Interventional Pain Procedures in Patients on Anticoagulants


The American Society of Regional Anesthesia (ASRA) guidelines on interventional pain procedures are more stringent compared to their guidelines on regional anesthesia. This is because of the type of patients seen in the pain clinic and the surgical nature of some of the interventions. The recommendations in the two guidelines are basically the same with respect to the older anticoagulants but different with regards to aspirin, NSAIDs, and the novel oral anticoagulants (NOACs). For the NOACs, a five half-life interval between discontinuation of the anticoagulant and the planned procedure is recommended to assure near complete elimination of the drug. Twenty four hours are usually recommended before the drug is resumed after the procedure. There have been advances in the laboratory monitoring and reversal of the NOACs.


anticoagulation, guidelines, hematoma, interventional pain procedures, monitoring, reversal



The American Society of Regional Anesthesia and Pain Medicine (ASRA) has published guidelines on regional anesthesia in patients on anticoagulants. These guidelines have been used by clinicians in the United States and as a template for guidelines of regional anesthesia organizations in other countries. The latest edition (the third edition) of the regional anesthesia guidelines was published in 2010 and is presently being updated. In the absence of guidelines on interventional procedures to guide them, pain physicians have used the ASRA regional guidelines in their practice. Unfortunately, there have been case reports of spinal anesthesia after epidural steroid injections (ESIs) or spinal cord stimulator placements not only when the regional anesthesia guidelines were followed but also when coagulation studies were normal ( Table 85.1 ).

TABLE 85.1

Reports of Spinal Hematoma After Epidural Steroid Injections or Spinal Cord Stimulator Placements

Report Patient Profile Outcome Other Drugs; Comment
Williams et al. 63, male, CESI Full recovery Indomethacin and allopurinol
Ghaly 57, male, CESI Full recovery Diclofenac, amitriptyline
Reitman and Watters 62, female, CESI Partial recovery Normal coagulation studies
Ain and Vance 85, female, LESI Motor recovery, residual numbness of left toes Warfarin resumed the night after ESI, enoxaparin restarted 24 h later, INR 1.2 the next day
Xu et al. 78, female, LESI Full recovery Warfarin restarted 8 h after ESI, enoxaparin resumed 30 h later
Giberson et al. 2 cases: 53, male; 70, male Patient 1: Residual left leg weakness
Patient 2: Complete recovery
Patient 1 took aspirin the day the leads were pulled; patient 2 had not taken aspirin for 7 days before the trial
Buvanendran and Young 73, female Complete recovery Took low-dose aspirin daily
Takawira et al. 52, male Spontaneous recovery Patient not on anticoagulant. Symptoms on third day of trial; hematoma possibly from lead migration.
Smith et al. 2 cases: 44, male; 66, female Patient 1: No recovery
Patient 2: Residual numbness below T8
Coagulation studies normal
Kloss et al. 50, male Partial recovery No mention of medications or recovery

CESI, Cervical epidural steroid injection; LESI, lumbar epidural steroid injection.

The occurrence of spinal hematoma even when the coagulation studies were normal or when the ASRA guidelines were adhered to is due to several reasons. Pain patients are elderly, a known risk factor in exaggerated response to anticoagulants. Older patients have spinal stenosis wherein the epidural space is decreased. Smaller amounts of hematoma can result in compression of the spinal cord, and progression of symptoms can be more rapid because of the diminished epidural space. The presence of numbness and/or weakness in patients with radiculitis, before the injection, confuses the clinical picture of spinal hematoma. Low back pain patients who had laminectomy can have scarring, further limiting the space where the hematoma can spread. With regard to the placement of spinal cord stimulators, large-bore needles are placed and several advancements and retractions are required to optimally place the electrodes. These result in a higher risk of trauma to the vascular structures. Finally, pain clinic patients may be on medications usually considered as not altering the clotting of blood. These drugs include the anticonvulsants carbamazepine, oxcarbazepine, valproate, and levetiracetam, and the selective serotonin reuptake inhibitors (SSRIs). In addition to these medications, elderly pain patients may be on antiplatelets for cardiovascular or central nervous system (CNS) thrombotic/embolic prophylaxis. The intake of several antiplatelet drugs is a known risk of spinal hematoma after a neuraxial injection.

Aware of the case reports of spinal hematoma when their guidelines on regional anesthesia were adhered to, the Board of Directors of the ASRA commissioned a Writing Committee to promulgate guidelines that are applicable to interventional pain procedures. These guidelines were endorsed by the European Society of Regional Anesthesia and Pain Therapy, American Academy of Pain Medicine, International Neuromodulation Society, North American Neuromodulation Society, and the World Institute of Pain.

