Regional Anesthesia & Pain Medicine:
ASRA Practice Advisory on neurologic complication
ASRA Practice Advisory on Neurologic Complications in Regional Anesthesia and Pain Medicine
Neal, Joseph M. M.D.a,b; Bernards, Christopher M. M.D.a,b; Hadzic, Admir M.D.c; Hebl, James R. M.D.d; Hogan, Quinn H. M.D.e; Horlocker, Terese T. M.D.d; Lee, Lorri A. M.D.b; Rathmell, James P. M.D.f; Sorenson, Eric J. M.D.h; Suresh, Santhanam M.D.g; Wedel, Denise J. M.D.d
aDepartment of Anesthesiology, Virginia Mason Medical Center, Seattle, WA
bDepartment of Anesthesiology, University of Washington, Seattle, WA
cDepartment of Anesthesiology, St. Luke's–Roosevelt Hospital Center, New York, NY
dDepartment of Anesthesiology, Mayo Clinic. Rochester, MN
hDepartment of Neurology, Mayo Clinic. Rochester, MN
eDepartment of Anesthesiology, Medical College of Wisconsin, Milwaukee, WI
fDepartment of Anesthesiology, Massachusetts General Hospital, Harvard University, Boston, MA
gDepartment of Anesthesiology, Childrens' Memorial Hospital, Northwestern University, Chicago, IL
No funds were received for this work. Honorio T. Benzon, M.D. served as Acting Editor‐in‐Chief for this report.
No reprints will be available.
Accepted for publication July 6, 2008
Neurologic complications associated with regional anesthesia and pain medicine practice are extremely rare. The ASRA Practice Advisory on Neurologic Complications in Regional Anesthesia and Pain Medicine addresses the etiology, differential diagnosis, prevention, and treatment of these complications. This Advisory does not focus on hemorrhagic and infectious complications, because they have been addressed by other recent ASRA Practice Advisories. The current Practice Advisory offers recommendations to aid in the understanding and potential limitation of neurologic complications that may arise during the practice of regional anesthesia and pain medicine.
The American Society of Regional Anesthesia and Pain Medicine (ASRA) convened a group of experts to develop a Practice Advisory on Neurologic Complications in Regional Anesthesia and Pain Medicine. The goal of this Practice Advisory is to provide information for practitioners of regional anesthesia and pain medicine regarding the etiology, differential diagnosis, prevention, and treatment of neurologic complications. This Practice Advisory focuses on neurologic injuries apart from those caused by hemorrhagic1 or infectious2‐6 complications, both of which have been the subject of other recent ASRA‐sponsored Practice Advisories. The current report is a summation of the Practice Advisory's findings and recommendations. Anesthesiologists are strongly encouraged to read the manuscripts that accompany the present summary document, because they contain the details upon which recommendations are based. The accompanying manuscripts represent most of the subtopics discussed at the conference.
The ASRA Practice Advisory on Neurologic Complications in Regional Anesthesia and Pain Medicine was convened on April 23, 2005 at the ASRA Annual Meeting in Toronto, Ontario, Canada. The project was approved by the ASRA Education Committee and Board of Directors. The panelists were chosen by ASRA based on demonstrated expertise in issues related to neurologic injury, and included all of the authors of this manuscript in addition to a malpractice attorney.* Panelists received no compensation for their contributions to the Practice Advisory, nor did any declare a conflict of interest pertinent to the topic. Panelists were charged with performing an extensive review of the literature, summarizing and presenting their findings at the conference, and producing a manuscript based on their scholarly work. During the Toronto conference, panelists and attendees discussed several issues related to neurologic injury in an open forum format. All subsequent recommendations were reviewed and approved by members of the panel. Manuscripts were first peer reviewed internally by 3 members of the expert panel and subsequently peer reviewed externally using this journal's standard review process.
The recommendations contained within this Practice Advisory represent the opinions of a small group of expert panelists and are based upon their clinical experience, review of limited scientific literature (anatomic and physiologic studies of animals and human cadavers, case reports, retrospective case series, and non‐randomized trials), and open forum discussion. Importantly, in this imperfect setting of controversial topics, limited data, and bias inherent to expert opinion, the Panel consistently tended towards making conservative recommendations. We acknowledge that other experts analyzing the same information may arrive at recommendations different than ours. The recommendations contained herein are not intended to define standard practice or to ensure the avoidance of adverse outcomes. Furthermore, the recommendations are subject to change as new information becomes available. Recommendations from the Practice Advisory are not intended to replace clinical judgment, and specific risk‐to‐benefit discussions as they pertain to individual patients.
