Therapeutic Agents for Spine Injection

Published on 27/02/2015 by admin

Filed under Anesthesiology

Last modified 27/02/2015

Print this page

This article have been viewed 2311 times

Chapter 2 Therapeutic Agents for Spine Injection

Local Anesthetics, Steroids, and Contrast Media

Matthew J. Pingree, Marc A. Huntoon

Chapter Overview

Chapter Synopsis: Spinal and epidural injections may be performed for therapeutic, diagnostic, and imaging purposes. This chapter explores the evolution of these techniques and the injectable agents in particular. Historically, the local anesthetic procaine was the most commonly injected drug followed by a wave of corticosteroid injection that began in the 1960s in Europe. Fluoroscopic guidance with injected dyes followed. Among local anesthetics, amide-type drugs have become the preferred class, which includes lidocaine and bupivacaine. Metabolism of the ester-type anesthetics (including procaine) produces para-aminobenzoic acid as a byproduct and presents greater risk of adverse allergic reaction. Local anesthetic systemic toxicity presents a separate serious complication risk. Glucocorticoids are useful in the control of pain because of their inhibitory action on inflammatory agents. Complication risks include Cushing’s syndrome—a suppression of the hypothalamic–pituitary–adrenal axis—and infarct from particulate steroids. Other more specific blockers of cytokines and other inflammatory mediators have also been investigated for epidural injection. Contrast media injected to aid in fluoroscopic imaging typically contain iodine. Adverse events are for the most part mild and easily treated. Anxiety can play a significant role in adverse reactions from injection of any substance.

Important Points:

Local anesthetics are rarely associated with immediate hypersensitivity (allergic) reactions, and the amide-type local anesthetics are particularly safe.

Local anesthetic systemic toxicity is an emergency that requires additional help and may be ameliorated with lipid emulsion therapy.

Corticosteroid agents for spinal injection have significant dose-related complications with repetitive use. Users should be aware that multiple practitioners may be using these agents, with no consistent dosing guidelines.

Particulate steroids have been associated with catastrophic spinal cord and brainstem infarcts, presumably from embolization into critical arteries, and caution suggests that nonparticulate agents may be more appropriate in specific procedures (e.g., cervical transforaminal epidurals).

Patients with iodine or shellfish allergies are not at greater risk for adverse reactions to contrast agents.

In cases of previous reaction to iodinated contrast, gadolinium is an acceptable alternative with extremely low incidence of adverse reactions.

Patients with previous anaphylactoid reactions, those with asthma, and certain food allergies may benefit from pretreatment with histamine receptor 1 and 2 antagonists and systemic steroids.

Clinical Pearls:

In patients who have had prior anaphylaxis due to contrast dyes, one might consider not injecting contrast agent if it is not critical to the outcome.

Because local anesthetic injection does not contribute to the long-term therapeutic outcome in most spinal injections, and may cause toxicity or fall risk, its use should be limited.

Doses of corticosteroid are not standardized, and thus dose selection should always be the lowest possible to achieve a therapeutic effect.

Clinical Pitfalls:

Practitioners need not ask about shellfish allergies, as the information has little clinical significance when choosing contrast agents.

Particulate corticosteroids should be avoided for most anterior and lateral cervical injections.

Introduction

Therapeutic spinal injections date back to the early part of the twentieth century before World War II. Over the past several decades, corticosteroids have emerged as the preferred class of spinally administered drugs for presumed inflammatory causes of pain. Common usage has been based largely on case series demonstrating efficacy and select double-blind placebo controlled trials (see Chapter 1). Surface landmark–based technique has also changed over time, with most authors currently endorsing very specific fluoroscopically controlled and contrast-proven injections. Despite these proscriptions, there have been no large-scale head-to-head trials that demonstrate outcome-based superiority of the more technologically advanced techniques. This chapter focuses on the agents that are commonly being used to perform therapeutic epidural and spinal injections, as well as the appropriate and safe use of contrast media (e.g., iohexol or gadodiamide agents) and some of the experimental agents being considered for future use.

