, the segmental arteries assume a greater role in supplying blood to the anterior and posterior spinal cord.

Paired radicular arteries originate from the thoracic aorta, enter the neuroforamina, penetrate the dura, and irrigate the spinal cord along with other local structures. In the thoracic region, these arteries bifurcate into anterior and posterior branches that join the anterior spinal artery and posterior spinal arteries, respectively. In the adult, many of these segmental arteries have become vestigial, with the cord being supplied by a few larger radicular arteries. Although somewhat controversial, the largest radicular artery is known as the artery of Adamkiewicz, or the arteria radicularis magna. The artery of Adamkiewicz has been reported to originate from the left in 80% of the population, from between T5–8 in 15%, T9–12 in 75%, and L1–2 in 10%. Biglioli et al. in a 2000 cadaveric study reported that the artery of Adamkiewicz originated from the left in 68% of cases and from between T12–L3 in 84%. This leaves areas prone to watershed infarcts around T1, T5, and T8–9. Cheshire et al. reported a relative hypovascular zone from T4–T8.

The neural structures of the spine are at risk from direct injury during spinal procedures, especially if instrumentation or hardware is used. Nerve roots pass medially and inferiorly to the pedicles in the spine and are at risk during pedicle screw placement, particularly in the thoracic spine due to relatively small pedicles. In one cadaveric study, Ebraheim et al. found that inferiorly, the greatest distance between the pedicle and nerve root was at T6–T9 (3.1–3.7 mm) and that the smallest distance was at T1–T2 (1.7–1.8 mm). They found no difference between the pedicle and thecal sac medially at any level.

When contemplating surgical intervention on patients with spinal deformity, care must be taken to appreciate all aspects of the deformity because the patient’s spine may have aspects of coronal and sagittal as well as rotational deformity. Understanding these nuances could be a book in and of itself; therefore, we will touch only briefly on these topics in this chapter. Patients with scoliosis often have rotation of the spine toward the convexity of the deformity. This leads to the pedicle screw on the concavity requiring a more medialized trajectory, and the opposite for the convexity. Furthermore, the neural elements will be shifted toward the concavity, leaving a relatively large free space lateral to the thecal sac at the convexity. Patients with large coronal or sagittal deformities are at increased risk of neurologic compromise during corrective maneuvers, and deformities in all planes must be measured out preoperatively for surgical planning and risk assessment.

Red Flags

Of the potential postoperative causes for neurologic decline after spinal surgery, postoperative epidural hematoma that compresses the spinal cord or cauda equina may be the most important to identify quickly and to intervene in surgically to restore a patient’s neurologic function. However, this complication is fairly rare, occurring in 0.1% to 0.24% of spinal surgical cases. Despite the relatively low number of symptomatic postoperative epidural hematomas, radiographic epidural hematomas have been reported in 33% to 100% of patients. In a 2005 study, Awad et al. described several risk factors for postoperative epidural hematoma, including age >60 years old, the use of preoperative nonsteroidal antiinflammatories, Rh-positive blood group, hypertension, coagulopathy (hepatitis B, hepatitis C, liver cirrhosis, thrombocytopenia, history of easy bruising, pernicious anemia, and neoplasms), and tobacco use using univariate analysis. When a logistic multivariate regression model was used, hypertension and medical coagulopathy proved not to be significant. When intraoperative variables were examined, risk factors for postoperative epidural hematoma included more than five operative levels, a hemoglobin <10 g/dL, and blood loss >1 L. Postoperatively, the only risk factor for development of an epidural hematoma was an international normalized ratio (INR) >2.0 within 48 hours after surgery. Sokolowski et al. reported in 2008 that age >60 years old, elevated preoperative INR, multilevel procedures, and higher levels of blood loss were associated with postoperative epidural hematomas. In 2002, Kou et al. also discussed risk factors for the development of spinal epidural hematoma postoperatively. They identified undergoing a multilevel surgery as well as having a preoperative coagulopathy as increasing the risk of epidural hematoma. Groen and Ponssen report larger exposures of the epidural space may increase the likelihood of bleeding from the venous plexus.

In one series, Moufarrij reported an increased incidence of symptomatic postoperative epidural spinal hematomas after implantation of spinal cord stimulator paddle leads versus thoracic laminectomies without the implantation of leads (2.60% vs 0.84%). None of these patients had identifiable risk factors for bleeding. Therefore, extra care should be taken when planning the placement of epidural paddle leads, and the patient should be counseled on this potential risk.