In this chapter, the following topics will be discussed: stratification of the risks of bleeding according to the type of procedure, antiplatelet medications, older anticoagulants, novel oral anticoagulants (NOACs), antidepressants, and herbal medications. Differences between the ASRA regional and interventional pain procedures will be pointed out. The GPIIb/IIIa inhibitors will not be discussed, as at present the use of these drugs is rare and, if used, the patients are usually placed on P2Y12 inhibitors for maintenance therapy.

Stratification of Risks

One distinguishing feature that the pain guidelines offer, in contrast to the regional guidelines, is stratification of the risks of bleeding according to the type of pain procedure ( Table 85.2 ). This is because the risks of bleeding, progression, and consequence of the hematoma from lack of compressibility vary with the different procedures. On one hand, the abdominal and pelvic sympathetic blocks are close to large vascular structures, cannot be easily compressed, and symptoms associated with hematoma from these injections are vague. Intravertebral and perivertebral procedures are associated with compression of the spinal cord with devastating consequences. On the other hand, trigger point injections, joint injections, and superficial nerve blocks (e.g., occipital nerve blocks in the scalp) are avascular, superficial, and associated with minor bleeding complications. Mild anticoagulation is probably acceptable when the risk and consequences are small. This is to prevent the patient from having a stroke, myocardial infarction, or venous thromboembolism (VTE).

TABLE 85.2

Pain Procedures According to the Potential Risks for Serious Bleeding a

High-Risk Procedures Intermediate-Risk Procedures Low-Risk Procedures
SCS trial and implant Interlaminar ESIs Peripheral nerve blocks
IT catheter and pump implant Facet MBB and RFA Peripheral joints and musculoskeletal injections
Vertebroplasty, kyphoplasty Paravertebral blocks Trigger point and piriformis injections
Epiduroscopy and epidural decompression Intradiscal procedures Sacroiliac joint injections
Sympathetic blocks Sacral LBBs
Pocket revision and IPG/ITP replacement

ESIs, Epidural steroid injections; IPG, intrathecal pump generator; LBB, lateral branch block; MBB, medial branch block; RFA, radiofrequency ablation.

a Patients who are at high risk for bleeding undergoing low- or intermediate-risk procedures should be treated as intermediate or high risk, respectively. Patients with high risk for bleeding include the elderly, concurrent use of anticoagulants, low body weight, or advanced liver and renal disease.

Aspirin Phosphodiesterase Inhibitors and Nonsteroidal Antiinflammatory Drugs

There are several mechanisms of aspirin’s antiplatelet effect. Aspirin preferentially blocks COX-1 over COX-2. It irreversibly inactivates COX-1 through the acetylation of the amino acid serine. By inactivating COX-1 and blocking thromboxane production, aspirin inhibits platelet activation, platelet aggregation, and thrombosis. Aspirin inactivates the COX-1 in the megakaryocytes in the bone marrow, and megakaryocytes are responsible for platelet production. The drug also influences coagulation through non-TXA2-mediated effects. These include impairment of secondary hemostasis and stability of the thrombus by acetylating fibrinogen and enhancing fibrinolysis. Unlike nonsteroidal antiinflammatory drugs (NSAIDs), aspirin decreases thrombin formation. At higher doses, aspirin prevents endothelial cell prostacyclin production by inhibiting COX-2. Prostacyclin not only inhibits platelet coagulation but also stimulates vasodilation.

Aspirin (ASA) is rapidly absorbed from the gastrointestinal tract, peak levels occur 30 minutes after ingestion, and significant platelet inhibition is present at 1 hour. For enteric-coated aspirin, peak plasma levels may be delayed until 3–4 hours after ingestion. COX activity does not return for 48 hours after aspirin intake, and this has been interpreted as the duration of effect of aspirin on megakaryocytes. The average lifespan of a platelet is 7–10 days. Approximately 10% of the circulating platelet pool is replaced every day. This means that at 5 days, approximately 50% of the circulating platelets are not affected by the aspirin.

Similar to the ASRA guidelines on regional anesthesia, the patient can continue to take the aspirin when low-risk pain procedures are performed. However, the two guidelines have different recommendations with regard to neuraxial injections. Whereas the regional guidelines allow neuraxial injections, the pain guidelines do not. This is in view of the case reports of spinal hematoma after ESI in patients on aspirin (see Table 85.1 ). The length of discontinuation depends on the reason for the patient’s taking the aspirin: 6 days for primary prevention (aspirin in patients with no overt cardiovascular disease) and 4 days for secondary prophylaxis. The drug can be resumed 24 hours later.