Because neurologic injuries related to anesthesia and pain medicine practice are extremely rare, standard tools of evidence‐based medicine such as randomized controlled trials, meta‐analysis, and prospective human studies rarely exist—and are unlikely to be available in the future. Therefore, recommendations contained within this Practice Advisory could not be easily graded based on traditional stratification methods, such as those used by the United States Agency for Health Care Policy and Research.7 Consequently, the Panel chose to stratify its recommendations based on a simplified schema used by other medical specialties facing similar circumstances (Table 1). 8
Overview of Anesthesia‐Related Neurologic Injury
Because of the infrequency at which anesthesia‐ and pain medicine‐related neurologic injuries occur, it is extremely difficult to obtain reliable and consistent incidence data. Medicolegal and insurance‐based data are biased by the very presence of injury, whereas underreporting may bias clinical‐based data. Therefore, medicolegal data such as that provided by the American Society of Anesthesiologists (ASA) Closed Claims Project may overestimate the occurrence of injury, while clinical studies potentially underestimate the true incidence. Academic regional anesthesiologists are reasonably accurate in estimating the incidence of permanent neuropathy from neuraxial or peripheral blocks.9 However, it is perhaps worrisome that the risk of permanent injury is only included in 42% to 77% of their informed consent discussions with patients.9,10
Recent clinical data from Sweden suggest that the frequency of hematoma is 1.3 to 2.7 per 100,000 (95% confidence interval [CI]) neuraxial anesthetics.11 Neuraxial infectious complications (epidural abscess and meningitis) range from 0 infections in over 70,000 epidural and spinal anesthetics12 to 1 abscess per 1,930 epidural anesthetics.13 Hematoma and infectious complications are relatively common as compared with conditions such as anterior spinal artery syndrome (ASAS), and direct needle trauma. For instance, no cases of ischemic spinal cord injury were reported in over 70,000 neuraxial anesthetics performed in France over a 10 month period.12 Only 12% of neuraxial injuries in the ASA Closed Claims Database were associated with ASAS or spinal cord infarction; only 6 of 821 neuraxial claims involved suspected direct spinal cord trauma.14 In contrast to the often transient nature of peripheral nerve injuries, most neuraxial injuries are permanent. In fact, permanent nerve injury has been shown to range from 15% in French surveillance studies12 to 80% to 100% in the ASA Closed Claims Database—with the frequency being highly dependent on the type of neural lesion.14
In the early postoperative period, mild paresthesias may be present in up to 15% of patients that undergo peripheral nerve block.15 Most of these symptoms resolve within days to weeks, with over 99% completely resolving by 1 year.16,17 Serious neurologic injury reported in a large prospective study occurred in 2.4 per 10,000 peripheral nerve blocks.18
Risk Factors Unrelated to Anesthesia
Proper evaluation of perioperative nerve injuries includes a search for causes unrelated to anesthetic technique. The plethora of factors that can contribute in whole or in part to perioperative neural injury underlies the difficulties encountered in both the academic study and clinical evaluation of this phenomenon. In brief, patient‐specific risk factors for perioperative nerve injury include pre‐existing neurologic disorders, diabetes mellitus, extremes of body habitus, male gender, and advanced age. Surgical risk factors include direct surgical trauma or stretch, compressive dressings or casts, tourniquet inflation, hematoma or abscess formation, perioperative inflammation, and improper patient positioning.17
ASA Closed Claims Perspective
Excluding obstetric and pain management cases, the ASA Closed Claims Database (1980‐1999) showed that neuraxial blocks were associated with temporary nerve injury in 38% and permanent nerve injury in 26% of cases with complications reported to the database. Of those neuraxial injuries believed to be block‐related (41%), 45% were attributed to the block technique, including direct needle or catheter damage to the neuraxis or adjacent nerves. Most permanent injuries involved the lumbosacral nerve roots or the thoracolumbar spinal cord. Over half of the neuraxial injuries were caused by an epidural hematoma, with the remaining injuries being caused by assorted conditions such as ASAS, epidural abscess formation, or meningitis.14
Of all claims associated with neuraxial chronic pain management, 25% involved nerve damage. Injuries associated with treatment for chronic pain were most likely to result from an infectious cause, with hematoma and direct needle injury occurring less frequently. The most striking finding stemming from the ASA analysis of chronic pain claims was that use of local anesthetic and/or opioid during epidural steroid injection was associated with 9 deaths or brain damage, while there were no deaths or brain damage when local anesthetic and/or opioid was not used as part of the neuraxial injectate. The authors warn that injection of neuraxial local anesthetic and/or opioid for pain treatment in the ambulatory setting should be accompanied by the same close monitoring, and the ability to perform resuscitative maneuvers that are available to those patients receiving neuraxial local anesthetic and/or opioid in the operating room or obstetric settings.19,20
Most claims involving peripheral nerve injury were for temporary injury (56%) with half of the deficits believed to be block‐related. The brachial plexus is the most commonly injured peripheral nerve structure in the Closed Claims Database, followed by the median, ulnar, and radial nerves. Lower extremity nerve injuries are rare in the Closed Claims Database, most likely reflecting the less frequent practice of lower extremity regional anesthesia. A notable increase in claims involving eye blocks performed by anesthesiologists has occurred since 1980.21 The most serious injuries from peripheral nerve or facet blocks placed for chronic pain involved blocks performed in the neck or near the neuraxis that ultimately injured neuraxial structures.20
Pathophysiology of Neuraxial Injury
Neuraxial injury may involve the spinal cord, spinal nerve roots, spinal nerves, or spinal vasculature. With the exception of obvious compressive lesions such as an epidural hematoma or abscess, the etiology of neuraxial injuries is rarely apparent. Neuraxial injuries are typically linked to needle‐ or catheter‐related mechanical damage to the neural or vascular components, to space‐occupying lesions competing for limited cross‐sectional area within the spinal canal, to ischemia, or to drug‐related neurotoxicity.