History

One of the first reports of spinal injections for pain control was trans-sacral (caudal) injections of the local anesthetic procaine.¹ The first use of epidural corticosteroids came out of the European literature.² One of the first large series was published in 1961, in which Goebert and colleagues³ administered 121 injections to 113 patients spanning a 5-year period. The majority of these injections were caudal epidural injections with only three cervical epidural injections. Large-volume injections of a mixture of 1% procaine with 125 mg of hydrocortisone acetate in 30-mL volumes were administered over a consecutive 3-day period.³ Subsequently, in the modern era after the 1970s, interlaminar epidural injections became standard. Winnie and colleagues⁴ published recommendations that are still popular today, including corticosteroid dose limitations, 2-week dosing intervals, and three epidural procedures in series. Later, many physicians adopted a transforaminal fluoroscopically guided approach.

Local Anesthetics

Local anesthetics block voltage-gated sodium channels and interrupt propagation of axonal impulses, but their action is not only limited to those biological actions. There are two classes of local anesthetics based on structure–activity relationships: amino esters and amino amides. Because of the popularity and much more common use of amide-type local anesthetics for spinal diagnostic and therapeutic procedures, the focus of this review is mainly on the amide-type local anesthetics lidocaine and bupivacaine.^5,⁶

Important properties of local anesthetics that pertain to clinical use include potency, speed of onset, and duration of action. The potency of a local anesthetic is related to its lipid solubility, which is usually defined by the octanol-buffer coefficient. The molecule must diffuse into the nerve membrane and bind at a partially hydrophobic site on the sodium channel.⁷ The more hydrophobic or lipophilic the local anesthetic, the more quickly it will permeate neuronal membranes, which increases its sodium channel binding affinity. Bupivacaine, for example, is many times more lipophilic than lidocaine (Table 2-1).

Table 2-1 Effects of pKa and Hydrophobicity on Local Anesthetic Action

The speed of onset of most local anesthetics directly relates to the dissociation constant, or pKa, of the compound as well as the pH of the local tissues. The pKa is the pH at which half of the compound is ionized or protonated; the other half is in the un-ionized or neutral form that more readily crosses the nerve membrane. This makes the local anesthetic with the pKa that is closest to physiological pKa of faster onset. The pH of the local anesthetic preparation also affects the onset time, and some commercially available preparations containing a vasoconstrictor (e.g., epinephrine) have an adjusted pH that is acidic because of the addition of hydrochloric salts, enhancing the stability of the vasoconstrictor (Table 2-2).^8,⁹ In vivo, other factors such as dose or concentration can also affect the onset of action.⁷ If faster onset is desired, then the addition of a small amount of sodium bicarbonate (1 : 20 NaHCO₃-to-anesthetic volume) can help adjust the pH closer to physiological conditions. Caution should be taken not to adjust the pH greater than 7 because of the possibility that drug precipitation will increase.

Table 2-2 Pharmacology of Selected Amide Local Anesthetics

Defining the duration of action is somewhat more difficult because it depends on multiple variables such as the location of injection, the lipophilicity of the local anesthetic, the dose, and the presence or absence of a vasoconstrictor. Longer acting local anesthetics are more lipophilic and are more slowly “washed out” from the lipophilic membrane.¹⁰ In humans, the peripheral vascular effects of the local anesthetics themselves also affect duration. Many agents have a biphasic effect on vascular smooth muscle with a vasoconstrictive response at low concentrations and vasodilatation at higher concentrations. These effects are complex and vary according to the concentration, time, and location of injection.⁷ In general, the more vascular the location, the more rapidly the agent is absorbed, metabolized, and excreted.

Metabolism of local anesthetics, not surprisingly, is dependent on its amide or ester structure. Ester type agents undergo rapid metabolism via plasma pseudocholinesterase, of which para-aminobenzoic acid (PABA) is a by-product. The ester type local anesthetics include procaine and benzocaine, which are two of the more commonly used agents. Conversely, amide-type agents such as lidocaine and bupivacaine are metabolized through the hepatic cytochrome P450 enzyme system as well as via conjugation. The byproduct of the ester-type metabolism, PABA, is thought to be an allergen involved in many local anesthetic allergic reactions. Moreover, given the dependence on the liver for the metabolism of amide-type local anesthetics, caution should be exercised when using these agents in patients with liver dysfunction.