The commonly used phosphodiesterase (PDE) inhibitors are dipyridamole and cilostazol. These drugs impede PDE isoenzymes expressed by platelets. PDE inhibitors thwart cyclic adenosine monophosphate (cAMP) and cyclic guanosine monophosphate (cGMP) levels. PDE-3 inhibitors increase cAMP levels, whereas PDE-5 inhibitors increase cGMP levels. Cilostazol inhibits PDE-3, while dipyridamole inhibits PDE-3 and PDE-5; dipyridamole has been combined with aspirin for a more pronounced effect. By blocking the secondary messengers cAMP, cGMP, and PDEs, these drugs reduce platelet aggregation and increase vasodilatation. These drugs decrease platelet aggregation by approximately 40%. Dipyridamole has a half-life of 10 hours. The extended-release dipyridamole, used in combination with aspirin, has a half-life of 13.6 hours. Cilostazol reaches peak plasma concentrations at 2 hours, and maximum platelet aggregation occurs at 6 hours. Similar to dipyridamole, cilostazol has an elimination half-life of 10 hours. Five half-lives, where 97% of the drug is eliminated, is equivalent to 50 hours. A discontinuation of 2 days should thus be adequate, and the drugs can be resumed 24 hours later.

NSAIDs, like aspirin, inhibit the COX enzyme. Such a blockade is reversible, compared with the irreversible effect of aspirin. The Writing Committee used the drug’s half-life to guide the duration of discontinuation of the NSAID before a pain procedure. As noted previously, an interval of five half-lives is adequate, as this represents 97% elimination of the drug. This results in a discontinuation of 1–2 days for ibuprofen, diclofenac, ketorolac, and indomethacin; 4 days for naproxen and meloxicam; 6 days for namebutone; and 10 days for oxaprozin and piroxicam. The drug can be resumed a day after the procedure ( Table 85.3 ).

TABLE 85.3

Duration of Discontinuation of Aspirin, Phosphodiesterase Inhibitors, and Nonsteroidal Antiinflammatory Drugs

From Patrick J, Dillaha L, Armas D, Sessa WC: A randomized trial to assess the pharmacodynamics and pharmacokinetics of a single dose of an extended-release aspirin. Postgrad Med . 127:573–580, 2015.

Drug Half-Life (Hours) Discontinuation, in Days, Based on Five Half-Lives
Aspirin 20–30 min a 4 for secondary prophylaxis, 6 for primary prophylaxis
Cilostazol 10 h 2
Dipyridamole 10 h 2
Diclofenac 1–2 1
Ibuprofen 2–4 1
Ketorolac 5–6 1
Etodolac 6–8 2
Indomethacin 5–10 2
Naproxen 12–17 4
Meloxicam 15–20 4
Namebutone 22–30 6
Oxaprozin 40–60 10
Piroxicam 45–50 10

a Aspirin has an irreversible effect on platelets; stoppage of drug is based on platelet recovery.

P2Y12 Inhibitors Clopidogrel, Prasugrel, and Ticagrelor

The P2Y12 receptors are mostly found on the surface of platelets. ADP released from platelet-dense granules binds to two platelet G-protein-coupled receptors, the P2Y 1 and P2Y 12 receptors. P2Y 1 is a Gq-coupled receptor that initiates ADP-induced platelet aggregation via the stimulation of phospholipase C and phosphatidylinositol-signaling pathway. P2Y 12 is a Gi-coupled seven-transmembrane domain receptor, which mediates platelet activation by inhibiting the adenylate cyclase-mediated signaling pathway and decreasing intracellular cAMP levels. The decrease in intracellular cAMP levels and activation of the GPIIB/IIIa receptor results in reduced platelet aggregation and increased vasodilation. P2Y 12 is also involved in platelet secretion induced by strong agonists. The significant role of the P2Y 12 receptor in platelet activation and formation of a stable thrombus has made it an important target in the management and prevention of arterial thrombosis. The P2Y12 inhibitors include clopidogrel, prasugrel, ticagrelor, and cangrelor. Patients who have cardiovascular problems are usually prescribed combined P2Y12 inhibitor and aspirin, the so-called “dual antiplatelet therapy,” complicating the stoppage of these drugs before a pain procedure.