Direct needle or catheter trauma to the spinal cord may be associated with inaccurate determination of vertebral levels, anatomical variation in the terminal portion of the conus medullaris, incompletely fused ligamentum flavum, or progressive (caudad to cephalad) posterior‐to‐anterior epidural space narrowing. Laterally directed needles intended to enter the dorsal epidural space may injure spinal nerves or vasculature near the medial aspect of the intervertebral foramina, while needles advanced medially during transforaminal injection may injure these same structures within or just lateral to the intervertebral foramina. Direct trauma is further complicated by inconsistent and unreliable signals when needles or catheters contact or enter the spinal cord. The spinal cord has no sensory receptors and sensory input from the meninges is inconsistent. This may explain, in part, reported cases in which needle entry into the spinal cord was not recognized even in awake patients. Actual injection into the spinal cord appears to more consistently elicit a painful response, most likely because of pressure‐related stimulation of afferent neurons.22
Another mechanism leading to neuraxial injury is the presence of a mass lesion that compresses the spinal cord within the fixed cross‐sectional area of the spinal canal. Intradural or extradural compression of the cerebrospinal fluid (CSF) or the spinal cord itself, especially when it develops rapidly, may generate pressures sufficient to compromise blood flow and cause ischemia or infarction. Hematoma and abscess are relatively common mass lesions that are of sufficient size to impair spinal cord function. In contrast, abnormal epidural fat, tumor, hypertrophic ligamentum flava, bony overgrowth of the vertebral canal or foramina, temporary mass effect from local anesthetic, or a relative change in cross‐sectional area secondary to various surgical positions may occur less often. Although these conditions have the potential to compress the neuraxis, their primary importance lies in further reducing available spinal canal cross‐sectional area should a more acute process such as an epidural hematoma occur.22
Although isolated case reports describe what is speculated to be local anesthetic‐induced spinal injury in presumably normal individuals, local anesthetics are rarely neurotoxic when administered in recommended concentrations and dosages. The potential for toxicity is heightened when mechanical damage to the spinal cord or neural fibers damages their normally protective connective tissue barriers, when injection or catheter insertion physically disrupts the structural integrity of the spinal cord, or when vasoconstrictors impede local anesthetic clearance. Some neuraxial components may theoretically be more susceptible to local anesthetic neurotoxicity—such as the cauda equina because it is partially unmyelinated and has a large surface area, and the spinal nerve roots contained within the dural root cuffs because they lack the mechanical and metabolic protection afforded peripheral nerves.22
Vascular injury within the neuraxis during regional anesthesia or pain medicine is also extremely rare and arguably more difficult to understand. Although hypotension and vasoconstrictors are often cited as the “diagnosis of exclusion” of unexplained perioperative spinal cord injury, neither factor seems to be supported by physiologic principles. Spinal cord blood flow (SCBF) is autoregulated within the same physiological range as the brain. Therefore, blood pressure within the normal autoregulatory range has no adverse effect on SCBF. There are no human data to support diminution of SCBF from vasoconstrictors such as epinephrine.23 Spinal cord blood supply can be tenuous, particularly to the anterior cord and the lower thoracic and lumbosacral segments. Although mechanical damage to the arterius radicularis magnus could compromise spinal cord circulation, this vessel and the anterior spinal artery system are quite distant from needles used for neuraxial and perineuraxial pain procedures.22 However, transforaminal approaches may have a higher potential for encountering major spinal arteries as they traverse the intervertebral foramen.24
In summary, the pathophysiology of neuraxial injury involves various, and perhaps combined mechanical, ischemic, and neurotoxic insults. Although a definitive etiology is apparent in the case of hematoma or infection, the link between specific factors and most other neuraxial injuries are associative—rather than causative—in nature. Our inability to prospectively identify patients at risk suggests that most of these injuries are neither completely predictable nor preventable. Furthermore, serious injuries are known to occur in healthy patients who have received competent and standard care.
Pathophysiology of Peripheral Nerve Injury
Similar to neuraxial injury, peripheral nerve injury is commonly linked to trauma from needles or catheters, ischemia, drug toxicity, compression, or neural stretch. Central to prognosis is whether or not the axon is preserved. Neuropraxic lesions, which damage the myelin sheath but preserve the axon, are typically associated with compressive or stretch injuries and are perhaps more likely to resolve. Conversely, if the axon is completely disrupted, recovery is slower and more likely to be incomplete. The double crush theory25 and some clinical evidence16,26 suggest that patients with pre‐existing peripheral nerve injury—even if subclinical—are more likely to sustain further nerve damage if a second subclinical or obvious injury occurs.27
Permanent peripheral nerve injury is not completely preventable—even in healthy patients receiving competent, standard care. Nerve injury theory suggests that more than 1 insult is often necessary to cause damage. Peripheral nerve injury appears to require the breach of connective tissue barriers such as the perineurium that surrounds individual nerve fascicles and protects them from the external milieu. Disruption of the perineurium from needle or catheter trauma is remarkably difficult to accomplish. Animal studies and experience with ultrasound‐guided nerve localization has shown that nerves tend to move away from approaching needles. When nerves are impaled, the needle may pass harmlessly into or through connective tissue, which constitutes up to 70% of a nerve's cross‐sectional area.28 However, if the fascicle is penetrated, neurons are exposed to local anesthetics that can cause time‐ and concentration‐dependent injury. Vasoconstrictors play a role by limiting local anesthetic clearance and thereby enhancing the time‐dependent component of injury. Decreased neural blood flow from edema or mass effect can also potentiate cytotoxicity. In summary, peripheral nerve injury associated with regional anesthesia is likely caused by a combination of insults to the nerve's internal milieu. However, the exact sequence and relative importance of these insults remain unknown.29
Diagnosis and Treatment
Suspected compressive lesions of the neuraxis require rapid diagnosis and treatment. The likelihood of full or partial recovery in these circumstances rapidly diminishes as time to decompression approaches or exceeds 8 hours.1 In most cases, magnetic resonance imaging (MRI) is the preferred imaging modality for spinal canal pathology, although diagnosis should not be delayed if only computed tomography (CT) is available.
