• Spinal cord stimulation trial and implant
• Interlaminar epidural steroid injections
• Peripheral nerve blocks
• Intrathecal catheter and pump implant
• Transforaminal epidural steroid injections
• Peripheral joint and musculoskeletal injections
• Vertebral augmentation (vertebroplasty and kyphoplasty)
• Facet medial branch nerve blocks and radiofrequency ablation
• Trigger point injections
• Epiduroscopy and epidural decompression
• Paravertebral blocks
• Piriformis muscle injections
• Intradiscal procedures
• Sacroiliac joint injection and sacral lateral branch blocks
• Sympathetic nerve blocks
• Peripheral nerve stimulation trials and implants
• Pocket revisions for IPG or ITP replacement
The guidelines writing committee recognized the additional risk attributable to advanced age, historical bleeding tendencies, advanced hepatic disease, and renal insufficiency. In the presence of one or more of these known risk factors, patients who would initially fall into a low-risk category should be advanced to an intermediate-risk category, and intermediate-risk patients should be advanced to the high-risk category . Surgical history should also be considered in providing a risk assessment since fibrous adhesions can develop after spine surgery thereby distorting the anatomy of the epidural vessels, potentially increasing the likelihood of bleeding complications due to development of scar tissue that compromises the capacity of the epidural space . For the sum of these variables, a comprehensive assessment of bleeding risk should account for both procedural and patient-specific variables , and necessitates physician judgment rather than rote categorization.
Increases in population longevity, increased surgical volume, the broader indications for the use of anticoagulant therapies, the increased utilization of regional anesthesia and interventional pain procedures for the treatment of both acute and chronic pain conditions place regional anesthesiologists and interventional pain physicians into a consultant role for anticoagulated patients in the perioperative and periprocedural periods. Though there is no substitute for clinician judgment, a broad understanding of the indications, mechanism of action, half-life, and general recommendations about the use of anticoagulants is warranted for each agent.
Nonsteroidal Anti-inflammatory Drugs
Nonsteroidal Anti-Inflammatory Drugs (NSAIDs) exert their analgesic effects by blunting prostaglandin production and inhibiting cyclooxygenase. Cyclooxygenase-1 (COX-1) is constitutively expressed while Cyclooxygenase-2 (COX-2) is induced in the presence of inflammation. Platelet function is affected when NSAIDs exert their effect on COX-1 preventing the formation of prostaglandin H2. Commonly utilized NSAIDs include salicylates, cyclooxygenase nonspecific agents, and COX-2 selective NSAIDs.
Aspirin is taken by more than 50 million Americans for the prevention of cardiovascular events and utilized considerably less often for pain in light of the advent of an expanding pharmacopeia of alternative analgesic agents . Aspirin has an irreversible affinity for COX-1 and blocks the production of thromboxane for the entire 7–10 day life span of a platelet . Platelet aggregation and thrombosis is inhibited within hours of aspirin administration . Support exists for the ongoing use of aspirin in patients to prevent recurrence of cardiovascular events, termed secondary prevention, and cessation of chronic salicylate use demonstrates an increased risk of cardiovascular events in these subjects [23–25]. This is attributed to a “rebound phenomenon” wherein thromboxane production increases and fibrinolysis decreases when chronic aspirin therapy is discontinued. Because 10 % of the platelet pool is replenished each day, complete restoration of platelet function does not occur until 10 days after cessation of aspirin therapy, though significant variability has been observed among clinical subsets. While low-dose aspirin was shown to increase the incidence of bleeding by a factor of 1.5, it did not lead to adverse clinical outcomes outside of intracranial surgery in a large review and meta-analysis . Nonetheless, isolated case reports and case series implicate aspirin in the development of bleeding complications associated with spinal cord stimulator lead placement, removal, or spinal anesthetic injections [27–30].