Adverse Reactions

Reactions involving the use of local anesthetics can result from a number of different sources, including toxicity from the medication itself, a reaction to a preservative or added vasoconstrictor, or even an allergic reaction. Outside of the discussion of PABA as a potential allergen, immediate hypersensitivity reactions to amide local anesthetics and their pathomechanism remain largely unidentified. Clinically, the allergic response corresponds with anaphylaxis (manifesting as tachycardia; hypotension; and feelings of weakness, heat, or vertigo) even though immunologically mediated reactions have rarely been observed. Other ingredients in the local anesthetic commercial preparations such as certain preservatives or even anxiety need to be considered as potential sources of the adverse reaction as well.¹¹

Toxicity, especially systemic toxicity, is usually the result of inadvertent intravascular injection, overdosage, or increased uptake from the area of injection. Systemic toxicity is a significant source of morbidity and mortality in the practice of anesthesia and especially the subspecialty regional anesthesia. Recently, the American Society of Regional Anesthesia and Pain Medicine published a practice advisory dealing with local anesthetic systemic toxicity (LAST). LAST was first recognized in the 1880s with the introduction of cocaine into clinical practice. Today, there is a focus on the development of less cardiotoxic local anesthetics based on structural changes or chirality (e.g., ropivacaine or levobupivacaine) as well as new and improved treatment for the cardiovascular effects of LAST with lipid emulsion therapy.

Epidemiological studies report a range of incidence of LAST that varies from 0 to 79 per 10,000.¹²^–¹⁴ In general, the cardiac toxicity results from the binding and inhibition of sodium channels, however, and correlates with local anesthetic potency especially in regards to inhibition of cardiac conduction. In addition, there is a vast array of other inotropic and metabolic signaling systems as well as mitochondrial metabolism implicated as potential targets for local anesthetics that would help explain the local anesthetic agents variable role in LAST.¹⁵

One way to reduce the incidence of LAST is to prevent it. Unfortunately, there is no single intervention that has proven to prevent this potentially life-threatening response. Obviously, using the least amount of local anesthetic required, incremental injection, frequent aspiration looking for the return of blood, using an intravascular marker such as epinephrine, and the use of ultrasound guidance are all recommended to try to prevent LAST.¹⁵

Classically, LAST presents in a predictable sequence with subjective central nervous system (CNS) symptoms of auditory changes, circumoral numbness, and metallic taste. Signs can then progress to seizures, coma, respiratory arrest, and ultimately cardiac toxicity that include cardiac excitation followed by cardiac depression at greatly increased blood concentrations. Unfortunately, in reality, the presentation can be extremely variable and is atypical in about 40% of cases with LAST. Even with an atypical presentation, the first symptoms presented in less than 5 minutes after injection 75% of the time. Seizure was the most common presenting symptom, and fewer than 20% of these seizures presented without any of the classic prodromal symptoms.¹⁵

Treatment of LAST, given the possible serious morbidity, should be quick and aggressive. Priorities include obtaining an airway, circulatory support, and diminishing the systemic effects as much as possible. Seizures should be rapidly treated with benzodiazepines if possible. Initiating the clinical algorithm as part of advanced cardiac life support (ACLS) is also important, although LAST presents a very different clinical scenario than that usually addressed by ACLS. The cause of circulatory arrest in LAST means that vasopressin and epinephrine may have less of a role or not be recommended. In fact, animal studies indicate that lipid emulsion treatment had better outcomes than both epinephrine and vasopressin.^15,¹⁶ (Please refer to the Practice Advisory sheets included in Table 2-3.)