Diagnosis of suspected peripheral nerve injury is guided by presenting symptoms, history, and physical examination. Complete or progressive neural deficits should prompt urgent evaluation by a neurologist or peripheral nerve surgeon. Mild and/or resolving symptoms without objective evidence of neural deficit typically indicate excellent prognosis and require only patient reassurance. If symptoms fail to progressively improve, neurological consultation should be sought in 2 to 3 weeks. Incomplete lesions with evidence of moderate or severe deficit are an indication for early neurological consultation and consideration of neurophysiologic testing (nerve conduction studies and electromyography) or MRI imaging of the nerve(s). Neurophysiologic testing can help quantify nerve damage and, together with MRI, can establish injury location. Although neurophysiologic changes are most apparent 14 to 21 days after injury, earlier testing may be indicated to rule out pre‐existing (including bilateral) disease, establish baseline, and aid in prognosis. After initial evaluation, incomplete and unresolved lesions should be followed up in 3 to 5 months. Improvement and evidence of reinnervation suggest further conservative management. However, no improvement or reinnervation might prompt referral to a peripheral nerve surgeon, although there is no consensus with regards to when an injured nerve warrants surgical exploration.27
Recommendations of the Practice Advisory Panel
As previously noted, the following recommendations are based on existing scientific literature and expert opinion, and tend toward conservative interpretation of this information. They are intended to promote understanding of neurologic complications associated with regional anesthesia and pain medicine. However, the recommendations cannot ensure the absence of adverse outcomes nor should they supersede physician judgment in specific clinical situations. The strength of each recommendation is stratified according to the evidence‐based grading scheme described in Table 1.
Neuraxial injuries may be limited by possessing a detailed knowledge of those anatomic conditions that place the spinal cord, spinal nerves, or spinal vasculature at risk for mechanically induced injury from needles, catheters, improper positioning, space‐occupying lesions, or drug‐induced neurotoxicity. When neurologic injury is suspected, diagnosis and treatment must proceed without delay to avoid adverse or incomplete recovery. Recommendations are listed in Table 2.
Although peripheral nerve injuries may be reduced by minimizing trauma to neural fibers, no nerve localization or monitoring technique has been shown to be clearly superior in terms of reducing the frequency of clinical injury. These techniques include paresthesia‐seeking, peripheral nerve stimulation, defined minimal or maximal milliamperage for acceptance of a motor response,30,31 ultrasound guidance, or monitoring of injection pressures.31‐35 Ultrasound guidance, advances in peripheral nerve stimulation,36 and pressure monitoring in particular are relatively new technologies; therefore, clinical experience and scientific study regarding their potential role in injury avoidance are limited.37 In patients with pre‐existing clinical or suspected subclinical peripheral nerve injury, consideration should be given to modifying the regional technique to minimize the introduction of additional potential neural insults. Reducing the dose, concentration, or potency of local anesthetic, or eliminating or reducing the concentration of vasoconstrictive additives are potential considerations, based on moderately extensive animal data.17,23 Limited human data neither support nor refute this recommendation. The diagnosis and prognosis of peripheral nerve injury largely depends on the severity of symptoms upon presentation. Recommendations are listed in Table 3.
Regional Anesthesia and the Patient With Pre‐Existing Neurologic Disease
Pre‐existing disorders of the central nervous system (CNS) and peripheral nervous system, such as multiple sclerosis, amyotrophic lateral sclerosis, and postpolio syndrome, present challenges to patients and anesthesiologists who desire to use neuraxial anesthetic techniques. Because these diseases involve damage to the CNS, the concern is that perioperative stress or further mechanical trauma, or drug‐induced toxicity could exacerbate the underlying disease process or lead to a relapse of symptoms. The rarity of these diseases results in a paucity of data that are often conflicting. For instance, an isolated case report describes a patient with postpolio syndrome who did not develop progression of his neurologic symptoms after spinal anesthesia.38 Small case series note an increase in the relapse rate of multiple sclerosis after spinal anesthesia in surgical patients39 vs. a statistically similar incidence of relapse rates in women who received neuraxial anesthesia and analgesia for obstetric indications.40 A retrospective investigation of 139 patients with pre‐existing CNS disease (multiple sclerosis, postpolio syndrome, traumatic spinal cord injury, etc.) who underwent neuraxial anesthesia revealed no subsequent exacerbation of disease process (0.0%‐0.3%, 95% CI).41 Another retrospective study of neuraxial anesthesia in patients with pre‐existing sensorimotor neuropathy or diabetic polyneuropathy noted the onset of new or progressive neurologic deficit in 0.4% (0.1% to 1.3%, 95% CI) of patients—an incidence that is higher than expected within the general population.42 However, the existing literature neither confirms nor refutes the safety of neuraxial anesthesia in patients with CNS or peripheral nervous system neurologic disorders, nor does it definitively address the relative safety of spinal vs epidural anesthesia or analgesia in these patients. Recommendations regarding patients with pre‐existing CNS disorders are listed in Table 4.
Peripheral nerve block may also be problematic in patients with pre‐existing clinical or suspected subclinical neurologic disease from diabetes mellitus, neurotoxic chemotherapeutic agents (e.g., vincristine or cisplatin), or in patients who are scheduled to undergo surgery on a nerve within the expected distribution of neural block. Although wide clinical experience suggests that neural block in these patients rarely exacerbates pre‐existing nerve injury, animal data and isolated human reports give some cause for concern. For instance, diabetic rats are more prone to the neurotoxic effects of local anesthetic.43 Patients undergoing ulnar nerve transposition are at no increased risk of postoperative neural deficit regardless of whether they receive general or regional anesthesia. However, neural deficits that did occur within the regional anesthesia group were associated with either an ulnar nerve paresthesia or motor response during block placement.44 In addition, 2 case reports describe patients with presumed subclinical neuropathy from cisplatin26 and multiple sclerosis37,45 who developed a new neural injury after uncomplicated brachial plexus block. Similar to CNS disease, the literature is unable to definitively support or refute the use of regional anesthesia in patients with pre‐existing peripheral neuropathy. Recommendations for these patients are listed in Table 4.