Patient-related and procedural risk stratification should occur prior to modifying chronic salicylate therapy in light of the prothrombotic rebound phenomenon. Routine discontinuation of aspirin should be avoided for procedures characterized as low risk (Table 8.1). A shared assessment and risk stratification should occur for intermediate-risk procedures, especially when known vascular anomalies occur in close proximity to the target, such as a variant course of the vertebral artery seen on cervical imaging prior to performing a stellate ganglion block. For intermediate-risk procedures in patients where aspirin is prescribed for secondary prevention, discontinuation of chronic salicylate therapy should represent the exception rather than the rule, and be supported by documentation indicating the reason for cessation of therapy. In high-risk procedures, an assessment of the indication for chronic salicylate therapy should be made; specifically, is the agent prescribed for primary or secondary cardiac prevention. Recognizing the importance of aspirin as a secondary preventive, a shared assessment and risk stratification should occur in conjunction with the managing physician and the proceduralist or consultant anesthesiologist. Among patient’s prescribed chronic salicylate therapy for primary prevention undergoing a high-risk procedure (as classified in Table 8.1), chronic salicylate therapy should stop 6 days prior to performing the elective interventional procedure and can be resumed 24 h after conclusion of the procedure . In patient’s prescribed chronic salicylate therapy for secondary prevention undergoing a high-risk procedure, chronic salicylate therapy should terminate only 4 days prior to performing an elective interventional procedure and can be resumed 24 h after conclusion of the procedure .
Nonaspirin NSAIDs reversibly inhibit cyclooxygenase, and the degree of inhibition of COX-1 parallels the increase of periprocedural bleeding risk. Because nonaspirin COX-1 NSAIDs inhibit platelet function reversibly, resumption of physiologic coagulation is dependent on the terminal half-life of the NSAID. NSAIDs such as meloxicam or etodolac have a stronger inhibitory effect on the COX-2 pathway and have a theoretical advantage of mitigating bleeding risk when compared to nonselective agents . The COX-2 selective agent celecoxib does not inhibit platelet aggregation or hemostasis at therapeutic or supratherapeutic clinical doses and does not increase surgical blood loss [32–35]. Neither acetaminophen nor celecoxib requires dose modification prior to administering a regional anesthetic or performing an interventional pain procedure deemed high, medium, or low risk.
Because NSAIDs do not confer a cardiac or cerebrovascular protective advantage, these agents can be readily stopped without consideration of their effects on cardiac or cerebrovascular risk prevention. For procedures stratified as posing an intermediate or low risk of bleeding (as classified in Table 8.1), no modification of chronic NSAID therapy is indicated prior to, or following, performance of an interventional pain procedure or administering a regional anesthetic. Nonaspirin NSAIDs should be discontinued prior to performing an elective high-risk interventional pain procedure based on the specific half-life of the agent. In general, a five half-life recommendation should be followed based upon the terminal pharmacokinetic elimination of agents at a steady state . Table 8.2 provides an indication of the duration of NSAID cessation prior to performing a high-risk elective interventional pain procedure. NSAID therapy can be resumed 24 h following completion of the procedure.
Recommendations for the duration of cessation of chronic NSAID therapy in patients undergoing high-risk elective interventional pain procedures
Approximate duration of five half-lives (days)
Warfarin affects coagulation by inhibiting the gamma carboxylation of the Vitamin K-dependent coagulation Factors II, VII, IX, X, Protein C, and Protein S. Because Factor VII has the shortest half-life, the initial anticoagulant effects following a single dose of warfarin are attributable to depletion of Factor VII . With ongoing use, all of the Vitamin K-dependent factors are inhibited once a steady-state drug concentration is reached. Response to warfarin dosing is heterogeneous and wide variations in therapeutic response are exhibited among treated patients, implicating warfarin as a notoriously challenging drug to initiate and maintain. This is compounded by the relatively narrow therapeutic index required for efficacy of this agent . It is known that select patient subsets, including the elderly and women, require less warfarin to achieve a therapeutic INR, though significant variability still occurs among matched patients .