Patients with preexisting compressive lesions in the cervical or thoracic spine may be at significant risk for neurologic deterioration after a spinal surgery if the proximal lesion is not addressed first. Patients presenting with signs and/or symptoms of myelopathy should be further examined with MR imaging of the cervical and/or thoracic spine before surgery. Sensory abnormalities of the trunk may be a tip-off of thoracic pathology that can be masked in the setting of lumbar stenosis. Therefore a careful neurologic examination and thorough documentation are a must for all patients undergoing planned spinal surgery.

Although postoperative urinary retention does not necessarily represent neurologic decline, it is a significant problem in spinal surgery patients. It has been reported in 8.8% of elective spine patients in a recent study. Risk factors include posterior lumbar surgery, patients with preexisting benign prostatic hypertrophy, chronic constipation, prior history of urinary retention, long operative times, and the use of a patient-controlled analgesia pump postoperatively.

Patients undergoing planned corrective procedures for spinal deformity may also be at an increased risk of postoperative neurologic decline, particularly in patients with sagittal imbalance. Patients with >80 degrees of coronal or sagittal deformity are at significant risk of neurologic compromise during corrective procedures. Lenke et al. report an incidence of new lower-extremity weakness of 22% in patients with coronal or sagittal deformities of >80 degrees. They also report a relatively high risk for patients undergoing a 3-column osteotomy such as a vertebral column resection or a pedicle subtraction osteotomy. The presence of preoperative motor deficits also significantly increased the risk of new postoperative deficits. Patients with cardiopulmonary comorbidities may also be at higher risk, secondary to either systemic hypotension or to vascular injury causing ischemia of the spinal cord. There are several physiologic factors that determine perfusion of the spinal cord, including systemic blood pressure or mean arterial pressure, hemoglobin concentration of the blood, and intravascular volume.

Prevention

Prevention of postoperative spinal epidural hematomas can best be achieved by limiting preoperative risk factors. Coagulopathies should be identified and corrected before surgery. Preoperative testing should include platelet count, prothrombin time/INR, and partial thromboplastin time. Patients should be counseled on when to stop medications such as anticoagulant, antiplatelet medication, and nonsteroidal antiinflammatory medications. Nonmodifiable risk factors such as advanced age, high anticipated blood loss, and a planned multilevel procedure should be acknowledged so the surgical team can maintain vigilance in the perioperative period. Postoperatively, intravenous anticoagulation use should be avoided for at least 12 hours. Furthermore, INR should be maintained <2.0 for 48 hours postoperatively.

Postoperative neurologic decline has also been associated with missed thoracic and/or cervical lesions. Thorough neurologic examination and documentation—including sensory testing of the trunk because thoracic myelopathy may be masked in the face of lumbar stenosis—before any planned surgery is essential to identify any additional compressive lesions that may compromise neurologic functioning during or after a surgery. If there are any concerning signs or symptoms of a proximal compressive lesion causing myelopathy, magnetic resonance imaging (MRI) scans of the cervical and/or thoracic spine should be obtained before any planned surgeries.

As spinal cord ischemia is another potential cause of postoperative neurologic decline, the patient should be optimized preoperatively to minimize the risk of spinal cord ischemia during surgery. Labs and vital signs should be noted to ensure adequate blood pressure, intravascular volume, and hemoglobin/hematocrit. Also, preparations should be made preoperatively so the surgical team is prepared in the case of excessive blood loss.

When planning a spinal surgery, intraoperative neuromonitoring (IONM) should be considered. Although IONM has not been shown to decrease neurologic complications in all types of spinal surgery, when performing a surgery to correct a spinal deformity, the use of multimodality monitoring has become commonplace. The sensitivity and specificity values for detecting neurologic complications using multimodality IONM have been reported to be up to 100%.

Although postoperative wound drains are common, either subfascial or suprafascial, several studies have shown no difference in the incidence of postoperative epidural hematoma between patients with postoperative wound drains versus no drains. There is one recent report, however, that showed less epidural hematomas on postoperative day 1 on MRI in patients with a subfascial drain.

Only gold members can continue reading. Log In or Register to continue