Table 2-3 Practice Advisory on Treatment of Local Anesthetic Systemic Toxicity for Patients Experiencing Signs or Symptoms of Local Anesthetic Systemic Toxicity

• Get help

• Initial focus

• Airway management: ventilate with 100% oxygen

• Seizure suppression: benzodiazepines are preferred

• Basic and advanced cardiac life support (BLS/ACLS) may require prolonged effect

• Infuse 20% lipid emulsion (values in parenthesis are for a 70-kg patient)

• Bolus 1.5 mL/kg (lean body mass) IV over 1 min (∼100 mL)

• Continuous infusion at 1.25 mL/kg/min (≈18 mL/min; adjust by roller clamp)

• Repeat bolus once or twice for persistent cardiovascular collapse

• Double the infusion rate to 0.5 mL/kg/min if blood pressure remains low

• Continue infusion for at least 10 min after attaining circulatory stability

• Avoid vasopressin, calcium channel blockers, β-blockers, and local anesthetics

• Alert the nearest facility with cardiopulmonary bypass capability

• Avoid propofol in patients with signs of cardiovascular instability

• Post the LAST event at www.lipidrescue.org and report use of lipid to www.lipidregistry.org

IV, intravenous; LAST, local anesthetic systemic toxicity.

Neal JM, Bernards CM, Butterworth JF IV, et al: ASRA practice advisory on local anesthetic systemic toxicity. Reg Anesth Pain Med 35:152-161, 2010.

Corticosteroids

Cortisone as a purified glucocorticoid preparation was first introduced in 1949; later, in 1952, its application was described for epidural use.¹⁷^–¹⁹ Since then, the use of steroids has been applied in the field of interventional pain management with varying degrees of success and complications.

Glucocorticoids, of which the injectable corticosteroids are a part, are produced in the zona fasciculate of the adrenal cortex and function under negative feedback from the hypothalamus and pituitary gland as part of the hypothalamus–pituitary–adrenal axis.²⁰ Glucocorticoids are used in interventional pain procedures because of their effects on inflammation. Glucocorticoids have significant inhibitory effects on cytokines and chemokines that are generated at sites of inflammation, as well as suppressive effects on leukocyte concentration, distribution, and function. Glucocorticoids have potential effects on most cells of the body through interactions with glucocorticoid receptors. Normally, intracellular glucocorticoid receptors are in a stabilized form coupled with two elements of heat-shock protein 90 (HSP90) and other proteins. Binding of glucocorticoid to its receptor allows the glucocorticoid to enter the cell where dissociation of the proteins occurs, and the glucocorticoid–receptor complex binds to the glucocorticoid response element of a target gene. Resultant transcription activity via RNA polymerase is thus altered, eventually leading to alterations in messenger RNA (mRNA) and new protein production, which leads to the hormonal response.²¹

Table 2-4 lists the antiinflammatory potency of some of the more commonly used neuraxial steroids.

Table 2-4 Antiinflammatory Potency of Commonly Used Neuraxial Steroids

Complications

Multiple complications of corticosteroids are possible and largely relate to unwanted side effects (e.g., iatrogenic Cushing’s syndrome). Tuel and colleagues ²² described a case of iatrogenic Cushing’s syndrome occurring in a 24-year-old man after a single dose of 60 mg of methylprednisolone. Laboratory evidence of suppression of the hypothalamic–pituitary axis, 20-lb weight gain, and cushingoid features (moon facies, stria) persisted for 12 months. In another report, two doses of 80 mg of methylprednisolone resulted in Cushing’s syndrome with peripheral edema, moon facies, a “buffalo hump,” and purpura in a 63-year-old woman.²³ These cases illustrate that doses that are well within the normal guidelines for epidural steroid injections may still result in adverse consequences.

Perhaps the most dreaded recent complications have been attributed to particulate steroids. A large survey of members of the American Pain Society ²⁴ identified 78 cases of either spinal cord or brainstem infarcts that were known of by those members that responded to the survey. Of these, 14 were fatal cases, and all involved particulate steroids. Tiso et al²⁵ and Benzon, et al²⁶ both studied the microscopic appearances of commonly used steroids. Although discrepancies exist between the two studies, they are useful to any discussion of the potential pathophysiology of injury. One notable difference is that dexamethasone was found to be nonparticulate in the study by Benzon et al²⁶ (Fig. 2-1). A subsequent animal study demonstrated that dexamethasone acted like a nonparticulate in that direct, intentional injection of the vertebral artery in a porcine model resulted in ischemic brain injury only in the animals receiving particulate steroids.²⁷