A spinal hematoma developed after a single dose of warfarin administered in advance of neuraxial anesthetic placement in at least one elderly female undergoing total knee arthroplasty, though other authors posit that safe levels of hemostasis are observed during the first 12–16 h following warfarin administration [36, 39]. When considering a change to therapeutic anticoagulation , a procedure-related bleeding risk assessment should be carried out as described in Table 8.1. For low-risk procedures, several authors believe that these injections can be performed in the presence of a therapeutic INR <3.0, while others recommend a shared assessment be undertaken with the prescribing physician and the interventionalist [7, 40, 41]. A careful discussion of risk, benefits , and alternatives should be undertaken in all cases prior to performing intermediate- and high-risk procedures. Typically, warfarin should be stopped for 5 days prior to performance of an intermediate- or high-risk procedure, and an International Normalized Ratio should be quantified in advance of the injection. In patients who are at high risk for thrombus formation, consideration can be made for “bridge therapy”: a technique wherein a short-acting low-molecular-weight heparin is initiated in the interim when warfarin levels are waning. By utilizing an agent with a short half-life, the abstinence period of therapeutic periprocedural anticoagulation is shortened. Following low-, intermediate-, or high-risk procedures, warfarin therapy can be restarted 24 h after completing the interventional procedure (Table 8.3).
Commonly encountered anticoagulants and recommended period of abstinence for interventional procedures deemed high, intermediate, or low risk
Time interval to stop agent prior to procedure
Time to restart agent
High bleeding-risk proceduresa
Primary prophylaxis: 6 days
Shared risk assessment
Secondary prophylaxis: shared risk assessment
See Table 8.2
5 days, normalized INR
5 days, normalized INR
Shared risk assessment
Low-molecular-weight heparin: prophylactic dosing
12–24 h for medium-/high-risk procedures
4 h for low-risk procedures
Low-molecular-weight heparin: therapeutic dosing
12–24 h for medium-/high-risk procedures
4 h for low-risk procedures
Shared risk assessment
4–5 days +
Shared risk assessment
Shared risk assessment
Clinically used heparins are available in an unfractionated or a low-molecular-weight form (enoxaparin, dalteparin) and can be administered subcutaneously or intravenously. Unfractionated heparin inactivates Factor IIa, Factor Xa, and Factor IXa and its anticoagulant effect can be reversed with the administration of protamine . The incidence of spinal hematoma is increased when patients are heparinized within 1 h of dural puncture, are administered aspirin concomitantly, or experience a traumatic dural puncture . ASRA recommends that intravenous heparin be stopped 2–4 h prior to a neuraxial intervention, and that heparin be avoided for at least 1 h after placement or removal of a neuraxial catheter [6, 43]. Regional anesthesiologists are more likely to encounter patients treated with intravenous heparin than their interventional pain physician counterparts given the typically elective nature of chronic pain procedures. Intravenous heparin should be stopped for 4 h prior to a low-, medium-, or high-risk procedure as defined in Table 8.1. Subsequently, intravenous heparin can be restarted as soon as 2 h following the procedure, or 24 h later if an intermediate- or high-risk procedure was performed and the neuraxial intervention was noted to be bloody . Subcutaneous heparin at a dose of 5000 units 2–3 times a day inhibits coagulation via factor Xa. The anticoagulant effects of subcutaneous heparin are observed within 1 h of administration and dissipate 6 h later. In most subcutaneous heparin dosing regimens, the PTT remains within the normal range, and the ASRA guidelines do not dictate a contraindication to placement of neuraxial anesthetics on patients receiving subcutaneous heparin [6, 44]. Despite this observation, at least two cases of spinal hematoma have been reported in patients receiving subcutaneous heparin [45, 46]. For this reason, in parallel with the elective nature of interventional pain procedures, the consensus committee broadly recommends an 8–10 h period of abstinence from subcutaneous heparin prior to an interventional procedure deemed low, medium, or high risk. Subsequently, subcutaneous heparin can be restarted 2 h following the injection (Table 8.3) .