Patients with a history of prior spinal surgery present unique technical challenges for neuraxial regional techniques. Various spine stabilization and fusion procedures can involve autologous bone grafts, screws placed into pedicles, and anterior or posterior instrumentation. Most instrumentation does not alter the architectural integrity of the posterior spinal elements. Therefore, neuraxial anesthesia or pain medicine procedures are frequently possible in these patients.46 Plain radiographs or fluoroscopy can be beneficial in planning or executing neuraxial techniques within these patients. Furthermore, imaging may decrease the frequency of multiple attempts at block placement or unsuccessful identification of the epidural space.47 Because of surgically induced epidural space scarring, local anesthetic spread may be abnormal in these patients; thus spinal anesthesia may be the preferred neuraxial technique when appropriate.48 Prior lumbar spine surgery may also increase the risk of paraplegia associated with transforaminal epidural steroid injections (Table 4).49,50
In summary, existing scientific literature and expert opinion can neither confirm nor refute an adverse effect of regional anesthesia in patients with pre‐existing neurologic disease. However, most existing data suggest that any increased risk, if present, is likely of minimal magnitude. More importantly, patients with pre‐existing neurologic disease may develop new or progressive deficits during the perioperative period independent of anesthetic choice — from surgical factors, perioperative stress, or their underlying comorbidity.
Regional Anesthesia in Anesthetized or Heavily Sedated Patients
One of the most controversial areas of regional anesthesia and pain medicine practice is whether or not it is advisable to perform blocks in patients that are either anesthetized (under general anesthesia) or sedated to the point of being unable to recognize and/or report any sensation that the physician would interpret as atypical during block placement (hereafter referred to as heavy sedation). This seemingly simple question is indeed tremendously complex. Central to its answer is whether or not the subjective sensations that herald systemic local anesthetic toxicity or impending neural injury are specific and/or sensitive enough to constitute reliable, consistent warning signs. If these symptoms are indeed reliable, do they substantively differ depending on the patient's degree of wakefulness? Finally, should recommendations regarding block placement in anesthetized or heavily sedated patients differ based on patient age, block site, or risk‐to‐benefit ratio?
Avoiding systemic (cardiac or central nervous system) local anesthetic toxicity is predicated on early detection of intravascular injection of local anesthetic. This is accomplished using objective measures such as test doses or the awareness of subjective symptoms of local anesthetic toxicity—including auditory enhancement, circumoral numbness, or metallic taste. Seizures after local anesthetic injection into arteries directly supplying the brain occur too rapidly for any warning to be effective. However, seizures or cardiac signs consequent to tissue absorption typically happen after detection measures are complete. Because subjective signs of local anesthetic toxicity are unreliable even in mildly sedated patients,51 only imperfect objective measures such as epinephrine‐containing test doses52 can potentially detect intravenous local anesthetic injection, regardless of patient wakefulness. Furthermore, anesthetic and sedative agents can lower the risk of systemic local anesthetic toxicity by raising seizure threshold and altering untoward hemodynamic manifestations of local anesthetic toxicity (Table 5). 53
Patient wakefulness as a factor related to neuraxial or peripheral nerve injury is more complicated. Proponents for only performing blocks in patients that are mildly to moderately sedated argue that such patients are able to recognize and communicate to the anesthesiologist worrisome sensations or pain, which in turn may modify the anesthetic procedure and prevent or lessen the severity of injury. Proponents for performing blocks in anesthetized or heavily sedated patients argue that this practice promotes the benefits of regional anesthesia to a wider range of patients who may otherwise refuse the procedure because of anxiety, or may reduce the likelihood of unintentional movement during procedures near critical structures. The latter point is debatable, because disinhibited patients may also move at inopportune times. Furthermore, these proponents note that neural injury has been reported in both anesthetized and awake patients. There is no scientifically valid answer to this question; randomized controlled trials are unlikely because of the large numbers of patients that would be required for statistical validity. Existing animal and human data on this topic suggest the following: (1) the experience of a paresthesia or pain on injection during block placement is neither sensitive nor specific for nerve injury, even though it may be sensitive for needle‐to‐nerve proximity; (2) neuraxial and peripheral nerve injury has been reported both in awake patients who experienced no atypical sensation, and in those who experienced severe paresthesia or pain on injection (with or without subsequent injection of local anesthetic); and (3) awake patients most commonly suffer no injury regardless of the presence or absence of atypical sensation during nerve block placement. Thus, the presence of general anesthesia or heavy sedation eliminates patient recognition and/or report of abnormal sensation, but may or may not impact the actual occurrence of injury.
While acknowledging the conflicting literature relevant to this topic, the Panel's conservative opinion is that general anesthesia or heavy sedation removes all opportunity for the patient and/or physician to recognize, report, and respond to an atypical symptom during block placement. This opinion is in part based on a major unknown in anesthesia‐related nerve injury—how often, when a patient reports an atypical sensation or the anesthesiologist perceives something atypical during block performance, does the anesthesiologist consciously or unconsciously modify technique, possibly avoiding injury? Therefore, the weight of the Panel's opinion is that regional anesthetic or pain blocks should not be performed in adults with concurrent general anesthesia or heavy sedation except in those circumstances when the physician and patient conclude that benefit clearly outweighs risk. The Panel further acknowledges that the risk of injuring an anesthetized, heavily sedated, or awake patient may differ as a function of block site. For instance, most reports of injury involve interscalene block in anesthetized or heavily sedated patients, thus the Panel makes a separate recommendation specific to interscalene block (Table 5). However, there are relatively fewer reports of injury to lower extremity nerves in anesthetized or heavily sedated patients.54 Whether this variation in reported injury is a function of relative block frequency, practice patterns, or reflects an inherent difference in risk is unknown. Finally, data are limited with regards to the risk, if any, of performing peripheral nerve blocks in the setting of an incomplete proximal peripheral nerve or plexus block, or an unresolved neuraxial block. However, several studies have linked selective nerve “rescue blocks” to peripheral nerve injury.55,56
There is no localization or monitoring device, or approach to the neuraxis, that ensures protection from nerve injury. Some studies have compared paresthesia with peripheral nerve stimulation as indicators of needle‐to‐nerve contact,31,32 while another has compared paresthesia or peripheral nerve stimulation with ultrasound‐guided nerve localization.33 These studies clearly illustrate the differences between localization modalities, but fail to determine superiority. Importantly, none of these observations have been clinically linked to fewer neural injuries. The use of a peripheral nerve stimulator does not prevent nerve injury in anesthetized patients.57 There are no data to establish the safety of performing ultrasound‐guided blocks in anesthetized or heavily sedated patients. Similarly, recent data suggest that high pressures during the injection of intrafascicular local anesthetic is frequently, but not always, associated with neural injury in dogs,34 and methods have been proposed to avoid high injection pressure.35 In part because ultrasound and pressure monitoring are relatively new technologies, there are no human data to confirm or refute these findings. Finally, there is no clinical evidence in nonanesthetized patients, that approaching the epidural space at the lumbar level is more or less safe than at the thoracic level;58,59 therefore, there is no reason to assume that either approach is inherently safer in the anesthetized or heavily sedated patient.