Clopidogrel and prasugrel are irreversible P2Y12 inhibitors. Both are prodrugs, requiring a two-step metabolism to be converted to their active metabolite. Clopidogrel is susceptible to genetic polymorphisms, and 9% of patients do not respond to the drug. Clopidogrel causes a 60% inhibition of platelets, while prasugrel results in 90% inhibition. For this reason, a 7-day stoppage is recommended for medium and high-risk procedures ( Table 85.4 ). For clopidogrel specifically, the European and Scandinavian guidelines respectively recommended a 7- and 5-day stoppage of clopidogrel before regional anesthesia. A 5-day duration may be reasonable, as a study showed that most patients have no platelet inhibition after 5 days, and the rest had minimal inhibition. Since a trial of spinal cord stimulator entails several days, most pain clinicians stop the clopidogrel for 5 days. A test of platelet activity should be performed if only a 5-day stoppage is observed to document that platelet recovery is significant. The commonly used tests of antiplatelet activity of the P2Y12 receptors include either the VerifyNow P2Y12 assay or the platelet mapping portion of the thromboelastography.

TABLE 85.4

Recommended Intervals Between Stoppage and Resumption of the P2Y12 Inhibitors and Medium- and High-Risk Pain Procedures

Drug Interval Between Discontinuation and Pain Procedure Resumption of the Drug
Clopidogrel 7 days, 5 days if platelet function studies show adequate recovery 12 h when loading dose is given, 24 h for the usual maintenance daily dose
Prasugrel 7 days 24 h
Ticagrelor 5 days 24 h
Cangrelor 1–2 h a

a Patients who are given cangrelor are usually maintained on oral P2Y12 inhibitors.

Ticagrelor, in contrast to clopidogrel and prasugrel, is a direct-acting P2Y12 inhibitor. It inhibits the aggregation of platelets by 90%; platelet function recovery is adequate 5 days after stopping the drug. Cangrelor, another direct P2Y12 inhibitor, is given intravenously. The drug is approved for clinical use in patients undergoing percutaneous coronary intervention. It has a fast onset of a few minutes, and its half-life is 3–7 minutes. Platelet function has been shown to normalize 1 hour after it is stopped. If a pain procedure has to be done, at least a 1-hour interval should be observed. This situation is rarely encountered, as patients are usually maintained on oral P2Y12 inhibitors.

Prasugrel and ticagrelor should be resumed 24 hours after the procedure, as these drugs take effect within 2–4 hours. Clopidogrel, if given in its usual 75-mg dose, can be started 12 hours after the pain procedure, since it takes 1 or 2 days to take full effect. However, a 300–600-mg loading dose of clopidogrel takes effect after a few hours and should be started 24 hours later.

Older Anticoagulants: Warfarin, Heparin, Low-Molecular-Weight Heparin, Fibrinolytic Agents

Warfarin inhibits the γ-carboxylation of the vitamin K–dependent coagulation factors. These factors have different half-lives: 6–8 hours for factor VII, 20–24 hours for factor IX, 20–42 hours for factor X, and 48–120 hours for factor II. The inhibition of clotting factor VII causes the initial anticoagulation from warfarin. This effect, however, is antagonized by a decrease in anticoagulant protein C, making the International Normalized Ratio (INR) unreliable during the early phase of warfarin therapy. The full anticoagulant effect of warfarin does not occur until 4 days later, when the levels of factor II are significantly decreased. Concentrations of clotting factors of 40% or more are considered adequate for hemostasis ; levels below 20% are associated with bleeding. While warfarin is used in the United States, acenocoumarol is prescribed in Europe. The difference is the normalization of the INR between the two drugs: 5 days for warfarin and 3 days for acenocoumarol.

The ability of warfarin to affect the extrinsic pathway of the coagulation cascade (by inhibiting the activity of factor VII), the intrinsic pathway (factor IX), factor X (junction of the two pathways), and the end product of coagulation (factor II) makes it a powerful anticoagulant. However, warfarin is difficult to dose, as it has a narrow therapeutic index, there is wide interpatient dosing variability, and genetic factors account for a large proportion of the variations in dose requirements. Although patients with variations in their CYP2C9 and/or VKORC1 require lower doses of warfarin, studies on genetic-based dosing were not uniform in their findings. The American College of Cardiology recommended against pharmacokinetic-based dosing, pending clinical studies.

In patients who have not taken warfarin, an INR of 1.4 is acceptable for neuraxial injections. However, in patients who took warfarin, the ASRA guidelines recommend normalization of the INR (1.2 or less) after stoppage of warfarin for 5 days before medium- and high-risk procedures are performed. This is in contrast to the European and Scandinavian guidelines that accept an INR of 1.4 or less as long as the warfarin was stopped for 5 days. A slightly prolonged INR of 1.3–1.4 may not be safe, as a study showed the levels of clotting factor X and II to be below 40% in two patients.