Low-molecular-weight heparin has a more predictable bioavailability than standard heparin and demonstrates a dose-dependent antithrombotic effect mediated by the inhibition of Factor Xa. Rarely is laboratory monitoring of factor Xa activity required because of predictable bioavailability, making low-molecular-weight heparin comparatively easy to dose when compared to conventional unfractionated heparin. Low-molecular-weight heparins can be administered once daily or every 12 h for thromboembolic prophylaxis or therapeutic anticoagulation , respectively. A cumulative review of data demonstrates increased risk of bleeding complications in conjunction with low-molecular-weight heparins among females, the elderly, those with anomalies of the spinal cord or vertebral column, renal insufficiency, those with an indwelling catheter, or those in whom neuraxial cannulation was difficult or bloody [3, 47]. The ASRA guidelines recommend a 12 h interval after prophylactic low-molecular-weight heparin is administered before performing a neuraxial intervention or a 24 h period of abstinence if therapeutic anticoagulation doses are utilized. If blood is encountered during the placement of a neuraxial anesthetic, the guidelines recommend abstaining from a subsequent dose of low-molecular-weight heparin for an additional 24 h (Table 8.3). Should a neuraxial catheter be utilized, a minimum interval of 4 h should elapse before removing the neuraxial catheter based upon an FDA Safety Communication published since the 2010 ASRA guidelines [6, 48]. More conservatively, and supported by a multispecialty group, 12 h should elapse between dosing prophylactic low-molecular-weight heparin and the performance of procedures dictated as low, intermediate, or high risk. Similarly, an additional 12 h should elapse between resuming low-molecular-weight heparin. When therapeutic low-molecular-weight heparin dosing is utilized, a period of 24 h should elapse prior to, and following procedures characterized by low, intermediate, or high risk .
ADP Receptor Blockers
The active metabolites of clopidogrel and prasugrel irreversibly block ADP receptors on the platelet surface, thereby preventing activation of the GPIIb/IIIa receptor complex and reducing platelet aggregation. These agents are becoming increasingly adopted in the treatment of coronary vascular disease, cerebrovascular ischemia, and peripheral vascular disease. Because these agents offer cardiac and/or cerebrovascular protection, there remains risk with abrupt discontinuation of ADP receptor blockers, and an assessment of the bleeding risk posed by the interventional procedure should be quantified as dictated in Table 8.1. For low-risk procedures, no change to the dosing strategy is indicated prior to or following the injection. Prior to undertaking a medium- or high-risk procedure, consultation with the prescribing physician should provide a bleeding risk assessment and account for age, comorbidities, and concomitant antiplatelet agents administered in conjunction with the ADP receptor blocker . Among most patients slated for a medium- or high-risk procedure (Table 8.3), a period of clopidogrel abstinence spanning 7 days should precede the injection. Among patients deemed high risk for thromboembolic event and undergoing a medium or high bleeding risk injection, a 5-day clopidogrel abstinence period should precede the administration of a regional anesthetic or interventional pain procedure. Following the injection, 12 h should elapse before resuming the usual daily dose of clopidogrel. Among patients undergoing medium- or high-risk procedures on prasugrel, the abstinence interval should span 7–10 days .
A new generation of anticoagulants continues to enter the marketplace and carry the advantage that INR monitoring is not routinely required. These agents, including apixaban , dabigatran, and rivaroxaban, are not subject to fluctuating pharmacodynamics based upon dietary intake, and there are no specific antagonists to universally reverse the anticoagulant effects of these agents [50–52]. Like warfarin, these newer anticoagulants confer cardiac protection and may pose a risk of thrombosis during a period of abstinence (Table 8.3); hence a careful assessment in collaboration with the prescribing physician is indicated prior to performing an elective regional anesthetic or recommending changes to the chronic use of these agents. Because of the relatively recent introduction of these newer anticoagulants, a substantial body of clinical evidence demonstrating bleeding risk is absent from the literature, though this should not be misinterpreted as a demonstration of their long-term safety given that the incidence of spinal or epidural hematomas is infrequent, and an incident case may not have yet manifested. In the absence of a substantial body of data, the broad recommendation is to discontinue the agent for five half-lives prior to an injection carrying bleeding risk. The ideal time to resume these anticoagulant agents following a neuraxial injection is also unknown, though authors typically recommend a 24–48 h interval following the interventional procedure [50, 53, 54].