Recognizing the absence of incontrovertible clinical data and the conflicting nature of case reports and case series regarding neural injury as a function of patient wakefulness,53 the Practice Advisory's recommendations are listed in Table 5.
Performing regional anesthesia or pain procedures in anesthetized or heavily sedated pediatric patients represents a distinct subset of recommendations, which in part recognizes that this represents standard practice for many anesthesiologists. The Panel acknowledges that there are no data to confirm or refute the concept that performing blocks in anesthetized or heavily sedated pediatric patients is inherently riskier or safer than doing so in adults. Particularly with infants and younger children, the ability to recognize and verbalize specific block‐related sensations may be absent regardless of wakefulness. Conversely, a struggling child may well increase the risk of an untoward event during needle placement. Thus, the Practice Advisory recommends that performing neuraxial and peripheral nerve block in anesthetized or heavily sedated pediatric patients may constitute an acceptable benefit‐to‐risk ratio (Table 5). Although ultrasound can accurately determine the skin‐to‐ligamentum flavum distance in infants and children,60 there is no clinical evidence that such preprocedural measurements will increase safety.
Transforaminal Injection of Steroids
Transforaminal injection of steroids, often used in the treatment of acute radicular pain, has been linked to cases of spinal cord infarction, cortical blindness, paralysis, and death.24,50,61 The presumed mechanism of these complications involves unintentional needle entry into a small artery that traverses the intervertebral foramen to join the arterial supply to the spinal cord or posterior circulation of the brain. This can occur at various levels, including the vertebral artery anterior to the cervical intervertebral foramina, or the spinal medullary or radicular arteries within the foramina at variable levels within the cervical,62,63 thoracic, lumbar, and sacral portions of the spine. Subsequent injection of particulate steroid preparations can result in occlusion of the distal arterioles within the spinal cord or brain and lead to infarction.61 In vitro studies note that methylprednisolone has the largest particles, betamethasone the smallest, and dexamethasone has no particulate matter.64 While evidence of unintended injection into perispinal vessels during transforaminal injection has been reported, direct evidence to confirm or refute the role of particulate steroids in causing subsequent neuronal injury is lacking (Table 6).
Neurologic complications associated with regional anesthesia and pain medicine are rare—particularly those complications that do not involve hematoma or infection. Understanding the pathophysiology and risk factors associated with neuraxial and peripheral nerve injury may allow anesthesiologists to minimize the number of adverse neurologic outcomes. Unfortunately, even with flawless care of otherwise healthy patients by well trained physicians, these complications are neither completely predictable nor preventable. This Practice Advisory offers a number of recommendations specific to common clinical scenarios encountered in everyday practice.
1. Horlocker TT, Wedel DJ, Benzon H, Brown DL, Enneking FK, Heit JA, Mulroy MF, Rosenquist RW, Rowlingson J, Tryba M, Yuan C-S. Regional anesthesia in the anticoagulated patient: Defining the risks (The second ASRA consensus conference on neuraxial anesthesia and anticoagulation). Reg Anesth Pain Med 2003;28:172-197.
2. Hebl J. Importance and implications of aseptic techniques during regional anesthesia. Reg Anesth Pain Med 2006;31:311-323.
3. Hebl JR, Neal JM. Infectious complications: A new practice advisory. Reg Anesth Pain Med 2006;31:289-290.
4. Horlocker T. Regional anesthesia and the immunocompromised patient. Reg Anesth Pain Med 2006;31:334-345.
5. Rathmell JP. Infectious risks of chronic pain treatments. Reg Anesth Pain Med 2006;31:346-352.
6. Wedel DJ. Regional anesthesia in the febrile or infected patient. Reg Anesth Pain Med 2006;31:324-333.
7. Research United States Department of Health and Human Services Agency for Health Care Policy and Research. Acute pain management: Operative or medical procedures and trauma. The Agency 1993; Clinical Practice Guideline No. 1; No. 92-0023-0107.
8. Eagle KA, Berger PB, Calkins H, Chaitman BR, Ewy GA, Fleischmann KE, Fleisher LA, Froehlich JB, Gusberg RJ, Leppo JA, Ryan T, Schlant RC, Winters WL Jr, Gibbons RJ, Antman EM, Alpert JS, Faxon DP, Fuster V, Gregoratos G, Jacobs AK, Hiratzka LF, Russell RO, Smith SC Jr, American College of Cardiology, American Heart Association. ACC/AHA guideline update for perioperative cardiovascular evaluation for noncardiac surgery–executive summary: A report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee to Update the 1996 Guidelines on Perioperative Cardiovascular Evaluation for Noncardiac Surgery) [Erratum in 2006;47:2356]. J Am Coll Cardiol 2002;39:542-553.