A low-molecular-weight heparin (LMWH) “bridge” therapy may be considered in patients who are at high risk for VTE. A study in patients who had atrial fibrillation, however, showed that the discontinuation of warfarin without bridge therapy was noninferior to bridge therapy with dalteparin. The incidence of arterial thromboembolism was not different between the two groups, while the incidence of major bleeding was significantly higher in the bridging group. This study may only be applicable to patients who have atrial fibrillation, the subject of the study, and not to patients with prosthetic cardiac valves. Another study showed patients who were bridged before their total hip arthroplasty had a slower healing time (in terms of their wounds) and also had a longer hospital stay.

Heparin inactivates thrombin (factor IIa), factor IXa, and Xa. The anticoagulant effect of intravenous heparin is immediate, whereas subcutaneous heparin takes 1 hour. Its therapeutic effect ceases 4–6 hours after its administration. The half-life of heparin increases with increased dose: 30 minutes for the 25 U/kg dose, 60-minute half-life of 100 U/kg, and 150 minutes for the 400 U/kg dose. Monitoring is via the activated partial thromboplastin time (aPTT), and therapeutic anticoagulation is achieved when the aPTT is 1.5–2.5 times the initial value. Reversal is achieved with protamine, with the dose being 1 mg of protamine per 100 U of heparin.

The performance of pain procedures in patients on intravenous heparin should be avoided, since interventional pain procedures are usually elective procedures. Noninterventional procedures such as medications (opioids, anticonvulsants, antidepressants) and adjunctive therapy can be performed to alleviate the patient’s pain. If an interventional procedure has to be done, it can be done at least 5 hours after the IV heparin is stopped. The 5-hour interval is based on the five half-lives (60-minutes) of the usual 100 U/kg clinical dose of intravenous heparin. As the half-lives of intravenous heparin varies with the dose, the interval between the last intravenous heparin dose and the interventional pain procedure should be adjusted accordingly.

The ASRA guidelines on regional anesthesia allow the continued use of BID subcutaneous heparin, while the pain guidelines recommended that interventional procedures not be performed until at least 8 hours after the subcutaneous heparin was given. This 8 hours is longer than the five-half-life (5–6 hours) of the drug. This was to accommodate patients with exaggerated response to heparin and the slightly more prolonged absorption of subcutaneous administration. The drug can be restarted a minimum of 2 hours after the procedure. Interventional pain procedures are not recommended in patients on TID subcutaneous heparin. Two studies showed the probable safety of doing neuraxial interventions with patients on subcutaneous TID heparin. No epidural hematomas occurred after removal of epidural catheters in 714 patients on TID heparin, and 20 patients had PTTs greater than 35 seconds when their epidurals were removed. Another study showed no spinal hematoma after thoracic epidurals in 928 patients; activated thromboplastin time was greater than 40 seconds in 115 patients. The small number of patients in these two studies is not adequate to change the recommendations.

For LMWH, the ASRA pain (and regional) guidelines recommend a 12-hour interval between stoppage of prophylactic dose of enoxaparin and interventional pain procedures. For therapeutic doses of enoxaparin (i.e., 1 mg/kg) and for dalteparin, a 24-hour interval is recommended. The drug can be resumed 4 hours after a low-risk pain procedure, an interval similar recommended by a recent FDA Drug Safety Communication.

In cases where the moderate- or high-risk procedures are bloody, the resumption of warfarin, heparin, and LMWH should be delayed for 24 hours after the procedure (see Table 85.5 ). As in the regional anesthesia guidelines, drugs that affect hemostasis (antiplatelet, NSAIDs, SSRIs, and other anticoagulants) should preferably not be concomitantly used in patients on these anticoagulants.

TABLE 85.5

Recommended Intervals Between Stoppage and Resumption of the Old Anticoagulants Interventional Pain Procedures

Drugs Drug Discontinuation and Pain Procedure Resumption of Drug
Warfarin 5 days, INR back to normal 24 h
IV heparin 4 h 2–24 h
BID subcutaneous heparin 8–24 h 2–4 h
TID subcutaneous heparin a
LMWH, prophylactic dose 12 h 4–24 h
LMWH, therapeutic dose 24 h 4–24 h
LMWH, dalteparin 24 h 4–24 h
Fibrinolytic agents a

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

Sep 21, 2019 | Posted by in PAIN MEDICINE | Comments Off on Interventional Pain Procedures in Patients on Anticoagulants
Premium Wordpress Themes by UFO Themes