9. Brull R, McCartney CJL, Chan VWS, Liguori GA, Hargettt MJ, Xu D, Abbas S, El-Beheiry H. Disclosure of risks associated with regional anesthesia: A survey of academic regional anesthesiologists. Reg Anesth Pain Med 2007;32:7-11.
10. Domino KB. Informed consent for regional anesthesia: What is necessary?. Reg Anesth Pain Med 2007;32:1-2.
11. Moen V, Dahlgren N, Irestedt L. Severe neurological complications after central neuraxial blockades in Sweden 1990-1999. Anesthesiology 2004;101:950-959.
12. Auroy Y, Narchi P, Messiah A, Litt L, Rouvier B, Samii K. Serious complications related to regional anesthesia. Results of a prospective survey in France. Anesthesiology 1997;87:479-486.
13. Wang LP, Hauerberg J, Schmidt JR. Incidence of spinal epidural abscess after epidural analgesia. Anesthesiology 1999;91:1928-1936.
14. Lee LA, Posner KL, Domino KB, Caplan RA, Cheney FW. Injuries associated with regional anesthesia in the 1980s and 1990s. Anesthesiology 2004;101:143-152.
15. Liguori GA. Complications of regional anesthesia: Nerve injury and peripheral nerve blockade. J Neurosurg Anesthesiol 2004;16:84-86.
16. Borgeat A, Ekatodramis G, Kalberer F, Benz C. Acute and nonacute complications associated with interscalene block and shoulder surgery. A prospective study. Anesth Analg 2001;95:875-880.
17. Neal JM, Hebl JR, Gerancher JC, Hogan QH. Brachial plexus anesthesia: Essentials of our current understanding. Reg Anesth Pain Med 2002;27:402-428.
18. Auroy Y, Benhamou D, Bargues L, Ecoffey C, Falissard B, Mercier F, Bouaziz H, Samii K. Major complications of regional anesthesia in France. The SOS regional anesthesia hotline service. Anesthesiology 2002;97:1274-1280.
19. Fitzgibbon DR, Posner KL, Domino KB, Caplan RA, Lee LA, Cheney FW. Chronic pain management: American Society of Anesthesiologists Closed Claims Project. Anesthesiology 2004;100:98-105.
20. Lee LA, Caplan RA, Domino KB. The ASA Closed Claims Project: Regional Anesthesia and Pain Medicine. In: Neal J.M, Rathmell J.P, eds. Complications in Regional Anesthesia and Pain Medicine. Philadelphia: Elsevier; 2007:343-352.
21. Lee L, Posner KL, Chaney FW, Caplan RA, Domino KB. Complications associated with eye blocks and peripheral nerve blocks: An ASA closed claims analysis. Reg Anesth Pain Med 2008;33:416-422.
22. Neal JM. Anatomy and pathophysiology of spinal cord injuries associated with regional anesthesia and pain medicine. Reg Anesth Pain Med 2008;33:423-434.
23. Neal JM. Effects of epinephrine in local anesthetics on the central and peripheral nervous systems: Neurotoxicity and neural blood flow. Reg Anesth Pain Med 2003;28:124-134.
24. Rathmell JP, Benzon HT. Transforaminal injection of steriods: Should we continue?. Reg Anesth Pain Med 2004;29:397-399.
25. Upton AR, McComas AJ. The double crush in nerve entrapment syndromes. Lancet 1973;2:359-362.
26. Hebl JR, Horlocker TT, Pritchard DJ. Diffuse brachial plexopathy after interscalene blockade in a patient receiving cisplatin chemotherapy: The pharmacologic double crush syndrome. Anesth Analg 2001;92:249-251.
27. Sorenson EJ. Neurological injuries associated with regional anesthesia. Reg Anesth Pain Med 2008;33:442-448.
28. Moayeri N, Bigeleisen PE, Groe GJ. Quantitative architecture of the brachial plexus and surrounding compartments, and their possible significance for plexus blocks. Anesthesiology 2008;108:299-304.
29. Hogan QH. Pathophysiology of peripheral nerve injury during regional anesthesia. Reg Anesth Pain Med 2008;33:435-441.
30. Karaca P, Hadzic A, Yufa M, Vloka JD, Brown AR, Visan A, Sanborn K, Santos AC. Painful paresthesiae are infrequent during brachial plexus localization using low-current peripheral nerve stimulation. Reg Anesth Pain Med 2003;28:380-383.
31. Urmey WF, Stanton J. Inability to consistently elicit a motor response following sensory paresthesia during interscalene block administration. Anesthesiology 2002;96:552-554.
32. Choyce A, Chan VWS, Middleton WJ, Knight PR, Peng P, McCartney CJL. What is the relationship between paresthesia and nerve stimulation for axillary brachial plexus block?. Reg Anesth Pain Med 2001;26:100-104.
33. Perlas A, Niazi A, McCartney C, Chan V, Xu D, Abbas S. The sensitivity of motor responses to nerve stimulation and paresthesia for nerve localization as evaluated by ultrasound. Reg Anesth Pain Med 2006;31:445-450.
34. Hadzic A, Dilberovic F, Shah S, Kulenovic A, Kapur E, Zaciragic A, Cosovic E, Vuckovic I, Divanovic KA, Mornjakovic Z, Thys DM, Santos AC. Combination of intraneural injection and high injection pressure leads to fascicular injury and neurologic deficits in dogs. Reg Anesth Pain Med 2004;29:417-423.
35. Tsui BC, Knezevich MP, Pillay JJ. Reduced injection pressures using a compressed air injection technique (CAIT): An in vitro study. Reg Anesth Pain Med 2008;33:168-173.
36. Urmey WF, Grossi P. Use of sequential electrical nerve stimuli (SENS) for location of the sciatic nerve and lumbar plexus. Reg Anesth Pain Med 2006;31:463-469.
37. Hebl JR. Ultrasound-guided regional anesthesia and the prevention of neurologic injury: Fact or fiction?. Anesthesiology 2008;108:186-188.
38. Higashizawa T, Sugiura J, Takasugi Y. Spinal anesthesia in a patient with hemiparesis after poliomyelitis [in Japanese]. Masui 2003;52:1335-1337.
39. Bamford C, Sibley W, Laguna J. Anesthesia in multiple sclerosis. Can J Neurol Sci 1978;5:41-44.
40. Bader AM, Hunt CO, Datta S, Naulty JS, Ostheimer GW. Anesthesia for the obstetric patient with multiple sclerosis. J Clin Anesth 1988;1:21-24.
41. Hebl JR, Horlocker TT, Schroder DR. Neuraxial anesthesia and analgesia in patients with preexisting central nervous system disorders. Anesth Analg 2006;103:223-228.
42. Hebl JR, Kopp SL, Schroder DR, Horlocker TT. Neurologic complications after neuraxial anesthesia or analgesia in patients with preexisting peripheral sensorimotor neuropathy or diabetic polyneuropathy. Anesth Analg 2006;103:1294-1299.
43. Kalichman MW, Calcutt NA. Local anesthetic-induced conduction block and nerve fiber injury in streptozotocin-diabetic rats. Anesthesiology 1992;77:941-947.
44. Hebl JR, Horlocker TT, Sorenson EJ, Schroeder DR. Regional anesthesia does not increase the risk of postoperative neuropathy in patients undergoing ulnar nerve transposition. Anesth Analg 2001;93:1606-1611.
45. Koff MD, Cohen JA, McIntyre JJ, Carr CF, Sites BD. Severe brachial plexopathy after an ultrasound-guided single-injection nerve block for total shoulder arthroplasty in a patient with multiple sclerosis. Anesthesiology 2008;108:325-328.
46. Viscomi CM, Rathmell JP, Monsey RD. Radiographic appearance of the lumbar spine after lumbar fusion. Reg Anesth Pain Med 2000;25:286-290.
47. Calleja M. Extradural analgesia and previous spinal surgery. Anaesthesia 1991;46:946-947.
48. Sharrock N, Urqhart B, Mineo R. Extradural anaesthesia in patients with previous lumbar spine surgery. Br J Anaesth 1990;65:237-239.
49. Houten JK, Errico TJ. Paraplegia after lumbosacral nerve root block: Report of three cases. Spine J 2002;2:70-75.
50. Huntoon MA, Martin DP. Paralysis after transforaminal epidural injection and previous spinal surgery. Reg Anesth Pain Med 2004;29:494-495.
51. Moore JM, Liu SS, Neal JM. Premedication with fentanyl and midazolam decreases the reliability of intravenous lidocaine test dose. Anesth Analg 1998;86:1015-1017.
52. Mulroy MF, Norris MC, Liu SS. Safety steps for epidural injection of local anesthetics: Review of the literature and recommendations. Anesth Analg 1997;85:1346-1356.
53. Bernards CM, Hadzic A, Suresh S, Neal JM. Regional anesthesia in anesthetized or heavily sedated patients. Reg Anesth Pain Med 2008;33:449-460.
54. Capdevila X, Pirat P, Bringuier S, Gaertner E, Singelyn FJ, Bernard N, Choquet O, Bouaziz H, Bonnet F. Continuous peripheral nerve blocks in hospital wards after orthopedic surgery. Anesthesiology 2005;103:1035-1045.
55. Selander D, Edshage S, Wolff T. Paresthesiae or no paresthesiae?. Nerve lesions after axillary blocks. Acta Anaesthesiol Scand 1979;23:27-33.
56. Stan TC, Krantz MA, Solomon DL, Poulos JG, Chaouki K. The incidence of neurovascular complications following axillary brachial plexus block using a transarterial approach. Reg Anesth 1995;20:486-492.
57. Benumof JL. Permanent loss of cervical spinal cord function associated with interscalene block performed under general anesthesia. Anesthesiology 2000;93:1541-1544.
58. Giebler R, Scherer R, Peters J. Incidence of neurologic complications related to thoracic epidural catheterization. Anesthesiology 1997;86:55-63.
59. Tanaka K, Watanabe R, Harado T, Dan K. Extensive application of epidural anesthesia and analgesia in a university hospital: Incidence of complications related to technique. Reg Anesth 1993;18:34-38.
60. Kil HK, Cho JE, Kim WO, Koo BN, Han SW, Kim JY. Prepuncture ultrasound-measured distance: An accurate reflection of epidural depth in infants and children. Reg Anesth Pain Med 2007;32:102-106.
61. Rathmell JP, April C, Bogduk N. Cervical transforaminal injection of steroids. Anesthesiology 2004;100:1595-1600.
62. Huntoon MA. Anatomy of the cervical intervertebral foramina: Vulnerable arteries and ischemic neurologic injuries after transforaminal epidural injections. Pain 2005;117:104-111.
63. 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.
64. Benzon HT, Chew TL, McCarthy R, Benzon HA, Walega DR. Comparison of the particle sizes of the different steroids and the effect of dilution: A review of the relative neurotoxicities of the steroids. Anesthesiology 2007;106:331-338.
* Ritchie E. Berger, Esq., Burlington, VT. Cited Here...
Key Words:: Complications of anesthesia; Nerve injury; Spinal anesthesia; Epidural anesthesia; Peripheral nerve block; Regional anesthesia; Pain medicine; Transforaminal block
©2008 American Society of Regional Anesthesia and Pain Medicine
Highlight selected keywords in the article text.